Fetal Cardiology: Embryology, Genetics, Physiology, Echocardiographic Evaluation, Diagnosis and Perinatal Management of Cardiac Diseases (2nd Edition) (Series in Maternal-Fetal Medicine)

Yagel Silverman Gembruch

Fetal Cardiology

This second edition is, like the first, the product of the combined efforts of many professionals – obstetricians, pediatric cardiologists, sonographers, molecular biologists, and medical physicists – and is a comprehensive guide intended for anyone interested in fetal organ scanning. The field of fetal ultrasonographic and echocardiographic imaging and fetal cardiac therapies and interventions is rapidly changing and developing, and over the very few years between editions considerable strides have been made in all aspects of fetal cardiology. Many chapters have therefore been added – including cardiac impact of twin-to-twin transfusion syndrome, cerebral outcomes in CHD, and congestive heart failure – and virtually all chapters have been updated. In two areas advances have been particularly noteworthy: the identification of additional genes implicated in congenital heart disease, and the advances in three- and four-dimensional ultrasound capabilities in fetal cardiac imaging, which has opened exciting avenues for clinical applications and research endeavors.

Fetal Cardiology

Second Edition Simcha Yagel MD Department of Obstetrics & Gynecology, Hadassah University Hospital, Mount Scopus, Jerusalem, Israel

Norman H. Silverman MD DSc FACC Division of Pediatric Cardiology, Lucile Packard Children’s Hospital, Stanford University Medical Center, Palo Alto, California, USA

Ulrich Gembruch MD PhD Department of Obsterics and Prenatal Medicine, Center of Obstetrics and Gynecology, University of Bonn, Germany

Fetal Cardiology

Embryology, Genetics, Physiology, Echocardiographic Evaluation, Diagnosis and Perinatal Management of Cardiac Diseases

An integral DVD contains over 50 video clips illustrating points in the text.

www.informahealthcare.com

Embryology, Genetics, Physiology, Echocardiographic Evaluation, Diagnosis and Perinatal Management of Cardiac Diseases

Edited by Second Edition

Published in association with the Journal of Maternal-Fetal & Neonatal Medicine.

Includes DVD

Simcha Yagel Norman H Silverman Ulrich Gembruch

Second Edition

Fetal Cardiology

SERIES IN MATERNAL-FETAL MEDICINE Published in association with the Journal of Maternal-Fetal & Neonatal Medicine Editors-in-Chief: Gian Carlo Di Renzo and Dev Maulik

Available 1 Howard Carp, Recurrent Pregnancy Loss: Causes, Controversies and Treatment ISBN 9780415421300 2 Vincenzo Berghella, Obstetric Evidence Based Guidelines ISBN 9780415701884 3 Vincenzo Berghella, Maternal-Fetal Evidence Based Guidelines ISBN 9780415432818 4 Moshe Hod, Lois Jovanovic, Gian Carlo Di Renzo, Alberto de Leiva, Oded Langer, Textbook of Diabetes and Pregnancy, second edition ISBN 9780415426206

Of related interest Joseph J Apuzzio, Anthony M Vintzelos, Leslie Iffy, Operative Obstetrics ISBN 9781842142844 Isaac Blickstein, Louis G Keith, Prenatal Assessment of Multiple Pregnancy ISBN 9780415384247 Tom Bourne, George Condous, Handbook of Early Pregnancy Care ISBN 9781842143230 Gian Carlo Di Renzo, Umberto Simeoni, The Prenate and Neonate: The Transition to Extrauterine Life ISBN 9781842140444 Asim Kurjak, Guillermo Azumendi, The Fetus in Three Dimensions: Imaging, Embryology and Fetoscopy ISBN 9780415375238 Asim Kurjak, Frank A Chervenak, Textbook of Perinatal Medicine, second edition ISBN 9781842143339 Catherine Nelson-Piercy, Handbook of Obstetric Medicine, third edition ISBN 9781841845807 Dario Paladini, Paolo Volpe, Ultrasound of Congenital Fetal Anomalies ISBN 9780415414449 Donald M Peebles, Leslie Myatt, Inflammation and Pregnancy ISBN 9781842142721 Felice Petraglia, Jerome F Strauss, Gerson Weiss, Steven G Gabbe, Preterm Birth: Mechanisms, Mediators, Prediction, Prevention and Interventions ISBN 9780415392273 Ruben A Quintero, Twin-Twin Transfusion Syndrome ISBN 9781842142981 Baskaran Thilaganathan, Shanthi Sairam, Aris T Papageorghiou, Amor Bhide, Problem Based Obstetric Ultrasound ISBN 9780415407281

Fetal Cardiology Embryology, Genetics, Physiology, Echocardiographic Evaluation, Diagnosis and Perinatal Management of Cardiac Diseases Second Edition Edited by Simcha Yagel MD Obstetrics and Gynecology Ultrasound Center Center for Human Placenta Research Department of Obstetrics and Gynecology Hadassah – Hebrew University Medical Centers Mount Scopus and Department of Obstetrics and Gynecology Hebrew University – Hadassah Faculty of Medicine Jerusalem, Israel Norman H Silverman MD Dsc FACC Division of Pediatric Cardiology Lucile Packard Children’s Hospital Stanford University Medical Center Palo Alto, California, USA Ulrich Gembruch MD PhD Department of Obstetrics and Prenatal Medicine Center of Obstetrics and Gynecology University of Bonn Bonn, Germany Associate Editor Sarah Margalyt Cohen Department of Obstetrics and Gynecology Hadassah – Hebrew University Medical Centers Mount Scopus Jerusalem, Israel

Informa Healthcare USA, Inc. 52 Vanderbilt Avenue New York, NY 10017 © 2009 by Informa Healthcare USA, Inc. Informa Healthcare is an Informa business No claim to original U.S. Government works Printed in the United States of America on acid-free paper 10 9 8 7 6 5 4 3 2 1 Library of Congress Cataloging-in-Publication Data Fetal cardiology : embryology, genetics, physiology, echocardiographic evaluation, diagnosis, and perinatal management of cardiac diseases / edited by Simcha Yagel, Norman H. Silverman, Ulrich Gembruch ; associate editor, Sarah Margalyt Cohen. – 2nd ed. p. ; cm. – (Series in maternal-fetal medicine) Includes bibliographical references and index. ISBN 978-0-415-43265-8 (hb : alk. paper) 1. Fetal heart–Diseases. 2. Fetal heart. I. Yagel, Simcha. II. Silverman, Norman H. III. Gembruch, Ulrich. IV. Series. [DNLM: 1. Heart Diseases–diagnosis. 2. Heart Diseases–therapy. 3. Fetal Heart– physiopathology. 4. Infant, Newborn. 5. Prenatal Diagnosis. WS 290 F4193 2009] RG618.F465 2009 618.3'261–dc22 2008049158

This book contains information obtained from authentic and highly regarded sources. Reprinted material is quoted with permission, and sources are indicated. A wide variety of references are listed. Reasonable efforts have been made to publish reliable data and information, but the author and the publisher cannot assume responsibility for the validity of all materials or for the consequence of their use. No part of this book may be reprinted, reproduced, transmitted, or utilized in any form by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying, microfilming, and recording, or in any information storage or retrieval system, without written permission from the publishers. For permission to photocopy or use material electronically from this work, please access www.copyright.com (http://www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC) 222 Rosewood Drive, Danvers, MA 01923, 978-750-8400. CCC is a not-for-profit organization that provides licenses and registration for a variety of users. For organizations that have been granted a photocopy license by the CCC, a separate system of payment has been arranged. Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. For Corporate Sales and Reprint Permissions call 212-520-2700 or write to: Sales Department, 52 Vanderbilt Avenue, 16th floor, New York, NY 10017. Visit the Informa Web site at www.informa.com and the Informa Healthcare Web site at www.informahealthcare.com

Composition by Exeter Premedia Services Pvt Ltd, Chennai, India Printed and bound in the United States of America.

To our wives, Neomie Yagel, Gabi Gembruch, and Heather Silverman, this volume is lovingly dedicated. ‘The only labor worth laboring for is a labor of love.’ —JS Gillette

Contents

List of video clips List of contributors Preface

x xii xvi

1. Cardiovascular development Arnold CG Wenink

1

2. Cardiac morphogenesis Adriana C Gittenberger–de Groot and Robert E Poelmann

9

3. Cardiac anatomy and examination of specimens Cornelia Tennstedt

19

4. Placental implantation and development Simcha Yagel and Debra S Goldman-Wohl

27

5. Placental circulations Eric Jauniaux and Graham J Burton

41

6. The physics of ultrasound imaging Zvi Friedman

57

7. Technical advances in fetal echocardiography Boris Tutschek and David Sahn

83

8. Epidemiology of congenital heart disease: etiology, pathogenesis, and incidence Julien IE Hoffman

101

9. Indications for fetal echocardiography: screening in low- and high-risk populations Ulrich Gembruch and Annegret Geipel

111

10.

Circulation in the normal fetus and cardiovascular adaptations to birth Abraham M Rudolph

131

11.

Development of fetal cardiac and extracardiac Doppler flows in early gestation Ahmet A Baschat and Ulrich Gembruch

153

12.

The examination of the normal fetal heart using two-dimensional echocardiography Rabih Chaoui

173

13.

First and early second trimester fetal heart screening Simcha Yagel, Sarah M Cohen, Baruch Messing, and Reuwen Achiron

185

14.

Four-dimensional ultrasound examination of the fetal heart by spatiotemporal image correlation Luís F Gonçalves, Jimmy Espinoza, Juan Pedro Kusanovic, Wesley Lee, and Roberto Romero

197

15.

Three- and four-dimensional ultrasound in fetal echocardiography: a new look at the fetal heart Simcha Yagel, Sarah M Cohen, Israel Shapiro, and Dan V Valsky

219

16.

Cardiac malpositions and syndromes with right or left atrial isomerism Rabih Chaoui

239

viii

Contents

17.

Anomalies of the right heart Jean-Claude Fouron

251

18.

Pulmonary atresia with intact ventricular septum Julene S Carvalho

269

19.

Intracardiac shunt malformations Einat Birk and Norman H Silverman

281

20.

Left heart malformations Lindsey D Allan

291

21.

Ventricular outflow tract anomalies: so-called conotruncal anomalies Shi-Joon Yoo, Fraser Golding, and Edgar Jaeggi

305

22.

Aortic arch anomalies Shi-Joon Yoo, Timothy Bradley, and Edgar Jaeggi

329

23.

Developments in diagnosis of transposition of the great arteries Laurent Fermont

343

24.

Abnormal visceral and atrial situs and congenital heart disease Shi-Joon Yoo, Mark K Friedberg, and Edgar Jaeggi

347

25.

Diseases of the myocardium, endocardium, and pericardium during fetal life Paulo Zielinsky, Antonio Piccoli Jr, João Luiz Manica, and Luiz Henrique Nicoloso

363

26.

Cardiomyopathy in the fetus John M Simpson

375

27.

Ultrasound examination of the fetal coronary circulation Ahmet A Baschat and Ulrich Gembruch

385

28.

Fetal cardiac tumors Mary T Donofrio and Gerard R Martin

401

29.

The fetal venous system: normal embryology, anatomy, and physiology and the development and appearance of anomalies Simcha Yagel, Zvi Kivilevitch, Dan V Valsky, and Reuwen Achiron

413

30.

Extracardiac Doppler investigation in fetuses with congenital heart disease Annegret Geipel, Ulrich Gembruch, and Christoph Berg

427

31.

Electrophysiology for the perinatologist Edgar Jaeggi

435

32.

Fetal bradydysrhythmia Klaus G Schmidt

449

33.

Fetal tachyarrhythmia Ulrich Gembruch

461

34.

Cardiac diseases in association with hydrops fetalis Ulrich Gembruch and Wolfgang Holzgreve

483

35.

Mending the tiniest hearts: an overview Thomas Kohl

515

36.

Fetal cardiac function in normal and growth-restricted fetuses Giuseppe Rizzo, Alessandra Capponi, and Domenico Arduini

531

37.

Venous flow in intrauterine growth restriction and cardiac decompensation Torvid Kiserud

547

38.

Congestive heart failure in the fetus James C Huhta

561

Contents

ix

39.

Congenital cardiovascular malformations and the fetal and neonatal circulation Abraham M Rudolph

579

40.

Twin–twin transfusion syndrome: impact on the cardiovascular system Jack Rychik

597

41.

Genetics and cardiac anomalies Eran Bornstein, David Seubert, and Mark I Evans

609

42.

Cardiac defects in chromosomally abnormal fetuses Jon Hyett and Alex Gooi

621

43.

Associated anomalies in congenital heart disease Christoph Berg, Ulrich Gembruch, and Annegret Geipel

635

44.

The neonate with congenital heart disease – medical and interventional management Ulrike Herberg

659

45.

Infants with congenital heart disease in the first year of life Andrew J Parry and Frank L Hanley

691

46.

Genetic counseling in families with congenital heart defects Klaus Zerres and Sabine Rudnik-Schöneborn

705

47.

Intrapartum evaluation of fetal well-being Yoram Sorokin and Sean C Blackwell

713

48.

Cardiac disease in pregnancy David Planer, Haim D Danenberg, and Chaim Lotan

725

49.

Maternal diseases and therapies affecting the fetal cardiovascular system Salim Kees and Eyal Schiff

737

50.

Congenital heart disease and the central nervous system: a perinatal perspective Amanda Shillingford and Jack Rychik

749

Index

759

List of Video clips

Chapter 7 7.1

Dynamic colorization

87

7.2

High-resolution real-time three-dimensional view of a beating normal fetal heart

87

7.3

Multiplanar reconstruction of a virtual cardiac cycle

88

7.4

Topographic imaging of a virtual cardiac cycle

89

7.5

Topographic imaging of a normal heart and a heart with d-transposition of the great arteries

89

7.6

Color tissue Doppler of a normal fetal heart at mid-trimester

93

7.7

Pulsed-wave tissue Doppler study of a normal fetal heart

93

7.8

Longitudinal strain in the fetal heart measured using speckle tracking

95

7.9

Circumferential strain in the fetal heart measured using speckle tracking

96

Chapter 12 12.1

Two-dimensional echocardiography: a parallel sweep from the upper abdomen to the large thorax

173

12.2

Apical four-chamber plane

177

12.3

Sagittal view of aortic arch

180

Chapter 13 13.1

MPR screen of a volume data set acquired with STIC

190

Chapter 14 14.1

Two-dimensional ultrasound scan of the four-chamber view

198

14.2

Standard planes of section

199

14.3

Tomographic ultrasound imaging of a normal volume dataset

201

14.4

Demonstration of technique to systematically visualize outflow tracts

203

14.5

Demonstration of transposition of great arteries

203

14.6

Demonstration of technique to visualize the aortic and ductal arches

203

14.7

‘Thick slice’ rendering of the atrioventricular valves of a normal fetus

206

14.8

‘Thick slice’ rendering of the atrioventricular valves in a fetus with Ebstein anomaly

208

14.9

‘Thick slice’ rendering in a case of absent pulmonary valve syndrome

209

14.10

‘Thick slice’ rendering of the right ventricle and pulmonary artery in a case of absent pulmonary valve syndrome

209

14.11

Technique to obtain rendered images of the outflow tracts

210

14.12

Crisscrossing of the outflow tracts

210

Video clips

xi

14.13

Technique to render the aortic and ductal arches

210

14.14

Four-dimensional visualization of the aortic and ductal arches

211

Chapter 15 15.1

B-flow of normal heart and aortic arch

220

15.2

Normal heart and great vessels

221

15.3

STIC acquisition combined with IM and VOCAL analysis for fetal cardiac ventrical volumetry

227

15.4

STIC acquisition in a case of TAPVC

229

15.5

B-flow of the heart and great vessels in a fetus with interrupted inferior vena cava with azygos continuation

229

15.6

IVS ‘virtual plane’ with color Doppler in evaluation of VSD

230

15.7

B-flow modality showing the parallel great vessels in a case of transposition

231

Chapter 19 19.1

Four-chamber view of an atrioventricular septal defect

285

19.2

Atrioventricular septal defect without attachment to interventricular septum

286

19.3

Subarterial doubly committed ventricular septal defect

288

19.4

Three-dimensional reconstruction of a perimembranous ventricular septal defect

288

Chapter 23 23.1

Inlet ventricular septal defect

343

Chapter 28 28.1

Two-dimensional echocardiographic imaging of a cardiac tumor

401

28.2

Three-dimensional echocardiographic imaging of a cardiac tumor

401

28.3

Two-dimensional echocardiographic imaging of a cardiac tumor

402

28.4

Color Doppler imaging of a cardiac tumor

402

28.5

Postnatal two-dimensional echocardiographic imaging of a cardiac tumor

402

28.6

Postnatal color Doppler imaging of a cardiac tumor

402

28.7

Postnatal three-dimensional ultrasound imaging of a cardiac tumor

404

28.8

Two-dimensional echocardiographic imaging of an atrial tumor

405

28.9

Postnatal two-dimensional echocardiographic imaging of an atrial tumor

405

Chapter 29 29.1

Fetal intra-abdominal umbilical vein varix

417

29.2

B-flow image of fetal intra-abdominal umbilical vein varix

417

29.3

Interrupted inferior vena cava with azygos continuation

418

29.4

B-flow image of interrupted inferior vena cava with azygos continuation

418

29.5

Three-dimensional ultrasound inversion mode of persistent right umbilical vein anomaly

419

29.6

Four-dimensional ultrasound B-flow image of agenesis of ductus venosus

420

29.7

Total anomalous pulmonary venous connection anomaly shown with high definition power Doppler and STIC acquisition

422

Contributors

Reuwen Achiron Department of Obstetrics and Gynecology Chaim Sheba Medical Center Tel Hashomer and Sackler School of Medicine Tel Aviv University Tel Aviv, Israel Lindsay D Allan Harris Birthright Centre for Fetal Medicine King’s College Hospital London, United Kingdom Domenico Arduini Department of Obstetrics and Gynecology Università Roma Tor Vergata Ospedale Fatebenefratelli Rome, Italy Ahmet A Baschat Department of Obstetrics, Gynecology, and Reproductive Sciences University of Maryland Baltimore, Maryland, USA Christoph Berg Department of Obstetrics and Perinatal Medicine University of Bonn Bonn, Germany Einat Birk Perioperative Pediatric Cardiology Heart Institute Schneider Children’s Medical Center Petach Tikva, Israel Sean C Blackwell Division of Maternal – Fetal Medicine Department of Obstetrics and Gynecology University of Texas Health Sciences Houston, Texas, USA

Eran Bornstein Department of Obstetrics and Gynecology Lenox Hill Hospital New York, New York, USA Timothy J Bradley University of Toronto Faculty of Medicine and Division of Cardiology Department of Paediatrics The Hospital for Sick Children Toronto, Ontario, Canada Graham J Burton Academic Department of Obstetrics and Gynaecology, Royal Free and University College London Medical School London and Department of Physiology, Development, and Neuroscience University of Cambridge Cambridge, United Kingdom Alessandra Capponi Department of Obstetrics and Gynecology Ospedale GB Grassi Rome, Italy Julene S Carvalho Brompton Fetal Cardiology Royal Brompton Hospital and Fetal Medicine Unit, St George’s Hospital London, United Kingdom Rabih Chaoui Center for Prenatal Diagnosis and Human Genetics Berlin, Germany Sarah Margalyt Cohen Department of Obstetrics and Gynecology Hadassah – Hebrew University Medical Centers Mount Scopus Jerusalem, Israel Haim D Danenberg Heart Institute Hadassah – Hebrew University Medical Center Jerusalem, Israel

Contributors

Mary T Donofrio Children’s National Heart Institute Children’s National Medical Center Washington, DC, USA Jimmy Espinoza Wayne State University Department of Obstetrics and Gynecology Detroit, Michigan, USA Mark I Evans Institute for Genetics and Fetal Medicine St Luke’s Roosevelt Hospital Center and Department of Obstetrics and Gynecology and Genetics Columbia University New York, New York, USA Laurent Fermont Institut de Puériculture Paris, France Mark K Friedberg The Hospital for Sick Children Toronto, Ontario, Canada Jean-Claude Fouron Unité de Cardiologie Foetale, CHU Sainte Justine Université de Montréal Montréal, Québec, Canada Zvi Friedman GE Healthcare Inc. Haifa, Israel Annegret Geipel Department of Obstetrics and Prenatal Medicine University of Bonn Bonn, Germany Ulrich Gembruch Department of Obstetrics and Prenatal Medicine Center of Obstetrics and Gynecology University of Bonn Bonn, Germany Adriana C Gittenberger–de Groot Department of Anatomy and Embryology Leiden University Medical Center Leiden, The Netherlands Fraser Golding Echocardiography Laboratory Division of Cardiology Department of Paediatrics The Hospital for Sick Children Toronto, Ontario, Canada

Debra S Goldman-Wohl Center for Human Placenta Research Department of Obstetrics and Gynecology Hadassah – Hebrew University Medical Centers Mount Scopus Jerusalem, Israel Luís F Gonçalves Perinatology Research Branch NICHD/NIH/DHHS Detroit, Michigan, USA Alex Gooi Prince Charles Hospital Brisbane, Queensland, Australia Frank L Hanley Division of Pediatric Cardiac Surgery University of California San Francisco, California, USA Ulrike Herberg Department of Pediatric Cardiology University of Bonn Bonn, Germany Julien IE Hoffman Cardiovascular Research Institute University of California San Francisco, California, USA Wolfgang Holzgreve Department of Obstetrics and Gynecology University of Basel Kantonspital Basel, Switzerland James C Huhta Perinatal Cardiology University of South Florida College of Medicine St Petersburg, Florida, USA John Hyett Royal Brisbane and Women’s Hospital Brisbane, Queensland, Australia Edgar T Jaeggi University of Toronto Faculty of Medicine and Fetal Cardiac Program Division of Cardiology Department of Paediatrics The Hospital for Sick Children Toronto, Ontario, Canada

xiii

xiv

Contributors

Eric Jauniaux Academic Department of Obstetrics and Gynaecology UCL EGA Institute for Women’s Health Royal Free and University College London (UCL Campus) London, United Kingdom Salim Kees Department of Obstetrics and Gynecology Chaim Sheba Medical Center Ramat Gan, Israel Torvid Kiserud Haukeland University Hospital Department of Clinical Medicine University of Bergen Bergen, Norway Zvi Kivilevitch Maccabi Health Services Women’s Health Center Ultrasound Center Beer Sheba, Israel Thomas Kohl Department of Obstetrics and Prenatal Medicine Bonn University Hospital Bonn, Germany Juan Pedro Kusanovic Perinatology Research Branch NICHD/NIH/DHHS Detroit, Michigan, USA

Luiz Henrique Nicoloso Fetal Cardiology Unit Institute of Cardiology of Rio Grande do Sul Porto Alegre, Brazil Andrew J Parry Department of Paediatric Surgery Bristol Royal Hospital for Children Bristol, United Kingdom Antonio Piccoli Jr Fetal Cardiology Unit Institute of Cardiology of Rio Grande do Sul Porto Alegre, Brazil David Planer Heart Institute Hadassah – Hebrew University Medical Center Jerusalem, Israel Robert E Poelmann Department of Anatomy and Embryology Leiden University Medical Center Leiden, The Netherlands Giuseppe Rizzo Department of Obstetrics and Gynecology Università Roma Tor Vergata Ospedale Fatebenefratelli Rome, Italy

Wesley Lee William Beaumont Hospital Royal Oak, Michigan, USA

Roberto Romero Perinatology Research Branch NICHD/NIH/DHHS Detroit, Michigan, and Bethesda, Maryland, USA

Chaim Lotan Heart Institute Hadassah – Hebrew University Medical Center Jerusalem, Israel

Sabine Rudnik-Schöneborn Institute of Human Genetics University of Technology Aachen, Germany

João Luiz Manica Fetal Cardiology Unit Institute of Cardiology of Rio Grande do Sul Porto Alegre, Brazil

Abraham M Rudolph Cardiovascular Research Institute University of California San Francisco, California, USA

Gerard R Martin Children’s National Medical Center Division of Cardiology Washington, DC, USA Baruch Messing Department of Obstetrics and Gynecology Hadassah – Hebrew University Medical Centers Mount Scopus Jerusalem, Israel

Jack Rychik Fetal Heart Program Cardiac Center Children’s Hospital of Philadelphia Philadelphia, Pennsylvania, USA David J Sahn Oregon Health and Sciences University Portland, Oregon, USA

Contributors

Eyal Schiff Department of Obstetrics and Gynecology Sheba Hospital Ramat Gan, Israel Klaus G Schmidt Division of Pediatric Cardiology and Pneumology Children’s Hospital University Medical Center Düsseldorf Düsseldorf, Germany David Seubert Department of Perinatology Bay State Medical Center Springfield, Massachusetts, USA Isabel Shapiro Bnai Zion Medical Center Haifa, Israel Amanda Shillingford Fetal Heart Program Cardiac Center Children’s Hospital of Philadelphia Philadelphia, Pennsylvania, USA Norman H Silverman Pediatric and Echocardiography Laboratory Division of Pediatric Cardiology Lucile Packard Children’s Hospital Stanford University Medical Center Palo Alto, California, USA John M Simpson Fetal and Paediatric Cardiology Department of Congenital Heart Disease Guy’s and St Thomas’ NHS Trust London, United Kingdom

Boris Tutschek Department of Obstetrics and Gynecology University of Düsseldorf Düsseldorf, Germany Dan V Valsky Department of Obstetrics and Gynecology Hadassah – Hebrew University Medical Centers Mount Scopus Jerusalem, Israel Arnold CG Wenink Department of Anatomy and Embryology Leiden University Medical Center Leiden, The Netherlands Simcha Yagel Obstetrics and Gynecology Ultrasound Center Center for Human Placenta Research Department of Obstetrics and Gynecology Hadassah – Hebrew University Medical Centers Mount Scopus and Department of Obstetrics and Gynecology Hebrew University – Hadassah Faculty of Medicine Jerusalem, Israel Shi-Joon Yoo Department of Medical Imaging Faculty of Medicine University of Toronto and Division of Cardiac Imaging Department of Diagnostic Imaging The Hospital for Sick Children Toronto, Ontario, Canada

Yoram Sorokin Department of Obstetrics and Gynecology Maternal Fetal Medicine Fellowship Program Wayne State University School of Medicine Detroit, Michigan, USA

Klaus Zerres Institute of Human Genetics University of Technology Aachen, Germany

Cornelia Tennstedt Department of Pathology Charité Medical Faculty of the Humboldt University Berlin, Germany

Paulo Zielinsky Fetal Cardiology Unit Institute of Cardiology of Rio Grande do Sul Porto Alegre, Brazil

xv

Preface

This second edition of Fetal Cardiology: Embryology, Genetics, Physiology, Echocardiographic Evaluation, Diagnosis and Perinatal Management of Cardiac Diseases is, like the first, the product of the combined efforts of many professionals: obstetricians, pediatric cardiologists, sonographers, molecular biologists, and medical physicists. The field of fetal ultrasonographic and echocardiographic imaging and fetal cardiac therapies and interventions is rapidly changing and developing, and is therefore in constant need of updating. Over the very few years between the first and second editions, considerable strides have been made in all aspects of fetal cardiology. Chapters have been added on many aspects of fetal cardiology, including the cardiac impact of the twinto-twin transfusion syndrome, cerebral outcomes in congenital heart disease (CHD), and congestive heart failure, and virtually all have been updated. In two areas advances have been particularly noteworthy: the identification of additional genes implicated in CHD, and the advances in three- and four-dimensional (3D/4D) ultrasound (US) capabilities in fetal cardiac imaging. The 3D/4D revolution has opened exciting avenues for clinical applications and research endeavors in fetal cardiology, and is perhaps the main impetus for the second edition. Motion gating technology has overcome much inherent difficulty in 3D fetal cardiac scanning, and now allows near real-time 3D/4D heart examination. The burgeoning 3D/4D US–based literature shows that these modalities contribute to our understanding of the normal and anomalous fetal heart, to interdisciplinary management team consultation, to parental counseling, and to professional training. 3D/4D ultrasound applications have been shown to facilitate screening programs, and through offline networking capabilities to improve healthcare delivery systems, to extend the benefits of advanced prenatal cardiac screening to patients far from

well-equipped centers. 3D/4D US has added virtual planes to fetal cardiac scanning: views of the fetal heart not generally accessible with a standard 2D approach. The broad classification congenital heart disease includes the most common congenital malformations, commonly estimated to be 8:1000 live births. As many more are thought to die in utero or abort spontaneously. This is an important aspect of fully integrated comprehensive fetal echocardiography in every targeted organ scan. Many books have been devoted to detection of fetal anomalies, and many to Doppler techniques. Most of these volumes devote only one chapter to fetal cardiac malformations. This is a comprehensive guide intended for anyone interested in fetal organ scanning. Anyone performing targeted organ scans or having an interest in the fetal heart, we believe, will find it useful. It is our hope that this volume will bridge among the specialties: obstetricians, pediatric cardiologists, and general cardiologists. To this end we have included chapters on treatment options and pharmacological or surgical interventions available to affected fetuses, as well as all life stages of heart disease, from embryology and the genes involved in cardiac development, to the reproductive health of women with CHD, and counseling of families affected with CHD. We also discuss the often-ignored placenta, an indivisible part of the cardiovascular system, which receives 50% of fetal blood volume. The subject of maternal status and how the mother’s treatments will affect the fetal heart is included, as well as future fetal treatments only now being developed.

Simcha Yagel Norman H Silverman Ulrich Gembruch November 2008

Fetal Cardiology

1 Cardiovascular development Arnold CG Wenink Introduction The importance of the knowledge of cardiovascular embryology has changed over the years. The field of embryology has been useful in the understanding of details of the anatomy of congenital malformations, although erroneous developmental explanations have hindered the understanding of congenital lesions.1 Embryological–pathological correlations can be made only when both the development and the pathological anatomy have been adequately described. Currently, the fields of fetal cardiology, genetics, and molecular biology are developing rapidly. Much of this knowledge has been acquired from in vitro experiments using cell or tissue cultures, but in vivo studies of genetically manipulated mice have become gradually more important. Integration of the information should lead to full understanding of the development within the embryo. In the present chapter, histological and gross anatomical features of the embryonic heart are discussed in an attempt to link them with prenatal cardiac function. The description of the mature heart in normal individuals and in the case of congenital malformations has greatly benefited from the sequential segmental approach.2 This approach is also useful for the description of consecutive developmental stages, because it highlights the course of the bloodstream and, therefore, has functional implications. After a sequential description of the development, some aspects will be separately highlighted.

Segments and intersegmental transitional zones in the embryonic heart Prior to septation, which leads to the four-chambered organ, the heart is a simple tube in which the blood is pumped through a number of serially connected segments. Although the tube starts off as a straight one, arranged in a caudocranial direction in the embryonic body, the

individual segments are best described when the looping process of the tube has taken place and the heart has adopted an S-shape. Extrapolation of the segmental organization to the straight heart tube stage is disputable, because at that stage not all components have yet been recognized using labeling experiments.3 Distinct atrial and ventricular cell lineages have, however, been described before looping.4 The most reliable criteria to distinguish the segments in the looped heart are unfortunately still based on general morphology, because specific markers for each segment, which would remain recognizable throughout development, have not yet become available. Several molecular markers are known to distinguish cardiac segments, but their specificity is restricted to limited periods of embryonic development.5 In the looped heart, cardiac segments can be distinguished by the presence of a series of constrictions in the myocardial tube and by the internal profile of the myocardium, which may be smooth or trabeculated, and in slightly advanced stages by the accumulation of endocardial cushion masses on the inside and subepicardial tissue externally. Following the bloodstream from the venous to the arterial poles, the segments are the venous sinus, the atrium, the ventricular inlet segment, and the ventricular outlet segment. The venous sinus collects all the embryonic veins, and the ventricular outlet segment is connected to the arterial trunk (aortic sac). Between these segments the intersegmental transitional zones are the sinoatrial, atrioventricular, and primary interventricular transitions (Figure 1.1).

The venous sinus The venous sinus is the most caudal segment, which is anchored within the transverse septum. Left and right sinus horns contribute to the segment and these collect all the veins of the left and right sides of the embryo. There has been considerable discussion about the relationship of the pulmonary veins to the venous sinus. According to classical descriptions, the primitive common pulmonary vein develops by sprouting from the dorsal wall of the left

2

Fetal Cardiology

la

ra

(a)

r

l

p

av s i

(b)

av

Figure 1.1 pf

in

(c)

pout (d) ap

dout

(e)

atrium.6 More recent accounts, using HNK-1 (human natural killer antigen-1) expression as a marker for the sinoatrial boundary in the chicken and the rat, describe the pulmonary vein connecting originally to the venous sinus.7–9 Observations in the mouse embryo still claim the primary connection to be established beyond the venous sinus.10 This controversy is a clear example of the need for distinct cardiac segmental markers in addition to more classical methods of distinguishing between the segments, as mentioned above. During development the venous sinus is incorporated into the atrium. Most of it contributes to the posterior wall of the mature right atrium, but a small portion forms the posterior left atrial wall and will contain the pulmonary venous orifices. A problem with nomenclature is apparent in this description; what is known is that the atrium in the mature heart is built from two distinct embryonic cardiac segments called the venous sinus and the atrium.

Schematic dissection of the embryonic heart after looping. (a) The venous sinus, which receives the systemic veins in its right (r) and left (l) sinus horns and the common pulmonary vein (p) separately, opens into the common atrium with its left (la) and right (ra) atrial appendages. (b) From the atrium, the blood continues into the atrioventricular canal (av) with superior (s) and inferior (i) endocardial cushions. (c) The atrioventricular canal (av) completely drains into the ventricular inlet segment (in). On its right side, the inlet segment shows an opening, the primary foramen, which is encircled by the primary fold (pf). Note the close proximity of the primary fold and the atrioventricular canal, which can be seen through the primary foramen (arrow). (d) The proximal outlet segment (pout) receives the blood from the primary foramen (arrow). (e) The most downstream portion of the ventricular mass is the distal outlet segment (dout) which has started to be septated by two endocardial outlet ridges (arrows). Within their core, these ridges contain the extremities of the aortopulmonary septum (ap) which are attached to both sides of the myocardial wall of the distal outlet segment.

In the mature heart, those parts that are derived from the embryonic venous sinus are easily distinguished by their smooth internal wall, in contradistinction to the wall of the embryonic atrial compartment, which carries pectinate muscles. In particular on the right side, the two embryonic compartments can be distinguished by the presence of structures derived from the sinoatrial transitional zone (Figure 1.2).

The sinoatrial transitional zone With the incorporation of the venous sinus into the atrium, the sinoatrial transition gradually becomes more apparent. Although its left side, which delimits the pulmonary venous orifice, is never very prominent,7 invagination into the right part of the atrium leads to the formation of large valves, the venous valves. The left venous valve is no longer discernible

Cardiovascular development

MAW

3

by the left and right atrial appendages. These remain recognizable by the pectinate muscles. The relatively early development of these pectinate muscles in the embryonic atrial compartment may imply an early contractile function of the atria, just as has been described for the trabeculations in the embryonic ventricles.

SCV m

rp

ms tc l

ss

OF cs ICV

tv ev

Figure 1.2 Diagram of the opened right atrium of the mature heart to show embryonic derivation of the septal structures. The right side of the sinoatrial transition is almost circular. Starting at the membranous septum (ms), it is represented by a muscular band (m) between the atrial appendage (MAW, medial atrial wall) and the entrance of the superior caval vein (SCV), from where it continues first into the terminal crest (tc) and then into the valve (ev) of the inferior caval vein (ICV) and the valve (tv) of the coronary sinus (cs). Only the valve of the oval foramen (OF) and the limbus (l) of the oval fossa are of embryonic atrial origin (reproduced with permission from reference 8).

in the mature heart because it is incorporated into the atrial septal structures. The right venous valve, however, remains present as the valve of the inferior vena cava (Eustachian valve) and the valve of the coronary sinal orifice (Thebesian valve) (Figure 1.2). The Eustachian valve continues into the terminal crest, which courses along the posterior atrial wall to surround the entrance of the superior vena cava. Much of this intersegmental transitional myocardium is involved in the development of the conducting tissues. The described position at the entrance of the superior vena cava coincides with the sinus node. The terminal crest has been implicated in descriptions of ‘internodal tracts’. The basal portion of this transition may be responsible for the formation of atrial inputs to the atrioventricular node. Distinction of the venous sinus and the sinoatrial transition from the embryonic atrial myocardium is possible because of the HNK-1 expression, which is always absent from the atrium.8

The atrium The atrium starts off as a relatively large unseptated cavity. Most of its remnants in the mature heart are represented

The atrioventricular canal The atrioventricular canal is the transitional zone that connects the atrium with the ventricular portion of the heart loop. In preseptation stages this canal forms an important proportion of the total myocardium, but its growth is relatively slow, and in advanced fetal stages the canal has become no more than the boundary between atrial and ventricular structures.11 Internally, the atrioventricular canal is invested with relatively large endocardial cushions which derive their cells from the covering endocardium by a process of endocardial–mesenchymal transformation.12 Also, neural crest cells have been demonstrated to migrate into the cushions.13 Because they are compressed by atrioventricular contraction, the cushion masses prevent regurgitation, and they are indeed the precursors of the atrioventricular valves.14 Externally, the atrioventricular canal is represented by a groove, which becomes filled with subepicardial mesenchyme. The originally single atrioventricular canal is septated into two separate atrioventricular valve orifices. Fusion of the endocardial cushions in the center of the canal forms the basis of this septation process. Because the superior cushion is immediately related to the base of the atrial septum and the inferior cushion is attached to the developing ventricular inlet septum, the same process also brings atrial and ventricular septal structures into continuity. It is not probable that this septation process forms the basis for the creation of separate left and right bloodstreams. Such bloodstreams already exist in early preseptation stages,15 and they are necessary for the creation of the inlet portion of the mature right ventricle and the splitting off of the ventricular inlet septum from the primary fold (see below). The general process of atrioventricular valve formation14 (Figure 1.3) does not explain the detailed final morphology of these valves. It does explain the general architecture, with annular attachment, a fibrous veil, and chords attached to papillary muscles, but the morphological differences between mitral and tricuspid valves are related to the morphology of the ventricles in which they develop. Along the same lines, many congenital malformations of the atrioventricular valves are not the result of defective valve formation but of preexisting abnormalities of the underlying ventricular myocardium. Straddling valves and the common valve in atrioventricular septal defects belong to this category. On the other hand, congenital valve pathology based on deviation of the general

4

Fetal Cardiology

a

s vl

c

c m

tc

v

(a)

p

(b)

(c)

(d)

(e)

Figure 1.3 Diagrams to show the mechanism of atrioventricular valve formation. (a) In the early stages, the myocardium of the atrium (a) and ventricle (v) is continuous at the atrioventricular canal. Externally, the atrioventricular groove is filled with subepicardial sulcus tissue (s). Internally, the atrioventricular myocardium is covered by endocardial cushion tissue (c). (b) The continuity of atrial and ventricular myocardium is disrupted. (c) The inner layer of ventricular myocardium is delaminated, which process creates a primitive flap valve consisting of cushion tissue (c) on its atrial side and of myocardium (m) on its ventricular side. (d) The myocardial layer on the ventricular side of the valve gradually disappears and retracts towards the atrioventricular canal and towards the ventricular apex. (e) The myocardial layer of the valve has completely disappeared. The cushion tissue has thinned out to form the valve leaflet (vl) and the tendinous chords (tc). The remaining delaminated myocardium constitutes the papillary muscle (p).

developmental mechanism is extremely rare. Ebstein’s anomaly is an example of this category. The myocardium of the atrioventricular canal is involved in the formation of the atrioventricular node and the proximal His bundle. These structures have been recognized by their HNK-1 content, as in the sinoatrial transitional myocardium.8 Any relationship between this immunohistochemical marker and the specific function of the conducting tissues has, however, not yet been elucidated.

The ventricular inlet segment Before septation, all of the atrial blood is propelled into a single cavity, the ventricular inlet segment. The embryonic heart may be described to have a univentricular atrioventricular connection. Roughly, the inlet segment may be compared with the mature left ventricle, but the term ‘primitive left ventricle’ is not used here, because not all of the tissues of the mature left ventricle are derived from this segment. The future ascending aorta and pulmonary trunk are connected to the more distal outlet segment of the heart. This means that the process that brings the aorta above the mature left ventricle also makes the outlet segment contribute to the left ventricle. The transfer, therefore, concerns the subaortic outflow tract in addition to the aortic orifice. The embryonic inlet segment has a sponge-like appearance because of its many trabeculations. These trabeculations are thought to be the main structures that cause ventricular systole.16 Their degree of myofibril assembly is considerably advanced when compared to that of the compact free wall.17 In early stages, the myocardial volume of

the ventricular inlet segment is considerably larger than that of the outlet segment, making the inlet segment largely responsible for ventricular force.11

The primary fold The subdivision of the ventricular mass in the looped heart is made by an external groove, which is called the primary groove. It corresponds to an internal myocardial profile, the primary fold, which delimits the communication between inlet and outlet segments. The communication has been described as an ‘interventricular foramen’.18,19 This term might cause confusion, because any interventricular communication would logically be closed by the ventricular septation process, which is not the case for this foramen. In the present account, the communication between the two segments within the ventricular mass is called the primary foramen. To guarantee continuity of the bloodstream from the inlet segment into the outlet segment and the great arteries, the primary foramen will never close. Most of the primary fold will form the muscular ventricular septum. In the inner curvature of the heart loop, the fold establishes the boundary between the arterial and atrioventricular regions of the ventricular mass (also known as the ventriculoinfundibular fold). A special process is involved in the formation of the ventricular inlet septum, i.e. that part of the ventricular septum which separates the bloodstreams coming from the two mature atrioventricular orifices.20 Initially, no other cavity than the outlet segment is present to the right of the primary fold. Then, the right atrioventricular

Cardiovascular development

AV

IN

PF OUT

IS

OUT

Figure 1.4 Diagram to show development of the inlet portion of the right ventricle. Initially, the atrioventricular canal (AV) drains completely into the ventricular inlet segment (IN). The blood has to pass the primary fold (PF) to reach the ventricular outlet segment (OUT). The blood coming from the right part of the atrioventricular canal excavates a new cavity within the tissue of this fold which splits the fold into an inlet septum (IS) and the septomarginal trabeculation. The latter is pushed forward and has always to be passed by the bloodstream to reach the outlet segment (second arrow in bottom figure) (reproduced with permission from reference 20).

bloodstream seems to create an excavation in the posterior part of the fold and to split off a separate inlet septum to its left side. This excavation will enlarge, and is present in the mature heart as the inlet portion of the right ventricle. The enlargement of this inlet portion increasingly separates the right part of the primary fold from the inlet septum. This part of the primary fold will continue to indicate the boundary between inlet and outlet segments, and the displaced portion is found in the mature heart as the septomarginal trabeculation between inlet and outlet parts of the right ventricle (Figure 1.4).

The ventricular outlet segment The right side of the ventricular part of the heart loop is initially constituted by only the ventricular outlet segment. The term ‘primitive right ventricle’ should be avoided, because rather complex processes still have to lead to formation of the mature right ventricle. These processes involve the excavation of a right-sided inlet cavity and the

5

splitting of the primary fold into an inlet septum and a septomarginal trabeculation (see above). The ventricular outlet segment is composed of two subdivisions: the proximal outlet segment and the distal outlet segment. The proximal outlet segment looks very much like the ventricular inlet segment, having initially a thin compact myocardial wall and showing many trabeculations. Before septation, this segment is considerably smaller than the inlet segment.11 Hemodynamically, in these early stages the heart might be compared to the congenital lesion called univentricular atrioventricular connection to the left ventricle, in which a comparatively small right ventricle is usually present. It may be that the ventricular inlet segment in these stages propels the blood directly into the next portion of the heart, the distal outlet segment. The distal outlet segment has a smooth myocardial wall and is invested internally with thick endocardial ridges, which contribute to septal structures and to the arterial valves. The cells within these ridges have originated from the covering endocardium by endocardial–mesenchymal transformation.12 Another source of cells in these cushions is the neural crest.13 Before the presence of the arterial valves, the myocardium of the distal outlet segment compresses the endocardial ridges to prevent regurgitation. This myocardium is precociously differentiated and has particular functional properties.17,21 As the atrioventricular canal, the distal outlet segment shows only temporary growth, and its proportional volume and length diminish gradually.11,22

The tissues of the embryonic heart The heart tube, which develops as a specialized portion of the vascular system, consists of three layers of different cellular composition. The main constituent is the myocardium, which is lined internally by the endocardium and externally by the epicardium.

The myocardium The myocardial constituent of the embryonic heart is derived from the splanchnic mesoderm, which forms a single horseshoe-shaped structure in front of the buccopharyngeal membrane.23 The limbs of this horseshoe contribute to the venous pole of the heart, its central portion giving rise to the ventricular structures.

The endocardium The splanchnic mesoderm also gives rise to a single endothelial plexus, which forms the internal endocardial

6

Fetal Cardiology

tube.23,24 Initially, the space between the endocardium and myocardium is occupied by the cardiac jelly.25 This jelly becomes invaded by mesenchymal cells, which are derived from the endocardium.12 The invasion leads to the formation of local masses of endocardial cushion tissue which are restricted to the atrioventricular canal and the distal outlet segment. In addition to the endocardial–mesenchymal transformation, other sources contribute to the cellular composition of the cushions. It should be realized that the endocardial tube forms part of the continuous system of endothelial structures forming the veins as well as the arteries. At the same time, the myocardial wall of the heart develops as part of the coelomic wall, which shows reflections into the pericardium at the venous and arterial poles. Therefore, the space between the myocardial and endocardial layers can be invaded by cells from extracardiac sources. A main source of extracardiac contribution is the embryonic neural crest.26 Indeed, neural crest cells have been demonstrated to invade the endocardial cushions in the distal outlet segment and to contribute to outflow tract septation.27 Neural crest cells also invade the inflow tract.13 A second extracardiac source of cushion cells is the epicardium, which has been demonstrated to send its derivatives into the atrioventricular cushions.28 They also form the substrate for the smooth muscle cells and fibroblasts of the coronary vasculature.29

The epicardium As described above, the coelomic epithelium forms the myocardium and the pericardium, and these tissues are continuous at the venous and arterial poles. The epicardium is of secondary origin. At the site of the transverse septum the epicardial organ starts to protrude into the pericardial cavity and sends out vesicles. These gradually cover the entire myocardial mantle, where they give rise to the epicardial lining and the subepicardial mesenchyme, including the early subepicardial vasculature.30–33 The epicardial vesicles have been described to carry mesenchymal cells from within the transverse septum.34

Septation and valve formation Complex processes transform the looped heart tube into the four-chambered organ with valves at the various transitions. All segments and intersegmental transitional zones are septated into left and right portions, although the processes and the tissues involved are variable. Evidently, all septal structures have to fuse to guarantee continuous left and right bloodstreams. For this reason, it is impractical to describe the septation processes individually for each separate segment. In the text to follow, the septation

processes will be dealt with in an integrated fashion, only subheading the venous and arterial portions of the heart separately.

Septation and valve formation at the venous pole Because most of the venous sinus is incorporated into the atrium, it is vital to analyze the mature atrial septal structures to find out which of them belong to the atrium proper and which are of sinal or sinoatrial origin. It appears that only a very small portion of the mature atrial septum, contained within the rims of the oval fossa, is of embryonic atrial origin, all other structures being derived from the sinoatrial transitional zone.8 The invagination of the venous sinus into the embryonic atrium leads to the formation of several distinct profiles. The sinus part, receiving the systemic veins, invaginates deeply into the right portion of the atrium, which process creates prominent double structures called the venous valves. These flaps can be distinguished to consist of sinal and atrial layers. Towards the pulmonary portion of the sinus, the invaginated profiles gradually diminish and the pulmonary venous orifice is only temporarily bordered by distinct elevations. It is this less well developed part of the sinoatrial transitional zone that is cut off from its right-sided counterpart by the atrial septum primum. This muscular septum is completely contained within the embryonic atrium, but when it descends it compresses the sinoatrial junction just to the right of the pulmonary vein.7 The posterior portion of its lower rim fuses with the inferior atrioventricular endocardial cushion, which is also the site where the inferior parts of the venous valves become fixed. As a result, the systemic part of the venous sinus is kept to the right of the atrial septum primum. To the left of the left venous valve, between it and the atrial septum primum, the atrial wall still invaginates somewhat to form an internal profile which is called the atrial septum secundum. This structure never becomes a full septum, but forms a semicircular rim that eventually fuses with the left venous valve. At birth, it forms the material to which the flap of the atrial septum primum is pressed to close the interatrial communication.

Septation and valve formation at the arterial pole Septation of the outflow tract takes place in immediate continuity with septation of the aortic sac35 (Figure 1.1e). The extracardiac mesenchyme, which descends from the region between the fourth and sixth pharyngeal arch arteries, not only forms the aortopulmonary septum between the ascending aorta and pulmonary trunk, but

Cardiovascular development

also sends two extensions toward the myocardial wall of the distal outlet segment. These extensions are completely covered by the endocardial outlet ridges. During expansion of the outflow tract the parts of its myocardial wall that are attached to the aortopulmonary septum seem to be kept together, and the attached myocytes will later invade the outflow tract septum which initially consisted of mesenchymal tissues only. The downstream portions of the endocardial outlet ridges form the semilunar valves.36

References 1. Anderson RH, Wenink ACG. Thoughts on concepts of development of the heart in relation to the morphology of congenital malformations. Experientia 1988; 44: 951–60. 2. Becker AE, Anderson RH. Diagnosis of congenital heart disease: the sequential segmental approach. In: Becker AE, Anderson RH et al, eds. Cardiac Pathology. Edinburgh: Churchill Livingstone, 1983: 9/2–12. 3. De La Cruz MV, Sánchez-Gómez C, Palomino MA. The primitive cardiac regions in the straight tube heart (Stage 9) and their anatomical expression in the mature heart: an experimental study in the chick embryo. J Anat 1989; 165: 121–31. 4. Yutzey KE, Bader D. Diversification of cardiomyogenic cell lineages during early heart development. Circ Res 1995; 77: 216–19. 5. Franco D, Lamers WH, Moorman AFM. Patterns of expression in the developing myocardium: towards a morphologically integrated transcriptional model. Cardiovasc Res 1998; 38: 25–53. 6. Los JA. Embryology. In: Watson H, ed. Paediatric Cardiology. London: Lloyd-Luke, 1968: 1–28. 7. DeRuiter MC, Gittenberger-de Groot AC, Wenink ACG et al. In normal development pulmonary veins are connected to the sinus venosus segment in the left atrium. Anat Rec 1995; 243: 84–92. 8. Wenink ACG, Symersky P, Ikeda T et al. HNK-1 expression patterns in the embryonic rat heart distinguish between sinuatrial tissues and atrial myocardium. Anat Embryol 2000; 201: 39–50. 9. Blom NA, Gittenberger-de Groot AC, DeRuiter MC et al. Development of the cardiac conduction tissue in human embryos using HNK-1 antigen expression: possible relevance for understanding of abnormal atrial automaticity. Circulation 1999; 6: 800–6. 10. Webb S, Brown NA, Wessels A, Anderson RH. Development of the murine pulmonary vein and its relationship to the embryonic venous sinus. Anat Rec 1998; 250: 325–34. 11. Knaapen MWM, Vrolijk BCM, Wenink ACG. Growth of the myocardial volumes of the individual cardiac segments in the rat embryo. Anat Rec 1995; 243: 93–100. 12. Markwald RR, Mjaatvedt CH, Krug EL, Sinning AR. Inductive interactions in heart development: role of cardiac adherons in cushion tissue formation. Ann NY Acad Sci 1990; 588: 13–25.

7

13. Poelmann RE, Gittenberger-de Groot AC. A subpopulation of apoptosis-prone cardiac neural crest cells targets to the venous pole. Multiple functions in heart development? Dev Biol 1999; 207: 271–86. 14. Oosthoek PW, Wenink ACG, Vrolijk BCM et al. Development of the atrioventricular valve tension apparatus in the human heart. Anat Embryol 1998; 198: 317–29. 15. Hogers B, DeRuiter MC, Baasten AMJ et al. Intracardiac blood flow patterns related to the yolk sac circulation of the chick embryo. Circ Res 1995; 76: 871–7. 16. Challice CE, Virágh S. The architectural development of the early mammalian heart. Tissue Cell 1973; 6: 447–62. 17. Wenink ACG, Knaapen MWM, Vrolijk BCM, VanGroningen JP. Development of myocardial fiber organization in the rat heart. Anat Embryol 1996; 193: 559–67. 18. Van Mierop LHS, Alley RD, Kausel HW, Stranahan A. Pathogenesis of transposition complexes. I. Embryology of the ventricles and great arteries. Am J Cardiol 1963; 12: 216–25. 19. Goor DA, Edwards JE, Lillehei CW. The development of the interventricular septum of the human heart; correlative morphogenetic study. Chest 1970; 58: 453–67. 20. Wenink ACG, Wisse BJ, Groenendijk PM. Development of the inlet portion of the right ventricle in the embryonic rat heart: the basis for tricuspid valve development. Anat Rec 1994; 239: 216–23. 21. DeJong F, Opthof T, Wilde AAM et al. Persisting zones of slow impulse conduction in developing chicken hearts. Circ Res 1992; 71: 240–50. 22. Thurkow EW, Wenink ACG. Development of the ventriculoarterial segment of the embryonic heart: a morphometrics study. Anat Rec 1993; 236: 664–70. 23. DeRuiter MC, Poelmann RE, VanderPlas-de Vries I et al. The development of the myocardium and endocardium in mouse embryos. Anat Embryol 1992; 185: 461–73. 24. Steding G, Seidl W, Kluth D, Schulze M. Die Enstehung des Endocards. Untersuchungen an Hühnerembryonen. Verh Anat Ges 1980; 74: 365–7. 25. Davis CL. Development of the human heart from its first appearance to the stage found in embryos of twenty paired somites. Carnegie Inst Contr Embryol 1927; 19: 245–84. 26. Morris-Kay G, Ruberte E, Fukiishi Y. Mammalian neural crest and neural crest derivatives. Ann Anat 1993; 175: 501–7. 27. Poelmann RE, Mikawa T, Gittenberger-de Groot AC. Neural crest cells in outflow tract septation of the embryonic chicken heart: differentiation and apoptosis. Dev Dyn 1998; 212: 373–84. 28. Gittenberger-de Groot AC, Vrancken Peeters MPFM, Mentink MMT et al. Epicardium-derived cells contribute a novel population to the myocardial wall and the atrioventricular cushions. Circ Res 1998; 82: 1043–52. 29. Vrancken Peeters MPFM, Gittenberger-de Groot AC, Mentink MMT, Poelmann RE. Smooth muscle cells and fibroblast of the coronary arteries derive from epithelial– mesenchymal transformation of the epicardium. Anat Embryol 1999; 199: 367–78. 30. Virágh S, Challice CE. The origin of the epicardium and the embryonic myocardial circulation in the mouse. Anat Rec 1981; 201: 157–68. 31. Virágh S, Gittenberger-de Groot AC, Poelmann RE, Kálmán F. Early development of quail heart epicardium and

8

Fetal Cardiology

associated vascular and glandular structures. Anat Embryol 1993; 188: 381–93. 32. Vrancken Peeters MPFM, Mentink MMT, Poelmann RE, Gittenberger-de Groot AC. Cytokeratins as a marker for epicardial formation in the quail embryo. Anat Embryol 1995; 191: 503–8. 33. Muñoz-Chápuli R, Macías D, Ramos C et al. Development of the subepicardial mesenchyme and the early cardiac vessels in the dogfish (Scylorhinus canicula). J Exp Zool 1996; 275: 95–111.

34. VandenEijnde SM, Wenink ACG, Vermeij-Keers C. Origin of subepicardial cells in rat embryos. Anat Rec 1995; 242: 96–102. 35. Bartelings MM, Gittenberger-de Groot AC. The outflow tract of the heart. Embryologic and morphologic correlations. Int J Cardiol 1989; 22: 289–300. 36. Hurle JM, Colvee E. Changes in the endothelial morphology of the developing semilunar heart valves. Anat Embryol 1983; 167: 67–83.

2 Cardiac morphogenesis Adriana C Gittenberger–de Groot and Robert E Poelmann

Introduction Cardiovascular development and the regulatory mechanisms underlying this major embryonic event have become essential knowledge for the fetal cardiologist. The increase in ultrasound technology to detect morphology and pathomorphology of the growing heart requires more insight into the morphogenetic and epigenetic pathways underlying normal and abnormal development. The ever-increasing information from experimental (transgenic) animal models allows insight into the role of many genes and proteins. This research, however, is mostly not directed by the need to understand and solve the basis of cardiac malformations. It is more often the case that scientists searching for important genetic pathways in development unravel unexpectedly a role for a gene in question in cardiac development. It is essential to link genetic and epigenetic (environmental) clues from patient material to basic genetic knowledge to make progress in understanding the complicated interactive processes that direct heart development. The crucial processes in human cardiac development take place within the first 6 weeks of embr yogenesis, and as such cannot be followed using in vivo diagnostics. It is, therefore, still imminent that essential knowledge is incorporated from animal models such as mouse, chicken, and, more recently, zebrafish, as basic principles of heart formation can be compared between various animal models and human development. One has to take into account, however, important species differences such as, for instance, a right-sided aortic arch system in birds, as compared to a left-sided system in mammals, and the lack of cardiac septation processes in fish with only a two-chambered heart tube as a final result. Furthermore, the various converging fields of research have resulted in a confusing use of terminology, which is not easily solved1 and which will undoubtedly continue with current new discoveries. This chapter will describe in brief the major events in cardiac development.2 There will be a focus specifically on the continuous recruitment of myocardium from the second heart field3 and on

extracardiac cellular contributions to the heart and their modulatory role.4

Primary cardiogenesis The primary heart tube develops from the precardiogenic mesoderm (Figure 2.1a) located bilaterally in the splanchnic layer of the lateral plate mesoderm of the embryo. These cardiogenic plates fuse rostrally in the midline and form a crescent-shaped structure, which constitutes the primary myocardial heart tube (Figure 2.1b). The inner lining of this tube is formed by cardiac jelly and endocardial cells that are connected to the endothelium of the embryonic vascular plexus. There is most probably a dual origin of the endocardial cells inside the myocardial tube, consisting of differentiated precardiac mesoderm and splanchnic mesoderm-derived cells.2,5 The endoderm of the underlying developing primary gut plays an important inductive role6 through signaling molecules such as the bone morphogenetic protein (BMP) and fibroblastic growth factor (FGF) families.7 The formed primary heart tube is never completely symmetric (Figures 2.1b and 2.1c), and genetic determinants of sidedness8 and cardiac looping9 are present. Already in the cardiogenic plate stage, predetermination of future cardiac segments can be distinguished.10 Data on the primitive cardiac segments are somewhat confusing, but most recent data based on extensive and minute tracing studies are in favor of a small atrial compartment, an atrioventricular canal, lined on the inside by atrioventricular cushions,11 and a primitive left ventricle (Figures 2.1c and 2.2a).12 In the human embryo this primary heart tube already starts to beat with peristaltic contractions at 3 weeks of development. Knockout studies of genes that are essential for primary cardiogenesis lead to early embryo lethality in mice. Heterozygous mutations of some of these genes in the human population can lead to congenital malformations such as, for example, those described for Nkx2.5 mutations.13

10

Fetal Cardiology

as well as elegant tracing of cell clones in mouse embryos12 have proved that essential parts of the cardiac myocardium are newly recruited. The splanchnic mesoderm forming the primary heart tube is referred to as the first heart field, and the newly recruited myocardium derives from the second heart field (Figures 2.2a and 2.3). While the dorsal mesocardium is interrupted in its mid-portion (Figure 2.2a), this myocardium can only reach the heart tube at the arterial pole and the venous pole. The emergence of genes and gene patterns during development allows us to distinguish some of them as marker genes for specific cardiac progenitor lines. For determination of the second heart field, the Isl-1 gene is an important marker.3 LacZ reporter gene experiments have shown that part of the outflow tract as well as the major part of the atria are derived from this heart field. Other experimental studies show a refinement of this myocardial addition, and it is necessary to describe details at the outflow tract (arterial pole) and inflow tract (venous pole) separately.

brain

paa pc

pc

(a)

(b)

oft

SCV

Ao Pu

sa

la ra

lv

Recruitment of myocardium at the arterial pole

Ao

a lv (c) rv icv (d)

Figure 2.1 (a) Primitive streak (*) stage embryo showing the right and left precardiac mesoderm (pc). (b) The primary heart tube is fed by the sinus venosus with the left and right cardinal veins (blue). The arterial pole is depicted in red, showing the first pharyngeal arch arteries (paa). (c) The segmental organization of the heart tube allows one to distinguish the primitive atrium (a), primitive left ventricle (lv), and the outflow tract (oft). The intersegmental transitional rings are depicted in various colors. (d) In an adult heart the transitional rings can still be traced. Note development of the nodal tissue, sinus node (sa), and atrioventricular node (*). Segmentation is complete. Ao, aorta; icv, inferior cardinal vein; la, left atrium; lv, left ventricle; Pu, pulmonary trunk; ra, right atrium; rv, right ventricle; SCV, superior cardinal vein.

Secondary cardiogenesis and organogenesis That the primary heart tube is not a small homunculus in which all future segments of the heart are already present is becoming more and more clear. Many textbooks have indicated such a schematic representation for simplicity.14 Earlier marker experiments in chicken embryos by De La Cruz et al15 and more recent work of Moreno-Rodriguez et al16

Dependent on the marker experiments and reporter gene constructs, several temporospatial patterns for recruitment of myocardium of the second heart field have been distinguished in recent literature: (1) the complete right ventricle including the complete arterial outflow tract, (2) the complete arterial outflow tract with proximal and distal outflow tract cushions (the anterior heart field), and (3) the arterial outflow tract lined by the distal outflow tract cushion (secondary heart field).17 Figure 2.3 provides a representation of this confusing nomenclature, explaining that the secondary heart field is the distal part of the anterior heart field and that both are part of the second heart field. The latter also comprises the area of the developing pharyngeal arch arteries forming, for example, the future aortic arch. To fully understand the possible clinical relevance of the above information it is necessary to introduce an essential further extracardiac cell population. This player is the neural crest cell derived from the crest of the neural tube, and migrates through the splanchnic mesoderm of the second heart field. Until very recently, based on neural crest ablation experiments18 this cell type was held responsible for many cardiac outflow tract malformations such as persistent truncus arteriosis, pulmonary stenosis and atresia, tetralogy of Fallot, double outlet right ventricle, and aortic arch malformations. This spectrum is ideally exemplified in the human 22q11 deletion syndrome that also shows other neural crest cell-influenced abnormalities in, for example, the face and thymus. The most essential gene in the 22q11 deletion syndrome is Tbx1.19 This gene is expressed in the second heart field mesenchyme at the arterial pole and

Cardiac morphogenesis

rv

11

paa paa

oft

shf a

lv pc

dAo peo

ls

g

(a)

(b)

Figure 2.2 (a) Mid-sagittal view of the cardiac region. a, atrium; dAo, dorsal aorta; g, gut; ls, left sinus horn; lv, left ventricle inflow; paa, pharyngeal arch arteries; peo, proepicardial organ; pc, pericardial cavity; rv, right ventricle outflow; shf, second heart field. Note that the posterior heart field has contributed to both the outflow tract (oft) and atrial myocardium. These are separated by a breakthrough of the pericardial cavity (*), the oblique sinus. (b) Neural crest cells depicted in blue have been carried along by the shf, contributing to both poles of the heart tube. Modified after H Lie-Venema et al. Scientific World Journal 2007; 7: 1777–98.

Cardiac neural crest PAA Neural plate

Secondary heart field

Anterior heart field

DOT

OFT

POT RV

ggL Isl-1First heart field

LV AVC atria

Splanchnic mesoderm Isl-1+ Second heart field

Posterior heart field

Atria IFT SAN CCS SV PV&CV

CCS

PEO

Figure 2.3 This illustration incorporates the various lineages as derived from the literature, and shows the contributions from the first and second heart fields as well as the neural crest to the definitive heart. AVC, atrioventricular canal; CCS, central conduction system; CV, cardinal veins; DOT, distal outflow tract; ggL, cardiac ganglia; IFT, inflow tract; LV, left ventricle; OFT, outflow tract; PAA, pharyngeal arch arteries; PEO, proepicardial organ; POT, proximal outflow; PV, pulmonary veins; RV, right ventricle; SAN, sinoatrial node; SV, sinus venosus.

12

Fetal Cardiology

not in the neural crest cells. This leads to the conclusion that disturbed interaction of second heart field and neural crest cells is essential for the spectrum of anomalies. It explains that a great number of genes expressed in either cell population can lead to comparable cardiac malformations, broadening immensely our scope of understanding of the pathomorphogenesis of outflow tract anomalies. Research is now focusing on the cell–cell and cell–matrix interactions of outflow tract populations. A last epigenetic candidate that is emerging as an important factor in outflow tract remodeling is the role of shear stress on the wall and pharyngeal arch arteries.20 Essential genes such as endothelin1, Kruppel-like factor 2 (KLF2),21 and transforming growth factor (TGFβ1)22 might have their effect through mechanical sensors, including the possible role of the cytoskeleton through an endocardial/endothelial cilia system.23

Relevance for understanding outflow tract congenital malformations The above findings imply that many genetic as well as environmental pathways can lead to the spectrum of outflow tract malformations. Hampering myocardial recruitment at the outflow tract such as seen in Tbx1 mutations will lead to outflow tract shortening and abnormal remodeling of the inner curvature myocardium. As a consequence, the aortic orifice is not properly wedged in between the atria and not brought into correct position while connecting to the left ventricle. The resulting abnormalities

are dextroposition of the aorta with a double outlet right ventricle as its most severe form (the most common anomaly in animal models). In these cases, neural crest cells can still migrate normally and reach the pharyngeal arch arteries and the endocardial outflow tract cushions (Figure 2.2b). However, through abnormal interaction with the surrounding pharyngeal arch mesenchyme and endothelium, the outflow tract myocardium and its endocardial cushions do not induce normal aortic arch formation and outflow tract septation, such as seen in the TGFβ2 knockout mouse.24 If the neural crest cells are primarily abnormal, as in the Pax3 knockout mouse, no outflow tract septation occurs at all, and a persistent truncus arteriosus (PTA) or common arterial trunk will be the result.25 A recent observation of our own group26 sheds some light on development of the tetralogy of Fallot with subaortic stenosis. In a vascular endothelial growth factor (VEGF)120/120 model27 an aberrant upregulation of VEGF120 in the subpulmonary outflow tract is present. This area shows increased myocardial apoptosis and abnormal expression in the Notch and Jagged pathway. The subpulmonary outflow tract is shortened, and shows hyperplastic outflow tract cushions leading to marked subpulmonary stenosis. In combination with factors already described for dextroposition of the aorta, the development of tetralogy of Fallot can be clarified. The specific sensitivity of the pulmonary outflow tract myocardium might relate to distinct genetic coding areas in the subpulmonary and subaortic outflow tract region28 (Figure 2.4 shows VEGF120/120 three-dimensional (3D) reconstruction). Epigenetic factors such as hyperglycemia in diabetic pregnancies may likewise lead to similar malformations, most probably affecting second heart field mesenchyme as well as neural crest cells.29,30

VEGF 120 120

VEGF+/+ Pu Pu

lv rv

(a)

(b)

Figure 2.4 (a) Ventral view of a 3D reconstruction of a 12.5-day-old vascular endothelial growth factor (VEGF)+/+ mouse embryo. The pulmonary trunk (Pu) is depicted in orange, the outflow tract cushions in blue.(b) The same view of a VEGF120/120 mouse. Note the decreased size of the pulmonary trunk. The area with apoptotic cells is represented in pink.

Cardiac morphogenesis

Recruitment of cardiac cells at the venous pole At the venous pole the growth of the atria including the incorporation of sinus venosus myocardium in the dorsal wall of the atria is an important mechanism (Figures 2.2a and 2.3). Tracing of Isl1-positive cells shows the extent of the incorporation.3 In contrast to the outflow tract this area and the derived sinus venosus myocardium have specific gene expression patterns including Tbx18,31 Shox2,32 podoplanin,33 and BMPs. Therefore, in parallel with the anterior heart field at the outflow tract, we have reserved the term posterior heart field for this region. Recently, we and others have discovered that this sinus venosus myocardium is initially Nkx2.5-negative31,33 (Figure 2.5). Nkx2.5-negative but podoplanin-positive areas form a U-shaped band that partly surrounds the cardinal veins on both left and right sides and includes the central area of the venous confluence (Figure 2.5).

Neural crest cells At the inflow tract there are two extracardiac cell populations that deserve some special attention. The first of these are the neural crest cells (Figure 2.2b) that play a less conspicuous part compared to the outflow tract. Major congenital abnormalities have not been reported from avian ablation studies18 or neural crest cell-specific knockout mice.25 Our own studies have shown that neural crest cells reach the inflow tract, where they are important for development of the sympathetic and parasympathetic innervation.34

13

Furthermore, they form a marked ring around the pulmonary venous anlage.35 We have also described a population of neural crest cells with undetermined differentiation, migrating through the dorsal mesocardium into the atrioventricular cushion mesenchyme that takes up position surrounding the developing atrioventricular conduction system.36 These cells go into apoptosis, and are postulated to have an inductive effect upon differentiation of the atrioventricular conduction system. Migration of the neural crest cells to the inflow tract, as determined by retroviral lacZ tracking studies in chicken embryos,36 has been contested in the literature, even in neural crest reporter mice such as the Wnt-1 Cre mouse,37 but was recently confirmed in two mouse reporter strains.38

Epicardium A more prominent role at the inflow tract is seen for the development of the epicardium. In our opinion, the epicardial cells derive from the posterior heart field and its covering coelomic wall mesothelium (Figures 2.2a and 2.3). Following a process of epithelial-to-mesenchymal transformation (EMT),11 these mesothelial cells differentiate into the already described sinus venosus myocardium and form an epithelial structure referred to as the proepicardial organ (PEO) in the avian embryo.39 The epicardial cells migrate over the naked myocardial heart tube.40 It is evident that retinaldehyde dehydrogenase (RALDH) and retinoic acid play an important role in guiding this process.41 After covering the heart, the epicardial cells undergo EMT in which a mesenchymal subepicardial layer is formed from epicardium-derived cells (EPDCs)42 (reviewed in references 43 and 44). Subsequently, these EPDCs migrate

cv cv

oft la ra

pv (a)

(b)

Figure 2.5 (a) A 3D reconstruction of an 11.5-day-old mouse from dorsal. The Nkx2.5-negative area (green) is seen as a U-shaped part of the mesoderm connecting the left and right cardinal veins. The left and right sinus nodes are indicated (*). Note the size difference, the right one being larger. cv, cardinal veins (blue); la/ra, left and right atria; oft, ventricular outflow tract (brown); pv, pulmonary vein (pink). (b) A 13.5-day-old mouse embryo from dorsal. There is still a large Nkx2.5-negative area (green) between the cardinal veins and wrapping around the sinus venosus. The left sinus node has become inconspicuous, whereas the right node (*) has expanded considerably.

14

Fetal Cardiology

atrium

av

avc ventricle

pc ec

Endocardium

smc fb cm

Cardiomyocytes Epicardium

Figure 2.6 Part of the myocardial wall covered by the epicardium. The epicardium-derived cells (EPDCs) in gray have migrated between the cardiomyocytes and have differentiated into cardiac fibroblasts. In the inset: EPDCs have gathered around the coronary endothelial tubes (ec) to differentiate into the smooth muscle cells (smc) and adventitial fibroblasts (fb). The myocardial-derived Purkinje cells (pc) are depicted in green and are in close association with EPDCs. av, atrioventricular groove; avc, atrioventriucular cushion; cm, cardiomyocytes.

into the atrial and ventricular myocardium to form the interstitial fibroblast and also take up a subendocardial position (Figure 2.6). A second wave of EMT is seen when the coronary capillary plexus is remodeled into an arterial and venous system in which the EPDCs are the source of smooth muscle cells and periarterial fibroblasts. The differentiation capacities of the EPDCs are still under investigation, with a main controversial issue being whether they are also the source of the coronary endothelial cells45 or whether the endothelial cells use the subepicardial space as a migration matrix but are otherwise derived from the liver endothelium.46 We have shown by PEO inhibition and rescue studies that EPDCs are essential for formation of the compact myocardium,47 formation of the main coronary arteries,48,49 and differentiation of the Purkinje network50 (Figure 2.6). As EPDCs migrate through the atrioventricular sulcus into the endocardial cushions, we are currently studying their role in these areas. There is an indication that they are important for embryonic atrioventricular dissociation and formation of the atrioventricular fibrous annulus, in which expression of the matrix protein periostin by EPDCs plays a prominent role.51

Relevance for the understanding of congenital malformations Severe developmental inflow tract malformations are seen in atrial sidedness. Related genes such as Pitx2 are being studied, and reveal interesting information on atrial isomerism and differentiation of the sinus venosus myocardium, including the sinoatrial node.31 From our observations it can be deduced that both formation of the atrial septum, in which the left venous valve is incorporated, and the development of parts of the conduction system might be disturbed when sinus venosus myocardium is not properly differentiated.33 The sinus venosus myocardium is transformed during embryonic development into what we refer to as the sinoatrial conduction system (Figure 2.1d). This encompasses the anlage of a left sinoatrial nodal region, surrounding the coronary sinus, part of the myocardial cells surrounding the pulmonary veins (Figure 2.5), and the right (future terminal crest) and left (incorporated into the atrial system) venous valves and their confluence, the septum spurium. It is very important to emphasize that use of the term

Cardiac morphogenesis

sinoatrial conduction system only links developmentally correlated structures. A functional capacity has only been reported during normal development for the right sinoatrial nodal region after an initial start on the left side. Clinically, a spectrum of rhythm disturbances can be explained from the disposition of the sinoatrial conduction system, as the above described areas are clearly recognizable as arrhythmogenic foci in the adult heart and might reflect dedifferentiation of embryonic cells reinitiating conduction system capacities52 that become aberrant. The inductive and formative role of EPDCs also sheds light on a number of cardiac malformations. Most prominent are the anlage and differentiation of the coronary vascular system. We have shown that partial inhibition of EPDC differentiation can lead to abnormalities of the main coronary arteries, including pinpoint orifices and complete absence of main coronary stems.48,49 In the latter case, aberrant connections with the cardiac lumen, so-called fistulae or ventriculocoronary arterial communication (VCAC), will develop to sustain myocardial perfusion.48,49 There is no major role for the persistence of embryonic ventricular communications, because they do not exist in normal early life, as suggested.53 This notion might be the erroneous origin of the so-called Thebesian vessels. In a clinical setting this background knowledge is important for differentiating between cases with pulmonary atresia without ventricular septal defect (VSD) that can present with or without VCAC. Fetal ultrasound diagnostics revealed the early presence of fetal VCAC in combination with pulmonary stenosis that later on developed into pulmonary atresia.54 Histopathology showed already severe coronary malformations with intimal thickening and even complete closure of vessels, and the presence of large VCAC.55 We hypothesize that these fetuses suffer from a primary coronary vessel (EPDC-derived) anomaly, whereas in cases with primary pulmonary atresia and a normal coronary vasculature the original problem might have been second heart field- or neural crest cell-derived.33 The role of EPDCs in formation of the compact myocardium might be of relevance for the understanding of some cardiomyopathies, including spongy myocardium or myocardial non-compaction, as well as abnormally thin myocardium. In cardiomyopathies the interaction of myocardium and the EPDC-derived interstitial fibroblast is essential. Defects in either or both might lead to abnormal myocardial differentiation. Recently, this was shown to be effective in the RXRα knockout mouse,56 and also by our own observations in the SP3 knockout mouse.57 The last aspect of EPDC involvement is in differentiation of the fibrous heart skeleton, the atrioventricular valves, and the inductive role in Purkinje fiber differentiation. We have shown that the gene periostin, expressed by EPDCs, might play an important role in differentiation of the fibrous heart skeleton and formation of the atrioventricular fibrous annulus. This annulus differentiates slowly, and myocardial cells most probably transdifferentiate into

15

a fibrous phenotype influenced by periostin-producing EPDCs.58 Formation of the fibrous annulus and its electrical isolating capacities might be essential for the switch from base-to-apex conduction propagation.51 Relatively late progress of this isolation process during formation of the right atrioventricular orifice (Chapter 1 and reference 59) might explain transient late fetal rhythm disturbances and Wolff–Parkinson–White syndrome. A similar disturbance of EPDC function on the borderline of the atrioventricular cushion and the underlying myocardium might inhibit delamination of the tricuspid valve and subsequent development of the Ebstein malformation.

Concluding remarks In conclusion, there will be the necessity to merge the amount of data from molecular cardiac development programs with the increasing knowledge from large genomic data screens from human cardiac malformations. Our current feeling is that in the development of cardiac malformations we should more prominently include environmental (epigenetic) factors as seen in diabetes29,30 and hyperhomocysteinemia60 as well as hemodynamic factors.21 The greatest challenge is to sort out cause and effect, as once a cardiac abnormality develops it will automatically evoke flow disturbances. Advancing refinement of prenatal ultrasound diagnostics and the tools to actually quantitate flow and shear stress during development will broaden our mechanistic insight.

Acknowledgments Nynke van den Akker, Edris Mahtab, and Nathan Hahurij are greatfully acknowledged for help with some of the figures. Ron Slagter and Jan Lens are acknowledged for artwork, and Joke van Benten for secretarial assistance.

References 1. Pexieder T, Wenink ACG, Anderson RH. A suggested nomenclature for the developing heart. Int J Cardiol 1989; 25: 255–64. 2. Gittenberger-de Groot AC, Bartelings MM, DeRuiter MC et al. Basics of cardiac development for the understanding of congenital heart malformations. Pediatr Res 2005; 57: 169–76. 3. Cai CL, Liang X, Shi Y et al. Isl1 identifies a cardiac progenitor population that proliferates prior to differentiation and contributes a majority of cells to the heart. Dev Cell 2003; 5: 877–89.

16

Fetal Cardiology

4. Gittenberger-de Groot AC. Mannheimer Lecture. The quintessence of the making of the heart. Cardiol Young 2003; 13: 175–83. 5. Noden DM. Origins and patterning of avian outflow tract endocardial tissues and cushion mesenchyme. Anat Rec 1990; 226: 72A–3A. 6. Waldo K, Zdanowicz M, Burch J et al. A novel role for cardiac neural crest in heart development. J Clin Invest 1999; 103: 1499–507. 7. Lough J, Sugi Y. Endoderm and heart development. Dev Dyn 2000; 217: 327–42. 8. Levin M, Pagan S, Roberts DJ et al. Left/right patterning signals and the independent regulation of different aspects of situs in the chick embryo. Dev Biol 1997; 189: 57–67. 9. Olson EN, Srivastava D. Molecular pathways controlling heart development. Science 1996; 272: 671–6. 10. Yutzey C, Rhee JT, Bader D. Expression of the atrial-specific myosin heavy chain AMHC1 and the establishment of anteroposterior polarity in the developing chicken heart. Development 1994; 120: 871–83. 11. Eisenberg LM, Markwald RR. Molecular regulation of atrioventricular valvuloseptal morphogenesis. Circ Res 1995; 77: 1–6. 12. Meilhac SM, Esner M, Kelly RG et al. The clonal origin of myocardial cells in different regions of the embryonic mouse heart. Dev Cell 2004; 6: 685–98. 13. Benson DW, Silberbach GM, Kavanaugh-McHugh A et al. Mutations in the cardiac transcription factor NKX2.5 affect diverse cardiac developmental pathways. J Clin Invest 1999; 104: 1567–73. 14. Gittenberger-de Groot AC, DeRuiter MC, Bartelings MM et al. Embryology of congenital heart disease. In: Crawford MH, DiMarco JP, eds. Cardiology. London: Mosby International, 2001: 1217–27. 15. De La Cruz M, Sanchez-Gomez C, Palomino MA. The primitive cardiac regions in the straight tube heart (Stage 9) and their anatomical expression in the mature heart: an experimental study in the chick heart. J Anat 1989; 165: 121–31. 16. Moreno-Rodriguez RA, Krug EL, Reyes L et al. Bidirectional fusion of the heart-forming fields in the developing chick embryo. Dev Dyn 2006; 235: 191–202. 17. Kelly RG. Molecular inroads into the anterior heart field. Trends Cardiovasc Med 2005; 15: 51–6. 18. Kirby ML, Waldo KL. Neural crest and cardiovascular patterning. Circ Res 1995; 77: 211–15. 19. Lindsay EA, Vitelli F, Su H et al. Tbx1 haploinsufficieny in the DiGeorge syndrome region causes aortic arch defects in mice. Nature 2001; 410: 97–101. 20. Hogers B, DeRuiter MC, Gittenberger-de Groot AC et al. Unilateral vitelline vein ligation alters intracardiac blood flow patterns and morphogenesis in the chick embryo. Circ Res 1997; 80: 473–81. 21. Groenendijk BCW, Hierck BP, Vrolijk J et al. Changes in shear stress-related gene expression after experimentally altered venous return in the chicken embryo. Circ Res 2005; 96: 1291–8. 22. Gittenberger-de Groot AC, Azhar M, Molin DG. Transforming growth factor beta-SMAD2 signaling and aortic arch development. Trends Cardiovasc Med 2006; 16: 1–6.

23. Van der Heiden K, Groenendijk BCW, Hierck BP et al. Monocilia on chicken embryonic endocardium in low shear stress areas. Dev Dyn 2006; 235: 19–28. 24. Bartram U, Molin DGM, Wisse LJ et al. Double-outlet right ventricle and overriding tricuspid valve reflect disturbances of looping, myocardialization, endocardial cushion differentiation, and apoptosis in TGFβ2-knockout mice. Circulation 2001; 103: 2745–52. 25. Conway SJ, Henderson DJ, Kirby ML et al. Development of a lethal congenital heart defect in the splotch (Pax3) mutant mouse. Cardiovasc Res 1997; 36: 163–73. 26. van den Akker NMS, Molin DGM, Peters PPWM et al. Tetralogy of Fallot and alterations in VEGF- and Notchsignalling in mouse embryos solely expressing the VEGF120 isoform. Circ Res 2007; 100: 842–9. 27. Stalmans I, Lambrechts D, Desmet F et al. VEGF: a modifier of the del22q11 (DiGeorge) syndrome? Nat Med 2003; 9: 173–82. 28. Bajolle F, Zaffran S, Kelly RG et al. Rotation of the myocardial wall of the outflow tract is implicated in the normal positioning of the great arteries. Circ Res 2006; 98: 421–8. 29. Molin DG, Roest PAM, Nordstrand H et al. Disturbed morphogenesis of cardiac outflow tract and increased rate of aortic arch anomalies in the offspring of diabetic rats. Birth Defects Res A Clin Mol Teratol 2004; 70: 927–38. 30. Roest PAM, van Iperen L, Vis L et al. Exposure of neural crest cells to elevated glucose leads to congenital heart defects, an effect that can be prevented by N-acetylcysteine. Birth Defects Res A Clin Mol Teratol 2007; 79: 231–5. 31. Christoffels VM, Mommersteeg MT, Trowe MO et al. Formation of the venous pole of the heart from an Nkx2-5negative precursor population requires Tbx18. Circ Res 2006; 98: 1555–63. 32. Blaschke RJ, Hahurij ND, Kuijper S et al. Targeted mutation reveals essential functions of the homeodomain transcription factor Shox2 in sinoatrial and pacemaking development. Circulation 2007; 115: 1830–8. 33. Gittenberger-de Groot AC, Mathab EAF, Hahurij ND et al. Nkx2.5-negative myocardium of the posterior heart field and its correlation with podoplanin expression in cells from the developing cardiac pacemaking and conduction system. Anat Rec 2007; 290: 115–22. 34. Verberne ME, Gittenberger-de Groot AC, VanIperen L et al. Distribution of different regions of cardiac neural crest in the extrinsic and the intrinsic cardiac nervous system. Dev Dyn 2000; 217: 191–204. 35. Poelmann RE, Jongbloed MR, Molin DG et al. The neural crest is contiguous with the cardiac conduction system in the mouse embryo: a role in induction? Anat Embryol (Berl) 2004; 208: 389–93. 36. Poelmann RE, Gittenberger-de Groot AC. A subpopulation of apoptosis-prone cardiac neural crest cells targets to the venous pole: multiple functions in heart development? Dev Biol 1999; 207: 271–86. 37. Jiang X, Rowitch DH, Soriano P et al. Fate of the mammalian cardiac neural crest. Development 2000; 127: 1607–16. 38. Nakamura T, Colbert MC, Robbins J. Neural crest cells retain multipotential characteristics in the developing valves and label the cardiac conduction system. Circ Res 2006; 98: 1547–54.

Cardiac morphogenesis

39. Viragh S, Gittenberger-de Groot AC, Poelmann RE et al. Early development of quail heart epicardium and associated vascular and glandular structures. Anat Embryol (Berl) 1993; 188: 381–93. 40. Vrancken Peeters M-PFM, Mentink MMT, Poelmann RE et al. Cytokeratins as a marker for epicardial formation in the quail embryo. Anat Embryol (Berl) 1995; 191: 503–8. 41. Xavier-Neto J, Shapiro MD, Houghton L et al. Sequential programs of retinoic acid synthesis in the myocardial and epicardial layers of the developing avian heart. Dev Biol 2000; 219: 129–41. 42. Gittenberger-de Groot AC, Vrancken Peeters M-PFM, Mentink MMT et al. Epicardium-derived cells contribute a novel population to the myocardial wall and the atrioventricular cushions. Circ Res 1998; 82: 1043–52. 43. Winter EM, Gittenberger-de Groot AC. Epicardium-derived cells (EPDCs) in cardiogenesis and cardiac regeneration. Cell Mol Life Sci 2007; 64: 692–703. 44. Perez-Pomares JM, Phelps A, Munoz-Chapuli R et al. The contribution of the proepicardium to avian cardiovascular development. Int J Dev Biol 2001; 45: S155–6. 45. Perez-Pomares JM, Carmona R, Gonzalez-Iriarte M et al. Origin of coronary endothelial cells from epicardial mesothelium in avian embryos. Int J Dev Biol 2002; 46: 1005–13. 46. Poelmann RE, Gittenberger-de Groot AC, Mentink MMT et al. Development of the cardiac coronary vascular endothelium, studied with antiendothelial antibodies, in chickenquail chimeras. Circ Res 1993; 73: 559–68. 47. Gittenberger-de Groot AC, Vrancken Peeters M-PFM, Bergwerff M et al. Epicardial outgrowth inhibition leads to compensatory mesothelial outflow tract collar and abnormal cardiac septation and coronary formation. Circ Res 2000; 87: 969–71. 48. Eralp I, Lie-Venema H, DeRuiter MC et al. Coronary artery and orifice development is associated with proper timing of epicardial outgrowth and correlated Fas ligand associated apoptosis patterns. Circ Res 2005; 96: 526–34. 49. Lie-Venema H, Gittenberger-de Groot AC, van Empel LJP et al. Ets-1 and Ets-2 transcription factors are essential for normal coronary and myocardial development in chicken embryos. Circ Res 2003; 92: 749–56. 50. Eralp I, Lie-Venema H, Bax NA et al. Epicardiumderived cells are important for correct development of the

51.

52.

53.

54.

55.

56.

57.

58.

59.

60.

17

Purkinje fibers in the avian heart. Anat Rec 2006; 288A: 1272–80. Kolditz DP, Wijffels MCEF, Blom NA et al. Persistence of functional atrioventricular accessory pathways in postseptated embryonic avian hearts: implications for morphogenesis and functional maturation of the cardiac conduction system. Circulation 2006; 115: 17–26. Jongbloed MRM, Schalij MJ, Poelmann RE et al. Embryonic conduction tissue: a spatial correlation with adult arrhythmogenic areas? Transgenic CCS/lacZ expression in the cardiac conduction system of murine embryos. J Cardiovasc Electrophysiol 2004; 15: 349–55. Wearn JT, Mettier SR, Klumpp TG et al. The nature of the vascular communications between the coronary arteries and the chambers of the heart. Am Heart J 1933; 9: 143–64. Chaoui R, Tennstedt C, Göldner B et al. Prenatal diagnosis of ventriculo-coronary communications in a secondtrimester fetus using transvaginal and transabdominal color Doppler sonography. Ultrasound Obstet Gynecol 1997; 9: 194–7. Gittenberger-de Groot AC, Tennstedt C, Chaoui R et al. Ventriculo coronary arterial communications (VCAC) and myocardial sinusoids in hearts with pulmonary atresia with intact ventricular septum: two different diseases. Prog Pediatr Cardiol 2001; 13: 157–64. Jenkins SJ, Hutson DR, Kubalak SW. Analysis of the proepicardium-epicardium transition during the malformation of the RXRalpha-/- epicardium. Dev Dyn 2005; 233: 1091–101. Van Loo PF, Mahtab EA, Wisse LJ et al. Transcription factor Sp3 knockout mice display serious cardiac malformations. Mol Cell Biol 2007; 27: 8571–82. Lie-Venema H, Eralp I, Markwald RR et al. Periostin expression by epicardium-derived cells is involved in the development of the atrioventricular valves and fibrous heart skeleton. Differentiation 2008; 76: 809–19. Jongbloed MR, Wijffels MC, Schalij MJ et al. Development of the right ventricular inflow tract and moderator band: a possible morphological and functional explanation for Mahaim tachycardia. Circ Res 2005; 96: 776–83. Boot MJ, Steegers-Theunissen RP, Poelmann RE et al. Cardiac outflow tract malformations in chick embryos exposed to homocysteine. Cardiovasc Res 2004; 64: 365–73.

3 Cardiac anatomy and examination of specimens Cornelia Tennstedt Cardiac anatomy Owing to the quality of prenatal ultrasound and the expanded experience of prenatal diagnosticians, it is possible to observe congenital heart malformations in increasingly greater detail and at an ever earlier stage of gestation.1 Since it is on the basis of ultrasound findings that decisions to terminate pregnancies are made, there is a critical need for monitoring and confirmation of the prenatal diagnosis. This need can be adequately met only by autopsy. The diagnosis of congenital heart malformations is considered to be difficult. In fact, with the proper systematic approach, most cases of congenital heart malformation are relatively easy to diagnose, although there are cases that are extremely difficult.2,3 For this reason, it is necessary that the system for describing and diagnosing such conditions be both simple and comprehensive, on the one hand, and applicable to all cases, on the other. The goal of a descriptive system of the heart is to allow the heart atria and ventricles, the aorta, and the pulmonary trunk to be identified and distinguished from each other. In this chapter, the characteristic features of the various components of the normal fetal heart are described.

The morphologically right atrium In the normal heart, the morphologically right atrium forms the right, front part of the cardiac mass (Figure 3.1). The morphologically right atrium consists of the following components: the venous component, the vestibule, the septum, and the appendage. The external wall of the atrium consists of two parts: the venous component (sinus venarum cavarum, or posterior part), into which the superior and inferior caval veins, as well as the coronary sinus, open; and an anterior part, the appendage, which extends forward and surrounds the right wall of the aorta. The morphologically right appendage has a broad, triangular shape. Internally, it is connected with the smooth-walled venous component of the atrium. This

connection is marked in the morphologically right appendage by a prominent crest with pectinate muscles. The ostium of the tricuspid valve points diagonally to the right and can be considered as the base of the vestibule. The ostium is smooth-walled and the leaflets of the tricuspid valve are attached to its edge (Figure 3.2). The terminal crest runs laterally, forming the connection between the venous component and the right appendage. It emerges from the anterior part of the septal surface and curves in front of the orifice of the superior caval vein. The superior caval vein is located in the roof of the atrium and enters the right atrium between the terminal crest and the superior rim of the oval fossa. The border of the crest is reinforced by fibrous structures. These structures separate the orifices of the inferior caval vein and the coronary sinus from the atrial appendage, becoming the venous valves of the inferior caval vein (Eustachian valve) and the coronary sinus (Thebesian valve). The Eustachian valve can be prominent, but it can also be completely lacking; in individual cases it can form a network with pronounced fenestration, which is called the Chiari network. The Eustachian valve is usually small; it protects the mouth of the inferior caval vein and may extend to the inlet of the superior caval vein. The tendon of Todaro is the continuation of the venous valves. The triangle between the tendon and the tricuspid valve marks the location of the atrioventricular conduction tissue.4 The septal surface consists of the floor of the oval fossa and the atrioventricular septum. The coronary sinus opens above the posterior interventricular groove into the right atrium. The so-called ‘septum secundum’ does not belong to the septum proper, but simply forms the surrounding wall of the atrial chambers.

The morphologically left atrium The morphologically left atrium is the most posterior chamber of the heart. It consists of the following components: the venous component, a septal surface, the vestibule, and an appendage.

20

Fetal Cardiology

Figure 3.2 Dissection showing the typical atrioventricular junction of the right heart. FO, fossa ovalis; CS, coronary sinus; TV, tricuspid valve.

internally the left appendage is trabeculated, but the trabeculations are finer.

The morphologically right ventricle Figure 3.1 Frontal view of the heart, showing the various components of the normal fetal heart. AAO, ascending aorta; DA, ductus arteriosus; DAO, descending aorta; PT, pulmonary trunk; RA, right atrium; LA, left atrium; RV, right ventricle; LV, left ventricle.

At the top, the venous component admits the four pulmonary veins, two on each side. It is considerably larger than the appendage. The septal surface of the left atrium is at an angle; it is in part uneven and consists of the left atrial surface of the oval fossa. The valve overlaps the surrounding atrial wall (the septum secundum) from above. The surface of the vestibule of the left atrium is smooth; the leaflets of the mitral valve are attached to it. The pectinate muscles are less pronounced here than in the right atrium, and do not project into the body of the atrium. In contrast with the right atrium, the pectinate muscles do not extend into the area of the atrioventricular junction. The morphologically left appendage has a tubular, hooked shape. It has characteristic grooves at the edge and partly surrounds the pulmonary artery. The connection between the smooth-walled venous component and the left appendage is narrower than in the right atrium, and it is not limited by a crest. As with the right appendage,

The morphologically right ventricle occupies the greatest part of the ventral mass of the heart. It consists of the following: the inlet, the apical trabecular, and the outlet components, which go from the lower right to the upper left in the ventricular mass. The inlet component surrounds and provides support for the leaflets and the tension apparatus of the tricuspid valve, which extends dorsally to the crux cordis. The leaflets of the tricuspid valve are in the septal, anterior–superior, and inferior (or mural) position. All the leaflets arise at the atrioventricular junction. The most characteristic feature of the tricuspid valve is the presence of tendinous cords, which fix the septal leaflet to the ventricular septum. The border between the inlet and the trabecular component is formed by the attachment of the papillary muscles. The trabecular zone reaches the apex of the heart. Characteristic of the apical trabecular component of the right ventricle are coarse trabeculations (Figure 3.3), which provide the best criterion for identification of the morphologically right ventricle in cases in which there is no inlet component. The inlet and outlet components are separated from each other in the roof of the ventricle by a prominent muscular crest, the crista supraventricularis. The outlet component (infundibulum) of the right ventricle is a muscular tube, which generally has smoother walls than the trabecular component. The three leaflets of the pulmonary valve are attached to a completely muscular infundibulum. The posterior wall of the infundibulum is separated from the aorta by an extracardiac space.

Cardiac anatomy and examination of specimens

21

Figure 3.3 Section showing the typical coarse trabeculations of the right ventricle.

The septum of the normal heart has a muscular and a membranous component, with the membranous component being very narrow. The trabecula septomarginalis is a powerful ridge that abuts the right-ventricular surface of the septum.

The morphologically left ventricle As with the right ventricle, the left ventricle consists of an inlet, an apical trabecular component, and an outlet component. The inlet component contains and surrounds the mitral valve, which has an aortic (anterior or septal) leaflet and a posterior (or mural) leaflet. The leaflets are separated from each other by the anterolateral and posteromedial commissures. The two mitral valve leaflets are connected to two groups of papillary muscles, which occupy a posteromedial and anterolateral position below the commissural areas. The most characteristic feature of the mitral valve is that it has no cordal attachments to the ventricular septum. The trabecular component of the left ventricle extends from the origin of the papillary muscles to the apex of the heart. The apical trabecular part of the left ventricle has fine criss-crossed trabeculations (Figure 3.4). The inlet and outlet components of the left ventricle form an acute angle, with the inlet and outlet components being separated by the anterior of the mitral valve. The outlet component of the left ventricle contains the aortic valve. The component consists partially of muscular, partially of fibrous tissue. Dorsally it is incomplete, so that the mitral and aortic valves are connected to each other by fibrous tissue. The aortic valve has three semilunar leaflets called the right coronary, left coronary, and non-coronary leaflets, which are attached to the root of the aorta.

Figure 3.4 Specimen in which the left ventricle is displayed to show the typical fine criss-crossed trabeculations. AO, aorta; MV, mitral valve.

The interventricular septum The interventricular septum consists mainly of muscular tissue, called the muscular septum, and only to a small extent of fibrous tissue, the membranous septum (Figure 3.5). The muscular septum is divided into the inlet, apical trabecular component, and outlet component. On the right-ventricular side, the inlet area septum is bordered by the cordal attachments of the septal tricuspid leaflet. On the left-ventricular side, this line is connected with the dorsal end of the smooth-walled septum. On the rightventricular side, the trabeculations of the apical trabecular component of the septum are rough, whereas they are fine on the left-ventricular side. The outlet component of the septum lies below the distal part of the infundibulum, separating the outlet paths of the right and left ventricles. The membranous septum lies at the point where the inlet, apical trabecular component, and outlet component of the muscular septum come together.

The aorta The aorta leaves the heart in the cranial direction (this section is called the ascending aorta), curves dorsally to the left (this section is called the aortic arch), and then runs caudally from in front of the thoracic spine (this section is called the descending aorta). The ascending aorta emerges from behind the pulmonary trunk out of the conus arteriosus of the left ventricle, and then runs intrapericardially somewhat to the right. The ascending aorta has only two branches: the right and left coronary arteries. These emerge immediately behind the aortic valve in the area of the bulbus aortae, the left

22

Fetal Cardiology

Figure 3.6 Longitudinal section through the aortic valve (AV) showing the ostium of the left coronary artery (LCA) below the left coronary leaflet and the ostium of the right coronary artery (RCA) below the right coronary leaflet.

Figure 3.5 Specimen sectioned along the cardiac long axis, showing the four cardiac chambers, the two atrioventricular valves, and the interventricular septum. meS, membranous septum; S, septum; muS, muscular septum.

coronary artery in the sinus aortae below the left coronary leaflet, and the right coronary artery in the sinus aortae below the right coronary leaflet (Figure 3.6). The ascending aorta passes over into the aortic arch. The great branches of the aorta emerge from the convex side of the aortic arch: the brachiocephalic artery (which divides into the right subclavian artery and the right common carotid artery), the left common carotid artery, and the left subclavian artery. The descending aorta designates that part of the vessel which extends from the ostium of the arterial duct to the aortic bifurcation.

The pulmonary trunk The pulmonary trunk emerges from the conus arteriosus of the right ventricle, terminating there at the pulmonary valve. Below the aortic arch it divides into right and left pulmonary arteries. In the fetal circulation the arterial duct emerges from the bifurcation of the pulmonary trunk, which connects the pulmonary trunk with the descending aorta. The arterial duct demarcates the isthmus of the aorta between the site of emergence of the left subclavian artery and the aortic insertion of the duct. The arterial duct is structurally different from the aorta and the pulmonary artery. The wall of the arterial duct is thicker

than that of both the aorta and the pulmonary trunk. The luminal surface of the duct in the newborn is less smooth than that of the great arteries, and has irregular ridges running lengthwise along it. After birth, the duct closes and becomes the arterial ligament. The longer right pulmonary artery passes under the aortic arch and behind the inferior vena cava to the right bronchus. The shorter left pulmonary artery reaches the left bronchus over a direct path in front of the ascending aorta.

Examination of specimens By applying the method of sequential segmental analysis of the heart, a relatively simple and reliable diagnosis of complex congenital heart malformations can be achieved.5,6 This requires a description of the atrial arrangement, the atrioventricular junction, and the ventriculoarterial junction, in this order. The normal heart consists of three segments: the atria, the ventricular mass, and the great arteries. Each of these three sections of the normal heart has a right and a left side. Congenital heart anomalies can affect one or more of these segments or the great veins. To determine which of the cardiac chambers are normal, the structure of their constant components has to be evaluated morphologically.7 This is followed by a description of the position of the heart in the thorax, the orientation of the apex of the heart, and any associated malformations.

Atrial arrangement The most constant component of the atrium is the atrial appendage. The right and left atrial appendages are

Cardiac anatomy and examination of specimens

23

Figure 3.7 Diagram showing the four possible arrangements of the atria (reproduced with permission from reference 9).

different.8 The shape of the morphologically right appendage is that of a broad triangle, whereas the left appendage is tubular and hooked. The most reliable feature differentiating between the right and left appendages, however, is the nature of the junction between the appendage and the smooth-walled venous component of the atrium. In the morphologically right appendage this junction is wide, and marked by an extensive crest with pectinate muscles extending all round the atrioventricular junction; in the morphologically left appendage this junction is narrow, and has no crest, and no pectinate muscles. The first step in the sequential analysis is the determination of the arrangement of the atrial chambers. There are four different possibilities here (Figure 3.7). In the usual arrangement, the so-called situs solitus, the morphologically right atrium is located to the right of the morphologically left atrium. The mirrored position, the situs inversus, is rare. In some cases the lateralization of the atria is lacking, so that there are two morphologically right appendages (right isomerism) or two morphologically left appendages (left isomerism). Generally, left- or right-atrial isomerism occurs in cases of visceral heterotaxy. Since the judgment as to whether isomerism exists is normally made on the basis of the atrial appendages, the expressions ‘isomerism of the right atrial appendage’ and ‘isomerism of the left atrial appendage’ are also used. The atrial arrangement almost always corresponds to the bronchial morphology. There are four different possibilities here (Figure 3.8). This is especially important in cases in which the arrangement of the atrial chambers is not clear. The identification of the bronchi is based on the fact that the morphologically left bronchus is almost twice as long as the morphologically right bronchus. In addition, the morphologically left bronchus is crossed by the left pulmonary artery before the bronchus divides, which is

Figure 3.8 Diagram showing the four variants of bronchial morphology usually correlating with atrial arrangement (reproduced with permission from reference 9).

not the case for the morphologically right bronchus. Therefore, a long hyparterial bronchus on the left means situs solitus, whereas a long hyparterial bronchus on the right is a sign of situs inversus. Bronchi of equal length are an indication of atrial isomerism. The relationship of the bronchi to the pulmonary arteries is the basis for distinguishing between right and left isomerism.

Variation of the atrioventricular junction This section describes the junction (also called ‘connection’) of the atria with the ventricles, the morphology of the atrioventricular valves and of the ventricles, and the interrelationship of the ventricles. There are two groups of atrioventricular junction: biventricular and univentricular. In biventricular connections each atrium is connected to one ventricle. If the morphologically right atrium is connected to the morphologically right ventricle and the morphologically left atrium is connected to the morphologically left ventricle, both atrioventricular connections are called concordant. An atrioventricular connection is called discordant when the morphologically right atrium

24

Fetal Cardiology

Figure 3.10 Figure 3.9 Diagram showing concordant and discordant atrioventricular connections in the usual and mirror image arrangements (reproduced with permission from reference 9).

is connected to the morphologically left ventricle or the morphologically left atrium is connected to the morphologically right ventricle (Figure 3.9). Atrioventricular concordance and discordance can each occur together with either situs solitus or situs inversus of the heart. If the atrial appendages are isomeric, the biventricular connections cannot be classified in this way and are called ambiguous. In cases of an ambiguous atrioventricular connection the ventricular topology should be described. Here one has to distinguish between a right-hand and a left-hand topology (Figure 3.10). In describing the spatial relationship of the ventricles to each other the following terms are used for the position of the right ventricle with regard to the left ventricle: anterior, posterior, superior, inferior, right, and left. The second group of atrioventricular junctions is called ‘univentricular.’ This connection is characterized by a double ventricle inlet (i.e. both atria open into a single ventricle) or the absence of an atrioventricular junction on the right or the left side. The atria can be connected to a dominant left ventricle (the right ventricle is rudimentary or incomplete); to a dominant right ventricle (the left ventricle is rudimentary or incomplete); or to a solitary ventricle of indeterminate morphology. Rudimentary right ventricles are normally located anterosuperior; rudimentary left ventricles are normally in the posteroinferior

Diagram showing the isomeric arrangement of the atrial appendages. The atrioventricular junctions are concordant, but they are ambiguous (reproduced with permission from reference 9).

position within the ventricular mass. In such cases there are either two atrioventricular valves or a single common valve. If there are two atrioventricular valves, one of them can be stenotic, regurgitant, imperforate or straddling, or sit astride the septum. If there is one atrioventricular valve sitting astride the septum, the heart is intermediate between having a completely biventricular and a completely univentricular connection. There are no defined intermediate categories. If less than 50% of one valve overrides the septum the connection is described as biventricular. The connection is called univentricular only when more than 50% of both valves open into one ventricle.10 Frequently the valves can be stenotic, incompetent, or imperforate. If one atrioventricular connection is missing, there is only a solitary valve that can be incompetent, straddling, or overriding. The relationship between the ventricles does not depend on the atrioventricular connection.

Variation of the ventriculoarterial junction The connection of the arterial trunks to the ventricular mass is here described. The possible relationships of the

Cardiac anatomy and examination of specimens

25

Figure 3.11 Diagram showing concordant and discordant ventriculoarterial junctions. When the aorta and pulmonary trunk arise from the appropriate ventricles, these are concordant. Origins from inappropriate ventricles produce discordant junctions (reproduced with permission from reference 9).

arterial valves to one another and the relationships of the arterial trunks, as well as the morphology of the infundibular structures, are described. The arterial trunks are the third segment of the heart. They are identified by their branching pattern. In normal hearts the aorta gives rise to the systemic and the coronary branches and the pulmonary trunk divides into the right and left pulmonary arteries. Terms used for describing the ventriculoarterial junctions of the heart are: concordant, discordant, double outlet, and single outlet (Figure 3.11). In a concordant or discordant ventriculoarterial connection, each great arterial trunk originates in a separate ventricle. If one of the great arterial valves overrides the ventricular septum, it is assigned to the ventricle associated with its greater part. The ventriculoarterial connection is concordant when the morphologically right ventricle is connected to the pulmonary trunk and the morphologically left ventricle is connected to the aorta. The term ‘ventriculoarterial discordance’ is used to describe the reverse connection. When more than half of both arterial valves are connected to the same ventricle it is called a double-outlet ventriculoarterial connection (or, simply, ‘double outlet’).9 The ventricle may be of right, left, or indeterminate morphology. If the ventricular mass is connected to only one arterial trunk, the arterioventricular connection is called a single-outlet ventriculoarterial connection (or, simply, ‘single outlet’), i.e. it is connected to a single common arterial trunk or to an aorta with an atretic pulmonary trunk lacking a connection to the ventricle, or to a pulmonary trunk

Figure 3.12 Diagram showing the four variants of a single arterial trunk from the heart (reproduced with permission from reference 9).

with an atretic aorta lacking a connection to the ventricle (Figure 3.12). With regard to the arrangement of the ventriculoarterial junction, the arterial valves can be perforate, or one of the arterial valves can be imperforate. One or both of the arterial valves may override the ventricular septum. If there is a common arterial valve, either it overrides the ventricular septum or it is committed to one ventricle. The structure separating the two arterial valves and the two ventricle outlet paths is the infundibular septum. The most frequent arrangement with regard to the infundibular morphology is that the right ventricle has a complete, muscular infundibulum which is lower than its arterial valve. A fibrous continuity exists between the arterial valve issuing from the left ventricle and the left atrioventricular valve. Another possibility is that on both sides a complete, muscular infundibulum lies below the arterial valves or there is a fibrous continuity on both sides between the arterial valves and the atrioventricular valve.

Position of the heart On describing the position of the heart in the chest it is necessary to give the basic position of the heart and the orientation of the cardiac apex. The heart can lie primarily within the left chest (levocardia), primarily within the right chest (dextrocardia), or in the midline. The cardiac apex can point to the left, to the right, or to the middle.

26

Fetal Cardiology

The orientation of the cardiac apex is independent of the position of the heart. An unusual cardiac position does not represent an abnormality of the cardiac morphology. A morphologically normal heart can be abnormally located. The basic position of the heart should be described separately from the cardiac morphology. In very rare cases the heart can have an extrathoracic position (ectopia cordis).

References 1. Chaoui R, Tennstedt C, Göldner B, Bollmann R. Prenatal diagnosis of ventriculo-coronary communication in a second trimester fetus using transvaginal and transabdominal color-Doppler-sonography. Ultrasound Obstet Gynecol 1997; 9: 194–7. 2. Anderson RH, Becker AE. Anatomie des Herzens, Ein Farbatlas. Stuttgart: Georg Thieme Verlag, 1982.

3. Anderson RH, Becker AE. Pathologie des Herzens, Ein Farbatlas. Stuttgart: Georg Thieme Verlag, 1985. 4. Ho SY, Anderson RH. Conduction tissue in congenital heart surgery. World J Surg 1985; 9: 550–67. 5. Anderson RH, Becker AE, Freedom RM et al. Sequential segmental analysis of congenital heart disease. Pediatr Cardiol 1984; 5: 281–8. 6. Van Praagh R. Terminology of congenital heart disease: glossary and commentary. Circulation 1977; 56: 139–43. 7. Van Praagh R, David I, Wright GB, Van Praagh S. Large RV plus small LV is not single RV. Circulation 1980; 61: 1057–8. 8. Sharma S, Devine W, Anderson RH, Zuberbuhler JR. The determination of atrial arrangement by examination of appendage morphology in 1842 heart specimens. Br Heart J 1988; 60: 227–31. 9. Anderson RH, Becker AE, eds. The Heart. London: Gower Medical, 1992. 10. Pacifico AD, Kirklin JM, Bargeron LM, Soto B. Surgical treatment of double outlet left ventricle with an intact ventricular septum. Clinical and autopsy diagnosis and developmental implications. Circulation 1973; 48 (Suppl III): 19–23.

4 Placental implantation and development Simcha Yagel and Debra S Goldman-Wohl

Introduction The placenta, it must always be remembered, is fetal in origin. Any aberration affecting the placenta has serious consequences for the developing fetus. The size and function of the placenta have a profound effect on normal fetal growth and development, as the organ responsible for the supply of oxygen to the fetus. Therefore, placental pathology is most relevant to fetal well-being, and has far-reaching effects on healthy cardiovascular function and development. This makes the understanding of normal and pathological placental implantation and function indispensable in a textbook dealing with fetal cardiology. The fetal heart pumps blood not only to the fetal brain and other organs, but also – in similar quantity to that within the fetal venous system – to the placenta. The placenta functions as the fetal lung in gas exchange, as well as the fetal kidney in waste removal. In this chapter we in no way attempt to provide an exhaustive review of placental form and function: whole textbooks have been devoted to this topic; rather, we focus our discussion on selected aspects of placental development having particular influence on the fetal cardiovascular system, among them intrauterine growth restriction (IUGR) and preeclampsia (PE). Other chapters in this book deal with the physiology of placentation and placental function, and their effects on fetal cardiac development and function (Chapters 10, 11, 36, 44, 47, and 49), while here we focus on the cellular and molecular processes involved in placental development and implantation.

Villous development Twelve to 18 days post-conception (pc), trophoblastic trabeculae begin to proliferate and form finger-like protrusions, the primary villi, into the maternal blood surrounding them. Two days later, these primary villi are invaded by embryonic connective tissue and thus transformed into secondary villi. From days 18 to 20, the first

fetal capillaries appear. At their appearance, the tertiary villi begin to develop. The first generation of tertiary villi make up the mesenchymal villi, which are the first structures to provide surface area for maternal–fetal exchange of nutrients, gases, and wastes. Between days 20 and 42 pc, the first generations of mesenchymal villi begin vasculogenesis, i.e. new capillaries are formed from mesenchymal precursor cells. The mesenchymal villi form the only pool for subsequent villous sprouting and development of the villous trees.1–9 The mesenchymal villi can differentiate into several types of specialized villi. Increased diameter and numerous stromal channels characterize the immature intermediate villi. These eventually become stem villi through stromal fibrosis. Mature intermediate villi begin to differentiate from the mesenchymal villi from about pregnancy week 23. These differ from immature intermediate villi in that they do not mature into stem villi. Rather, they produce many terminal villi along their surfaces. The terminal villi are highly capillarized, and are efficient vessels for maternal–fetal diffusional exchange. Some mesenchymal villi and immature intermediate villi are retained in the centers of the villous trees, forming a sort of growth reserve (Figure 4.1).1,10–17 The process of branching morphogenesis begins with selection of a cell subgroup induced towards motility and invasivity. This invasiveness may be a general characteristic of branching morphogenesis or its driving force, as branches enter surrounding tissues.18 Fibroblast growth factor (FGF) signals, through their receptor tyrosine kinases (RTKs), coordinate various biological processes: angiogenesis, proliferation, differentiation, and branching morphogenesis. We examined human placental branching morphogenesis through the expression of FGF receptors (FGFR) 1–4 and FGF10 in placentas and decidua. We found that FGFR 1–4 are expressed in placenta but not in the decidua, while FGF10 is expressed by decidual cells and by the placenta, especially the extravillous trophoblasts (EVTs).19 We also showed that Spry2 is expressed through the three trimesters of pregnancy by the placental villous macrophages (Hofbauer cells)20 and that FGF10 induces

28

Fetal Cardiology

1st and 2nd Trimester

3rd Trimester

Trophoblastic sprout Villous sprout Mesenchymal villus

Hot spot Immature intermediate villus

Stem villus

Spry2 expression in Hofbauer cell culture, suggesting that mesenchymal–epithelial interaction and cross-talk among the different placenta cell types may play a role in the regulation of placental development.20 Exogenous FGF10 promotes invasion and outgrowth of villi in organ culture, and upregulates metalloprotease activity in trophoblast cell culture, as well as increasing the migration of single cell trophoblasts through MatrigelTM. Silencing of Spry2 expression by siRNA also enhanced the outgrowth of trophoblasts.21 Spry2 was shown to be a potential regulator of FGF10 activity; FGF10 is in turn a strong inducer of cell migration and collagenolytic activity. We speculate that trophoblast outgrowth and invasion at the fetal–maternal interface are in part positively regulated by FGF10 and by Sprouty 2 negatively.19,20

Invasion and implantation Trophoblasts are the first cells to differentiate in the embryo. It is these cells that adhere to the uterus and initiate the implantation process. Trophoblasts retain a stem cell population of villous cytotrophoblasts throughout pregnancy, and differentiation of trophoblasts into two major cell lineages, the syncytiotrophoblast and the invasive trophoblast, also continues to the end of pregnancy. The syncytiotrophoblast of the chorionic villi is responsible for placental nutrient and gas exchange, as well as most placental hormone and growth factor production. It is within the chorionic villi that the fetal placental arteries and veins develop.

Mature intermediate villus with terminal villi

Figure 4.1 Schematic representation of the formation and differentiation of placental villi during early and late pregnancy. Histological characteristics of the various villus types and their typical topographical relationships. Note the immature intermidiate villus (left) showing a ‘hot spot’ which subsequently (top of the left villus) develops via trophblastic and villous sprouts into a new mesenchymal villus. Hot spots correspondingly mark the sites of future villous branching (modified from reference 1).

The second, invasive, lineage derived from the cytotrophoblasts may be defined as the endovascular or interstitial lineage. Early in pregnancy, the cytotrophoblasts begin to proliferate and form cell columns. It is from these cell columns that extravillous trophoblasts emanate and from which the interstitial and endovascular invasive trophoblasts are derived. The former migrate through and invade the uterine tissue and anchor the placenta to the uterus, while the latter migrate to the maternal uterine spiral arteries. It is in the spiral arteries that the endovascular invasive trophoblasts begin the work of conversion of the spiral arteries, by displacing and replacing the endothelial cell lining of the vessels. This aids the process of forming a vessel of low resistance and high capacitance, equipped to meet the ever-increasing demand for blood flow required to maintain the growing pregnancy. This trophoblast-mediated conversion of the spiral arteries must occur by the end of the first trimester for a healthy pregnancy to continue. Failure of the trophoblasts to invade the uterus appropriately, whether through shallow trophoblast invasion, fewer invasive trophoblasts, or failure to convert the spiral arteries, is often observed in placentas of preeclampsia.17,22–25 Conversion of the maternal spiral arteries is mediated by trophoblast invasion. This is coincident with loss of the muscle and elastic tissue surrounding the arteries and replacement of the endothelial cells with trophoblasts.26 Interstitial endovascular invasion may be superficial in preeclampsia, not penetrating to one-third of the myometrial layer proximal to the decidua. Spiral artery mean diameter remains low: the persistence of unmodified narrow spiral arteries results in reduced placental perfusion.

Placental implantation and development

Some light has recently been shed on the mechanisms of cytotrophoblast induction of vessel remodeling in a transplantation model of placental villi into severe combined immunodeficiency disease (SCID) mice. Investigators showed that trophoblasts mediated both maternal endothelial cell, and vascular smooth muscle cell, apoptosis. Conversion of the spiral arteries depends on these modeling steps.27 Other molecular mechanisms necessary for trophoblast invasion include transcription factor expression and differentiation to the invasive phenotype, mechanisms of migration and invasion, digestion of the extracellular matrix (ECM), expression of angiogenic factors involved in spiral artery remodeling and, ultimately, trophoblast immune cell cross-talk and induction of chemokines and cytokines that govern trophoblast migration and angiogenesis.

Extracellular matrix degradation Trophoblast invasion and digestion of the extracellular matrix is governed by intrinsic trophoblast cell programming as well as interaction with the maternal cellular environment. To achieve successful invasion, trophoblasts must induce the repertoire of genes involved in digestion of the extracellular matrix. For example, MMP (92-kDa matrix metalloproteinase, getatinase B) is closely associated with the invasive phenotype of trophoblasts.28,29 When trophoblast invasion is shallow (for example in preeclampsia), gene expression and the activity of several molecules involved in degradation of the extracellular matrix are abnormal,30 including expression of MMP, which fails to be upregulated in preeclampsia. The enzymatic activities of urokinase plasminogen activator and plasminogen inhibitor are also altered in preeclampsia, suggesting a role for these molecules in invasion and migration of trophoblasts.31 In their pioneering work, Genbacev and colleagues32 showed that the 5–8% O2 conditions of the late first trimester fetomaternal interface favor the invasive trophoblast phenotype. On the other hand, under 2% O2 conditions, trophoblast proliferation is preferentially observed. It has also been shown that under hypoxic conditions trophoblasts retain adhesion molecule properties mimicking those of trophoblasts of preeclampsia.25,33 It has also been demonstrated34 that hypoxia-inducible factor (HIF1-α) and its downregulating response to higher oxygen tension can adjust trophoblast invasion through inhibition of transforming growth factor β3 (TGFβ3), thus enhancing the invasive phenotype. Furthermore, expression of the tumor suppressor protein von Hippel–Lindau (pVHL), which is critical for HIF1-α and HIF1-β regulation,35 was shown to be downregulated as trophoblasts enter the more invasive oxygen-enriched environment.

29

Expression of adhesion molecules by trophoblasts Adhesion molecules and integrins play an important role in cell migration. Extravillous trophoblasts express the stage-specific repertoire of adhesion molecules (aVβ4, α1β1, vascular endothelial (VE)-cadherin, vascular cell adhesion molecule 1 (VCAM-1), and platelet endothelial cell adhesion molecule 1 (PECAM-1)). Certain adhesion molecules characterizing the stem cell population of villous cytotrophoblasts inhibit invasion: these molecules are downregulated as trophoblasts enter the invasive pathway of differentiation (α6β4, αVβ6, and E-cadherin).25,33,36 Invasive trophoblasts undergo an epithelial to endothelial cell transformation and express endothelial cell-specific adhesion molecules. If the endovascular invasive trophoblasts do not undertake the function of the endothelial cells which normally line the spiral artery, inadequate spiral arteries result, leading to deficient placentation and its sequelae. Defects in the molecules responsible for trophoblast invasion and degradation of the extracellular matrix, through which trophoblasts must migrate, and failure to acquire endovascular integrin markers are indicative of a problem in a fetal pathway of trophoblast migration (Figure 4.2).37

Placental vasculogenesis and angiogenesis Angiogenesis is the formation of new vascular beds, and is critical to normal tissue growth and development.38–40 Placentation includes angiogenesis in both fetal and maternal tissues to ensure sufficiently increased uterine and umbilical blood flow41–47 to afford the most favorable surroundings for meeting the demands of the developing fetus. Later in pregnancy, angiogenesis factors influence physiological exchange;41,42,48 indeed, decreased vascular development and increased vascular resistance have been shown to be associated with early embryo loss.49,50 One of the earliest stages in embryonic development is the establishment of functional circulation.51–53 Increased transplacental exchange, which makes possible the rapid growth of the fetus in the second half of pregnancy, is supported by the extensive increase in uterine and umbilical blood flow.41,54 Major angiogenesis factors have been identified, including those involved in placental vascularization processes. Among them are vascular endothelial growth factor (VEGF), fibroblast growth factor (FGF), and the angiopoietin (ANG) protein families, and their respective receptors.39,55–58 Indeed, VEGF and FGF are involved in most of the heparinbinding angiogenic activity produced by both ovarian59 and placental60–64 tissues. VEGF s specific stimulation of

30

Fetal Cardiology

Cytotrophoblast Syncytiotrophoblast

Bilaminar embryonic disk

Blastocyst cavity

Amniotic cavity

Decidualized stroma Oxygen tension

Transcription factors

Growth factors and cytokines

Regulation of trophoblast proliferation and differentiation Regulation of placental growth Blood vessel

Endometrial glands

Regulation of Proteinases trophoblast adhesion- and inhibitors molecule expression

Regulation of cytotrophoblast invasion and vascular mimicry

Endometrial capillaries

Figure 4.2 Blastocyst Implantation. The diagram shows an invading blastocyst (about 9 to 10 days after conception) and the processes necessary for trophoblast invasion (reproduced with permission from reference 37).

vascular permeability and vascular endothelial cell protease production and migration all influence the angiogenic process.38–40,55,59 VEGF has also been shown to increase angiogenesis in both in vivo and in vitro models,65–67 and has been implicated in the primary regulation of angiogenesis in normal and pathological processes such as luteal growth, wound healing, coronary ischemia, and tumor growth.55,68 Mouse models have been used to study various aspects of placentation, including angiogenesis. During early pregnancy, VEGF mRNA is expressed in fetal placental, more than in maternal (endometrial) placental tissues, while basic FGF (bFGF) mRNA is expressed more in endometrial than in fetal placental tissues. In late pregnancy, VEGF mRNA is increasingly expressed in the placentome and

intercotyledonary fetal membranes, while bFGF is greatest in the intercotyledonary fetal membranes.69 Gene knockout studies show a central role for VEGF in fetal and placental angiogenesis. In mice, homozygous knockouts of the genes for VEGFRs led to defects in the initial formation of placental vasculature and angiogenesis, which led to fetal demise before mid-gestation.70,71 Defects included abnormal vascular formation, organization, patterning, and endothelial morphology. Homozygous knockouts of VEGF were lethal by day 11 and led to dramatic cardiovascular defects: delayed and abnormal development of the heart, aorta, major vessels, and extraembryonic vasculature, including the yolk sac and the placenta.72,73 Heterozygous knockout embryos with decreased VEGF expression showed similar defects and also died by day 11–12,72,73

Placental implantation and development

by mid-gestation.56,87 Null mutations of Tie1, Tie2, and ANG1 die later than VEGFR knockouts, pointing to a later role for Tie and ANG in vascular development. VEGFR knockouts lack primary vascular growth and organization, while ANG and Tie knockouts are affected at the stage of vascular remodeling.91 ANG does not affect endothelial cell proliferation, but increases microvascular organization and endothelial cell survival.89,92–95 In addition, ANG1 is produced by periendothelial cells, and partners VEGF in the angiogenic process. The endothelial cell-specific receptors Tiel and Tie2 and the angiopoietin ligands of Tie2 play fundamental roles in angiogenesis and in remodeling of vessel structure, but not in cell proliferation. Specifically, Ang2, an antagonist ligand for Tie2, is involved in the widening of vessels and disruption of vessel integrity.90 We have described the expression patterns of Tie2 and its antagonist ligand ANG2 at the fetal–maternal interface.91 Ang2 is expressed in the syncytiotrophoblast and Tie2 in both fetal and maternal endothelial cells (Figure 4.3).91,96 This receptor–ligand interaction and fetal–maternal cross-talk has the potential to mediate widening of both fetal and maternal vessels. The ephrin subfamily of endothelial cell ligands and receptors plays a role in endothelial cell and nerve cell targeting and migration but not proliferation.96 The downregulation of EPHB4 and upregulation of ephrin-B1 in

which may point to a threshold level of VEGF expression for normal vascular development to occur. FGFs are also potent angiogenesis factors, and have been shown to stimulate in vivo and in vitro uterine arterial and fetal placental arterial endothelial cell proliferation.38,39,74,75 FGFs are unique among the angiogenic growth factors in that they are pleiotropic, and influence angiogenesis and other developmental and differentiated functions.76 VEGF and FGF regulate placental blood flow. Upregulation of endometrial expression of VEGF and bFGF mRNA, as well as increased uterine vascularization and blood flow, were demonstrated following estrogen treatment in ovariectomized ewes and mice.47,77–79 Both VEGF and bFGF stimulate endothelial production of nitric oxide (NO), a local vasodilator, which has been shown to mediate an estrogeninduced increase in uterine blood flow.74,80–84 NO also regulates the expression of VEGF and bFGF.85,86 ANGs have also been shown to be major angiogenic factors that regulate (both increase and decrease) vascular growth and development.56,87–89 Like VEGF, ANG1 and ANG2 appear to be vascular-specific growth factors, because the receptor Tie2 is present primarily on endothelial cells.56,90 ANG2 is a natural Tie2 antagonist, leading to vascular regression and modulation of vascular growth. ANG1 is a Tie2 agonist, critical in embryonic vascular development. Lack of ANG1 leads to significant cardiovascular defects, and demise

Ephrin-B1 PIGF

VEGF-A VEGF-B VEGF-C VEGF-D

31

Ang 1

Ang 2

+

–

Ang 3

Ang 4

–

+

+

Ephrin-B2

Ephrin-A1

?

?

(a)

VEGFR-1 VEGFR-2 VEGFR-3 (KDR/Flk-1) (Flt-4) (Flt-1)

Tie1 (b)

EphB2

Tie2

EphB3

EphB4

EphA2

(c)

Figure 4.3 Schematic representation of three families of vascular growth factors and their receptor interactions. (a) Vascular endothelial growth factors (VEGFs); (b) angiopoietins; (c) ephrins. The four factors that are discussed in detail in this review are highlighted in red. In (b), ‘+’ or ‘–’ indicates whether the particular angiopoietin activates or blocks the Tie2 receptor, whereas ‘?’ indicates that a potential interaction has not yet been confirmed experimentally. In (c) only those members of the large ephrin ligand family (and only their counterpart Eph receptors) that have been implicated in vascular growth are shown (reproduced with permission from reference 96).

32

Fetal Cardiology

cytotrophoblasts is associated with endovascular migration.97 We investigated whether the ligand, ephrin-Al, is involved in trophoblast targeting as well as angiogenesis in the placenta. Using RNA in situ hybridization analysis, we found ephrin-Al expressed in cell columns of extravillous trophoblast in first trimester placenta.98 In late first trimester and in second and third trimesters, ephrin-Al is found in the extravillous trophoblast cells that have invaded the decidua. We did not observe ephrin-Al in syncytiotrophoblast nor in cytotrophoblasts. The cell-specific distribution of ephrin-Al suggests that it may play a role in migration of trophoblasts and in the vascular remodeling induced by the invading extravillous trophoblasts. EphrinAl’s role in the defective cell invasion observed in preeclampsia remains to be investigated.

The role of natural killer cells: when killers become builders Natural killer cell (NK cell) cytotoxicity is regulated by both inhibitory and activating receptors.99 Groundbreaking population-based genetic studies100 showed evidence that certain combinations of fetal and maternal genotypes, specifically human leukocyte antigen (HLA)-C2 genotypes in the fetus and specific genotypes of maternal NK KIR receptors (killer immunoglobulin-like inhibitory receptors), KIR-AA, increased the risk for preeclampsia. Conversely, the presence of the maternal KIR-B haplotype was protective for preeclampsia. This overinhibitory model necessitated an intellectual shift among researchers, that genotypes that encode greater inhibition of NK cells are associated with the development of preeclampsia, whereas previous theories of preeclampsia had focused on the need to inhibit the NK cell response as a prerequisite for allowing trophoblast invasion. These observations100 can be reconciled through the pioneering work in the mouse model.101 Research into animal models over the past decade laid the foundation for a theory of uterine NK cells’ pivotal role in angiogenesis and artery remodeling at the fetal–maternal interface.101 The role of decidual NK cell interactions with extravillous trophoblasts, growth factor secretion, cytokines, and angiogenic stimulators that have the potential to enhance spiral artery widening and trophoblast invasion102 demonstrate these observations at the human fetomaternal interface. Furthermore, we demonstrated that the genetic interactions described by Hiby et al100 cause overinhibition of NK cells, which, in turn, increases the risk for preeclampsia, and leads to decreased secretion of specific cytokines interleukin 8 and interferon γ inducible protein-10 (IL-8, IP-10) as well as angiogenic factors vascular endothelial growth factor and placental growth factor (VEGF and PLGF)102 (Figure 4.4).

This NK cell regulation of spiral artery conversion can extend to the effect that they have on trophoblast attraction and migration throughout the decidua. The molecular signals partly responsible for trophoblast attraction are being revealed through chemokine receptor expression on invading trophoblasts and chemokine ligand expression by the NK cells populating the maternal–fetal interface. Specifically, Hanna et al102 demonstrated that trophoblasts expressing the chemokine receptors CXCR1 and CXCR3 are attracted to decidual NK cells. These express their respective ligands, IL-8 and IP-10. In addition, activated decidual NK cells express higher levels of IP-10 and IL-8. It yet remains to be demonstrated that the maternal– fetal immune genetic interactions that predispose a pregnancy to preeclampsia actually result in narrow, unconverted spiral arteries, establishing a placental insufficiency ripe for the development of preeclampsia. This maternal NK cell and fetal trophoblast cross-talk would exemplify how immune interactions between the mother and fetus at the maternal–fetal interface could contribute to the development of preeclampsia by both failure of trophoblasts to invade and lack of the angiogenenic signals necessary to convert the spiral arteries leading to uteroplacental blood flow insufficiency.26

Maternal immune tolerance to the fetus Mother and fetus are distinct genetically, and maternal and fetal tissues are in intimate contact at the placenta and maternal decidua across the maternal–fetal interface. This intimate contact could be expected to generate a maternal immune rejection response, but in most cases it does not. Indeed, maternal immune rejection of the developing fetus has been implicated in pathologies of pregnancy such as recurrent miscarriage and preeclampsia.103–105 Since trophoblasts are the placental cell population most in contact with the maternal immune system, research focuses on maternal immune recognition of these cells. The mechanisms involved in placental trophoblast, and thereby fetal, avoidance of maternal immune surveillance remain inadequately understood. Trophoblasts invade maternal tissue while eluding maternal immune surveillance. Protection of trophoblasts from attack by NK cells is of critical importance, since in a normal pregnancy trophoblasts closely associate with NK cells at the implantation site, where they represent the major lymphocyte population.106 Several mechanisms work to maintain the delicate balance between immune tolerance and activation, which could lead to rejection of the embryo by the decidual lymphocytes. Both extravillous trophoblast (EVT) and decidual lymphocytes

Placental implantation and development

VEGF

PLGF

IL-8

33

IP-10

Concentration (pg/ml)

** 1000 750

LIR– KIR2DL1– KIR2DS4+ dNK clone

500 250 0 721.221

Concentration (pg/ml)

(a)

721.221 –Cw4

721.221 –Cw6

Figure 4.4

1000 750 LIR– KIR2DL1+ KIR2DS4– dNK clone

**

500 250 0 721.221

(b) Concentration (pg/ml)

721.221 –HLA-G

721.221 –HLA-G

721.221 –Cw4

721.221 –Cw6

1000 750 **

500

LIR+ KIR2DL1– KIR2DS4– dNK clone

250 0 721.221

(c)

721.221 –HLA-G

721.221 –Cw4

721.221 –Cw6

MHC class I-recognizing receptors regulate growth factor production by dNK cells. (a–c): Concentrations of IL-8, IP-10, VEGF, and PLGF in supernatants from dNK clones having the indicated phenotypes that were coincubated for 72 hours with irradiated 721.221 cells mock-transfected or transfected with the indicated MHC class I molecule. The average concentrations of cytokines secreted by a representative clone out of four to eight dNK clones from each receptor subgroup are shown. Values are mean ± SD for each sample. **P < 0.05, Student t test. One representative data set is shown out of two experiments performed (reproduced with permission from reference 102).

Fetal blood vessel Intervillous space

Syncytiotrophoblast HLA-G negative trophoblast

HLA-G positive trophoblast

NK cell Trophoblast lysis

Spiral artery

Decidua

Normal

Nitabuch’s layer

Preeclampsia

Figure 4.5 Model of HLA-G expression in the placenta. HLA-G is normally expressed only in anchoring extravillous trophoblasts, not in syncyciotrophoblast nor in cytotrophoblasts. The extravillous trophoblasts normally invade the uterus and the maternal spiral arteries, and remodel the arteries for increased blood flow. HLA-G protects these invasive cells from lysis by natural killer (NK) cells and is the laissez-passer necessary for successful trophoblast endovascular and interstitial invasion. When HLA-G expression (dark knobs on cell) is lacking, as it is in preeclampsia, trophoblasts invading the decidua are lysed by NK cells. Spiral arteries are not remodeled and remain narrow, resulting in a poorly perfused placenta (reproduced with permission from reference 130).

34

Fetal Cardiology

Normal Blood Cytotrophoblast Fetal blood vessel Syncytiotrophoblast

Intervillous space

Anchoring cell column Endovascular invasive trophoblast Interstitial invasive trophoblast Endothelial cell

Decidua

Spiral artery

Fetal pathway Defects in: trophoblast differentiation, migration, invasion (ECM degradation, HIF-α, etc.), adhesion, molecule switching, HLA-G expression, angiogenesis, NK cell–trophoblast cross-talk

Maternal pathway Thrombophilic disorders: genetic (factor V, MTHFR, etc.), acquired (aPLS); immune factors, risk factors for cardiovascular disease

Intervillous space

Shallow trophoblast invasion

Arterial thrombosis and atherosis

Unmodified spiral artery

Shallow trophoblast invasion unmodified spiral artery

Arterial thrombosis and atherosis Placental insufficiency

Endothelial cell dysfunction Maternal syndrome

Figure 4.6 Schematic representation of normal and defective fetal and maternal pathways in the development of preeclampsia. In normal pregnancy, invasive trophoblasts enter both the decidua and maternal spiral arteries. They replace the endothelial cells of maternal uterine spiral arteries. The muscle and elastic tissue surrounding the vessels is destroyed, widening the artery and increasing blood flow to the intrauterine space. There are several possibilities indicative of a defect in fetal trophoblast development. In this case, invasion of trophoblasts is insufficient and the spiral artery never converted to a widened vessel. The striations beside the spiral artery in the fetal pathway diagram depict a vessel that has maintained muscle and elastic tissue. This may be indicative of defects in trophoblast differentiation, migration, invasion (ECM degradation, HIF1-α, etc.), adhesion molecule switching, HLA-G expression, angiogenesis or cross-talk between trophoblasts and decidual NK cells. In the case of a defective maternal pathway in placental development, the spiral artery may have undergone conversion but is narrowed because of thrombosis or atherosis of the vessel. This may be caused by defects in trophoblast invasion or angiogenesis, thrombophilic disorders, genetic (factor V, MTHFR, etc.) or acquired (aPLS) immune factors, or risk factors for cardiovascular disease. Both the fetal and maternal pathways lead to narrow spiral arteries which, in turn, results in uteroplacental blood flow insufficiency and ultimately the maternal cascade of events of preeclampsia (reproduced with permission from reference 26).

are involved in these mechanisms. One of the chief activities of leukocytic and non-leukocytic cells at the maternal–fetal interface is to produce and release cytokines. The cytokine cross-talk occurring across this interface has been the subject of extensive research; cytokines help to facilitate communication between the host and guest cells. Modulation of the local cytokine profile107 is thought to control EVT invasion. Therefore, the cytokine release of decidual lymphocytes is closely

regulated. EVT also expresses two non-classical class I major histocompatibility complex (MHC) proteins, HLA-E108 and HLA-G,109 along with the classical HLA-C protein, but does not express HLA-A and HLA-B proteins.110 This unique pattern of expression of class I MHC may prevent rejection of the semi-allogeneic fetus by the mother’s immune system, as most of the cytotoxic T lymphocytes (CTLs) are directed against HLA-A and -B proteins.

Placental implantation and development

Activated T cells (Th0 cells) produce a variety of cytokines; activated CD4+ T cells (Th1 and Th2 cells) are grouped according to the cytokines they produce. Th1 cytokines, interleukin 2 (IL-2) and interferon γ, help macrophages and cytotoxic T lymphocyte-mediated responses. Th2 cells help antibody production by B cells. These two types of response may be mutually exclusive.111–113 Cytokine secretion by fetal and maternal tissues at the interface seems to be high. Receptors for a number of cytokines have been found on fetal trophoblasts.114,115 Cytokines or growth factors produced by the decidua may potentially affect trophoblast growth and invasion. The question of local suppression of Th1-type cytokines in human pregnancy remains open. Interferon γ is expressed in the placenta and found in amniotic fluid at term.116,117 IL-2, however, is absent from uteroplacental tissues. Since large granulated lymphocyte cytotoxicity against trophoblasts is induced in vitro only following activation by IL-2, its absence may be a method of large granulated lymphocyte cytolytic behavior regulation.114–121 Of the Th2 cytokines, IL-4 and IL-6 have been found in maternal and fetal tissues at the fetomaternal interface, and cytotrophoblasts and maternal lymphocytes produce IL-10. The role of cytokines at the maternal–fetal interface is not yet fully understood.122–124 There is epidemiological evidence to indicate that primipaternity, insemination with donor sperm, and pregnancy without prior cohabitation increase the risk of preeclampsia, which suggests an immune component for development of the disease.125–127 HLA-G expression is absent or reduced in preeclampsia as compared to normal placenta.30,128–130 It has been suggested that HLA-G, perhaps in conjunction with HLA-E, protects invasive trophoblasts from attack by NK cells.131 When invasive trophoblasts lacking HLA-G encounter decidual NK cells they are destroyed (Figure 4.5). The decidua is heavily infiltrated by NK cells, and attention has focused on maternal NK cell response to HLA-G.107 It is possible, if the maternal natural killer cell recognition of HLA-G is defective, that HLA-Gpositive trophoblasts will be unable to infiltrate the decidua. Reduced HLA-G expression in preeclampsia and vulnerability to attack by the maternal NK cells supports the idea that the disruption of trophoblast invasion would lead to failure of spiral artery conversion, and thereby initiate the cascade of events observed in preeclampsia.132,133 This preeclampsia model in turn exemplifies the balancing act of escaping maternal immune surveillance.130 The Fas receptor and its ligand FasL belong to the tumor necrosis factor and nerve growth factor receptor family. They are known to be involved in immune response regulation. Kauma et al134 showed that FasL is expressed on trophoblasts at the maternal–fetal interface, and that these cells are able to induce apoptosis in activated lymphocytes. These findings may also help to elucidate

35

a mechanism for maternal immune tolerance of the developing fetus.103–105,134–136 Preeclampsia may be considered as a two-step disease, with its initiating factor being narrow spiral arteries, which result in insufficient placental blood flow, leading to a hypoxic uterine environment and endothelial dysfunction, and finally resulting in the maternal syndrome. Many of the mechanisms discussed above, and in addition maternal factors such as risk for atherosclerosis, would lead to narrow spiral arteries. That is, narrow spiral arteries and poor placental perfusion may be the consequence of a fetal pathway of shallow trophoblast invasion and unconverted spiral arteries or a maternal pathway of normally converted spiral arteries that become blocked; these pathways converge in the maternal disease of preeclampsia (Figure 4.6).26

Conclusions In the last decade, tremendous progress has been made in our knowledge of the cellular and molecular mechanisms precipitating these events, which are followed by vasculogenesis and angiogenesis of fetal and maternal blood vessels at the fetal–maternal interface. However, many key questions are still open for further investigation, which will better elucidate the complex processes involved in placental implantation and development.

References 1.

2. 3.

4.

5.

6.

7.

8. 9.

Castellucci M, Kosanke G, Verdenelli F et al. Villous sprouting: fundamental mechanisms of human placental development. Hum Reprod Update 2000; 6: 485–94. Boyd JD, Hamilton WJ. The Human Placenta. Cambridge: Heffner, 1970. Enders AC, King BF. Formation and differentiation of extraembryonic mesoderm in the rhesus monkey. Am J Anat 1988; 181: 327–40. King BF. Ultrastructural differentiation of stromal and vascular components in early macaque placental villi. Am J Anat 1987; 178: 30–44. Demir R, Kaufmann P, Castellucci M et al. Fetal vasculogenesis and angiogenesis in human placental villi. Acta Anal 1989; 136: 190–203. Castellucci M, Kaufmann P. A three-dimensional study of the normal human placental villous core. II Stromal architecture. Placenta 1982; 3: 269–86. Castellucci M, Scheper A, Scheffen I et al. The development of the human placental villous tree. Anat Embryol 1990; 181: 117–28. Benirschke K, Kaufmann P. The Pathology of the Human Placenta, 4th edn. New York: Springer, 2000. Kaufmann P, Kingdom JC. Development of the vascular system in the placenta. In: Risau W, Rubanyi G, eds.

36

10.

11.

12. 13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24. 25.

Fetal Cardiology

Morphogenesis of Endothelium. Amsterdam: Harwood, 2000: 255–75. Kaufmann P, Sen DK, Schweikhart G. Classification of human placental villi. I. Histology. Cell Tissue Res 1979; 200: 409–23. Kaufmann P, Castellucci M. Development and anatomy of the placenta. In: Fox H, Wells M, eds. Haines and Taylor’s Obstetrical and Gynaecological Pathology. Edinburgh: Churchill Livingstone, 1995: 1437–76. Salvatore CA. The placenta in acute toxemia. A comparative study. Am J Obstet Gynecol 1968; 102: 347–53. Macara L, Kingdom JC, Kohnen G et al. Elaboration of stem villous vessels in growth restricted pregnancies with abnormal umbilical artery Doppler waveforms. Br J Obstet Gynaecol 1995; 102: 807–12. Macara L, Kingdom JC, Kaufmann P et al. Structural analysis of placental terminal villi from growth-restricted pregnancies with abnormal umbilical artery Doppler waveforms. Placenta 1996; 17: 37–48. Todros T, Sciarrone A, Piccoli E et al. Umbilical Doppler waveforms and placental villous angiogenesis in pregnancies complicated by fetal growth restriction. Obstet Gynecol 1999; 93: 499–503. Kingdom JC, Kaufmann P. Oxygen and placental villous development: origins of fetal hypoxia. Placenta 1997; 18: 613–26. Brosens IA, Robertson WB, Dixon HG. The role of the spiral arteries in the pathogenesis of preeclampsia. Obstet Gynecol Annu 1972; 1: 177–91. Affolter M, Bellusci S, Itoh N et al. Tube or not tube: remodeling epithelial tissues by branching morphogenesis. Dev Cell 2003; 4: 11–18. Anteby EY, Natanson-Yaron S, Greenfield C et al. Human placental Hofbauer cells express sprouty proteins: a possible modulating mechanism of villous branching. Placenta 2005; 26: 476–83. Anteby EY, Natanson-Yaron S, Hamani Y et al. Fibroblast growth factor-10 and fibroblast growth factor receptors 1–4: expression and peptide localization in human decidua and placenta. Eur J Obstet Gynecol Reprod Biol 2005; 119: 27–35. Natanson-Yaron S, Anteby EY, Greenfield C et al. FGF 10 and Sprouty 2 modulate trophoblast invasion and branching morphogenesis. Mol Hum Reprod 2007; 13: 511–19. Gerretsen G, Huisjes HJ, Elema JD et al. Morphological changes of the spiral arteries in the placental bed in relation to preeclampsia and fetal growth retardation. Br J Obstet Gynaecol 1981; 88: 876–81. Khong TY, De Wolf F, Robertson WB, Brosens I. Inadequate maternal vascular response to placentation in pregnancies complicated by preeclampsia and by small-for-gestational age infants. Br J Obstet Gynaecol 1986; 93: 1049–59. Moodley J, Ramsaroop R. Placental bed morphology in black women with eclampsia. S Afr Med J 1989; 75: 376–8. Zhou Y, Damsky CH, Fisher SJ. Preeclampsia is associated with failure of human cytotrophoblasts to mimic a vascular adhesion phenotype. One cause of defective endovascular invasion in this syndrome? J Clin Invest 1997; 9: 2152–64.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38. 39. 40. 41.

42.

Goldman-Wohl DS, Yagel S. Examination of distinct fetal and maternal molecular pathways suggests a mechanism for the development of preeclampsia. J Reprod Immunol 2007; 76: 54–60. Red-Horse K, Rivera J, Schanz A et al. Cytotrophoblast induction of arterial apoptosis and lymphangiogenesis in an in vivo model of human placentation. J Clin Invest 2006; 116: 2643–52. Librach CL, Werb Z, Fitzgerald ML et al. 92-kD type IV collagenase mediates invasion of human cytotrophoblasts. J Cell Biol 1991; 113: 437–49. Shimonovitz S, Hurwitz A, Dushnik M et al. Developmental regulation of the expression of 72 and 92 kd type IV collagenases in human trophoblasts: a possible mechanism for control of trophoblast invasion. Am J Obstet Gynecol 1994; 171: 832–8. Lim K-H, Zhou Y, Janatpour M et al. Human cytotrophoblast differentiation/invasion is abnormal in pre-eclampsia. Am J Pathol 1997; 151: 1809–18. Graham CH, McCrae KR. Altered expression of gelatinase and surface-associated plasminogen activator activity by trophoblast cells isolated from placentas of preeclamptic patients. Am J Obstet Gynecol 1996; 175: 555–62. Genbacev O, Zhou Y, Ludlow JW, Fisher SJ. Regulation of human placental development by oxygen tension. Science 1997; 277: 1669–72. Zhou Y, Fisher SJ, Janatpour M et al. Human cytotrophoblasts adopt a vascular phenotype as they differentiate. A strategy for successful endovascular invasion? J Clin Invest 1997; 99: 2139–51. Caniggia I, Mostachfi H, Winte RJ et al. Hypoxiainducible factor-1 mediates the biological effects of oxygen on human trophoblast differentiation through TGFβ(3). J Clin Invest 2000; 105: 577–87. Genbacev O, Krtolica A, Kaelin W, Fisher SJ. Human cytotrophoblast expression of the von Hippel-Lindau protein is downregulated during uterine invasion in situ and upregulated by hypoxia in vitro. Dev Biol 2001; 233: 526–36. Damsky CH, Fitzgerald ML, Fisher SJ. Distribution patterns of extracellular matrix components and adhesion receptors are intricately modulated during first trimester cytotrophoblast differentiation along the invasive pathway, in vivo. J Clin Invest 1992; 89: 210–22. Norwitz ER, Schust DJ, Fisher SJ. Implantation and the survival of early pregnancy. N Engl J Med 2001; 345: 1400–8. Folkman J, Klagsbrun M. Angiogenic factors. Science 1987; 233: 442–7. Klagsbrun M, D’Amore PA. Regulators of angiogenesis. Annu Rev Physiol 1991; 53: 217–39. Reynolds LP, Killilea SD, Redmer DA. Angiogenesis in the female reproductive system. FASEB J 1992; 6: 886–92. Reynolds LP, Redmer DA. Utero-placental vascular development and placental function. J Anim Sci 1995; 73: 1839–51. Assheton R. The morphology of the ungulate placenta, particularly the development of that organ in sheep, and notes upon the placenta of the elephant and hyrax. Phil Trans R Soc Lond Ser B 1906; 198: 143–200.

Placental implantation and development

43. 44.

45.

46.

47.

48.

49.

50.

51. 52. 53. 54.

55. 56.

57. 58.

59. 60.

61.

62.

63.

Greiss FC, Anderson SG. Uterine blood flow during early ovine pregnancy. Am J Obstet Gynecol 1970; 106: 30–8. Reynolds LP, Magness RR, Ford SP. Uterine blood flow during early pregnancy in ewes: interaction between the conceptus and the ovary bearing the corpus luteum. J Anim Sci 1984; 58: 423–9. Reynolds LP. Utero-ovarian interactions during early pregnancy: role of conceptus-induced vasodilation. J Anim Sci 1986; 62 (Suppl 2): 47–61. Reynolds LP, Redmer DA. Growth and microvascular development of the uterus during early pregnancy in ewes. Biol Reprod 1992; 47: 698–708. Magness RR. Matemal cardiovascular and other physiological responses to the endocrinology of pregnancy. In: Bazer FW, ed. The Endocrinology of Pregnancy. Totowa, NJ: Humana Press, 1998: 507–39. Bell AW, Kennaugh TM, Battaglia FC et al. Metabolic and circulatory studies of fetal lamb at midgestation. Am J Physiol 1986; 250: E538. Meegdes HLM, Ingenhoes R, Peelers LLH, Exalto N. Early pregnancy wastage: relationship between chorionic vascularization and embryonic development. Fertil Steril 1988; 49: 216–20. Bassil S, Magritte JP, Roth J et al. Uterine vascularity during stimulation and its correlation with implantation in in-vitro fertilization. Hum Reprod 1995; 10: 1497–501. Ramsey EM. The Placenta, Human and Animal. New York: Praeger, 1982. Witschi E. Development of Vertebrates. Philadelphia: WB Saunders, 1956. Patten BM. Foundations of Embryology, 2nd edn. New York: McGraw-Hill, 1964. Meschia G. Circulation to female reproductive organs. In: Shepherd JT, Abboud FM, eds. Handbook of Physiology, section 2, vol III, part 1. Bethesda, MD: American Physiologocal Society, 1983: 241–69. Ferrara N, Davis-Smyth T. The biology of vascular endothelial growth factor. Endocr Rev 1997; 18: 4–25. Breier G, Damert A, Plate KH, Risau W. Angiogenesis in embryos and ischemic diseases. Thromb Haemost 1997; 78: 678–83. Hanahan D. Signaling vascular morphogenesis and maintenance. Science 1997; 277: 48–50. Neufeld G, Cohen T, Gengrinovitch S, Poltorak Z. Vascular endothelial growth factor (VEGF) and its receptors. FASEB J 1999; 13: 9–22. Reynolds LP, Grazul-Bilska AT, Redmer DA. Angiogenesis in the corpus luteum. Endocrine 2000; 12: 1–9. Reynolds LP, Millaway DS, Kirsch JD et al. Angiogenic activity of placental tissues of cows. J Reprod Fertil 1987; 81: 233–40. Reynolds LP, Redmer DA. Secretion of angiogenic activity by placental tissues of cows at several stages of gestation. J Reprod Fertil 1988; 83: 497–502. Millaway DS, Redmer DA, Kirsch JD et al. Angiogenic activity of maternal and fetal placental tissues of ewes throughout gestation. J Reprod Fertil 1989; 86: 689–96. Zheng J, Redmer DA, Killilea SD, Reynolds LP. Characterization of heparin-binding endothelial mitogens produced by the ovine endometrium during early pregnancy. Biochem Cell Biol 1998; 76: 89–96.

64.

65.

66.

67.

68.

69. 70.

71.

72.

73.

74.

75.

76. 77.

78.

79.

80.

37

Zheng J, Magness RR, Redmer DA, Reynolds LP. Angiogenic activity of ovine placental tissues: immunoneutralization with FGF-2 and VEGF antisera. J Soc Gynecol Investig 1995; 2: 289. Leung DW, Cachianes G, Kuang WJ et al. Vascular endothelial growth factor is a secreted angiogenic mitogen. Science 1989; 246: 1306–9. Nicosia RF, Nicosia SV, Smith M. Vascular endothelial growth factor, platelet-derived growth factor, and insulinlike growth factor-1 promote rat aortic angiogenesis in vitro. Am J Pathol 1994; 145: 1023–9. Phillips GD, Stone AM, Jones BD et al. Vascular endothelial growth factor (rhVEGF165) stimulates direct angiogenesis in the rabbit cornea. In Vivo 1995; 8: 961–5. Fraser HM, Dickson SE, Lunn SF et al. Suppression of luteal angiogenesis in the primate after neutralization of vascular endothelial growth factor. Endocrinology 2000; 141: 995–1000. Reynolds LP, Redmer DA. Angiogenesis in the placenta. Biol Reprod 2001; 64: 1033–40. Fong GH, Rossant J, Gertsenstein M, Breitman ML. Role of the flt-1 receptor tyrosine kinase in regulating the assembly of vascular endothelium. Nature 1995; 376: 66–70. Shalaby F, Rossant J, Yamaguchi TP et al. Failure of blood island formation and vasculogenesis in flk-1 deficient mice. Nature 1995; 376: 62–6. Carmeliet P, Ferreira V, Breier G et al. Abnormal blood vessel development and lethality in embryos lacking a single VEGF allele. Nature 1996; 380: 435–9. Ferrara N, Carver-Moore K, Chen H et al. Heterozygous embryonic lethality induced by targeted inactivation of the VEGF gene. Nature 1996; 380: 439–42. Zheng J, Bird IM, Melsaether AN, Magness RR. Activation of the nitrogen-activated protein kinase cascade is necessary but not sufficient for basic fibroblast growth factor- and epidermal growth factor-stimulated expression of endothelial nitric oxide synthase in ovine fetoplacental artery endothelial cells. Endocrinology 1999; 140: 1399–407. Cale JM, Millican DS, Itoh H et al. Pregnancy induces an increase in the expression of glyceraldehyde-3-phosphate dehydrogenase in uterine artery endothelial cells. J Soc Gynecol Investig 1997; 4: 284–92. Gospodarowicz D. Biological activities of fibroblast growth factors. Ann NY Acad Sci 1991; 638: 1–8. Cullinan-Bove K, Koos RD. Vascular endothelial growth factor/vascular permeability factor expression in the rat uterus: rapid stimulation by estrogen correlates with estrogen-induced increases in uterine capillary permeability and growth. Endocrinology 1993; 133: 829–37. Reynolds LP, Kirsch JD, Kraft KC, Redrner DA. Timecourse of the uterine response to estradiol-17β in ovariectomized ewes: expression of angiogenic factors. Biol Reprod 1998; 59: 613–20. Reynolds LP, Kirsch JD, Kraft KC et al. Time-course of the uterine response to estradiol-17β in ovariectomized ewes: uterine growth and microvascular development. Biol Reprod 1998; 59: 606–12. Babaei S, Teichert-Kuliszewska K, Monge JC et al. Role of nitric oxide in the angiogenic response in vitro to basic fibroblast growth factor. Circ Res 1998; 82: 1007–15.

38

81.

82.

83.

84.

85.

86.

87.

88.

89.

90.

91.

92.

93.

94.

95.

96.

Fetal Cardiology

Hood JD, Meininger CJ, Ziche M, Granger HJ. VEGF upregulates ecNOS message, protein, and NO production in human endothelial cells. Am J Physiol 1998; 274: HI054–8. Kroll J, Waltengberger J. VEGF-A induces expression of eNOS and iNOS in endothelial cells via VEGF receptor-2 (KDR). Biochem Biophys Res Commun 1998; 252: 743–6. Rosenfeld CR, Cox BE, Roy T, Magness RR. Nitric oxide contributes to estrogen-induced vasodilation of the ovine uterine circulation. J Clin Invest 1996; 98: 2158–66. Vagnoni KE, Shaw CE, Phernetton TM et al. Endothelial vasodilator production by uterine and systemic arteries. III. Ovarian and estrogen effects on NO synthase. Am J Physiol 1998; 275: HI845–56. Sasaki K, Hattori T, Fujisawa T et al. Nitric oxide mediates interleukin-1-induced gene expression of matrix metalloproteinases and basic fibroblast growth factor in cultured rabbit articular chondrocytes. J Biochem 1998; 123: 431–9. Frank S, Stallmeyer B, Kampfer H et al. Differential regulation of vascular endothelial growth factor and its receptor fms-like-tyrosine kinase is mediated by nitric oxide in rat renal mesangial cells. Biochem J 1999; 338: 367–74. Suri C, Jones PF, Patan S et al. Requisite role of angiopoietin-1, a ligand for the Tie2 receptor, during embryonic angiogenesis. Cell 1996; 87: 1171–80. Lindahl P, Hellstrom M, Kalen M, Betsholtz C. Endothelial-perivascular cell signaling in vascular development: lessons from knockout mice. Curr Opin Lipidol 1998; 9: 407–11. Patan S. Tie 1 and Tie 2 receptor tyrosine kinases inversely regulate embryonic angiogenesis by the mechanism of intussusceptive micro-vascular growth. Microvasc Res 1998; 56: 1–21. Maisonpierre PC, Suri C, Jones PF et al. Angiopoietin-2, a natural antagonist for Tie2 that disrupts in vivo angiogenesis. Science 1997; 277: 55–60. Goldman-Wohl DS, Ariel I, Greenfield C et al. Tie-2 and angiopoietin-2 expression at the fetal-maternal interface: a receptor ligand model for vascular remodelling. Mol Hum Reprod 2000; 6: 81–7. Hayes AJ, Huang WQ, Mallah J et al. Angiopoietin-1 and its receptor Tie-2 participate in the regulation of capillarylike tubule formation and survival of endothelial cells. Microvasc Res 1999; 58: 224–37. Kwak HJ, So JN, Lee SJ et al. Angiopoietin-1 is an apoptosis survival factor for endothelial cells. FEBS Lett 1999; 448: 249–53. Papapetropoulos A, Garcia-Cardena G, Dengler TJ et al. Direct actions of angiopoietin-1 on human endothelium: evidence for network stabilization, cell survival, and interaction with other angiogenic growth factors. Lab Invest 1999; 79: 213–23. Thurston G, Suri C, Smith K et al. Leakage-resistant blood vessels in mice trans-genically overexpressing angiopoiectin-1. Science 1999; 286: 2511–14. Yancopoulos GD, Davis S, Gale NW et al. Vascular specific growth factors and blood vessel formation. Nature 2000; 407: 242–8.

97.

98.

99. 100.

101.

102.

103.

104.

105.

106.

107.

108.

109.

110.

111.

112.

113.

114.

Red-Horse K, Kapidzic M, Zhou Y et al. EPHB4 regulates chemokine-evoked trophoblast responses: a mechanism for incorporating the human placenta into the maternal circulation. Development 2005; 132: 4097–106. Goldman-Wohl D, Greenfield C, Haimov Kochman R et al. Eph and ephrin expression in normal placental development and preeclampsia. Placenta 2004; 25: 623–30. Hanna J, Mandelboim O. When killers become helpers. Trends Immunol 2007; 28: 201–6. Hiby SE, Walker JJ, O Shaughnessy KM et al. Combinations of maternal KIR and fetal HLA-C genes influence the risk of preeclampsia and reproductive success. J Exp Med 2004; 200: 957–65. Croy BA, Esadeg S, Chantakru S et al. Update on pathways regulating the activation of uterine natural killer cells, their interactions with decidual spiral arteries and homing of their precursors to the uterus. J Reprod Immunol 2003; 59: 175–91. Hanna J, Goldman-Wohl D, Hamani Y et al. Decidual NK cells regulate key developmental processes at the human fetal-maternal interface. Nat Med 2006; 12: 1065–74. Gill T, Wegmann T. Immune recognition, genetic regulation, and the life of the fetus. In: Gill T, Wegmann T, Nisbet-Brown E, eds. Immunoregulation and Fetal Survival. New York: Oxford University Press, 1987: 3–11. Cross J, Werb Z, Fisher S. Implantation and the placenta: key pieces of the development puzzle. Science 1994; 266: 1508–18. McIntyre JA, Faulk WP, Nichols-Johnson VR, Taylor CG. Immunologic testing and immunotherapy in recurrent spontaneous abortion. Obstet Gynecol 1986; 67: 169–75. Loke YW, King A. Immunology of human placental implantation: clinical implications of our current understanding. Mol Med Today 1998; 3: 153–9. King A, Hiby SE, Gardner L et al. Recognition of trophoblast HLA class I molecules by decidual cell receptors. Placenta 2000; 21: s81–5. King A, Allan DS, Bowen M et al. HLA-E is expressed on trophoblast and interacts with CD94/NKG2 receptors on decidual NK cells. Eur J Immunol 2000; 30: 1623–31. Rouas-Friess N, Marchal-Bras Goncalie R, Manier C et al. Direct evidence to support the role of HLA-G in protecting the fetus from maternal uterine natural killer cytolysis. Proc Natl Acad Sci USA 1997; 94: 11520–5. King A, Burrows TD, Hiby SE et al. Surface expression of HLA-C antigen by human extravillous trophoblasts. Placenta 2000; 21: 376–87. Vince GS, Johnson PM. Growth factors and cytokines at the maternal/fetal interface. Biochem Soc Trans 2000; 28: 191–5. Robertson SA, Seamark RF, Guilbert LJ, Wegmann TG. The role of cytokines in gestation. Crit Rev Immunol 1994; 14: 239–92. Wegmann TG, Lin H, Guilbert L, Mosmann TR. Bidirectional cytokine interactions in the maternal-fetal relationship: is successful pregnancy a TH2 phenomenon? Immunol Today 1993; 14: 353–6. Hampson I, McLaughlin PJ, Johnson PM. Low-affinity receptors for tumour necrosis factor-alpha, interferongamma and granulocyte-macrophage colony-stimulating

Placental implantation and development

115.

116.

117.

118.

119.

120.

121.

122.

123.

124.

125.

factor are expressed on human placental syncytiotrophoblast. Immunology 1993; 79: 485–90. Yelavarthi KK, Hunt JS. Analysis of p60 and p80 tumor necrosis factor-alpha receptor messenger RNA and protein in human placentas. Am J Pathol 1993; 143: 1131–41. Bulmer JN, Morrison L, Johnson PM, Meager A. Immunohistochemical localization of interferons in human placental tissues in normal, ectopic, and molar pregnancy. Am J Reprod Immunol 1990; 22: 109–6. Olah KS, Vince GS, Neilson JP et al. Interleukin-6, interferon-gamma, interleukin-8, and granulocytemacrophage colony stimulating factor levels in human amniotic fluid at term. J Reprod Immunol 1996; 32: 89–98. King A, Jokhi PP, Smith SK et al. Screening for cytokine mRNA in human villous and extravillous trophoblasts using the reverse-transcriptase polymerase chain reaction (RT-PCR). Cytokine 1995; 7: 364–71. Ferry BL, Sargent IL, Starkey PM, Redman CW. Cytotoxic activity against trophoblast and choriocarcinoma cells of large granular lymphocytes from human early pregnancy decidua. Cell Immunol 1991; 132: 140–9. Grabstein KH, Eisenman J, Shanebeck K et al. Cloning of a T cell growth factor that interacts with the beta chain of the interleukin-2 receptor. Science 1994; 264: 965–8. Armitage RJ, Macduff BM, Eisenman J et al. IL-15 has stimulatory activity for the induction of B cell proliferation and differentiation. J Immunol 1995; 154: 483–90. de Morales-Pinto MI, Vince GS, Flanagan BF et al. Localization of IL-4 and IL-4 receptors in the human term placenta, decidua and amniochorionic membranes. Immunology 1997; 90: 87–94. Nishino E, Matsuzaki N, Masuhiro K et al. Trophoblastderived interleukin-6 (IL-6) regulates human chorionic gonadotropin release through IL-6 receptor on human trophoblasts. J Clin Endocrinol Metab 1990; 71: 436–41. Roth I, Corry DB, Locksley RM et al. Human placental cytotrophoblasts produce the immunosuppressive cytokine interleukin 10. J Exp Med 1996; 184: 539–48. Need JA. Preeclampsia in pregnancies by different fathers: immunological studies. Br Med J 1975; 1: 548–9.

126.

127.

128.

129.

130.

131.

132.

133.

134.

135.

136.

39

Robillard PY, Hulsey TC, Perianin J et al. Association of pregnancy-induced hypertension with duration of sexual cohabitation before conception. Lancet 1994; 344: 973–5. Smith GN, Walker M, Tessier JL et al. Increased incidence of preeclampsia in women conceiving by intrauterine insemination with donor versus partner sperm for treatment of primary infertility. Am J Obstet Gynecol 1997; 177: 455–8. Colburn GT, Chiang MH, Main EK. Expression of the nonclassic histocompatibility antigen HLA-G by preeclamptic placenta. Am J Obstet Gynecol 1994; 170: 1244–50. Hara N, Fujii T, Yamashita T et al. Altered expression of human leukocyte antigen G (HLA-G) on extravillous trophoblasts in preeclampsia: immunohistological demonstration with anti-HLA-G specific antibody ‘87G’ and anti-cytokeratin antibody ‘CAM5.2’. Am J Reprod Immunol 1996; 36: 349–58. Goldman-Wohl DS, Ariel I, Greenfield C et al. Lack of HLA-G expression in extravillous trophoblasts is associated with preeclampsia. Mol Hum Reprod 2000; 6: 88–95. Navarro F, Llano M, Mellon T et al. ILT2 (LIRI) and CD94/NKG2A NK receptors respectively recognize HLA-G1 and HLA-E molecules co-expressed on taget cells. Eur J Immunol 1999; 29: 277–83. Dekker GA, Sibai BM. Etiology and pathogenesis of preeclampsia; current concepts. Am J Obstet Gynecol 1998; 179: 1359–75. Thellin O, Coumans B, Zorzi W et al. Tolerance to the foeto-placental ‘graft’: ten ways to support a child for nine months. Curr Opin Immunol 2000; 12: 731–7. Kauma SW, Huff TF, Hayes N, Nilkaeo A. Placental Fas ligand expression is a mechanism for maternal immune tolerance to the fetus. J Clin Endocrinol Metab 1999; 84: 2188–94. Marti JI, Hermann U. Immnogestosis: a new etiologic concept of ‘essential’ EPH gestosis, with special consideration of the primigravid patient. Am J Obstet Gynecol 1977; 128: 489–93. Nagata S. Fas and Fas ligand: a death factor, and its receptor. Adv Immunol 1994; 57: 129–44.

5 Placental circulations Eric Jauniaux and Graham J Burton

Introduction The placenta in mammals is the essential interface between the maternal circulation carrying O2-rich blood and nutrients and the fetal circulation. Human implantation is almost unique amongst mammals in that it is highly invasive, and the conceptus embeds itself completely within the maternal uterine endometrium and superficial myometrium. The chorionic villi, the basic structures of the early placenta, form during the 4th and 5th weeks postmenstruation,1,2 and surround the entire gestational sac until 8–9 weeks of gestation. Between the 3rd and 4th months, the villi at the implantation site become elaborately branched and form the definitive placenta, whereas the villi on the opposite pole degenerate to form the placental membranes (Figure 5.1). Toward the end of pregnancy the villi present a surface area of 12–14 m2, providing an extensive and intimate interface for maternofetal exchange.1 Ample dilatation of the uteroplacental circulation together with rapid villous angiogenesis are the key factors necessary for adequate placental development and function and subsequent fetal growth. The uterus loses its innervations during pregnancy3 and the placenta and cord are not innervated at all.4 These findings imply that the development of a low resistance to blood flow in the placental circulation is essentially the result of anatomical transformations and/or biochemically induced vasomotor mechanisms. The correlation of Doppler ultrasound findings with anatomical and physiological features suggests that the establishment of high flow–low resistance circulation in both placental circulations is primarily the consequence of the considerable increase of the diameter of the corresponding vascular bed, the length of the vascular network and the blood viscosity having a much smaller influence. In this chapter we have reviewed the basic hemodynamic concepts of the placental circulations and their relationships to maternal cardiac adaptation to pregnancy and embryonic and subsequently fetal cardiac development.

Figure 5.1 Diagram of a complete gestational sac at 8 weeks of gestation showing the presence of the trophoblastic shell and intravascular spiral artery plugs in the center of the placental bed, whereas in the periphery an intervillous circulation is established (arrows).

Normal development of the uteroplacental circulation The human uterine vasculature is made of a complex vessel network which anastomoses with branches of the ovarian and vaginal arteries to establish a vascular arcade perfusing the internal genital organs.5 The tortuous ascending uterine artery gives off approximately 8–10 arcuate branches which extend inward for about one-third of the thickness of the myometrium and envelop the anterior and posterior walls of the uterus. From this network arise

42

Fetal Cardiology

the radial arteries followed by the spiral arteries which enter the endometrium.

Physiological transformation of the uterine circulation At term, the spiral arteries which, in the non-pregnant state, transport just a few milliliters of blood per minute need to carry around 600 ml per minute.6 The basis of the adjustment of the maternal placental flow rates is the transformation of the uterine vasculature, which is associated with the peripheral widening of the supplying arteries by tissue growth and remodeling of the arterial wall. It is a gradual process which starts at implantation and which is then linked to the trophoblastic infiltration of the endometrium and superficial myometrium. Anatomical and radiographic studies including uterine perfusion experiments have demonstrated that the uterine vascular network elongates and dilates steadily throughout pregnancy.6 When the blastocyst attaches to the uterine wall, trophoblastic cells infiltrate the decidua from the proliferating tip of the anchoring villi and from the trophoblastic shell.7–10 These extravillous trophoblastic cells, which are non-proliferative, can be first found both within and around the spiral arteries in the central area of the placenta.11 They gradually extend laterally, reaching the periphery of the placenta around midgestation. Depth-wise the changes normally extend as far as the inner third of the uterine myometrium within the central region of the placental bed, but the extent of invasion is progressively shallower toward the periphery.8,12 These trophoblastic cells penetrate the myometrium via the intercellular ground substance, affecting its mechanical and electrophysiological properties.8,13 Human placentation is also characterized by a remodeling of the spiral arteries. The architecture of their decidual and myometrial parts is disrupted with the loss of myocytes from the media and the internal elastic lamina, and these essential arterial components are progressively replaced by fibrinoid material.7,9,10 Consequently these vessels lose their responsiveness to circulating vasoactive compounds. In normal pregnancies, the transformation of spiral arteries into uteroplacental arteries is described as completed around mid-gestation. However, there is a gradient in the infiltration of the trophoblast along the spiral artery, and even in normal pregnancy, not all spiral arteries are completely transformed.14 This transformation, termed ‘physiological changes’, results in the metamorphosis of small-caliber spiral vessels into flaccid distended uteroplacental arteries. Within this context the uteroplacental circulation differs from other vascular beds in that the diameter of the vessels increases, rather than decreases, as they approach their target organ. From radiologic measurements, Burchell15 observed that the diameter of the uterine artery doubled by 6.5 weeks of

pregnancy, and that by mid-pregnancy the diameter of the arcuate arteries exceeded that of the uterine arteries. Indeed, by term some of the arcuate arteries were twice the diameter of the uterine arteries, and equal to that of the internal iliac arteries. The main aim of these vascular changes is to optimize the distribution of maternal blood into a low-resistance uterine vascular network and ultimately inside the placental intervillous chamber. However, the physiological conversion may not be so important in terms of volume of intervillous blood flow, but it may play a pivotal role in affecting the quality of that flow in terms of the perfusion pressure, the pulsatility and rate of blood flow, and the consistency of the flow. In the past, it has been assumed that the principal function of the placenta is to supply the fetus with as much oxygen as possible, and to a large extent that is still the case in the second half of pregnancy when fetal weight gain is greatest. Our recent combined in vivo–in vitro investigations have resulted in a new understanding of the maternofetal relationship during the first trimester of pregnancy, and have led to the hypothesis that the placenta limits, rather than facilitates, oxygen supply to the fetus during the period of organogenesis.16,17

Development of the intervillous circulation The intervillous circulation in the hemochorial placenta has been referred to as an open system compared to other circulatory beds where the blood is retained within arteries, through capillary beds to veins.18,19 Because the spiral arteries open into essentially a large lake of blood and the intervillous space does not impose any impedance to flow, the human placenta has been considered to act as a large arteriovenous shunt. Modern anatomic and in vivo studies have shown that human placentation is in fact not truly hemochorial in early pregnancy. From the time of implantation, the extravillous trophoblast not only invades the uterine tissues but also forms a continuous shell at the level of the decidua (Figure 5.1). The cells of this shell anchor the placenta to the maternal tissue but also form plugs in the tips of the uteroplacental arteries.20–23 The shell and the plugs act like a labyrinthine interface that filters maternal blood, permitting a slow seepage of plasma but no true blood flow into the intervillous space. This is supplemented by secretions from the uterine glands, which are discharged into the intervillous space until at least 10 weeks.24 Recently, a combination of sonographic in vivo investigation, vascular casting, and oxygen measurements has shown conclusively that extensive shunting occurs within the myometrium under the placental bed.25 Whether the formation of these shunts is related to trophoblast invasion is not clear, but they are not observed in the opposite wall of the uterus.

Placental circulations

In the human, placental development is precocious, and the conceptus is fully embedded in the uterine wall before the primitive streak has formed. Consequently, other strategies need to be employed to restrict exposure of the fetus to oxygen, and as a result the human placenta is exposed to major fluctuations in O2 concentration from conception to delivery.26–28 In normal pregnancies this is a very well controlled phenomenon that has to provide a delicate balance between the metabolic needs of the fetus and its placenta and the potential danger of oxygen free radicals (OFR). During that period, the placental villi display only a few capillaries and fetal erythrocytes are nucleated,1,2,16,17 suggesting that the fetal blood is extremely viscous and

43

consequently the fetoplacental blood flow limited. Furthermore, during the first trimester the villous membrane is twice the thickness it will be in the second, and the early placenta and fetus are separated by the exocoelomic cavity (Figure 5.2), which occupies most of the space inside the gestational sac.29 These features support the concept that in normal pregnancies, the earliest stages of development take place in a low oxygen environment, reflecting to some extent the evolutionary path. Placental trophoblastic cells are extremely sensitive to oxidative stress because of their extensive cell divisions and the concomitant exposure of their DNA.28 Thus, it is well established that maternal metabolic disorders such as

(a)

(b)

(c)

(d)

Figure 5.2 Two-dimensional (2D) (a) and 3D (b) color flow mapping of the main uterine artery crossing the iliac vessels. Spectral analysis of blood velocity waveforms obtained from the main uterine artery at 14 weeks (c) and 20 weeks (d) of gestation. Note the protodiastolic notch at 14 weeks (c).

44

Fetal Cardiology

diabetes, which are associated with an increased generation of OFR, are known to be associated with a higher incidence of miscarriages, vasculopathy, and fetal structural defects,28,30 indicating that the mammalian conceptus can be irreversibly damaged by oxidative stress. The physiological hypoxia of the first trimester gestational sac may protect the developing fetus against the deleterious and teratogenic effects of OFR. Recent evidence also indicates that it is necessary to maintain stem cells in a fully pluripotent state,31 for at physiological levels free radicals regulate a wide variety of cell functions, in particular transcription factors.28 Excessive production of OFR results in oxidative stress, and there are two examples when this occurs physiologically during human pregnancy. First, at the end of the first trimester, a burst of oxidative stress is evidenced in the periphery of the early placenta.26 The underlying uteroplacental circulation in this area is never plugged by the trophoblastic shell, allowing limited maternal blood flow to enter the placenta from 8–9 weeks of gestation (Figure 5.1). This leads to higher local oxygen concentrations at a stage of pregnancy when the trophoblast possesses low concentrations and activities of the main antioxidant enzymes superoxide dismutase, catalase, and glutathione peroxidase. Focal trophoblastic oxidative damage and progressive villous degeneration trigger the formation of the fetal membranes,32 which is an essential developmental step enabling vaginal delivery. The oxidative stress may also stimulate the synthesis of various trophoblastic proteins. The second example involves an ischemia–reperfusion (I/R) phenomenon. Angiographic studies of the uterine vasculature of the rhesus monkey have demonstrated that during normal pregnancy flow from spiral arteries into the intervillous space is often intermittent, arising from spontaneous vasosonstriction.33 Although equivalent studies have not been performed in the human, the general similarity of the uteroplacental vasculature and delivery of blood into the intervillous space to those of the rhesus monkey has led to the assumption that intermittent perfusion of the intervillous space also occurs in our species. Placental inflow may also be compromised by external compression of the arteries during uterine contractions in both the rhesus and human, and even by postural changes.33 Some degree of I/R stimulus may therefore be a feature of normal human pregnancy, especially toward term when the fetus and placenta are extracting large quantities of O2 from the intervillous space.34 This chronic stimulus could lead to upregulation of the anti-OFR defense in the placenta, thus reducing oxidant stress. As in early pregnancy, this well-controlled oxidative stress may play a role in continuous placental remodeling and essential placental function such as transport and hormonal synthesis. At the end of the first trimester the trophoblastic plugs are progressively dislocated, allowing maternal blood to flow progressively freely and continuously within the intervillous space. During the transitional phase of

10–14 weeks’ gestation, two-thirds of the primitive placenta disappears, the exocoelomic cavity is obliterated by the growth of the amniotic sac, and maternal blood flows progressively throughout the entire placenta.32 These events bring the maternal blood closer to the fetal tissues, facilitating nutrient and gaseous exchange between the maternal and fetal circulations.

Maternal hemodynamic changes and regulation of the uteroplacental circulation The maternal plasma volume increases gradually from the first month of gestation to a plateau in the third trimester, about 40% above the non-pregnant level.35 The mechanisms regulating these changes in human pregnancy are still unknown, and the gradual pattern of plasma volume expansion is out of phase with the rapid pattern of increase in maternal cardiac output and drop in arterial pressure. Most of the 30–60% increase in cardiac output and half of the 10% fall in arterial pressure is accomplished in the first trimester,36 and cardiac output continues to increase even in late pregnancy.37 It has been suggested that the fall in peripheral systemic vascular tone is the main factor triggering the rise in cardiac output in early pregnancy.38 The resulting rapid fall in preload and afterload leads to a compensatory increase in heart rate and activation of the volume restoring mechanisms. It is only from the end of the first trimester onward that this increased cardiac output is due to an increase in stroke volume. Blood viscosity is directly related to the hematocrit. Maternal total red cell volume increases steadily throughout gestation.35 The increment varies from 18 to 30% of non-pregnant values, and its extent is considerably influenced by oral iron intake. However, at the same time plasma volume increases by about 40%, resulting in hemodilution and, therefore, in a decrease of maternal blood hematocrit and blood viscosity. A lower maternal blood viscosity potentiates the fall in peripheral vascular resistance. A variety of endocrine substances are known to influence the uterine artery resistance. Their role is emphasized by the fact that in the human, the decrease in uterine vascular resistance starts during the luteal phase of the menstrual cycle.39 Furthermore, in some animal species, such as horses, pigs, and cows, maternal tissue is not infiltrated by trophoblastic cells and the conceptus is adequately perfused by maternal blood. It is well established that estrogen produces important changes in uterine blood flow in both human and various animal species. For example, direct intra-arterial uterine infusion of estradiol in ewes induces a dramatic increase of the uterine blood flow in both the pregnant and non-pregnant uterine vascular bed, while progesterone partially inhibits the vascular effect of estradiol.40 A similar response is observed in postmenopausal

Placental circulations

women receiving hormone replacement therapy.41 At physiological levels estradiol decreases the resistance to flow in the uterine circulation, and this is partially reversed by progestogen. Maternal serum 17β-estradiol levels may also have a significant influence on uterine resistance to blood flow during pregnancy.42 In addition, independent of gestational age, maternal serum levels of relaxin may also be a significant contributor to uterine resistance to blood flow as assessed by Doppler indices.43 Another study, published around the same time, showed negative correlation of estradiol and a positive correlation of progesterone with uterine artery impedance but no correlation with spiral artery flow characteristics.44 Discrepancy between different studies suggests that the relationship of sex steroids to uterine Doppler velocimetry during pregnancy is more complex than previously thought, and requires further investigation. A close temporal relationship has been found between the Doppler detection of a continuous intervillous flow and the human chorionic gonadotropin (hCG) peak.45 Although there is no obvious role for the hCG molecule in the regulation of uteroplacental hemodynamics, this finding suggests that hCG secretion may be influenced by changes in intraplacental hemodynamics and possibly by changes in placental oxygen tension.

Ultrasound/Doppler features of the uteroplacental circulation The various branches of the uterine circulation can be differentiated by means of color Doppler imaging, and the overall Doppler mapping features correlate well with the classical and modern anatomic findings.46 However, the uteroplacental circulation is a dynamic model in which the magnitude of blood flow through a single vessel may vary importantly (Figures 5.2 and 5.3). Thus, the evaluation of blood flow in single uteroplacental vessels is often difficult to interpret, and of limited value in understanding the pathophysiology of placental-related pregnancy disorders. In non-pregnant women and during the first half of normal pregnancy, flow velocity waveforms (FVWs) from the main uterine arteries are characterized by a well-defined protodiastolic ‘notch’ (Figure 5.2). End-diastolic flow (EDFs) increase in the main uterine arteries and their branches during the second half of the menstrual cycle, and this increase continues as pregnancy advances. In 85% of pregnancies, the protodiastolic notch disappears before 20 weeks of gestation,47 and may reflect the end of the implantation process and its associated physiological changes. Blood flows in the spiral (Figure 5.3) arteries are characterized in pregnancy by a low-impedance irregular flow pattern which shows no significant changes in shape throughout pregnancy.46 Doppler studies have demonstrated a progressive decrease in the downstream resistance to blood flow in the uterine circulation from implantation to term.48–54 This decrease

45

can be observed in all segments of the uterine circulation. Impedance to blood flow through spiral arteries in the second trimester is lower in the central area of the placental bed than in peripheral areas.51 These Doppler data are in agreement with histologic data.6,8 The resistance index (RI) or the pulsatility index (PI) measured from the FVWs recorded at the level of the main uterine arteries reflects the downstream flow impedance in the whole uterine circulation.2 Both left and right main arteries must be investigated at the same time, as unilateral measurement may provide erroneous results concerning true uterine perfusion. Between 12 and 14 weeks of gestation, there is a rapid increase from 50 to 120 cm/s in the mean peak systolic velocity (PSV) of the main uterine artery. A decrease of the resistance and pulsatility indices from the main uterine artery towards the spiral arteries can also be demonstrated at different stages of pregnancy.49 When comparing Doppler features of the placental circulation at different gestational ages, we found that a nonpulsatile signal corresponding to maternal intraplacental blood flow could not be identified inside the intervillous space before 10 weeks of gestation.2 This finding has been confirmed by other authors,52,55 provoking a vigorous debate involving both ultrasonographers53,56 and anatomists57,58 about the status of the maternal circulation during the first trimester.59,60 In view of the potential danger from oxidative stress, the debate has become obsolete, and it is likely that onset of maternal blood flow to the placenta is normally a progressive phenomenon, with communication between the uteroplacental arteries and the intervillous space being established in a small number of vessels at a time from the end of the second month of pregnancy onward. Our recent data support the concept that in normal pregnancies the intervillous circulation is gradually established between the beginning of the third month and the end of the fourth month of gestation.32,61 They also demonstrate that, in normal pregnancies, onset of maternal blood flow is most often initiated in the peripheral regions of the placenta, and this may be related to regional differences in the extent of plugging of the maternal arteries. The evidence available suggests, therefore, that the normal establishment of a continuous intervillous circulation is an incremental phenomenon, starting in the periphery and expanding progressively to the rest of the placenta thereafter. This concept is supported by the immunohistochemical and morphological evidence of temporospatial differences in the degree of trophoblastic oxidative stress.32

Abnormal development of the uteroplacental circulation Placental-related disorders of pregnancy are almost unique to the human species. These disorders, which affect around

46

Fetal Cardiology

(a)

(b)

(c)

(d)

Figure 5.3 2D (a) and 3D (b) color flow mapping of the placental bed. Spectral analysis of blood velocity waveforms obtained from spiral artery at 10 weeks of gestation (c) and from the intervillous space at 14 weeks of gestation (d). Note the discontinuous blood flow with a venous pattern (i.e. intervillous flow) in (d).

a third of human pregnancies, primarily include miscarriage and preeclampsia. In other mammalian species, the incidence of both disorders is extremely low. Although epidemiological data on animals living in the wild, such as monkeys, are limited, laboratory rodents are known to have postimplantation pregnancy loss rates of less than 10%,62 and spontaneous preeclampsia has only been exceptionally described in ‘patas’ monkeys63 and guinea pigs.64 There is mounting evidence that oxidative stress or an imbalance in the oxidant/antioxidant activity in uteroplacental tissues plays a pivotal role in the development of placental-related diseases. In about two-thirds of early pregnancy failures there is anatomical evidence of defective

placentation, which is mainly characterized by a thinner and fragmented trophoblast shell, reduced cytotrophoblast invasion of the endometrium, and incomplete plugging of the lumen at the tips of the spiral arteries.12,20,21,32 This is associated with the absence of physiological changes in most spiral arteries, and leads to a premature onset of the maternal circulation throughout the entire placenta (Figure 5.4). Independent of the etiology of the miscarriage, the excessive entry of maternal blood into the intervillous space has two effects: a direct mechanical effect on the villous tissue which becomes progressively enmeshed inside large intervillous blood thrombi, and indirect O2mediated widespread trophoblastic oxidative damage and increased apoptosis.12,32,65 Overall, the consequences are

Placental circulations

(a)

47

endothelial cell dysfunction. The nature of the inflammatory response and oxidative stress seen in preeclampsia is not intrinsically different from that seen in normal pregnancies. The difference is one of degree, which strongly suggests that preeclampsia is at the extreme end of a continuum of changes seen in every pregnancy.

In vivo features of abnormal placentation (b)

(c)

Figure 5.4 Diagrams showing placentation in a normal ongoing pregnancy (a), in an early pregnancy failure (b), and in a complete hydatidiform mole (c). Note in (a) the continuous trophoblastic shell, the plugs in the lumens of the spiral arteries, and the interstitial migration of the extravillous trophoblast through the decidua down to the superficial layer of the myometrium; in (b) the discontinuous trophoblastic shell, the absence of plugs, and the reduced migration of extravillous trophoblastic cells; in (c) the absence of trophoblastic plugs and interstitial migration.

placental degeneration with complete loss of syncytiotrophoblast function and detachment of the placenta from the uterine wall. This mechanism is common to all miscarriages, the time at which it occurs in the first trimester depending on the etiology.17 Preeclampsia stems from a similar, though lesser, defect in early trophoblast invasion, for while invasion is sufficient to anchor the conceptus it is insufficient to fully convert the spiral arteries into low-resistance channels.10 Incomplete conversion of the spiral arteries results in retention of smooth muscle cells within their walls, and so some vasoreactivity persists in 30–50% of the placental vascular bed. This may lead not only to diminished perfusion of the intervillous space, but more importantly to intermittent perfusion. Since the placenta and fetus continually extract oxygen, it is expected that transient hypoxia will result,34 and that consequently the placenta suffers a chronic low-grade ischaemia–reperfusion type injury. These findings have led us to suggest that preeclampsia is a threestage disorder, with the primary pathology being an excessive or atypical maternal immune response.17 This would impair the placentation process, leading to chronic oxidative stress in the placenta and finally to diffuse maternal

The ability of transvaginal color Doppler to detect flow velocity waveforms from small vessels such as from the terminal part of the uteroplacental circulation has given rise to much enthusiasm from clinicians interested in predicting early and late pregnancy complications related to an abnormal placentation. Failure of conversion by itself should not influence the volume of maternal blood to the placenta, as this is mainly determined by changes in the radial and arcuate arteries. Rather, maternal blood flow will enter the intervillous space at greater velocity and pressure than normal, a phenomenon that we described as the ‘hose effect’.66–68 On ultrasound the inflow appears as jet-like streams surrounded by turbulence, and the force is sufficient to drive apart the villous branches and form intervillous lakes, also called maternal lakes, in a manner similar to the formation of the central cavity of a lobule at the start of the second trimester. In the most severe cases, there is evidence of an abnormal intervillous circulation from the beginning of the second trimester, which is characterized by major placental anatomical changes that have been described on ultrasound as ‘jelly like’.68 These features refer to the overall appearance of the placenta, with its chorionic plate being pushed up by jet-like bloodstreams, with an overall reduction in the placental mass echogenicity. These ultrasound anomalies are commonly found in association with raised maternal serum α-fetoprotein levels and with early fetal growth restriction (FGR), suggesting damage to the villous membranes and in particular to the trophoblastic layer.68 Several Doppler screening studies, both in the second and more recently in the first trimester of pregnancy, have demonstrated an association between increased impedance to flow in the uterine arteries and subsequent development of preeclampsia, FGR, and perinatal death.69 Most studies have focused on the fact that an increased impedance to flow in the uteroplacental circulation and persistent uterine artery notching at mid-gestation predict the development of placental-related pregnancy disorders. Initial screening by continuous-wave Doppler was carried out at 18–20 weeks of gestation, and the examination was repeated in those with increased impedance to flow at around 24 weeks. The differences in Doppler technique employed in the different studies might partly explain their discrepant results. Other factors such as the definition of abnormal flow, the populations, the gestational age

48

Fetal Cardiology

at which women were examined, and the criteria for the diagnosis of preeclampsia and FGR may also have contributed to the wide variation in detection rates. Recent data suggest that increased impedance at 23–24 weeks identifies about 40% of those that subsequently develop preeclampsia and about 20% of those that develop FGR.69 Overall, abnormal Doppler is better at predicting severe disease, and the sensitivity of abnormal Doppler features for preeclampsia requiring delivery prior to 34 weeks of gestation is about 80% and for FGR it is about 60%. There are only a few first trimester screening studies, and most of these have shown that, even at an early stage, impedance to flow in the uterine arteries can be increased in pregnancies that subsequently develop preeclampsia and FGR.69 In women with increased impedance to flow, the pooled likelihood ratio (LR) for the development of preeclampsia was about 5, and for those with normal Doppler results the LR was about 0.5. The pooled LRs for FGR were about 2 and 0.9, respectively.69 In complicated early pregnancies, the placenta appears on color flow mapping to be hypervascularized well before the end of the first trimester.70 Unlike the findings on the development of the intervillous circulation in normal pregnancies, this observation has not been disputed, and the differences between the Doppler signals and the histopathological findings in early pregnancy failure compared to normal pregnancies are so striking that they constitute one of the strongest arguments in favor of the Hustin and Schaaps hypothesis (Figure 5.5). In a study correlating ultrasonographic with pathological features, we found extensive dislocation of the trophoblastic shell and a massive infiltration of the intervillous space by maternal blood in cases presenting with a continuous intervillous blood flow on color mapping before 12 weeks of gestation.12,70 We have recently shown that at 8–9 weeks and 10–11 weeks an intervillous flow is significantly more commonly detected inside the placenta of abnormal pregnancies, in particular in the central area.32,61,71 These findings suggest that in early pregnancy failure the initial central trophoblastic migration and vascular plugging is insufficient, allowing the entry of larger than normal quantities of maternal blood into the placenta. As a consequence, oxidative damage to the trophoblast is significantly increased, and this will prevent the normal villous trees from developing and hence compromise the anchoring of the placenta. The premature entry of maternal blood into the intervillous space at this stage of pregnancy will disrupt the placental shell, and is probably the mechanical cause of abortion. This mechanism is common to all miscarriages, the time at which it occurs in the first trimester depending on the etiology.12,17,32,61,71 The predictive value of Doppler measurements in early pregnancy is limited. All Doppler studies in the first trimester have failed to demonstrate abnormal blood flow indices in the uteroplacental circulation of pregnancies that subsequently ended in spontaneous abortion or in anembryonic gestation. However, the premature entry of a larger than physiological quantity of

maternal blood inside the placenta could be associated with a significant increase in second trimester complications such as preeclampsia.50

Normal development of the umbilicoplacental circulation The development of the fetal vasculature begins during the 3rd week post-conception (5th week of pregnancy) with the de novo formation of hemangioblastic cell cords within the villous stromal core. By the beginning of the 4th week the cords have developed lumens and the endothelial cells become flattened. Surrounding mesenchymal cells become closely apposed to the tubes, and differentiate to form pericytes.72,73 During the next few days, connections form between neighboring tubes to create a plexus, and this ultimately unites with the allantoic vessels developing in the connecting stalk to establish the fetal circulation to the placenta. Around 28 days post-ovulation (6 completed menstrual weeks), the villous vasculature is connected with the primitive heart and the vascular plexus of the yolk sac via the vessels of the connecting stalk.1 Around the end of the 5th week of gestation, the primitive heart begins to beat, and this pivotal phenomenon has been documented in utero as early as 36 days’ menstrual age.74 From 6 to 9 weeks, there is a rapid increase of the mean heart rate to a plateau in the second and third trimesters. The fetoplacental circulation is established from around 8 weeks of gestation.66 The development of the fetoplacental circulation is characterized by a progressive rise in blood flow and decrease in vascular resistance to flow. Because of the absence of innervation beyond the proximal 1–2 cm of the umbilical cord, the umbilicoplacental circulation is considered to be a passive circulation in which the flow rate is determined by the mean effective perfusion pressure.75

Development of the villous circulation Early in pregnancy the capillary network is labile and undergoes considerable remodeling. Angiogenesis continues until term through a series of different phases which most probably reflect different concentrations and combinations of growth factors induced by the changing intrauterine environment.76,77 From 25 weeks onwards, terminal capillary loops are produced; indeed it is thought that the differential elongation of the capillary network to that of the containing villus causes vascular loops to obtrude from the surface, so creating a new terminal villus.78 The caliber of the fetal capillaries is not constant within intermediate and terminal villi, and frequently on the apex of a tight bend the capillaries become greatly

Placental circulations

(a)

(b)

(c)

(d)

49

Figure 5.5 (a) Color flow mapping of a gestational sac containing a 5-mm embryo with a heart activity and showing the placental bed circulation. (b) Histological view of a decidual biopsy collected from the area under the definitive placenta showing trophoblastic plugs inside the lumen of the tips of the spiral arteries at 9 weeks of gestation (arrows). (c) Color flow mapping of a gestational sac in a missed miscarriage at 9 weeks (last menstrual period) showing diffuse intervillous signals (arrow). (d) Histological view of a decidual biopsy in a similar case of missed miscarriage at 9 weeks showing the poor transformation of the tip of the spiral arteries and the absence of trophoblastic plugs (arrows).

dilated, forming sinusoids. These regions may help to reduce vascular resistance and so facilitate distribution of fetal blood flow through the villous trees.78,79 Placental capillary formation is only completed by mid-gestation. De novo formation of capillaries from the transformation of mesenchymal cells is rare in mature placental tissue, and only occurs in persisting mesenchymal villi.80 The villous circulation of the definitive placenta is composed of muscularized stem arteries (750 μm), branching more than 10 times and ending in long capillary

loops, 15–20 μm in diameter, and perivascular capillary networks.

Hemodynamics of the umbilicoplacental circulation During the second half of pregnancy, the vascular network with the lowest resistance in the entire fetal circulatory

50

Fetal Cardiology

system is at the level of the placental vascular bed.81 According to Poiseuille’s law, a decrease in blood viscosity and/or in the vascular system length and an increase in the mean radius of the vascular system collectively decrease the resistance to flow.82 The latter effect is particularly strong, the resistance depending on the radius to the fourth power. The number of fetal capillaries per villous profile and the proportion of the villi (volume fraction) occupied by the fetal capillaries remains low (1–2/villi) during the first 2 months of pregnancy.83 Furthermore, up to the middle of the second month of gestation, all embryonic erythrocytes are nucleated,1,2 indicating that the fetal blood viscosity is high in early pregnancy. Subsequently, the number of nucleated erythrocytes decreases rapidly, and by 12 weeks of gestation, these cells account for less than 10% of the total red cell population in the fetal circulation. Thus, as the mean radius of the villous vascular system is low and the viscosity high, the resistance to flow in the early umbilicoplacental circulation must be high.16,17 This suggests that for at least the first 2 months of gestation, the extraembryonic circulation is mainly vitelline (as opposed to placental). The mechanism by which a reduction in vascular impedance through the fetoplacental circulation occurs during the second trimester is not known. Since umbilical vessels are not innervated, vasodilatation must occur through one of two mechanisms: through a direct vasorelaxant paracrine effect of an agent acting upon vascular smooth muscle, or through angiogenesis. From an anatomical perspective, the villous capillary bed has the greatest potential to influence total umbilicoplacental vascular impedance. However, the formation of villous capillaries is progressive, and does not appear to be directly related to the rapid drop in umbilical artery impedance observed between 12 and 14 weeks’ gestation.2 Several humoral factors are known to modulate the tone of the cord and villous vessels of the term placenta. Vasoconstrictor substances include thromboxane A2 produced by the platelets and the umbilical vessels, angiotensin II produced by the fetal kidney, and endothelin-1 synthesized locally within the placenta.84 Possible vasodilators produced by the fetus are the atrial natriuretic peptide and prostacyclin.85 Maternal serum hCG and relaxin levels have an independent correlation with the fetal heart rate,86 but the influence of other maternal or placental factors on the hemodynamics of the developing umbilicoplacental circulation remains uncertain. NO and cyclic guanosine monophosphate (cGMP) concentrations are positively correlated with umbilical artery impedance between 9 and 15 weeks’ gestation.87 These data suggest that NO and cGMP may play an important part in maintaining flow through the early first trimester fetoplacental circulation. As gestation advances, the reduction in impedance occurs despite a decrease in NO production, indicating that another vascular-endothelial mechanism, or new villous vessel formation, is responsible for reducing umbilical artery resistance at this critical stage in the development of

the fetoplacental circulation. The passage of fetal blood through an anatomically high-resistance circuit could lead to endothelial stimulation of nitric oxide synthase (NOS) activity, thus maintaining vasodilatation within the umbilicoplacental circulation until anatomical changes occur. Shear stresses may be highest in the first trimester umbilicoplacental circulation, prior to the development of new villous blood vessels and the appearance of a low-impedance, positive end-diastolic flow umbilical artery velocity waveform. This changing pattern of blood flow after 12 weeks could lead to reduced endothelial stimulation, and consequently to a rapid reduction of both NOS and GMP levels from the villous trophoblast.87

In vivo features of the umbilicoplacental circulation In the first 3 months of gestation, umbilical arteries show a high degree of vascular resistance to blood flow expressed by narrow systolic waveforms, absence of EDF, and high PI values (Figure 5.6). PI values from the umbilical artery remain high, which suggests minor changes in umbilical placental vascular resistance until the end of the first trimester.2,88 Between 12 and 14 weeks, the end-diastolic flow (EDF) develops rapidly but is incomplete and/or inconsistently present. Diastolic frequencies throughout the entire cardiac cycle are recorded in the umbilical artery of the normally developing fetus from 14 weeks onward (Figure 5.6). After that period, the trend in the RI or PI shows a gradual decrease until the end of the third trimester. End-diastolic velocities are present at the level of intracerebral arteries 2 weeks earlier than in the umbilical artery.64 In early pregnancy, there is no relationship between umbilical artery Doppler impedance indices, fetal heart rate,89 or villous angiogenesis.2 There is also no correlation between umbilical artery resistance to flow and fetal blood hematocrit or umbilical cord length later in pregnancy.90 In early pregnancy, the rapid appearance of the EDF in the umbilical circulation and the drop in umbilical artery PI value coincide with the time of the establishment of the intervillous circulation. Changes in the pressure gradient due to the expansion of the intervillous space and/or a modification of the local concentration of vasodilators could also influence relaxation of the small placental arteries. The addition of color flow mapping2,91 to Doppler equipment, three-dimensional (3D) power Doppler, multigate spectral velocimetry, and more recently spectral Doppler index mapping92–95 have facilitated the visualization of smaller intraplacental vessels, and clearly visualize umbilical cord coiling (Figure 5.7). As for the uteroplacental circulation, Doppler signals can be obtained from the different segments of the umbilicoplacental circulation (Figure 5.7), and spectral analysis also shows a decline in resistance indices with advancing gestation and toward the

Placental circulations

(a)

(b)

(c)

(d)

51

Figure 5.6 Color Doppler mapping and spectral analysis of the umbilical cord blood flow at 7 weeks (a), 11 weeks (b), 14 weeks (c), and 16 weeks (d). Note the progressive appearance of the end-diastolic flow.

intraplacental arterioles.49 However, Doppler investigation of the branches of the umbilicoplacental circulation is limited to chorionic and main stem villous (approximately 750 μm in internal diameter) vessels, and the villous capillary bed cannot be directly explored with these techniques.

Abnormal villous vascular development Primary anomalies of villous angiogenesis are rare, and include mainly chorioangioma and molar villi. Secondary anomalies are more common, and are found within the

context of FGR associated with congenital infection or chromosomal abnormalities and disorders of uteroplacental circulation leading to chronic fetal hypoxia. It has been proposed that the placenta is hyperoxic, rather than hypoxic as is commonly assumed, in cases of severe intrauterine growth restriction.96 This theory may explain the basis for many of the morphological changes observed, but does not account for how the hyperoxia is initiated. From analysis of the umbilical waveform it is possible to assess the impedance to placental blood flow, and to accurately predict fetal hypoxia.97–99 Various attempts have been made to correlate the Doppler abnormalities with placental structural changes in order to provide a mechanistic explanation for their origin. The results have been

52

Fetal Cardiology

(a)

(b)

(c)

(d)

Figure 5.7 Color Doppler mapping of the umbilical cord placental insertion at 12 weeks (a) and 20 weeks of gestation (b). Note the entry of one of the umbilical arteries inside the placenta (a) and the branching of the chorionic vessels on the surface of the placenta (b). 2D and 3D color Doppler mapping of a free loop of umbilical cord at 16 weeks (c) and 32 weeks (d). Note the difference in coiling of the cord between the second and the third trimester views.

varied, ranging from claims of a reduction in the number of arteries within the supporting stem villi to a reduction in the capillary vascular bed within the terminal villi, the principal site of gaseous exchange. The underlying cause of the placental lesions is not known, although the fact that Doppler changes in the umbilical circulation are invariably seen subsequent to similar changes in the uterine arteries strongly suggests that they are a secondary phenomenon. The absence and a reverse of EDF in the umbilical arteries are common findings in pregnancies complicated by severe

fetal growth restriction. Histomorphometric features are consistent with previous findings that increasing severity of abnormal Doppler waveforms in the uterine and umbilical circulations is associated with fetal distress and hypoxia.100 Abnormal intraplacental blood flow at 28–34 weeks of gestation is also strongly associated with FGR.95 As Doppler abnormalities of the umbilical circulation are rarely seen in the absence of uterine arterial abnormalities, it is most likely that they are a secondary phenomenon. Deficient trophoblast invasion during early pregnancy

Placental circulations

leads to incomplete conversion of the spiral arteries. These vessels remain of higher resistance than normal, and the retention of smooth muscle within the spiral arteries exacerbates their normal contractility, resulting in longer periods of vasoconstriction and hence greater fluctuations in oxygen tension. This in turn promotes a mild ischemia– reperfusion injury in the placental tissues,34 leading to oxidative stress in the fetal vasculature. Oxidative stress is a powerful inducer of endothelial cell apoptosis, and repeated insults during mid-pregnancy may lead to regression of the capillaries, particularly as a high percentage are not stabilized by a pericyte covering.100 Such regression would increase vascular impedance in a reverse of the pattern seen during normal pregnancy, and so account for the changes in umbilical waveform observed. The intermediate and terminal villi are the principal sites of gaseous exchange, and decreased vascularization will inevitably impair placental exchange. This will lead to fetal hypoxia and growth retardation, but also reduced oxygen extraction from the intervillous space and so hyperoxia on the venous side of the placenta as a tertiary event.100 There have been several reports of a fatal outcome of chromosomally normal fetuses between 11 and 18 weeks of gestation following the discovery of absent or reversed EDF in the umbilical artery earlier in pregnancy.67,101,102 The umbilical artery PI is also abnormally high in some fetuses investigated with trisomy 18 and triploidy.102 In these cases, the EDF at 13 weeks of gestation was either incomplete or absent, suggesting an abnormal development of the villous circulation. These findings indicate that an abnormal EDF in early pregnancy can also be an ominous sign of adverse fetal outcome before mid-gestation as it is during the second half of pregnancy. Blood flow resistance indices are lower in the superficial and deep placenta compared with the cord insertion area.49,66 Absent or opposite gradient between the umbilical artery and the placental vessels was associated with adverse pregnancy outcome.103 Elevated intraplacental resistance to blood flow indicates increased maternal intraplacental impedance as early as week 8 of gestation.104 Overall, there is no doubt that the Doppler ultrasound study of umbilical artery waveforms helps to identify the compromised fetus in ‘high risk’ pregnancies, appears to improve a number of obstetric care outcomes, and appears promising in helping to reduce perinatal deaths.105 Conversely, routine Doppler ultrasound in low risk or unselected populations does not confer benefit on mother or baby.106

References 1. 2.

Boyd JD, Hamilton WJ. The Human Placenta. Cambridge: Heffer and Sons, 1970. Jauniaux E, Jurkovic D, Campbell S. In vivo investigations of anatomy and physiology of early human placental circulations. Ultrasound Obstet Gynecol 1991; 1: 435–45.

3.

4.

5.

6. 7.

8.

9.

10.

11.

12.

13. 14.

15. 16.

17.

18.

19.

20.

21.

53

Morizaki N, Morizaki J, Hayashi RH, Garfield RE. A functional and structural study of the innervation of the human uterus. Am J Obstet Gynecol 1989; 160: 218–28. Reilly FD, Russell PT. Neurohistochemical evidence supporting an absence of adrenergic and cholinergic innervation in the human placenta and umbilical cord. Anat Rec 1979; 188: 277–86. Itskovitz J, Lindenbaum ES, Brandes JM. Arterial anastomosis in the pregnant human uterus. Obstet Gynecol 1980; 55: 67–70. Ramsey EM, Donner NW. Placental Vasculature and Circulation. Stuttgart: Georg Thieme, 1980. Pijnenborg R, Dixon G, Robertson WB, Brosens I. Trophoblastic invasion of human decidua from 8 to 18 weeks of pregnancy. Placenta 1980; 1: 3–19. Pijnenborg R, Bland JM, Robertson WB, Dixon G, Brosens I. The pattern of interstitial invasion of the myometrium in early human pregnancy. Placenta 1981; 2: 303–16. Pijnenborg R, Bland JM, Robertson WB, Brosens I. Uteroplacental arterial changes related to interstitial trophoblast migration in early human pregnancy. Placenta 1983; 4: 397–414. Pijnenborg R, Vercruysse L, Hanssens M. The uterine spiral arteries in human pregnancy: facts and controversies. Placenta 2006; 27: 939–58. Kaufmann P, Burton GJ. Anatomy and genesis of the placenta. In: Knobil E, Neill JD, eds. The Physiology of Reproduction, 2nd edn. New York: Raven Press, 1994: 441–84. Hempstock J, Jauniaux E, Greenwold N, Burton GJ. The contribution of placental oxidative stress to early pregnancy failure. Hum Pathol 2003; 34: 1265–75. Wood C. The expansile behaviour of the human uterus. J Obstet Gynaecol Br Commonw 1964; 71: 615–20. Meekins JW, Pijnenborg R, Hanssens M, McFayden IR, van Asshe A. A study of placental bed spiral arteries and trophoblast invasion in normal and severe pre-eclamptic pregnancies. Br J Obstet Gynaecol 1994; 101: 669–74. Burchell C. Arterial blood flow in the human intervillous space. Am J Obstet Gynecol 1969; 98: 303–11. Jauniaux E, Gulbis B, Burton GJ. The human first trimester gestational sac limits rather than facilitates oxygen transfer to the foetus: a review. Placenta 2003; 24 (Suppl A): S86–93. Jauniaux E, Poston L, Burton GJ. Placental-related diseases of pregnancy, human evolution and oxygen. Hum Reprod Update 2006; 12: 747–55. Moll W, Kunzel W, Herberger J. Hemodynamic implications of hemochorial placentation. Eur J Obstet Gynecol Reprod Biol 1975; 5: 67–74. Moll W. Structure adaptation and blood flow control in the uterine arterial system after hemochorial placentation. Eur J Obstet Gynecol Reprod Biol 2003; 110: S19–27. Hustin J, Jauniaux E, Schaaps JP. Histological study of the materno-embryonic interface in spontaneous abortion. Placenta 1990; 11: 477–86. Hustin J, Jauniaux E. Morphology and mechanisms of abortion. In: Barnea E, Hustin J, Jauniaux E, eds. The First Twelve Weeks of Gestation: A New Frontier for Investigation and Intervention. Heidelberg: Springer-Verlag, 1992: 469–85.

54

22.

23.

24.

25.

26.

27.

28.

29. 30.

31.

32.

33.

34.

35.

36.

37.

38.

Fetal Cardiology

Hustin J, Schaaps JP. Echographic and anatomic studies of the maternotrophoblastic border during the first trimester of pregnancy. Am J Obstet Gynecol 1987; 157: 162–8. Burton GJ, Jauniaux E, Watson AL. Maternal arterial connections to the placental intervillous space during the first trimester of human pregnancy: the Boyd collection revisited. Am J Obstet Gynecol 1999; 181: 718–24. Burton GJ, Watson AL, Hempstock J, Skepper JN, Jauniaux E. Uterine glands provide histiotrophic nutrition for the human fetus during the first trimester of pregnancy. J Clin Endocrinol Metab 2002; 87: 2954–9. Schaaps JP, Tsatsaris V, Giffin F et al. Shunting the intervillous space: new concepts in human uteroplacental vascularization. Am J Obstet Gynecol 2005; 192: 323–32. Jauniaux E, Watson AL, Hempstock J et al. Onset of maternal arterial blood flow and placental oxidative stress; a possible factor in human early pregnancy failure. Am J Pathol 2000; 157: 2111–22. Jauniaux E, Watson AL, Burton GJ. Evaluation of respiratory gases and acid-base gradients in fetal fluids and uteroplacental tissue between 7 and 16 weeks. Am J Obstet Gynecol 2001; 184: 998–1003. Burton GJ, Hempstock J, Jauniaux E. Oxygen, early embryonic metabolism and free radical-mediated embryopathies. Reprod Biomed Online 2003; 6: 84–96. Jauniaux E, Gulbis B. Fluid compartments of the embryonic environment. Hum Reprod Update 2000; 6: 268–78. Jawerbaum A, Gonzales E. The role of alterations in arachidonic acid metabolism and nitric oxide homeostasis in rat models of diabetes during early pregnancy. Curr Pharm Des 2005; 11: 1327–42. Ezashi T, Das P, Roberts RM. Low O2 tensions and the prevention of differentiation of hES cells. Proc Natl Acad Sci USA 2005; 102: 4783–8. Jauniaux E, Hempstock J, Greenwold N, Burton GJ. Trophoblastic oxidative stress in relation to temporal and regional differences in maternal placental blood flow in normal and abnormal early pregnancies. Am J Pathol 2003; 162: 115–25. Martin CB, McGaughey HS, Kaiser IH et al. Intermittent functioning of the uteroplacental arteries. Am J Obstet Gynecol 1964; 90: 819–23. Hung TH, Skepper JN, Burton G. In vitro ischemia-reperfusion injury in term human placenta as a model for oxidative stress in pathological pregnancies. Am J Pathol 2001; 159: 1031–43. Blackburn ST, Loper DL. Maternal, Fetal and Neonatal Physiology: A Clinical Perspective. Philadelphia: Saunders, 1992. Clapp J, Seaward BL, Sleamaker RH, Hisser J. Maternal physiologic adaptation to early human pregnancy. Am J Obstet Gynecol 1988; 159: 1456–60. Mabie WC, DiSessa TG, Crocker LG, Sibai BM, Arheart KL. A longitudinal study of cardiac output in normal human pregnancy. Am J Obstet Gynecol 1994; 170: 849–56. Duvekot JJ, Cheriex EC, Pieters FA, Menheere PA, Peeters LL. Early pregnancy changes in hemodynamic and volume homeostasis are consecutive adjustments triggered by a primary fall in systemic vascular tone. Am J Obstet Gynecol 1993; 169: 1382–92.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

50.

51.

52.

53.

54.

Jurkovic D, Jauniaux E, Campbell S. Doppler ultrasound investigations of pelvic circulation during the menstrual cycle and early pregnancy. In: Barnea E, Hustin J, Jauniaux E, eds. The First Twelve Weeks of Gestation: A New Frontier for Investigation and Intervention. Heidelberg: Springer-Verlag, 1992: 78–96. Resnik R, Brink GW, Plumer MH. The effect of progesterone on estrogen-induced uterine blood flow. Am J Obstet Gynecol 1977; 128: 251–4. de Ziegler D, Bessis R, Frydman R. Vascular resistance of uterine arteries: physiological effects of estradiol and progesterone. Fertil Steril 1991; 55: 775–9. Jauniaux E, Jurkovic D, Delogne-Desnoek J, Meuris S. Influence of human chorionic gonadotropin, oestradiol and progesterone on uteroplacental and corpus luteum blood flow in normal early pregnancy. Hum Reprod 1992; 7: 1467–73. Jauniaux E, Johnson MR, Jurkovic D et al. The role of relaxin in the development of the uteroplacental circulation in early pregnancy. Obstet Gynecol 1994; 84: 338–42. Dickey RP, Hower JF. Effect of ovulation induction on uterine blood flow and oestradiol and progesterone concentrations in early pregnancy. Hum Reprod 1995; 10: 2875–9. Meuris S, Nagy AM, Delogne-Desnoeck J, Jurkovic D, Jauniaux E. Relationship between the establishment of intervillous blood flow and the human chorionic gonadotropin early pregnancy peak. Hum Reprod 1995; 10: 947–50. Jauniaux E, Jurkovic D, Campbell S, Kurjak A, Hustin J. Investigation of placental circulations by color Doppler ultrasound. Am J Obstet Gynecol 1991; 164: 486–8. Bower S, Bewley S, Campbell S. Improved prediction of preeclampsia by two-stage screening of uterine arteries using the early diastolic notch and color Doppler imaging. Obstet Gynecol 1993; 82: 78–83. Jurkovic D, Jauniaux E, Kurjak A et al. Transvaginal color Doppler assessment of utero-placental circulation in early pregnancy. Obstet Gynecol 1991; 77: 365–9. Jauniaux E, Jurkovic D, Campbell S, Hustin J. Doppler ultrasound features of the developing placental circulations: correlation with anatomic findings. Am J Obstet Gynecol 1992; 166: 585–7. Dickey RP, Gasser RF, Hower JF et al. Relationship of uterine blood flow to chorionic sac and embryo growth rate. Hum Reprod 1995; 10: 2676–9. Matijevic R, Meekins JW, Walkinshaw SA, Neilson JP, McFadyen IR. Spiral artery blood flow in the central and peripheral areas of the placental bed in the second trimester. Obstet Gynecol 1995; 86: 289–92. Coppens M, Loquet P, Kollen M, De Neubourg F, Buytaert P. Longitudinal evaluation of uteroplacental and umbilical blood flow changes in normal early pregnancy. Ultrasound Obstet Gynecol 1996; 7: 114–21. Valentin L, Sladkevicius P, Laurini R, Soderberg H. Marsal K. Uteroplacental and luteal circulation in normal first trimester pregnancies: Doppler ultrasonographic and morphologic study. Am J Obstet Gynecol 1996; 174: 768–75. Makikallio K, Tekay A, Jouppila P. Uteroplacental hemodynamics during early human pregnancy: a longitudinal study. Gynecol Obstet Invest 2004; 58: 49–54.

Placental circulations

55.

56.

57. 58.

59.

60.

61.

62.

63.

64.

65.

66.

67.

68.

69.

70.

71.

72.

Jaffe R, Woods JR. Color Doppler imaging and in vivo assessment of the anatomy and physiology of the early uteroplacental circulation. Fertil Steril 1993; 60: 293–7. Kurjak A, Kupesic S. Doppler assessment of the intervillous blood flow in normal and abnormal early pregnancy. Obstet Gynecol 1997; 89: 252–6. Carter AM. When is the maternal placental circulation established in Man? Placenta 1997; 18: 83–7. Craven CM, Ward K. Syncytiotrophoblastic fragments in first-trimester decidual veins: evidence of placental perfusion by the maternal circulation early in pregnancy. Am J Obstet Gynecol 1999; 181: 455–9. Jaffe R, Jauniaux E, Hustin J. Maternal circulation in the first trimester human placenta: myth or reality. Am J Obstet Gynecol 1997; 176: 695–705. Jauniaux E. Intervillous circulation in the first trimester: the phantom of the color Doppler obstetric opera. Ultrasound Obstet Gynecol 1996; 8: 73–6. Jauniaux E, Greenwold N, Hempstock J, Burton GJ. Comparison of ultrasound and Doppler mapping of the intervillous circulation in normal and abnormal early pregnancies. Fertil Steril 2003; 79: 100–6. Wilmut I, Sales DI, Ashworth CJ. Maternal and embryonic factors associated with prenatal loss in mammals. J Reprod Fertil 1986; 76: 851–64. Palmer AE, London WT, Sly DL, Rice JM. Spontaneous preeclamptic toxaemia of pregnancy in the patas monkey (Erythrocebus patas). Lab Anim Sci 1979; 29: 102–6. Seidl DC, Hughes HC, Bertolet R, Lang CM. True pregnancy toxaemia (preeclampsia) in the guinea pig (Cavia pocellus). Lab Anim Sci 1979; 29: 472–8. Kokawa K, Shikone T, Nakano R. Apoptosis in human chorionic villi and decidua during normal embryonic development and spontaneous abortion in the first trimester. Placenta 1998; 19: 21–6. Jauniaux E, Jurkovic D, Campbell S. In vivo investigation of the placental circulations by Doppler echography. Placenta 1995; 16: 323–31. Jauniaux E, Nicolaides KH. Placental lakes, absent umbilical artery diastolic flow and poor fetal growth in early pregnancy. Ultrasound Obstet Gynecol 1996; 7: 141–4. Jauniaux E, Ramsay B, Campbell S. Ultrasonographic investigation of placental morphology and size during the second trimester of pregnancy. Am J Obstet Gynecol 1994; 170: 130–7. Papageorghiou AT, Yu CKH, Nicolaides KH. The role of uterine Doppler in predicting adverse pregnancy outcome. Best Pract Res Clin Obstet Gynaecol 2004; 18: 383–96. Jauniaux E, Zaidi J, Jurkovic D, Campbell S, Hustin J. Comparison of color Doppler features and pathologic findings in complicated early pregnancy. Hum Reprod 1994; 9: 2432–7. Greenwold N, Jauniaux E, Gulbis B et al. Relationships between maternal serum, endocrinology, placental karyotype and intervillous circulation in early pregnancy failure. Fertil Steril 2003; 79: 1373–9. Jaffe R, Dorgan A, Abramowicz JS. Color Doppler imaging of the uteroplacental circulation in the first trimester: value in predicting pregnancy failure or complication. AJR Am J Roentgenol 1995; 164: 1255–8.

73. 74. 75.

76.

77.

78.

79.

80.

81.

82.

83.

84.

85.

86.

87.

88.

89.

90.

55

Dempsey EW. The development of capillaries in the villi of early human placentas. Am J Anat 1972; 134: 221–38. Hustin J, Jauniaux E. Curing the human embryo - curing the placenta. Hum Reprod 1993; 8: 1966–82. Berman W, Goodlin RC, Heyman MA, Rudolph AM. Relationship between pressure and flow in the umbilical and uterine circulations of the sheep. Circ Res 1976; 38: 262–6. Charnock Jones DS, Kaufmann P, Mayhew TM. Aspects of human fetoplacental vasculogenesis and angiogenesis. I. Molecular recognition. Placenta 2004; 25: 103–13. Kaufmann P, Mayhew TM, Charnock Jones DS. Aspects of human fetoplacental vasculogenesis and angiogenesis. II. Changes during normal pregnancy. Placenta 2004; 25: 114–26. Kaufmann P, Bruns U, Leiser R et al. The fetal vascularisation of term placental villi. II. Intermediate and terminal villi. Anat Embryol 1985; 173: 203–14. Jauniaux E, Burton GJ, Jones CPJ. Early human placental morphology. In: Barnea E, Hustin J, Jauniaux E, eds. The First Twelve Weeks of Gestation: A New Frontier for Investigation and Intervention. Heidelberg: Springer-Verlag, 1992: 45–64. Demir R, Kaufmann P, Castellucci M, Erbengi T, Kotowski A. Fetal vasculogenesis and angiogenesis in human placental villi. Acta Anat 1989; 136: 190–203. Fouron JC, Teyssier G, Maroto E, Lessard M, Marquette G. Diastolic circulatory dynamics in the presence of elevated placental resistance and retrograde diastolic flow in the umbilical artery: a Doppler echographic study in lambs. Am J Obstet Gynecol 1991; 164: 195–203. Maulik D. Hemodynamic interpretation of the arterial Doppler waveform. Ultrasound Obstet Gynecol 1993; 3: 219–27. Jauniaux E, Burton GJ, Moscoso GJ, Hustin J. Development of the early human placenta: a morphometric study. Placenta 1991; 12: 269–76. Kingdom JCP, Burrell SJ, Kaufmann P. Pathology and clinical implications of abnormal umbilical artery Doppler waveforms. Ultrasound Obstet Gynecol 1997; 9: 271–86. McCarthy AL, Woolfson RG, Evans BJ et al. Functional characteristics of small placental arteries. Am J Obstet Gynecol 1994; 170: 945–51. Johnson MR, Jauniaux E, Ramsay B et al. Maternal relaxin: a determinant of fetal heart rate? Br J Obstet Gynaecol 1994; 101: 1003–4. Lees C, Jauniaux E, Jurkovic D, Campbell S. The relationship between nitric oxide placental production and umbilical artery vascular impedance in early pregnancy. Obstet Gynecol 1998; 91: 761–5. Loquet P, Broughton-Pipkin F, Symonds EM, Rubin PC. Blood velocity waveforms and placental vascular formation. Lancet 1988; 2: 1252–3. Huisman TWA, Stewart PA, Wladimiroff JW. Doppler assessment of normal early fetal circulation. Ultrasound Obstet Gynecol 1992; 2: 300–5. Wright JW, Ridgway LE. Sources of variability in umbilical artery systolic/diastolic ratios: implication of the Poiseuille equation. Am J Obstet Gynecol 1990; 163: 1788–91.

56

91.

92.

93.

94.

95.

96.

97.

98.

Fetal Cardiology

Kirkinen P, Kurmanavichius J, Huch A, Huch R. Blood flow velocities in human intraplacental arteries. Acta Obstet Gynecol Scand 1994; 73: 220–4. Pretorius DH, Nelson TR, Baergen RN et al. Imaging of placental vasculature using three-dimensional ultrasound and color power Doppler: a preliminary study. Ultrasound Obstet Gynecol 1998; 12: 45–9. Haberman S, Bracero LA, Byrne DW. Spectral Doppler index mapping of the umbilicoplacental circulation and pregnancy outcome. Gynecol Obstet Invest 2004; 58: 1–7. Yagel S, Anteby EY, Shen O et al. Simultaneous multigate spectral Doppler imaging of the umbilical artery and placental vessels: novel ultrasound technology. Ultrasound Obstet Gynecol 1999; 14: 256–61. Yagel S, Anteby EY, Shen O et al. Placental blood flow measured by similtaneous multigate spectral Doppler imaging in pregnancies complicated by placental vascular abnormalities. Ultrasound Obstet Gynecol 1999; 14: 262–6. Kingdom JCP, Kaufmann P. Oxygen and placental villous development: origins of fetal hypoxia. Placenta 1997; 18: 613–21. Bilardo CM, Nicolaides KH, Campbell S. Doppler measurements of fetal and uteroplacental circulations: relationship with umbilical venous blood gases measured at cordocentesis. Am J Obstet Gynecol 1990; 162: 115–20. Gudmundsson S, Lindblad A, Marsal K. Cord blood gases and absence of end-diastolic blood velocities in the umbilical artery. Early Hum Dev 1990; 24: 231–7.

99.

100.

101.

102.

103.

104.

105.

106.

Weiner CP. The relationship between the umbilical artery systolic/diastolic ratio and umbilical blood gas measurements in specimens obtained by cordocentesis. Am J Obstet Gynecol 1990; 162: 1198–202. Kuzmina IY, Humina-Vakulic GJ, Burton GJ. Placental morphometry and Doppler flow velocimetry in cases of chronic human fetal hypoxia. Eur J Obstet Gynecol Reprod Biol 2005; 120: 139–45. Ariyuki Y, Hata T, Kitao M. Reverse end-diastolic umbilical artery velocity in a case of intrauterine fetal death at 14 weeks’ gestation. Am J Obstet Gynecol 1993; 169: 1621–2. Jauniaux E, Gavriil P, Khun P et al. Fetal heart rate and umbilicoplacental doppler flow velocity waveforms in early pregnancies with a chromosomal abnormality and/ or increased nuchal translucency thickness. Hum Reprod 1996; 11: 435–9. Daniel-Spiegel E, Weiner Z, Weiner E, Shalev E. Umbilicalplacental blood flow gradient during the early second trimester of pregnancy. J Matern Fetal Neonatal Med 2005; 17: 133–6. Makikallio K, Jouppila P, Tekay A. First trimester uterine, placental and yolk sac haemodynamics in preeclampsia and preterm labour. Hum Reprod 2004; 19: 729–33. Neilson JP, Alfirevic Z. Doppler ultrasound for fetal assessment in high risk pregnancies. Cochrane Database Syst Rev 2000; (2): CD000073. Bricker L, Neilson JP. Routine doppler ultrasound in pregnancy. Cochrane Database Syst Rev 2000; (2): CD001450.

6 The physics of ultrasound imaging Zvi Friedman Introduction The purpose of this chapter is to review the physics of ultrasound imaging and the various parameters that define image quality, and subsequently show how the image can be optimized using the system controls.

Image quality Confident and reliable ultrasonographic diagnosis depends on the ability to consistently produce excellent images. Table 6.1 summarizes the basic concepts that will be used in this chapter to define a ‘good’ image. We shall later in this chapter see how these depend upon the various system parameters, as well as the system controls that could be used to affect them.

Generation of the ultrasonic image In the following, we shall describe how images are formed by state-of-the-art ultrasound systems. A short pulse of ultrahigh frequency sound (between 2 and 20 MHz), lasting about 1 microsecond (one millionth of a second) is transmitted into the body. The pulse is ‘focused’ and portions of it are reflected back toward the transducer. By ‘focused’ we mean that the pulse is somehow made to travel along a very thin beam. The point in the body where the beam is thinnest will be referred to as the ‘focal point’ and the total length where the beam is still ‘thin’ is called the ‘focal depth of field’. If we assume that we know the velocity of propagation of sound in the tissue, we can assign the received echo at a specific time to a definite location within the body. The quality of the image will be affected by the beam parameters detailed in Table 6.2.

Propagation of sound in the body Interference Sound propagates in the body in the form of pressure waves. Beams are formed utilizing the physical phenomenon of interference. These waves, a succession of high- and lowpressure regions, travel through the tissue at the speed of sound – about 1500 m/s. The wavelength is defined as follows. Take a snapshot of the wave at any instant. The wavelength is the distance between two adjacent points along the beams having the same phase, e.g. distance between two successive minimums or two successive maximums. Molecules crowding into a region in the tissue will generate a high-pressure zone. The molecules leaving it generate a low-pressure region. This leads to the physical concept of interference. When two or more sources send sound waves across the same volume in the tissue, the resultant sound pressure in a given region is determined by the net inflow of molecules. It is therefore determined by the algebraic sum of the individual pressures. We may also distinguish between destructive and constructive interference.

Destructive interference If the high-pressure region from one wave coincides with the low-pressure region from another wave traveling across the same region, the net resultant pressure change in that region will be less in magnitude than any of the individual pressures.

Constructive interference If the high-pressure regions of two waves coincide, the sound pressure at the region where the two waves cross will be higher in magnitude than that of each of the individual waves. This is called constructive interference.

58

Fetal Cardiology

Table 6.1 Parameters that define image quality Concept

Explanation

Detail resolution

The ability to visualize minute anatomic detail The ability to adequately visualize minute anatomic detail deep in the patient The ability to clearly and easily differentiate between tissue types (e.g. fetal lungs should be easily distinguishable from fetal liver) Non-echogenic objects in the image (e.g. amniotic fluid or simple cysts) should appear as totally ‘black’ Spurious echoes resulting from a variety of artifacts might cause these areas to appear as echogenic and lead to misdiagnosis Uniformity of all the above throughout the entire image

Sensitivity

Contrast resolution

Low noise

Consistency

Beam forming The beam forming in modern ultrasound systems is accomplished using one- or two-dimensional transducer arrays. Beam forming allows us to better visualize minute anatomical details in the patient. The ‘transmit beam’ is our ‘searchlight’. The receive focus is the ‘microscope’ with which we examine the lighted area. With one- or twodimensional transducer arrays, beam forming is accomplished electronically, as will be elaborated below. As can be intuitively understood, two-dimensional arrays can provide much narrower beams.

Frequency and bandwidth When measuring the pressure at any given point in the body, we shall find a periodic temporal behavior. The number of pressure maximums (or minimums) per second at any given location within the body is defined as the frequency of the sound wave. An ultrasonic pulse will not persist, however, at any given point longer than 1 microsecond, (this is the typical pulse length used in diagnostic ultrasound). In order to measure the frequency, we simply measure the time difference between two adjacent maximums (or minimums). Thus, if the time difference between two adjacent maximums is T microseconds, the wave frequency is f = 1/T MHz (millions of cycles per second). The above description is quite crude. The main reason is that for a time signal to be described as a single frequency signal (a ‘pure’ sine wave), it must be long

Table 6.2 quality

Beam parameters that affect image

Parameter

Explanation

Beam width – transmit

The width of the beam along which the transmitted beam travels The width of the beam along which the received (reflected) beam travels The length (time duration) of the ultrasonic beam transmitted into the body The length (time duration) of the electric pulse generated by the transducer when the ultrasonic beam impinges on it Approximately 1540 meters per second The fact that this velocity is not the same for all tissues is one of the reasons for degradation in image quality

Beam width – receive

Pulse length – transmit

Pulse length – receive

Velocity of sound in the tissue

enough in time. The longer the signal lasts, the ‘purer’ is its frequency. Thus, for example, a 3.5-MHz signal lasting 1 microsecond (about three cycles) will contain signals in the frequency range 2.5–3.5 MHz. (The frequency range is inversely proportional to the time duration of the signal.) The situation is illustrated in Figure 6.1. To get a rough idea of the actual frequency content of the signal, we measure time differences between all adjacent peaks (maximums or minimums). Each measurement will result in a slightly different value for the frequency, thus defining the frequency range in the signal.

Attenuation As the signal travels across the body its intensity is continuously decreased. Intensity is lost due to absorption, scattering, and specular reflection. The three mechanisms are summarized in Table 6.3.

Specular reflection and scattering Only two out of three mechanisms listed in Table 6.3 will eventually result in an ultrasound wave getting back to the transducer: specular reflection and scattering. Specular (mirror-like) reflection occurs when the reflector is a large extensive (relative to the wavelength – 0.44 mm at 3.5 MHz) surface. Examples of such a surface could be the boundaries of a kidney, the surface of the placenta, the diaphragm, the skin of a fetus, etc. Specular reflections are

The physics of ultrasound imaging

highly directional and will occur only if the reflecting surface is precisely perpendicular to the direction of the beam. The amplitude of the reflected beam depends only upon the properties of the tissues on both sides of the surface. The amplitude of the reflected wave will be large for tissues with very different properties and will be very direction dependent. Scattering is generated by parenchymal structures. It occurs when the ultrasonic wave is reflected from particles that are of the size of one wavelength (0.44 mm at 3.5 MHz) or smaller. The intensity of the reflection will depend upon the size of the reflector relative to the wavelength. The larger is the reflector, the stronger is the reflection. It has

been shown experimentally that the intensity of the reflection is roughly proportional to the square of the ratio between the particle size and the wavelength. This means that if the parenchymal architecture that gives rise to echoes in one organ is three times larger than that in another, the scatter will be nine times more intense. Similarly, doubling the frequency of the ultrasound (halving the wavelength) increases the relative echo strength by a factor of 4.

Absorption Absorption is a significant factor in attenuation. Unlike specular reflection and scatter, it does not result in a wave that gets back to the transducer. Rather, it represents a series of processes through which the energy in the ultrasonic wave is eventually converted into heat. In the frequency range usually used in diagnostic ultrasound, the absorption coefficient is proportional to the wave frequency. Thus, for example, a 7.0-MHz wave will lose 50% of its original intensity at a depth that is about half the depth for which a 3.5-MHz wave loses 50% of its original intensity.

0.4

Pressure [MPa]

59

0

–0.4

Acoustic impedance –0.8 0

0.4

0.8

1.2

1.6

t [μs]

Figure 6.1 Pressure wave generated by a 3.5-MHz probe.

Table 6.3

2

As stated above, when an ultrasonic beam meets an interface between two types of tissues, part of the beam is reflected. The proportion of the sound that is reflected depends on the acoustic properties of the two tissues. As a matter of fact, the acoustic properties affecting reflection can be related to a single parameter in the tissue, sometimes

The three mechanisms causing sound attenuation in the body

Mechanism

Description

Angle dependence

Frequency dependence

Absorption

Kinetic energy of particles moving into high-pressure regions and out of low-pressure regions is converted into heat due to viscous friction

Absorption is angle independent

The intensity of the beam is decreased with depth, following an exponential decay law

Scattering

Part of the beam is scattered (in all directions) by small microscopic particles (smaller than the wavelength)

Scattering is angle independent

Scattered intensity is proportional to the 4th power of the frequency

Specular reflection

Part of the beam is reflected at the surfaces interfacing between two regions that have different acoustic properties

Specular (mirror-like) reflection is strongly angle dependent

Specular reflection is not frequency dependent

The distance at which the beam intensity is reduced by half is inversely proportional to the frequency

60

Fetal Cardiology

referred to as the acoustic impedance of the tissue. This parameter is a function of the density of the tissue and its stiffness.

Shadowing and enhancement Ideally, an ultrasound image would be one for which the gray-level at a given point on the screen represents unambiguously the reflection coefficient of the corresponding point in the tissue. In reality this is almost never the case due to several artifacts. Two of these sonographic artifacts will be discussed here while others will be discussed later. When an echo is so large that the sound continuing into the tissue is highly attenuated, an acoustic shadow will be cast,

as in the example in Figure 6.2 showing a transthoracic image of a fetus. The shadows cast by the fetal ribs are clearly seen. A similar situation will occur when the attenuation is caused by absorption rather than by reflection, as can be seen in Figure 6.3. The reverse phenomenon will occur when the sound traverses through a medium with relatively small attenuation, i.e. no absorption and no scattering. Typical examples could be a cyst or a pocket of amniotic fluid. Because the beams passing through the cyst are less attenuated than the rest of the beams that pass through tissue with higher attenuation, the area behind will be brighter. This is called posterior enhancement, and is sometimes used to differentiate between fat (highly absorbing) and liquid (nonabsorbing). The difference between the two situations is illustrated in Figures 6.4 and 6.5.

Figure 6.2

Figure 6.4

Fetal chest. Note the acoustic shadowing formed by the ribs.

Acoustic enhancement posterior to a cystic mass.

Figure 6.3

Figure 6.5

Acoustic shadowing formed by highly absorbing tissue.

Acoustic shadowing by a fatty mass. Note the lack of posterior enhancement.

The physics of ultrasound imaging

Resolution Spatial (detail) resolution The concept of resolution relates to the amount of detail that can be perceived in an image. In a high resolution image small structures can easily be seen. A simplified definition of system resolution is the separation between two point test targets that are just resolvable (can be seen in the image as two distinct points). The axial resolution is limited by the pulse length, and is due to the fact that two targets along the beam direction will be separable only if the echo from the first target dies off before echo from the second target starts arriving. The lateral resolution on the other hand is determined by the beam width. Both axial and lateral resolutions depend on the relative intensity (reflection coefficient) of the targets. Thus, for example, for two equal targets along the beam the resolution will be relatively high, since echoes arriving from the second target will be separable from those arriving from the first target after the latter are decreased by a factor of approximately 2. If, on the other hand, the first target is 1000 times stronger than the second target (a situation that very often occurs in diagnostic ultrasound), one needs to wait until the echoes from the first target are less than one-thousandth of their peak value. When evaluating equipment, one should be very careful in examining and comparing resolution numbers quoted by the manufacturers. These numbers must also state the relative intensities of the targets in order to have any meaning at all. In most state-of-the-art ultrasound systems the axial resolution is much smaller than the lateral resolution. Both axial and lateral resolution degrade with imaging depth. The degradation of the axial resolution with the depth is a result of attenuation in the tissue, and is related to the fact that higher frequencies are attenuated faster than lower frequencies. The lateral resolution for a given sound frequency and a given transducer aperture is inversely proportional to the distance. Thus, if the resolution at a distance of 50 mm is 1 mm, it will be degraded to 2 mm at a distance of 100 mm. In actuality the situation will be even worse, due to the dependence of the attenuation on the depth. Thus, for example, the frequency that can be used for imaging at a depth of 100 mm is half the frequency that could have been used for imaging at a depth of 50 mm. Most modern systems attempt to provide a uniform resolution throughout the entire image. In order to accomplish this objective, the transducer aperture is increased with the depth of imaging. This partly takes care of the dependence of the lateral resolution on the depth. However, it hignores the effects of the attenuation.

Contrast resolution Contrast resolution is the ability to perceive tissue contrast differences in gray-scale images. Diagnostically important

61

image contrast results from small differences in the acoustic properties of tissue, sometimes referred to as subject contrast. It is mainly determined by the spatial distribution, concentration, and reflectivity of tissue structures within the scan plan. The ultrasound system translates these minute differences into low amplitude shades of gray.

Speckles One of the essential tasks of diagnostic ultrasound imaging is the detection of focal lesions of low contrast against background tissue. The capability to detect a target of any given shape (e.g. circular) depends on the ‘contrast’ between the target and the background, upon the target size, and upon the background ‘noise’. The main ‘noise’ factor in diagnostic ultrasound imaging is the so-called ‘speckle noise’. This noise is superimposed on the image and appears in the form of gray ‘stains’ of random sizes and random intensities. This noise ‘camouflages’ the target, mainly by ‘breaking’ its contours. It has been called speckle because of its similarity to the equivalent optical phenomenon, laser speckle. The phenomenon of speckles results directly from the use of coherent radiation for imaging. It occurs when structures in the object (tissue), which are on a scale too small to be observed by the imaging system (smaller than the imaging wavelength), cause interference to occur between different parts of the wave received from the object region that correspond to a given point in the image. The detection of a lesion within background tissue is a psychophysical process, carried out by the eye–brain system of the observer. The observer performs in his brain an integration process by which all signals resulting from echoes from the target area are summed together. The target lesion will be discriminated from the background if: (1) the number of individual speckles within the target area is large enough (i.e. the average speckle size is much smaller than a typical target dimension); (2) the average brightness of individual speckles in the target area is different enough from the average brightness of individual speckles in the background area. Statement 1 can also be phrased mathematically, stating that the signal to noise is proportional to m1/2, where m is the average number of speckles in the target area. The average speckle size mainly depends on the properties of the ultrasonic beam: the ultrasonic frequency, the transducer aperture, the distance from the transducer, and the pulse length. The image sampling, i.e. the number of beams across the field of view that are used to generate the image (image line density), as well as the number of samples along the beam, can also affect the speckle pattern. Modern well-designed systems are sampled densely enough so that sampling will have minimal effects on the image. The speckle intensity will depend on the average number of scatterers per unit volume in the tissue, the difference in

62

Fetal Cardiology

acoustic properties between the scatterers and the tissue in which they are immersed, their size, and the uniformity of their distribution, as well as on the properties of the beam. However, since the amplitude and distribution of the speckles are related not only to the ultrasonic beam properties but also to the parenchymal tissue architecture, the appearance of the speckles is fairly constant for each type of tissue and differs between tissues. This allows us to use the speckle information to discriminate between tissues, despite the fact that the echo pattern is random in nature.

Under-sampling and aliasing As previously stated, the sampling rate (in both axial and lateral directions) must be high enough to comply with the speckle density, i.e. the distance between two samples must be much smaller than the average speckle size or the average distance between two speckles in the image. One of the most severe effects of ‘under-sampling’, i.e. sampling at a rate which is not high enough, on the image is called aliasing. Speckles are caused by the interference of echoes from densely packed tiny parenchymal reflecting structures that are closer than the basic system resolution. Thus, there is no one-to-one correspondence to any identifiable structure in the body, which is relatively large. Small movements of the probe will cause continuous small movements of such structures. The speckle pattern will, however, change abruptly, giving the effect of a moving background noise. Of course, the larger is the change the more disturbing is the effect. The amount of change will depend first on the ratio between the sampling rate and the natural resolution of the system. Another important factor will be the gain setting of the system, where the effect will be most annoying at high gain.

Speckle reduction Because of the undesirable effect of speckles on the capability to distinguish lesions from the surrounding tissue, it would be desirable to make the speckles ‘finer’ (decrease their average size), as well as make them less intense. The first objective can be reached by increasing the frequency bandwidth of the ultrasonic beam; this results in reducing the interference. The speckle intensity can be reduced by image compounding. The object is imaged from various aspects. The displayed image is then an average of the images from the various aspects. In one commercial implementation the object is viewed from eight directions. Obviously, the various directions must be far enough apart in order for the speckles to ‘average out’. The effect of this implementation is illustrated in Figure 6.6.

Reverberations Consider yourself in a room that has mirrors on two opposite walls. Now try to estimate your distance from one of the mirrors by just looking at it. Multiple reflections at the opposing mirrors will probably make such an attempt useless. A similar situation occurs in acoustics. As discussed above, when the sound beam is perpendicular to an interface, part of it will be reflected. The amount of the reflected wave will depend upon the intensity of the incident wave. It will, however, also depend on the difference in the acoustic properties of the two tissues. The reflected wave will then hit the face of the transducer. Since the transducer material has acoustic properties that are very different from those of the patient tissue, only part of the wave will enter the transducer. A large portion of the wave will bounce

Figure 6.6 The effect of image compounding. Images taken in compound mode (a) and in non-compound mode (b).

The physics of ultrasound imaging

back into the patient. It will travel to the reflecting surface and back to the transducer. Since the time for such a multiple round trip between the transducer and the patient is a multiple of the time for a single round trip, and since the system registers the axial location of an object according to the time of a single round trip, the same object will be registered in multiple positions in the image.

Image formation in premium high resolution diagnostic ultrasound systems Introduction Premium high resolution ultrasound systems have become increasingly sophisticated and include many new signalprocessing features. The main factors affecting diagnostic image quality have already been identified to be detail resolution, sensitivity, contrast resolution, noise, and consistency. In this section the entire system signal-processing chain will described in detail and the effects of the various blocks and system features on the diagnostic image quality will be elaborated.

63

by the interaction of the ultrasound beam with the tissue. It ends in the eye–brain system of the of the sonographer. The machine must, however, do a larger amount of electronic signal-processing before the echo information it receives can be displayed as a clinically useful diagnostic image. The signal-processing functions include: improving the lateral resolution by focusing techniques, correcting for the attenuation (in order to make the image brightness uniform), adjusting the amplitude of the echoes further to emphasize particular features of interest, estimating the amplitudes of the missing echoes, inserting the results in the right places to generate a smooth, pleasant image, and rejecting misleading information.

Focusing To generate a high resolution image, the origin of each echo must be accurately determined. For this to be achieved, one needs to (1) use high frequency, very short pulses; and (2) make the beam as narrow as possible. Modern scanners employ multielement electronic beamforming and focusing. In order to understand the operation and performance of these beam-forming techniques, let us first review the ‘fixed lens‘ focusing method.

Fixed lens focusing

The signal-processing chain The signal-processing chain is schematically described in Figure 6.7. The signal processing chain begins with the scanned crosssection in the patient, the features of which are determined

Older technology, such as single-element mechanical sector scanners, used a fixed lens to focus the beam. Focusing is accomplished by attaching a lens – a piece of plastic material, usually shaped spherically, as shown in Figure 6.8 – in which the sound wave travels more slowly than in the tissue. Since the wave from the edge of the transducer

Figure 6.7 Signal-processing chain.

64

Fetal Cardiology

starts traveling in the tissue before the wave from the center of the transducer, and since the speed of sound in the lens is slower than that in the transducer, the wavefront will bend and become spherical.

Focusing with an array transducer The same principle is used in array transducers. The effect of curving the wavefront is accomplished by dividing the transducer into a large number of very small transducers (elements).

Focusing (lens) r2 c2

r2

c2

c1

r1

r0

Figure 6.8 Focusing with a fixed focus lens.

+

r1 c1

=

r0 c0

On transmit The pulse is fired from each of the elements separately. The edge of the array is fired first and the central element is fired last. By properly adjusting the time delays between the elements, a spherical wavefront is produced, similar to that produced by the fixed lens. The concept is described schematically in Figure 6.9.

Confocal imaging on transmit With an electronic transducer array the focusing point is determined by the set of delays between the pulses. This results in a beam with a focal point at which lateral resolution is best. Areas close to the focal point portray good image quality. Other areas are out of focus and have much lower image resolution. In order to improve image resolution in other areas, more focal points are used. To generate an ultrasonic line, several consecutive beams are fired. Each of the beams is focused at a different point. For each beam, only a segment close to the focal point is used, whereas the rest of the beam is ignored. Combining the segments around each of the focal points then generates the entire image. The fact that many beams need to be transmitted in order to generate a single line in the image severely reduces the frame rate. Sometimes the frame rate is further reduced, since in order to avoid mixing of reflections of the present beam from the near zone with reflections of the previous beam by strong reflectors deep in the body, one should allow more time for all the echoes from a pulse to arrive before transmitting the next pulse. This extra time can be saved if consecutive beams are transmitted along separated lines. This sort of technique allows, in some systems, one to achieve clinically acceptable frame rates with more than five transmit focal zones.

Total delay First

1 2

Wave front

3 4 5 6

Figure 6.9

7

To focus the beam, the pulses are first applied to the outside elements and then later to the elements that are situated nearer the center of the array. Each element is a center of a circular wave. The circular wave fronts spreading from the outer elements propagated a greater distance because they were fired earlier. The varying distances of propagation result in crossing points (where constructive interference occurs) lying along a circular curve.

8

Last

9 10 11 12 13 14 15 First

16

The physics of ultrasound imaging

Dynamic focusing on receive Focusing a transducer array on receive is very similar to focusing it on transmit. One simply has to delay the output of each of the elements properly and then sum them together. Owing to the fact that echoes from deeper points in the patient arrive later at the surface of the transducer, one can easily optimize the image by dynamically varying the delays as function of time so that all echo sources are focused (summed in phase to generate a constructive interference). At the summation point, echoes from all other points will not be in phase, resulting in destructive interference.

Out-of-plane focusing and beam-width artifacts Most array transducers available in the market today employ a single row of 128–192 very small elements. This allows excellent focusing of the beam in the plane of scanning, which is the plane perpendicular to the transducer surface and which cuts the center points of all the elements. The beam width in the perpendicular direction is focused with a cylindrical lens along the length of the array. This allows us to focus the beam at some distance from the array. Since, due to practical limitations, the transducer aperture in this direction is rather limited (typically 10–15 mm), and since the focus is fixed, the beam is relatively wide in this direction. The results of a wide beam in the out-of-plane may be painful. Let us consider two examples. The first example comes from gynecology. It is well known that diagnosing an adnexal mass as a simple cyst may be crucial to determine its benign nature. A simple cyst appears black (non-echoic) since it does not absorb and does not scatter. If the beam is too wide, one of the portions of the beam may be interacting with the fluid (non-echoic) structure while another portion of the beam interacts with adjacent soft tissue. This could create echoes registered within the cystic structure. A second example is when one desires to detect a low contrast lesion within a tissue, where the contrast is even further reduced due to the width of the beam.

Side lobes and grating lobes One of the more serious problems in multielement array transducers is related to image degrading artifacts referred to as side lobes and grating lobes. Generally, side lobes will appear close to the main beam, whereas grating lobes appear at a relatively large distance from the main lobe.

Side lobes Although the majority of the sound produced by a simple transducer propagates directly away from the transducer

65

face to generate the so-called main beam, a small portion of the energy will be concentrated outside the main central beam. These secondary, side lobes are produced because at the lateral margins of the sound source a portion of the sound energy is transmitted radially away from the beam axis. Spatial variation in side lobe intensity occurs because of interference between sound energies arising from opposite sides of the transducer. Interference can be either constructive or destructive. As side lobes are mainly an edge effect, their intensity relative to the intensity of the main lobe will decrease with increasing size of the transducer (the size is most conveniently expressed in units of the wavelength). Also, since side lobes are caused by interference, their intensity will decrease if the pulse length is decreased. A second source of side lobes in array transducers comes from the discontinuous nature of the aperture. The ideal shape of the aperture is a circular curve, which we now approximate with a set of linear segments. The approximation will naturally improve as we decrease the length of each of the segments and increase their number. The contribution of this source to the side lobes will therefore decrease as we increase the number of elements and decrease their size. Side lobes mainly disturb the possibility to see a weak target next to a strong target, since the side lobe intensity of the strong target might be larger than the echo of the second target. The effect of each of the above parameters on the side lobe intensity is summarized in Table 6.4.

Grating lobes Grating lobes are caused by the constructive interference of laterally directed energy from edges of the individual array elements. Their intensity and angle from the main beam depends on the center to center spacing (pitch) of the elements, on the spacing between edges of adjacent elements, upon the wavelength (or sound frequency), and upon the size and curvature of the array. The effect of each of the above parameters on the grating lobe intensity and position is summarized in Table 6.5. Modern, high-performance systems should not produce noticeable grating lobes. As can be seen from Table 6.5,

Table 6.4 Effect of various array parameters on side lobe intensity Parameter change

Effect on side lobe intensity

Increase total array aperture

Decrease

Increase sound frequency

Decrease

Decrease element size, but keep total aperture fixed (by increasing number of elements)

Decrease

66

Fetal Cardiology

Table 6.5 Effect of various array parameters on grating lobe intensity Parameter change

Effect on grating lobe intensity and position

Increase total array aperture

Increase

Increase sound frequency

Increase: grating lobes will appear closer to the main lobe

Decrease element size, but keep total aperture fixed (by increasing number of elements)

Decrease: grating lobes will be shifted far away from the main lobe

Increase probe radius of curvature

Decrease

50

100

150

200

In order to extract all the clinically relevant information contained in the image and display it to the diagnostician, the acoustic data must be electronically manipulated so that all the clinically relevant information is faithfully displayed on the limited range monitor. It will be convenient to base the operation of the system on the concept of dynamic range. The concept of dynamic range is most easily explained using Figure 6.11. As can be seen from Figure 6.11, all ultrasound signals start from the zero level and end at the system saturation level. The lower portion of the signal lies in the range within the noise level and is therefore obscured by it. The system reject level is set just above the noise level so that only the echoes above the reject level are displayed. The ratio between the saturation and reject level is defined as the system dynamic range. In most modern systems, the dynamic range of the echoes exceeds 100 dB (a ratio of 100 000:1). Since the system cannot display this whole range of information, several steps are usually taken in order to reduce it with minimal diagnostic penalty. This will be most effectively accomplished by noting that the huge dynamic range of 100 000:1 is mainly a result of two effects. The range of each of the effects is approximately 300:1. The first effect is the result of tissue attenuation, which is negligible at close range but is very high deep in the image, whereas the second effect is a result of the differences in reflection coefficients. An ideal ultrasound image is a two-dimensional map of reflection coefficients regardless of depth. Our first task would therefore be to minimize the effects of attenuation by the tissue. This task is achieved using a system function called time gain compensation (TGC), to be described below.

250 50

100

150

200

250

Figure 6.10 Shades of gray representing a dynamic range of 64:1. The brightest shade (white) represents the value 64 and the darkest shade (black) represents the value 1.

increasing the imaging frequency for a given transducer will produce grating lobes.

Dynamic range: gray-scale The acoustic transducers used in diagnostic ultrasound are capable of recording echoes over a range of pressures in excess of 100 000:1 (100 dB). Presently available state-ofthe-art monitors are capable of displaying a range of no more than approximately 32:1 (30 dB). Generally, the information is displayed in the form of gray-levels as shown in Figure 6.10.

Time gain compensation The effect of attenuation by the tissue can be at least partly compensated for by progressively increasing the gain of the system as echoes come deeper from the body. When sound is attenuated by a large homogeneous object such as the liver, the reduction in intensity is exponential. In order to compensate for this type of attenuation, one would need to increase the gain logarithmically with time, since time is equivalent to depth. Since the attenuation coefficient may be patient dependent, one could fairly well compensate for this type of attenuation by adjusting a single parameter. The situation in obstetrics is, however, far more complicated. The ultrasonic beam will traverse through fat, muscles (e.g. from the maternal abdominal wall), amniotic fluid, placenta, fetal bones and tissues, etc. The attenuation is extremely heterogeneous and can by no means be represented using a single parameter. In this case, a multiple gain adjustment with some 8–12 sliding potentiometers is implemented. The image is then optimized by trial and error.

The physics of ultrasound imaging

67

Signal dynamic range

Figure 6.11 The dynamic range of a representative ultrasound system. All ultrasonic signals begin at a zero signal level and can increase in amplitude until they reach the system ‘saturation level’. Some of the low-level signals fall in the region of the background noise and will therefore be obscured. All modern systems have a built-in system-reject, which eliminates both system noise and the low-intensity echoes that lie just above the noise level. The dynamic range of the system is the ratio between the system saturation level and the system reject level.

Noise

Displayed brightness

Saturation level

Zero level Noise

Signal dynamic range

Compression

Post-processing In addition to the logarithmic compression, an additional post-processing function is usually applied to the resultant curve. This function is application specific. Examples of such functions are shown in Figure 6.13. The selection of the post-processing function is crucial to obtaining the desired appearance of the image. Different image appearances are optimally tuned for different diagnostic requirements and applications. Image appearance that is best suited to emphasize outlines of objects is different from the image appearance that would maximize the contrast between similar tissues. The first example is characterized by large echo differences (e.g. between the heart chambers and the myocardium). In this case, the information required could be extracted even if the 30 shades of gray provided by the display are equally divided between the 300 input levels, as described in Figure 6.13. On the other hand, when imaged at such a setting, the tissue part of the image will appear ‘grainy’, i.e. large, high contrast speckles.

250 200 Output level

After having reduced the original 100 000:1 input dynamic range by the TGC to a value of about 300:1, the dynamic range needs to be further compressed until a displayable range of 30:1 is reached. Most ultrasound systems use some kind of logarithmic function for compression of the dynamic range, as described in Figure 6.12.

300

150 100 50 0

0

1

2

3

4

5

6

7

8

9

10

Input level (arbitrary units) x 105

Figure 6.12 Logarithmic compression of the dynamic range.

For the second application, the requirement is to differentiate between tissue structures with only slight difference in structure. Echoes from such tissue structures are very weak to begin with, as they mainly result from scattering from tiny scatterers. The requirement, therefore, is that the gain for low amplitude signals should be relatively high, so that similar tissue structures with only slight differences in echo signal would still look different on the screen. On the other hand, we are less interested in the highly echogenic structures, so that the fact that they are saturated (appear as white on the screen, without any

68

Fetal Cardiology

capability to resolve any structures in them) is less important. This might have an effect on the image ‘cosmetics’ but will not affect the diagnosis. Note, however, that we still do not wish to emphasize the extremely low signals that usually represent noise, or are obscured by noise (see Figure 6.11). The gain for these levels should thus be kept very low. The overall gain curve will therefore have a shape of a logarithmic S-curve as shown in Figure 6.13. This second application presents a challenge, and calls for a high quality, high dynamic range system, in which the noise level due to image artifacts such as side lobes, grating lobes, reverberations, etc. is low enough so as not to interfere with the information that needs to be displayed. The effect of the dynamic range control is illustrated in Figure 6.14, showing the same image taken at four different settings of the dynamic range.

Persistence As seen above, even very slight movements of the probe could change the speckle pattern in tissue texture echoes, but will cause much slighter changes in appearance in the larger outlines. Averaging of successive frames will therefore smooth the speckles but will hardly affect important outline information. During rapid search scanning or when imaging very fast moving objects such as heart valves, this temporal averaging is undesirable. This calls for an implementation that would allow full control of the degree of averaging. In most common implementations, the averaging is an iterative process in which the new pixel (picture element) information is averaged with the old information, with

White

Black

user selectable weights. In some implementations, weights are allowed to change automatically according to the rate of change of image data between successive frames. Weights are also made to vary between different regions in the image.

Recent developments Tissue harmonic imaging Reverberations, side lobes, and grating lobes are very obvious and disturbing in obstetrics and in echocardiography. In these two applications, one desires to accurately outline borders between tissue and large fluid compartments, e.g. the borders between the endocardium and the heart chambers. The chambers should be presented as black, since blood is a very weak reflector. Reverberations, side lobes, and grating lobes will generally appear as image noise and clutter that will obscure the boundary between blood and tissue. All these are substantially reduced with tissue harmonic imaging. Tissue harmonic imaging is based on the fact that an ultrasonic wave propagates in a fluid (or tissue), and if the acoustic amplitude is sufficiently high, it is distorted, with the amount of distortion continuously increasing as the wave propagates deeper into the tissue. This distortion is a manifestation of the deviation of the wave propagation in tissue from linearity. A simplistic way of describing the distortion is given below. Ultrasound propagates in a fluid or soft tissue as longitudinal waves of alternate compressions and rarefactions. For high enough acoustic pressure, the compressional

Brightness

S Shaped Linear Logarithmic

Signal amplitude

Figure 6.13 Examples of three possible postprocessing functions: logarithmic, linear, and S-shaped.

The physics of ultrasound imaging

69

Figure 6.14 Image taken at very low (upper left), low (upper right), high (lower left), and very high (lower right) dynamic ranges.

phases travel at a speed that is higher than the average sound velocity in this medium, whereas the rarefactional phases travel at a speed that is lower than the average speed of sound in the medium. This is a simple result of the fact that the speed of sound is larger for the more compressed (denser) regions of the tissue. We thus get a distortion that will cause a waveform that is initially sinusoidal to become more like a sawtooth, as described in Figure 6.15. The amount of distortion is proportional to the distance (measured in units of the wavelength) and to the 2nd power of the acoustic pressure. In terms of frequency content, the waveform distortion is equivalent to second as well as higher harmonics generation (integer multiples of the original frequency). The second harmonic image is then generated by filtering out the fundamental component in the receive signal. Since the distorted and scattered energy caused by reverberation, side lobe, and grating lobe artifacts is much weaker than the transmit energy, the harmonics generated will be much weaker. As a result, the tissue harmonic image contains minimal noise and clutter compared to fundamental

imaging. The effect is illustrated in Figure 6.16, comparing fetal images in the linear and second harmonic modes of imaging.

Coded excitation Higher resolution imaging can be accomplished at high ultrasonic frequencies. However, high frequency ultrasound is highly attenuated by the tissue. In theory, increasing the amplitude of the transmitted pressure wave can compensate for this effect. Unfortunately, the amplitude of the transmitted wave is limited in order not to cause adverse biological effects in the tissue. This limitation was recently greatly relieved by one of the manufacturers, applying coded excitation to ultrasonic imaging. Coded excitation encodes a special ‘signature’ to the ultrasonic beam by repeating a specific pattern of ones and zeros. The received beam is then decoded accordingly, with a resultant signal which is several times larger than the basic signal. The decoded signal amplitude is proportional to the

70

Fetal Cardiology

1 0.8 0.6 Amplitude

0.4 0.2 0 –0.2 –0.4 –0.6 –0.8

Amplitude

–1

0

1

2

3

4 ωt/2(rads)

5

6

7

0

1

2

3

4 ωt/2(rads)

5

6

7

1 0.8 0.6 0.4 0.2 0 –0.2 –0.4 –0.6 –0.8 –1

Figure 6.15 Initial waveform (top) and distorted waveform (bottom).

Figure 6.16 A fetal image in tissue harmonic-imaging mode (a) and in linear mode (b).

length of the code. Using this technique, deep abdominal imaging can be obtained at frequencies that are double the frequencies obtained using conventional techniques. The effect of coded excitation is illustrated in Figure 6.17, comparing coded excitation and conventional non-coded excitation imaging.

Gray-scale system controls and their effect on image quality As already discussed above, due to inherent physical and technological limitations an ‘ideal system’ does not exist, and the user will generally face painful compromises.

The physics of ultrasound imaging

Thus, for example, one would like to image at the highest possible frequency in order to improve resolution. However, high frequency ultrasound is highly attenuated. This would mean that for ‘heavy patients’ one needs to image at lower frequencies in order to gain penetration. In most modern systems, the imaging parameters are preset for the various applications in most expected conditions. The situations in ‘real life’ will, however, always be somewhat different from those for which the presets were determined. Some adjustments by the operator will therefore almost always be required. The system controls and their effect on overall image quality are summarized in Table 6.6.

Figure 6.17 An image in coded excitation mode (left) and in conventional (non-coded-excitation) mode (right).

Table 6.6

71

Doppler Introduction Ultrasonic imaging allows us not only to see how organs look, but also to a certain degree to analyze how they function. We can easily detect and analyze motion of blood (down to the arteriole level) or tissue, e.g. myocardial motion. In the first example, although the scattering from blood cells is extremely low, we can still ‘see’ them based upon the differences in velocity compared with the surrounding tissue. Motion analysis is based upon the physical fact that when a sound wave is reflected from a moving target, the reflected wave changes its frequency. The frequency change is proportional to the velocity component of the moving target along the direction of the beam, and to the beam frequency. The frequency change of the reflected beam will be positive (the received reflected beam will have a higher frequency than the incident beam) if the target is moving toward the transducer and will be negative (the received reflected beam will have a lower frequency than the incident beam) if the target moves away from the transducer. The situation is described schematically in Figure 6.18. Doppler motion analysis is performed within a ‘sample volume’. The term ‘sample volume’ relates to a certain region of the image, and the objective is to provide a statistical description of the motion of the various constituents within this region. The way the motion is described is called a ‘Doppler spectrum’. A Doppler spectrum is actually a histogram that describes the distribution of velocities (along the insonating beam) within the ‘sample volume’. When a certain ‘sample volume’ in the body is insonated, there will generally be a multitude of reflected waves at a variety of ultrasonic frequencies. Each of the red blood cell clusters, moving at a certain velocity, will scatter an

Gray-scale system controls

Control change

Favorable effect

Adverse effect

Increase gain

See weak echoes

Strong echoes will be saturated

Set time gain compensation

Optimize gain at all depths

Increase imaging frequency

See more detail in the near field

Lose the image at the far field

Increase system field of view

Improve orientation Improve the ability to measure large objects

Reduce frame rate, and smear of fast moving objects

Increase persistence

Reduce speckle and temporal noise

Smearing of fast moving objects

Add transmit focus

Increase image area that can be ‘seen better’

Reduce frame rate

Move transmit focus to region of interest

Improve image resolution of object of interest

Select required gray-scale curve

Optimize image for the desired diagnosis

72

Fetal Cardiology

Moving reflector

Transmitted beam

Reflected beam

Figure 6.18 An ultrasonic beam, emitted by the transducer, is reflected by moving red blood cells. The change in frequency of the wave is proportional to the velocity of the red blood cells. When the reflected signal reaches the transducer, the difference between the transmitted and the returned frequency is measured. This difference in frequency is called the Doppler frequency shift. The magnitude of the Doppler frequency shift is dependent on the angle between the ultrasonic beam and the red cell velocity vector.

ultrasonic beam with the corresponding Doppler shift. The intensities of the individual waves at given Doppler shifts will be proportional to the number of red blood cells moving at this particular velocity. A Doppler signal is therefore composed of a range of frequencies with varied amplitudes. Unfortunately, not only the red blood cells but also surrounding tissues such as the arterial walls are also moving reflectors that contribute to the Doppler signal. Since the vessel walls are usually extended in size, the reflected signals from them could be much larger in amplitude than those of blood cells. In extreme situations, when the beam is perpendicular to the wall, intensities of signals from the vessel walls might be as much as 100 000 times stronger than reflections from the blood cells themselves. Luckily, in most cases the velocities of the arterial walls are significantly lower than the velocities of the blood cells. This fact is used to discriminate between the two types of signal. There are, however, cases such as coronary flow where this is not true. There is practically no way to measure blood flow in the coronaries using conventional Doppler ultrasound techniques. Another example is capillary flow. In both cases, novel methods utilizing ultrasonic contrast agents are employed. Contrast agents are, however, not permitted in obstetric applications and are therefore beyond the scope of this book.

Doppler modes Three modes of operation will be discussed: continuous wave (CW) spectral Doppler, pulsed wave (PW) spectral Doppler, and Doppler color flow imaging.

Continuous wave Doppler Continuous wave (CW) Doppler is performed using a transducer composed of two separate elements. The first element functions continuously as the transmitter

while the second element functions continuously as the receiver. In CW Doppler all blood flow velocities detected along the axis of the transducer will contribute to the Doppler signal. In practice this means that the operator cannot determine the precise origin of the Doppler signal along the ultrasonic beam. In other words, the ‘sample volume’ in this case consists of the entire beam. This type of operation was used in obstetrics more than 20 years ago. It is still used in monitoring equipment and in echocardiology scanners. However, it is no longer found in modern imaging equipment for obstetrics and gynecology.

Pulsed wave Doppler In this mode of operation, the same transducer elements are used for both transmitting and receiving the ultrasonic beam, and in most cases also for gray-scale imaging as well as for color flow imaging (see explanation below). The transmitted signal is a relatively short pulse, although it may be several times longer than the pulse in gray-scale imaging. The returned signal is ‘gated’. This means that only the received signal at a predefined time gate (which corresponds to a specific range gate) is processed and included in the spectral analysis. The Doppler measurement starts by positioning a sample volume on the two-dimensional gray-scale image prior to activating the PW Doppler beam. Only the signals returning from the ranges of the sample volume will be analyzed. The sample volume can be positioned anywhere on the image and may assume almost any desired size. Another difference between the two modes is that unlike CW Doppler where insonation is continuous, insonation in PW Doppler generally lasts less than 1% of the time. This allows the use of lower signal intensities, which means lower hazard to the patients.

The physics of ultrasound imaging

73

Pulse repetition frequency After a pulse is transmitted, the transducer will function as a receiver until the signal has returned from the specified depth. Only then can the next pulse be transmitted. This defines a minimum time between consecutive pulses or a maximum rate or repetition frequency. The depth range thus limits the pulse repetition frequency. High pulse repetition frequencies (PRFs) are required for measuring high velocities, resulting in high Doppler frequency shifts. It can be shown mathematically that only when the Doppler frequency shift is less than one-half the sampling rate (PRF/2) can the velocity be properly assessed. The PRF is determined from the time difference between two consecutive pulses, which depends upon the range. This imposes constraints on the maximum range: high velocities can only be measured at close ranges. When the Doppler frequency shift exceeds PRF/2 a folding artifact will occur, as illustrated in Figure 6.19. Actually, for a given PRF, the measured Doppler frequency shift may be anywhere between –PRF/2 and PRF/2. This defines a measurable velocity range. The spectrum in Figure 6.19 can be ‘unfolded’ by moving the ‘baseline’. The baseline is the Doppler frequency shift (corresponding to the velocity component along the ultrasonic beam) which corresponds to the center of the spectrum. The maximum measurable velocity range is therefore dependent on the PRF, meaning that large velocity ranges may be measured at close image ranges only.

the Doppler shift signals in each of them, a two-dimensional distribution of Doppler shift signals can be obtained. In order to achieve high enough frame rates, the ‘interrogation time’ at each gate is rather limited. A detailed spectral analysis at each gate would be too long if reasonable accuracy is required and is therefore not applicable. As a result, the analysis is usually limited to the determination of the average velocity at each gate. The interrogation time at each gate, however, cannot be made too short. Slowly moving tissue will also cause a Doppler signal, which must be filtered out in order to eliminate ‘flash’ artifacts. Between six and 12 ultrasonic beam transmissions are required at each sample volume in order to distinguish blood flow from slowly moving underlying tissue, employing relatively sophisticated filtering. The Doppler frequency shifts and the amplitude of the Doppler signal are then analyzed in each individual gate. The data are used to assess either the average velocity (magnitude and direction) or the total amplitude of the signal in each of the gates. There are therefore two forms of presenting real-time blood flow information: the first, sometimes called color flow imaging, in which a colorcoded velocity map is displayed on top of the gray-scale image; and the second, sometimes referred to as power Doppler or ‘ultrasound angio’, in which the Doppler signal amplitude, which is proportional to the density of the blood cells, is displayed as a color map on top of the grayscale image.

Doppler color flow imaging

Color velocity imaging In this mode a real-time image representing blood-flow velocities is displayed as an overlay on top of the gray-scale two-dimensional image. Color flow imaging is based on a

By placing multiple gates (sample volumes) across an entire area in the gray-scale image and separately processing

Figure 6.19 Spectral folding caused by aliasing.

74

Fetal Cardiology

color scale. Similar to the gray-scale presentation used in two-dimensional imaging, the velocities measured at each Doppler gate are assigned a corresponding shade of color. The scale is designed to optimize high or low velocity display. Usually a red and blue color scheme is used. According to this scheme, blood flowing toward the transducer is colored red while blood flowing away from the transducer is colored blue. An example of a color velocity image is provided in Figure 6.20, demonstrating blood perfusion in the kidney.

Power Doppler imaging This is a relatively new way to image blood flow. What is displayed in this case is not the average velocity in the sample volume, but only the Doppler intensity. This is essentially the intensity of the reflection from moving

Figure 6.20 A color velocity image of a native kidney.

blood particles. Most physical phenomena affecting Doppler ultrasound are not relevant in this case. There is almost no angle dependence and no aliasing effect, so that very low pulse repetition rates can be used to detect very low flow without risking the artifacts associated with too low PRFs. Since power Doppler is velocity independent, values do not change significantly over the heart cycle. In particular, since we are not trying to trace the fast velocity changes during the systolic phase, temporal averaging techniques can be employed to reduce noise. This generally results in a three-fold increase in sensitivity in the capability to demonstrate low flow. Figure 6.21 shows a side-by-side comparison of renal flow as imaged by color velocity mapping and by power Doppler.

Doppler signal processing The chain of the Doppler signal processing is described in Figure 6.22. The first function in the signal processing chain is called ‘demodulation’. The problem we are trying to solve in detecting Doppler ‘frequency shifts’ (the frequency difference between the incident ultrasonic wave and the waves reflected from the moving objects) is actually quite complicated. The frequency shifts involved are usually in the range of a few hundred Hz to tens of thousands of Hz. These are to be compared to the carrier frequency (the frequency of the incident ultrasonic beam) which is of the order of several (2–10) MHz. The task of determining these relatively minute frequency shifts is generally accomplished by ‘mixing’ (multiplying point by point) the received signal with a reference ‘master clock’ signal that is of the frequency of the transmitted wave. This will result in a signal that could be written as a sum of two components having two main frequencies. The first component is

Figure 6.21 Color velocity images of a native kidney (a) and power Doppler image of the same kidney (b).

The physics of ultrasound imaging

a high frequency signal. Its frequency equals the sum of the transmitted and reflected signals. The second component is a low frequency signal that equals the difference between the frequencies of the transmitted and reflected signals. By passing the resultant signal through a low pass filter, the high frequency component is removed (filtered out). A high pass filter is also almost always used in order to filter out every frequency component that originates from arterial wall and tissue motions. It can be shown that this process will give us only the magnitude of the frequency shift. In order to get also the sign of the shift, we also ‘mix’ the signal with a second ‘master clock’ reference signal. The second signal is identical to the first master clock signal, but shifted by 90° (delayed in time by exactly one-quarter of the cycle of the signal). It can be shown that using the resultant two demodulated signals, the direction of the flow can also be determined. Our next step is to determine the frequency content of the demodulated signal. An example of such a signal is described in Figure 6.23. One can obtain a rough idea of the frequency content of the signal simply by measuring all time differences, t1, t2, t3 … etc. between adjacent peaks. From these one can then calculate the corresponding frequency shifts f1 = 1/t1, f2 = 1/t2, f3 = 1/t3 … etc. The number of particles in each frequency shift range is approximately proportional to the number of measurements with frequencies in that range. This process must take place over a long enough period of time T. The longer is the ‘measurement time’ T, the more ‘peaks’ there are in the signal, the more data points there are, and the higher is the accuracy. As a rule of thumb, T ∼ 1/Δf, where Δf is the required resolution in

Transducer

Transmitter

Oscillator

Receiver

Demodulator

Spectral analyzer

Spectral display

Figure 6.22 Schematic description of the Doppler processing chain.

75

frequency. However, T cannot be made too large. The reason is that the frequency content of the signal changes over time. Thus, for example, during systole, when the velocity of the particles changes rapidly, T must be small enough so that the frequency content of the acoustic Doppler signal remains constant for the period of measurement T. Sometimes, as will be explained below, T is severely limited by other system requirements, as is the case for example in color flow imaging. In this case, the frequency resolution becomes so coarse that a spectral analysis of the signal becomes meaningless. In situations like these, only the average frequency over a given period of time is measured. With the simplistic approach described above, we simply average out f1, f2, f3 … etc. The estimate is quite accurate even for a short measuring period T, as the number of frequency measurements (N = Τ, where ιs the average frequency shift) may still be quite large. In most practical cases, N is of the order of 10. For detailed spectral analysis, much larger values of T (N = 64, 128, or 256) are used.

Spectral analysis In practice, the methods employed for spectral analysis are much more sophisticated and accurate. Spectral analysis is mostly performed using a highly efficient computer coded algorithm, called FFT (fast Fourier transform). Usually there is also a certain overlap between data used for consecutive spectra. This overlap is required in order to update the spectra often enough, in order to follow quick changes such as during systole. As discussed above, spectral Doppler analysis is usually performed at a single Doppler sample volume (‘gate’), for consecutive periods of time 2–10 seconds, in order to include several heart cycles. It is therefore ‘non-real-time’ in nature.

The display of spectral results The information obtained from each spectrum is quite extensive. One actually obtains for each velocity (or Doppler frequency shift) the spectral density function. The spectral density function at each velocity is proportional to the number of blood cells moving at that velocity (and reflecting a sound wave with the corresponding frequency shift). Modern systems usually repeat the calculation every 2 milliseconds. This results in a huge amount of extensive information that must be properly displayed so that it can be easily interpreted by the sonographer. The method commonly accepted in all commercial systems is described below. As illustrated in Figure 6.24, each ‘spectrum’ (the spectral density function as a function of the frequency shift) is displayed as a narrow vertical bar. The height along the vertical axis is proportional to the velocity (or frequency

76

Fetal Cardiology

(a) 1 0.8 0.6 0.4 0.2 0 -0.2 -0.4 -0.6 -0.8 -1 0

20

40

60

80

100

120

140

20

40

60

80

100

120

140

20

40

60

80

100

120

140

(b) 1 0.8 0.6 0.4 0.2 0 -0.2 -0.4 -0.6 -0.8 -1 0 (c) 1 0.8 0.6 0.4 0.2 0 -0.2 -0.4 -0.6 -0.8 -1 0

Figure 6.23 Examples of the transmitted signal (a), the signal from a reflector moving toward the transducer (b), and the demodulated signal (c).

4 140

Figure 6.24 Velocity (arbitrary units)

Spectral density (arbitrary units)

120 100 80 60 40

0

20 0 –4

–3

–2

–1

0

1

Velocity (arbitrary units)

2

3

4 –4

An illustration of a display of a spectrum (right-hand side). The spectrum is displayed as a column of points and is just a different way of displaying the velocity (in the direction of the beam) distribution in the left-hand side of this figure. The point at the bottom represents the lowest velocity whereas the point at the top represents the highest velocity. The entire velocity range in this illustration has been divided into 128 velocities. The vertical position of each of the bars represents its velocity. The brightness of each bar is proportional to the number of blood cells moving at this velocity.

The physics of ultrasound imaging

shift) and the brightness at each point is proportional to the number of particles moving with this velocity. The next spectrum is now displayed next to the previous one, and so on. This is repeated until the whole area that is dedicated to the spectral display is full. When this happens the next spectrum will be displayed in the leftmost position, replacing the ‘old’ spectrum displayed there; the next one will be displayed to its right, and so on.

77

shall concentrate on the hemodynamic information that can be extracted from these frequency shift measurements. In doing so we are giving up the pretension to measure volume flow. The data provided by the Doppler shifts alone is however extensive enough to allow clinical assessment of the hemodynamic system. Obviously, the Doppler data allow a simple recognition of the existence of flow. In the following we shall show how the information embedded in the spectral envelope (maximum frequency shift curve) can be used in the evaluation of the vascular system.

The spectral ‘envelope’ Often, the entire overwhelming amount of information is not actually used for clinical diagnosis. In some cases, only the envelope is required in order to provide the relevant hemodynamic information. The envelope actually describes the velocity of the fastest cluster of cells as a function of time. In order to obtain this function one needs for each spectrum to mark the highest velocity (frequency shift) for which the spectral density function is not zero. It is called the envelope because historically this curve was manually inscribed by the sonographer as an envelope to the spectral display (after the image was frozen). In today’s modern equipment, the envelope is automatically computed online (during the scan). Figure 6.25 is an example of a spectral display describing the flow at a certain point of a carotid artery. The envelope, as provided automatically by the system, is marked in red.

Wave form analysis Note that Doppler frequency shifts do not measure velocity. They are only proportional to velocity. In the following we

Doppler waveform analysis and the Doppler indexes As discussed above, the maximum frequency shift curve (the spectral envelope) represents the temporal changes in the velocities of the fastest moving blood cells during the cardiac cycle. Perhaps the most important feature of the Doppler waveform is its pulsatility. Pulsatile flow, such as is often found in arteries, is characterized by a pulse-like shape with a systolic peak. This is distinctly different from the flat constant velocity waveform usually found in veins. The amount of pulsatility carries a lot of information regarding the vascular system. The pulsatility of the waveform can be represented quantitatively in many different ways according to the vascular property under investigation. The indexes defined below are mostly relevant for evaluating downstream resistance. Several different indexes have been defined in the literature. They are all based on the peak systolic frequency shift (S), the end-diastolic frequency shift (D), and the temporal mean frequency shift over exactly one cardiac cycle (A) (Figure 6.26).

Figure 6.25 Spectral display and envelope.

78

Fetal Cardiology

circulation in 1978. It has been shown that an elevated pulsatility index in the umbilical artery is indicative of increased downstream resistance within the placenta, associated with an obliterative process. The clinical diagnostic value of the method is still questionable due to the following reasons:

1.4 S

Velocity (arbitrary units)

1.2 1 0.8

A

0.6

D

0.4 0.2 0 0

1

2

3

4

5

6

7

8

9

10

Time (arbitrary units)

Figure 6.26 Typical time–velocity waveforms. S is the maximal value of the maximum frequency Doppler shift, D is the minimal value of the maximum frequency Doppler shift, and A is the average of the maximum frequency Doppler shift over a heart cycle.

It has been found that elevated values of these indexes are associated with increased downstream resistance. This has led to a multitude of clinical studies aimed at the development of clinical tools for evaluation of the fetoplacental, uteroplacental, and fetal cerebral circulation. Due to technical limitations, all presently available methods are based on measuring the Doppler indexes in a very small number of sites, typically 1–3. Usually, these sites are conveniently selected in locations where the Doppler signals are relatively easy to acquire, e.g. at the umbilical artery, the uterine artery, or the middle cerebral artery. As a result, only major changes in the waveforms such as reverse or no end-diastolic flow in the umbilical artery or a diastolic notch in the uterine circulation are really clinically significant. Following recent advances in computer and multimedia technologies, it has become possible to perform simultaneous acquisition and analysis of a multitude of Doppler waveforms in a very short period of time. This has stimulated a wealth of new promising ideas and methods that could potentially improve the accuracy of Doppler waveform analysis as a clinical tool for evaluating various vascular systems. The limitations of the present methods and some of the new ideas will be reviewed in the context of fetoplacental circulation.

Umbilical–fetoplacental circulation State of the art Umbilical Doppler waveform analysis was suggested as a non-invasive tool for the assessment of the fetoplacental

1. It has been shown that only when two-thirds of the placental microvasculature is affected will a significant change in the pulsatility index be noticed. 2. Pulsatility values depend also upon additional factors such as the fetal cardiac output, fetal heart rate, fetal cardiac stroke volume, etc. 3. Although all Doppler indexes defined above do not depend explicitly on direction, there still is a residual dependence, mainly due to the high pass filter, originally designed to filter out the arterial wall and tissue signals. When a vessel is approximately perpendicular to the flow direction, slow diastolic flow might be more affected by the filter than the relatively fast systolic flow. In the case of flow in the umbilical artery, the flows at the center of the arteries are always faster than the flows closer to the vessel walls, especially during diastole. One therefore expects a distribution of pulsatility values in the umbilical artery, resulting from variations in the position and orientation of the beam and Doppler sample volume relative to the vessel walls.

Recent developments Tissue Doppler and two-dimensional strain (speckle tracking imaging) One of the most important recent technical developments in the field of Doppler imaging is the invention of tissue Doppler (spectral and imaging). Tissue Doppler is used for the analysis of moving tissues. Although the principles are not different from those used to analyze blood flow, this mode of imaging has been implemented only recently. The main reason is that for tissue Doppler imaging, especially in clinically relevant applications, mainly cardiological, extremely high frame rates are required. Such high frame rates have recently become feasible with the introduction of systems with digital beam forming. These systems allow the required high frame rates by receiving several beams for each beam transmitted. Applications include studies of regional function, diastolic function analysis, cardiomyopathy, cardiac resynchronization imaging, and more. However, Doppler-only-based techniques are limited due to the angle dependence of the signal. As a result, certain myocardial areas are excluded. Until recently, only magnetic resonance imaging (MRI) could provide full, two-dimensional motion analysis through tracking of magnetic tags.

The physics of ultrasound imaging

However, MRI is not widely available for clinical use because it is expensive and time-consuming. Other limitations include relatively low spatial (∼2–5 mm) and temporal resolution (at best ∼30 ms) of the magnetic tags, difficulties in analyzing the whole cardiac cycle due to the short persistence of the tagging, and its inability to analyze beat-to-beat variability. General Electric introduced into the market a new diagnostic tool, similar in concept to MRI tagging, that allows objective analysis of the complete myocardial motion throughout the entire cardiac cycle. Similar in concept to MRI tagging, two-dimensional (2D) strain analyzes motion by tracking ‘tags’ (natural acoustic markers) in the ultrasonic image in two dimensions. These natural markers are used in a way similar to the magnetic tags in MRI. As with tagged MRI, the tags are short-lived; one cannot expect the natural acoustic markers to persist throughout the entire cardiac cycle, mainly due to their movement in and out of the imaging plane. However, unlike in MRI, in which the entire tagging fades out and limits the analysis time to only part of the heart cycle, ultrasound’s new acoustic markers keep coming in as some of the previous markers fade out. This is illustrated in Figure 6.27. Myocardial motion and velocities are then analyzed by calculating frame-to-frame changes. 2D strain is in a way a natural extension of one-dimensional Doppler motion analysis. Similar to one-dimensional Doppler, myocardial motion is characterized in terms of tissue velocity and tissue deformation parameters, such as strain and strain rate. Indications for functional evaluation of the fetal heart include: intrauterine growth restriction, where cardiac failure may be present, ischemia, placental dysfunction/insufficiency, congenital heart defects such as hypertrophic cardiac myopathies, fetal anemia and other causes of hydrops fetalis, fetal arrhythmias, twin-to-twin transfusion

79

syndrome, and other malformations of the fetal cardiovascular system and malformations impacting on cardiac function (e.g. vein of Galen aneurysm), maternal diabetes, and other fetal/maternal problems that may affect the fetal heart.

Machine controls An optimal Doppler signal requires optimization of quite a large number of system parameters. Thus, for example, if the PRF selected is too low, high velocities will result in aliasing. If, on the other hand, the PRF selected is too high, the overall measurement time will be short so that the measurement resolution will be low. As also discussed above, tissue and vessel wall motion need to be filtered out. This again is quite tricky. Too much filtering will also eliminate ‘legitimate’ low velocity blood flow signals. Color flow imaging still presents more problems. The ‘quality’ of the measurement depends on the number of ultrasonic beams that are transmitted in a given direction, which defines at each gate the number of data points that are available for the calculation. Increasing the ‘quality’ will improve the accuracy of the calculation, but will also adversely affect the frame-rate. Table 6.7 below lists the various controls used in the operation of the various Doppler modes. As seen from Table 6.7, the number of controls that need to be optimized is quite large. In order to allow rapid optimization of the gray-scale image, Doppler, and color, most systems will use presets for all clinical applications. Proper use of the presets saves time, reduces button pushing, and ensures diagnostic results. Note, however, that presets are usually designed for average set of conditions in the given application. Fine adjustments are usually for the individual cases.

Figure 6.27 ‘Natural acoustic tagging’. New features (blue circles) keep coming into the image as old ones (yellow circles) fade away.

80

Fetal Cardiology

Table 6.7

The system controls available to the operator for optimizing Doppler measurements

Control

Pulsed wave (spectral) Doppler

Color flow imaging

Total gain

Increases signal received from moving objects of low reflectivity. Noise is also amplified

Increases signal received from moving objects of low reflectivity. Noise is also amplified

Wall motion filter

Decreases the signal from slowly moving objects such as vessel walls or surrounding tissue (irrelevant in tissue Doppler). Effect will increase with higher setting, but signal from slowly moving blood particles will not be observed

Decreases the signal from slowly moving objects such as vessel walls or surrounding tissue (irrelevant in tissue Doppler). Effect will increase with higher setting, but signal from slowly moving blood particles will not be observed

Baseline

Shifting the baseline will increase the measurement range of positive (negative) velocities and reduce the measurement range of negative (positive) velocities

Shifting the baseline will increase the measurement range of positive (negative) velocities and reduce the measurement range of negative (positive) velocities. Not provided by some manufacturers. Irrelevant in power Doppler mode

Sample volume position

Defines the precise location on the anatomical gray-scale image from where Doppler data are sampled

NA

ROI position

NA

Defines the part of the image where blood flow will be imaged

Sample volume size

Increasing the sample volume will increase signal intensity, as long as the sample volume size does not exceed the boundaries of the vessel

NA

Size of the ROI

NA

The size and position of the area on the image where flow will be demonstrated. Increasing the size of the ROI will reduce the frame rate

Pulse repetition frequency (PRF)

The PRF determines the maximum velocity that can be measured without causing aliasing. The maximum PRF is depth range dependent. Aliasing can sometimes be indicated by ‘cropped’ spectral displays

The PRF determines the maximum velocity that can be measured without causing aliasing. The maximum PRF is depth range dependent. Aliasing can sometimes be recognized when colors are mixed within the same vessel

Color ‘quality’

NA

Defines the number of beams transmitted in each direction, for computation of the velocities along the vector. Increasing the ‘quality’ may be required for demonstrating very low flow, but will result in reduced frame rates

ROI, region of interest; NA, not applicable.

Bibliography Atkinson P, Woodcock JP. Doppler Ultrasound and Its Use in Clinical Measurement. London: Academic Press, 1982. Burckhardt CB. Speckles in ultrasound B-mode scans. IEEE Trans Sonics Ultrasonics 1978; 25: 1–6. Castillo E, Lima JA, Bluemke DA. Regional myocardial function: advances in MR imaging and analysis. Radiographics 2003; 23 (Spec No): S127–40. Haberman S, Friedman Z. A new technique for improved diagnosis of local placental abnormalities: Fourier analysis of intraplacental waveforms. Gynecol Obstet Invest 1993; 36: 211–20.

Kremkau WF. Diagnostic Ultrasound: Principles, Instrumentation and Exercises. Orlando: Grune & Stratton, 1984. Leitman M, Lysyansky P, Sidenko S et al. Two-dimensional strain – a novel software for real-time quantitative echocardiographic assessment of myocardial function. J Am Soc Echocardiogr 2004; 17: 1021–9. Maslak S. Computed sonography. In: Sanders R, Hill M, eds. Ultrasound Annual 1985. New York: Raven Press, 1985; 1–16. Moore CC, McVeigh ER, Zerhouni EA. Quantitative tagged magnetic resonance imaging of the normal human left ventricle. Top Magn Reson Imaging 2000; 11: 359–71. Scanlan KA. Sonographic artifacts and their origins. AJR Am J Roentgenol 1991; 156: 1267–72.

The physics of ultrasound imaging

Smith SW, Wagner RF, Sandrick JM, Lopez H. Low contrast detectability and contrast detail analysis in medical ultrasound. IEEE Trans Sonics Ultrasonics 1983; 30: 164–73. Wagner RF, Smith SW, Sandrick JM, Lopez H. Statistics of speckles in ultrasound B scans. IEEE Trans Sonics Ultrasonics 1983; 30: 156–63.

81

Wells PNT. Instrumentation including color flow mapping. In: Taylor KJW, Burns PN, Wells PNT, eds. Clinical Applications of Doppler Ultrasound. New York: Raven Press, 1988; 26–45. Wells PNT, Halliwell M. Speckle in ultrasonic imaging. Ultrasonics 1981; 19: 225–9.

7 Technical advances in fetal echocardiography Boris Tutschek and David Sahn

Introduction Structural congenital heart disease (CHD) is among the most frequently missed anomalies in prenatal ultrasound studies. Prenatal detection can significantly improve perinatal outcome at least in certain types of CHD.1–4 Various technical advances in ultrasound systems have promised improvements specifically for fetal echocardiography, but prenatal detection rates vary considerably, and depend on factors such as operator training and experience.5–9 Screening for fetal heart disease and detailed diagnosis of fetal heart disease require technologies of different complexity.10 This chapter attempts to review recent and imminent technical advances that have impacted or may affect fetal echocardiography. Technical improvements in transducer technology and image processing as well as a basic description of three-dimensional (3D) technology are complemented with an outlook on the techniques for studying fetal cardiac mechanics.

that the wave fronts they generate converge at the level of the selected depth (or focus). This electronic focusing has been used in one-dimensional array transducers over many years. However, it affects the image only in the plane of the ultrasound elements. Perpendicular to the plane of the ultrasound elements the focus zone cannot be changed in one-dimensional arrays: the ‘focus’ of the beam perpendicular to the long axis of the elements is fixed. Recently, two-dimensional matrix array transducers that can overcome this limitation have become available. Using matrix array transducers, focusing perpendicular to the long axis of the transducer is possible, avoiding artifacts from structures immediately adjacent to the insonated plane. Figure 7.2 shows a phantom with cystic objects insonated with either a conventional one-dimensional or a new two-dimensional array transducer capable of electronic focusing in two planes.

Improved acquisition, processing, and display Electronic focusing and two-dimensional matrix array transducers Lateral resolution describes the minimal distance between neighboring objects in the plane of the ultrasound beam that can be resolved. Resolution depends on the distance between adjacent transducer elements and also on the lateral width of the ultrasound beam (slice thickness) used to interrogate a certain depth. Figure 7.1 shows an ultrasound phantom insonated with good and poor lateral discrimination. Focusing of the ultrasound beam can be achieved by electronic ‘steering’: adjacent elements are activated at different time points so

(a)

(b)

Figure 7.1 Lateral resolution of 0.3-mm nylon threads in a phantom insonated with good (a) and poor (b) lateral resolution (image kindly provided by H Dudwiesus, GE Healthcare).

84

Fetal Cardiology

(a)

(b)

Figure 7.2 Improved resolution using a two-dimensional matrix array transducer (b) versus a conventional, one-dimensional array transducer (a). Using a matrix transducer, focusing is possible also perpendicular to the transducer’s longitudinal axis. The small cystic structures in the phantom are resolved better and over a greater depth field (image kindly provided by H Dudwiesus, GE Healthcare).

Tissue harmonic imaging Lower frequency ultrasound signals have a better tissue penetration. Decreasing the insonation frequency, however, also lowers image resolution. Tissue penetration of ultrasound waves also depends on tissue properties such as tissue density, pressure, temperature, and others; emitted sound waves themselves also exceed pressure and, therefore, change as they travel through tissue, are reflected, and are received. The further the ultrasound waves travel through the tissue, the more their waveform changes from the emitted pure sinus wave to a compound waveform, consisting of the original and of additional waveforms of smaller amplitude, but higher frequency. These lower amplitude, higher frequency waves occur at multitudes of the insonation (fundamental) frequency and are called

harmonic frequencies.11 Signals recorded at twice the insonation frequency are called first harmonic frequencies. Harmonic imaging removes the fundamental frequencies using filters and typically shows a narrower main lobe, yielding better axial and lateral resolution, as well as better side-lobe suppression.12 Tissue harmonic imaging can improve imaging of fluid-filled structures, providing better border definition and contrast as well as decreasing artifacts. Harmonic imaging can be implemented without increasing power output. Tissue harmonic imaging (THI) shows a significant improvement in gray-scale imaging, especially in difficult scanning conditions as for example in obese patients.13,14 THI improved the resolution in half the pregnant patients studied by Treadwell et al,15 in particular in obese women. Kovalchin et al16 compared fundamental and harmonic imaging in a group of fetuses from mothers, 71% of whom had difficult scanning conditions. Image quality and visualization of the ventricles, valves, and the aortic and ductal arches were better using harmonic imaging. The authors concluded that harmonic imaging improved the image quality and that it was a useful adjunct to fundamental imaging in fetal echocardiography. Paladini et al17 performed a detailed study of harmonic imaging in fetal echocardiography. They studied 50 women in three groups receiving fetal echocardiography with either fundamental or harmonic imaging. In general the diagnostic results were equal, but resolution from harmonic imaging was poorer, leading the authors to the conclusion that in pregnant women of normal weight, fundamental imaging is still the technique of choice for fetal echocardiography. In obese women and those with inadequate imaging in fundamental imaging, however, harmonic imaging performed better (‘rescue’ modality).

Real-time compound imaging In conventional ultrasound imaging, the ultrasound beam is emitted at right angles to the elements. If this beam hits the interface between structures of different echogenicity (a reflector) at a 90° angle, most echoes are returned to the transducer. At other angles reflections are diverted away from the transducer, reducing the received signal intensity. In the worst case, when the beam is parallel to the interface, little or no reflection occurs. Real-time spatial compound imaging uses electronic beam steering of a transducer array to acquire several overlapping scans of an object from different view angles (Figure 7.3). Individual views of the same plane, but acquired from slightly different angles (up to nine different angles are being used), are compounded into a multiangle image in real-time. Reflections or signal losses occurring in particular on cystic structures or irregular shaped borders are reduced using compound imaging (CI; for an example of compound imaging of the fetal heart see Figure 7.4). The actual frame rate, however, drops, because views are acquired from

Technical advances in fetal echocardiography

different angles, then compounded (computed) before the image can be displayed. CI has been marketed under various names (e.g. Philips: SonoCT; GE Healthcare: CrossX Beam™ imaging). CI reduces ultrasonic artifacts such as speckle, promising improved contrast resolution and tissue differentiation.18 In a clinical study of breast tissue, for example, CI increased signal-to-noise ratio, but also increased the apparent target width.19 A study of compound and

85

harmonic imaging in fetal diagnostic scanning at 11–14 weeks’ gestation reported improved visualization in various anatomical regions for the combination of compound and harmonic imaging. For the fetal heart there was a small (but probably not significant) improvement over conventional B-mode versus compound plus harmonic imaging.20 Compound imaging alters contrast ratios,21 similar to using higher frequency transducers,22 which must be kept in mind when qualitative assessment, for example of fetal bowel echogenicity or cardiac valve or myocardium appearance, is to be based on subjective contrast impression using harmonics and compound imaging. In addition, averaging multiple frames for CI invariably reduces the frame rate, limiting its practical use.

Speckle reduction

(a)

(b)

Figure 7.3 Conventional ‘single line of sight’ (a) and real-time compound imaging (b). By electronically steering the ultrasound beams to ‘look’ at objects from different angles, true reflectors can be better differentiated from artifacts because their reflections appear consistently in several of the angled views.

(a)

Speckles are artifactual ultrasound signals caused by the interference of reflected ultrasound energy from scatters that are too small and too close to be resolved with the frequency used. Speckles appear as a pseudo-structure in histologically homogeneous tissue due to wave interference of reflected signals.23 Speckle reduces the spatial and contrast resolution in ultrasound images and results in an artificial ‘sono-texture’ that becomes apparent when highest magnifications are used (Figure 7.5). A prominent speckle structure in a diagnostic image can obscure true objects with little contrast to the neighboring tissue. To reduce speckles, techniques for (1) improved resolution, including higher frequency transducers, coded excitation, matrix-array transducers, and harmonic imaging, (2) temporal averaging and spatial compounding, and (3) postprocessing approaches, involving different types of filters,

(b)

Figure 7.4 A normal fetal heart (at 23 weeks’ gestation) imaged without (a) and with compound imaging (b) (but otherwise with identical settings).

86

Fetal Cardiology

have been employed.23 Speckle reduction algorithms aim to remove the distracting speckle pattern without reducing the detail in the ultrasound image (Figure 7.6). The techniques of geometric filtering to reduce speckle were first applied to radar images and later also to ultrasound images.24 Post-processing algorithms for speckle reduction have been introduced under different names, for

example, Philips: XRES, ContextVision: GOPView;25 GE Healthcare: speckle reduction imaging (SRI).23 It should be mentioned that, using harmonic and also compound imaging, small structures may appear different in size from fundamental imaging, a fact that has been noted as well regarding other subtle anatomical details such as nuchal translucency or smallest distance measurements. In an ultrasound phantom designed to measure distance below 1 mm, fundamental imaging combined with speckle reduction, but without compound and harmonic imaging, showed the most accurate distance measurements (Figure 7.7).26

The use of colors in B-mode imaging

Ultrasound speckles. Interference of the reflected ultrasound waves at the level of the resolution limit causes an artifactual pattern appearance (see alternating white and dark areas indicated by the arrows) without histological correlate in the tissue.

The physiology of human visual perception involves two types of receptors in the retina. The rods are specialized for perception of low light intensities, the so-called scotopic vision, providing sharp vision in low light surroundings. Photopic vision is performed by the cones, which are the dominant receptors in the fovea centralis where they provide highest resolution perception. Rods typically resolve between 20 and 60, under ideal circumstances up to 250, gray levels, but cones, utilizing color, saturation, and intensity, enable differentiation of up to seven million colors in photopic vision.27 So-called ‘photopic ultrasound imaging’ was available on a commercial ultrasound system also used for fetal studies (Elegra; Siemens Medical Ultrasound, Erlangen, Germany). In this implementation, the wide dynamic range available from raw ultrasound data

(a)

(b)

Figure 7.5

Figure 7.6 Speckle reduction. Post-processing algorithms in modern ultrasound systems reduce speckles by real-time image analysis and interposition of gray tones to ameliorate the speckled appearance. (a) High magnification shows the artifactual pattern of the tissue (speckles); (b) speckle reduction applied to the identical still image.

Technical advances in fetal echocardiography

a

1

b

87

c

1

Figure 7.7 14 MHZ THI

14 MHZ nativ Abb 14

1 L 0.03 cm

1 L 0.02 cm

was transformed in real time from the conventional shades of a gray-scale (scotopic) into a photopic image, enabling much finer delineation over a wide dynamic range. Photopic imaging has been studied in internal medicine, but no fetal studies have been published. This ultrasound system providing photopic imaging is now no longer in production, but color to enhance the visual perception, for example monochromatic coloring of B-mode images, is available in many ultrasound systems. Rendered 3D images are often produced using a colored display. In one commercial system (Aplio; Toshiba Europe Medical Systems) there is a palette using several colors in substitution for the conventional simple gray-scale. Another way of exploiting colors to display ultrasound data has been implemented for three-dimensional ultrasound in a new commercial ultrasound system (iE33; Philips Medical Systems) to enhance the depth perception: ‘dynamic colorization’ is a 3D display modality coloring elements in the foreground (‘closer to the viewer’) differently from those in the background. Video clips 7.1 and 7.2 show the application of dynamic colorization for fetal 3D studies.28

Advanced methods for motion detection B-flow Conventional Doppler techniques in vascular imaging tend to exaggerate the real size of a vessel (‘bleeding’ of the color). In addition, the morphological information detected and displayed by gray-scale B-mode imaging may be lost by the overlying color or power Doppler signals. B-mode imaging can be extended to detect blood flow independent of Doppler-derived signals by using digitally encoded ultrasound. An ultrasound beam is encoded in

10 MHZ THI

Measuring accuracy of fundamental and tissue harmonic imaging in an ultrasound phantom with membranes 0.3 mm apart. (a) Fundamental imaging at 14 MHz, (b, c) harmonic imaging at 14 and 10 MHz (reproduced with permission from reference 26).

two separate beams: the reflection of the first is used to reconstruct the cross-sectional imaging, while the reflections of the second, amplified beam are analyzed for the moving blood particles. Both signals are displayed in the same gray-scale image. When compared to fetal Doppler ultrasound, B-flow has higher resolution and frame rates, and disadvantages of Doppler-based flow detection such as aliasing and signal dropout at orthogonal scanning angles are avoided. B-flow alone or in combination with spatiotemporal image correlation (STIC) allows visualization of fine small vessels with low velocity, such as pulmonary veins, offering a potential for the detection of small cardiac vessels, such as total anomalous pulmonary venous return.29,30 Volpe et al31 showed that B-flow in combination with STIC was superior to 2D gray-scale and color Doppler alone in the detection of small pulmonary vessels (major pulmonary collateral arteries), which may crucially aid in diagnosing complex cardiac malformations with pulmonary vascular involvement.

Directional power Doppler The reflected signal from Doppler insonation contains different information that can be used in various display modalities. The frequency shift of the reflected signals indicates the velocity of the reflectors, for example the cells in the moving blood, whereas the amplitude of the reflected signal correlates with the reflected energy. Traditionally, color Doppler has been used to encode motion direction, for example using red and blue denoting flow towards and away from the transducer and velocity encoded in the brightness of the color signal. Power Doppler, however, displays the amplitude component of the reflected signals. The combination of both yields a sensitive motion display with directional information. Heling et al32 studied ‘advanced dynamic flow (ADF)’, a sensitive

88

Fetal Cardiology

flow modality based on power Doppler imaging with directional information for fetal studies including fetal echocardiography. They concluded that ADF was superior to conventional color Doppler vascular imaging with higher resolution, good lateral discrimination, and higher sensitivity, providing flow information almost in B-mode image quality.

Navigation and display of ultrasound volume data Reconstructed fetal 3D echocardiography Reconstructed 3D echocardiography currently is the dominant clinical technique for fetal 3D echocardiography. A sequence of sonographic data from a series of crosssectional images is acquired while an automated probe sweeps the ultrasound plane across the fetal heart. Offline reconstruction then generates a virtual cardiac cycle comprising data from multiple heartbeats which are rearranged using an algorithm that composes volume data from multiple cross-sectional images33 (for review see reference 34). Clinical systems commercially available today use STIC for gray-scale and color Doppler,35,36 but the reconstructed nature of the volume leaves room for motion artifacts. Other approaches have also been studied (for reviews see references 37 and 38), but are not available yet for broad clinical use. STIC was first introduced into a commercially available ultrasound system by Kretz Ultrasound (now GE Healthcare), and is today incorporated into other

(a)

ultrasound systems (Philips, Medison). Volume data from STIC displayed as multiplanar imaging, reconstruction of 3D views (so-called ‘rendering’39,40), the combination with color Doppler, tomographic, and inversion modes,36,41–44 and the ability to quantify fetal cardiac volumes45–47 have provided fascinating insights into and very graphic three-dimensional representation of the fetal heart. STIC is described extensively in Chapter 14. Therefore, we will only briefly describe the display modalities available from volume echocardiographic data. Three-dimensional sequences of the fetal heart can be displayed in various forms. One (or several) cross-sectional plane(s) can be placed anywhere within a volume, enabling interactive scrolling through the heart offline, either in the plane of acquisition (highest spatial resolution) or in any other (reconstructed) plane. Multiplanar imaging describes a display with three orthogonal planes (Figure 7.8). If a whole cardiac cycle has been reconstructed, all structures can be viewed at different times in the cardiac cycle. Inspecting adjacent cross-sectional planes in a volume scan is particularly useful for examination of the great arteries. Using two dimensions only, the outflow tracts can be visualized by moving the transducer toward the fetal head to obtain cross-sections, and by rotation and slight angulation to obtain views along the vessels. This may, however, be difficult due to an unfavorable fetal position, movements, or lack of operator experience.48 In contrast, a stored volume (sequence) can be manipulated digitally, overcoming some of these limitations. Because of the given anatomical position and relation of the normal structures, views of the great arteries can be derived virtually from the standard four-chamber view, provided that the volume covers these structures. Failure to be able to do so may even hint at structural heart disease. For example, if a fetal

(b) Diaphragm

RV

RV

RA

MS

MS LV

Pulm.v LA Pulm.v dAo

Left lung Stomach

MS

RA μs PV

Figure 7.8 Multiplanar imaging of a normal fetal heart at 23 weeks’ gestation. (a) The display shows three orthogonal cross-sections with four-chamber view (top left), short-axis view (top right), and en face view of the interventricular septum (IVS). (b) Annotated image with the intersection of all three planes represented by the red dot; grey shaded area: IVS; RV/RA, right ventricle/atrium; LV/LA, left ventricle/atrium; pulm.v, pulmonary vein(s) PV, pulmonary valve; IAS, interatrial septum (see also corresponding Video clip 7.3)

Technical advances in fetal echocardiography

cardiac volume has been acquired in the four-chamber view insonation, cross- or longitudinal sections of great arteries can be extracted from the data set following a simple algorithm (e.g. the ‘spin’ technique48). Typical 2D echocardiography aligns its standard planes with one of the ventricular axes, while primarily digital imaging modalities (magnetic resonance imaging (MRI) and computed tomography (CT)) display structures in relation to standard anatomical orientations (sagittal, coronal, transverse) and in parallel planes or cross-sections. Volume echocardiography can combine the conventional 2D imaging and orientation standards. In tomographic imaging several parallel cross-sectional planes through the same volume are displayed similar to a typical CT or MRI study (Figure 7.9 and Video clip 7.4). Tomographic displays of ultrasound volume data have been given different commercial names (‘multislice’, ‘tomographic

–3

89

ultrasound imaging/TUI’, ‘iSlice’). Tomographic imaging of 3D fetal echocardiographic data can elegantly display normal or structurally abnormal planes from one insonation angle (Figure 7.10 and Video clip 7.5). Tomographic imaging has successfully been applied to the heart at different gestations and to CHD.49,50 A semiautomatic algorithm helps to display the relevant cardiac structures in one or several tomographic or multiplanar panels.51–53 In the current release of the Voluson ultrasound system (GE Healthcare), an automated approach (‘volume computeraided diagnosis’, VCAD) is now available commercially.54 Rendering modes are an alternative to the cross-sectional modes. Rendering displays either external or internal surfaces of organs from volume data. Using particular settings for display thresholds, apparent tissue transparency, and shading techniques, three-dimensional representations can be displayed on a two-dimensional computer

–2

7.0 mm –1

2

1

3

Figure 7.9 Tomographic display of a fetal volume containing parallel cross-sections of the upper fetal abdomen and thorax. Panel –3 shows the fetal stomach; –2: hepatic veins converging toward inferior vena cava (IVC); –1: IVC passing through the diaphragm; *: indicates the level of the four-chamber view; 1: origin of the aorta; 2: origin and branching of the main pulmonary artery; 3: cross-section of the head and neck vessels in the upper mediastinum. The top left image shows a sagittal section orthogonal to the other views, indicating the spatial relationship of the horizontal sections (see also Video clip 7.4).

90

Fetal Cardiology

Normal 4CV

Normal 4CV

Ao arises from LV and above 4CV

PA originates cranially and from RV

(a)

PA arises from LV and above 4CV

Ao ventrally and from RV (wide arch)

(b)

Figure 7.10 Tomographic imaging of a normal heart (a) and a dextrotransposition of the great arteries (b). For the same gestational age, identical settings of slice distances at three different levels at and above the four-chamber view (4CV) display the diagnostic sections (see also Video clip 7.5). Ao, aorta; LV, left ventricle; PA, pulmonary artery; RV, right ventricle.

monitor or in a printed image. Surface rendering was originally developed to display the outer surfaces of solid objects such as the fetal face or the skeleton in three dimensions.39,55 For rendered images of the fetal heart, cropping into the heart will display its internal surfaces,40,56 including cutting the heart in halves at the level of the four-chamber view, or looking from an en face view of the valve plane, generating a more plastic impression than the thin slice usually displayed in a conventional B-mode image (Figure 7.11). Rendering in inversion mode shows only fluid-filled spaces (Figure 7.11f), generating ‘digital casts’ of the fetal heart and vasculature in both normal and structurally abnormal cases.42,43,57,58 STIC can be combined with color and power Doppler, generating structural and functional cardiac information and angiography-like images.30,36,59,60 Due to its high sensitivity for low velocities, power Doppler can also demonstrate extracardiac structures such as pulmonary vasculature, teratoma, and chorioangioma.61,62 B-flow combined with STIC as well as these rendering techniques provides even more detailed visualization of small normal and aberrant vessels.31,42 The most recent extension to the color rendering modes from STIC is the color inversion mode (HD11 XE; Philips Medical Systems), which combines the anatomic detail of

fluid-filled structures in inversion mode with the dynamic information of color Doppler (Figure 7.12).

Semiautomatic quantification Functional and quantitative analysis of the fetal heart is yet another promising new area of volumetric fetal echocardiography research. Technically different 3D technologies have been applied to measure cardiac ventricular volumes and masses.45–47,63–67 The first report on 3D measurement of fetal cardiac ventricles by Chang et al,63 using a fast automated sweep, demonstrated a linear increase of the cardiac volume between 20 and 30 weeks and that 3D volumetry had a better reproducibility than 2D measurements. Initial experiments using non-gated 3D mode from free-hand acquisition confirmed the correlation between ventricular chamber volumes and gestational age.64,65 In vitro experiments using STIC demonstrated acceptable accuracy for volume and even mass estimations in the range comparable to mid- and late-gestation fetal hearts.45,46 3D inversion mode sonography combined with STIC represents another and possibly more reproducible method for estimating fetal cardiac ventricle volume.66 Live 3D

Technical advances in fetal echocardiography

(a)

91

(b)

Figure 7.11

(c)

(d)

(e)

(f)

mode using newer transducer technologies, i.e. twodimensional matrix array transducers, holds the promise of measurements in non-reconstructed data sets. Preliminary data show that non-reconstructed fetal cardiac volumetry and STIC technology exhibit reasonable concordance between these two measurement approaches.67

Cross-sectional and rendered views of a normal fetal heart at 29 weeks’ gestation (all images derived from a spatiotemporal image correlation (STIC) volume acquired in a four-chamber view plane). Cross-sections: (a) B-mode, (b) color duplex with biventricular inflow (in red); rendered images: (c) surface-rendered view, cropped to display only the ‘cranial half’ of the heart, (d) color-only rendering of the whole volume with display of hepatic veins and inferior vena cava (blue) and ventricular inflow, (e) ‘glass-body’: region of interest as in (d) but with transparent tissue rendering over color Doppler, (f) inversion mode (thin volume slice cropped to display only the four-chamber plane) (reproduced with permission from reference 28).

Finally, semiautomated virtual casting of fetal heart using live 3D ultrasound data algorithms have been developed that can expand ‘seeds’ placed into the cardiac cavities in a 3D volume automatically, performing both a segmentation and a volumetric analysis of cardiac chambers (Deng, personal communication).

92

Fetal Cardiology

(a)

(b)

Figure 7.12 Color inversion mode (normal fetal heart, 29 weeks). Full cardiac volume rendering (viewed cranially) in diastole (a) and systole (b) (reproduced with permission from reference 28).

Live three-dimensional ultrasound using matrix array transducers Matrix array transducer Transducers with more elements and broader frequency bandwidths improve both resolution and tissue penetration. A major goal in transducer technology has been the development of an electronic two-dimensional array, enabling full beam steering in all three dimensions.68

Sparse array matrix transducer The first 2D array transducer available was a sparse 2D array (Volumetric Medical Imaging Inc., Durham, NC, 1997).69–71 Sklansky et al72 were the first to evaluate a (sparse) array matrix transducer for real-time threedimensional fetal echocardiography to evaluate fetal cardiac anatomy and function in 10 human fetuses (four with congenital heart disease). The system used displayed the volumetric data as a series of four simultaneous planes. They concluded that fetal real-time three-dimensional echocardiography is a feasible, facile, and rapid new technique. Another group also used this 3D scanner

system for real-time 3D fetal echocardiography prospectively in 13 fetuses at mid-term, comparing 3D with 2D visualization rates. While standard planes could be visualized more easily in three dimensions, image quality was inferior to 2D mode.73 Quantification of cardiac ventricles was also attempted using such a system (Figure 7.13).74 Routine application of the initial devices to fetal echocardiography remained limited; while coarse volumetric assessment of the fetal cardiac ventricles was feasible, resolution even in the third trimester was generally insufficient for structural diagnosis.

Full matrix array transducer Recently, a full 2D array transducer with some 2800 individual elements, miniaturized multiplexers built into the probe handle and capable of generating in excess of 20 B-mode volumes per second, and even color duplex volumes, has been introduced (Philips Medical Systems, 2001). Maulik et al75 used such a 4-MHz 2D array transducer in 12 fetuses between 16 and 37 weeks. They concluded that the system allowed comprehensive visualization of fetal cardiac anatomy and color Doppler flow unattainable by two-dimensional approaches, and suggested that live 3D fetal echocardiography could be a significant tool for prenatal diagnosis and assessment of congenital heart disease.

Technical advances in fetal echocardiography

93

Figure 7.13 Real-time fetal 3D echocardiography using the first generation two-dimensional matrix array (Volumetrics, 2.5-MHz matrix transducer). Fetal echocardiography at 25 weeks’ gestation shows the acquisition orientation (four-chamber view on the top right, a section perpendicular to it (through the aortic outflow) in the lower right, and two perpendicular images from the elevation plane (cross-sections through the two ventricles in short axis)), (reproduced with permission from reference 28).

Sklansky et al55 studied 30 fetuses with this matrix system, focusing on rendered images of the fetal heart in comparison to conventional cross-sectional imaging. Volume-rendered displays identified all major abnormalities, and enabled offline retrieval of views not visualized during the actual scans; volume-rendered displays had only slightly inferior image quality compared with conventional two-dimensional images. Acar et al76 reported their experience with bi-plane (simultaneous display of two different cross-sections, e.g. simultaneous display of both outflow tracts) and live 3D (rendered) imaging with the same system in 60 fetuses between 22 and 34 weeks. In pathologic fetal hearts, 3D was helpful, for example for localizing multiple cardiac tumors, estimating size and function of the right and left ventricles, and evaluating mechanism of valvular regurgitation and pulmonary obstruction. At least three other transducers have been introduced for adult (3 MHz; Philips Medical Systems and Vingmed/GE Healthcare) or pediatric (7 MHz; Philips Medical Systems) volumetric imaging. Realtime 3D echocardiography should provide an accurate means of determining chamber volumes and cardiac mass.77 Preliminary data using these systems show that they can be used for fetal cardiac volumetry (Figures 7.14 and 7.15).67 It remains to be seen how they can be applied to the diagnosis of structural fetal cardiac disease.

Cardiac mechanics Motion detection, contractility Motion of the fetal heart as a whole and of its individual regions, as well as intracardiac blood flow, can be studied to describe the cardiac mechanical function globally or

locally. Sonographic measurement of cardiac function can be based on ventricular dimensions (fractional shortening calculated from M-mode or B-mode imaging to estimate the ejection fraction or from 3D measurement, as described above) or on velocities of intracardiac blood flow or cardiac structures (tissue Doppler echocardiography). Blood flow Doppler studies provide data from velocity waveform analysis across the atrioventricular valves, in the great arteries, or in the veins close to the heart; quantitative data include acceleration times, mechanical cardiac intervals, or computation of cardiac output from flow velocities and vessel diameters (for a recent review see reference 78).

Tissue Doppler Fetal tissue Doppler echocardiography (TDE) has been used to assess regional diastolic and systolic wall excursions and for functional assessment of the fetal heart.79–84 Figure 7.16 and the accompanying Video clip 7.6 show color tissue Doppler imaging of a normal fetal heart. Quantitative information can be obtained by using either color or pulsed wave Doppler TDE, also enabling diagnosis of fetal cardiac rhythm disturbances.84–86 Figure 7.17 shows an example of normal and abnormal fetal cardiac rhythm. The accompanying Video clip 7.7 shows a normal cardiac pulsed wave TDE study.

Strain and strain rate, speckle tracking Because overall tissue velocities alone have not been sufficiently sensitive in fetal cardiac studies, Di Salvo et al87

94

Fetal Cardiology

Figure 7.14 Real-time fetal 3D echocardiography using a fully electronically steerable two-dimensional matrix array transducer (x3-1/iE33 and software QLab 3DQ Advanced; Philips Medical Systems, Bothell, WA). The ultrasound images represent three cross-sections orthogonal to each other and a 3D niche view of a normal 26-week fetal heart.

LV

Septum um

pt

Se

Apical view

LV

Cranial

Apex

Cranial Lateral view

Base

Diastole

Systole

Figure 7.15 Virtual endocardial casts of the normal human left cardiac ventricle in a 28-week fetus in diastole and systole and endocardial volume curves, generated using semiautomatic volume calculation software from a volume loop acquired using real-time 3D echocardiography (QLab 3DQ Advanced and x3-1/iE33; Phillips Medical Systems, Bothell, WA).

Technical advances in fetal echocardiography

studied regional myocardial deformation (strain and strain rate) using tissue Doppler in 75 normal human fetuses. They described the characteristics of ventricular filling, and found an increase of longitudinal deformation with gestational age. Larsen88 obtained tissue Doppler sequences to calculate fetal strain and strain rate in three normal

+10.7

–10.7 cm/s

Figure 7.16 Color tissue Doppler image of a normal fetal heart (reproduced with permission from reference 83; see also Video clip 7.6).

35 175 55 95 110

12

95

fetuses and one fetus with aortic valve atresia, showing severely reduced strain rates in the affected ventricle. Doppler-based measurements, however, suffer from the limitation of angle dependency. In addition, quantitative analysis based on Doppler data is hampered by measurement errors and by its angle dependency. Strain rate imaging by speckle tracking imaging (STI) is less affected by these factors, and is the only sonographic measurement that allows rotational or twist analysis of the left ventricle of the heart. STI exploits the sonographic texture of the cardiac walls in a two-dimensional image sequence of the beating heart. Individual segments of the beating heart, i.e. regions along the myocardial walls in long- or short-axis views, can be tagged and tracked by a semiautomatic image analysis algorithm due to their ‘speckled’ appearance (Figure 7.18). The deformation (or stretching) of tissue, normalized to its original size or shape, is called cardiac strain. The rate at which this tissue deformation occurs is called the strain rate.89 The higher is the strain rate, the faster the deformation occurs between individual points in the myocardium, or, in other words, the faster the myocardium contracts or relaxes. Strain rate imaging has been used in various adult cardiac conditions including ischemic heart disease and myocardial dyssynchrony (for an extensive review see reference 90). STI can be used to assess longitudinal strain in the human fetus91 (see Video clip 7.8). One intriguing

Systole S‘

6 cm/s 95.00ms 0.00ms 0.00ms/s

E‘ A‘ Diastole 1sec

(a)

Diastole

A‘ E‘

Systole (b)

Extrasystole

S‘

Figure 7.17 (a) Normal pulsed wave (PW) tissue Doppler (TDE) tracing of the right ventricular valve annulus and (b) PW TDE showing an extrasystole (reproduced with permission from reference 83). The systolic (S’) and diastolic (E’ and A’) peaks can be seen and the segments of the mechanical action of the heart can be measured in milliseconds.

96

Fetal Cardiology

Figure 7.18 Speckle tracking uses the artifactual, but region-specific pattern generated by wave interference to track smallest tissue regions (magnified from a cross-sectional image sequence throughout the cardiac cycle).

aspect of the complex architecture of the left ventricle is the ability to produce torsion of the ventricle, similar to wringing a towel dry. This fundamental component of left ventricular function is caused by the helical arrangement of myocardial layers.92 STI can also be used to study left ventricular torsion in the human fetus91 (see Video clip 7.9). Prospective studies are required to establish reproducibility of fetal STI and possible uses.

Future developments Automated 3D/4D studies of ventricular volumes and chamber mechanics It is likely that in the near future the ability to acquire and compute and measure 3D/4D images of the fetal heart will continue to improve. Automated and tailored methods for edge detection should substantially improve the ability to integrate cardiac volume measurements into the clinical realm of fetal echocardiography. Likewise, high frame rate 3D/4D acquisition of high resolution for speckle cluster mechanics should yield methods for defining cardiac mechanics, strain, twist, and torsion measurements of the fetal ventricles, and the assessment of right and right ventricular interaction taking into account the effect of through-plane motion. A break point will occur that clearly sets different data format and analysis formats for fetal cardiac imaging, as distinct from the other 3D methods used in perinatology.

A bifurcation in the pathway for matrix array implementation It is likely that matrix array development will proceed in two directions: for transabdominal imaging of the fetal

Figure 7.19 Capacitive micromachined ultrasound transducers (CMUTs) (scanning electron microsope image).

heart, larger aperture, more complex, or even curved plane matrix arrays will be developed; for denser acquisitions, high-resolution fetal cardiac imaging will correct aberrations which occur along the imaging pathway. The growing interest and understanding of the potential for changing the natural history of heart disease with prenatal cardiac interventions will lead to developments of miniaturized devices for intrauterine applications. These will provide 4D near-field imaging and potentially offer integrated therapeutic options such as radiofrequency, laser, or high-intensity focused ultrasound therapies. They will incorporate the latest non-ceramic

Technical advances in fetal echocardiography

ultrasound sensors – such as capacitive micromachined ultrasound transducers (CMUTs) (Figure 7.19) – which have the agility for controlling frequency and power output93 far beyond what exists today.

Conclusion Recent technological ultrasound advances have increased the image quality available for fetal cardiology. The possibilities of volume ultrasound, in particular together with offline processing, have pushed the field forward tremendously. Currently, STIC is the clinically dominant 3D modality, but continuing advances in array technology will eventually lead to non-reconstructed, truly live 3D imaging. The next breakthrough in this sense will be the introduction of full aperture matrix transducers ideally suited for transabdominal scanning. Pushing the limits at another front, highest-frequency transvaginal probes, together with the general trend toward earlier prenatal diagnosis, are becoming available now, and their use in fetal echocardiography is imminent. It can be expected that these developments will be accompanied by further automation, including detection algorithms that enable tracking of tissue borders, automatic assessment of relaxation and contraction, and automatic volume and mass quantification and flow measurements. Despite all the technological improvements, the main factor likely to improve the detection rate of fetal cardiac defects remains the operator’s ability to obtain diagnostic views. Especially offline image or volume analysis methods and novel ways of analyzing and presenting fetal echocardiography data will improve the general level of scanning techniques and personal expertise.

Legends for the DVD Video clip 7.1 Dynamic colorization. Dynamic colorization refers to the differential coloring of fore- and background objects in threedimensional rendering. In this example of a real-time 3D study of a normal fetus at 16 weeks’ gestation the fetal legs and umbilical cord in the foreground are colored in amber while the structures further away from the viewer are shaded in a blue. (Reproduced with permission from reference 28.)

Video clip 7.2 High-resolution real-time three-dimensional view of a beating normal fetal heart (22 weeks’ gestation). Using a two-dimensional matrix array transducer, instantaneous or real-time threedimensional scanning with sufficiently high spatial and temporal resolution is possible.

97

Video clip 7.3 Multiplanar reconstruction (MPR) of a virtual cardiac cycle acquired using spatial-temporal image correlation (STIC). In this video three orthogonal sections through a looping virtual cardiac cycle are shown; the intersection of planes is denoted by little yellow or red dots. The top left panel (A plane) is the acquisition plane, adjusted to display the four-chamber view, the top right panel B the short axis view of the ventricles and the bottom left panel C displays the interventricular septum en face.

Video clip 7.4 Tomographic imaging of a virtual cardiac cycle (from a STIC acquisition). A stack of sections parallel to the four-chamber view is displayed from a virtual cardiac cycle (captured using STIC). By adjusting the center of the volume in the middle panel and selecting the appropriate distance between the sections several relevant planes can be displayed dynamically and at once (top left panel: short axis view displaying the “slice” levels). (Reproduced with permission from reference 28.)

Video clip 7.5 Tomographic imaging of a normal heart (left panel) and a heart with d-transposition of the great arteries (TGA, right panel; reduced playback speed). Both hearts are displayed using the same settings to demonstrate the striking differences in analogous planes. (Reproduced with permission from reference 28.)

Video clip 7.6 Color tissue Doppler of a normal fetal heart at mid-trimester. By reducing the wall filter, pulse repetition frequency and gain the tissue motion is shown in color. Note the synchronicity of the atria and ventricles, the foramen ovale flap and how the septum moves with the free wall of the right ventricle.

Video clip 7.7 Pulsed-wave tissue Doppler study of a normal fetal heart. The fetal heart is imaged in an apical four chamber view and the PW sample volume placed to cover the excursion of the lateral part of the tricuspid valve annulus. The wall filter and pulse repetition frequencies as well as the gain are lowered and the sweep speed increased. The resulting PW tracing shows the tissue excursion with the typical two-peak diastolic and the systolic (E’ and A’, movement away from transducer, negative velocities) and the systolic peak (S’, towards the transducer, positive velocity). Arrhythmias show distinctly altered typical patterns, and timing of the mechanical segment of the cardiac cycle can be measured from the PW tracing. (Reproduced with permission from reference 84.)

Video clip 7.8 Longitudinal strain in the fetus measured using speckle tracking in an apical four-chamber view. The colors in the top left panel indicate longitudinal shortening and lengthening (tissue deformation, strain). The graph on the right shows longitudinal strain over time during one cardiac cycle.

Video clip 7.9 Application of speckle tracking to measure circumferential strain of the fetal right ventricle in a short axis view. The colors in the top left panel indicate circumferential shortening and lengthening (tissue deformation, strain). The graph on the right shows circumferential strain over time during one cardiac cycle.

98

Fetal Cardiology

References 1.

2.

3.

4.

5.

6.

7. 8.

9.

10. 11. 12.

13.

14.

15.

16.

17.

Crane JP, LeFevre ML, Winborn RC et al. A randomized trial of prenatal ultrasonographic screening: impact on the detection, management, and outcome of anomalous fetuses. The RADIUS Study Group. Am J Obstet Gynecol 1994; 171: 392–9. Bonnet D, Coltri A, Butera G et al. Detection of transposition of the great arteries in fetuses reduces neonatal morbidity and mortality. Circulation 1999; 99: 916–18. Tworetzky W, McElhinney DB, Reddy VM et al. Improved surgical outcome after fetal diagnosis of hypoplastic left heart syndrome. Circulation 2001; 103: 1269–73. Fuchs IB, Muller H, Abdul-Khaliq H et al. Immediate and long-term outcomes in children with prenatal diagnosis of selected isolated congenital heart defects. Ultrasound Obstet Gynecol 2007; 29: 38–43. Sharland GK, Allan LD. Screening for congenital heart disease prenatally. Results of a 2 1/2-year study in the South East Thames Region. Br J Obstet Gynaecol 1992; 99: 220–5. Carvalho JS, Mavrides E, Shinebourne EA et al. Improving the effectiveness of routine prenatal screening for major congenital heart defects. Heart 2002; 88: 387–91. Simpson LL. Screening for congenital heart disease. Obstet Gynecol Clin North Am 2004; 31: 51–9. Tegnander E, Eik-Nes SH. The examiner’s ultrasound experience has a significant impact on the detection rate of congenital heart defects at the second-trimester fetal examination. Ultrasound Obstet Gynecol 2006; 28: 8–14. Tegnander E, Williams W, Johansen OJ, Blaas HGK, Eik-Nes SH. Prenatal detection of heart defects in a nonselected population of 30 149 fetuses – detection rates and outcome. Ultrasound Obstet Gynecol 2006; 27: 252–65. Ville Y. Fetal imaging: a brief history of the future. Ultrasound Obstet Gynecol 2007; 29: 1–5. Forsberg L. Ultrasonic biomedical technology; marketing versus clinical reality. Ultrasonics 2004; 42: 17–27. Fowlkes JB, Averkiou M. Contrast and tissue harmonics imaging. In: Goldman LW, Fowlkes JB, eds. Categorial Course in Diagnostic Radiology Physics: CT and US Cross-sectional Imaging. Oak Brook, IL: RSNA, 2000: 77. Shapiro RS, Wagreish J, Parsons RB et al. Tissue harmonic imaging sonography: evaluation of image quality compared with conventional sonography. AJR Am J Roentgenol 1998; 171: 1203–6. Tranquart F, Grenier N, Eder V, Pourcelot L. Clinical use of ultrasound tissue harmonic imaging. Ultrasound Med Biol 1999; 25: 889–94. Treadwell MC, Seubert DE, Zador I, Goyert GL, Wolfe HM. Benefits associated with harmonic tissue imaging in the obstetric patient. Am J Obstet Gynecol 2000; 182: 1620–2. Kovalchin JP, Lewin MB, Bezold LI, Altman CA, Ayres NA. Harmonic imaging in fetal echocardiography. J Am Soc Echocardiogr 2001; 14: 1025–9. Paladini D, Vassallo M, Tartaglione A et al. The role of tissue harmonic imaging in fetal echocardiography. Ultrasound Obstet Gynecol 2004; 23: 159–64.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29. 30.

31.

32.

33.

Entrekin RR, Porter BA, Sillesen HH et al. Real-time spatial compound imaging: application to breast, vascular, and musculoskeletal ultrasound. Semin Ultrasound CT MR 2001; 22: 50–64. Dahl JJ, Soo MS, Trahey GE. Clinical evaluation of combined spatial compounding and adaptive imaging in breast tissue. Ultrason Imaging 2004; 26: 203–16. von Kaisenberg CS, Kuhling-von Kaisenberg H, Fritzer E et al. Fetal transabdominal anatomy scanning using standard views at 11 to 14 weeks’ gestation. Am J Obstet Gynecol 2005; 192: 535–42. Lee HJ, Cho JY. Quantitative assessment of fetal bowel echogenicity: comparison of harmonic, compound, and fundamental sonographic images. J Clin Ultrasound 2003; 31: 302–7. Vincoff NS, Callen PW, Smith-Bindman R, Goldstein RB. Effect of ultrasound transducer frequency on the appearance of the fetal bowel. J Ultrasound Med 1999; 18: 799–803. Milkowski A, Yadong L, Becker D, Ishrak SO. Speckle reduction imaging. 2003, GE Medical Systems. www. gehealthcare.com/euen/ultrasound/docs/education/ whitepapers/whitepaper_SRI.pdf, accessed June 2007. Busse LJ, Crimmins TR, Fienup JR. A model based approach to improve the performance of the geometric filtering speckle reduction algorithm. Presented at 1995 IEEE Ultrasonics Symposium, 95CH35844, Seattle, WA, 7–10 November, 1995: 1353–6. Floreby L. GOPView™ US-Ultrasound image enhancement in real-time. 2005. www.contextvision.com, accessed June 2007. Wunsch R, Dudwiesus H, Reinehr T. [Prospective comparison of different ultrasound modalities to measure thicknesses less than 1 mm]. Fortschr Röntgenstr 2007; 179: 65–71. [in German] Gebel M, Bleck J, Schmitt KJ, Breitkopt P, Sutcliffe P. Photopic Ultrasound Imaging – Ein adaptives Verfahren basierend auf unserer Sehphysiologie. Electromedica 2000; 68: 59–64. Tutschek B, Sahn DJ. Three-dimensional echocardiography for studies of the fetal heart: present status and future directions. Cardiol Clin 2007; 25: 341–55. Pooh RK. New application of B-flow sono-angiography in perinatology. Ultrasound Obstet Gynecol 2000; 15: 163. Yagel S, Cohen S, Shapiro I, Valsky DV. 3D and 4D ultrasound in fetal cardiac scanning: a new look at the fetal heart. Ultrasound Obstet Gynecol 2007; 29: 81–95. Volpe P, Campobasso G, Stanziano A et al. Novel application of 4D sonography with B-flow imaging and spatio-temporal image correlation (STIC) in the assessment of the anatomy of pulmonary arteries in fetuses with pulmonary atresia and ventricular septal defect. Ultrasound Obstet Gynecol 2006; 28: 40–6. Heling KS, Chaoui R, Bollmann R. Advanced dynamic flow - a new method of vascular imaging in prenatal medicine. A pilot study of its applicability. Ultraschall Med 2004; 25: 280–4. Nelson TR, Pretorius DH, Sklansky M, Hagen-Ansert S. Three-dimensional echocardiographic evaluation of fetal heart anatomy and function: acquisition, analysis and display. J Ultrasound Med 1996; 15: 1–9.

Technical advances in fetal echocardiography

34. 35.

36.

37.

38. 39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

Deng J, Rodeck CH. New fetal cardiac imaging techniques. Prenat Diagn 2004; 24: 1092–103. DeVore GR, Falkensammer P, Sklansky MS, Platt LD. Spatio-temporal image correlation (STIC): new technology for evaluation of the fetal heart. Ultrasound Obstet Gynecol 2003; 22: 380–7. Chaoui R, Hoffmann J, Heling KS. Three-dimensional (3D) and 4D color Doppler fetal echocardiography using spatio-temporal image correlation (STIC). Ultrasound Obstet Gynecol 2004; 23: 535–45. Deng J, Rodeck CH. Current applications of fetal cardiac imaging technology. Curr Opin Obstet Gynecol 2006; 18: 177–84. Sklansky M. Advances in fetal cardiac imaging. Pediatr Cardiol 2004; 25: 307–21. Riccabona M, Pretorius DH, Nelson TR, Johnson D, Budorick NE. Three-dimensional ultrasound display modalities in obstetrics. J Clin Ultrasound 1997; 25: 157–67. Yagel S, Benachi A, Bonnet D et al. Rendering in fetal cardiac scanning: the intracardiac septa and the coronal atrioventricular valve planes. Ultrasound Obstet Gynecol 2006; 28: 266–74. Goncalves LF, Espinoza J, Lee W, Mazor M, Romero R. Three and four-dimensional reconstruction of the aortic and ductal arches using inversion mode: a new rendering algorithm for visualization of fluid-filled anatomical structures. Ultrasound Obstet Gynecol 2004; 24: 696–8. Goncalves LF, Espinoza J, Lee W et al. A new approach to fetal echocardiography. Digital casts of the fetal cardiac chambers and great vessels for detection of congenital heart disease. J Ultrasound Med 2005; 24: 415–24. Espinoza J, Goncalves LF, Lee W Mazor M, Romero R. A novel method to improve prenatal diagnosis of abnormal systemic venous connections using three- and four-dimensional ultrasonography and ‘inversion mode’. Ultrasound Obstet Gynecol 2005; 25: 428–34. Goncalves LF, Lee W, Espinoza J, Romero R. Examination of the fetal heart by four-dimensional (4D) ultrasound with spatiotemporal image correlation (STIC). Ultrasound Obstet Gynecol 2006; 27: 336–48. Bhat AH, Corbett V, Carpenter N et al. Fetal ventricular mass determination on threedimensional echocardiography: studies in normal fetuses and validation experiments. Circulation 2004; 110: 1054–60. Bhat AH, Corbett VN, Liu R et al. Validation of volume and mass assessments for human fetal heart imaging by 4-dimensional spatiotemporal image correlation echocardiography: in vitro balloon model experiments. J Ultrasound Med 2004; 23: 1151–9. Messing B, Valsky DV, Cohen SM et al. 3D inversion mode combined with STIC: a novel technique for fetal heart ventricle volumetry in congenital heart disease. Ultrasound Obstet Gynecol 2006; 28 (Suppl): 397. DeVore GR, Polanco B, Sklansky MS, Platt LD. The ‘spin’ technique: a new method for examination of the fetal outflow tracts using three-dimensional ultrasound. Ultrasound Obstet Gynecol 2004; 24: 72–82. Paladini D, Vassallo M, Sglavo G, Lapadula C, Martinelli P. The role of spatio-temporal image correlation (STIC) with tomographic ultrasound imaging (TUI) in the sequential

50.

51.

52.

53.

54.

55.

56.

57.

58.

59.

60.

61.

62.

63.

64.

99

analysis of fetal congenital heart disease. Ultrasound Obstet Gynecol 2006; 27: 555–61. Goncalves LF, Espinoza J, Romero R et al. Four-dimensional ultrasonography of the fetal heart using a novel Tomographic Ultrasound Imaging display. J Perinat Med 2006; 34: 39–55. Abuhamad A. Automated multiplanar imaging: a novel approach to ultrasonography. J Ultrasound Med 2004; 23: 573–6. Abuhamad AZ, Falkensammer P. Automated sonography: defining the spatial relationships of standard diagnostic fetal cardiac planes in the second trimester of pregnancy. Ultrasound Obstet Gynecol 2006; 28: 359–411. Espinoza J, Kusanovic JP, Goncalves LF et al. A novel algorithm for comprehensive fetal echocardiography using 4 dimensional ultrasonography and tomographic imaging. J Ultrasound Med 2006; 25: 947–56. Reichartseder F, Falkensammer P. VCAD Volume-aided diagnosis. An automated approach to visualize standard views of the fetal heart. GE Healthcare, Zipf, Austria. http://brochureorder.geultrasound.net/xtcommerce/ pdfs/300-06-U026E.pdf, accessed June 2007. Sklansky MS, DeVore GR, Wong PC. Real-time 3-dimensional fetal echocardiography with an instantaneous volume-rendered display: early description and pictorial essay. J Ultrasound Med 2004; 23: 283–9. Vinals F, Pacheco V, Giuliano A. Fetal atrioventricular valve junction in normal fetuses and in fetuses with complete atrioventricular septal defect assessed by 4D volume rendering. Ultrasound Obstet Gynecol 2006; 28: 26–31. Ghi T, Cera E, Segata M et al. Inversion mode spatiotemporal image correlation (STIC) echocardiography in three-dimensional rendering of fetal ventricular septal defects. Ultrasound Obstet Gynecol 2005; 26: 679–86. Lee W, Goncalves LF, Espinoza J, Romero R. Inversion mode: a new volume analysis tool for 3-dimensional ultrasonography. J Ultrasound Med 2005; 24: 201–7. Goncalves LF, Romero R, Espinoza J et al. Fourdimensional ultrasonography of the fetal heart using color Doppler spatiotemporal image correlation. J Ultrasound Med 2004; 23: 473–81. Yagel S, Valsky DV, Messing B. Detailed assessment of fetal ventricular septal defect with 4D color Doppler ultrasound using spatio-temporal image correlation technology. Ultrasound Obstet Gynecol 2005; 25: 97–8. Ruano R, Benachi A, Aubry MC et al. Prenatal diagnosis of pulmonary sequestration using three-dimensional power Doppler ultrasound. Ultrasound Obstet Gynecol 2005; 25: 128–33. Sciaky-Tamir Y, Cohen SM, Hochner-Celnikier D et al. Three-dimensional power Doppler (3DPD) ultrasound in the diagnosis and follow-up of fetal vascular anomalies. Am J Obstet Gynecol 2006; 194: 274–81. Chang FM, Hsu KF, Ko HC et al. Fetal heart volume assessment by three-dimensional ultrasound. Ultrasound Obstet Gynecol 1997; 9: 42–8. Meyer-Wittkopf M, Cole A, Cooper SG, Schmidt S, Sholler GF. Three-dimensional quantitative echocardiographic assessment of ventricular volume in healthy human fetuses and in fetuses with congenital heart disease. J Ultrasound Med 2001; 20: 317–27.

100

65.

66.

67.

68.

69.

70.

71.

72.

73.

74.

75.

76.

77.

78. 79.

Fetal Cardiology

Esh-Broder E, Ushakov FB, Imbar T, Yagel S. Application of free-hand three-dimensional echocardiography in the evaluation of fetal cardiac ejection fraction: a preliminary study. Ultrasound Obstet Gynecol 2004; 23: 546–51. Messing B, Cohen SM, Valsky DV et al. Fetal cardiac ventricle volumetry in the second half of gestation assessed by 4D ultrasound using STIC combined with inversion mode. Ultrasound Obtet Gynecol 2007; 30: 142–51. Tutschek B, Hui L, Robertson PA et al. Fetal cardiac ventricular volumes derived from 3D/4D echo: definitive data from two different types of 3D ultrasound systems. J Am Coll Cardiol 2007; 49 (Suppl A): 254A. Light ED, Davidson JO, Fiering TA, Hruschka TA, Smith SW. Progress in 2-D arrays for real-time volumetric imaging. Ultrason Imag 1998; 20: 235. Stetten G, Ota T, Ohazama C et al. Real-time 3D ultrasound: a new look at the heart. J Cardiovasc Diagn Procedures 1998; 15: 73–84. Ota T, Fleishman CE, Strub M et al. Real-time, threedimensional echocardiography: feasibility of dynamic right ventricular volume measurement with saline contrast. Am Heart J 1999; 137: 958–66. Shiota T, Jones M, Chikada M et al. Real-time threedimensional echocardiography for determining right ventricular stroke volume in an animal model of chronic right ventricular volume overload. Circulation 1998; 97: 1897–900. Sklansky MS, Welson T, Strachan M et al. Real-time three-dimensional fetal echocardiography: initial feasibility study. J Ultrasound Med 1999; 18: 745–52. Scharf A, Geka F, Steinborn A et al. 3D real-time imaging of the fetal heart. Fetal Diagn Ther 2000; 15: 267–74. Tutschek B, Buck T, Reihs T et al. Real-time threedimensional fetal echocardiography. Ultrasound Obstet Gynecol 2000; 16: 53–4. Maulik D, Nanda NC, Singh V et al. Live three-dimensional echocardiography of the human fetus. Echocardiography 2003; 20: 715–21. Acar P, Dulac Y, Taktak A et al. Real-time three-dimensional fetal echocardiography using matrix probe. Prenat Diagn 2005; 25: 370–5. Rusk RA, Mori Y, Davies CH et al. Comparison of ventricular volume and mass measurements from B- and C-scan images with the use of real-time 3-dimensional echocardiography: studies in an in vitro model. J Am Soc Echocardiogr 2000; 13: 910–17. Simpson J. Echocardiographic evaluation of cardiac function in the fetus. Prenat Diagn 2004; 24: 1081–91. Twining P. Myocardial motion imaging: a new application of power color flow and frequency-based color flow

80. 81.

82.

83.

84.

85.

86.

87.

88.

89.

90. 91.

92. 93.

Doppler in fetal echocardiography. Ultrasound Obstet Gynecol 1999; 13: 255–9. Harada K, Tsuda A, Orino T et al. Tissue Doppler imaging in the normal fetus. Int J Cardiol 1999; 71: 227–34. Paladini D, Lamberti A, Teodoro A et al. Tissue Doppler imaging of the fetal heart. Ultrasound Obstet Gynecol 2000; 16: 530–5. Pacileo G, Paladini D, Russo MG et al. Echocardiographic assessment of ventricular filling pressure during the second and third trimesters of gestation. Ultrasound Obstet Gynecol 2000; 16: 128–32. Tutschek B, Zimmermann T, Buck T, Bender HG. Fetal tissue Doppler echocardiography: detection rates of cardiac structures and quantitative assessment of the fetal heart. Ultrasound Obstet Gynecol 2003; 21: 26–32. Tutschek B, Schmidt KG. Utility of tissue Doppler echocardiography (TDE) in analysing fetal arrhythmias. Ultrasound Obstet Gynecol 2004; 24: 229. Rein AJJT, Levine JC, Nir A. Use of high-frame rate imaging and Doppler tissue echocardiography in the diagnosis of fetal ventricular tachycardia. J Am Soc Echocardiogr 2001; 14: 149–51. Rein AJJT, O’Donnell C, Geva T et al. Use of tissue velocity imaging in the diagnosis of fetal cardiac arrhythmias. Circulation 2002; 106: 1827–33. Di Salvo G, Russo MG, Paladini D et al. Quantification of regional left and right ventricular longitudinal function in 75 normal fetuses using ultrasound-based strain rate and strain imaging. Ultrasound Med Biol 2005; 31: 1159–62. Larsen LU, Petersen OB, Norrild K et al. Strain rate derived from color Doppler myocardial imaging for assessment of fetal cardiac function. Ultrasound Obstet Gynecol 2006; 27: 210–13. D’hooge J, Heimdal A, Jamal F et al. Regional strain and strain rate measurements by cardiac ultrasound: principles, implementation and limitations. Eur J Echocardiography 2000; 1: 154–70. Stoylen A. Strain rate imaging. http://folk.ntnu.no/ stoylen/strainrate/, accessed June 2007. Tuschek B, Hui L, Robertson PA et al. Fetal cardiac ventricular volumes derived from 3D/4D echo: definitive data from two different types of 3D ultrasound systems. J Am Coll Cardiol 2007; 49(Suppl A): 254A. Thomas JD, Popovic ZB. Assessment of left ventricular function by cardiac ultrasound. JACC 2006; 48: 2012–25. Yeh DT, Oralkan O, Wygant IO, O’Donnell M, Khuri-Yakub BT. 3-D ultrasound imaging using a forward-looking CMUT ring array for intravascular/ intracardiac applications. IEEE Trans Ultrason Ferroelect Freq Contr 2006; 53: 1202–11.

8 Epidemiology of congenital heart disease: etiology, pathogenesis, and incidence Julien IE Hoffman

Introduction Congenital heart disease (CHD) is, by definition, cardiovascular disease present at birth. Most CHD is due to gross structural developmental cardiovascular anomalies such as septal defects, stenosis or atresia of valves, hypoplasia or absence of one or other ventricle, or abnormal connections between great vessels and the heart. A few children are born with arrhythmias (mainly conduction defects), and hypertrophic or dilated cardiomyopathy, although these usually present later in childhood or adulthood. Even though asphyxial heart disease is present at birth, it is not included as CHD. As defined, CHD is one of the most common serious congenital anomalies, occurring in up to 2% of liveborn children, and in an even higher percentage of fetuses.1,2 Cardiovascular development involves a series of complex processes, each component of which has to occur at the right time under the orchestration of a cascade of genes and gene products.3–5 Many of these have been found in studies of chick and mouse heart development, some have been identified in humans, but vast numbers are yet to be discovered. In general, the further upstream (closer to initiation of development) a gene is, the more its malfunction will affect major cardiac architecture. The formation of the heart begins with the differentiation of two specialized groups of cells – the primary and secondary heart fields.6–11 The mammalian heart develops partly from the primary heart tube formed by myocardial and endothelial cells that have differentiated from splanchnic mesoderm and will form the atria and left ventricle. Anterior to these cells are myocardial precursors that form the secondary or anterior heart field, and cells from this field develop into the outflow tract and right ventricle. The two cell populations fuse early in development, but remain clonally distinct. Integration of the secondary and primary fields depends in part on the influence of cells in the neural crest.12,13

This fused heart precursor begins as a straight tube that contains in sequence from the caudal end segments that will develop into the atria, the left ventricle, the right ventricle, and the truncus arteriosus. The tube elongates and twists into a d-loop, so that the right ventricle moves to the right side of the left ventricle and the normal asymmetry of the heart is initiated. Recently, studies have revealed that the earliest asymmetry occurs in Hensen’s node, and is produced by the unidirectional swirling of the monocilia that dip from the nodal cells into the extracellular embryonic fluid.14–16 The proteins that drive the cilia – kinesin and dynein – are subject to mutations that can alter normal asymmetry. This asymmetrical rotation affects some early expressed genes (for example, lefty, nodal, fibroblastic growth factor 8, zic3); faulty expression of these genes will produce dextrocardias, l-loops, situs abnormalities (heterotaxies), and complex heart disease.17,18 After normal d-looping has occurred, the future atria are still connected to the left ventricle, and this is connected to the right ventricle which leads to the truncus arteriosus, the precursor of the future aorta and main pulmonary artery. The next major developmental changes are the formation of the primary atrial and ventricular septa, and the movement of the atrioventricular and semilunar valve rings. These changes allow the right and left atria to connect to the right and left ventricles, respectively, and the ventricles to communicate with their respective great arteries. Coordinating these changes are genes involved in cell migration and the formation of the extracellular matrix, so that their malfunction produces gross structural anomalies indicating developmental arrest at a primitive stage, such as double inlet left ventricle (single ventricle), double outlet right ventricle, and truncus arteriosus.12,19 These abnormalities are in part related to abnormal function of the endocardial cushions, and other abnormalities of these cushions lead to the relatively gross distortions of architecture found in atrioventricular septal (endocardial cushion) defects.

102

Fetal Cardiology

Given the importance of precisely orchestrated spatial and temporal gene expression in cardiovascular development, it is not surprising that specific genetic abnormalities are being found in increasing numbers in CHD. These abnormalities may take the form of chromosomal rearrangements, microdeletions (both of which affect many genes), or isolated point mutations. In addition, there are also well known environmental causes of CHD, and some of these may act by interfering with the action of specific genes. Finally, with the complexity of cardiovascular development, it would not be surprising to find that non-specific factors and random errors in cell migration produce various defects or influence the severity of others.

Chromosomal defects Chromosomal defects vary from gross abnormalities such as trisomies or monosomies to deletion syndromes and microdeletions that involve contiguous genes. About 0.30–2.27% (median 0.67%) of all live births are associated with chromosomal defects; these are usually the aneuploidies: trisomy 21, 18, and 13 and the sex polysomies (e.g. XXX and XYY) that increase with maternal age, and monosomy X that is inversely related to maternal age.1 These and other less common abnormalities are listed in Table 8.1. Infants with these abnormalities are survivors of a much larger cohort of fetuses with chromosomal defects; in first trimester fetuses aborted for non-medical reasons, 2.6–6.4% (median 4.5%) have chromosomal defects. Therefore, only about 15% of fetuses with chromosomal defects survive to birth. Among spontaneous abortions, the incidence of chromosomal abnormalities is much higher, ranging from 15.7 to 69.8% (median 46.6%). The chances of survival to birth depend on the defect. It is about 1% for trisomy 21, 18, and 13, and for X monosomy, but as high as 80% for the sex polysomies (Table 8.2); fortunately, sex polysomies do not have an increased incidence of CHD. One other aspect of chromosomal anomalies is important for the echocardiographer. The percentage of aborted fetuses with chromosomal anomalies is greatest in the youngest fetuses, and decreases to 6–15.9% (median 11.7%) beyond 20 weeks’ gestational age. Therefore, the earlier the echocardiographic study is done, the greater is the chance of finding a chromosomal abnormality in a fetus with CHD. These considerations indicate the desirability of karyotyping all early fetuses found to have CHD. Because all these chromosomal defects except for sex polysomies have a very high association with CHD (Table 8.1), the incidence of CHD in general depends on how many of these fetuses are conceived by older mothers and how many of the affected fetuses reach term alive. As a group, chromosomal defects account for about 6% of all

CHD in liveborn infants,21 but this figure could change dramatically depending on increased survival of the affected fetuses on the one hand or the frequency of therapeutic abortion on the other.

Genetic abnormalities Chromosomal abnormalities involve excesses or deficiencies of multiple genes. On the other hand, single gene defects are also common, and currently account for at least 3% of all CHD. This figure is likely to grow rapidly. Some gene mutations characteristically cause defects in more than one organ system and produce recognizable syndromes. It is likely that the genes responsible for these syndromes are general or ‘upstream’, and function early in development to affect several systems. For example, the Holt–Oram syndrome, due to mutations in the Tbx5 transcription factor, is associated with left-sided defects including septal defects, atrioventricular septal defects, and hypoplastic left heart syndrome, and abnormalities of the radius or thumb.22 The Noonan syndrome, about half due to a mutation of the PDPN11 gene on chromosome 12, shows characteristic body habitus and facies as well as dysplastic pulmonary valves and hypertrophic cardiomyopathy.23 Some of the better known genetic syndromes with CHD are listed in Table 8.3; a more detailed list is published by Burn.24 Other genes, however, are specific or ‘downstream’, and function later in development to affect only one organ or part of one organ. Many gene mutations have been found in CHD, usually not as part of a general syndrome.5 None of these genetic defects to date are very frequent, nor should we expect them to be. In the development of any part of the cardiovascular system there is usually a cascade of genes and gene products involved. Mutation in any one of the members of the cascade would cause a specific cardiac defect, and there is no reason why one particular gene should always be involved. Almost certainly, some genes will be more susceptible to mutations than will other genes in the cascade, so that some mutations might be expected to be more common than others. Nevertheless, it is likely that any given defect can arise from mutations in one of many genes; multiple genotypic abnormalities may converge on a similar phenotype. In addition, there are gene mutations with major effects that are modified by other genes with subtle differences in timing or level of expression that will influence the phenotype.28 This makes screening for genetic causes of CHD complicated and expensive. In general, patients with one congenital abnormality often have abnormalities of other systems. For example, abnormalities of the genitourinary tract occur in about 30% of patients with congenital heart disease, and patients with tetralogy of Fallot often have omphaloceles.29–31 These are not usually regarded as syndromes, and it is not

Epidemiology of congenital heart disease

Table 8.1

103

Chromosomal defects and congenital heart disease (CHD). Based on reference 20

Chromosomal defect

Incidence/1000 live births Percentage with CHD

Predominant types of CHD

Trisomies 21 (Down syndrome)

1–1.5

50–60

AVSD, VSD

18 (Edward syndrome)

0.2–0.3

95

VSD, PDA

13 (Patau syndrome)

0.1–0.2

90

VSD, ASD

Duplications 3q26–27 duplication (Cornelia de Lange)

VSD

4p

10–15

5p (Opitz)

10

8

20

9p

Low

10q

50

11p

Low

12p (Pallister–Killian; Fryn)

25

22p duplication (cat eye)

VSD VSD+, coarctation, PDA, ASD, AS, absent pericardium TAPVC, ToF

Monosomy X (Turner)

0.1–0.2

50

Coarctation of the aorta; aortic stenosis or bicuspid aortic valve

4p− (Wolf–Hirshhorn)

50

ASD

4q−

60

VSD, PDA, PPS, AS, tricuspid atresia, ASD, coarctation, ToF

5p− (cri-du-chat)

30

Variable

9p− (CHARGE)

30–50

11q− (Jacobsen)

60

13q−

50

18p−

10

18q−

Low

20p11− (Alagille)

High

Deletion syndromes

VSD, PDA, PS

PPS

Microdeletion syndromes 22q11 (DiGeorge: CATCH-22); Shprintzen (velocardiofacial) 7q11.23 (Williams) 16p13.3 (Rubinstein–Taybi)

Aortic arch anomalies, interrupted aortic arch, truncus arteriosus, ToF High 25

Supravalvar AS, PPS PDA, VSD, ASD

AS, aortic stenosis; ASD, atrial septal defect (secundum); AVSD, atrioventricular septal defect (endocardial cushion or atrioventricular canal defect); PDA, patent ductus arteriosus; PS, pulmonic stenosis (valvar); PPS, peripheral pulmonic stenosis; TAPVC, total anomalous pulmonary venous connection; ToF, tetralogy of Fallot; VSD, ventricular septal defect.

104

Fetal Cardiology

Table 8.2 Survival of fetuses with chromosomal defects. Reproduced with permission from reference 1 Group

Aneuploidy Sex chromosomes

Autosomes

Monosomy Polysomy Trisomy Liveborn

Polyploidy Euploidy Miscellaneous Percentage total Total pregnancies aberrations

17

127

107

1

205

38

Spontaneous abortion

4434

31

7601

2848

489

330

Total

4451

158

7708

2849

694

368

Percentage survival

0.38

80.4

1.4

0.0

29.5

10.3

0.71

70 000

52.4

30 000 1 00 000

3.1

Table 8.3 CHD caused by gene mutations (syndromic and non-syndromic). Based on data from references 24–27. In many of these disorders the specific gene mutations have been identified Syndrome

Chromosome

CHD

Noonan

12q24.1, 12p12.1

Pulmonic stenosis; hypertrophic cardiomyopathy

Apert

10q26

VSD, coarctation of aorta, PS

Holt–Oram

12q24.1

ASD+, VSD, AVSD, truncus arteriosus

Ellis–van Creveld

4p16

Single atrium, AVSD

Marfan

15q21

Dilatation and rupture of aorta; aortic or mitral regurgitation

Ehlers–Danlos type IV

2q31–32.3

Arterial rupture

Osteogenesis imperfecta 17, 7

Aortic root dilatation, aortic or mitral incompetence

Pseudoxanthoma elasticum

MVP, coronary arterial disease, restrictive cardiomyopathy

16p3.1

Mucopolysaccharidoses 4p16.3; 5q1–13; Xq27.3

Aortic or mitral incompetence; coronary artery narrowing

Hypertrophic cardiomyopathy

Asymmetric ventricular hypertrophy due to a large variety of mutations in sarcomeric proteins: myosin heavy and light chains, actin, tropomyosin, titin, troponin, caveolin, and myosin binding protein C

1q32, 2q24.3, 3p, 3p25, 7q31, 11p11.2, 12q23–24, 14q12, 15q14, 15q22.1, 19q13.4, 20q13.3

Dilated cardiomyopathy Xp21.2, Xq28, 1q21, 1q32, 1q42–43, Dilated (congestive) cardiomyopathy due mainly to abnormal cytoskeletal but also sarcomeric proteins: dystrophin, tafazzin, 2q31, 2q35, 5q33, 6q22.1, 10q212.3–q23,2, 10q22–q23, 11p11, lamin A/C, cardiac troponin T, α-actinin, titin, desmin, δ-sarcoglycan, phospholamban, metavinculin, myosin binding 11p15.1, 14q12, 15q14, 15q22 protein C, muscle LIM protein, β-myosin heavy chain, cardiac actin, α-tropomyosin Non-compaction of left Xq28, 18q12.1–12.2, 11p15 ventricle

Cardiomyopathy

Osler–Rendu–Weber

9q34.1; also 5q31.3, 12q11

Pulmonary arteriovenous fistulae

Long QT

QT1–11p15.5, QT2–12p11.1, QT3–3p21, QT4–4q25–27, QT5–21q22.1, QT6–21q22.1

Long QT interval, arrhythmias, sudden death

Supravalvar AS

7q11.23

Supravalvar AS, PPS

AS, aortic stenosis; ASD, atrial septal defect (secundum); AVSD, atrioventricular septal defect (endocardial cushion or atrioventricular canal defect); MVP, mitral valve prolapse; PDA, patent ductus arteriosus; PS, pulmonic stenosis (valvar); PPS, peripheral pulmonic stenosis; VSD, ventricular septal defect.

Epidemiology of congenital heart disease

clear whether these associations stem from a common genetic origin or from some other disturbance during fetal development. Much has been written about multifactorial inheritance in CHD24,32 and in cardiovascular disease in general.33,34 The concept is that certain lesions may be due to the interaction of several genes (polygenic inheritance), the outcome being modulated by environmental factors. This combination of events, for example, could explain why CHD could be genetically caused without manifesting classical Mendelian genetic frequencies. For example, if one parent (especially the mother) has had CHD, there is an increased risk for CHD in the children, but this risk is usually no more than 5–10%, unlike the 50% risk found for autosomal dominant disease or the 25% risk associated with recessive disease. Criteria needed to indicate a multifactorial model have been described by Burn.24 Recurrence risk to sibs and offspring is approximately the square root of the population incidence. The risk to sibs is comparable to the risk to offspring. The only form of CHD shown to fit the multifactorial model well is the patent ductus arteriosus (Zetterqvist, cited in reference 24) If this is true, then, how do we explain the other forms of CHD? One possible answer lies in experiments reported by Kurnit and colleagues.35,36 During endocardial cushion development, certain cell adhesion molecules such as platelet endothelial cell adhesion molecule are downregulated when endocardial cells undergo mesenchymal transformation. The quintessential lesion related to endocardial cushion defects, the atrioventricular septal defect, is particularly common in trisomy 21. Studies have shown that in trisomy 21 fibroblasts cultured from the lung have abnormally increased adhesiveness. Kurnit and colleagues developed a computer model of embryological development in which they programmed differing degrees of adhesiveness, random migration, and certain rules for when migration and cell division would cease. With normal cell adhesiveness, the atrioventricular canal region developed normally in their model. With abnormally increased adhesiveness, some of the atrioventricular canals were abnormally formed, just as in an atrioventricular canal defect. What was important in their study was that not all the canals were abnormal. Therefore the abnormality produced in their model was due to the abnormal adhesiveness, but its expression depended in part upon random events in cell migration. Thus, failure to fit classical Mendelian genetics is not an argument against a genetic cause of CHD.

Environmental factors CHD has been associated with several environmental toxic or infectious factors (Table 8.4). We do not know whether most of these act by affecting gene expression directly or

105

Table 8.4 Environmental causes of CHD. Data taken from references 24, 37, and 38 Environmental factor

CHD

Frequency of CHD (%)

Rubella virus

PDA, PPS, PS, ASD, VSD

> 35

Mumps

Endocardial fibroelastosis

Lithium

Mitral and tricuspid incompetence, Ebstein syndrome, ASD

Diabetes in pregnancy

Outflow tract lesions, especially TGA, coarctation of the aorta

Alcohol

VSD, ASD, ToF

Retinoic acid

Conotruncal anomalies

3–5

25–70

Phenylketonuria ToF, VSD, coarctation, PDA, SV, ASD

25–50

Trimethadione

TGA, ToF, HLH

15–30

Phenytoin

PS, AS, PDA, coarctation of the aorta

Systemic lupus erythematosus

Complete heart block

Coumadin

PDA, PPS

Thalidomide

Truncus arteriosus, ToF, septal defects, coarctation, PS, TGA, TAPVC

2–3 20–40

5–10

AS, aortic stenosis; ASD, atrial septal defect (secundum); AVSD, atrioventricular septal defect (endocardial cushion or atrioventricular canal defect); HLH, hypoplastic left heart syndrome; DA, patent ductus arteriosus; PS, pulmonic stenosis (valvar); PPS, peripheral pulmonic stenosis; SV, single ventricle; TAPVC, total anomalous pulmonary venous connection; TGA, complete transposition of the great arteries; ToF, tetralogy of Fallot; VSD, ventricular septal defect.

by blocking the action of the gene product. None of them are known to affect the genome itself. Phenylketonuria, a genetic defect itself, affects the fetus through the increased maternal blood levels of phenylalanine and phenylpyruvic acid. An example of a link between environmental and genetic events occurs with retinoic acid and its metabolites. As Kirby first demonstrated,12 the development of the aorticopulmonary and truncal septa depends on the migration into the embryonic heart of cells from the cranial neural crest. If these cells are removed experimentally, there is a high incidence of conotruncal defects such as ventricular septal defects, double outlet right ventricles, and truncus arteriosus. Since these neural crest cells also aid in the formation of the pharyngeal arches and pouches

106

Fetal Cardiology

(from which the thymus and parathyroid glands are derived) and the aortic arches, this may explain why aortic arch anomalies and truncus arteriosus are so frequently associated with DiGeorge (CATCH-22: cardiac defects, abnormal facies, thymic hypoplasia, cleft palate, hypocalcemia, resulting from 22q 11 deletions) syndrome.39–42 Certain chemicals are now known to interfere with migration of these neural crest cells. Thus, giving bis-diamine, isotretoin, or all-trans-retinoic acid produces lesions resembling those found in Kirby’s experiments.5 It is thus possible that some outflow tract lesions might be due either to genetically or to environmentally determined defects of neural crest cell migration. Here, the environmental agent, retinoic acid, phenotypically and perhaps genotypically mimics DiGeorge syndrome. About 35% of dilated cardiomyopathies are familial and genetically determined, and the majority of hypertrophic cardiomyopathies are also due to isolated gene mutations43–49 (Table 8.3). Although most of these cardiomyopathies are not present in the fetus or even in the neonate, they do have a genetic origin. So do certain arrhythmias such as the long QT syndrome,25,50–52 Brugada syndrome,53–55 and those associated with arrhythmogenic right ventricular dysplasia.56–58 A normal echocardiogram in the face of an abnormal family history of one of these diseases does not exclude the genetic defect.

Incidence of congenital heart disease To determine the true incidence of CHD there must be a medical system in which pediatric cardiologists can diagnose CHD accurately and objectively, and in which the whole population has easy access to these cardiologists. These conditions are met in several countries today. One of the changes in recent years is the availability of echocardiography for diagnosis. Not only is this very accurate in experienced hands, but it can be applied to children with minimal heart disease who previously would not have had the diagnosis proven because they would not have been submitted to cardiac catheterization. Even if the above criteria are met, the incidence of CHD will be determined accurately only if ascertainment of the disease is complete. Two barriers exist. Some CHD causes death in the first few days after birth. A specific diagnosis might not have been made in these infants by the time of death, and without an autopsy examination the incidence of these serious malformations might be seriously underestimated.59 Conversely, children with very mild lesions such as minimal pulmonic stenosis, or small atrial or ventricular septal defects, might never be included in any cardiologist’s practice, so that their frequency will also be underestimated. Even though these mild lesions seldom cause clinical problems, failing to include them in a

database reduces our ability to study the factors that cause these lesions. There are several other factors that influence our ability to estimate the true incidence of CHD: 1. Patent ductus arteriosus of prematurity, a maturational disorder rather than a cardiovascular anomaly, is very common. An unknown number of these may be included in any collected series, thereby inflating the incidence of the anomaly.60 Unfortunately, many reports do not specifically exclude this form of patent ductus arteriosus. 2. Another major problem is whether or not the bicuspid aortic valve is included in the series. Some data suggest that the bicuspid aortic valve occurs in 1% (possibly more) of the population.61 If even a few patients with this lesion are included under the title of aortic stenosis it will alter the apparent incidence of the latter lesion drastically. Alternatively, if subjects with bicuspid valves and small pressure gradients across them are not classified as aortic stenosis, the incidence of aortic stenosis will appear to be smaller than it should be. 3. Mitral valve prolapse is also common, and has been thought to occur in 4–5% of the population.62 The prolapse does not appear to be present at birth,63 although this does not exclude a developmental or genetic origin for the prolapse. Most series do not include prolapse as a congenital lesion, but obviously if even a few with mitral valve prolapse and mitral regurgitation are included under the heading of mitral valve lesions, they will inflate their incidence. 4. Atrioventricular septal (endocardial cushion canal) defects are disproportionately frequent in children with trisomy 21, and trisomy 21 is more frequent in mothers over 35 years old. The frequency of this lesion in any series, therefore, depends on two factors. One is how many older mothers there are in any series, and this may well differ in different societies. The other is how many pregnancies of fetuses with trisomy 21 will be terminated medically, thus reducing the incidence of this lesion at birth. 5. Many serious forms of CHD are now detected by fetal echocardiography, and the parents may choose to abort these fetuses.64 In some series65,66 as many as 50% of pregnancies with fetuses with CHD were electively terminated. In communities where this practice occurs we have to take account of these aborted fetuses if we wish to determine the true incidence of CHD. 6. Recent studies of infants in the newborn nursery have shown that as many as 4–5% of them have tiny muscular ventricular septal defects.67 Many of these infants had no murmurs. About 95% of these defects close by themselves within 6–12 months after birth. Therefore, the incidence of ventricular septal defects and indeed of all CHD (because ventricular septal defects are the most common form of CHD) depends upon how many of

Epidemiology of congenital heart disease

these trivial defects are included in the series.2 If they are all included, the incidence of all forms of CHD might be 5–6% of all live births. If they are excluded, the incidence drops to about 1% of live births. In fact, over the last 10 years the incidence of CHD has been rising slowly, not because of true increase in all congenital heart lesions but because of an increasing number of patients with small ventricular septal defects who are now being included in the series.68–71 7. In one study, the investigators specified that they did not include mild pulmonic stenosis with systolic gradients across the pulmonary valve of under 25 mmHg.72 Studies such as this will then underestimate the incidence of pulmonary stenosis which is often mild. 8. Fetal echocardiography has shown that certain lesions, particularly ventricular septal defects, may be detected in utero but have disappeared at the time of birth. In terms of the need for medical services this is a very

107

desirable outcome, but in terms of understanding frequency and etiology this leads to the underestimation of the incidence of these lesions. From these caveats it is clear that it is very difficult to assess accurately the incidence of CHD at birth. It should be clear, too, that it might be better to describe the absolute incidence of each specific lesion per 1000 or per 100 000 of the population, rather than to describe its incidence as a proportion of all CHD. For example, consider that in a series examined after a year of age that excludes patent ductus arteriosus in premature infants, mitral valve prolapse, and bicuspid aortic valves there are 50 patients with CHD who have tetralogy of Fallot, 400 children with a ventricular septal defect, and 550 children with all other forms of CHD. From these data, the incidence of tetralogy will be 5% and of ventricular septal defects will be 40% of all CHD. If in that same series the investigators decided

Table 8.5 Incidence of CHD in liveborn children. Data on percentages from reference 73, data on incidence per million from reference 2 Lesion

Percentage of all congenital heart disease∗ Lowest

Ventricular septal defect**

25%

Median

75%

Per million liveborn children

Highest

Lowest

25%

Median

75%

Highest

16.4

27.1

32.4

42.3

50.2

987

1773

2829

4482

6616

Patent ductus arteriosus

0.8

5.2

6.8

11.0

16.0

60

324

567

782

2108

Atrial septal defect

3.4

6.2

7.5

10.8

14.5

135

372

564

1059

2112

Atrioventricular septal defect

1.3

2.8

3.8

5.2

19.6

85

242

340

396

791

Pulmonic stenosis

2.2

5.2

7.0

8.8

14.3

160

355

532

836

1155

Aortic stenosis

0.3

2.7

3.9

5.8

12.0

40

161

256

388

1425

Coarctation of the aorta

0.0

3.6

4.8

5.7

9.8

0

289

356

492

620

Transposition of the great arteries

2.1

3.5

4.4

5.4

8.4

176

226

303

364

560

Tetralogy of Fallot

2.2

3.8

5.2

7.6

10.4

167

302

356

577

633

Persistent truncus arteriosus

0.0

0.6

1.4

1.7

3.8

0

61

94

136

344

Hypoplastic left heart

0.0

1.6

2.8

3.4

5.7

0

154

226

279

347

Hypoplastic right heart

0.0

1.5

2.2

3.2

5.7

0

105

160

224

347

Double inlet left ventricle

0.0

0.8

1.5

1.9

2.7

0

54

85

136

277

Double outlet right ventricle

0.6

1.0

1.8

3.0

4.3

51

82

115

188

263

Total anomalous pulmonary venous connection

0.0

0.6

1.0

1.9

2.8

0

52

91

120

155

Miscellaneous

2.6

7.6

10.1

14.6

23.9

∗25% and 75% are the lower and upper quartiles, respectively. ∗∗Excludes one study in which 5% of all neonates had a small ventricular septal defect (VSD), but other forms of CHD were not studied. This study would give a high value for VSD of 50 000 per million liveborn children, and about 93% of all CHD. In addition to the data shown above, the bicuspid aortic value occurs in about 10–12 per thousand live births, equivalent to the incidence of all other forms of congenital heart disease combined.

108

Fetal Cardiology

to include another 4000 children with small ventricular septal defects present at birth, then the same 50 children with tetralogy of Fallot would be included in a group of 5000 children with CHD (4400 of whom would have ventricular septal defects), for a percentage of 1% tetralogy of Fallot. It would be better to focus on the 50 with tetralogy of Fallot and relate them to the more accurately determined numbers of live births. Some of the current data on the incidence of CHD are given in Table 8.5, both as percentages of all CHD and as numbers per million live births.2 Even excluding the studies of neonates with an extremely high percentage of ventricular septal defects, a ventricular septal defect is still the most frequent form of CHD. Not only is this true in general, but a ventricular septal defect is the most frequently observed congenital cardiac defect in subjects with chromosomal and genetic defects. This probably does not carry any great implications. The ventricular septum is formed from multiple portions of the developing heart,5 so that it is reasonable that many different ways of interfering with cardiac development will end up with a ventricular septal defect.

Conclusions Worldwide, CHD is estimated to occur in about 1 500 000 live births annually.74 Today, because of advances in its treatment, most subjects with CHD survive to become adults and to raise their own families. This progress comes at considerable economic cost; furthermore, CHD remains a major cause of morbidity and premature death. Because the children of these parents have an increased incidence of CHD,23 the total incidence of CHD is likely to increase slowly, generation by generation. Therefore, the ultimate goal should be to prevent CHD or reduce its incidence, and this can be accomplished only after we have a better understanding of its genetic and environmental causes.

4.

5. 6. 7.

8.

9.

10.

11.

12.

13.

14. 15.

16.

17.

Acknowledgments

18.

I thank Dr Harold Bernstein and Dr James Bristow for valuable advice and criticism.

19.

References

20.

1. 2. 3.

Hoffman JIE. Incidence of congenital heart disease: II. Prenatal incidence. Pediatr Cardiol 1995; 16: 155–65. Hoffman JIE, Kaplan S. The incidence of congenital heart disease. J Am Coll Cardiol 2002; 39: 1890–900. Bristow J. The search for genetic mechanisms of congenital heart disease. Cell Mol Biol Res 1995; 41: 307–19.

21.

Gittenberger-De Groot AC, Poelmann RE. Normal and abnormal cardiac development. In: Moller JH, Hoffman JIE, eds. Pediatric Cardiovascular Medicine. New York: Churchill Livingstone, 2000: 3–14. Harvey RP, Rosenthal N, eds. Heart Development. San Diego: Academic Press, 1999. Kelly RG. Molecular inroads into the anterior heart field. Trends Cardiovasc Med 2005; 15: 51–6. Maeda J, Yamagishi H, McAnally J et al. Tbx1 is regulated by forkhead proteins in the secondary heart field. Dev Dyn 2006; 235: 701–10. Phan D, Rasmussen TL, Nakagawa O et al. BOP, a regulator of right ventricular heart development, is a direct transcriptional target of MEF2C in the developing heart. Development 2005; 132: 2669–78. Verzi MP, McCulley DJ, De Val S et al. The right ventricle, outflow tract, and ventricular septum comprise a restricted expression domain within the secondary/anterior heart field. Dev Biol 2005; 287: 134–45. Waldo KL, Hutson MR, Ward CC et al. Secondary heart field contributes myocardium and smooth muscle to the arterial pole of the developing heart. Dev Biol 2005; 281: 78–90. Zaffran S, Kelly RG, Meilhac SM et al. Right ventricular myocardium derives from the anterior heart field. Circ Res 2004; 95: 261–8. Kirby ML. Contribution of neural crest to heart and vessel morphology. In: Harvey RP, Rosenthal N, eds. Heart Development. San Diego: Academic Press, 1999: 1179–93. Waldo KL, Hutson MR, Stadt HA et al. Cardiac neural crest is necessary for normal addition of the myocardium to the arterial pole from the secondary heart field. Dev Biol 2005; 281: 66–77. Brueckner M. Cilia propel the embryo in the right direction. Am J Med Genet 2001; 101: 339–44. McGrath J, Somlo S, Makova S et al. Two populations of node monocilia initiate left-right asymmetry in the mouse. Cell 2003; 114: 61–73. Schneider H, Brueckner M. Of mice and men: dissecting the genetic pathway that controls left-right asymmetry in mice and humans. Am J Med Genet 2000; 97: 258–70. Casey B, Kosaki K. Genetics of human left-right axis malformations. In: Harvey RP, Rosenthal N, eds. Heart Development. San Diego: Academic Press, 1999: 479–89. Malumder K, Overbeek PA. Left-right asymmetry and cardiac looping. In: Harvey RP, Rosenthal N, eds. Heart Development. San Diego: Academic Press, 1999: 391–402. Mjaadvedt CH, Yamamura H, Wessels A et al. Mechanisms of segmentation, septation, and remodeling of the tubular heart: endocardial cushion fate and cardiac looping. In: Harvey RP, Rosenthal N, eds. Heart Development. San Diego: Academic Press, 1999: 159–77. Burn J, Goodship J. Congenital heart disease. In: Emery AE, Rimoin DL, eds. Principles and Practice of Medical Genetics. Edinburgh: Churchill Livingstone, 1997: 767–828. Greenwood RD, Rosenthal A, Parisi L et al. Extracardiac abnormalities in infants with congenital heart disease. Pediatrics 1975; 55: 485–92.

Epidemiology of congenital heart disease

22.

23.

24.

25. 26. 27.

28.

29.

30.

31.

32.

33. 34.

35.

36.

37. 38.

39.

Bruneau BG, Logan M, Davis N et al. Chamber-specific cardiac expression of Tbx5 and heart defects in HoltOram syndrome. Dev Biol 1999; 211: 100–8. Jongmans M, Sistermans EA, Rikken A et al. Genotypic and phenotypic characterization of Noonan syndrome: new data and review of the literature. Am J Med Genet A 2005; 134: 165–70. Burn J. The aetiology of congenital heart disease. In: Anderson RH, Baker EJ, Macartney FJ et al, eds. Paediatric Cardiology, 2nd edn. London: Churchill Livingstone, 2002: 141–213. Ackerman MJ. Molecular basis of congenital and acquired long QT syndromes. J Electrocardiol 2004; 37: 1–6. Towbin JA. Molecular genetics of hypertrophic cardiomyopathy. Curr Cardiol Rep 2000; 2: 134–40. Hamosh A, Scott AF, Amberger JS et al. Online Mendelian Inheritance in Man (OMIM), a knowledgebase of human genes and genetic disorders. Nucleic Acids Res 2005; 33(Database issue): D514–7. Benson DW, Silberbach GM, Kavanaugh-McHugh A et al. Mutations in the cardiac transcription factor NKX2.5 affect diverse cardiac developmental pathways. J Clin Invest 1999; 104: 1567–73. Greenwood RD. Cardiovascular malformations associated with extracardiac anomalies and malformation syndromes. Patterns for diagnosis. Clin Pediatr (Phila) 1984; 23: 145–51. Kramer HH, Majewski F, Trampisch HJ et al. Malformation patterns in children with congenital heart disease. Am J Dis Child 1987; 141: 789–95. Zahka KJ. Associated abnormalities in children with congenital heart disease. In: Emmanouilides GC, Riemenschneider TA, Allen HD, Gutgesell HP, eds. Heart Disease in Infants, Children and Adolescents, Including the Fetus and Young Adult, 5th edn. Baltimore: Williams & Wilkins, 1989: 614–28. Nora JJ. Multifactorial inheritance hypothesis for the etiology of congenital heart diseases. The geneticenvironmental interaction. Circulation 1968; 38: 604–17. Sabatine MS, Seidman JG, Seidman CE. Cardiovascular genomics. Circulation 2006; 113: e450–5. van Asselt KM, Kok HS, van der Schouw YT et al. Role of genetic analyses in cardiology: part II: heritability estimation for gene searching in multifactorial diseases. Circulation 2006; 113: 1136–9. Kurnit DM, Aldridge JF, Matsuoka R et al. Increased adhesiveness of trisomy 21 cells and atrioventricular canal malformations in Down syndrome: a stochastic model. Am J Med Genet 1985; 20: 385–99. Kurnit DM, Layton WM, Matthysse S. Genetics, chance, and morphogenesis. Am J Hum Genet 1987; 41: 979–95. Jackson BT. The pathogenesis of congenital cardiovascular anomalies. N Engl J Med 1968; 279: 80–9 concl. Michels VV, Riccardi VM. Congenital heart defects. In: Emery AE, Rimoin DL, eds. Principles and Practice of Medical Genetics. Edinburgh: Churchill Livingstone, 1990: 1207–37. Hong R. The DiGeorge anomaly. Clin Rev Allergy Immunol 2001; 20: 43–60.

40.

41.

42.

43. 44. 45.

46. 47. 48.

49.

50. 51.

52.

53. 54.

55. 56.

57.

58.

59.

60.

109

Lindsay EA, Baldini A. Congenital heart defects and 22q11 deletions: which genes count? Mol Med Today 1998; 4: 350–7. Momma K, Matsuoka R, Takao A. Aortic arch anomalies associated with chromosome 22q11 deletion (CATCH 22). Pediatr Cardiol 1999; 20: 97–102. Wilson DI, Goodship JA, Burn J et al. Deletions within chromosome 22q11 in familial congenital heart disease. Lancet 1992; 340: 573–5. Baty C, Watkins H. Familial hypertrophic cardiomyopathy: man, mouse and cat. QJM 1998; 91: 791–3. Burch M, Blair E. The inheritance of hypertrophic cardiomyopathy. Pediatr Cardiol 1999; 20: 313–16. Fatkin D, MacRae C, Sasaki T et al. Missense mutations in the rod domain of the lamin A/C gene as causes of dilated cardiomyopathy and conduction-system disease. N Engl J Med 1999; 341: 1715–24. Piano MR. Familial hypertrophic cardiomyopathy. J Cardiovasc Nurs 1999; 13: 46–58. Towbin JA, Bowles NE. Molecular genetics of left ventricular dysfunction. Curr Mol Med 2001; 1: 81–90. Towbin JA, Bowles NE. Dilated cardiomyopathy: a tale of cytoskeletal proteins and beyond. J Cardiovasc Electrophysiol 2006; 17: 919–26. Towbin JA, Solaro RJ. Genetics of dilated cardiomyopathy: more genes that kill. J Am Coll Cardiol 2004; 44: 2041–3. Ackerman MJ. Cardiac channelopathies: it’s in the genes. Nat Med 2004; 10: 463–4. Ching CK, Tan EC. Congenital long QT syndromes: clinical features, molecular genetics and genetic testing. Expert Rev Mol Diagn 2006; 6: 365–74. Modell SM, Lehmann MH. The long QT syndrome family of cardiac ion channelopathies: a HuGE review. Genet Med 2006; 8: 143–55. Antzelevitch C. Brugada syndrome. Pacing Clin Electrophysiol 2006; 29: 1130–59. Coronel R, Berecki G, Opthof T. Why the Brugada syndrome is not yet a disease: Syndromes, diseases, and genetic causality. Cardiovasc Res 2006; 72: 361–3. Napolitano C, Priori SG. Brugada syndrome. Orphanet J Rare Dis 2006; 1: 35. Yang Z, Bowles NE, Scherer SE et al. Desmosomal dysfunction due to mutations in desmoplakin causes arrhythmogenic right ventricular dysplasia/cardiomyopathy. Circ Res 2006; 99: 646–55. Beffagna G, Occhi G, Nava A et al. Regulatory mutations in transforming growth factor-beta3 gene cause arrhythmogenic right ventricular cardiomyopathy type 1. Cardiovasc Res 2005; 65: 366–73. Rampazzo A, Beffagna G, Nava A et al. Arrhythmogenic right ventricular cardiomyopathy type 1 (ARVD1): confirmation of locus assignment and mutation screening of four candidate genes. Eur J Hum Genet 2003; 11: 69–76. Abu-Harb M, Wyllie J, Hey E et al. Antenatal diagnosis of congenital heart disease and Down’s syndrome: the potential effect on the practice of paediatric cardiology. Br Heart J 1995; 74: 192–8. Anderson CE, Edmonds LD, Erickson JD. Patent ductus arteriosus and ventricular septal defect: trends in reported frequency. Am J Epidemiol 1978; 107: 281–9.

110

61.

62.

63.

64.

65.

66.

67.

Fetal Cardiology

Hoffman JIE, Kaplan S, Liberthson RR. Prevalence of congenital heart disease. Am Heart J 2004; 147: 425–39. Freed LA, Levy D, Levine RA et al. Prevalence and clinical outcome of mitral-valve prolapse. N Engl J Med 1999; 341: 1–7. Nascimento R, Freitas A, Teixeira F et al. Is mitral valve prolapse a congenital or acquired disease? Am J Cardiol 1997; 79: 226–7. Allan LD, Cook A, Sullivan I et al. Hypoplastic left heart syndrome: effects of fetal echocardiography on birth prevalence. Lancet 1991; 337: 959–61. Allan LD, Sharland GK, Milburn A et al. Prospective diagnosis of 1,006 consecutive cases of congenital heart disease in the fetus. J Am Coll Cardiol 1994; 23: 1452–8. Daubeney PE, Sharland GK, Cook AC et al. Pulmonary atresia with intact ventricular septum: impact of fetal echocardiography on incidence at birth and postnatal outcome. UK and Eire Collaborative Study of Pulmonary Atresia with Intact Ventricular Septum. Circulation 1998; 98: 562–6. Roguin N, Du ZD, Barak M et al. High prevalence of muscular ventricular septal defect in neonates. J Am Coll Cardiol 1995; 26: 1545–8.

68.

69.

70.

71.

72. 73.

74.

Fixler DE, Pastor P, Chamberlin M et al. Trends in congenital heart disease in Dallas county births 1971–1984. Circulation 1990; 81: 137–42. Meberg A, Otterstad JE, Froland G et al. [Children with congenital heart defects in Vestfold 1982–88. Increase in the incidence resulting from improved diagnostics methods]. Tidsskr Nor Laegeforen 1990; 110: 354–7. [in Norwegian] Spooner EW, Hook EB, Farina MA et al. Evaluation of a temporal increase in ventricular septal defects: estimated prevalence and severity in northeastern New York, 1970–1983. Teratology 1988; 37: 21–8. Wren C, Richmond S, Donaldson L. Presentation of congenital heart disease in infancy: implications for routine examination. Arch Dis Child Fetal Neonat Ed 1999; 80: F49–53. Bound JP, Logan WF. Incidence of congenital heart disease in Blackpool 1957–1971. Br Heart J 1977; 39: 445–50. Hoffman JIE. Incidence, mortality, and natural history. In: Anderson RH, Baker EJ, Macartney FJ et al, eds. Paediatric Cardiology, 2nd edn. London: Churchill Livingstone, 2002: 111–39. Hoffman JIE. Reflections on the past, present, and future of pediatric cardiology. Cardiol Young 1994; 4: 208–23.

9 Indications for fetal echocardiography: screening in low- and high-risk populations Ulrich Gembruch and Annegret Geipel

Introduction Technical advances in ultrasound technology over the past decade and the introduction of fetal echocardiography into the prenatal ultrasound examination have further improved the antenatal detection of congenital heart disease (CHD). Nevertheless, cardiac anomalies are the most frequently overlooked lesions during prenatal ultrasound evaluation, and this has profound medical, psychological, socioeconomic, and medicolegal consequences. There are several controversial issues in this field, and the usefulness of fetal echocardiography as a screening instrument in unselected populations has remained a matter of debate. Congenital heart defects constitute a major segment of birth malformations. The reported incidence rate per 1000 liveborn infants increases from 3.3 at birth to 4.0 at the end of the first week of neonatal life, 5.2 by the end of the first month, and 7.8 by the end of the first year. In the Baltimore–Washington Infant Study between 1981 and 1989, 60% of cases with CHD were diagnosed by 4 weeks of age, 80% by 12 weeks, and 90% by 24 weeks.1 Cardiac defects are seen four to five times more frequently in stillbirth.2 However, the true incidence among fetuses is difficult to evaluate. The in utero development of some types of congenital heart defect is still partially unknown, and is currently under intensive study.3 The likelihood of detecting a fetal cardiac defect is closely related to the experience of the ultrasonographer, the timing of the examination, and the equipment used.4–6 To perform a complete and accurate examination, highresolution ultrasound with pulsed and color Doppler imaging capabilities is required. As some cardiac lesions can evolve in terms of their echocardiographic characteristics with the progress of pregnancy, a certain group of defects cannot be reliably excluded by a single scan, particularly if it is performed in early pregnancy. Owing to

widely different levels of the obstetric scanning experience of examiners, there is a large discrepancy in reported study results. To improve expertise in the diagnosis and management, and knowledge of the natural history of CHD, tertiary centers have been created, obtaining a high level of sensitivity and specificity. After accurate diagnosis, counseling of the patient is mandatory about the kind of anomaly, possibilities for treatment, and management options. The perinatal management will depend on associated extracardiac malformations or chromosomal disorders, the prognosis and outcome of the cardiac defect, and gestational age-related differences in national laws regarding termination of pregnancy, as well as the wishes of the patient. A referral center should offer up-to-date information about current methods and results of treatment with the backup of a pediatric cardiologist and surgeon. In recent years, efforts have predominantly focused on developing screening programs for prenatal diagnosis of congenital cardiac anomalies. Specific indications and risks have been classified, and patients considered at high risk have been offered detailed fetal echocardiography, commonly at a center with appropriate expertise and facilities. However, most children are born to mothers who have no known risk features during pregnancy. Screening in the low-risk population has been reported to have lower accuracy than in the high-risk population.7,8 The impact of these screening programs on the mortality and morbidity of children with prenatally diagnosed cardiac anomalies and of the general population has been evaluated with different results. Fetuses with lesions that require intervention early in the neonatal period might especially benefit from closer care and the planning of delivery.9–11 Furthermore, the possibility of intrauterine cardiac intervention, which is established for antiarrhythmic treatment but is still in the experimental stage for others, might further improve fetal outcome12 (Chapter 33).

112

Fetal Cardiology

Risk groups and indications for fetal echocardiography Recommendations on indications for fetal echocardiography are closely linked to the recognition of possible etiologic factors and risk groups in respect of CHD. Three main etiologic groups have been identified on the basis of empirical, animal experimental, and genetic data.13–18 These are primary genetic factors (chromosomal aberrations, 5%; monogenic inherited defects, 3%), primary exogenous factors (maternal infections, 1%; teratogens/maternal metabolic diseases, 1%), and multifactorial inheritance (90%). In cases of multifactorial inheritance the cardiac anomaly is considered to result from the interaction between unspecified genetic and environmental factors. This includes a genetic predisposition and the influence of environmental triggers at a vulnerable period of embryogenesis; the most sensitive period is approximately 6 weeks.17 With advances in molecular genetics, the concept of multifactorial etiology is being increasingly questioned, as numbers of congenital cardiac defects have been identified as a result of inherited or new mutations of single genes and of inherited or spontaneous microdeletions.18–20 However, at present the etiology for the majority of congenital cardiac anomalies is still under research, and therefore in clinical practice the identification of high-risk populations is based on patient history and sonographic anomalies or markers (Table 9.1).

Examination of the high-risk population There are sufficient numbers of referral centers for highrisk patients where detailed echocardiography can be performed. High-resolution ultrasound with color Doppler facilities should be used in addition to two-dimensional echocardiography. Extended echocardiographic examination includes visualization of International Society of Ultrasound in Obstetrics and Gynecology (ISUOG) guidelines26: • the standard four-chamber view • the left and right ventricular outflow tracts with crossing of the aorta and the pulmonary artery • the ‘three-vessel–trachea view’ • the short-axis view of the ventricles and great arteries • the aortic arch and ductus arteriosus • the superior and inferior venae cavae. Using this technique, the detection rate of congenital cardiac anomalies in referral centers is about 80–90%.3,27 While nearly all cardiac anomalies can be diagnosed by two-dimensional echocardiography, some cases might be exclusively diagnosed by color-coded Doppler echocardiography. In some cases, the latter can provide essential

diagnostic and prognostic information, especially through enhanced demonstration of the great arteries, the pulmonary veins, valvular stenoses, insufficiencies, and intracardiac shunts.21,28–31 A novel approach in the assessment of the fetal cardiac anatomy are four-dimensional (4D) sonography, B-flow imaging, and spatiotemporal image correlation (STIC).32 Although the majority of detailed cardiac examinations is performed between 20 and 24 weeks’ gestation, specialized centers provide increasingly the possibility of early fetal echocardiography performed from around 12–14 weeks of gestation. Indications are the same as described for second trimester fetal echocardiography. However, examination early in pregnancy has lower detection rates compared to second trimester echocardiography, and should therefore always be followed by a second trimester examination33–37 (Chapter 13).

Indications for fetal echocardiography Although only approximately 10% of fetuses with cardiac anomalies have known risk factors, detailed fetal echocardiography, including first trimester evaluation, should be offered to this high-risk population (Table 9.1). The proportion of fetuses affected is likely to be risk factor-specific, and therefore each risk category should be assessed independently. The most common indication is a family history of CHD – with a previous child affected, a parent or a second-degree relative with a heart defect, or an occurrence of a certain familial genetic syndrome (Table 9.2). The recurrence rate is about 2–4% for parental CHD and 2% for siblings. Some lesions, however, have a slightly higher recurrence rate than others, especially left heart obstructive lesions and heterotaxy syndromes.20,39,40 In general, the recurrence risk of CHD as an isolated multifactorial disorder is obtained by taking the square root of the incidence in the specific population.41 In a series of 6640 pregnancies with a first-degree family history of CHD, a recurrence was seen in 2.7%. Exact concordance was seen in 37% and group concordance in 44% of cases. Concordance was highest (55%) in families with two or more recurrences and in cases with isolated atrioventricular septal defects (80%) and laterality defects (64%).42 More congenital heart defects occurred in the offspring of affected mothers (3–5%) than in those of affected fathers (2%).14,40,42,43 Furthermore, the rate of miscarriage in offspring of affected women (20%) was significantly higher than in those of affected fathers (9%).14,17,43 Paternal CHD has been considered a stronger risk factor in one study,44 whereas other studies found no statistical differences compared to siblings.43 Besides the possibility of mitochondrial inheritance,17 imprinted genes may influence cardiac development, and

Indications for fetal echocardiography

113

Table 9.1 Indications for fetal echocardiography (high-risk group for congenital heart disease (CHD)). Modified from references 13, 15, and 20–25 Risk (%)

Indication Increased risk based on history 1. Positive family history of CHD • one previous child affected • two previous siblings affected • maternal CHD • paternal CHD • syndromes or malformations associated with cardiac anomalies (see Table 9.2) 2. Maternal diseases • diabetes mellitus • phenylketonuria • connective tissue disease and/or autoantibodies (risk for atrioventricular block) 3. Exposure to teratogens in pregnancy • teratogenic drugs: alcohol, amphetamine, anticonvulsants (carbamazepine, hydantoin, phenobarbital, phenytoin, trimethadione, valproic acid), lithium, retinoic acid (vitamin A), warfarin • intrauterine infections: rubella, myocarditis from cytomegalovirus, coxsackievirus, and parvovirus • high doses of ionizing radiation Documented fetal anomaly 1. Abnormal obstetric sonogram • suspected cardiac anomaly, abnormal cardiac position • extracardiac malformations frequently associated with cardiac anomaly (see Tables 9.3 and 9.4) • arrhythmia extrasystoles tachyarrhythmia complete heart block • non-immune hydrops fetalis • hygromata colli • increased thickness of nuchal translucency ≥ 3.5 mm • early (before 32 weeks) and/or predominant symmetrical growth restriction • amniotic fluid abnormality of moderate and severe degrees 2. Diagnosed chromosome anomaly (see Tables 9.4 and 9.6) 3. Twin gestation • monozygous twins • conjoined twins

2* 10* 4* 2*

4–6** 12–16** 2–3

50–60 13–14 3–6 1–2 1–2 30–40 15–20 15–20 6 10 5 25–30 7

Decline of invasive prenatal diagnosis (‘genetic sonography’) In cases with increased risk due to • advanced maternal age • abnormal biochemical parameters in the maternal serum (AFP, hCG, PAPP-A, uE3) • familial risk AFP, α-fetoprotein; hCG, human chorionic gonadotropin; uE3, unconjugated estriol, PAPP-A, pregnancy-associated plasma protein A. *The risk of recurrence is also dependent on the type of CHD in the index case. **The risk of CHD is dependent on the metabolic control in the first weeks of pregnancy.

normal heart formation might depend on the maternal copy of the gene.43 Furthermore, a high recurrence risk of around 14% was found in mothers with an atrioventricular septal defect and aortic stenosis, whereas the risk was significantly lower in mothers with tetralogy of Fallot and transposition of the great arteries, and also with coarctation of the aorta (2.5%, 0%, and 4%, respectively),14,41,43 suggesting monogenic inheritance for some defects. However, the predominant effect of prenatal diagnosis, and

especially of fetal echocardiography, is early reassurance of patients with a previous child born after undiagnosed severe cardiac anomaly, often resulting in death. After a normal examination they are calmed, and face the remaining pregnancy with great relief.15,21,39 A number of drugs (Table 9.1) and high doses of ionizing radiation have been implicated as causes of various malformations, including heart anomalies.13,22 Exposure to these teratogens in the first 6–8 weeks of pregnancy increases the risk of

114

Fetal Cardiology

Table 9.2 and 38

Syndromes and conditions frequently associated with CHD. Modified from references 18, 22,

Syndrome

Associated CHD

Risk (%)

Apert Bourneville–Pringle (tuberous sclerosis) Cornelia de Lange CHARGE* association DiGeorge/Shprintzen (microdeletion del 22q11) Ehlers–Danlos Ellis–van Creveld Goldenhar (hemifacial microsomia) Holt–Oram Klippel–Feil Marfan Meckel–Gruber Noonan Rubinstein–Taybi Thrombocytopenia–absent radius (TAR) VACTERL** association Williams–Beuren (microdeletion)

VSD, coarctation, ToF Rhabdomyoma VSD, ASDII VSD, ASD, ToF, DORV, TAC, CoA Conotruncal defects (VSD, ToF, DORV, TAC), CoA, IAA Dilated aortic root, MvP with MR AS ToF, VSD, CoA, IAA, right AA, ASD ASD, VSD VSD, TGA Dilated aortic root, MvP with MR ASD, VSD, CoA, PS PS, VSD, ASD, ToF, AS PDA, ASD, VSD ToF, ASD, dextrocardia HLH, VSD, ToF Supravalvular AS, CoA, IAA, PS, ASD, VSD, ToF

10 50 30 50–65 85–95 50 50 20–30 50–85 25–40 60–90 25 55–65 25 30–33 50 100

*Coloboma, heart defects, atresia of the choanae, retardation of growth and development, genital and urinary abnormalities, ear abnormalities and hearing loss. **Vertebral anomalies, anal atresia, cardiac defect, tracheo-esophageal fistula, renal abnormalities, limb abnormalities. AA, aortic arch; AS, aortic stenosis; ASD, atrial septal defect (type II); AVSD, atrioventricular septal defect; CHB, complete heart block; CoA, coarctation of the aorta; DORV, double outlet right ventricle; HLH, hypoplastic left heart; IAA, interrupted aortic arch; MvP, mitral valve prolapse; MR, mitral regurgitation; PDA, patent ductus arteriosus; PS, pulmonary stenosis; TAC, truncus arteriosus communis; TGA, transposition of great arteries; ToF, tetralogy of Fallot; VSD, ventricular septal defect.

occurrence of CHD. In this context, alcohol, antiepileptics, lithium, and retinoic acid are the most important drugs.45 Coexisting maternal diseases, such as insulin-requiring diabetes mellitus and phenylketonuria, are associated with an elevated risk of fetal cardiac lesions, but good metabolic control of maternal blood sugar and phenylalanine levels, respectively, periconceptionally and during the first 8 weeks of pregnancy might decrease the risk of the affected mothers to near the level of the normal population.46 In cases of poor metabolic control during early pregnancy, the risk for fetal CHD in maternal diabetes is between 4 and 6%,47,48 and in phenylketonuria up to 14%.49 A predominance of conotruncal lesions has been reported in both maternal diseases.48–50 If there are maternal autoantibodies, specifically anti-Ro (anti-SSA) and/or anti-La (anti-SSB), which may occur in mothers with connective tissue diseases such as systemic lupus erythematosus and Sjögren syndrome, but also in healthy women, repetitive echocardiographic examinations of the fetal heart should be performed in the second and third trimesters for early detection of autoantibodyinduced fetal heart block and cardiomyopathy. Some viral intrauterine infections, such as maternal rubella infection early in pregnancy, may be associated with an elevated risk of fetal cardiac lesions, but an increased risk

for fetal structural cardiac lesions is not well documented in other infections. Sometimes, parvovirus B19, coxsackievirus, adenovirus, and cytomegalovirus can cause fetal myocarditis. A much higher incidence of cardiac anomalies is found in fetuses with sonographically suspected cardiac and extracardiac anomalies (Tables 9.1 and 9.3).23,24,27,55 In patients referred for an ‘abnormal’ four-chamber view, a cardiac anomaly can be expected in about 50–60% of patients, depending on the training level of the referring physicians, yielding by far the most cases of severe CHD.5,24,56 Fetuses with diagnosed extracardiac anomalies constitute another major group of patients in whom fetal echocardiography is warranted, as the detection of a cardiac anomaly fundamentally affects the prognosis and might influence perinatal care strategies. A variety of malformations of the central nervous system, the mediastinum, and gastrointestinal and urogenital systems are frequently associated with heart defects (Table 9.3, Chapter 43); the risk does not seem to be increased for gastroschisis. If these anomalies are associated with a cardiac malformation, the likelihood of a chromosomal anomaly is markedly increased. The overall incidence of extracardiac malformations with congenital heart disease varies from 27 to 42%, with a higher

Indications for fetal echocardiography

Table 9.3 Extracardiac malformations frequently associated with a cardiac anomaly. Modified from references 22, 38, and 51–54 Central nervous system (2–15%) hydrocephalus microcephalus agenesis of the corpus callosum encephalocele Dandy–Walker malformation Mediastinum (10–40%) diaphragmatic hernia esophageal atresia (VACTERL association) Gastrointestinal (12–22%) duodenal atresia abnormal situs visceralis anorectal anomalies Abdominal wall (14–30%) omphalocele ectopia cordis Genitourinary (5–40%) hydronephrosis renal agenesis renal dysplasia horseshoe kidney Vascular (5–10%) single umbilical artery persistent right umbilical vein agenesis of ductus venosus

frequency if detected during pregnancy than at birth.22,51,52 There are some types of cardiac lesions where non-cardiac abnormalities are commonly associated, such as atrioventricular septal defect or tetralogy of Fallot, whereas others only rarely occur with extracardiac malformation (Table 9.4).52,55 However, the high incidence of extracardiac malformations of 30–40% is biased by a selected referral to the tertiary center of high-risk fetuses with already detected anomalies. For an effectively screened low-risk population the incidence of extracardiac malformations is assumed to be lower, at around 20%, and is higher the earlier in pregnancy the screening is performed. Fetal echocardiography should be performed in all cases of fetal arrhythmia, not only for differentiation, but also for exclusion of a cardiac anomaly which may be present in 30–40% of cases with complete atrioventricular block and in 2–5% of cases with extrasystoles and tachyarrhythmias.58 Cardiovascular anomalies are often observed in fetuses with non-immune hydrops fetalis (Chapter 34). If brady- and tachyarrhythmias are included as anomalies, the anomaly rate is between 20 and 30%. If only true cardiac anomalies are included, the incidence is between 15 and 20% of cases. This is for two reasons. First, some cardiac defects can result in intrauterine congestive heart

115

Table 9.4 Frequency of extracardiac and chromosomal anomalies associated in single types of CHD diagnosed prenatally.* Based on the data of references 22, 48, 52, and 57 Type of congenital heart disease Atrioventricular septal defect Ventricular septal defect Atrial septal defect Tetralogy of Fallot Double outlet right ventricle Hypoplastic left heart Truncus arteriosus communis Transposition of great arteries Coarctation of the aorta Tricuspid atresia Ebstein’s anomaly Aortic stenosis Pulmonary stenosis/ atresia

Chromosomal anomaly (%)

Extracardiac anomaly (%)

35–47

30–50

37–48

30–37

1–3 27 12–45

16 25–30 19–20

4–10 14–33

1 15–21

0–3

15–26

21–30 2–9 0–3 1–15 4–5

12–20 15–34 6 13 20–26

*In most prenatal series the number of chromosomal and non-chromosomal extracardiac anomalies seems to be distinctly overrepresented.

failure with signs of hydrops, as in atrioventricular septal defect, Ebstein’s anomaly, tricuspid valve dysplasia, and obstructions of the right and left ventricular outflow tracts, resulting in atrioventricular–valvular insufficiencies. Essential to the pathomechanism is the development of atrioventricular–valvular dysfunction in combination with a relative restriction of the transatrial flow through the fossa ovalis, which results in increased right atrial volume and pressure with consequent elevation of hydrostatic venous pressure with resultant edema.59–61 Intrauterine congestive heart failure with hydrops and associated extracardiac malformations has been reported as a negative prognostic factor for fetal outcome.55,57,59,62 Second, chromosome anomalies may result in hydrops by non-cardiac mechanisms. Therefore, in many cases with non-immune hydrops and a heart defect, this relationship is coincidental rather than causal, for example in Turner syndrome, where hydrops and hygroma result from a lymphatic drainage disorder, and a coarctation of the aorta only may be associated. Furthermore, in early pregnancy spontaneous fetal demise frequently occurs in hydropic fetuses often associated with a chromosomal abnormality. Because cardiac defects are more common in this group of fetuses, the incidence of the association between hydrops, chromosomal anomaly,

116

Fetal Cardiology

and cardiac defect is higher in first- and early secondtrimester studies.36 Abnormalities of the cardiovascular system are found at increased frequency in the group of fetuses with thickened nuchal translucency detected by screening between 11 and 14 weeks of gestation.25,63,64 Nuchal translucency (NT) screening selects about 70–80% of fetuses with chromosomal abnormalities at a 5% false-positive rate, including a high proportion of euploid fetuses with cardiac defects or genetic syndromes.36,65–68 While initial reports suggested that narrowing of the aortic isthmus with overperfusion of the head and neck is an underlying pathophysiological mechanism, these initial observations were not confirmed by studies published subsequently.25,64,69 With regard to etiology, in addition to a cardiac cause, other hypotheses include abnormalities of the extracellular matrix, mediastinal compression, and most recently a delay in development of the lymphatic vessels of the neck.70 As a variety of cardiac lesions with different hemodynamic flow pattern can be found among fetuses with increased NT, there is no evidence that this is directly related to the structural cardiac abnormality itself.64,71 However, a temporary impairment of cardiac function in early pregnancy, related to increased NT, tricuspid regurgitation, and/or increased pulsatility of the flow velocity waveforms in the ductus venosus seems to be causal for the majority of cases. The prevalence of major cardiac defects increases exponentially with fetal NT thickness.25,72 In a recent study by Atzei and colleagues including 6921 fetuses, major cardiac defects were identified in 132 (19.1/1000). The prevalence increased from 4.9 per 1000 in those with NT below the median, to 8.7 for NT between the median and ≤ 95th centile, 18.2 for NT between the 95th and 99th centiles, to 35.2, 64.4, and 126.7 for respective NTs of 3.5–4.4 mm, 4.5–5.4 mm, and ≥ 5.5 mm.25 There are six series (Table 9.5) with a combined total of 6982 fetuses, reporting the prevalence of major cardiac defects in fetuses with increased NT. Cardiac abnormalities were found in 16/1000 of those with NT of 2.5–3.4 mm and 64/1000 in those with NT of 3.5 mm and more. As the latter group corresponds to a 2–3 times higher risk for cardiac disease than those with a positive family history, these pregnancies constitute a new high-risk group and should receive detailed fetal

Table 9.5

echocardiography at mid-gestation and, preferentially, even at 12–14 weeks. Whereas increased nuchal translucency often represents a transient anomaly, hygroma colli is often persistent or even progressive and accompanied by pleural effusion and/or general hydrops. Chromosomal abnormalities are reported in up to 50%, in most cases representing Turner syndrome.75 Coarctation of the aorta and other obstructions of the left ventricular outflow tract are the most common cardiac malformations associated with Turner syndrome in utero as well as postpartum.75,76 Another indication for fetal echocardiography is fetal growth restriction, possibly accompanied by abnormalities of amniotic fluid.27,51 Early intrauterine growth restriction with manifestation before 32 weeks of gestation as well as the gestational age-independent symmetric type might raise suspicion of chromosomal abnormality, especially trisomy 18 and triploidy, or less commonly, of nonchromosomal syndromes. Because echocardiographically detectable cardiac anomalies are often present in these disorders, complete cardiac examination should be performed in all fetuses with early growth restriction, and demonstration of a cardiac defect should be followed by karyotyping. Finally, twin pregnancies must be mentioned; an increased risk (4–7%) for cardiac anomalies is predominantly seen in monozygous twins.22,77 A high intrauterine mortality rate is reported in those fetuses with prenatally detected heart disease and associated chromosomal pathology.51,52,55 Therefore, the frequency of an associated abnormal karyotype (20–30%) is far higher in prenatal studies than in liveborn infants.51,55,78 Another cause for the high frequency of chromosomal aberrations in prenatal series with CHD is the referral bias that primarily selects fetuses with an extracardiac malformation, hydrops, and early growth restriction for detailed cardiac examination; in these settings the incidence of chromosomal anomalies is high. As an isolated heart defect diagnosed in utero might be the only sign of a chromosomal abnormality, karyotyping has to be discussed. The identification of a lethal chromosomal aberration might have an enormous impact on the further obstetric and neonatal management.27,78 In Table 9.6, the most common chromosomal disorders associated with CHD are summarized. In contrast to lesions

Prevalence of cardiac defects in fetuses with increased nuchal translucency (NT)

Study Hyett et al, 199769 Ghi et al, 200172 Lopes et al, 200373 McAuliffe et al, 200535 Bahado-Singh et al, 200574 Atzei et al, 200525 Total

n

NT 2.5–3.4 mm

NT > 3.5 mm

1389 1319 275 177 378 3444 6982

6/1102 (0.5%) 18/722 (2.5%) 2/185 (1.1%) 5/122 (4.1%) 2/335 (0.6%) 43/2365 (1.8%) 76/4831 (1.6%)

9/287 (3.1%) 42/597 (7.0%) 11/90 (12.2%) 8/55 (14.5%) 1/43 (2.3%) 64/1079 (5.9%) 135/2151 (6.3%)

Indications for fetal echocardiography

Table 9.6 and 79

117

Chromosomal abnormalities associated with CHD. Modified from references 15, 18, 38, 53, Risk (%)

Karyotype

Associated heart defects

Trisomy 21 Trisomy 18

AVSD, VSD, ASD, ToF, CoA Conotruncal defects (VSD, ToF, DORV), AVSD, BiAv, AS, BiPv, PS, HLH VSD, ASD, HLH, ToF TAPVC, VSD, ASD VSD CoA (mostly tubular), AS, HLH, ASD

40–50 99

VSD, ASD, PDA VSD, ASD, PDA Conotruncal defects (VSD, ToF, DORV, TAC), CoA, IAA

40–60 30–60 75–85

Trisomy 13 Partial trisomy 22 (cat-eye syndrome) Triploidy Turner syndrome Partial monosomy 4p− (Wolf–Hirschhorn syndrome) 5p− (cri-du-chat syndrome) Microdeletion 22q11

80–90 40 60 30–40

AS, aortic stenosis; ASD, atrial septal defect; AVSD, atrioventricular septal defect; BiAv, bicuspid aortic valve; BiPv, bicuspid pulmonary valve; CoA, coarctation of the aorta; DORV, double outlet right ventricle; HLH, hypoplastic left heart; IAA, interrupted aortic arch; PDA, persistent ductus arteriosus; PS, pulmonary stenosis; TAC, truncus arteriosus communis; TAPVC, totally anomalous pulmonary venous connection; ToF, tetralogy of Fallot; VSD, ventricular septal defect.

such as tricuspid atresia, transposition of the great arteries, and Ebstein’s anomaly with rarely associated aneuploidies, there is an accumulation of cases with chromosomal anomalies in fetuses with atrioventricular septal defect, ventricular septal defects, tetralogy of Fallot, and coarctation of the aorta (Table 9.4).55,57,79,80 An increasing proportion of patients requiring fetal echocardiography are those who want to avoid invasive prenatal diagnosis despite given indications such as advanced maternal age, abnormal serum biochemistry, or other risk factors for fetal chromosomal anomalies. In the second trimester, each chromosomal disorder manifests its own phenotypic expression. This includes a variety of major malformations as described for trisomy 18, trisomy 13, and triploidy, and more subtle anomalies associated with trisomy 21. In addition to the detailed sonographic examination, fetal echocardiography is of particular importance, as approximately 50% of newborns with trisomy 21, 99% of those with trisomy 18, 90% of those with trisomy 13, and 35% of those with Turner syndrome have a cardiac defect (Table 9.6).81 In contrast to the majority of sonographic markers for trisomy 21, heart defects are usually not transient, as is nuchal translucency, are not dependent on gestational age-related cut-off levels, as are nuchal thickness and pyelectasis, and on center-specific reference images and differences between ethnic groups, as are long bone and nasal bone lengths, and are independent of observer and scanning conditions, in contrast to hyperreflectoric bowel, for instance. Offering second trimester ‘genetic sonography’ to women at increased risk for fetal trisomy 21 is associated with a high utilization rate. In the presence of a normal ultrasound examination,

the risk for fetal Down syndrome was reduced by 83% in patients with advanced maternal age and 88% in patients with abnormal triple screen. This approach clearly decreased the rate of genetic amniocentesis based on age-related risk and/or abnormal biochemical markers, with avoidance of unnecessary amniocentesis-induced fetal loss. In appropriate clinical settings it was found to be cost-effective, with an increase in the detection rate of trisomy 21.82–84 However, reported sensitivities of second trimester identification of fetal Down syndrome differ widely. One of the main reasons for these variations is the underdetection of congenital heart defects. In a targeted echocardiographic study, Paladini et al reported a 56% incidence of CHD in fetuses with known Down syndrome, which was an atrioventricular septal defect (AVSD) in 44% of cases. Conversely, 53% of cases with AVSD and normal visceral situs were associated with Down syndrome.85 In contrast, most studies evaluating second trimester screening for trisomy 21 detect fewer than 20% of heart malformations. The targeted echocardiographic examination and the addition of color-coded Doppler imaging for better visualization has been shown to markedly increase the likelihood of detecting trisomy 21 during ‘genetic sonography’. By combination of non-cardiac markers with the four-chamber view, the sensitivity was increased from 60 to 75%, with a falsepositive rate of 6.4%. By evaluating the heart with regard to ventricular septal defects, atrioventricular defects, pericardial effusion, tricuspid regurgitation, and chamber discordance, the sensitivity of ‘genetic sonography’ was further increased to 91%, with a false-positive rate of 14%.86

118

Fetal Cardiology

Screening examination of low-risk patients There has been increasing demand for the assessment of the fetal heart during the routine screening program for fetal abnormalities, as about 90% of fetuses with congenital heart disease present without any known risk factors, warranting detailed fetal echocardiography. Approximately 50% of cardiac malformations are considered easily amenable to surgery, while the other half, i.e. 4/1000, are considered severe. These account for approximately 20% of neonatal deaths; 50% of infant deaths are attributed to congenital anomalies. Thus, cardiac anomalies have at birth a 6.5 times higher incidence than that of chromosomal anomalies and are four times as common as neural tube defects. Screening programs have been established in many countries, with conflicting results. The common criteria for screening are much more easily fulfilled by fetal echocardiographic screening compared to screening for chromosomal anomalies and neural tube defects, as in the latter the intention is mostly the diagnosis of anomalies with subsequent pregnancy termination. According to the World Health Organization, the value of any screening is assessed in terms of the following criteria: 1. The underlying disease should have a sufficient prevalence and severity. With a birth prevalence of 8/1000, congenital heart defects represent the most common congenital malformations. Furthermore, it is the most lifethreatening condition in the first month of life, accounting for approximately 20% of perinatal deaths.87–89 2. The disease should have a fixed spectrum of symptoms. A fetal heart abnormality is likely if an abnormal four-chamber view or abnormal cardiac position is present. Exclusion of cardiac lesions is recommended in all malformations frequently associated with heart defects, in fetuses with increased nuchal translucency ≥ 3.5 mm, in the presence of arrhythmia, nonimmune fetal hydrops, and fetal growth restriction (Table 9.1). 3. The screening methods should be simple and acceptable. With appropriate training the four-chamber view is the easiest of the echocardiographic views to obtain. However, it has to be interpreted correctly.90 The general use of ultrasound and the screening for malformations is widely accepted because this offers reassurance in most pregnancies and has been considered safe for both mother and fetus.91,92 4. Screening should be accurate, with a low incidence of false-positive and false-negative results. Fetal echocardiography performed by skilled operators is known to be a reliable method for the diagnosis of congenital heart disease.3,5,93 However, most of these studies with high detection rates have been carried out in university hospitals with level-three experience.27,94

Rather disappointing results have been reported in screening a low-risk population outside these referral centers.7,95–97 While false-positive diagnoses are rarities in most studies,3,94 some have had extremely high falsenegative rates.95,96 Therefore, continued educational support in fetal echocardiography is a critical issue in improving both sensitivity and specificity.5,90,98 5. Methods of confirmation and further follow-up observation of the disease should be available. In referring the patient to appropriate centers this criterion can be met. In all fetuses with predicted heart disease, the diagnoses should be confirmed postpartum by echocardiography, by cardiac catheterization, or at cardiac surgery. In cases of termination of pregnancy, intrauterine fetal death, or neonatal death, autopsy results should be obtained.5 6. The disease should be treatable. There are therapeutic options for most cardiac anomalies. The management options will depend on the type of cardiovascular anomaly, the gestational age at diagnosis, the associated major malformations, and the ethical considerations of the patient. Counseling involving a multidisciplinary team should provide information on the cardiac anomaly, its natural history, and the further management possibilities. However, non-directive counseling of parents can only barely be a reality, as personal opinion and past experiences of the counselor enter counseling decisions.99 Most parents’ experience conflicts with decisions such as terminating or continuing the pregnancy, and choosing aggressive cardiac surgery or deciding against surgical intervention. In series of prenatally diagnosed CHD, generally with a poorer prognosis than CHD diagnosed after birth, about 30–40% of parents elected to interrupt the affected pregnancy (Table 9.7). In particular, the prognosis is unfavorable if there are chromosomal and non-chromosomal extracardiac anomalies and if the CHD is associated with fetal congestive heart failure, especially in fetuses with atrioventricular septal defect, left atrial isomerism, and complete heart block. Screening using the four-chamber view yields a more severe spectrum of CHD than is found in unselected live births.55,80 However, owing to the increasing detection rate in fetal cardiac scanning for isolated and minor cardiac defects and also owing to significant improvements in interventional cardiology and cardiac surgery during recent decades, the percentage of parents opting for termination of pregnancy has been continuously decreasing in most centers.99,104 For the most serious CHD, opinions on treatment options amongst pediatric cardiologists and pediatric cardiosurgeons differ widely. Non-directive counseling with pregnancy termination is one option, with the Norwood operation and heart transplantation being alternatives in hypoplastic left heart syndrome. For other cardiac lesions with predominant palliative treatment, opinions differ, as do the accepted treatment opportunities in various countries. As long-term prognosis of some operation

Indications for fetal echocardiography

Table 9.7

119

Outcome of CHD (studies with at least 100 fetuses included only) Cases of continuing pregnancy (n = 2291)

Study Sharland et al, 1991100 Respondek et al, 199451 Yagel et al, 19973 Fesslová et al, 199957 Paladini et al, 200252 Brick and Allan, 2002101 Levi et al, 2003102 Fesslová et al, 2003103 Total

Number of fetuses

Abnormal karyotype

Extracardiac anomaly

TOP

442

42

71

220

100

14

42

168

16

847

Death postpartum

Alive*

57

87

78

26

5

24

45

10

25

15

25

103

149

162

245

72

259

271

400

104

118

150

16

85

149

408

39

63

98

19

73

185

271

50

135

77

43

47

104

670

98

159

174

50

167

279

3306

512 (16%)

760 (23%)

1015 (31%)

Intrauterine fetal death

277 (12%)

767 (34%)

Lost/ ongoing

33

1214 (53%)

*Variable from 6 days after birth to median follow-up of 3 years. TOP, termination of pregnancy.

techniques (Fontan operation, Norwood operation, heart transplantation) is not fully established and there is the expectation of further innovations in the therapy of complex cardiac anomalies, counseling is even more difficult.99,104,105 In this context, all previous studies on the prognosis of prenatally diagnosed cardiac anomalies are not helpful, since the results were influenced by many factors, including patient selection. Apart from fetuses with hydrops, associated anomalies, heart failure, and growth restriction, screening based on the four-chamber view mostly detects cardiac anomalies with an unfavorable prognosis, which present not only the most severe lesions, but also situations rarely observed postnatally. These include tricuspid valve dysplasia, Ebstein’s anomaly, left atrial isomerism with atrioventricular septal defect, complete heart block, and cardiac tumors.55,58,106 If termination of pregnancy is excluded, survival rates of 40–60%, including those after surgery, are reported.3,101,103 Associated anomalies and chromosome disorders are poor prognostic factors associated with a higher intrauterine and postpartum mortality rate.51,57,80 The outcome of several studies of congenital heart disease is summarized in Table 9.7. 7. Screening should result in an improved outcome and a positive cost–benefit ratio. Cost–benefit and costeffectiveness analyses from a background of hightechnology medicine seem artificial, since in these circumstances the possible improvement in mortality and long-term prognosis in a few children justifies the

expense of prenatal diagnosis. Prenatal diagnosis of a cardiac anomaly is thought to optimize perinatal management of the affected child, as knowledge of the condition should help to avoid the hazards of neonatal transfer and clinical worsening with the possibility of hypoxia and acidosis causing multiorgan failure and long-term neurological damage, which may otherwise result from mismanagement due to postpartum delays. This is of particular importance for anomalies dependent on a patent ductus arteriosus, or other critical anomalies requiring immediate therapeutic and palliative interventions such as prompt infusion of prostaglandin E1, early balloon atrial septostomy, or balloon valvuloplasty.11,107–110 However, survival advantage as a result of prenatal diagnosis has been difficult to demonstrate conclusively.111,112 Retrospective studies on the outcome of all neonates with CHD could not demonstrate marked improvements in the early physiologic condition, the length of hospitalization, or the in-hospital mortality rate by prenatal diagnosis and following delivery and management in a tertiary care center.9,112,113 However, most postnatal series do not take into account neonates with CHD who die without diagnosis, neonates with postnatal diagnosis but dying before reaching the hospital or after early discharge, and cases considered unsuitable for cardiac surgery because of low birth weight or serious extracardiac anomalies.9,105,106,113 Also, the possibility of pregnancy termination after prenatal diagnosis may cause the prenatal selection of a group of

120

Fetal Cardiology

parents predisposed towards more aggressive neonatal management even in the most serious CHD.112,114 In specific lesions, such as hypoplastic left heart (HLH), transposition of the great arteries, and coarctation of the aorta, better outcomes resulting from prenatal diagnosis of CHD, and delivery and postnatal care at a specialized center could clearly be demonstrated. This applies especially to pre- and postoperative morbidity, but not uniformly to mortality.10,109,115,116 In neonates with dextrotransposition of the great arteries, studies could demonstrate that the preoperative mortality109 and preoperative morbidity,9,109 especially the appearance of critical hemodynamic compromise with severe hypoxemia, profound metabolic acidosis, and multiorgan failure, could be significantly decreased if transposition of the great arteries was diagnosed before birth.9,109 A lower postoperative mortality has also been observed in one of these series.109 Inadequate interatrial mixing might require urgent balloon atrial septostomy shortly after birth in around 10% of newborns with transposition of the great arteries.109 In the small high-risk group of neonates with dextrotransposition of the great arteries, the complicated preoperative course as a result of a restrictive foramen ovale, ductus arteriosus constriction, or pulmonary hypertension may already be predicted by fetal echocardiography.110,117 Accordingly, prenatal diagnosis of coarctation of the aorta has been shown to reduce preoperative hemodynamic instability and improve survival.116 For fetuses with hypoplastic left heart syndrome, improvement of survival after prenatal diagnosis has been shown for neonates with cardiac defects amenable to biventricular repair, but not for neonates with univentricular hearts.112 A lower surgical mortality of first-stage Norwood surgery after prenatal diagnosis has been reported in one study,115 whereas other authors could not show any reduction of mortality by prenatal diagnosis of hypoplastic left heart syndrome.9,10,108,118 Recently, a small study reported improved hospital survival in cases with HLH which underwent intrauterine left atrial decompression for restrictive atrial septum.11 Most studies, however, clearly demonstrated a decreased preoperative morbidity, with avoidance of severe hemodynamic compromise by hypoxemia, profound metabolic acidosis, multiorgan failure, and neurologic events.9,10,108,118 It can be speculated that a good preoperative condition resulting from prenatal diagnosis, optimized perinatal care, and further improvements in the surgical treatment and extracorporeal bypass technique may avoid significant long-term neurological impairment in infants with hypoplastic left heart syndrome after the Norwood procedure,119 in which the first-stage survival rate in full-term neonates with ideal anatomy and without other anomalies is above 80%.10,120 It is important to realize that the largest decrease in survival after Norwood palliation occurs in the first month, and is

caused by various factors such as non-chromosomal and chromosomal extracardiac anomalies, prematurity, and additional intracardiac findings such as a restrictive foramen ovale resulting in severe preoperative obstruction to pulmonary venous return or severe tricuspid dysplasia.114,120 In some of these cases, neonatal heart transplantation may be an alternative option. Therefore, at the time of prenatal diagnosis, detailed cardiac and extracardiac assessment is an important prerequisite for effective counseling of parents and for optimizing perinatal management. In order to prevent impaired left ventricular growth and to avoid a postnatal univentricular situation, intrauterine transthoracic balloon valvuloplasty has been advocated in critical aortic stenosis, but success has been limited so far.12,121,122 It is obvious that postnatal treatment in complex CHD is more expensive than the average cost of termination of pregnancy. This option in complex uncorrectable cardiac anomalies and those associated with severe chromosomal and non-chromosomal extracardiac anomalies drastically cuts the cost of neonatal management by avoidance of unnecessary cardiologic and cardiosurgical intervention, and might contribute to a decrease in perinatal mortality and morbidity. A considerably lower incidence of hypoplastic left heart in liveborn children,123,124 as well as of pulmonary atresia with intact ventricular septum,125 has already been demonstrated as a result of sonographic screening. However, there is only a small decrease, by 2%, in the overall incidence of live births with cardiac anomalies, leading to a small reduction of infant deaths due to congenital cardiac anomalies and cardiosurgical procedures for congenital cardiac anomalies if the detection rate of screening for cardiac anomalies remains low.126 If the detection rate of fetal cardiac screening, even if performed as simple four-chamber view screening, were higher, a profound effect on the prevalence and spectrum of CHD at term would be seen, because many affected pregnancies are terminated if the serious CHD is diagnosed early in gestation127 (Table 9.7). In 1995, Rustico and co-workers calculated the total cost of screening for cardiac anomalies in 7024 women (repeated twice in 15% of cases) as being US$323 104, based on US$40 for a 30-minute ultrasound examination of the fetus, including two-dimensional echocardiography.107 Consequently, the cost for each cardiopathy diagnosed correctly (23 cases) was US$14 048. The cost averted by each severe cardiopathy resulting in pregnancy termination (13 cases) was US$24 854. In the United States the average hospital cost for newborns with hypoplastic left heart is US$57 400 (range US$0–759 000), for first-stage surgical reconstruction is US$126 600 (range US$5600–551 000), and is US$126 695 for cardiac transplantation.128 Having spared a family of a child with complex congenital heart disease, with many social, psychological, and emotional consequences, is an emotive subject, but has to be taken

Indications for fetal echocardiography

into consideration. Nevertheless, to calculate cost savings in favor of pregnancy termination seems questionable and ethically unacceptable. A potential advantage of prenatal diagnosis of heart disease is to allow planning of the delivery in a unit with all neonatal facilities, to avoid emergency transfer and separation from parents in this situation. A last profound benefit of prenatal diagnosis in general, and sonographic screening for cardiac anomalies in particular, which is generally not considered in such cost-effectiveness analyses, is the exclusion of an anomaly for the vast majority of pregnant women, providing reassurance and reducing anxiety. The associated relief facilitates bonding between the mother and the unborn fetus.

Four-chamber view screening Many screening models for cardiac anomalies based on demonstration of the four-chamber view have been incorporated into level 1 screening in many countries.129 It has been suggested that this would result in a detection rate of 2 per 1000 heart defects. The influence of gestational age in depicting a correct four-chamber view has been demonstrated by several investigators. Before 18 weeks of gestation, the four chambers cannot be seen adequately in a large proportion of fetuses. For this reason, and as some lesions present secondary structural changes resulting from in utero development, screening between 18 and 23 weeks (preferentially ≥ 20 weeks) of gestation using the four-chamber view improves the identification considerably. At this time the four-chamber view can be obtained in 95–98% during the first examination, generally within 1–2 minutes.5,98,130 With appropriate educational support and practical training the examiner is sufficiently skilled to obtain optimal images, reducing false-negative and in particular false-positive results.6,95,98,131 Higher-frequency probes will improve the likelihood of recognition of small defects: a ≥ 5-MHz transducer generally seems to be superior to a 3.5-MHz probe. Marked hampering of ultrasound transmisson by adipose tissue, scar tissue due to previous abdominal surgery, and, more rarely, fetal position and oligohydramnios have been reported as the main causes for unsatisfactory imaging in the four-chamber view.132 However, screening by using the four-chamber view reduces the examination to a small part of the fetal heart. This includes demonstration of both atria, the interatrial septum with the valve of the foramen ovale beating into the left atrium, both atrioventricular valves, which open and close, and the interventricular septum with the inflow tracts. Furthermore, the atrial and ventricular size as well as wall thickness can be assessed. Lesions usually associated with abnormal four-chamber view images are listed in Table 9.8. However, many conotruncal anomalies are potentially not detectable if screening is limited to the four-chamber view only. In addition, the cardiac axis

121

Table 9.8 Demonstration of cardiac anomalies by primary and secondary changes in the four-chamber view in the second trimester. Modified from references 98 and 104 Cardiac anomalies with abnormal four-chamber view At the atrioventricular junction • mitral atresia • tricuspid atresia • atrioventricular septal defect • severe Ebstein’s anomaly, tricuspid dysplasia At the ventricular–systemic junction • aortic atresia • severe aortic stenosis • pulmonary atresia with intact interventricular septum • severe pulmonary stenosis At the venous–atrial junction • complete anomalous pulmonary venous drainage without pulmonary venous obstruction Others • severe coarctation of the aorta • interrupted aortic arch • double inlet ventricle • large ventricular septal defect • large atrial septal defect Cardiac anomalies with usually normal four-chamber view • transposition of the great arteries • tetralogy of Fallot • pulmonary atresia with ventricular septal defect • double outlet right ventricle • truncus arteriosus communis • mild Ebstein’s anomaly • small and/or outlet ventricular septal defect • atrial septal defect of secundum type • mild/moderate aortic stenosis • mild/moderate pulmonary stenosis • mild/moderate coarctation of the aorta • partial anomalous pulmonary venous drainage

and position should be analyzed. The normal left-sided deviation of the heart is about 45 ± 20° (2 standard deviations (SD)).26 Some conotruncal malformations, often associated with chromosomal anomalies, may result in an increased left-sided deviation.133,134 Displacing intrathoracic anomalies and abdominal wall defects may also result in an abnormal cardiac position detectable in the fourchamber view. Studies on the efficacy of screening using the fourchamber view for the detection of cardiac malformations have differed widely, with sensitivities ranging between 10%96,97 and 80%.93,98,135 Buskens and co-workers136 described differences and deficiencies of various screening studies for the detection of fetal cardiac anomalies which are similar to those of many other screening studies for fetal anomalies. Some of the major differences are addressed here.

122

Fetal Cardiology

1. The reported incidence of congenital cardiac defects in the study populations was between 3 and 13 per 1000 births (Table 9.9). With an expected incidence of eight cardiac anomalies in 1000 live births it seems extremely unlikely that these differences reflect true incidences of cardiac anomalies. 2. Depending on the definition of a cardiac anomaly, smaller defects, such as muscular ventricular septal defects or mild valve dysfunction, were included in some and excluded in other studies. This holds true also for lesions considered as physiologic in utero, including the persistent patent ductus arteriosus and the type II atrium septum defect in the fossa ovalis region. The detection rate must be significantly lower if cases of ventricular septal defects and type II atrial septal defects are included, as demonstrated by Hafner and co-workers, whose detection rate rose from 43.8% for all CHD to 74.3%, if ventricular and atrial septal defects (type II) were excluded.138 However, the informative and important differentiation into severe and mild defects used in some studies should

not be applied uncritically to the efficacy assessment of prenatal screening. Mild cardiac defects are of equal importance, as they might be associated with chromosomal and non-chromosomal anomalies. 3. The studied population was different. This was partially based on selection factors – some studies reported on screening results in the general population, others included a mixed patient population of low-risk as well as high-risk patients. Tertiary centers seem to have had a higher detection rate, incorporating a more homogeneous level of expertise, better equipment, longer examination time, and a mostly preselected study population with a higher incidence of cardiac anomalies.5,140 4. Some studies were designed as retrospective and some as prospective studies. 5. The examination periods varied in studies. Although the time between 18 and 24 weeks of gestation is generally accepted as optimal for the assessment of the fetal heart, the majority of examinations were performed from 18 weeks onwards and some even as early as 16 weeks

Table 9.9 Data on different screening studies with detection rates of CHD Study Sharland and Allan, 199298 Vergani et al, 1992135 Buskens et al, 199696 Todros et al, 1997137 Westin, 200654

Incidence/ 1000

Population risk

Level of experience

Weeks of pregnancy

3

Mixed

1

—

5 8 5 8

Mixed Low Low Low

1 1 1 1

18–22 16–24 18–22 18

1

16–22

1–2

—

1–2

16–22 20–22

Tegnander et al, 199597

12

Low

Ott, 19957

12

Tegnander et al, 200680 Rustico et al, 1995107

14

Low High risk Low

9

Low

2

Hafner et al, 1998138 Carvalho et al, 20025

13 4

Low Low

2 2–3

Oggè et al, 2006139 Wong et al, 2003140

9 7

Low Low

Achiron et al, 199294

4

Low

2 2 3 3

18–24

Stümpflen et al, 199627

7

Mixed

3

18–28

Yagel et al, 19973

8

Mixed

3

13–16/20–22

(16–)22* 18–23 > 18 —

*Some fetuses were examined at 16–18 weeks, but the vast majority at 22 weeks. 4 CV, four-chamber view.

Cardiac examination 4 CV

Detection (%)

69 (detectable in 4 CV) 4 CV 81 4 CV 5 4 CV 15 4 CV 15 (major CHD only) Anomaly screening 18 (major CHD only) 4 CV 26 (major CHD only) 4 CV + left outflow tract 14 Extended echo 62 4 CV + outflow tract 15 (57 of major CHD) 4 CV + outflow tract 35 (61 of major CHD) Extended echo ± color 44 4 CV + outflow tract 75 (major CHD only) 4 CV + outflow tract 65 4 CV + outflow tract 21 4 CV + outflow tract 61 4 CV 48 Extended echo 78 4 CV 47 Extended echo + color 88 Extended echo 81

Indications for fetal echocardiography

of gestation. Comparison of studies is difficult, as the lower sensitivity as well as a higher incidence of cardiac anomalies at earlier gestational ages contributes to different results. 6. There were marked differences in the level of experience of the examiners, ranging from a 2-day instruction course for demonstration of the four-chamber view to many years of echocardiographic experience. 7. In some studies cardiac views were limited to the fourchamber view; others included visualization of the left ventricular outflow, complete two-dimensional echocardiography, or even color Doppler echocardiography (Table 9.9). 8. A critical factor that especially applied to cardiac anomalies was the availability of a thorough postnatal evaluation with respect to examination methods as well as the follow-up period.124 In certain studies, clinical examination was not performed in all newborns. Postnatal echocardiography was restricted to selected cases, whereas in other studies it was generally performed. The frequency of postnatal detection of atrial and ventricular septal defects and pulmonary stenoses is profoundly enhanced by the use of color Doppler echocardiography.124 There were different observation periods and follow-up information. The postmortem examination of spontaneous abortions, stillbirths, and postnatal deaths was incomplete in many studies. Major cardiac anomalies are usually detected in the first week of life, but the majority of defects – which are usually not critical – are commonly detected after months. Optimal follow-up for complete detection of all cardiac anomalies, which has rarely been performed in published studies, would include a pediatric cardiology follow-up examination with colorcoded Doppler echocardiography at the end of the first week of life and a repeated examination at 4–6 months. The findings of second-trimester ‘first level’ ultrasound studies in unselected populations by demonstration of the four-chamber view are sobering, as reported detection rates are between 5 and 20%.7,54,96,97 Tegnander and colleagues detected only 10% of all cardiac anomalies and 26% of critical ones, utilizing the four-chamber view in the second phase between 16 and 22 weeks of gestation.97 This is consistent with prospective sonographic screening studies for anomalies in general, in which a correspondingly small percentage of anomalies was detected.95,129,131,141,142 Without firm integration of the four-chamber view into sonographic screening, the detection rate of congenital cardiac defects is even lower.95,97 Higher sensitivity of screening with the four-chamber view and detection rates between 48 and 69% in unselected and mixed-risk populations has been reported in other centers, presumably owing to the higher level of training of the examiners.27,94 This means that 2 or 3 in 1000 live births or 1:300 to 1:500, respectively, have a cardiac anomaly that can be detected by demonstration of the four-chamber

123

view. The spontaneous loss rate between the second trimester and term, especially marked in fetuses with chromosomal anomalies, will result in slightly higher rates and different relative distributions of cardiac anomalies in the second trimester when compared to those in live births. To date, multiple studies have shown a further increase of 20–30% in the detection rate of cardiac anomalies by incorporating visualization of the outflow tracts and the great arteries into the examination, compared to using the four-chamber view alone.27,90,94,107,143 The results of different studies are summarized in Table 9.9. If we look at the different types of cardiac anomaly (Table 9.8), it becomes apparent that some are always detectable in the four-chamber view and some only with certain localization and size (e.g. ventricular septal defects). Others will manifest in the course of pregnancy in lesions of sufficient severity (pulmonary stenosis, aortic stenosis, coarctation of the aorta). Lesions in the region of the outflow tracts and the large vessels are usually not detectable via the four-chamber view, including some severe anomalies which become symptomatic early after birth (transposition of the great arteries, tetralogy of Fallot, pulmonary stenosis and pulmonary atresia with ventricular septal defect, double outlet right ventricle and truncus arteriosus communis). The latter defects account for 25% of all congenital cardiac anomalies in live births. Almost never diagnosable in the four-chamber view are the majority of ventricular septal defects, pulmonary stenoses, atrial septal defects of the secundum type, and, naturally, the patent ductus arteriosus which account for 30%, 10%, 10%, and 10% of all cardiac anomalies, respectively.2,16,124 If the last prenatally ‘physiologic’ anomalies are excluded, a maximum of 40–50% of congenital anomalies can be diagnosed by the four-chamber view. This potential, however, is not nearly approached in most of the general screening settings. Realistic estimates state that currently only 20% of those anomalies detectable in the second trimester fourchamber view are being diagnosed.54,96,126 but a significant increase in detection rate is seen in areas with a high training level of all participating examiners.127

Additional visualization of the ventricular outflow tracts Prenatal detection of the transposition of the great arteries would be of particular importance, since an improvement of long-term prognosis and mortality rates can be expected from prenatal diagnosis and the optimization of perinatal management. Furthermore, ventricular septal defects of the outflow tract, double outlet right ventricle, tetralogy of Fallot, and truncus arteriosus communis are common cardiac anomalies, found in the majority of fetuses with trisomies 13 and 18. These defects can be diagnosed

124

Fetal Cardiology

Table 9.10 Yield of CHD by indication for fetal echocardiography. Based on the combined data of references 5, 56, 144, and 145 Indication Family history of CHD Maternal diabetes Teratogen exposure Unsatisfactory/abnormal cardiac views Arrhythmia Extracardiac anomalies including hydrops Intracardiac echogenic focus Aneuploidy Increased nuchal translucency with normal karyotype Total

Referrals 2374 (26.2%) 1442 (16.0%) 666 (7.4%) 959 (10.6%) 1391 (15.4%) 1511 (16.7%) 243 (2.7%) 344 (3.8%) 105 (1.2%) 9035

only by imaging of the outflow tracts of both ventricles. Therefore, the examiner should be trained to visualize the four-chamber view as well as the ventricular outflow tracts with the origin and course of the great arteries, as this has been shown to improve prenatal detection rates considerably.5

Conclusions Many reasons stated above make the prenatal diagnosis of as many cardiac anomalies as possible desirable from the viewpoint of the perinatologist and the perinatal cardiologist. In particular, an improvement of postnatal outcome could be demonstrated for some subgroups of cardiac defects. Therefore, detailed fetal echocardiography should be performed in all high-risk pregnancies such as those with a positive family history for cardiac defects, with maternal diabetes, and with already detected cardiovascular and extracardiac fetal anomalies (Table 9.10). As nearly 90% of all pregnancies with fetuses identified as having a cardiac anomaly are not initially considered to be at increased risk, and as customary prenatal sonograms almost always miss an isolated cardiac defect, adequate detection of cardiac anomalies can be achieved only by cardiac screening of all fetuses in the second trimester coupled with referral of women with affected fetuses to centers with specialist cardiac experience. Demonstration of the four-chamber view allows the detection of 40–50% of congenital cardiac anomalies in a low-risk group of patients. Additional visualization of the outflow tracts and the great arteries increases the rate to 65–70% and includes all critical cardiac defects, especially transposition of the great arteries. Current ultrasonographic technology is sufficient for this purpose. However, general cardiac screening performed by many different examiners of varying experience results in detection rates that fall far below the potential. The full potential of fetal echocardiographic

CHD 23 (2.9%) 20 (2.6%) 1 (0.1%) 453 (56.7%) 22 (2.7%) 230 (28.8%) 1 (0.1%) 42 (5.2%) 7 (0.9%) 799

Prevalence/1000 for indication 9.7 14 1.5 472 16 152 4 122 67 88

screening can be attained only with further improvement in the level of expertise of all persons involved in screening; this can be achieved by permanent training and liaison between the screening examiner and the specialized referral center. For all high-risk patients, early fetal echocardiography should be offered at 13–14 weeks of gestation followed by an additional echocardiographic examination at 20–22 weeks of gestation at specialized centers. By achieving detection of serious CHD earlier in fetal life, termination of an abnormal fetus is more acceptable. In this context the increased thickness of nuchal translucency at 11–14 weeks of gestation seems to be an important marker for CHD even in chromosomally normal fetuses (Table 9.10). On the other hand, prenatal diagnosis of congenital heart defects with adequate perinatal management has been shown to improve the short- and long-term outcome for critical heart defects, in particular for the dextrotransposition of the great arteries and hypoplastic left heart.

References 1. Perry LW, Neill CA, Ferencz C, Rubin JD, Loffredo CA. Infants with congenital heart disease: the cases. In: Ferenz C, Rubin JD, Loffredo CA, Magee CA, eds. Epidemiology of Congenital Heart Disease. The Baltimore–Washington Infant Study 1981–1989. Perspectives in Pediatric Cardiology. Armonk, NY: Futura, 1993; 4: 33–62. 2. Hoffman JIE, Christianson R. Congenital heart disease in a cohort of 19502 births with long-term follow-up. Am J Cardiol 1978; 42: 641–7. 3. Yagel S, Weissman A, Rotstein Z et al. Congenital heart defects: natural course and in utero development. Circulation 1997; 96: 550–5. 4. Allan LD. Cardiac anatomy screening: what is the best time for screening in pregnancy? Curr Opin Obstet Gynecol 2003; 15: 143–6. 5. Carvalho JS, Mavrides E, Shinebourne EA, Campbell S, Thilaganathan B. Improving the effectiveness of routine

Indications for fetal echocardiography

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

prenatal screening for major congenital heart defects. Heart 2002; 88: 387–91. Tegnander E, Eik-Nes SH. The examiner’s ultrasound experience has a significant impact on the detection rate of congenital heart defects at the second-trimester fetal examination. Ultrasound Obstet Gynecol 2006; 28: 8–14. Ott WJ. The accuracy of antenatal fetal echocardiography screening in high- and low-risk patients. Am J Obstet Gynecol 1995; 172: 1741–9. Randall P, Brealey S, Hahn S, Khan KS, Parsons JM Accuracy of fetal echocardiography in the routine detection of congenital heart disease among unselected low risk populations: a systematic review. BJOG 2005; 112: 24–30. Kumar RK, Newburger JW, Gauvreau K, Kamenir SA, Hornberger LK. Comparison of outcome when hypoplastic left heart syndrome and transposition of the great arteries are diagnosed prenatally versus when diagnosis of these two conditions is made only postnatally. Am J Cardiol 1999; 86: 1649–53. Mahle WT, Clancy RR, McGaurn SP et al. Impact of prenatal diagnosis on survival and early neurologic morbidity in neonates with the hypoplastic left heart syndrome. Pediatrics 2001; 107: 1277–82. Vida VL, Bacha EA, Larrazabal A et al. Hypoplastic left heart syndrome with intact or highly restrictive atrial septum: surgical experience from a single center. Ann Thorac Surg 2007; 84: 581–5; discussion 586. Kohl T, Breuer J, Heep A et al. Fetal transesophageal echocardiography during balloon valvuloplasty for severe aortic valve stenosis at 28+6 weeks of gestation. J Thorac Cardiovasc Surg 2007; 134: 256–7. Koren G, Edwards MB, Miskin M. Antenatal sonography of fetal malformations associated with drugs and chemicals: a guide. Am J Obstet Gynecol 1987; 156: 79–85. Nora JJ, Nora AH. Update on counseling the family with a first degree relative with a congenital heart defect. Am J Med Genet 1988; 29: 137–42. Zerres K, Gembruch U, Schwanitz G et al. Fetale Echokardiographie und klinische Genetik—Eine enge Wechselbeziehung. Z Kardiol 1990; 79: 96–106. Boughman JA, Neill CA, Ferencz C, Loffredo CA. The genetics of congenital heart disease. In: C Ferencz, Rubin JD, Loffredo CA, Magee CA, eds. Epidemiology of Congenital Heart Disease. The Baltimore–Washington Infant Study 1981–1989. Perspectives in Pediatric Cardiology. Armonk, NY: Futura, 1993; 4: 123–67. Nora JJ. Causes of congenital heart diseases: old and new modes, mechanisms, and models. Am Heart J 1993; 125: 1409–19. Pierpont ME, Basson CT, Benson W et al. Genetic basis for congenital heart defects: current knowledge: a scientific statement from the American Heart Congenital Cardiac Defects Committee, Council on Cardiovascular Disease in the Young. Circulation 2007; 115: 3015–38. Payne R, Johnson M, Grant J, Strauss A. Towards a molecular understanding of congenital heart disease. Circulation 1995; 91: 494–504. Calcagni G, Digilio MC, Sarkozy A, Dallapiccola B, Marino B. Familial recurrence of congenital heart disease: an overview and review of the literature. Eur J Pediatr 2007; 166: 111–16.

125

21. Gembruch U. Prenatal diagnosis of congenital heart disease. Prenat Diagn 1997; 17: 1283–98. 22. Copel JA, Pilu G, Kleinman CS. Congenital heart disease and extracardiac anomalies; associations and indications for fetal echocardiography. Am J Obstet Gynecol 1986; 154: 1121–32. 23. Cooper M, Enderlein MA, Dyson DC, Roge CI, Tarnoff H. Fetal echocardiography: retrospective review of clinical experience and an evaluation of indications. Obstet Gynecol 1995; 86: 577–82. 24. Gembruch U, Chaoui R. Möglichkeiten und Grenzen eines Screeningprogramms Pränatale Diagnostik fetaler Herzfehler durch Untersuchung von ‘High-risk’ und ‘Low-risk’-Kollektiven. Gynäkologe 1997; 30: 191–9. 25. Atzei A, Gajewska K, Huggon IC, Allan L, Nicolaides KH. Relationship between nuchal translucency thickness and prevalence of major cardiac defects in fetuses with normal karyotype. Ultrasound Obstet Gynecol 2005; 26: 154–7. 26. International Society of Ultrasound in Obstetrics and Gynecology. Cardiac screening examination of the fetus: guidelines for performing the ‘basic’ and ‘extended basic’ cardiac scan. Ultrasound Obstet Gynecol 2006; 27: 107– 13. 27. Stümpflen I, Stümpflen A, Wimmer M, Bernaschek G. Effect of detailed fetal echocardiography as part of routine prenatal ultrasonographic screening on detection of congenital heart disease. Lancet 1996; 348: 854–7. 28. Gembruch U, Hansmann M, Redel DA, Bald R. Fetal twodimensional Doppler echocardiography (colour flow mapping) and its place in prenatal diagnosis. Prenat Diagn 1989; 9: 535–47. 29. Sharland GK, Chita SK, Allan LD. The use of colour Doppler in fetal echocardiography. Int J Cardiol 1990; 28: 229–36. 30. Copel JA, Morotti R, Hobbins JC, Kleinman CS. The antenatal diagnosis of congenital heart disease using fetal echocardiography: is color flow mapping necessary? Obstet Gynecol 1991; 78: 1–8. 31. Gembruch U, Chatterjee M, Bald R, Redel DA, Hansmann M. Color flow mapping of fetal heart. J Perinat Med 1991; 19: 27–32. 32. Volpe P, Campobasso G, Stanziano A et al. Novel application of 4D sonography with B-flow imaging and spatiotemporal image correlation (STIC) in the assessment of the anatomy of pulmonary arteries in fetuses with pulmonary atresia and ventricular septal defect. Ultrasound Obstet Gynecol 2006; 28: 40–6. 33. Weiner Z, Lorber A, Shalev E. Diagnosis of congenital cardiac defects between 11 and 14 weeks’ gestation in highrisk patients. J Ultrasound Med 2002; 21: 23–9. 34. Carvalho JS. Fetal heart scanning in the first trimester. Prenat Diagn 2004; 24: 1060–7. 35. McAuliffe FM, Trines J, Nield LE et al. Early fetal echocardiography – a reliable prenatal diagnosis tool. Am J Obstet Gynecol 2005; 193: 1253–9. 36. Becker R, Wegner RD. Detailed screening for fetal anomalies and cardiac defects at the 11–13-week scan. Ultrasound Obstet Gynecol 2006; 27: 613–18. 37. Smrcek JM, Berg C, Geipel A et al. Detection rate of early fetal echocardiography and in utero development of congenital heart defects. J Ultrasound Med 2006; 25: 187–96.

126

Fetal Cardiology

38. Drose JA. Scanning indications and technique. In: Drose JA, ed. Fetal Echocardiography. Philadelphia: WB Saunders, 1998: 15–57. 39. Allan LD, Crawford DC, Chita CK et al. Familial recurrence of congenital heart disease in a prospective series of mothers referred for fetal echocardiography. Am J Cardiol 1986; 58: 334–7. 40. Boughman JA, Berg KA, Astemborski JA et al. Familial risks of congenital heart defect assessed in a populationbased epidemiologic study. Am J Med Genet 1987; 26: 839–49. 41. Siu S, Chitayat D, Webb G. Pregnancy in women with congenital heart defects: what are the risks? Heart 1999; 81: 225–6. 42. Gill HK, Splitt M, Sharland GK, Simpson JM. Patterns of recurrence of congenital heart disease. An analysis of 6,640 consecutive pregnancies evaluated by detailed fetal echocardiography. J Am Coll Cardiol 2003; 42: 923–9. 43. Burn J, Little J, Holloway S et al. Recurrence risk in offspring of adults with major heart defects: results from first cohort of British collaborative study. Lancet 1998; 351: 311–16. 44. Buskens E, Stewart PA, Hess J et al. Efficiency of fetal echocardiography and yield by risk category. Obstet Gynecol 1996; 87: 423–8. 45. Kuehl KS, Loffredo CA. Genetic and environmental influences on malformations of the cardiac outflow tract. Expert Rev Cardiovasc Ther 2005; 3: 1125–30. 46. Shields LF, Gan EA, Murphy HF et al. The prognostic values of hemoglobin A1c in predicting fetal heart disease in diabetic pregnancies. Obstet Gynecol 1993; 81: 954–7. 47. Smith RS, Comstock CH, Lorenz RP et al. Maternal diabetes mellitus: which views are essential for fetal echocardiography? Obstet Gynecol 1997; 90: 575–9. 48. Meyer-Wittkopf M, Simpson JM, Sharland GK. Incidence of congenital heart defects in fetuses of diabetic mothers: a retrospective study of 326 cases. Ultrasound Obstet Gynecol 1996; 8: 8–10. 49. Rouse B, Azen C, Koch R et al. Maternal Phenylketonuria Collaborative Study (MPKUCS) offspring: facial anomalies, malformations, and early neurological sequelae. Am J Med Genet 1997; 69: 89–95. 50. Ferencz C, Rubin JD, McCarter RJ. Maternal diabetes and cardiovascular malformations: predominance of double outlet right ventricle and truncus arteriosus. Teratology 1990; 41: 319–26. 51. Respondek ML, Binotto CN, Smit S et al. Extracardiac anomalies, aneuploidy and growth retardation in 100 consecutive fetal congenital heart defects. Ultrasound Obstet Gynecol 1994; 4: 272–8. 52. Paladini D, Russo MG, Teodoro A et al. Prenatal diagnosis of congenital heart disease in the Naples area during the years 1994–1999 – the experience of a joint fetal-pediatric cardiology unit. Prenat Diagn 2002; 22: 545–52. 53. Cullen S, Sharland GK, Allan LD, Sullivan IA. Potential impact of population screening for prenatal diagnosis of congenital heart disease. Arch Dis Child 1992; 67: 775–8. 54. Westin M, Saltvedt S, Bergman G et al. Routine ultrasound examination at 12 or 18 gestational weeks for prenatal detection of major congenital heart malformations? A

55.

56.

57.

58.

59.

60.

61.

62.

63.

64.

65.

66.

67.

68.

69.

randomised controlled trial comprising 36,299 fetuses. BJOG 2006; 113: 675–82. Allan LD, Sharland GK, Milburn A et al. Prospective diagnosis of 1006 consecutive cases of congenital heart disease in the fetus. J Am Coll Cardiol 1994; 23: 1452–8. Simpson LL. Indications for fetal echocardiography from a tertiary-care obstetric sonography practice. J Clin Ultrasound 2004; 32: 123–8. Fesslová V, Nava S, Villa L; Fetal, Cardiology Study Group of the Italian Society of Pediatric Cardiology. Evolution and long term outcome in cases with fetal diagnosis of congenital heart disease: Italian multicenter study. Heart 1999; 82: 594–9. Berg C, Geipel A, Kamil D et al. The syndrome of left isomerism: sonographic findings and outcome in prenatally diagnosed cases. J Ultrasound Med 2005; 24: 921–31. Pavlova M, Fouran JC, Drblik SP et al. Factors affecting the prognosis of Ebstein’s anomaly during fetal life. Am Heart J 1998; 135: 1081–5. Meise C, Germer U, Gembruch U. Arterial Doppler ultrasound in 115 second- and third-trimester fetuses with congenital heart disease. Ultrasound Obstet Gynecol 2001; 17: 398–402. Fesslová V, Villa L, Nava S et al. Spectrum and outcome of atrioventricular septal defect in fetal life. Cardiol Young 2002; 12: 18–26. Gembruch U, Meise C, Germer U, Berg C, Geipel A. Venous Doppler ultrasound in 146 fetuses with congenital heart disease. Ultrasound Obstet Gynecol 2003; 22: 345–50. Huggon IC, Ghi T, Cook AC et al. Fetal cardiac abnormalities identified prior to 14 weeks’ gestation. Ultrasound Obstet Gynecol 2002; 20: 22–9. Makrydimas G, Sotiriadis A, Huggon IC et al. Nuchal translucency and fetal cardiac defects: a pooled analysis of major fetal echocardiography centers. Am J Obstet Gynecol 2005; 192: 89–95. Hyett JA, Moscoso G, Papapanagiotou G et al. Abnormalities in the heart and great arteries in chromosomally normal fetuses with increased nuchal translucency thickness at 11–13 weeks of gestation. Ultrasound Obstet Gynecol 1996; 7: 245–50. Snijders RJM, Noble P, Sebire N, Souka A, Nicolaides KH. UK multicentre project on assessment of risk of trisomy 21 by maternal age and fetal nuchal-translucency thickness at 10–14 weeks of gestation. Lancet 1998; 352: 343–6. Gasiorek-Wiens A, Tercanli S, Kozlowski P et al; German-Speaking Down Syndrome Screening Group. Screening for trisomy α1 by fetal nuchal translucency and maternal age: a multicenter project in Germany, Austria and Switzerland. Ultrasound Obstet Gynecol 2001; 18: 645–8. Souka AP, Von Kaisenberg CS, Hyett JA et al. Increased nuchal translucency with normal karyotype. Am J Obstet Gynecol 2005; 192: 1005–21. Hyett JA, Perdu M, Sharland GK et al. Increases nuchal translucency at 10–14 weeks of gestation as a marker for major cardiac defects. Ultrasound Obstet Gynecol 1997; 10: 242–6.

Indications for fetal echocardiography

70. Haak MC, van Vugt JM. Pathophysiology of increased nuchal translucency: a review of the literature. Hum Reprod Update 2003; 9: 175–84. 71. Carvalho JS. Opinion. The fetal heart or the lymphatic system or ...? The quest for the etiology of increased nuchal translucency. Ultrasound Obstet Gynecol 2005; 25: 215–20. 72. Ghi T, Huggon C, Zosmer N, Nikolaides KH. Incidence of major structural cardiac defects associated with increased nuchal translucency but normal karyotype. Ultrasound Obstet Gynecol 2001; 18: 610–14. 73. Lopes LM, Brizot ML, Lopes MA et al. Structural and functional cardiac abnormalities identified prior to 16 weeks’ gestation in fetuses with increased nuchal translucency. Ultrasound Obstet Gynecol 2003; 22: 470–8. 74. Bahado-Singh RO, Wapner R, Thom E et al. Elevated first-trimester nuchal translucency increases the risk of congenital heart defects. Am J Obstet Gynecol 2005; 192: 1357–61. 75. Ganapathy R, Guven M, Sethna F et al. Natural history and outcome of prenatally diagnosed cystic hygroma. Prenat Diagn 2004; 24: 965–8. 76. Papp C, Beke A, Mezei G et al. Prenatal diagnosis of Turner syndrome: report on 69 cases. J Ultrasound Med 2006; 25: 711–17. 77. Manning N, Archer N. A study to determine the incidence of structural congenital heart disease in monochorionic twins. Prenat Diagn 2006; 26: 1062–4. 78. Paladini D, Calabro R, Palmieri S, D’Andrea T. Prenatal diagnosis of congenital heart disease and fetal karyotyping. Obstet Gynecol 1993; 81: 679–82. 79. Gembruch U, Baschat AA, Knöpfle G, Hansmann M. Results of chromosomal analysis in fetuses with cardiac anomalies diagnosed by first- and early second-trimester echocardiography. Ultrasound Obstet Gynecol 1997; 10: 391–6. 80. Tegnander E, Williams W, Johansen OJ et al. Prenatal detection of heart defects in a non-selected population of 30,149 fetuses – detection rates and outcome. Ultrasound Obstet Gynecol 2006; 27: 252–65. 81. Geipel A, Gembruch U. Genetic ultrasound scan – second trimester. In: van Vugt JMG, Shulman LP, eds. Prenatal Medicine. New York: Informa Healthcare, 2006: 103–19. 82. Vintzileos AM, Guzman ER, Smulian JC et al. Down syndrome risk estimation after normal genetic sonography. Am J Obstet Gynecol 2002; 187: 1226–9. 83. Yeo L, Vintzileos AM. The use of genetic sonography to reduce the need for amniocentesis in women at high-risk for Down syndrome. Semin Perinatol 2003; 27: 152–9. 84. DeVore GR. Is genetic ultrasound cost-effective? Semin Perinatol 2003; 27: 173–82. 85. Paladini D, Tartaglione A, Agangi A et al. The association between congenital heart disease and Down syndrome in prenatal life. Ultrasound Obstet Gynecol 2000; 15: 104–8. 86. DeVore GR. The role of fetal echocardiography in genetic sonography. Semin Perinatol 2003; 27: 160–72. 87. Lee K, Khoshnood B, Chen L et al. Infant mortality from congenital malformations in the United States, 1970–1997. Obstet Gynecol 2001; 98: 620–7.

127

88. Chew C, Halliday JL, Riley MM, Penny DJ. Populationbased study of antenatal detection of congenital heart disease by ultrasound examination. Ultrasound Obstet Gynecol 2007; 29: 619–24. 89. Centers for Disease Control and Prevention (CDC). Hospital stays, hospital charges, and in-hospital deaths among infants with selected birth defects – United States, 2003. MMWR Morb Mortal Wkly Rep 2007; 56: 25–9. 90. Chaoui R. The four-chamber view: four reasons why it seems to fail in screening for cardiac abnormalities and suggestions to improve detection rate. Ultrasound Obstet Gynecol 2003; 22: 3–10. 91. Saari-Kemppainen A, Karjarlainen O, Yostalo P, Keinonen OP. Ultrasound screening and perinatal mortality: controlled trial of systematic one stage screening in pregnancy. Lancet 1990; 336: 387–91. 92. Newnham JP, Evans SF, Michael CA, Stenley FJ, Landau LJ. Effects of frequent ultrasound during pregnancy: a randomised controlled trial. Lancet 1993; 342: 887–91. 93. Copel JA, Pilu G, Green J, Hobbins JC, Kleinman CS. Fetal echocardiographic screening for congenital heart disease: the importance of the four-chamber view. Am J Obstet Gynecol 1987; 157: 648–55. 94. Achiron R, Glaser J, Gelernter I, Hegesh J, Yagel S. Extended fetal echocardiographic examination for detecting cardiac malformations in low risk pregnancies. Br Med J 1992; 304: 671–4. 95. Montana E, Khoury MJ, Cragan JD et al. Trends and outcomes of prenatal diagnosis of congenital cardiac malformations by fetal echocardiography in a well defined birth population. Atlanta, Georgia, 1990–1994. J Am Coll Cardiol 1996; 28: 1805–9. 96. Buskens E, Grobbee DE, Wladimiroff JW, Hess J. Routine screening for congenital heart disease: a prospective study in the Netherlands. In: Wladimiroff JW, Pilu G, eds. Ultrasound and the Fetal Heart. Carnforth, UK: Parthenon Publishing, 1996: 71–80. 97. Tegnander E, Eik-Nes SH, Johansen OJ et al. Prenatal detection of heart defects at the routine fetal examination at 18 weeks in a non-selected population. Ultrasound Obstet Gynecol 1995; 5: 372–80. 98. Sharland GK, Allan LD. Screening for congenital heart disease prenatally. Results of a 2 1/2-year study in the South East Thames region. Br J Obstet Gynaecol 1992; 99: 220–5. 99. Allan LD, Huggon IC. Counselling following a diagnosis of congenital heart disease. Prenat Diagn 2004; 24: 1136–42. 100. Sharland GK, Lockhart SM, Chita SK, Allan LD. Factors influencing the outcome of congenital heart disease detected prenatally. Arch Dis Child 1991; 66: 284–7. 101. Brick DH, Allan LD. Outcome of prenatally diagnosed congenital heart disease: an update. Pediatr Cardiol 2002; 23: 449–53. 102. Levi S, Zhang WH, Alexander S, Viart P, Grandjean H; The Eurofetus Study Group. Short-term outcome of isolated and associated congenital heart defects in relation to antenatal ultrasound screening. Ultrasound Obstet Gynecol 2003; 21: 532–8. 103. Fesslová V, Villa L, Kustermann A. Long-term experience with the prenatal diagnosis of cardiac anomalies in

128

104. 105. 106.

107.

108.

109.

110.

111.

112.

113.

114.

115.

116.

117.

118.

119.

120. 121.

Fetal Cardiology

high-risk pregnancies in a tertiary center. Ital Heart J 2003; 4: 855–64. Sharland G. Changing impact of fetal diagnosis of congenital heart disease. Arch Dis Child 1997; 77: P1–3. Simpson JM. Hypoplastic left heart. Ultrasound Obstet Gynecol 2000; 15: 271–8. Yates RS. The influence of prenatal diagnosis on postnatal outcome in patients with structural congenital heart disease. Prenat Diagn 2004; 24: 1143–9. Rustico MA, Benettoni A, D’Ottavio GD et al. Fetal heart screening in low-risk pregnancies. Ultrasound Obstet Gynecol 1995; 6: 313–19. Chang AC, Huhta JC, Yoon GY et al. Diagnosis, transport, and outcome in fetuses with left ventricular outflow tract obstruction. J Thorac Cardiovasc Surg 1991; 102: 841–8. Bonnet D, Coltri A, Butera G et al. Detection of transposition of the great arteries in fetuses reduces neonatal morbidity and mortality. Circulation 1999; 99: 916–18. Jouannic JM, Gavard L, Fermont L et al. Sensitivity and specificity of prenatal features of physiological shunts to predict neonatal clinical status in transposition of the great arteries. Circulation 2004; 110: 1743–6. Jaeggi ET, Sholler GF, Jones OD, Cooper SG. Comparative analysis of pattern, management and outcome of pre- versus postnatally diagnosed major congenital heart disease: a population-based study. Ultrasound Obstet Gynecol 2001; 17: 380–5. Copel JA, Tan ASA, Kleinman CS. Does a prenatal diagnosis of congenital heart disease alter short-term outcome? Ultrasound Obstet Gynecol 1997; 10: 237–41. Simpson LL, Harvey-Wilkes K, D’Alton ME. Congenital heart disease: the impact of delivery in a tertiary center on SNAP scores (scores for neonatal acute physiology). Am J Obstet Gynecol 2000; 182: 184–91. Allan LD, Apfel HD, Printz BF. Outcome after prenatal diagnosis of the hypoplastic left heart syndrome. Heart 1998; 79: 371–4. Tworetzky W, McElhinney DB, Reddy VM et al. Improved surgical outcome after fetal diagnosis of hypoplastic left heart syndrome. Circulation 2001; 103: 1269–73. Franklin O, Burch M, Manning N et al. Prenatal diagnosis of coarctation of the aorta improves survival and reduces morbidity. Heart 2002; 87: 67–9. Maeno YV, Kamenir SA, Sinclair B et al. Prenatal features of ductus arteriosus constriction and restrictive foramen ovale in d-transposition of the great arteries. Circulation 1999; 99: 1209–14. Eapen RS, Rowland DG, Franklin WH. Effect of prenatal diagnosis of critical left heart obstruction on perinatal morbidity and mortality. Am J Perinatol 1998; 15: 237– 42. Kern JH, Hinton VJ, Nereo NE et al. Early developmental outcome after the Norwood procedure for hypoplastic left heart syndrome. Pediatrics 1998; 102: 1148–52. Bove FJ. Surgical treatment for hypoplastic left heart syndrome. Jpn J Thorac Cardiovasc Surg 1999; 47: 47–56. Kohl T, Sharland G, Allan LD et al. World experience of percutaneous ultrasound-guided balloon valvuloplasty in human fetuses with severe aortic valve obstruction. Am J Cardiol 2000; 85: 1230–3.

122. Tworetzky W, Wilkins-Haug L, Jennings RW et al. Balloon dilation of severe aortic stenosis in the fetus: potential for prevention of hypoplastic left heart syndrome: candidate selection, technique, and results of successful intervention. Circulation 2004; 110: 2125–31. 123. Allan LD, Cool A, Sullivan I, Sharland GK. Hypoplastic left heart syndrome: effects of fetal echocardiography on birth prevalence. Lancet 1991; 337: 959–61. 124. Brenner JI. Prevalence and outcome of congenital heart disease in infancy: a 10-year population-based experience. In: Wladimiroff JW, Pilu G, eds. Ultrasound and the Fetal Heart. Carnforth, UK: Parthenon Publishing, 1996: 107–15. 125. Daubeney PFF, Sharland GK, Cook AC et al; UK and Eire Collaborative Study of PAIVS. Pulmonary atresia with intact ventricular septum. Impact of fetal echocardiography on incidence at birth and postnatal outcome. Circulation 1998; 98: 562–6. 126. Abu-Harb M, Wyllie J, Hey E, Richmond S, Wren C. Antenatal diagnosis of congenital heart disease and Down’s syndrome: the potential effect on practice of paediatric cardiology. Br Heart J 1995; 74: 192–8. 127. Bull C; British Paediatric Cardiac Association. Current and potential impact of fetal diagnosis on prevalence and spectrum of serious congenital heart disease at term in the UK. Lancet 1999; 354: 1242–7. 128. Gutgesell HP, Massaro TA. Management of hypoplastic left heart in a consortium of university hospitals. Am J Cardiol 1995; 76: 809–11. 129. Garne E, Stoll C, Clementi M; The Euroscan Group. Evaluation of prenatal diagnosis of congenital heart diseases by ultrasound: experience from 20 European registries. Ultrasound Obstet Gynecol 2001; 17: 386–91. 130. Tegnander E, Eik-Nes SH, Linker DT. Incorporating the four-chamber view of the fetal heart into the second-trimester routine fetal examination. Ultrasound Obstet Gynecol 1994; 4: 24–8. 131. Levi S, Schaaps JP, De Havay P, Coulon R, Defoort P. End-result of routine ultrasound screening for congenital anomalies: The Belgian Multicentric Study 1984–92. Ultrasound Obstet Gynecol 1995; 5: 366–71. 132. De Vore GR, Medcaris AL, Bear MB, Horenstein J, Platt LD. Fetal echocardiography: factors that influence imaging of the fetal heart during the second trimester of pregnancy. J Ultrasound Med 1993; 12: 659–63. 133. Crane JM, Ash K, Fink N, Desjardins C. Abnormal fetal cardiac axis in the detection of intrathoracic anomalies and congenital heart disease. Ultrasound Obstet Gynecol 1997; 10: 90–3. 134. Smith RS, Comstock CH, Kirk JS, Lee W. Ultrasonographic left cardiac axis deviation: a marker for fetal anomalies. Obstet Gynecol 1995; 85: 187–91. 135. Vergani P, Mariani S, Ghidini A et al. Screening for congenital heart disease with the four-chamber view of fetal heart. Am J Obstet Gynecol 1992; 167: 1000–3. 136. Buskens E, Grobbee DE, Hess J, Wladimiroff JW. Prenatal diagnosis of congenital heart disease; prospects and problems. Eur J Obstet Gynecol Reprod Biol 1995; 60: 5–11.

Indications for fetal echocardiography

137. Todros T, Faggiano F, Chiappa E et al. Accuracy of routine ultrasonography in screening heart diseases prenatally. Prenat Diagn 1997; 10: 901–6. 138. Hafner E, Scholler J, Schuchter K, Sterniste W, Philipp K. Detection of fetal congenital heart disease in low-risk population. Prenat Diagn 1998; 18: 808–15. 139. Oggè G, Gaglioti P, Maccanti S, Faggiano F, Todros T; Gruppo Piemontese for Prenatal Screening of Congenital Heart Disease. Prenatal screening for congenital heart disease with four-chamber and outflow-tract views: a multicenter study. Ultrasound Obstet Gynecol 2006; 28: 779–84. 140. Wong SF, Chan FY, Cincotta RB, Lee-Tannock A, Ward C. Factors influencing the prenatal detection of structural congenital heart diseases. Ultrasound Obstet Gynecol 2003; 21: 19–25. 141. Alembik Y, Stoll C. Routine fetal echocardiography and detection of congenital heart disease. Lancet 1996; 348: 1732.

129

142. Crane JP, LeFevre ML, Winborn RC et al; the RADIUS Study group. A randomized trial of prenatal ultrasonographic screening: impact on the detection, management, and outcome of anomalous fetuses. Am J Obstet Gynecol 1994; 171: 392–9. 143. Bromley B, Estroff JA, Sanders SP et al. Fetal echocardiography: accuracy and limitations in a population at high and low risk for heart defects. Am J Obstet Gynecol 1992; 166: 1473–81. 144. Hamar BD, Dziura J, Friedman A et al. Trends in fetal echocardiography and implications for clinical practice. 1985 to 2003. J Ultrasound Med 2006; 25: 197–202. 145. Friedberg MK, Silverman NH. Changing indications for fetal echocardiography in a University Center population. Prenat Diagn 2004; 24: 781–6. 146. Berg C, Kremer C, Geipel A et al. Ductus venosus blood flow alterations in fetuses with obstructive lesions of the right heart. Ultrasound Obstet Gynecol 2006; 28: 137–420.

10 Circulation in the normal fetus and cardiovascular adaptations to birth Abraham M Rudolph Fetal circulation Postnatally, oxygen uptake and carbon dioxide removal from the body occur in the lungs. Energy substrates are absorbed in the gastrointestinal tract and delivered to the liver through the portal venous system before entering the general circulation. In the mammalian fetus, oxygen uptake and carbon dioxide removal are accomplished in the placenta through the umbilical circulation. Energy substrates diffuse or are actively transported from the maternal circulation across the placental membrane and are transferred to the fetal body via the umbilical vein. Although a proportion of the substrates enter the hepatic circulation, an average of about 50% bypass the liver to enter the general fetal circulation via the ductus venosus

Course of the circulation Postnatally the pulmonary circulation is completely separated from the systemic circulation. Well-oxygenated arterial blood ejected by the left ventricle flows through the systemic arteries to supply all tissues of the body with oxygen and nutrients. Blood then enters the systemic venous system and returns to the right atrium and ventricle; it is ejected into the pulmonary arterial circulation and returns to the left atrium and ventricle through the pulmonary veins. Blood thus flows serially through the circulation and, apart from passage of a small amount of coronary venous blood into the left ventricular cavity through Thebesian veins, no mixing of arterial and venous blood occurs. In the fetus, oxygenated blood returns to the body through the umbilical venous system. This blood mixes with systemic venous blood before entering the cardiac ventricles to be ejected to perfuse the fetal body. As seen in Figure 10.1, the umbilical vein enters the porta hepatis and gives rise to several branches that are distributed to the left lobe of the liver.1 Distal to the branches supplying the left lobe, the ductus venosus originates and passes first posteriorly and then superiorly to connect with

the inferior vena cava. The umbilical vein then arches to the right lobe, where it is joined by the portal vein. After this junction, branches are distributed to the right lobe of the liver. The left hepatic vein drains into the inferior vena cava in close proximity to the ductus venosus. In the sheep fetus, the left hepatic vein and ductus venosus drain through a common orifice on the left posterior aspect of the vena cava. In the human, the two vessels drain through adjacent orifices separated by a sharp ridge. In the fetal lamb, a thin valve-like membrane covers the distal orifice of the ductus venosus and left hepatic vein.2,3 Although the function of this membrane has not been defined, I have proposed that it facilitates preferential streaming of blood flow, as mentioned below. The right hepatic vein drains separately into the right and posterior aspect of the inferior vena cava and the orifice is also partly covered by a distal valve-like membrane. Based on angiographic studies in which contrast medium was injected into the umbilical vein, Lind et al suggested that umbilical venous blood passing through the ductus venosus flowed largely to the left atrium through the foramen ovale.4 We have studied the distribution of umbilical and portal venous and inferior vena cava blood in fetal lambs by injecting radionuclide-labeled microspheres into each vessel and determining the distribution of the spheres in the fetal body.5,6 Umbilical venous blood is distributed to the left lobe of the liver, which receives about 90% of its blood supply from this source; the remaining 10% is derived from the descending aorta through the hepatic artery. Almost all the blood passing through the ductus venosus (92–95%) is from the umbilical vein; the remaining small amount is contributed by the portal vein. Umbilical venous blood is distributed to the right lobe of the liver through the arcuate connection to the portal vein. Most portal venous blood flows to the right lobe of the liver. Only a small amount passes through the ductus venosus and none is distributed to the left lobe of the liver. The ductus venosus acts as a bypass for umbilical venous blood. To some extent it reduces the resistance to the flow of umbilical venous blood to the inferior vena cava by diverting it away from the hepatic microcirculation.

132

Fetal Cardiology

Tricuspid valve SVC

Foramen ovale

LHV RHV Ductus venosus

Umbilical vein IVC Portal vein

Figure 10.1 Course of blood flow in the region of the porta hepatis. Umbilical venous blood is distributed to the left lobe of the liver. The ductus venosus arises from the umbilical vein, which then arches to the right to join the portal vein. Portal venous blood is largely distributed to the right liver lobe and only a small proportion passes through the ductus venosus (reproduced with permission from reference 1).

The proportion of umbilical venous blood that passes through the ductus varies greatly, both in the lamb and in the human fetus, from about 20 to 90%, with an average of about 50%.7,8 Ductus venosus blood is preferentially distributed through the foramen ovale into the left atrium and left ventricle, whereas abdominal inferior vena cava blood flows preferentially through the tricuspid valve to the right ventricle.The distribution of blood from the left and right hepatic veins was also examined using radionuclidelabeled microspheres in fetal sheep. Blood from the left hepatic vein, which enters the inferior vena cava through the same orifice as the ductus venosus, also flows preferentially across the foramen ovale. In contrast, right hepatic venous blood streams preferentially across the tricuspid valve, similarly to the flow of abdominal inferior vena caval blood (Figure 10.1). These streaming patterns of abdominal inferior vena cava blood and ductus venosus blood can be identified by direct observation of the thinwalled inferior vena cava in the thorax of the fetal lamb in utero. A stream of well-oxygenated blood from the ductus venosus and left hepatic vein can be identified on the left anterior aspect of the vena cava; a poorly oxygenated stream from the abdominal inferior vena cava and right hepatic vein is observed on the posterior right aspect of

the vena cava. Similar streaming has been observed in the inferior vena cava and atria in fetal sheep by color flow Doppler studies.9 The ductus venosus stream is directed largely through the foramen ovale, whereas distal inferior vena cava blood streams across the tricuspid valve. Ultrasound examinations of human fetuses have also shown similar flow patterns of blood in the ductus venosus and abdominal inferior vena cava.10,11 The mechanisms responsible for this selective streaming have not been fully defined. In the sheep fetus, the valvelike structures over the entrance of the ductus venosus and left hepatic vein may direct the blood draining from these vessels toward the foramen ovale. Similarly, right hepatic venous blood may be deflected by the valve toward the tricuspid valve. These valves are not well defined in the human fetus, so other mechanisms must account for preferential streaming. The inferior margin of the atrial septum separates the entrance of the inferior vena cava from the left atrium. The crista dividens, the crescentic edge of the superior portion of the atrial septum, overlies the inferior vena cava, so that the posterior left portion of the inferior vena cava connects directly through the foramen ovale to the left atrium. During phases of the cardiac cycle, the Eustachian valve and the lower portion of the atrial septum move in unison, either to the left, to facilitate blood through the foramen ovale, or to the right, to direct flow through the tricuspid valve.9 This will tend to direct blood on the left anterior aspect of the vena cava through the foramen ovale. Another mechanism that has been proposed is that different velocities of the streams tend to allow separation in the inferior vena cava.9–11 The mean velocity of the abdominal inferior vena cava stream is relatively low (about 15 cm/s). The ductus venosus mean velocity is, however, much higher (about 55–60 cm/s). It is suggested that this high velocity permits the stream to maintain a degree of separation from the abdominal inferior vena cava stream in the thoracic portion of the inferior vena cava and carry the blood through the foramen ovale. This preferential streaming of ductus venosus and left hepatic venous blood through the foramen ovale provides blood of higher oxygen saturation to the left atrium and ventricle and thus into the ascending aorta. Abdominal inferior vena cava blood and the right hepatic venous blood of lower oxygen saturation is preferentially distributed into the right ventricle and pulmonary artery. Almost all superior vena cava blood passes through the tricuspid valve and is distributed into the right ventricle. Normally about 5% or less flows through the foramen ovale into the left atrium. Ultrasound studies in fetal lambs demonstrated that the small amount of superior vena cava blood that enters the foramen ovale does so indirectly. It is first diverted in a retrograde direction into the upper portion of the inferior vena cava during atrial systole and then enters the foramen during the rapid inflow phase from the inferior vena cava.9

Circulation in the normal fetus and cardiovascular adaptations

65 70 m55 45

55

70 3 60 70 4

55

65

40

m3

m2

70 m55 45

70

Figure 10.2 Course of the circulation in the heart and great vessels in the late-gestation fetal lamb. The figures in circles within the chambers and vessels represent per cent oxygen saturation levels. The figures alongside the chambers and vessels are pressures in mmHg related to amniotic pressure level as zero. m, mean pressure (reproduced with permission from reference 1).

Right ventricular blood is ejected into the pulmonary trunk; a small proportion passes into the pulmonary circulation but the majority is directed through the ductus arteriosus to the descending aorta (Figure 10.2). Normally none of the blood flowing through the ductus arteriosus passes in a retrograde fashion across the aortic isthmus to the ascending aorta and its branches. The left atrium receives blood from the foramen ovale and pulmonary veins, and then empties into the left ventricle, which ejects into the ascending aorta. Most ascending aortic blood is distributed to the coronary circulation, head and cerebral circulation, and upper extremities; only a small proportion passes across the aortic isthmus into the descending aorta. Descending aortic blood is distributed to the abdominal organs and the tissues of the lower trunk and lower extremities, but a large amount enters the umbilical–placental circulation

Admixture of oxygenated and systemic venous blood In the adult circulation, there is essentially no mixing of oxygenated pulmonary venous and systemic venous blood.

133

However, oxygenated umbilical venous and poorly oxygenated systemic venous blood mix at several sites in the fetal circulation before being distributed to the systemic arteries. A proportion of umbilical venous blood mixes with portal venous blood and enters the central circulation through the right hepatic vein. Blood from the ductus venosus, left and right hepatic veins, and abdominal inferior vena cava all enter the thoracic portion of the inferior vena cava. Preferential streaming of blood from the ductus venosus and left hepatic vein to some extent separates well-oxygenated and poorly oxygenated blood. In the left atrium, blood entering the foramen ovale from the inferior vena cava is joined by pulmonary venous blood, which, in the fetus, has reduced oxygen saturation. Systemic venous blood is preferentially directed into the right ventricle, pulmonary trunk, and ductus arteriosus to the descending aorta and its branches to supply the arterial branches of the lower body, as well as the placenta. Thus, blood delivered to all fetal tissues and to the placenta is a mixture of oxygenated umbilical venous and systemic venous blood. Some umbilical venous blood is returned to the placenta without first being delivered to fetal tissues to permit oxygen uptake. This arrangement is inefficient because it imposes an additional workload on the heart to supply oxygen to the tissues. Similarly, blood returning to the heart from the superior and inferior vena cava that is distributed to the fetal tissues without first being delivered to the placenta for oxygenation contributes to the inefficiency of the fetal circulation. In the sheep fetus under normal conditions, about 45% of superior vena cava blood and 53% of inferior vena cava blood are returned to the fetal tissues without having had the opportunity to take up oxygen in the placenta. About 22% of umbilical venous blood is returned to the placenta without first passing through the systemic microcirculation. The inefficiency resulting from systemic venous and umbilical venous recirculation constitutes about 33% of the combined ventricular output of the fetal heart.

Fetal vascular pressures The fetus is surrounded by amniotic fluid in the uterus within the abdomen; it is customary to relate all vascular pressures to amniotic cavity pressure. Fetal pressures are thus raised by increases in intra-abdominal pressure as occurs with straining, gaseous distension, feeding, or uterine contraction. In the quietly standing ewe, intraamniotic pressure is usually about 8–10 mmHg above atmospheric pressure. All pressures are presented with amniotic pressure as baseline (Figure 10.2). Umbilical venous pressure is about 8–10 mmHg near the umbilical ring and 2–3 mmHg lower near the placenta. Normally the pressure shows a continuous flat contour, with no phasic change during atrial or ventricular systole.

134

Fetal Cardiology

This lack of pulsatile pressure extends into the porta hepatis, where the mean pressure is 5–6 mmHg. This is in contrast to the inferior and superior vena cava, which show variations of pressure with the cardiac cycle. Postnatally, the left atrial is higher than the right atrial mean pressure; the right atrial pressure contour shows a dominant a-wave, whereas the left atrium shows a dominant v-wave. In the fetus, the mean pressure in the superior and inferior vena cava and the right atrium is about 2–3 mmHg and a- and v-wave pressures are both about 4–5 mmHg. Left atrial pressure has a contour similar to that of the right atrium, and the mean pressure is 1–2 mmHg lower than right atrial pressure. Right and left systolic and end-diastolic pressures are similar. In the lamb fetus, the right ventricular and pulmonary arterial systolic pressure tends to exceed the left ventricular and aortic pressure by 5–8 mmHg during late gestation; this is probably the result of mild constriction of the ductus arteriosus. Aortic pressure increases with gestational age in the lamb fetus, from a mean level of 25–30 mmHg at about 60 days’ gestation to 60–70 mmHg close to term, at about 145 days’ gestation.

Blood gases and oxygen saturation Maternal arterial blood in the pregnant ewe has a PO (partial pressure of oxygen) of 90–100 mmHg and a PcO2 of about 35 mmHg. There is a large PO2 gradient across the placenta; the Po2 of umbilical venous blood is 32–35 mmHg. Umbilical venous blood PCO2 is about 40 mmHg and the pH is 7.40. The P50 (the PO 2 at which hemoglobin is 50% saturated with oxygen) for fetal blood in the sheep is considerably lower (∼27 mmHg) than that of adult blood (∼38 mmHg). Therefore, when PO2 of umbilical venous blood is 32–35 mmHg, oxygen saturation is about 90%. Oxygen saturations of blood in various cardiac chambers and great vessels measured in the fetal lamb are shown in Figures 10.2 and 10.3. Umbilical venous blood has an oxygen saturation of 80–90%. The left hepatic venous blood oxygen saturation is about 75%, whereas right hepatic venous blood oxygen saturation is lower, at about 65%. This is related to the difference in the blood supplying the left and right lobes of the liver, as discussed above. Blood in the abdominal inferior vena cava distal to the entrance of the ductus venosus and hepatic veins has a PO2 of about 12–14 mmHg and an oxygen saturation of 35–40%. The PO2 and oxygen saturation in superior vena cava blood are similar. The PO2 of right ventricular and pulmonary arterial blood is 18–20 mmHg and oxygen saturation is about 50%. Left ventricular and ascending aortic blood has a PO2 of about 25–28 mmHg and an oxygen saturation of about 65%, whereas descending aortic blood has a PO2 of 20–23 mmHg and an oxygen saturation of about 55%. Systemic arterial blood has a

LHV

73

63

DV 83

RHV PS

83 PV 35

DAo

40 UV

55

IVC

P

Figure 10.3 Oxygen saturations of blood in the vessels in the region of the porta hepatis and in the inferior vena cava and hepatic veins (reproduced with permission from reference 1).

PCO2 of 43–45 mmHg and a pH of about 7.38–7.39. Reliable values for blood gases and oxygen saturations for the human fetus in utero are not available.

Effects of administering oxygen to the mother Administering 100% oxygen to the ewe raises her arterial oxygen saturation to 100% and the PO2 to more than 400 mmHg. Umbilical venous blood PO2 in the fetal lamb increases to 40–50 mmHg and oxygen saturation reaches 95–100%. Arterial PO2 increases to only 30–35 mmHg, with an oxygen saturation of about 80%. The large oxygen difference between the maternal arterial and fetal umbilical venous PO2 is the result of diffusion limitation across the placental membrane. The separation between the maternal and fetal circulations in the sheep is fairly broad, because the sheep has a syndesmochorial placenta. It is possible that the maternal–fetal oxygen gradient is lower in the human fetus because the placental membrane has fewer layers.

Cardiac output and its distribution Postnatally blood is ejected by the left ventricle into the aorta and distributed to the tissues; it returns through the veins to the right atrium and is ejected by the right

Circulation in the normal fetus and cardiovascular adaptations

ventricle to the pulmonary circulation and returns to the left atrium and ventricle. In this circulation the volume of blood ejected by each ventricle is similar and is termed the cardiac output. In the fetus, as mentioned above, systemic and umbilical venous blood mix and the mixed blood is distributed to the various parts of the body and to the placenta; blood to many organs is derived from both ventricles. Unlike in the postnatal circulation, the volumes of blood ejected by left and right ventricles are different in the fetus. The output of the heart is usually expressed as combined ventricular output, the sum of the volumes ejected by the two ventricles. In chronically instrumented fetal lambs during the later months of gestation (term is about 145 days), the combined ventricular output is about 450 ml/min per kg fetal body weight.12,13 Umbilical– placental blood flow is about 200 ml/min per kg body weight, and blood flow to the fetal body is about 250 ml/ min per kg. The right ventricle ejects about two-thirds and the left ventricle about one-third of combined ventricular output in the fetal lamb (Figures 10.4 and 10.5). Umbilical–placental flow of about 200 ml/min per kg represents 40–45% of combined ventricular output (CVO). Although the proportions vary, about 55% of umbilical venous blood passes through the ductus venosus and 45%

135

through the hepatic circulation. Thus, about 110 ml/min per kg passes through the ductus venosus. The right and left lobes of the liver receive a total of about 90 ml/min per kg fetal body weight of blood from the umbilical vein, and the right lobe receives ∼30 ml/min per kg fetal body weight. Inferior vena cava blood distal to the entrance of the hepatic veins and ductus venosus (abdominal inferior vena cava) is derived from the lower body organs and the lower extremities as well as the lower portion of the trunk. In the fetal lamb, this is ∼30% of CVO or about 135 ml/min per kg. The blood entering the heart from the inferior vena cava includes blood from the ductus venosus, left and right hepatic veins, and abdominal inferior vena cava, and constitutes about 70% of CVO, or about 315 ml/min per kg (Figures 10.4 and 10.5). About 115 ml/min per kg or about 25% of CVO passes through the foramen ovale to the left atrium; this blood is derived predominantly from the ductus venosus. Venous return from the superior vena cava is 90–95 ml/min per kg, representing about 21% of CVO, and is largely directed across the tricuspid valve into the right ventricle. About 200 ml/min per kg of inferior vena cava blood as well as coronary venous blood enters the right ventricle. The right ventricle ejects about 66% of

90

21

31

45

10

265

59 66

3

7 300

34

66

150

69

27

21

35

15

90 7

195 15

115

310 35

3 69

310

Placenta 110

200

Figure 10.4 The percentages of the combined ventricular output that return to the fetal heart that are ejected by each ventricle and that flow through the main vascular channels. Figures represent values for late-gestation lambs (reproduced with permission from reference 1).

Figure 10.5 The volumes of blood flowing through cardiac chambers and great vessels for the late-gestation fetal lamb (ml/min per kg body weight) (reproduced with permission from reference 1).

Fetal Cardiology

CVO, or about 300 ml/min per kg. Only about 10–15% of the blood ejected by the right ventricle is directed to the pulmonary circulation; this constitutes about 8% of CVO or about 35 ml/min per kg fetal weight. The remaining 58% of CVO, about 265 ml/min per kg, passes through the ductus arteriosus. The left ventricle receives the 115 ml/min per kg of blood that passes through the foramen ovale as well as about 35 ml/min per kg from pulmonary venous return. It ejects about 150 ml/min per kg, representing about 33% of CVO. Less than a third of the blood ejected by the left ventricle passes across the aortic isthmus to the descending aorta. This represents about 10% of CVO or about 45 ml/min per kg. The coronary circulation receives about 3% of CVO; about 20% of CVO, about 90 ml/min per kg, is distributed to the head, brain, upper extremities, and upper portion of the trunk. The magnitude of blood flow through the major arteries is reflected by the relative diameters of these vessels. The pulmonary trunk is very large, and the ascending aorta somewhat narrower; the descending aorta is also very wide, whereas the isthmus of the aorta is much narrower than the ascending or descending aorta and the ductus arteriosus. The combined ventricular output is about 450 ml/min per kg fetal body weight. In the sheep fetus it is fairly constant in relation to fetal body weight from about 90 to 140 days’ gestation (term is 140–145 days). In the last few days of pregnancy there is a modest fall in CVO related to body weight. This could be related to uterine contraction or other unidentified factors. The proportions of CVO distributed to different organs change during gestational development. In the fetal lamb, the placenta receives about 42–44% of CVO at 75–90 days’ gestation, but the proportion falls to 36–38% near term. The percentage distributed to the brain and lungs increases with gestational age. This change during growth is more striking when the changes in blood flow per unit of organ tissue weight are examined. Changes in organ blood flow per 100 g of weight, during development of the fetal lamb, are shown in Figure 10.6. At about 110 days (0.75 gestation), blood flow per 100 g organ weight progressively increases to the brain, gut and lungs. The cause of the increase in flow to these organs is not known; it could be related to an increase in size of the vascular bed due to the growth of new vessels, or to increased metabolic activity with vasodilatation, or a combination of these factors.

Circulation in the human fetus Although the general course of the circulation in the human fetus is similar to that in the lamb, the proportions of combined ventricular output distributed to body organs differ, because the relative weights of some organs are very different. Perhaps the most important factor is brain size; in the mature human fetus the brain constitutes 12% of

300 Heart 250

200 Flow ml/100 g/min

136

Kidneys 150 Brain Lungs 100 Gut 50 Trunk 0 0.5

0.6

0.7 0.8 Gestation

0.9

1.0

Figure 10.6 Changes in blood flow to various organs in relation to tissue weight in the fetal lamb during the latter half of gestation (reproduced with permission from reference 1).

body weight, as compared with 3% in the lamb. It is reasonable to assume that the blood flow related to weight is similar in the two species. Near term, both human and sheep fetal body weights are about 3.5 kg, and brain weights are about 65 g in the sheep and 350 g in the human. If it is assumed that blood flows to the brain are similar in relation to tissue weight at 120 ml/min per 100 g, total brain flows would be 80 ml/min in the sheep and 420 ml/ min in the human, or 22 and 120 ml/min per kg, respectively. Limited data are available for blood flows in the human fetus, based on Doppler flow studies.14,15 The umbilical blood flow has been reported to be about 180 ml/ min per kg estimated fetal weight16, but recently it has been proposed that umbilical flow in the human fetus is lower, at about 120–140 ml/min per kg fetal weight.17 Although reports of left and right ventricular output vary considerably, generally the combined ventricular output appears to be similar to that in the sheep fetus at about 450 ml/min per kg estimated weight. However, the ratio of right to left ventricular output, which in the sheep is about 2:1, is about 1.2–1.3:1 in the human.14,15 I have attempted to estimate the quantities of blood flowing through the cardiac chambers and great vessels in the human fetus. Unfortunately, limited information is currently available, but one publication provides estimates of pulmonary blood flow.18 Based on this report, I have assumed pulmonary blood flow in the late-gestation fetus

Circulation in the normal fetus and cardiovascular adaptations

31

140

10

45

39

175

17

3 55

31

21

49

200

250

28

49

75

15

45

95

140

17

3

137

125

220 75

15 Placenta

9

220

40

Figure 10.7 The percentages of the combined ventricular output that return to the fetal heart that are ejected by each ventricle and that flow through the main vascular channels for the late-gestation human fetus (reproduced with permission from reference 1).

to be about 75 ml/min per kg fetal weight. Using this information, I have estimated the blood flow patterns in the late-gestation human infant. The percentages of CVO are shown in Figure 10.7 and the blood flows through various great vessels and ejected by each ventricle are depicted in Figure 10.8. The pattern of flow in the liver region of the human appears to be similar to that of the lamb fetus. In fetal lambs, Doppler flow studies demonstrated a blood flow velocity of 55–60 cm/s in the ductus venosus, whereas the velocity in the abdominal inferior vena cava was only about 16 cm/s.9 The ductus venosus stream was directed preferentially through the foramen ovale (see above). In the human fetus, blood flow velocity in the ductus venosus has been reported to be 65–75 cm/s, and the ductus venosus stream also preferentially flows through the foramen ovale.10,11

Flow velocity contours Arterial flow Patterns of blood flow in various sites of the circulation in the fetus have been studied in fetal lambs by electromagnetic

Placenta 40

180

Blood flows in human fetus (ml/min/kg)

Figure 10.8 The volumes of blood (ml/min per kg) flowing through cardiac chambers and great vessels of the late-gestation human fetus (reproduced with permission from reference 1).

or ultrasonic means, and in human fetuses by means of Doppler ultrasound. Velocity recordings in the ascending aorta and main pulmonary trunk are similar to those in the adult. However, in the fetal lamb, there are distinct differences between aortic and pulmonary blood flow patterns. The velocity in the main pulmonary trunk rises rapidly after the onset of ejection to reach its peak early in systole. It then falls rapidly initially, but the rapid decrease in velocity is interrupted and a definite incisura in the velocity profile is frequently noted (Figure 10.9). The ascending aortic velocity shows a slower rise at the onset of ejection, and peak velocity is achieved in about midsystole. The peak of aortic velocity is close to the incisura on the downslope of the pulmonary trunk velocity tracing. The mechanisms responsible for these differences in the velocity patterns have not been defined. One possible explanation is that there are marked differences in impedance in the two circulations. The left ventricle ejects into the ascending aorta, which in the fetal lamb carries only about a third of CVO. The aortic isthmus is relatively narrow and it transmits only about 10% of CVO; it imposes some degree of obstruction to flow from the ascending to

138

Fetal Cardiology

Pulmonary trunk

PA press

QDuctus

Ascending aorta

QLPA

Figure 10.9

Figure 10.10

Blood flow velocities were recorded simultaneously from the ascending aorta and the pulmonary trunk with electromagnetic flow transducers in a late-gestation fetal lamb in utero. The calibration for the two flows is identical. Note that stroke volume, the area under the curve, is almost twice as great from the right as from the left ventricle. The differences in flow contours are discussed in the text.

Simultaneous recordings of blood flow velocities in the ductus arteriosus and left pulmonary artery (LPA), using ultrasonic flowmeters and pulmonary arterial pressure (PA Press) in a fetal lamb in utero. See text for discussion (reproduced with permission from reference 1).

the descending aorta. The arterial system into which the left ventricle ejects has a relatively low compliance and this may explain the slow rise and late peaking of velocity. The right ventricle ejects predominantly through the ductus arteriosus to the descending aorta, from which the relatively low-resistance umbilical–placental circulation arises. The ejection into this highly compliant circulation could explain the rapid rise and early peaking of the velocity. The incisura on the downslope of the pulmonary trunk velocity tracing indicates that the fall in velocity is briefly interrupted. A possible explanation for this phenomenon is that when blood ejected by the left ventricle crosses the aortic isthmus to reach the ductus arteriosus, it interferes with flow from the pulmonary trunk through the ductus and thus slows the fall in velocity momentarily. This hypothesis is supported by the fact that the incisura coincides with the peak of the aortic velocity profile. The velocity profiles for the pulmonary trunk and ascending aorta have not been described in human fetuses. They could differ considerably from those of the lamb, because the differences in left and right ventricular stroke volumes are less marked. Also, the volume of blood flowing into the cerebral circulation in the human fetus is much greater than that in the lamb, and the compliance of the arterial system into which the left ventricle ejects is considerably greater than that in the lamb. The velocity pattern of blood flow in the branch pulmonary arteries is distinctive in the fetal lamb.19 In the normal lamb, velocity increases rapidly early in systole, but forward flow ceases about half-way through systole

and a variable amount of retrograde flow occurs through the remainder of systole (Figure 10.10). No antegrade flow is evident during diastole but a small amount of retrograde flow continues through much of diastole. The factors contributing to this pattern of flow have not been fully defined, but it appears to be related to the low compliance of the large pulmonary arteries, the high resistance of the pulmonary circulation and the presence of the ductus arteriosus. During the early phase of right ventricular systole, blood flows into the main and branch pulmonary arteries, as well as through the ductus arteriosus. Owing to the high vascular resistance, flow into the lung ceases, but flow through the ductus continues. The recoil of the large pulmonary arteries results in reversed flow of blood through the ductus into the low-resistance circulation of the lower body and placenta, resulting in retrograde flow at the site of recording of the velocity profile. The interruption of the rapid decline in forward flow in the ductus is evident in the velocity tracing of the ductus, coincident with the retrograde flow in the left pulmonary artery tracing. The velocity profile in the branch pulmonary arteries is markedly altered by changes in pulmonary vascular resistance. Administration of a pulmonary vasodilator such as acetylcholine to the fetal lamb results in an increase in pulmonary blood flow. The duration of the forward flow phase in the branch pulmonary arteries is prolonged, the degree depending on the magnitude of the vasodilatation and the decrease in the duration and amount of retrograde flow (Figure 10.11). An increase in pulmonary

Circulation in the normal fetus and cardiovascular adaptations

0

PA pressure mmHg

systole

0

Phasic flow pulmonary artery

Phasic flow pulmonary artery

PA pressure mmHg

systole 75

139

75

0

0

1 second

1 second Control

Ach infusion

Figure 10.11 The pattern of blood flow in the left pulmonary artery is shown before (left panel) and during (right panel) infusion of acetylcholine (Ach) into the pulmonary artery. See text for discussion.

vascular resistance resulting from induced hypoxia in the lamb markedly reduces the duration and magnitude of the forward flow phase and increases the duration and degree of retrograde flow. The velocity profile in the branch pulmonary arteries of the human fetus has not been well characterized. It is likely that pulmonary blood flow is relatively higher in the human than in the lamb fetus (see above) and the forward flow phase would be longer and the retrograde flow less significant than in the lamb. Defining the velocity profile in human fetuses might provide useful information about the status of pulmonary vascular resistance.

Tracheal pressure ECG

SVC flow IVC flow SVC pressure 1 second

Venous flow

Figure 10.12

Patterns of flow in the superior and inferior venae cavae are similar, and in the normal fetus do not differ significantly from postnatal flow patterns (Figure 10.12). The flow contours are related inversely to the pressure tracing. Coincident with atrial systole, there is a short phase of retrograde flow, followed by forward flow during ventricular systole; this forward flow is enhanced at the end of ventricular systole, coincident with the phase of rapid inflow into the ventricle. As is evident in Figure 10.12, venous flow is markedly altered during fetal respiratory movements; flow is greatly enhanced during inspiratory movements and reduced during expiration. The velocity profile is markedly affected by alterations in heart rate and vascular resistance. Bradycardia results in an exaggeration

of retrograde flow during atrial systole, as does an increase in peripheral vascular resistance. Flow in the umbilical vein is continuous, and the phasic changes associated with the cardiac cycle are normally not evident, so that the velocity profile is flat. However, induction of hypoxia in the lamb fetus results

Simultaneous recordings of blood flow velocities in the superior (SVC) and inferior vena cava (IVC) by electromagnetic flow transducers placed around the vessels in the thorax, and vena cava and intratracheal pressures in a fetal lamb in utero. Note the increase in forward flow in both vessels associated with inspiratory effort (decrease in intratracheal pressure). For detailed description see text.

140

Fetal Cardiology

in some degree of phasic change in umbilical flow velocity. This is probably related to the fact that a greater proportion of umbilical venous blood traverses the ductus venosus and dilates the vessel, allowing transmission of central venous pressure.

Determinants of cardiac output Cardiac ventricular output is the product of heart rate and stroke volume. Stroke volume is determined by preload, afterload, and myocardial contractility. Preload determines the degree to which the ventricular muscle is stretched immediately prior to contraction. In the intact heart, ventricular volume at end-diastole determines the length of the cardiac myocytes and thus sarcomere length. The greater is the length of the sarcomere, up to an optimal level, immediately before contraction occurs, the greater is the force generated during contraction. An increase in end-diastolic ventricular volume increases the force of contraction of the muscle and, in the intact heart, increases the stroke volume if other factors are unchanged. Afterload, or load on the heart muscle during development of active force, determines the degree of shortening of the sarcomeres and thus the volume ejected during systole. With the same force of ventricular contraction, the chamber will empty more if the afterload is low, and the ejection volume will be lower if the afterload is increased. In the intact circulation, afterload is influenced by several factors, such as arterial pressure, compliance of the arterial system, and peripheral vascular resistance. Contractility is the intrinsic force of contraction of the muscle; with isolated muscle, increased contractility increases the force developed and, in the intact heart, increases the stroke volume, or developed pressure. In the intact circulation, heart rate, preload, afterload, and contractility are interrelated, and a change in one factor may modify other parameters. It is therefore important to consider possible changes in these other parameters when assessing the effects of alteration of one regulatory factor.

Effects of heart rate In the adult, cardiac output is relatively constant over a wide range of heart rates. Increasing the heart rate to 150 beats/min or decreasing it to 50 beats/min from a resting rate of about 70 beats/min does not alter output. Greater increases in heart rate may decrease cardiac output because the reduction in diastolic filling time does not permit adequate filling to maintain stroke volume. With very slow heart rates, stroke volume is increased to maintain cardiac output, but when maximum diastolic filling has been achieved, further slowing results in a decrease of ventricular output.

In studies in fetal sheep, spontaneous increases in heart rate above the resting level of about 160 beats/min are associated with increases of ventricular output of up to 15–20%, and spontaneous decreases in heart rate result in a fall in output.20 In these studies, it was not apparent whether the tachycardia was directly responsible for the increase in cardiac output. It is possible that the factors inducing the increase in heart rate also affected loading conditions or contractility. The effects of electrical pacing of the right or left atrium to increase rates to 240–300 beats/min were studied in fetal lambs. Pacing the right atrium resulted in an increase of left ventricular output of up to 15%, with only a small increase or no change in right ventricular output. At rates above 300–320 beats/min, ventricular output fell progressively with increasing rate, presumably because diastolic filling time was greatly reduced. Pacing the left atrium increased right ventricular output modestly, but decreased left ventricular output. Normally in the fetus, the right atrial pressure is slightly higher than that in the left atrium throughout the cardiac cycle. During pacing, the left atrial pressure pulse is altered so that the left atrial pressure exceeds that in the right atrium during some phases of the cycle and interferes with flow through the foramen ovale into the left atrium, reducing the left ventricular filling and output. Vagal stimulation decreased the output of both ventricles by about 15–20%, associated with bradycardia. However, the decrease in output could not be ascribed entirely to the decrease in heart rate, because vagal stimulation increases systemic arterial pressure and elevates afterload; this could contribute to the fall in ventricular output.

Effects of preload and afterload Preload and afterload are discussed together because there is usually an interaction between them in the intact circulation. If afterload is increased, the volume ejected by the ventricle during systole is reduced and residual ventricular volume increases. If ventricular filling is maintained, preload is greater with the next beat. In utero studies of fetal lambs have been performed to assess the role of preload on cardiac output. In most of these studies, ventricular end-diastolic or atrial pressures have been used as an index of preload. Pressure measurements may not, however, be a reliable indicator of volume, because ventricular compliance determines the volume at any particular pressure. Studies in isolated myocardium and intact hearts have shown that fetal myocardium is less compliant than that of the adult.21 Rapid intravenous infusions of 0.9% NaCl solution raised cardiac output associated with an increase in atrial pressure in newborn lambs.22 Cardiac output increased progressively with elevation of atrial pressure to levels of about 15 mmHg. Several investigators have studied the

Circulation in the normal fetus and cardiovascular adaptations

1.0

600

LA-7 LA-3.5

LVSV ml/kg

CVO ml/min/kg

0.9

400

141

LA-10

0.8 0.7 0.6

200

0.5 30

4

8 RA pressure mmHg

12

16

Figure 10.13 Changes in combined ventricular output (CVO) associated with acute reduction of atrial pressure by blood removal and increase of atrial pressure by infusion of electrolyte solution in fetal lambs (reproduced with permission from reference 1).

effects of decreasing or increasing preload in fetal lambs in utero.23–26 Preload was decreased by reducing fetal blood volume by removal of blood, and increased by rapid intravenous infusion of electrolyte solution. A fall in right atrial and right ventricular end-diastolic pressure resulted in a marked decrease in cardiac output. Output rose when atrial pressure increased by 2–4 mmHg above resting levels, but further increases in pressure did not result in greater output by the ventricle (Figure 10.13). This response is distinctly different from that in the postnatal lamb, in which increases of atrial pressure to levels of 15–20 mmHg are associated with a progressive increase in ventricular output. Based on these studies, it was proposed by Gilbert23,24 that the fetal heart is normally operating near the top of its ventricular function curve; the elevation of cardiac output associated with an increase in preload is limited because myocardial performance, or contractility, is relatively poor in the fetus. However, a decrease in atrial pressure reduces preload, resulting in a fall in cardiac output. In these studies, the effects of rapid infusion of electrolytes on arterial pressure were not considered. Associated with the infusion, fetal arterial pressure also increases and thus changes afterload. We examined the effects of changing preload at various constant levels of arterial pressure.27 Arterial pressure elevation dramatically reduced left ventricular stroke volume at all levels of mean atrial pressure (Figure 10.14). At constant arterial pressure levels, progressive elevation of left atrial pressure increased left ventricular stroke volume even with atrial pressures of 10–12 mmHg. This study demonstrated that the fetal heart responds to increases in preload by increasing its output.

40

50 60 70 Arterial pressure (mmHg)

80

90

Figure 10.14 When systemic arterial pressure is regulated, an increase in pressure results in a fall of left ventricular stroke volume (LVSV) at fixed left atrial (LA) pressure. At any level of arterial pressure, an increase in left atrial pressure increases left ventricular stroke volume (reproduced with permission from reference 1).

It did not, however, resolve whether performances of the fetal and adult myocardium are comparable.

Myocardial performance Studies of isolated myocardium from fetal and adult sheep have demonstrated that fetal myocardium develops less active tension than adult myocardium at similar muscle lengths.28 Also, the maximum force that can be generated is considerably lower for fetal than for adult myocardium. Several differences in morphological and biochemical parameters of myocardium have been described that could account for the lesser contractility of fetal myocardium. It was suggested that fetal myocardium contains fewer sarcomeres, or contractile units, in each myocyte. Another factor that may be important is development of the sarcoplasmic reticulum, which regulates the movement of calcium ions, essential for myocardial contraction. The fetal myocardial sarcoplasmic reticulum is well developed, but the T-tubular system, representing the extension of the sarcoplasmic reticulum to provide closer relations with the contractile elements, is either poorly developed or absent in the immature myocardium. Not only are there structural differences in the sarcoplasmic reticulum, but, in studies with isolated sarcoplasmic reticulum vesicles, calcium uptake was found to be impaired in fetal myocardium.29 Local release of norepinephrine (noradrenaline) at sympathetic nerve endings is an important mechanism for increasing myocardial contractility. Sympathetic nerve endings are sparse or even absent in fetal myocardium.

142

Fetal Cardiology

The abundance of sympathetic nerve endings varies greatly during development in different species. In the guinea pig, myocardial sympathetic innervation is almost fully developed at birth,30 whereas in the rabbit and the rat there is almost no innervation at birth, but it develops within 14–21 days after birth.31 The sheep fetus has no detectable sympathetic innervation at 75 days (mid-gestation), but innervation begins to appear at 90–100 days, and is abundant but not yet fully developed just before birth.32 In addition to the difference in sympathetic innervation, β-adrenergic receptor concentration is lower in fetal than in adult myocardium.33 Although these differences in sympathetic innervation and β-adrenoreceptor concentration may not be important in the resting fetal heart, they could influence the ability to respond to stress.

Regulation of the circulation in the fetus In the adult the systemic and pulmonary circulations are separate. Each ventricle is subjected to a potentially different preload and afterload, and the stroke volume of each ventricle might vary greatly. The Frank–Starling mechanism is useful for adjusting the outputs of the two ventricles so that over a short period the ventricles eject similar volumes. A reduction in venous return to the right atrium reduces the filling pressure and end-diastolic volume of the right ventricle, resulting in a decrease of stroke volume. Pulmonary blood flow and venous return to the left atrium and ventricle is reduced and stroke volume falls. An increase in systemic arterial pressure will restrict left ventricular stroke volume; end-diastolic volume will increase so that, with the next beat, greater force is generated to increase stroke volume. In the fetus the presence of the foramen ovale tends to make right and left atrial pressures equal throughout the cardiac cycle. The ductus arteriosus provides a large communication between the aorta and the pulmonary artery, resulting in almost identical pressures in the two vessels. In view of the similar atrial pressures and similar aortic and pulmonary arterial pressures, differences in stroke volumes of the left and right ventricles in the fetal lamb are probably due to differences in afterload on the ventricles. The aortic isthmus, which in the fetus is narrower than the ascending and descending aorta, to some extent functionally separates the upper and lower body circulations. The left ventricle ejects into the ascending aorta and the vessels of the head and neck, a circulation that in the lamb fetus is poorly compliant and has a relatively high vascular resistance. The right ventricle ejects into the pulmonary trunk and directly through the large ductus arteriosus into the descending aorta and its branches. This circulation has a higher compliance and a lower resistance because it includes the umbilical–placental

vasculature. This functional separation of the aorta at the isthmus has been demonstrated in fetal lambs. A rapid reduction in peripheral vascular resistance in the lower body circulation, induced by a vasodilator, causes a decrease in descending aortic pressure and increase in right ventricular stroke volume for several beats, whereas the ascending aortic pressure and left ventricular output do not change. Similarly, injection of a vasodilator into the ascending aorta causes an evanescent decrease of ascending aortic pressure and increase in left ventricular stroke volume.

Reflex regulation Chemoreflexes Previous studies on the role of chemoreflexes in control of the fetal circulation were conflicting. Some investigators suggested that aortic and carotid chemoreceptors are relatively inactive in the fetus, but these studies were performed in anesthetized exteriorized fetal lambs.34 Other studies indicated that aortic receptors were important in causing a bradycardic response to hypoxia.35 More recent studies in fetal lambs have shown that they are active, at least in the last third of gestation.36 Responses to carotid chemoreceptor stimulation are much greater than to aortic receptor stimulation.37 The chemoreceptors are stimulated by hypoxemia, and experimentally can be activated by intravascular injection of small doses of sodium cyanide. The cardiovascular response dominates, with bradycardia and immediate hypotension, but respiratory gasps are noted. The bradycardia can be abolished if the lambs are pretreated with atropine, indicating that the bradycardia is induced by vagus nerve stimulation. Confirmation of the fact that the cyanide response is the result of chemoreceptor stimulation was obtained by demonstrating the loss of the cardiovascular and respiratory responses in fetal lambs in which sinoaortic denervation had been accomplished.37 In the adult, chemoreceptor stimulation results in reflex peripheral vasoconstriction. It is likely that the peripheral vasoconstriction induced by fetal hypoxemia is largely mediated by chemoreceptor stimulation. From the studies in fetal lambs it is apparent that their chemoreflex responses are different from those in the adult. The respiratory response in the adult animal dominates, whereas chemoreceptor stimulation in the fetus causes only a minor respiratory response. There is as yet no explanation for this difference in response.

Baroreflexes In the adult, arterial pressure is maintained over a fairly narrow range through the control of baroreceptors. Stimulation of aortic and carotid baroreceptors by a rise in arterial pressure induces bradycardia, depression of

Circulation in the normal fetus and cardiovascular adaptations

Birth-associated changes in the circulation Delivery of the fetus from the uterus disrupts the umbilical–placental circulation. The functions of oxygen uptake and carbon dioxide removal are transferred to the lungs. Pulmonary ventilation has to be established to provide gas exchange. During fetal life, pulmonary blood flow is relatively low, and has to increase to allow oxygen uptake adequate for postnatal survival. The well-oxygenated blood from the umbilical veins and poorly oxygenated blood from the vena cava mix partially and venous blood is diverted away from the lungs through the foramen ovale and the ductus arteriosus. In the adult, blood circulates in series. All venous blood is returned to the right atrium and ventricle, ejected into the lungs where it is oxygenated, and then passes to the left atrium and ventricle to be ejected into the systemic arterial circulation. Apart from the return of minor amounts of venous blood into the left ventricle via Thebesian veins, there is no mixing of arterial and venous blood. The foramen ovale and ductus arteriosus have to be closed functionally or anatomically to establish the adult circulation. As the fetus is delivered, several events occur in a short time period. Fluid in the fetal airways is removed either by expulsion through the mouth as a result of chest compression, or by absorption into the pulmonary circulation with the onset of breathing. Regular ventilation is established, and the umbilical–placental circulation is terminated by disruption or clamping of the umbilical

cord. Ventilation by room air is associated with an increase in the alveolar oxygen concentration and also with the rhythmic physical expansion of the lung and removal of alveolar fluid. It has been difficult to assess the role of each of these factors in contributing to the circulatory changes associated with birth, because they occur almost simultaneously. We developed a fetal lamb preparation to examine the individual role of birth events in these changes.40 Catheters were implanted in various fetal vessels, a tube was inserted into the trachea, and an inflatable balloon occluder was placed around the umbilical cord. All the catheters were exteriorized to the maternal flank and the ewe and fetus allowed to recover from surgery. Fetal vascular pressures and blood gases were monitored and blood flows were measured repeatedly by the radionuclidelabeled microsphere technique. The effect of rhythmic expansion of the lung was assessed by ventilating the fetus with a gas mixture of 5% carbon dioxide, 3% oxygen, and 92% nitrogen. This did not significantly change fetal blood gas levels of PO2 21 mmHg and Pco2 40 mmHg in descending aortic blood. The fetus was then ventilated with 100% oxygen; this raised fetal descending aortic PO2 to about 50 mmHg and oxygen saturation to above 90%. With the fetus well oxygenated, the effect of occluding the umbilical cord was then assessed. The proportions of CVO ejected by each ventricle and distributed to the major vessels are shown in Figure 10.7. Rhythmic ventilation without altering fetal blood gases produced a considerable increase in pulmonary blood flow (Figure 10.15) and decrease in pulmonary vascular resistance (Figure 10.16).41 The proportion of CVO passing to the lungs increased from 9 to 31%. Interestingly, pulmonary arterial pressure did not fall, suggesting that

2500 Pulmonary blood flow (ml/100 g/min)

myocardial contractility, and peripheral vasodilatation, all of which tend to decrease arterial pressure. Ablation of aortic and carotid baroreceptors, by bilateral section of the aortic and carotid afferent nerves, results in an initial increase in resting heart rate and arterial pressure, but within 1–2 days these parameters return to average levels that were present during the pre-denervation period. Wide swings of arterial pressure and heart rate occur around the average pressure and rate, in association with stimuli that produce only small changes in the normal animal.38 Arterial baroreceptors are functional in the fetus relatively early in gestation, but their importance in regulating fetal arterial pressure has been questioned. In fetal lambs, baroreflex sensitivity increases with gestational age from about 80 days’ gestation; near term gestation, the bradycardia induced by increased arterial pressure is equal to that noted postnatally. In fetal lambs, sinoaortic denervation results in the same wide variation in heart rate and blood pressure as observed in adult animals.39 It is thus apparent that baroreflexes are important in stabilizing arterial blood pressure in the fetus as well.

143

2000

1500

1000

500

0 Control

Lung expansion

Oxygenation

O2 + cord occlusion

Figure 10.15 Changes in pulmonary blood flow resulting from physical expansion of the lung, ventilation with oxygen, and umbilical cord occlusion in fetal lambs (reproduced with permission from reference 1).

144

Fetal Cardiology

1.2

18

PVR mmHg/ml/100 g/min

1.0

27 45

0.8 24 52

0.6

3

3

31

48

0.4 52

0.2

17

18 Control

Lung expansion

Oxygenation

O2 + cord occlusion

the ductus arteriosus was still widely patent and that aortic pressure was transmitted to the pulmonary artery. Associated with the increased pulmonary blood flow, pulmonary venous return to the left atrium increased and the proportion of CVO passing through the foramen ovale decreased (Figure 10.17). Although CVO did not change, right ventricular output constituted about 52% of CVO, as compared with 65% in the unventilated fetus. Also, only 24% of CVO passed through the ductus arteriosus to the descending aorta, compared with 57% in the control state. Left ventricular output increased from 34 to 48% of CVO, so that the outputs of the two ventricles were now similar. Ventilation with oxygen resulted in a further decline in pulmonary vascular resistance and rise in pulmonary blood flow (Figure 10.18). Only a minor proportion of blood ejected by the right ventricle passed through the ductus arteriosus to the descending aorta, almost all being distributed to the pulmonary circulation. The large venous return to the left atrium elevated left atrial pressure above that in the systemic veins and right atrium; this resulted in closure of the foramen ovale, with only insignificant flow from the right to the left atrium. Total CVO did not change significantly, but left ventricular output exceeded right ventricular output, with 55% of CVO contributed by the left and 45% by the right ventricle. This higher output by the left ventricle was the result of the development of a shunt from the aorta to the pulmonary artery through the still open ductus arteriosus, constituting about 10% of CVO. Pulmonary arterial pressure gradually and progressively decreased below aortic levels. This reflected the separation of systemic and pulmonary arteries by constriction of the ductus arteriosus.

31

3 48

Figure 10.16 Changes in pulmonary vascular resistance (PVR) resulting from physical expansion of the lung, ventilation with oxygen, and umbilical cord occlusion in fetal lambs (reproduced with permission from reference 1).

48

Figure 10.17 Proportions of combined ventricular output ejected by the right and left ventricles and flowing through great vessels in fetal lambs during ventilation without changing fetal blood gases (reproduced with permission from reference 40).

10

43 53 4 11 45

2

52

55 45

36 3

10

52

2 36

Figure 10.18 Proportions of combined ventricular output ejected by the right and left ventricles and flowing through great vessels in fetal lambs during ventilation with oxygen (reproduced with permission from reference 40).

Occlusion of the umbilical cord completely eliminated umbilical blood flow. It resulted in a modest increase in systemic arterial pressure and a small increase in the shunt through the ductus arteriosus from the aorta to the pulmonary artery. However, no other additional changes

Circulation in the normal fetus and cardiovascular adaptations

14

43 57

2 16

41

2

55

59 41

29 4

14

55

2 29

Figure 10.19 Proportions of combined ventricular output ejected by the right and left ventricles and flowing through great vessels in fetal lambs after occlusion of the umbilical cord during ventilation with oxygen (reproduced with permission from reference 40).

occurred and CVO was still similar to that in the control fetal state (Figure 10.19). From these studies it is apparent that the dramatic decrease in pulmonary vascular resistance resulting from ventilation of the lungs is the dominant factor contributing to the circulatory changes during the perinatal period. The elevation in left atrial pressure resulting from increased pulmonary venous return to the left atrium closes the foramen ovale. The cessation of umbilical venous return may also contribute to closure of the foramen. Constriction of the ductus arteriosus (vide infra) completes the separation between the left and right sides of the heart and the major arteries, resulting in the series circulation characteristic of the adult.

Perinatal changes in the pulmonary circulation Rhythmic physical expansion of the lungs and an increase in oxygen levels in the ventilating gas mixture have independent but complementary roles in pulmonary vasodilatation. The mechanisms by which these processes reduce pulmonary vascular resistance have been investigated, but are not yet fully resolved. There is considerable evidence suggesting that rhythmic expansion of the lungs results in the production of the prostaglandin, prostacyclin (PGI2), probably from endothelial cells.42 PGI2 is a pulmonary vasodilator and could be responsible for the effect of ventilation. Although inhibition of prostaglandin production

145

in the sheep fetus limits the degree of pulmonary vasodilatation with ventilation, it does not completely prevent it, suggesting that other factors may be involved. The possibility should be entertained that the removal of fluid present in fetal airways and the replacement with gas could contribute to a decrease in pulmonary vascular resistance based on physical phenomena alone. During fetal life, the alveoli contain fluid, and the positive pressure from the amniotic cavity is transmitted through the thorax to the lungs. This would tend to compress the pulmonary vessels alongside the alveoli and small bronchi. During delivery of the fetus, airway fluid is removed; spontaneous breathing results in the development of a negative intrapleural pressure, with a gradient from the airways outward to the pleura, and this tends to dilate pulmonary vessels. However, as shown experimentally in fetal lambs, positive-pressure ventilation with no change in blood gas concentrations also reduces pulmonary vascular resistance. Physical factors that could possibly contribute to this are changes in surface forces on the alveoli. During fetal life, there is a fluid–fluid interface on the alveolar surface with no significant surface tension forces. When the alveoli are filled with gas, a strong surface tension at the gas–fluid interface tends to collapse the alveoli. This would result in a force tending to dilate pulmonary vessels associated with the alveoli, and thus decrease pulmonary vascular resistance. During fetal life, the pulmonary vessels are exposed to the PO2 of blood in the pulmonary arteries, which, in fetal lambs, is about 18 mmHg. Fetal pulmonary vessels are markedly constricted when PO2 is reduced below and dilated by an increase of PO2 above control levels. Ventilation with air increases the PO2 in precapillary pulmonary vessels, because oxygen diffuses into these vessels from surrounding alveoli, resulting in vasodilatation. It has been suggested that the response to changes in PO2 may be due to a direct effect on smooth muscle cells, and could be related to transmembrane movement of potassium via oxygen-sensitive potassium channels. A reduction in PO2 blocks potassium channels and results in constriction, whereas a rise in PO2 opens the channels, causing vasodilatation. The role of potassium channels in the response is supported by studies in fetal lambs showing that potassium channel blockers cause pulmonary vasoconstriction.43 The vasodilator effect of oxygen on pulmonary arterioles has also been shown to be associated with nitric oxide (NO) mechanisms. NO production by endothelial cells, which is associated with an increase in PO2, induces relaxation of vascular smooth muscle cells, resulting in a fall in pulmonary vascular resistance. N-nitro-L-arginine inhibits NO production; in studies in fetal lambs, it markedly limited the reduction in pulmonary vascular resistance associated with oxygenation.44 In the lamb, the rapid increase in pulmonary blood flow with ventilation is associated with a drop in pulmonary

146

Fetal Cardiology

vascular resistance from the fetal level of 1.6 to 0.3 mmHg/ ml/min/kg. Functional responses of the pulmonary circulation are considerably greater in the fetus than in the adult, due to differences in morphology of the small pulmonary arteries. The pre-acinar arteries in the fetal lung have a thick wall with a prominent smooth muscle layer. The intra-acinar arteries, associated with bronchioli, are partly muscular or non-muscular, and more distal vessels do not have smooth muscle cells. The muscular medial layer does not change significantly during the latter half of gestation, but the number of intra-acinar and alveolar duct arteries increases with lung growth during fetal development. After birth, the pulmonary arterioles have a thinner muscular media as a result of an increase in lumen size. Subsequently the smooth muscle layer gradually regresses, so that pre-acinar arteries develop the morphological features of adult vessels with the thin wall and large lumen/wall ratio. These changes occur over several weeks. The normal development of the pulmonary circulation before and after birth may be influenced by alteration of the PO2 and the intraluminal pressure to which the arteries are exposed. Interference with normal oxygenation after birth may delay the normal regression of smooth muscle. Individuals born and continuing to live at high altitude are exposed to a lower partial pressure of oxygen in inspired air; they retain a greater amount of smooth muscle in pulmonary vessels and have higher pulmonary arterial pressures than individuals at sea level.45 An increase of pulmonary arterial pressure in the fetus, as may occur with constriction of the ductus arteriosus, results in increased development of the pulmonary arteriolar smooth muscle layer.46 The greater amount of smooth muscle may interfere with postnatal adaptation, and the fall in pulmonary arterial pressure may be slower than normal as a result of the higher pulmonary vascular resistance. After birth, pulmonary arterial pressure may not fall normally if there is a congenital cardiovascular malformation with a large communication between the left and right ventricles or the aorta and pulmonary artery. The large communication results in a tendency for pressures on the left and right sides of the heart to equalize, and thus pulmonary arterial pressure does not fall as in the normal infant. Persistence of the high pulmonary arterial pressure delays normal maturation of the pulmonary vessels and the smooth muscle component persists. The role of the pulmonary circulation in the hemodynamic and clinical manifestations of congenital cardiovascular malformations is discussed in Chapter 39.

postnatal levels, and pulmonary blood flow adequate for oxygen needs may not be established. This phenomenon has been named persistent pulmonary hypertension of the newborn’, and may result from several conditions. An inability to establish normal respiration after birth will interfere with the expansion of alveoli with air, and thus the PO2 will not be increased normally and pulmonary vasoconstriction will persist. This may occur in babies who are depressed as a result of sedative drugs given to the mother. It may also result from an obstruction to airways, as with meconium aspiration. In these infants, development of the lung and pulmonary vasculature may be normal, but pulmonary vasoconstriction persists because alveolar oxygen concentration does not increase. Relief of airway obstruction or stimulation of breathing results in a rapid fall of pulmonary vascular resistance. Failure of the normal postnatal decrease in pulmonary vascular resistance may be the result of abnormal prenatal development of the pulmonary resistance arteries. As mentioned above, an increase of pulmonary arterial pressure in the fetus, as may result from constriction of the ductus arteriosus, induces increased development of the pulmonary vascular smooth muscle.47 This may interfere with the achievement of normal postnatal pulmonary vascular resistance. These vessels are exquisitely sensitive to changes in PO2, and even mild degrees of hypoxemia may result in marked pulmonary vasoconstriction. The muscle does regress slowly postnatally, but normal pulmonary arterial pressure and flow after birth may not be achieved for several weeks. Persistent pulmonary hypertension of the newborn may occur in infants when the fetus has been exposed to indomethacin in utero, because prostaglandin inhibition causes ductus arteriosus constriction.46 Experimental studies have suggested that prolonged fetal hypoxia may also induce an increase in pulmonary vascular smooth muscle. Persistent pulmonary hypertension of the newborn may be associated with inadequate cross-sectional area of the pulmonary vascular bed. This is most commonly the result of an interference with lung development resulting from encroachment of a space-occupying lesion in the thorax, such as a large lung cyst, or by intestine herniating through a diaphragmatic hernia. It may also be the result of agenesis of a lung. Since pulmonary vascular development parallels the growth of alveolar units, lack of lung development will be associated with a reduced size of the pulmonary vascular bed. Adequate development of the pulmonary circulation after birth will occur only slowly, as new alveolar units are added with growth of the lung.

Persistent pulmonary hypertension of the newborn

Ductus arteriosus closure after birth

If pulmonary vascular resistance does not fall normally after birth, pulmonary arterial pressure will not drop to normal

The ductus arteriosus connects the pulmonary trunk, before origin of the left and right pulmonary arteries, to

Circulation in the normal fetus and cardiovascular adaptations

the descending aorta, just beyond the origin of the left subclavian artery, in the human fetus. The wall of the ductus is morphologically quite different from that of the aorta and pulmonary artery. Whereas the walls of these arteries are largely composed of elastic tissue, the predominant tissue in media of the ductus is smooth muscle. During fetal life, the ductus arteriosus diverts a major proportion of right ventricular output away from the lungs to the descending aorta. In the sheep fetus, right ventricular output is about 66% of combined ventricular output and almost 90% of right ventricular blood passes through the ductus arteriosus. Estimates of flow through the ductus arteriosus, measured by ultrasound in human fetuses, show considerable variation. The proportion of right ventricular output distributed to the lungs is greater in the human, and it is estimated that about 60% of the blood ejected by the right ventricle passes though the ductus. The ductus remains widely patent throughout gestation, and no pressure gradient between the pulmonary trunk and the descending aorta can be detected. However, during the latter weeks of gestation, mild constriction may occur, as evidenced by a 5–8-mmHg drop in systolic pressure across the ductus. After birth, the ductus constricts rapidly; the rate of constriction appears to vary in different species. In the rat, rabbit, and guinea pig it is essentially closed within minutes. In the sheep the process is somewhat slower, closure usually being achieved within an hour. In full-term human infants, functional closure usually occurs within 12–15 hours. During the first few hours, a bidirectional shunt may be detected by ultrasound, but after about 6 hours, only a small left to right shunt may occur for up to about 15 hours. Prior to complete closure, the ductus responds to a decrease of PO2 by dilatation and, with the increase of pulmonary arterial pressure resulting from pulmonary vasoconstriction, some degree of shunting from the pulmonary artery to the aorta may again occur. The PO2 to which the ductus arteriosus is subjected is an important determinant of the degree of constriction. In the fetus, right ventricular blood with a Po2 of about 18 mmHg in the lamb passes through the ductus to the descending aorta. The ductus arteriosus is constricted by an increase in oxygen levels. This has been observed in ductus rings in a tissue bath48 or in isolated perfused ductus preparations,49 as well as in fetuses in utero. In a tissue bath, ductus preparations are relaxed at Po2 of 25–30 mmHg; they show a progressive increase in the degree of constriction from PO2 of about 40 to 100 mmHg. With air breathing, the Po2 of arterial blood increases to 90–100 mmHg, but the mechanism by which oxygen constricts the ductus has not yet been resolved. The magnitude of response to oxygen is dependent on gestational age. The more immature is the fetus, the less is the constrictor response, and the level of PO2 required to initiate constriction is greater. The ductus arteriosus has also been shown to be very sensitive to prostagandins. The ductus is relaxed by

147

prostaglandin E2 and prostaglandin I2 (prostacyclin); both are produced by the ductus wall, but they are also produced elsewhere in the fetus, and circulating levels of PGE2 are elevated as compared with postnatally. Although large amounts of PGI2 are produced by the ductus, PGE2 is probably more important in regulating its tone, because the ductus muscle is much more sensitive to PGE2. Prostaglandins are synthesized from arachidonic acid by the cyclo-oxygenase enzymes. Inhibition of prostaglandin synthesis by agents such as aspirin or indomethacin in the lamb fetus results in constriction of the ductus, confirming the fact that prostaglandins are important in maintaining its patency prenatally. Blood PGE2 concentrations fall rapidly after birth, and the relative contributions of the removal of PGE2 and the increase in blood to ductus constriction after birth have yet to be resolved. Studies by Clyman et al indicate that the rise in PO2 is more important than the fall in PGE2 levels in late gestation, whereas in the second and early part of the third trimester, the ductus is more sensitive to the removal of prostaglandin than to the increase in PO2.50 Studies in fetal lambs indicate that the sensitivity of the ductus to oxygen and PGE2 can be matured by administering corticosteroids to the immature fetus. Recently, the role of nitric oxide in influencing the ductus has been demonstrated. Nitric oxide relaxes the ductus muscle. In prematurely delivered baboons, a cyclooxygenase inhibitor did not induce complete closure of the ductus, but combination with an inhibitor of nitric oxide production caused complete closure.51 The role of nitric oxide in normal regulation of the ductus is yet to be determined. The mechanisms by which permanent closure of the ductus arteriosus is achieved have not yet been fully resolved. Constriction results in thickening of the intima and the development of intimal mounds that encroach on the lumen. Disruption of the internal elastic lamina results in migration of endothelial and smooth muscle cells. Clyman et al have suggested that the middle layers of the ductus receive oxygen supply from blood in the lumen. Constriction results in thickening of the wall and severe hypoxia of the middle portion of the wall.52 This results in cell damage, with replacement by fibrous tissue. Permanent closure of the ductus is usually complete within a week, but may not occur for up to 3 weeks in some infants. While the ductus is still patent, a small shunt from the aorta to the pulmonary artery (left to right shunt) may be detected by ultrasound.

The ductus arteriosus in the preterm infant Postnatal closure of the ductus arteriosus is frequently delayed in infants born prematurely. The younger is the gestational age at birth, the more likely it is that the ductus

148

Fetal Cardiology

will not close soon after birth. In preterm infants with birth weights under 750 g, more than 80% have persistent patency of the ductus beyond the third day after birth. In infants with birth weights of 1000–1500 g, the incidence of persistent patency of the ductus is 40–50%, whereas in infants weighing 1500–1750 g at birth the incidence is about 20%. Most of the more immature infants also have severe respiratory distress syndrome. Several possible explanations for delayed closure of the ductus in preterm infants have been proposed. It was first suggested that the preterm infant did not achieve an adequate elevation of arterial PO2 due to poor ventilation, but this possibility has been excluded because, with assisted ventilation, PO2 may be raised to levels normal for mature infants and ductus patency persists. As mentioned above, the ductus arteriosus of the immature animal is more sensitive to the effect of removal of prostaglandins than to an increase of PO2, and even the high PO2 achieved may not induce complete closure. The possibility that PGE2 levels do not fall postnatally as rapidly as in the mature infant has also been examined. Results of these studies indicate that there is a delay in the fall of blood PGE2 concentrations after birth in preterm infants, and levels may be elevated for as long as 2–4 weeks postnatally. However, although PGE2 concentrations have been elevated in many infants with respiratory distress syndrome, only some had ductus arteriosus patency. The hemodynamic and clinical consequences of patency of the ductus in preterm infants are discussed in Chapter 39.

Postnatal changes in cardiac output The combined output (CVO) of the left and right ventricles is about 400–450 ml/min per kg body weight in term fetal lambs in utero. Measurements in newborn lambs showed cardiac output levels of 400–450 ml/min per kg. Since left and right ventricular outputs are the same postnatally, the output of each ventricle is about 400–450 ml/ min per kg and combined output is 800–900 ml/min per kg, about twice that in the fetus. In the fetus, right ventricular output is about 66% of CVO, or about 300 ml/min per kg; it increases by about 50% after birth. The left ventricle, which in the fetus ejects about 150 ml/min per kg, increases its output almost three-fold to 400–450 ml/ min per kg after birth. The factors contributing to this rise of cardiac output postnatally have not been fully assessed. The studies mentioned above in fetal lambs in utero, in which we sequentially induced physical expansion of the lungs, oxygenation and umbilical cord occlusion, indicated that none of these events resulted in a rise of CVO. The proportion of CVO ejected by the left ventricle increased, and that by the right ventricle decreased. The possibility was considered that delivery from the in utero

environment with a temperature of about 39°C into a room air environment with a temperature of about 25°C could contribute to the increase in CVO. In studies in which we measured CVO in non-breathing fetal lambs delivered into a water bath but with no alteration of umbilical–placental flow, changing the bath temperature from 37°C to 25°C did not result in a significant change of CVO.53 In these experimental studies, the fetuses did not breathe spontaneously and were not exposed to the stress of delivery. It has been proposed that fetal cardiac output is regulated by a high pericardial pressure transmitted from the intrauterine cavity across the thorax and fluidfilled lungs; this high pericardial pressure restricts filling of the ventricles and limits stroke volume. Spontaneous ventilation after birth results in the development of a negative intrapleural (relative to atmospheric) pressure, also creating a negative intrapericardial pressure, which facilitates filling of the ventricles. This greater diastolic filling of the ventricles would, if myocardial function is adequate, result in a higher stroke volume and greater ventricular output. Several hormones are recognized as having important roles in the perinatal circulatory adjustments. Plasma catecholamine concentrations increase during delivery and could, by their effect of increasing myocardial contractility, facilitate the increase in cardiac output occurring after natural vaginal delivery. In the experimental studies the fetuses were not exposed to the stress of delivery, and thus may not have experienced catecholamine stimulation of the myocardium. In sheep, plasma cortisol concentrations increase slowly from about 120 days’ gestation, and 2–3 days prior to delivery (at about 150 days) they rise sharply, several-fold. Cortisol plays an important role in maturing the myocardium in the perinatal period; it has been shown to reduce nuclear proliferation and to increase protein concentration in fetal myocytes; this could be a factor in providing an increase in cardiac output after birth.54 Thyroid hormone has long been known to affect the myocardium in adults. Deficiency of the thyroid results in a depression of myocardial performance. The number of β-adrenoreceptors in the myocardium is greatly reduced in adult animals with thyroid deficiency. Although decreased response to catecholamine stimulation resulting from reduced β-adrenoreceptors could well contribute to the decreased function, thyroid hormone may have additional effects by modifying heavy chain myosin expression. The role of thyroid hormone on cardiac performance in the perinatal period has been demonstrated in sheep. In fetal lambs, plasma tri-iodothyronine (T3) concentration is about 1.0 ng/ml. After vaginal delivery it rises to about 4.0 ng/ml within 30–60 min. This is unlikely to be a factor in the rapid increase in cardiac output, because, in adults, the effect of thyroid hormone is noted within days rather than minutes. In studies of fetal lambs, we observed

Circulation in the normal fetus and cardiovascular adaptations

that complete thyroidectomy performed just prior to delivery resulted in no increase in T3 concentrations, but the lambs showed the expected increase in cardiac output.55 However, if thyroidectomy was done about 10 days before delivery, fetal T3 levels were undetectable, and after delivery, the lambs showed a limitation of cardiac output as well as a blunted response to catecholamine infusion. This suggested that T3 was important prenatally for normal myocardial development, and we showed that β-adrenoreceptor numbers in ventricular muscle were significantly reduced.56

Morphological changes in the myocardium after birth Histological studies of adult and fetal myocardium show dramatic differences. The adult myocyte in the sheep heart has a diameter of about 15–20 μm, whereas in the fetal lamb heart myocytes are much smaller, with a diameter of 5–7 μm. The nucleus in adult myocytes is relatively small, and polyploidy is very common. In the fetal myocyte the nucleus is relatively larger, and most cells have a single nucleus. Observations in fetal lambs over the latter half of gestation show no significant change in myocyte diameter.1 Since the weight of the heart increases greatly, the increase in muscle mass is almost exclusively by an increase in cell numbers, or by hyperplasia. Postnatally, myocyte size increases dramatically and almost all growth is the result of hypertrophy; minimal mitosis occurs postnatally. Measurements of the DNA and protein content of the myocardium during fetal and postnatal life confirm this difference in growth patterns. DNA concentration reflects the number of nuclei in tissues; protein concentration reflects the total tissue mass. A high DNA/protein ratio suggests a relatively large number of nuclei, indicating that cells are small. The reverse suggests that the cells are large relative to nuclear numbers. During fetal life, the lamb heart shows a relatively high DNA/protein ratio, but after birth the DNA/protein ratio falls, reflecting the cessation of mitosis and the increase in myocyte size. Although all the factors responsible for the dramatic change in the pattern of myocardial growth after birth are not determined, cortisol at least appears to be important. In studies in fetal lambs, in which we infused cortisol into the left coronary artery for up to 96 hours in utero, the DNA/protein ratio of the myocardium fell, in a similar manner to that noted normally after birth.56

Postnatal changes in hepatic and ductus venosus blood flow after birth Prenatally, umbilical venous blood enters the porta hepatis; about 50% is distributed to the left and right lobes of

149

the liver and 50% passes through the ductus venosus.57 Portal venous flow in the fetal lamb is quite low, and almost all is distributed to the right lobe of the liver, with less than 10% passing through the ductus venosus. The liver receives a very large blood supply of almost 450 ml/min per 100 g liver weight. After birth, umbilical blood flow ceases, and apart from a small amount of flow from the hepatic artery, all hepatic flow is derived from the portal vein. After birth the hepatic blood flow is about 100 ml/min per 100 g liver weight; the flow increases to 140 ml/min per 100 g and then increases rapidly after feeding to about 300 ml/min per kg liver weight.58 The ductus venosus had been thought to react passively to the intraluminal pressure, but it was shown that prostaglandin is partly responsible for maintaining patency of the ductus venosus. Postnatally, removal of prostaglandins and cessation of flow from the umbilical vein contribute to closure of the ductus venosus. Soon after birth, considerable proportions of portal venous blood may pass through the ductus venosus, but by 3–4 days this becomes negligible, and by about 6–10 days after birth the ductus closes.

References 1. Rudolph AM. Congenital Diseases of the Heart. Armonk, NY: Futura, 2001. 2. Bristow J, Rudolph AM, Itskovitz J. A preparation for studying liver blood flow, oxygen consumption, and metabolism in the fetal lamb in utero. J Dev Physiol 1981; 3: 255–66. 3. Rudolph AM. Hepatic and ductus venosus blood flows during fetal life. Hepatology 1983; 3: 254–8. 4. Lind J, Stern L, Wegelius C. Human Foetal and Neonatal Circulation. Springfield, IL: CC Thomas, 1964. 5. Reuss ML, Rudolph AM, Heymann MA. Selective distribution of microspheres injected into the umbilical veins and inferior venae cavae of fetal sheep. Am J Obstet Gynecol 1981; 141: 427–32. 6. Edelstone DI, Rudolph AM. Preferential streaming of ductus venosus blood to the brain and heart in fetal lambs. Am J Physiol 1979; 237: H724–9. 7. Edelstone DI, Rudolph AM, Heymann MA. Liver and ductus venosus blood flows in fetal lambs in utero. Circ Res 1978; 42: 426–33. 8. Rudolph AM, Heymann MA, Teramo KAW, Barrett CT, Raiha NCR. Studies on the circulation of the previable human fetus. Pediatr Res 1971; 5: 452–65. 9. Schmidt KG, Silverman NH, Rudolph AM. Assessment of flow events at the ductus venosus–inferior vena cava junction and at the foramen ovale in fetal sheep by use of multimodal ultrasound. Circulation 1996; 93: 826–33. 10. Kiserud T. Hemodynamics of the ductus venosus. Eur J Obstet Gynecol Reprod Biol 1999; 84: 139–47.

150

Fetal Cardiology

11. Kiserud T. Fetal venous circulation—an update on hemodynamics. J Perinat Med 2000; 28: 90–6. 12. Rudolph AM, Heymann MA. The circulation of the fetus in utero. Methods for studying distribution of blood flow, cardiac output and organ blood flow. Circ Res 1967; 21: 163–8. 13. Rudolph AM, Heymann MA. Circulatory changes during growth in the fetal lamb. Circ Res 1970; 26: 289–99. 14. Sutton MG, Plappert T, Doubilet P. Relationship between placental blood flow and combined ventricular output with gestational age in normal human fetus. Cardiovasc Res 1991; 25: 603–8. 15. Meijboom EJ, Horowitz S, Valdes-Cruz LM et al. A Doppler echocardiographic method for calculating volume flow across the tricuspid valve: correlative laboratory and clinical studies. Circulation 1985; 71: 551–6. 16. Kiserud T, Rasmussen S, Skulstad S. Blood flow and the degree of shunting through the ductus venosus in the human fetus. Am J Obstet Gynecol 2000; 182: 147–53. 17. Kohl T, Silverman NH. Evaluation of umbilical venous blood flow by Doppler color flow mapping and conventional ultrasonographic methods. J Ultrasound Med 1996; 15: 465–73. 18. Rasanen J, Wood DC, Debbs RH et al. Reactivity of the human fetal pulmonary circulation to maternal hyperoxygenation increases during the second half of pregnancy: a randomized study. Circulation 1998; 97: 257–62. 19. Lewis AB, Heymann MA, Rudolph AM. Gestational changes in pulmonary vascular responses in fetal lambs in utero. Circ Res 1976; 39: 536–41. 20. Rudolph AM, Heymann MA. Cardiac output in the fetal lamb: the effects of spontaneous and induced changes of heart rate on right and left ventricular output. Am J Obstet Gynecol 1976; 124: 183–92. 21. Romero T, Covell J, Friedman WF. A comparison of pressure–volume relations of the fetal, newborn and adult heart. Am J Physiol 1972; 222: 1285–90. 22. Klopfenstein HS, Rudolph AM. Postnatal changes in the circulation and response to volume loading in the sheep. Circ Res 1978; 42: 839–45. 23. Gilbert RD. Control of fetal cardiac output during changes in blood volume. Am J Physiol 1980; 238: H80–6. 24. Gilbert RD. Effects of afterload and baroreceptors on cardiac function in fetal sheep. J Dev Physiol 1982; 4: 299–309. 25. Thornburg KL, Morton MJ. Filling and arterial pressures as determinants of RV stroke volume in the sheep fetus. Am J Physiol 1983; 244: H656–63. 26. Thornburg KL, Morton MJ. Filling and arterial pressures as determinants of left ventricular stroke volume in fetal lambs. Am J Physiol 1986; 251: H961–8. 27. Hawkins J, Van Hare GF, Schmidt KG, Rudolph AM. Effects of increasing afterload on left ventricular output in fetal lambs. Circ Res 1989; 65: 127–34. 28. Friedman WF. The intrinsic physiologic properties of the developing heart. Prog Cardiovasc Dis 1972; 15: 87–111. 29. Mahony L, Jones LR. Developmental changes in cardiac sarcoplasmic reticulum in sheep. J Biol Chem 1986; 261: 15257–65.

30. Lipp JA, Rudolph AM. Sympathetic nerve development in the rat and guinea-pig heart. Biol Neonate 1972; 21: 76–82. 31. Friedman WF, Pool PE, Jacobowitz D et al. Sympathetic innervation of the developing rabbit heart. Biochemical and histochemical comparisons of fetal, neonatal, and adult myocardium. Circ Res 1968; 23: 25–32. 32. Lebowitz EA, Novick JS, Rudolph AM. Development of myocardial sympathetic innervation in the fetal lamb. Pediatr Res 1972; 6: 887–93. 33. Birk E, Tyndall LC, Rudolph AM, Roberts JM. Effects of thyroid hormone on myocardial adrenergic beta-receptor responsiveness and function during late gestation. Pediatr Res 1992; 31: 468–73. 34. Jansen AH, Chernick V. Respiratory response to cyanide in fetal sheep after peripheral chemodenervation. J Appl Physiol 1974; 36: 1–5. 35. Dawes GS. Foetal and Neonatal Physiology. Chicago, IL: YearBook, 1968. 36. Itskovitz J, Rudolph AM. Cardiorespiratory response to cyanide of chemoreceptors in fetal lambs. Am J Physiol 1987; 252: H916–22. 37. Bartelds B, van Bel F, Teitel DF, Rudolph AM. Carotid, not aortic, chemoreceptors mediate the fetal cardiovascular response to acute hypoxemia in lambs. Pediatr Res 1993; 34: 51–5. 38. Cowley AW Jr, Liard JF, Guyton AC. Role of the baroreflex in daily control of arterial blood pressure and other variables in dogs. Circ Res 1973; 32: 564–76. 39. Itskovitz J, LaGamma EF, Rudolph AM. Baroreflex control of the circulation in chronically instrumented fetal lambs. Circ Res 1983; 52: 589–96. 40. Teitel DF, Iwamoto HS, Rudolph AM. Effects of birthrelated events on central blood flow patterns. Pediatr Res 1987; 22: 557–66. 41. Teitel DF, Iwamoto HS, Rudolph AM. Changes in the pulmonary circulation during birth-related events. Pediatr Res 1990; 27: 372–8. 42. Leffler CW, Hessler JR, Green RS. The onset of breathing at birth stimulates pulmonary prostacyclin synthesis. Pediatr Res 1984; 18: 938–42. 43. Mital S, Konduri GG. Vascular potassium channels mediate oxygen-induced pulmonary vasodilation in fetal lambs. Biol Neonate 2000; 77: 58–68. 44. Moore P, Velvis H, Fineman JR et al. EDRF inhibition attenuates the increase in pulmonary blood flow due to oxygen ventilation in fetal lambs. J Appl Physiol 1992; 73: 2151–7. 45. Sime F, Banchero N, Penaloza D et al. Pulmonary hypertension in children born and living at high altitudes. Am J Cardiol 1963; 11: 143–9. 46. Heymann MA, Rudolph AM. Effects of acetylsalicylic acid on the ductus arteriosus and circulation in fetal lambs in utero. Circ Res 1976; 38: 418–22. 47. Levin DL, Hyman AI, Heymann MA, Rudolph AM. Fetal hypertension and the development of increased pulmonary vascular smooth muscle: a possible mechanism for persistent pulmonary hypertension of the newborn infant. J Pediatr 1978; 92: 265–9. 48. Oberhansli-Weiss I, Heymann MA, Rudolph AM, Melmon KL. The pattern and mechanisms of response of the ductus

Circulation in the normal fetus and cardiovascular adaptations

49.

50.

51.

52.

arteriosus and umbilical artery to oxygen. Pediatr Res 1972; 6: 693–700. McMurphy DM, Heymann MA, Rudolph AM, Melmon KL. Developmental changes in constriction of the ductus arteriosus: responses to oxygen and vasoactive substances in the isolated ductus arteriosus of the fetal lamb. Pediatr Res 1972; 6: 231–8. Clyman RI, Mauray F, Heymann MA. Ductus arteriosus: developmental response to oxygen and indomethacin. Prostaglandins 1978; 15: 993–8. Keller RL, Tacy TA, Fields S et al. Combined treatment with a nonselective nitric oxide synthase inhibitor (l-NMMA) and indomethacin increases ductus constriction in extremely premature newborns. Pediatr Res 2005; 58: 1216–21. Kajino H, Goldbarg S, Roman C et al. Vasa vasorum hypoperfusion is responsible for medial hypoxia and anatomic remodeling in the newborn lamb ductus arteriosus. Pediatr Res 2002; 51: 228–35.

151

53. van Bel F, Roman C, Iwamoto HS, Rudolph AM. Sympathoadrenal, metabolic, and regional blood flow responses to cold in fetal sheep. Pediatr Res 1993; 34: 47–50. 54. Rudolph AM, Roman C, Gournay VA. Perinatal myocardial DNA and protein changes in the lamb: effect of cortisol in the fetus. Pediatr Res 1999; 46: 141–6. 55. Breall JA, Rudolph AM, Heymann MA. Role of thyroid hormone in postnatal circulatory and metabolic adjustments. J Clin Invest 1984; 73: 1418–24. 56. Birk E, Tyndall MR, Erickson LC et al. Effects of thyroid hormone on myocardial adrenergic β-receptor responsiveness and function during late gestation. Pediatr Res 1992; 31: 468–73. 57. Smolich JJ, Walker AM, Campbell GR, Adamson TM. Left and right ventricular morphometry in fetal, neonatal, and adult sheep. Am J Physiol 1989; 257: H1–9. 58. Townsend SF, Rudolph CD, Rudolph AM. Changes in ovine hepatic circulation and oxygen consumption at birth. Pediatr Res 1989; 25: 300–4.

11 Development of fetal cardiac and extracardiac Doppler flows in early gestation Ahmet A Baschat and Ulrich Gembruch Introduction The onset of rhythmic contractions of the primitive embryonic heart between 21 and 24 days after conception initiates an important sequence in the functional development of the embryonic cardiovascular system. The normal development of the embryonic cardiovascular system and the fetoplacental unit are necessary to ensure adequate blood flow, oxygen delivery, and gas and nutrient exchange at an organ and cellular level. As the placenta is the major respiratory organ in utero, the normal maturation of these two circulatory systems is important for adequate fetal growth and development. The first trimester is a period of rapid development in many organ systems coupled with exponential embryonic growth. Thereafter, fetal growth and development continue towards term in a more steady fashion. The cardiovascular system has to match these needs of the growing and developing embryo. It is therefore not surprising that there are important changes in fetal cardiac function which take place in the first and second trimesters. Since its introduction by FitzGerald and Drumm1 Doppler sonography has evolved as an important tool for the non-invasive examination of the fetal cardiovascular system in uncomplicated pregnancies and fetal disease. The wide application of Doppler sonography has greatly enhanced our knowledge of the maturation of the embryonic cardiovascular system. The study of these cardiovascular developmental changes has been of importance for several reasons. Studies investigating the normal and abnormal development of the human fetal cardiovascular system indicate that there are important differences from other mammalian species. Therefore, data that have been gathered in sheep and primate experiments may have to be applied with caution to the human fetus. Greater understanding of normal early vascular development allows Doppler flow studies to be integrated into prenatal diagnosis. In this context the integration of ductus venosus (DV) Doppler studies into first-trimester anomaly and aneuploidy screening, and application in pregnancies at high risk for

chromosomal defects, have been investigated.2,3 Previous observations have indicated that certain cardiac defects may produce a unique hemodynamic or structural impact in early gestation which ultimately contributes to their poor prognosis.4,5 Examination of DV blood flow is often abnormal under such circumstances. This explains why abnormal DV blood flow in chromosomally normal fetuses with increased nuchal translucency may be beneficial in identifying fetuses with underlying major cardiac defects.3 Preliminary results suggest that the application of this information and incorporation of DV Doppler screening into the nuchal translucency scan may result in a major reduction in the need for invasive testing.2 This example illustrates that the understanding of embryonic and fetal cardiovascular dynamics is likely to gain increasing importance in the future. This chapter outlines functional cardiovascular changes in the fetus with particular emphasis on the first and second trimesters.

Cardiovascular control mechanisms The heart undergoes a repetitive orderly sequence of cardiac contraction and valvular action, which is responsible for the antegrade delivery of blood. These events are summarized in the cardiac cycle, and the level of cardiac function determines the efficiency with which adequate blood flow can be provided to the body under physiological and pathological conditions. A detailed knowledge of the physiology underlying the cardiac cycle is helpful in the application and interpretation of Doppler waveforms obtained from the fetal circulation. The principal events that constitute the cardiac cycle are ventricular diastole, when the atrioventricular (AV) valves are open and the ventricles receive blood, and ventricular systole, when the aortic and pulmonary valves are open and blood is ejected into the circulation. In addition, discrete phases of the cardiac cycle have been identified which further subdivide the two principal events. Evaluation of fetal cardiac

154

Fetal Cardiology

function requires knowledge about the characteristics of these phases and their relationship to flow and pressure events. Since Doppler waveforms display only time and velocity information, any deductions about cardiovascular pressures are made from our understanding of the phases underlying the cardiac cycle. Nonetheless, important differences between fetal and adult cardiac function have been identified using the Doppler technique.

Cardiac cycle – diastole Ventricular diastolic filling in the adult is subdivided into a passive phase and an active phase. Initial passive filling is rapid and reaches a plateau (diastasis) which is followed by atrial contraction triggered by the electrical discharge at the sinoatrial node, initiating active filling. The initial rapid inflow of blood into the ventricles causes a proportionate fall in atrial pressure, which is reflected in a fall in the venous pulse (Y-descent). Subsequent contraction of the atria results in a rapid rise in the atrial pressure, which is reflected in a rise in the venous pulse

(a-wave). The two phases of ventricular inflow produce a biphasic flow velocity waveform across the AV valve consisting of an early peak (E-wave) and a second shorter peak during atrial contraction (A-wave). In the adult heart the majority of ventricular filling (90%) occurs during the E-wave while the A-wave contributes to the remainder when a person is at rest. However, if the heart rate is very high (e.g. during exercise), the atrial contraction may account for up to 40% of ventricular filling. In such situations the ‘atrial kick’ contributes to a larger proportion of ventricular filling and therefore the relationship of the E- and A-waves may reverse.6 Atrial contribution to ventricular filling therefore varies inversely with duration of ventricular diastole and directly with atrial contractility. After atrial contraction is complete, the atrial pressure begins to fall, causing a pressure gradient reversal across the AV valves. This causes the valves to float upward (pre-position) before complete closure during the beginning of systole. At this time, the ventricular volumes are maximal and this ventricular end-diastolic volume constitutes the ventricular preload (Figure 11.1).

B

B

C SV

A

A

SV

D

Ventricular volume Ventricular pressure Precordial venous flow velocity waveform

S-wave

D-wave a-wave

Electrocardiogram Diastole

IC

Systole

IR

Diastole

IC Systole

IR

Figure 11.1 Diagrammatic representation of the cardiac cycle. Shown is the relationship between intraventricular pressure and volume and precordial venous flow velocity waveforms during the cardiac cycle. Electrical activity precedes atrial and ventricular contractions. In the fetus a larger proportion of ventricular filling occurs by atrial contraction in late diastole. The rapid rise in atrial pressure is transmitted into the venous system, decreasing antegrade flow (a-wave). With onset of ventricular contraction atrioventricular valves close (A) and intraventricular pressure rapidly rises without ventricular shortening (IC, isovolumetric contraction) until it exceeds pressure in the great vessels, and semilunar valves open (B). Ventricular shortening during ejection of the stroke volume (SV) causes rapid descent of the atrioventricular valve ring, allowing increased precordial venous forward flow (S-wave). When ventricular pressure falls below diastolic pressures in the major vessels, the semilunar valves close (C). When intraventricular pressure falls below atrial pressure, the atrioventricular valves open (D) at the end of isovolumetric relaxation (IR). The rapid inflow into the ventricle is reflected by increased precordial venous flow (D-wave).

Development of fetal cardiac and extracardiac Doppler flows

Cardiac cycle – systole When the electrical impulses traverse the annulus fibrosus and reach the ventricular conduction system, ventricular myocardial contraction is initiated. Elevation of intraventricular pressures above the atrial pressures results in closure of the tricuspid and mitral valves. Following closure of the atrioventricular valves initially there is a period of rapid rise in intraventricular pressures (isovolumetric contraction). During this phase of isovolumetric contraction there may be bulging of the AV valves, resulting in a temporary increase in atrial pressures transmitted to the venous pressure waveform. Once the intraventricular pressures exceed those of the great arteries, the pulmonary and aortic valves open; left ventricular blood is ejected into the ascending aorta and right ventricular blood is ejected into the main pulmonary artery during this rapid ejection phase of ventricular systole. The rapid ventricular shortening results in a descent of the AV-ring which causes atrial pressures to fall below venous pressures (X-wave); atrial filling begins at this time. This period corresponds to the S-wave in the venous flow velocity waveform (see below). Following rapid ejection the rate of outflow from the ventricle decreases and the ventricular and aortic pressures start to decrease. At this point muscle fibers have shortened, are repolarizing, and can no longer contract forcefully. This causes ventricular active tension to decrease and the rate of ejection and ventricular emptying to fall. When the ventricular ejection falls to zero, the intraventricular pressures fall below the diastolic pressures in the major vessels, resulting in closure of the aortic and pulmonary valves. Ongoing ventricular relaxation ensures decreasing intraventricular pressures until these fall below atrial pressure and the atrioventricular valves open at the end of isovolumetric relaxation. The volume of blood that remains in the ventricle prior to the opening of the AV valves is called the end-systolic volume. The difference between the end-diastolic volume and the end-systolic volume is the stroke volume (Figure 11.1).

Indices of cardiac function Individual ventricular blood volumes ejected in 1 minute define the ventricular outputs representing the product of individual ventricular stroke volumes and heartbeats per minute. The combined cardiac output is simply the sum of right ventricular and left ventricular outputs. The parallel arrangement of the fetal circulation results in the unique feature that the relative contribution of the individual chambers to the combined cardiac output can change under physiological and pathological conditions. The blood pressure generated by the combined effects of ventricular contraction force, vessel wall resistance and downstream vascular resistance determines fetoplacental perfusion. For any organ the perfusion is passively

155

regulated by the pressure drop across the arterial and venous ends of its vascular supply. In addition, many organs have the potential to optimize perfusion by local changes in resistance vessel diameter by a process called autoregulation.7 The downstream resistance experienced by each ventricle is determined by the sum of blood flow resistance in individual downstream vascular beds. The product of the cardiac output and peripheral resistance determines the blood pressure generated. Although the mechanisms involved in the cardiac cycle appear relatively straightforward at first, it is important to realize the presence of many interacting factors. Changes in cardiac output may be due to variations in heart rate and/or stroke volume. Changes in the filling state of the heart (preload), changes of downstream resistance (afterload), and variations in myocardial contractility can achieve alterations of the stroke volume. Preload can be defined as the initial stretching of the cardiac myocytes prior to contraction, and is related to the sarcomere length. Sarcomere length cannot be determined in the intact heart, so other indexes such as ventricular end-diastolic volume or pressure are substituted. Neither of these measures is ideal, because they do not accurately reflect sarcomere length. Nevertheless, end-diastolic pressure, and particularly end-diastolic volume, are used as clinical indices of preload. Other factors influencing cardiac function are valvular competence, blood viscosity and inertia of the blood, and myocardial muscle mass. The heart rate is predetermined by the sinoatrial node, which is the cardiac pacemaker with the highest rate of intrinsic automaticity. There is superimposed modulation of heart rate and intracardiac conduction through the AV node by the autonomic nervous system. The overall control of vascular tone and blood pressure is integrated at the level of the vasomotor center.8 Under physiological circumstances, preload is primarily determined by venous return and blood volume. Heart rate, by affecting filling time, can have a pronounced inverse effect on preload. If preload is viewed as enddiastolic volume, then preload is determined ultimately by the end-diastolic pressure and the compliance of the ventricle. Therefore, a decrease in compliance, as occurs with ventricular hypertrophy, will lead to a reduction in preload unless there is a corresponding increase in end-diastolic pressure. In the adult heart an increase in preload or a decrease in afterload within a physiological range will result in an increase of stroke volume. The Frank–Starling mechanism describes the ability of the myocardium to increase stroke volume in response to increases in preload. The efficiency of the Frank–Starling mechanism is strongly influenced by diastolic and systolic myocardial properties. Degree and velocity of myocardial relaxation are the major determinants of ventricular compliance and therefore resistance of the ventricles to diastolic filling. A low ventricular compliance is associated with an exaggerated increase in

156

Fetal Cardiology

intraventricular pressure when preload is increased. Under such circumstances ventricular filling becomes increasingly dependent on atrial contraction, displacing the effective period of filling to late diastole. The decreased ventricular filling capacity is prohibitive for an increase in the stroke volume. In addition, myocardial contractility has to be of sufficient force to oppose the effects of afterload. Contractility and contraction velocity are the main determinants of stroke volume in circumstances of increased afterload. The functionality of the myocardium is one of the factors responsible for the efficiency with which cardiovascular control mechanisms can modulate cardiac output. Slower systolic contraction velocity and strength and low diastolic compliance have a significant impact on cardiac function and the efficiency of cardiovascular reflex mechanisms.

The fetal circulation The unique arrangement of the fetal circulation impacts significantly on cardiac function and distribution of cardiac output. According to data from sheep studies, approximately 50% of oxygenated blood of placental origin reaches the fetal heart via the umbilical vein through the DV after bypassing the hepatic circulation.8,9 The DV develops at approximately 7 weeks of gestation and, in contrast to the other veins, which grow with the embryo, shows little increase in size.10 The narrow diameter of the ductus is responsible for a marked acceleration of the entering blood.10,11 Blood with lower oxygen content enters the right atrium via the inferior and superior caval veins and the coronary sinus. Differential directionality of right atrial venous inflow and the crista dividens of the foramen ovale results in an interatrial right-to-left passage of well-oxygenated blood that was also confirmed for the human fetus.12,13 One of these bloodstreams originates in the umbilical sinus, reaching the left atrium via the left upper portion of the inferior vena cava after considerable flow acceleration in the DV; middle and left hepatic venous blood joins this stream. A second stream originating in the abdominal portion of the inferior vena cava is joined by right hepatic venous blood entering the right atrium via the right upper portion of the inferior vena cava. This stream meets with blood from the superior vena cava and coronary sinus and enters the right ventricle through the tricuspid valve. Pulmonary venous return reaches the left ventricle via the left atrium. The relative separation of right atrial inflow and little admixture from the pulmonary veins ensures that in the fetal lamb the left ventricle receives approximately 65% of the well-oxygenated blood, which has a 15–20% higher oxygen content than right ventricular blood. Because of the parallel arrangement of the fetal circulation the afterload acts separately on each ventricle. Right

ventricular afterload is predominantly determined by vascular resistance in the main pulmonary artery, the ductus arteriosus, and the descending aorta with its organ branches, and the combined resistance of the fetoplacental circulations. Vascular resistance in the ascending aorta and the brachiocephalic circulation predominantly determines left ventricular afterload. The relative separation of the venous ventricular inflows also has the effect that right ventricular preload comprises mainly the superior and inferior venae cavae, while left ventricular preload comprises the pulmonary veins and the left hepatic vein and DV. The characteristics of the fetal circulation have been extensively studied in the fetal lamb. Because of the parallel arrangement there is a differential distribution of right and left ventricular outputs with a predominant contribution of the right ventricle. In the fetal sheep the ratio of right-to-left ventricular output is 1.8:1. Owing to the high pulmonary vascular resistance and the orientation of the ductus arteriosus, only 13% of the right ventricular output is distributed to the lungs and the remaining 87% reaches the descending aorta.9,14,15 Approximately two-thirds of this blood reaches the placental vascular bed for oxygenation via the umbilical arteries. Left ventricular output is predominantly distributed to the coronary and brachiocephalic circulations. Of the 29% of left ventricular output reaching the descending aorta, two-thirds reaches the placenta. The right ventricle therefore has the role to deliver oxygen-poor blood to the placenta for oxygenation while the left ventricle delivers well-oxygenated blood to the brain and heart. In the fetal lamb the majority (41%) of the combined cardiac output is delivered to the placenta, while 22% supplies the upper part of the body, 8% the lungs, and 3% the myocardium.9,14,15 A change in distribution of the cardiac output with advancing gestation has been documented for the fetal lamb. The proportion of cardiac output reaching the brachiocephalic circulation increases from approximately 20% at midgestation to term. At this time the brachiocephalic circulation receives approximately 35% of the common cardiac output while the placenta and the remaining body receive 30% each. The high resistance in the fetal peripheral circulation and the constantly falling blood flow resistance in the placental bed ensure that a considerable proportion of aortic blood flow is diverted to the placenta via the umbilical arteries for oxygenation. The differential downstream distribution of blood volume passing through the descending aorta changes with gestation. At mid-gestation 59% of the blood flow reaches the placenta while 41% reaches the lower half of the body. This proportion changes toward term, when only 33% of blood flow reaches the placenta and 67% is distributed to the lower body9,14–16 (Figure 11.2). The non-invasive Doppler examination of the human fetal circulation seems to confirm the developmental changes observed in other mammalian species. However, there are some important differences that have been

Development of fetal cardiac and extracardiac Doppler flows

Right ventricle

Left ventricle

~65% of venous return IVC & SVC / coronary sinus

~35% of venous return: DV / pulmonary veins

Coronary artery 8% LVO Heart 3% CCO

Pulmonary artery 13% RVO

AAO 63% LVO Brain 26% CCO

Aortic isthmus 29% LVO (10% CCO)

Lungs 8% CCO

Ductus arteriosus 87% RVO (53% CCO)

Descending aorta (63% CCO)

10% LVO (4% CCO)

157

29% RVO (18% CCO)

19% LVO (6% CCO)

Lower body 22% CCO

identified in the human fetus. The dominance of the right ventricle contributing approximately 53% of the combined cardiac output has also been demonstrated for the human fetus.17–19 From approximately 20 weeks’ gestation onward, the ratio between the right and left ventricular output remains relatively constant, at 1.2:1. The difference from the sheep fetus is due to the fact that the human fetus has a relatively larger brain mass.17 Another important difference from the fetal lamb is that in the human fetus a higher proportion of umbilical blood is directed to the liver and less is shunted through the DV. Compared to the 50% shunting of umbilical blood through the DV found in animal experiments, the degree of shunting in the human fetus under physiological conditions is considerably less, suggesting a higher priority of the fetal liver than was previously realized.20,21 These findings indicate that there are important differences in regional blood flows in different mammalian species that may have important consequences in pathological states. Further studies are necessary to delineate these differences in greater detail to elucidate fetal adaptation to intrauterine life.

Structural and functional maturation of the fetal myocardium The immature fetal myocardium has decreased contractility and compliance, as well as slower contraction and

58% RVO (35% CCO) Placenta 41% CCO

Figure 11.2 Diagrammatic representation of the fetal circulation. Shown is the distribution of right and left ventricular output (RVO, LVO) and the combined cardiac output (CCO) in the circulation of the fetal lamb. The largest proportion of the CCO is distributed to the placenta for oxygenation. Through preferential streaming, well-oxygenated left ventricular blood supplies the brain and heart while right ventricular blood with lower oxygen content is predominantly distributed to the placenta. DV, ductus venosus; AAO, ascending aorta; IVC, inferior vena cava; SVC, superior vena cava.

relaxation rates than the adult or neonatal myocardium. The Frank–Starling mechanism is necessary for optimal cardiac output at various filling stages of the heart. It was long presumed that the human fetus was unable to increase cardiac output by the Frank–Starling mechanism. However, the demonstration of post-extrasystolic potentiation in cases of fetal arrhythmia suggests that the Frank– Starling mechanism is operational in the human fetus from 20 weeks onwards, but working at the upper limit.22,23 The fetal myocardial response to increases in afterload is another limitation when compared to neonatal and adult myocardium. Investigations in the fetal lamb have demonstrated that resting fetal cardiac function is normally near maximum, very sensitive to changes in afterload, and largely unaffected by baroreceptors.24,25 There is a sharp drop of cardiac output with minimal increases in afterload. In addition, cardiac output experiences only a slight increase if afterload falls. For these reasons, fetal cardiac function is described as being relatively afterload-sensitive.15,24,25 Cardiac output can be increased with volume loading and β-adrenoceptor stimulation, provided arterial pressure does not increase.25,26 This indicates that alteration of preload is an important determinant of cardiac output in utero. The majority of the fetal blood volume is contained in the placental circulation situated between the arterial and venous limbs of the circulation. Since this large proportion of the vascular volume is extracorporeal, the ability to increase preload by alteration of venous return is limited.27 Finally, the presence of central shunting through the foramen ovale

158

Fetal Cardiology

also results in equalization of pressures between the two ventricles (interdependence). Owing to this effect, preload of both ventricles is equally affected, impairing the ability for selective regulation of ventricular output by this mechanism. The sarcomere and its protein components actin, myosin, troponin, and tropomyosin constitute the contractile unit of the cardiac muscle. After β-adrenergic stimulation, electrical impulses travel across the sarcoplasmic reticulum resulting in calcium ion release from the T-tubule system. These calcium ions bind to troponin at the actin-to-myosin binding site. Conversion of adenosine triphosphate (ATP) to adenosine diphosphate (ADP) by a myosin-bound enzyme (ATPase) releases energy that is necessary for the relative motion between actin and myosin. Activation of troponin by the binding of calcium ions is permissive for this electromechanical coupling, which results in shortening of the sarcomere and ultimately contraction of the muscle fiber. The density of β-adrenoceptors, number of sarcomeres, development of the sarcoplasmic reticulum and the T-tubule system, structure of contractile proteins, and the number of calcium binding sites on the troponin influence the strength of contractility of the muscle fiber. In contrast, the contraction velocity is determined by the reaction rate of the myosin ATPase.28,29 Contraction velocity and maximal contractile force of isolated fetal cardiac muscle fibers are significantly lower than in the neonate or adult.15 Many mammalian experimental studies have documented structural and functional maturation of the myocardium in fetal and neonatal life. These changes presumably also occur in the human fetus. Fetal myocardium has approximately 30% less contractile elements per gram of muscle fiber28,29 as well as a lower concentration of contractile proteins.30,31 The myosin ATPase bound to the heavy chain myosin isoforms has a slower reaction rate than that bound to adult isoforms.32–34 Fetal troponin isoforms display a lower affinity and sensitivity for calcium ion binding than their adult counterparts.30,31–35 The fetal sarcoplasmic reticulum and the transverse tubular system are less well developed and have a lower capacity for calcium ions than in the adult or neonate.35,36 A number of ultrastructural changes have a significant impact on fetal myocardial function with advancing gestational age. An increase in myocardial adrenergic receptor density and maturation of the sarcoplasmic reticulum ultimately allow more effective stimulation of the sarcomere. Increased capacity and distribution of calcium ions within the sarcoplasmic reticulum enhance delivery of this important contractile substrate to troponin binding sites.30–37 There is a switch in the expression of isoforms of the contractile proteins to more mature adult forms, which contributes to an increase in contractility and maximal generated contractile force.38 The combination of enhanced calcium delivery in response to stimulation and changes in contractile elements results in a markedly

increased efficiency of electromechanical coupling.15 An important modulator of this maturational process is the concentration of thyroxine.39,40 The marked rise in thyroxine concentrations prior to term may contribute to improved cardiac function by an upregulation of myocardial β-adrenergic receptors.39,40 However, the sequence and timing of these changes has not been fully delineated. There is an up to 40-fold increase of thyroid hormones in the embryonic tissues from the 9th to the 12th week of gestation, and a steady further increase toward term.41–43 This increase may promote the development of cardiac adrenoceptors as well as switching to adult forms of myosin heavy chains, therefore having a more profound effect on cardiac functional development. The cardiac weight increases 200-fold between 8 and 20 weeks of gestation. Both ventricles are structurally identical at this time.15 Between 13 and 17 weeks of gestation there is a significant linear increase of all cardiac dimensions, which correlates with fetal growth and gestational age. The ratios between right and left ventricles and pulmonary trunk and aorta remain relatively constant at 1:1 and 1:1.1, respectively.44 In addition to the changes in cardiac dimensions, a significant change in embryonic heart rates with advancing gestation has been documented in response to peripheral and central maturation of autonomic cardiac control.

Development of the fetal cardiac conduction system and autonomic innervation Fetal cardiac activity is initiated by contracting ventricular myocytes on the caudal region of the straight heart tube between days 21 and 23 after conception. Development of the cardiac conduction system is ongoing during the folding and looping process of the embryonic heart and its maturation continues until after delivery. The sinoatrial and AV nodes persist as remnants of the primary myocardium of the sinus venosus and AV junction. As the atria and ventricles become electrically isolated by growth of the endocardial cushions, the AV node provides the only conductive pathway between these chambers. The ventricular conduction system develops concomitantly with ventricular septation and first becomes detectable in the myocardium surrounding the interventricular foramen. Cardiac contractions are initiated at the time of development when truncoventricular fusion commences, involving the atria and the sinus venosus. Therefore, the primitive heart pumps blood before rotation and septation. In normal pregnancies the mean heart rate is in the region of 110 beats/minute at 5 weeks of gestation. This is followed by a rise to approximately 170 beats/minute at

Development of fetal cardiac and extracardiac Doppler flows

9–10 weeks, with a subsequent fall to 150 beats/minute at 14–15 weeks and a continuing gradual further drop to term.44–47 The initial changes in heart rate are primarily attributed to the changes in cardiac morphology. The increasing heart rate in the early first trimester is predominantly due to the fusion of the sinus venous with the atria and ventricles and the development of the conductive tissue. This allows the establishment of the sinoatrial node as the primary cardiac pacemaker with the highest intrinsic spontaneous rate of rhythmicity.48,49 The subsequent decrease in heart rate is thought to reflect increasing parasympathetic modulation,50 improved ventricular contractility, increasing ventricular muscle mass51 and improved atrioventricular valvular function.15 The successive decline of baseline rate and increasing heart rate variability toward term reflect a relative increase in parasympathetic cardiac innervation and central maturation of autonomic control. In the human fetus the parasympathetic inhibition of the sinoatrial node appears to be established between 12 and 17 weeks, followed by sympathetic nerves between 22 and 24 weeks.5 Marked episodes of sinus bradycardia with excellent prognosis may be observed until this period in gestation. With formation of the AV node and the junctional connection to the ventricles, development of the conduction system is generally completed at 12 weeks. The complete formation of the annulus fibrosus may be disturbed or delayed, resulting in persistence of abnormal conduction pathways. These allow establishment of aberrant reentry conduction in some cases. Completion of the annulus fibrosus eventually leads to cessation of aberrant conduction.14,52 The sequential changes of embryonic heart rate are relatively constant. It has been realized that bradycardia is associated with a higher risk for chromosomal anomalies and/or spontaneous miscarriage.53 More recently it was demonstrated that an increase in the embryonic heart rate was associated with a higher risk for aneuploidies.54 It therefore appears that the human embryonic heart goes through an ultrastructural maturation sequence, which is coupled with the growth of the cardiac structures and the development of receptor and neural connections. The association between abnormal structural cardiac development and the presence of transient thickening of the nuchal fold suggests that intravascular volume handling in the human embryo and early fetus may be particularly vulnerable to cardiac dysfunction. Evidence of maturation of the autonomic innervation of the fetal heart has been gathered using Doppler ultrasound. Between 10 and 20 weeks of gestation, increasing variability of fetal heart rate and time-averaged velocity has been documented in the abdominal aorta.55 In addition, umbilical artery peak systolic velocity variability and fetal heart rate variability also increase in this period.56 The heart rate variability is considered as evidence of maturation of the parasympathetic nervous system, whereas peak

159

systolic velocity variability reflects the activation of a hemodynamic feedback mechanism.57,58 The documented inverse relationship between umbilical artery flow velocity and fetal heart rate during this time has been considered as evidence that the Frank–Starling mechanism regulates cardiovascular control as early as the late first and early second trimesters of pregnancy.55–58

Doppler investigation of the cardiovascular system The Doppler equation can be used to describe the relationship between the Doppler shift frequency and the absolute blood flow velocity. After resolution of the equation, the velocity can be calculated from the Doppler shift frequency if the velocity and angle of insonation are known. The processing of incoming signals via fast Fourier transformation allows the graphic display of the calculated velocities in a two-dimensional image. In such an image the flow velocity waveform is displayed, with the velocities on the ordinate and the time on the abscissa (Figure 11.3). The integral under the waveform corresponds to the maximal velocity in the time interval examined.59,60 The flow velocity waveform can be analyzed by quantitative or qualitative waveform analysis. Quantitative analysis utilizes formulas to calculate absolute velocities and volume flow, as shown in Figure 11.4. Quantitative analysis is predominantly utilized for intracardiac flow velocity waveforms. The primary parameters utilized include the absolute diastolic and systolic velocities, time averaged velocity (TAV), time averaged maximum velocity (TAMX), and time to peak velocity. The TAV can be used to calculate volume flow if the vessel diameter is known. Since inaccuracies in vessel diameter are raised to the square, measurement of absolute volume flow by Doppler requires stringent adherence to correct anatomic planes using an insonation angle close to 0°.59–61 The intracardiac flows that have been predominantly investigated include the diastolic filling across the atrioventricular valves, the systolic output across the aortic and pulmonary valves, and the flow across the foramen ovale. Extracardiac Doppler ultrasound predominantly uses qualitative waveform analysis based on angle-independent indexes. The indexes used in fetal medicine utilize the systolic, end-diastolic, and TAMX velocities. Using these indexes avoids the sources of error inherent in the use of quantitative waveform analysis and the calculation of absolute volume flow. The systolic/diastolic ratio, resistance index, and pulsatility index predominantly reflect downstream flow resistance and filling pressure of the arterial bed.60,62,63 Many authors prefer the use of the pulsatility index, since this allows ongoing waveform analysis in cases of absent or reversed end-diastolic blood

160

Fetal Cardiology

(a) Systolic peak velocity

Shaded = time averaged maximum velocity (TAMX)

Enddiastole

End-diastolic peak velocity

Time Acceleration time / time to peak velocity (b) Systolic peak velocity

Shaded = time averaged maximum velocity (TAMX) S D

Diastolic peak velocity

Peak velocity during atrial systole

a

Time

Figure 11.3 Arterial flow velocity waveform. (a) Diagrammatic representation of an arterial flow velocity waveform. The marked velocities are utilized for the calculation of indexes, while the acceleration time is used as an absolute measurement. (b) Diagrammatic representation of a venous flow velocity waveform. The triphasic waveform (S, systole; D, diastole; a, atrial contraction) reflects volume flow changes during the cardiac cycle (see text).

flow velocities.58 An increase of downstream vascular resistance results in a relative decrease of end-diastolic velocities and a subsequent increase of these three indexes (Figure 11.4). Since the beginning of the 1990s examination of the venous system has been increasingly incorporated into the assessment of the human fetus. These studies have allowed greater insight into the effects of many conditions on cardiac preload. The vessels most commonly utilized in

this context are the inferior vena cava, DV, hepatic veins, and the free umbilical vein. The flow velocity waveform of the inferior vena cava is influenced more by right ventricular function, while those of the left hepatic vein and the DV more closely represent left ventricular function.64–66 The venous flow velocity waveform consists of two peaks and troughs. The systolic and diastolic peaks (the S-wave and D-wave, respectively) are each followed by a trough. The S-wave is produced by descent of the AV-ring during ventricular systole, while the D-wave results from passive ventricular filling during early diastole. The trough that follows the S-wave is produced by ascent of the AV-ring as the ventricle relaxes. The sudden increase in right atrial pressure with atrial contraction in late diastole causes a variable amount of reverse flow, producing a second trough after the D-wave (the A-wave) (Figure 11.3b). The degree of reverse flow during atrial contraction varies considerably in individual veins. There may be reverse flow during atrial contraction in the inferior vena cava and hepatic veins. In contrast, normal blood flow in the DV is antegrade also during atrial systole. The most widely applied indexes for the assessment of venous Doppler-derived flow are the percentage of reverse flow,67 peak velocity index for veins, pulsatility index for veins,68 and the preload index.69 Venous indexes are believed primarily to reflect cardiac function and to a lesser extent cardiac afterload64–72 (Figure 11.4). Except for the preload index, all indexes mentioned increase when there is a decrease in antegrade flow, such as that associated with increased reverse flow during atrial systole. Such situations occur with increased atrial and ultimately increased central venous pressure. The umbilical venous flow velocity waveform has a constant pattern in the second and third trimesters. Pulsations in the umbilical vein are considered evidence of retrograde transmission of elevated central venous pressures. The pulsations correspond to the maximal reverse flow component of atrial contraction.73

Development of intracardiac flow velocity waveforms Diastolic ventricular filling is examined by positioning the Doppler gate immediately distal to the respective AV valves. The characteristic waveform has two peaks representing early diastolic ventricular filling (E-wave) and ventricular filling during atrial systole (A-wave) (Figure 11.5). The ratio between the E- and A-wave peak velocities is a generally accepted angle-independent index for quantification of the waveform across the AV valves. The index is thought to reflect changes in ventricular diastolic function, relaxation velocity and compliance, and pre- and afterload of the respective ventricle. In the adult,

Development of fetal cardiac and extracardiac Doppler flows

Cardiac indexes

161

Venous indexes

E/A ratio

=

E A % Reverse flow

=

= TAV x πr2

Volume flow (Q)

TVI (S+D) TVI a

S/A ratio

=

S a

Preload index

=

S–a S

Acceleration time

Arterial indexes Pulsatility index

=

S–D TAV

Peak velocity index =

S–a D

Resistance index

=

S–D S

Pulsatility index

S–a TAMX

Systolic diastolic ratio =

A-wave dominance A

E/A waves in utero

Figure 11.4 Formulas used for quantitative and qualitative waveform analysis.

E

E-wave dominance A

Velocity

E

S D

=

E/A waves after birth

Time

Figure 11.5 E/A diagram in the adult and fetus. Shown is a diagrammatic representation of the flow velocity waveform across atrioventricular valves. In the fetus a greater proportion of diastolic filling occurs during atrial systole, resulting in a higher peak velocity during this part of the cardiac cycle (A). After birth the relationship between velocities during early diastole (E) and A-wave reverses (see text).

diastolic filling occurs predominantly in the early passive phase of diastole resulting in an E/A ratio of > 1. In contrast, during fetal life, diastolic filling relies mainly on atrial contraction. Subsequently, the A-wave is generally predominant in utero, resulting in an E/A ratio of > 1. This A-wave dominance in utero is presumed to be due to the lower compliance and relaxation time of the fetal myocardium when compared to the adult. Low ventricular compliance is associated with impaired filling during the passive early phase of diastole, requiring atrial contraction

to contribute to the major part of ventricular filling. Therefore, a lower E/A ratio is thought to reflect these properties of the ventricle. Doppler measurements of the transatrioventricular flow velocity waveforms showed that monophasic AV flow velocity waveforms become biphasic as early as 8 weeks.74,75 A significant gestational age-dependent increase was observed for all AV waveform parameters, which showed a linear relationship with the crown–rump length.74–76 An increase in cardiac size and atrioventricular valve area contributes to an increase in atrioventricular blood flow volume and thus ventricular preload. In both the first and the second trimesters the transtricuspid blood flow volume exceeds that across the mitral valve.76,77 This is in keeping with the fact that the right ventricle contributes to a greater proportion of the combined cardiac output in utero. If the atrioventricular flow profiles are used for calculation of the ventricular output, reproducible results have been obtained, showing right ventricular dominance.17 With advancing gestation there is in an increase of both transtricuspid and transmitral peak E, but little change in peak A, resulting in a substantial increase in the E/A ratio from approximately 0.5 at 13 weeks of gestation to 0.8–0.9 at 36–38 weeks’ gestation.77–82 This is thought to reflect the combined effects of decreasing afterload and improved diastolic ventricular function and both myocardial relaxation and compliance, which are likely to result in a decrease of ventricular diastolic pressures, favoring passive filling during early diastole.77–82 During the second and third trimesters of pregnancy a lack of correlation between the Doppler echocardiographically assessed ventricular filling pressures, which were determined by measuring the ratio between flow velocity (E) and annular velocity (EA) in early diastole, and gestational age was shown.83 This suggests the absence of a significant change in ventricular filling pressure and a constant compliance, respectively.

162

Fetal Cardiology

Therefore, progressive enhancement of active relaxation and/or changes in loading conditions are more likely to explain variations in fetal AV blood flow than changes in ventricular compliance.83

Flow velocity waveforms in the outflow tracts Peak velocities in the ascending aorta and the main pulmonary artery are predominantly determined by cardiac contractility and afterload and, therefore, are accepted parameters of ventricular systolic performance. Measurement of outflow tract peak blood flow velocities has been performed as early as 10 weeks of gestation. In these examinations peak blood flow velocities in the ascending aorta were found to exceed those in the main pulmonary artery.77,78,82 The improvement in systolic cardiac function between 13 and 20 weeks results in an increase in output from both ventricles, which is also associated with increasing peak blood flow velocities in both outflow tracts. At 13 weeks, outflow tract velocities average 30 cm/s.84 By 20 weeks, the peak blood flow velocity in the ascending aorta averages 60 cm/s and in the main pulmonary artery 54 cm/s. With continuing gradual increase toward term these velocities increase to 60–120 cm/s in the ascending aorta and 50–110 cm/s in the main pulmonary artery.66,84,85 Calculation of the valvular area allows the estimation of the cardiac output when the TAV and heart rate are known or if the velocity integral and the heart rate are known, because a flat velocity profile can be suggested for the blood flow in the ascending aorta and the pulmonary valve directly behind the semilunar valves. In the human fetus it has been confirmed that the pulmonary arterial diameter consistently exceeds the aortic diameter, with both vessels showing a linear increase with gestational age.44 Since the pulmonary artery has a slightly larger diameter than the aorta (approximately 1.3:1) it is expected that the right ventricle contributes to a larger proportion of the cardiac output. This assumption, which was previously documented in an animal experiment, has also been confirmed in the human fetus. These findings confirm animal experimental data suggesting the intrauterine dominance of the right ventricle. A positive relationship between stroke volume and gestational age, which increases exponentially from 0.7 ml at 20 weeks to 7.6 ml at 40 weeks for the right and from 0.7 ml at 20 weeks to 5.2 ml at 40 weeks for the left ventricle, was found.84,85 Based on these measurements it has been calculated that the right ventricular stroke volume exceeds the left ventricular stroke volume by approximately 28% through gestation.85 In concordance with these data, combined cardiac outputs have been found to increase from approximately 100 ml/minute at 18 weeks’ gestation to approximately

1000 ml/minute at 38 weeks’ gestation. However, the relationship between cardiac output and fetal body weight remains relatively constant at 450 ml/kg/per minute throughout pregnancy.12,68 With increasing gestational age there is a relative shift in cardiac output towards the left ventricle. The ratio between right and left ventricular output decreases from 1.25 at 18 weeks’ gestation to 1.15 at 38 weeks’ gestation.61 The time to peak velocity (TPV) or acceleration time of the arterial flow velocity waveform is primarily determined by mean arterial pressure and to a lesser degree by ventricular contractility. There is an inverse relationship between the TPV and the mean arterial pressure. Several studies have come to different conclusions regarding the TPV, and there is still controversy about the utility and significance of TPV investigation in the human fetus.86 Machado et al found that between 16 and 30 weeks the TPV was significantly shorter in the pulmonary artery than in the aorta (32.1 ms compared to 43.7 ms), suggesting that in the mid-trimester fetus the mean pulmonary arterial pressure is higher than in the aorta.87 van Splunder et al74 have confirmed these findings for the first-trimester fetus. In contrast, Rizzo et al88 found no major differences between these two vessels, while Sutton et al documented a consistently higher TPV in the aorta.89 Chaoui et al were able to document that the TPV and the ratio between TPV and the ejection time increased significantly between 18 weeks and term.90 These changes throughout gestation are in accordance with animal experiments and suggest that the mean arterial pressure in the main pulmonary artery is higher than in the ascending aorta. In addition, the changes in ejection times and vascular resistances indicate an increase in perfusion as well as a decrease in the vascular resistance and pressure in the pulmonary circulation.90 The majority of authors agree that the difference in TPV between the aorta and the pulmonary artery becomes less marked and may be negligible close to term.

Flow velocity waveforms in the ductus arteriosus The ductus arteriosus provides a conduit between the main pulmonary artery and the descending aorta. It originates near the origin of the left pulmonary artery and ends in the descending aorta immediately distal to the left subclavian artery. From approximately 26 weeks of gestation, the ductus arteriosus exhibits increasing sensitivity toward prostaglandin.91 Most authors agree that right ventricular pressures are predominantly determined by the blood flow resistance in the ductus arteriosus. Systolic peak blood flow velocities in this vessel increase from 50 cm/s at 15 weeks’ gestation to 130–160 cm/s near term.92

Development of fetal cardiac and extracardiac Doppler flows

End-diastolic flow can first be detected between 13 and 14 weeks’ gestation and is generally established by 17 weeks’ gestation.93 The pulsatility index in the ductus arteriosus is relatively high (2.46 ± 0.52) and remains relatively constant throughout pregnancy.94 In constriction of the ductus arteriosus there is an increase of arterial peak blood velocities, which is more marked in diastole. This results in a fall of the pulsatility index below 1.9 in mild cases and as low as 1.0 with associated tricuspid insufficiency in severe constriction.91–94 Measurement of the timing of cardiac contractions has allowed a comparison to be made between the duration of individual cardiac cycle phases for individual ventricles. A longer diastolic filling time has been found on the left side of the heart (197 ms) compared with the right (174 ms). In contrast, the ejection time is shorter for the left ventricle (174 ms) than on the right side of the heart (189 ms).95,96 The significant decrease in heart rate between 10 and 20 weeks’ gestation results in a lengthening of the cardiac cycle from 373 to 406 ms. Concurrently, there is a significant increase of the diastolic filling time associated with a linear decrease in isovolumetric relaxation and ejection time. This development documents several maturation processes in fetal cardiac function including elevation of myocardial relaxation velocity, increasing compliance, improved ventricular contractility, and decreasing cardiac afterload.74,76

Extracardiac arterial Doppler indexes The blood flow resistance in the ascending aorta and therefore left ventricular afterload is predominantly determined by the brachiocephalic circulation (predominantly the common carotid arteries and to a lesser extent the left subclavian artery).97 Accordingly, changes in vascular resistance in the common carotid arteries reflect predominantly cerebral vascular blood flow resistance (external carotid, internal carotid, and cerebral arteries). The blood flow resistance in these vessels develops similarly, and therefore many authors prefer the middle cerebral artery, since it is readily accessible at close to 0° insonation. Right ventricular afterload is determined by the blood flow velocities in the pulmonary circulation, the ductus arteriosus, and the vascular beds distal to the ductus (descending aorta with its organ branches, fetoplacental circulation).

The cerebral circulation Intracerebral flow velocity waveforms have been visualized as early as 10 weeks of gestation. This early in pregnancy it

163

is often not possible to distinguish between the individual cerebral arteries.81,98 It is of note that in intracerebral vessels end-diastolic velocities are present between 11 and 13 weeks. The pulsatility index of the intracerebral vessels and the internal carotid artery is relatively constant until 20 weeks of gestation.81,97,98 Thereafter, a decrease in the pulsatility index is observed in all cerebral vessels toward term.81 The pulsatility index in the middle cerebral artery is higher than in all other cerebral vessels. Most authors report a modest rise in the pulsatility index until approximately 26 weeks, with a subsequent fall toward term.99,100 Since the pulsatility index is also influenced by filling pressure, an increase in cerebral blood flow volume may be an explanation for this observation. The decrease of the pulsatility index with advancing gestation suggests a decrease in cerebral blood flow resistance.

The pulmonary circulation The vascular connections between intrapulmonary arterioles and the main pulmonary vessels are established between 35 and 50 days after conception. After 16 weeks there is increasing arborization of the pulmonary vascular tree, which continues toward term.101,102 The intrapulmonary arteries possess a relatively thick muscular coat and therefore have a small lumen. In utero the pulmonary circulation is a low-flow, high-resistance vascular bed. In the fetal lamb only 3.7% of the common cardiac output reaches the pulmonary circulation in mid-gestation. This proportion increases to 7% at term.103 The high blood flow resistance in the pulmonary circulation is predominantly determined by the overall small cross-sectional area of the pulmonary vascular bed.103,104 Therefore, fetoplacental blood flow resistance is a more important determinant of the flow velocity in the main pulmonary artery.15 Several studies have been undertaken to investigate vascular development in the branch pulmonary arteries, and have yielded inconsistent results. The pulmonary blood flow pattern at the proximal and middle branch level is characterized by a rapid early systolic flow acceleration, a sharp mid-systolic deceleration, a deep notch or reverse flow in early diastole, and a low diastolic forward flow.105,106 The high pulmonary vascular resistance compared to the fetal systemic circulation and the preferential blood flow through the patent ductus arteriosus into the descending aorta may explain this blood flow pattern. The distal branches, however, shows a monophasic forward flow pattern with lower pulsatility and acceleration and deceleration velocities.105,106 The peak velocity and pulsatility of blood flow significantly decrease if the Doppler interrogation moves from the proximal to the distal arterial branches.105,106 A high pulsatility index has been documented by several investigators, but findings regarding the development of the pulsatility index from 15 weeks to term differ amongst investigators.105–109 Laudy et al found

164

Fetal Cardiology

no significant change in the pulsatility index in the proximal pulmonary branch artery, but a significant gestational age-related drop in the systolic/diastolic (S/D) ratio.108 Their findings suggest that analysis of the unique waveform of the branch pulmonary arteries using the pulsatility index may not accurately reflect downstream vascular resistance. It is likely that there is a fall in the pulmonary vascular impedance with gestational age. Rasanen et al found a decreasing pulsatility index with gestational age in the proximal and distal branch pulmonary arteries. This decrease was linear until approximately 31 weeks and then remained relatively unchanged.109 Laudy et al demonstrated a similar decrease in the pulsatility index between 20 and 30 weeks followed by a significant increase from 31 weeks until term.105 They concluded that fetal branch pulmonary arterial vascular impedance decreases significantly between 20 and 30 weeks and suggested an increase of pulmonary vascular resistance during the last 8–10 weeks of pregnancy.105–109 Recently, Sivon et al found a slight constant increase of pulsatility index in the proximal, middle, and distal segments of the branch pulmonary artery, but without statistical significance.106 The lack of change in branch pulmonary artery pulsatility index in later gestation could suggest that during this time lung growth and increase in vascular cross-sectional area result in an in overall increase in blood flow in this area.105–110 Indeed, Sutton et al found that blood flow through the lungs increased exponentially with gestational age by almost four-fold between 18 and 37 weeks, and was a mean of 22% of combined ventricular output.13 Foramen ovale blood flow increased exponentially three-fold, representing between 17 and 31% of combined ventricular output. The proportions of these flows remained unchanged through the second and third trimesters of pregnancy.13 Similarly, Rasanen et al established that, from 20 to 30 weeks of gestation, the proportion of pulmonary blood flow increased from 13 to 25% and maintained a constant pattern thereafter.111 However, they documented that the proportion of foramen ovale blood flow decreased from 34 to 18% with little change in the proportions after 30 weeks. At 38 weeks, the right ventricular output was 60% of the combined cardiac output, indicating that the development of the human fetal pulmonary circulation has an important role in the distribution of cardiac output.111 Despite this increase in total pulmonary flow during the second half of gestation, the TAMX and peak systolic velocity at the proximal, middle, and distal pulmonary branch level did not reveal significant gestational age-related change, suggesting that the increase of vessel diameter, rather than a rise of flow velocities, determines the observed increase of pulmonary blood flow.105,106 Further research into these relationships is necessary. It appears safe to assume that there is a progressive increase in pulmonary blood flow with advancing gestation in the human fetus. In the second trimester this may be due to a decrease in pulmonary vascular impedance to blood

flow, while in the third trimester an increase in lung volume and overall vascular cross-sectional area may be the major contributor to this development.90,105–111

Descending aorta and fetal visceral blood flow In the descending aorta, end-diastolic blood flow velocities become detectable between 13 and 15 weeks of gestation, therefore later than in the cerebral circulation. Between 16 and 20 weeks there is a rapid fall of the pulsatility index. After this time the pulsatility index in the descending aorta remains relatively constant toward term, despite a linear increase in blood flow volume.112 This is analogous to the development in the ductus arteriosus, indicating that there is relatively little change in blood flow resistance in these vessels after 20 weeks of gestation. This has been attributed to divergent development of blood flow resistance in distal vascular beds. Doppler sonography has been performed in many vascular territories originating from the descending aorta. These include splenic, hepatic, mesenteric, adrenal, renal, and iliac arteries, and, further distally, the femoral arteries113–119 (Figure 11.6). In the renal and adrenal arteries, a linear decrease in the pulsatility index has been observed from 20 to 38 weeks.113,114,119 The splenic artery shows a temporary rise in the pulsatility index in mid-gestation with a subsequent decline toward term,115 while the mesenteric and hepatic arteries represent high-impedance vascular beds.114,116,117 Distal to the bifurcation into the iliac arteries, the vascular beds behave differently with respect to the development of resistance indexes. The examination of the external iliac arteries and femoral arteries reflects development of vascular resistance in the lower body. In the femoral artery there is a linear increase in the pulsatility index with advancing gestation.114,118 Throughout gestation there is a diastolic flow reversal in this vessel and the external iliac artery, indicating the high blood flow resistance in these vessels.118 The blood flow pattern in the internal iliac artery is predominantly determined by placental and umbilical cord blood flow resistance and therefore shows a different development during gestation. End-diastolic blood flow is established at a similar time in gestation to that in the umbilical artery. Thereafter there is a continuous decrease in pulsatility index paralleling the development in the placental circulation. Since the pulsatility index in the descending aorta remains relatively constant, it is presumed that the sum of all these divergent changes in flow resistance overall cause little change in aortic blood flow resistance.112,114 Importantly, the different blood flow resistance is likely to have an impact on the regional distribution of blood flow in the vascular beds distal to the descending aorta.

Development of fetal cardiac and extracardiac Doppler flows

165

Afterload

PI femoral artery PI external iliac artery PI hepatic artery LV PI DA PI DAO RV PI splenic artery PI renal artery PI umbilical artery PI carotid arteries PI cerebral arteries

20

25

30

35

Gestational age (weeks)

Uteroplacental blood flow The development of the flow velocity waveform in the maternal and fetal compartments of the placenta is suggestive of a progressive fall in vascular resistance with advancing gestational age. The diastolic notch in the spiral artery flow velocity waveform disappears by 13 weeks’ gestation, followed by the arcuate arteries within a further 2 weeks; intervillous flow fully develops by 14 weeks’ gestation.120,121 To evaluate the placental blood flow resistance in the fetal compartment the umbilical arteries were examined. Increasing differentiation of the villous vascular tree into tertiary villi results in an overall decrease in placental blood flow resistance. End-diastolic flow is first appreciable at 13–15 weeks’ gestation, and there is a continuous rise in end-diastolic velocities contributing to the linear fall in umbilical artery pulsatility index toward term.121–123 Examination of the umbilical flow velocity waveform using Laplace transform techniques suggests that vessel wall tone also decreases at the beginning of the second trimester.124 These changes in flow velocity waveforms show the increased development of the villous vascular tree and the development of tertiary villi.

Extracardiac venous Doppler indexes The blood flow volume in the umbilical vein approximates 140–180 ml/kg per minute in the second trimester and decreases slightly to 110–170 ml/kg per minute toward term.123,125,126 It is of note that, although umbilical venous volume flow increases linearly with fetal weight, the

40

Figure 11.6 Development of ventricular afterload in gestation. With advancing gestation there is a steady decrease in the pulsatility index (PI) in the cerebral circulation and therefore left ventricular (LV) afterload. In contrast, there is little change in the pulsatility index in the ductus arteriosus (DA) and descending aorta (DAO). This may be due to the divergent development of blood flow resistance in downstream vascular beds. RV, right ventricle.

volume flow per unit body weight changes little with gestational age. In the first trimester the umbilical vein has a pulsatile flow pattern, and a constant Doppler flow pattern is generally established at approximately 13 weeks’ gestation.127 There are differences in flow velocity waveforms in precordial venous vessels. The inferior vena cava and hepatic veins have a flow velocity waveform with reverse flow during atrial contraction. In contrast, ductus venosus blood flow is antegrade throughout the cardiac cycle. This is partly due to the fact that there is marked antegrade acceleration of blood flow as umbilical venous blood enters the ductus venosus. Therefore, the highest venous blood flow velocities are found in the ductus venosus. These range between 65 and 75 cm/s during systole in the first trimester and increase linearly toward term. With increasing gestational age there is a continuous decrease in venous Doppler indexes, which is apparent for the precordial veins.128–130 This is predominantly due to an increased antegrade flow component during atrial systole. This decrease in indexes is most marked between 13 and 20 weeks’ gestation, and is likely to represent the combined effects of improved cardiac diastolic function and decreasing afterload, resulting in an overall decrease in intraventricular end-diastolic pressures. Concurrent with the decrease in venous Doppler indexes the proportion of fetuses with pulsatile flow in the free umbilical vein decreases, especially between 11 and 13 weeks of gestation.70,73,127–131 The flow velocity waveform in the pulmonary veins has a biphasic profile with a systolic peak and a second peak during the early part of ventricular diastole, and is inversely related to the left atrial pressure pulsations.132–134 In contrast to the adult133 there is continuous antegrade flow toward the left atrium in the human fetus.132–135 This continuous forward flow even during atrial systole suggests

166

Fetal Cardiology

that the fetal extraparenchymal pulmonary venous system is not spacious and compliant. Therefore, extraparenchymal pulmonary venous pressure may be higher than the left atrial pressure even at atrial systole.135 From approximately 20 weeks onward there is a marked increase in antegrade peak blood flow velocities, contributing to a significant decrease in the S/D ratio.132 This development is likely to be the result of several factors. There may be a significant increase in pulmonary blood flow volume during this time. In addition, there may be a decrease in the pressure gradient toward the left ventricle,132 presumably due to improved diastolic ventricular filling facilitating atrial emptying during diastole. Therefore, the decreasing S/D ratio in the pulmonary venous vessels documents improved left ventricular diastolic function and increased preload.133,134

disappearance of venous pulsation in the free umbilical portion of the umbilical vein. These changes indicate an enhanced antegrade venous flow toward the atrial chambers, which ultimately results in an increase in preload. Between 10 and 20 weeks of gestation there is a significant increase in cardiac cycle length due to a decrease in fetal heart rate. Despite this lengthening of the cardiac cycle, ejection time and isovolumic relaxation time show a significant decrease.136 The concomitant improvement of diastolic function is necessary for an efficient accommodation of this increased preload. Indeed, a significant decrease in the atrial contribution to ventricular filling has been demonstrated between 10 and 14 weeks of gestation, indicating improved diastolic function and the increase in cardiac output necessary to maintain adequate growth.136,137 Changes in ventricular afterload are of special significance in regulating organ flow in the fetus, since the ‘afterload sensitivity’ of the fetal ventricular myocardium limits the ability to regulate cardiac output by other mechanisms. At 6 weeks of gestation approximately 20% of the cardiac cycle is occupied by the isovolumic contraction, while isovolumic relaxation occupies approximately 16%.75 The isovolumic contraction time shortens progressively and is not measurable after 12 weeks, suggesting effective heart compliance by this time.75 The persistent decrease in right and left ventricular afterload in the first and early second trimesters associated with the increase in preload and growth of the fetal heart result in an exponential increase in combined cardiac output. During the second and third trimesters relative distribution of the right and left ventricular outputs is influenced by the changes in afterload of the individual ventricles.138 End-diastolic velocities are discernible in the cerebral circulation before the fetoplacental circulation. With ongoing development there is a

Sequential changes of flow velocity waveforms The immature fetal heart has limited diastolic and systolic function in the first trimester, limiting its ability to accommodate changes in cardiac preload and afterload. Cardiac development during the following weeks results in improved cardiac function throughout the cardiac cycle. Enhanced diastolic function is manifested by changes in intracardiac and extracardiac flow velocity waveforms – there is change from a monophasic to a biphasic flow velocity profile during ventricular inflow by 8–10 weeks, which is followed by elevation of the E/A ratio. The latter changes occur in association with a decrease of pulsatility in precordial venous Doppler flow velocity waveform and

Decreasing pulmonary venous indexes Decreasing precordial venous indexes UV pulsations

Constant UV flow Increasing impedance lower body

EDF UA EDF MCA

Decreasing placental impedance Decreasing cerebral impedance

Exponentially rising cardiac output

Steadily rising cardiac output

Exponentially rising E/A ratio

Steadily rising E/A ratio

Figure 11.7

Steady decrease in baseline heart rate Heart rate Steady increase in cardiac compliance and contractility 8

13

15

17

20

24

28

Gestational age (weeks)

32

36

40

Development of cardiac and extracardiac functionality with gestation. Illustrated is the approximate temporal relationship of arterial and venous flow velocity changes and cardiac function with advancing gestational age. UV, umbilical vein; EDF, end-diastolic flow; UA, umbilical artery; MCA, middle cerebral artery.

Development of fetal cardiac and extracardiac Doppler flows

progressive drop in the pulsatility indexes and presumably also blood flow resistance in the brachiocephalic circulation. At the same time, from 20 weeks onwards pulsatility indexes in the ductus arteriosus and the descending aorta remain relatively constant. This divergent development of ventricular afterload results in physiological redistribution of the cardiac output in favor of the left ventricle and therefore the upper body. In the vascular beds distal to the descending aorta, development of Doppler indexes is also divergent in pregnancy. While vascular resistance in the limbs and therefore the external iliac arteries increases, there is a steady decrease in the umbilical artery and therefore the internal iliac pulsatility index. Owing to this development, the major proportion of descending aorta blood flow is distributed to the placental vascular bed for oxygenation. The combined cardiac output increases steadily toward term. The increased contractility and preload and decrease in afterload are associated with increased peak blood flow velocities and an exponential increase of cardiac output during gestation. After 18 weeks’ gestation the relationship between cardiac output and fetal body weight remains relatively constant at 450 ml/kg per minute. Although placental blood flow increases from 115 ml/minute from 20 weeks to 415 ml/minute at term, it decreases in relation to fetal body weight, resulting in relative placental insufficiency at term.126 Improved resolution of modern ultrasound machines allows the study of small-caliber vessels. Studies of human fetuses with growth restriction suggest that the cerebral, adrenal, and coronary circulations are capable of autoregulation in the second trimester.113,139–142 In addition, the liver and spleen appear to be preferentially perfused organ systems.115,117 With delivery of the fetus, the fetal circulation gathers its postnatal functionality by closure of the central shunts143 (Figure 11.7).

References 1. FitzGerald DE, Drumm JE. Non-invasive measurement of human fetal circulation using ultrasound: a new method. Br Med J 1977; 2: 1450–1. 2. Matias A, Gomes C, Flack N, Montenegro N, Nicolaides KH. Screening for chromosomal abnormalities at 10–14 weeks: the role of ductus venosus blood flow. Ultrasound Obstet Gynecol 1998; 12: 380–4. 3. Matias A, Huggon I, Areias JC, Montenegro N, Nicolaides KH. Cardiac defects in chromosomally normal fetuses with abnormal ductus venosus blood flow at 10–14 weeks. Ultrasound Obstet Gynecol 1999; 14: 307–10. 4. Baschat AA, Gembruch U, Knöpfle G, Hansmann M. Firsttrimester fetal heart block: a marker for cardiac anomaly. Ultrasound Obstet Gynecol 1999; 14: 311–14. 5. Allan LD, Sharland GK, Tynan MJ. The natural history of the hypoplastic left heart syndrome. Int J Cardiol 1989; 25: 341–3.

167

6. Nagueh SF, Mikati I, Kopelen HA et al. Doppler estimation of left ventricular filling pressure in sinus tachycardia. A new application of tissue doppler imaging. Circulation 1998; 98: 1644–50. 7. Guyton AC, Ross JH, Carrier O. Evidence for tissue demand as the major factor causing autoregulation. Circ Res 1964; 15(Suppl I): 60–9. 8. Edelstone DI, Rudolph AM, Heymann MA. Liver and ductus venosus blood flows in fetal lambs in utero. Circ Res 1978; 42: 426–33. 9. Rudolph AM. Distribution and regulation of blood flow in the fetal and neonatal lamb. Circ Res 1985; 57: 811–21. 10. Chacko AW, Reynolds SR. Embryonic development of the sphincter of the ductus venosus. Anat Rec 1953; 115: 151–73. 11. Kiserud T, Eik-Nes SH, Blaas HG, Hellevik LR. Ultrasonographic velocimetry of the fetal ductus venosus. Lancet 1991; 338: 1412–14. 12. Wilson AD, Rao PS, Aeschlimann S. Normal fetal foramen flap and transatrial Doppler velocity pattern. J Am Soc Echocardiogr 1990; 3: 491–4. 13. Sutton MS, Groves A, MacNeill A, Sharland G, Allan L. Assessment of changes in blood flow through the lungs and foramen ovale in the normal human fetus with gestational age: a prospective Doppler echocardiographic study. Br Heart J 1994; 71: 232–7. 14. Long WA, Henry GW. Autonomic and central neuroregulation of fetal cardiovascular function. In: Polin RA, Fox WW, eds. Fetal and Neonatal Physiology. Philadelphia: WB Saunders, 1992: 629–45. 15. Sutton MS, Gill T, Plappert T. Functional anatomic development in the fetal heart. In: Polin RA, Fox WW, eds. Fetal and Neonatal Physiology. Philadelphia: WB Saunders, 1992: 598–609. 16. Teitel DF. Physiologic development of the cardiovascular system of the fetus. In: Polin RA, Fox WW, eds. Fetal and Neonatal Physiology. Philadelphia: WB Saunders, 1992: 609–19. 17. De Smedt MC, Visser GH, Meijboom EJ. Fetal cardiac output estimated by Doppler echocardiography during mid- and late gestation. Am J Cardiol 1987; 60: 338–42. 18. Allan LD, Chita SK, Al-Ghazali W, Crawford DC, Tynan M. Doppler echocardiographic evaluation of the normal human fetal heart. Br Heart J 1987; 57: 528–33. 19. Kenny J, Plappert T, Doubilet P, Salzman D, Sutton MG. Effects of heart rate on ventricular size, stroke volume, and output in the normal human fetus: a prospective Doppler echocardiographic study. Circulation 1987; 76: 52–8. 20. Kiserud T, Rasmussen S, Skulstad S. Blood flow and the degree of shunting through the ductus venosus in the human fetus. Am J Obstet Gynecol 2000; 182: 147–53. 21. Kiserud T. Fetal venous circulation – an update on hemodynamics. J Perinat Med 2000; 28: 90–6. 22. Tonge HM, Wladimiroff JW, Noordam MJ, Stewart PA. Fetal cardiac arrhythmias and their effect on volume blood flow in descending aorta of human fetus. J Clin Ultrasound 1986; 14: 607–12. 23. Reed KL, Sahn DJ, Marx GR, Anderson CF, Shenker L. Cardiac Doppler flows during fetal arrhythmias: physiologic consequences. Obstet Gynecol 1987; 70: 1–6.

168

Fetal Cardiology

24. Gilbert RD. Effects of afterload and baroreceptors on cardiac function in fetal sheep. J Dev Physiol 1982; 4: 299–309. 25. Hawkins J, Van Hare GF, Schmidt KG, Rudolph AM. Effects of increasing afterload on left ventricular output in fetal lambs. Circ Res 1989; 65: 127–34. 26. Moise KJ Jr, Mari G, Fisher DJ et al. Acute fetal hemodynamic alterations after intrauterine transfusion for treatment of severe red blood cell alloimmunization. Am J Obstet Gynecol 1990; 163: 776–84. 27. Indik JH, Reed KL. Variation and correlation in human fetal umbilical Doppler velocities with fetal breathing: evidence of the cardiac–placental connection. Am J Obstet Gynecol 1990; 163: 1792–6. 28. Barany M. ATPase activity of myosin correlated with speed of muscle shortening. J Gen Physiol 1967; 50(Suppl): 197–218. 29. Schwartz K, Lecarpentier Y, Martin J et al. Myosin isoenzymic distribution correlates with speed of myocardial contraction. J Mol Cell Cardiol 1981; 13: 1071–5. 30. Nassar R, Reedy MC, Anderson PA. Developmental changes in the ultrastructure and sarcomere shortening of the isolated rabbit ventricular myocyte. Circ Res 1987; 61: 465–83. 31. Sweeney LJ, Zak R, Manasek FJ. Transitions in cardiac isomyosin expression during differentiation of the embryonic chick heart. Circ Res 1987; 61: 287–95. 32. Cummins P, Price KM, Littler WA. Foetal myosin light chain in human ventricle. J Muscle Res Cell Motil 1980; 1: 357–66. 33. Nakanishi T, Okuda H, Kamata K et al. Development of myocardial contractile system in the fetal rabbit. Pediatr Res 1987; 22: 201–7. 34. Ingwall JS, Kramer MF, Woodman D, Friedman WF. Maturation of energy metabolism in the lamb: changes in myosin ATPase and creatine kinase activities. Pediatr Res 1981; 15: 1128–33. 35. Maylie JG. Excitation-contraction coupling in neonatal and adult myocardium of cat. Am J Physiol 1982; 242: H834–43. 36. Mahony L, Jones LR. Developmental changes in cardiac sarcoplasmic reticulum in sheep. J Biol Chem 1986; 261: 152–7. 37. Mahdavi V, Izumo S, Nadal-Ginard B. Developmental and hormonal regulation of sarcomeric myosin heavy chain gene family. Circ Res 1987; 60: 804–14. 38. Nadal-Ginard B, Mahdavi V. Molecular basis of cardiac performance. Plasticity of the myocardium generated through protein isoform switches. J Clin Invest 1989; 84: 1693–700. 39. Whitsett JA, Pollinger J, Matz S. Beta-adrenergic receptors and catecholamine sensitive adenylate cyclase in developing rat ventricular myocardium: effect of thyroid status. Pediatr Res 1982; 16: 463–9. 40. Williams LT, Lefkowitz RJ, Watanabe AM, Hathaway DR, Besch HR Jr. Thyroid hormone regulation of beta-adrenergic receptor number. J Biol Chem 1977; 252: 2787–9. 41. Nathanielsz PW, Silver M, Comline RS. Plasma triiodothyronine concentration in the foetal and newborn lamb. J Endocrinol 1973; 58: 683–4. 42. Costa A, Arisio R, Benedetto C, Cheng SY, Morkin E. Thyroid hormones in tissues from human embryos and fetuses. J Endocrinol Invest 1991; 14: 559–68.

43. Edwards JG, Bahl JJ, Flink IL, Cheng SY, Morkin E. Thyroid hormone influences beta myosin heavy chain (beta MHC) expression. Biochem Biophys Res Commun 1994; 199: 1482–8. 44. Gembruch U, Shi C, Smrcek JM. Biometry of the fetal heart between 10 and 17 weeks of gestation. Fetal Diagn Ther 2000; 15: 20–31. 45. Shenker L, Astle C, Reed K, Anderson C. Embryonic heart rates before the seventh week of pregnancy. J Reprod Med 1986; 31: 333–5. 46. Schats R, Jansen CA, Wladimiroff JW. Embryonic heart activity: appearance and development in early human pregnancy. Br J Obstet Gynaecol 1990; 97: 989–94. 47. Wisser J, Dirschedl P. Embryonic heart rate in dated human embryos. Early Hum Dev 1994; 37: 107–15. 48. Yamauchi A. Electron microscopic observations on the development of S-A and A-V nodal tissues in the human embryonic heart. Z Anat Entwicklungsgesch 1965; 124: 562–87. 49. James TN. Cardiac conduction system: fetal and postnatal development. Am J Cardiol 1970; 25: 213–26. 50. Friedman WF, Pool PE, Jacobowitz D, Seagren SC, Braunwald E. Sympathetic innervation of the developing rabbit heart. Biochemical and histochemical comparisons of fetal, neonatal, and adult myocardium. Circ Res 1968; 23: 25–32. 51. Clark EB, Hu N, Dummett JL et al. Ventricular function and morphology in chick embryo from stages 18 to 29. Am J Physiol 1986; 250: H407–13. 52. Kirby ML. Development of the fetal heart. In: Polin RA, Fox WW, eds. Fetal and Neonatal Physiology. Philadelphia: WB Saunders, 1992: 589–97. 53. Rempen A. Diagnosis of viability in early pregnancy with vaginal sonography. J Ultrasound Med 1990; 9: 711–16. 54. Hyett JA, Noble PL, Snijders RJ, Montenegro N, Nicolaides KH. Fetal heart rate in trisomy 21 and other chromosomal abnormalities at 10–14 weeks of gestation. Ultrasound Obstet Gynecol 1996; 7: 239–44. 55. Ursem NT, Clark EB, Keller BB, Hop WC, Wladimiroff JW. Assessment of fetal heart rate variability and velocity variability by Doppler velocimetry of the descending aorta at 10–20 weeks of gestation. Ultrasound Obstet Gynecol 1999; 14: 397–401. 56. Ursem NT, Struijk PC, Hop WC et al. Heart rate and flow velocity variability as determined from umbilical Doppler velocimetry at 10–20 weeks of gestation. Clin Sci 1998; 95: 539–45. 57. Montenegro N, Ramos C, Matias A, Barros H. Variation of embryonic/fetal heart rate at 6–13 weeks’ gestation. Ultrasound Obstet Gynecol 1998; 11: 274–6. 58. Ursem NT, Kempski MH, de Ridder MA, Clark EB, Wladimiroff JW. An estimate of fetal autonomic state by spectral analysis of human umbilical artery flow velocity waveforms. Cardiovasc Res 1998; 37: 601–5. 59. Kremkau F. Doppler Ultrasound – Principles and Instruments. Philadelphia: WB Saunders, 1995. 60. Maulik D, Yarlaggada P. In vitro validation of Doppler waveforms indices. In: Maulik D, McNellis D, eds. Doppler Ultrasound Measurement of Normal Fetal Hemodynamics. New York: Perinatology Press, 257–82. 61. Beeby AR, Dunlop W, Heads A, Hunter S. Reproducibility of ultrasonic measurement of fetal cardiac haemodynamics. Br J Obstet Gynaecol 1991; 98: 807–14.

Development of fetal cardiac and extracardiac Doppler flows

62. Gosling RG, Lo PTS, Taylor MG. Interpretation of pulsatility index in feeder arteries to low impedance vascular beds. Ultrasound Obstet Gynecol 1991; 1: 175–9. 63. Pourcelot L. Applications cliniques de l’examen Doppler transcutane. Velocimetrie Ultrasonore Doppler 1974; 34: 625–7. 64. Hecher K, Campbell S, Doyle P, Harrington K. Nicolaides K. Assessment of fetal compromise by Doppler ultrasound investigation of the fetal circulation. Arterial, intracardiac, and venous blood flow velocity studies. Circulation 1995; 91: 129–38. 65. Rizzo G, Arduini D, Romanini C, Mancuso S. Doppler echocardiographic assessment of atrioventricular velocity waveforms in normal and small-for-gestational-age fetuses. Br J Obstet Gynaecol 1988; 95: 65–9. 66. Rizzo G, Arduini D, Romanini C. Doppler echocardiographic assessment of fetal cardiac function. Ultrasound Obstet Gynecol 1992; 2: 434–45. 67. Rizzo G, Capponi A, Arduini D, Romanini C. The value of fetal arterial, cardiac and venous flows in predicting pH and blood gases measured in umbilical blood at cordocentesis in growth retarded fetuses. Br J Obstet Gynaecol 1995; 102: 963–9. 68. Hecher K, Campbell S, Snijders R, Nicolaides K. Reference ranges for fetal venous and atrioventricular blood flow parameters. Ultrasound Obstet Gynecol 1994; 4: 381–90. 69. Kanzaki T, Chiba Y. Evaluation of the preload condition of the fetus by inferior vena caval blood flow pattern. Fetal Diagn Ther 1990; 5: 168–74. 70. Reed KL, Chaffin DG, Anderson CF. Umbilical venous Doppler velocity pulsations and inferior vena cava pressure elevations in fetal lambs. Obstet Gynecol 1996; 87: 617–20. 71. Severi FM, Rizzo G, Bocchi C et al. Intrauterine growth retardation and fetal cardiac function. Fetal Diagn Ther 2000; 15: 8–19. 72. Reed KL, Anderson CF. Changes in umbilical venous velocities with physiologic perturbations. Am J Obstet Gynecol 2000; 182: 835–8. 73. Reed KL, Chaffin DG, Anderson CF, Newman AT. Umbilical venous velocity pulsations are related to atrial contraction pressure waveforms in fetal lambs. Obstet Gynecol 1997; 89: 953–6. 74. van Splunder P, Stijnen T, Wladimiroff JW. Fetal atrioventricular flow-velocity waveforms and their relation to arterial and venous flow-velocity waveforms at 8 to 20 weeks of gestation. Circulation 1996; 94: 1372–8. 75. Leiva MC, Tolosa JE, Binotto CN et al. Fetal cardiac development and hemodynamics in the first trimester. Ultrasound Obstet Gynecol 1999; 14: 169–74. 76. van Splunder IP, Wladimiroff JW. Cardiac functional changes in the human fetus in the late first and early second trimesters. Ultrasound Obstet Gynecol 1996; 7: 411–15. 77. Reed KL, Meijboom EJ, Sahn DJ et al. Cardiac Doppler flow velocities in human fetuses. Circulation 1986; 73: 41–6. 78. Wladimiroff JW, Stewart PA, Burghouwt MT, Stijnen T. Normal fetal cardiac flow velocity waveforms between 11 and 16 weeks of gestation. Am J Obstet Gynecol 1992; 167: 736–9. 79. Labovitz AJ, Pearson AC. Evaluation of left ventricular diastolic function: clinical relevance and recent Doppler echocardiographic insights. Am Heart J 1987; 114: 836–51.

169

80. Stoddard MF, Pearson AC, Kern MJ et al. Influence of alteration in preload on the pattern of left ventricular diastolic filling as assessed by Doppler echocardiography in humans. Circulation 1989; 79: 1226–36. 81. Rizzo G, Arduini D, Romanini C. Fetal cardiac and extracardiac circulation in early gestation. J Matern Fetal Invest 1991; 1: 73–8. 82. van der, Mooren K, Barendregt LG, Wladimiroff JW. Fetal atrioventricular and outflow tract flow velocity waveforms during normal second half of pregnancy. Am J Obstet Gynecol 1991; 165: 668–74. 83. Pacileo G, Paladini D, Russo MG et al. Echocardiographic assessment of ventricular filling pressure during the second and third trimesters of gestation. Ultrasound Obstet Gynecol 2000; 16: 128–32. 84. Sharkey A, Tulzer G, Huhta J. Doppler blood velocities in the first trimester of pregnancy. Am J Obstet Gynecol 1991; 164: 331–5. 85. Kenny JF, Plappert T, Doubilet P et al. Changes in intracardiac blood flow velocities and right and left ventricular stroke volumes with gestational age in the normal human fetus: a prospective Doppler echocardiographic study. Circulation 1986; 74: 1208–16. 86. Silverman NH. Acceleration time in the aorta and pulmonary artery measured by Doppler echocardiography in the midtrimester normal human fetus. Br Heart J 1988; 59: 639–40. 87. Machado MV, Chita SC, Allan LD. Acceleration time in the aorta and pulmonary artery measured by Doppler echocardiography in the midtrimester normal human fetus. Br Heart J 1987; 58: 15–18. 88. Rizzo G, Arduini D, Romanini C, Mancuso S. Doppler echocardiographic assessment of time to peak velocity in the aorta and pulmonary artery of small for gestational age fetuses. Br J Obstet Gynaecol 1990; 97: 603–7. 89. Sutton MS, Gill T, Plappert T, Saltzman DH, Doubilet P. Assessment of right and left ventricular function in terms of force development with gestational age in the normal human fetus. Br Heart J 1991; 66: 285–9. 90. Chaoui R, Taddei F, Rizzo G et al. Doppler echocardiography of the main stems of the pulmonary arteries in the normal human fetus. Ultrasound Obstet Gynecol 1998; 11: 173–9. 91. Huhta JC. The fetal ductus arteriosus. In: Copel J, Reed K, eds. Doppler Ultrasound in Obstetrics and Gynecology. New York: Raven Press, 1994: 325–32. 92. Huhta JC, Cohen AW, Wood DC. Premature constriction of the ductus arteriosus. J Am Soc Echocardiogr 1990; 3: 30–4. 93. Brezinka C, Huisman TWA, Stijnen T, Wladimiroff JW. Normal flow velocity waveforms in the fetal ductus arteriosus in the first half of pregancy. Ultrasound Obstet Gynecol 1992; 2: 397–401. 94. Tulzer G, Gudmundsson S, Sharkey AM et al. Doppler echocardiography of fetal ductus arteriosus constriction versus increased right ventricular output. J Am Coll Cardiol 1991; 18: 532–6. 95. Reed KL, Sahn DJ, Scagnelli S, Anderson CF, Shenker L. Doppler echocardiographic studies of diastolic function in the human fetal heart: changes during gestation. J Am Coll Cardiol 1986; 8: 391–5.

170

Fetal Cardiology

96. Anderson CF, Reed KL, Shenker L. Doppler measurements of diastolic filling and systolic ejection times in the human fetus. Presented at AIUM Annual Meeting, New Orleans, 1987. 97. Wladimiroff JW, Tonge HM, Stewart PA. Doppler ultrasound assessment of cerebral blood flow in the human fetus. Br J Obstet Gynaecol 1986; 93: 471–5. 98. van den Wijngaard JA, Groenenberg IA, Wladimiroff JW, Hop WC. Cerebral Doppler ultrasound of the human fetus. Br J Obstet Gynaecol 1989; 96: 845–9. 99. Arduini D, Rizzo G, Romanini C. Changes of pulsatility index from fetal vessels preceding the onset of late decelerations in growth-retarded fetuses. Obstet Gynecol 1992; 79: 605–10. 100. Vyas S, Nicolaides KH, Bower S, Campbell S. Middle cerebral artery flow velocity waveforms in fetal hypoxaemia. Br J Obstet Gynaecol 1990; 97: 797–803. 101. Hislop A, Reid L. Intra-pulmonary arterial development during fetal life – branching pattern and structure. J Anat 1972; 113: 35–48. 102. Levin DL, Rudolph AM, Heymann MA, Phibbs RH. Morphological development of the pulmonary vascular bed in fetal lambs. Circulation 1976; 53: 144–51. 103. Rudolph AM, Heymann MA. Circulatory changes during growth in the fetal lamb. Circ Res 1970; 26: 289–99. 104. Long W. Developmental pulmonary circulatory physiology. In: Long W, ed. Fetal and Neonatal Cardiology. Philadelphia: WB Saunders, 1990: 76–96. 105. Laudy JAM, de Ridder MAJ, Wladimiroff JW. Human fetal pulmonary velocimetry. Repeatability and normal values with emphasis on middle and distal pulmonary vessels. Ultrasound Obstet Gynecol 2000; 15: 479–86. 106. Sivan E, Rotstein Z, Lipitz S, Sevillia J, Achiron R. Segmentary fetal branch pulmonary artery blood flow velocimetry: in utero Doppler study. Ultrasound Obstet Gynecol 2000; 16: 453–6. 107. Emerson DS, Cartier MS, DeVore G et al. Distal pulmonary artery branches in the fetus: new observations with color flow and pulsed Doppler. J Ultrasound Med 1991; 10: S19. 108. Laudy JA, de Ridder MA, Wladimiroff JW. Doppler velocimetry in branch pulmonary arteries of normal human fetuses during the second half of gestation. Pediatr Res 1997; 41: 897–901. 109. Rasanen J, Huhta JC, Weiner S, Wood DC, Ludomirski A. Fetal branch pulmonary arterial vascular impedance during the second half of pregnancy. Am J Obstet Gynecol 1996; 174: 1441–9. 110. Rizzo G, Capponi A, Chaoui R et al. Blood flow velocity waveforms from peripheral pulmonary arteries in normally grown and growth-retarded fetuses. Ultrasound Obstet Gynecol 1996; 8: 87–92. 111. Rasanen J, Wood DC, Weiner S, Ludomirski A, Huhta JC. Role of the pulmonary circulation in the distribution of human fetal cardiac output during the second half of pregnancy. Circulation 1996; 94: 1068–73. 112. Tonge HM, Struijk PC, Wladimiroff JW. Blood flow measurements in the fetal descending aorta: technique and clinics. Clin Cardiol 1984; 7: 323–9. 113. Mari G, Uerpairojkit B, Abuhamad AZ, Copel JA. Adrenal artery velocity waveforms in the appropriate and small-

114.

115.

116.

117.

118.

119.

120.

121.

122.

123.

124.

125.

126.

127.

128.

129.

130.

for-gestational-age fetus. Ultrasound Obstet Gynecol 1996; 8: 82–6. Mari G, Copel A. Peripheral fetal circulation. In: Copel JA, Reed K, eds. Doppler Ultrasound in Obstetrics and Gynecology. New York: Raven Press, 1994: 125–32. Abuhamad AZ, Mari G, Bogdan D, Evans AT. Doppler flow velocimetry of the splenic artery in the human fetus: is it a marker of chronic hypoxia? Am J Obstet Gynecol 1995; 172: 820–5. Rhee E, Detti L, Mari G. Superior mesenteric artery flow velocity waveforms in small for gestational age fetuses. J Matern Fetal Med 1998; 7: 120–3. Kilavuz O, Vetter K. Is the liver of the fetus the 4th preferential organ for arterial blood supply besides brain, heart, and adrenal glands? J Perinat Med 1999; 27: 103–6. Mari G. Arterial blood flow velocity waveforms of the pelvis and lower extremities in normal and growth-retarded fetuses. Am J Obstet Gynecol 1991; 165: 143–51. Vyas S, Nicolaides KH, Campbell S. Renal artery flowvelocity waveforms in normal and hypoxemic fetuses. Am J Obstet Gynecol 1989; 161: 168–72. Coppens M, Loquet P, Kollen M, De Neubourg F, Buytaert P. Longitudinal evaluation of uteroplacental and umbilical blood flow changes in normal early pregnancy. Ultrasound Obstet Gynecol 1996; 7: 114–21. den Ouden M, Cohen-Overbeek TE, Wladimiroff JW. Uterine and fetal umbilical artery flow velocity waveforms in normal first trimester pregnancies. Br J Obstet Gynaecol 1990; 97: 716–19. Wladimiroff JW, Huisman TW, Stewart PA. Fetal and umbilical flow velocity waveforms between 10–16 weeks’ gestation: a preliminary study. Obstet Gynecol 1991; 78: 812–14. Gill RW, Warren PS. Doppler measurements of umbilical blood flow. In: Sanders JC, James AE, eds. The Principles and Practice of Ultrasonography in Obstetrics and Gynecology. Norwalk: Appleton-Century-Crofts, 1985: 87–97. Brackley KJ, Ramsay MM, Pipkin FB, Rubin PC. A longitudinal study of umbilical artery Doppler waveforms in normal pregnancy: analysis using Laplace transform techniques. Br J Obstet Gynaecol 1998; 105: 78–82. Lees C, Albaiges G, Deane C, Parra M, Nicolaides KH. Assessment of umbilical arterial and venous flow using color Doppler. Ultrasound Obstet Gynecol 1999; 14: 250–5. Sutton MS, Theard MA, Bhatia SJ et al. Changes in placental blood flow in the normal human fetus with gestational age. Pediatr Res 1990; 28: 383–7. Rizzo G, Arduini D, Romanini C. Umbilical vein pulsations: a physiologic finding in early gestation. Am J Obstet Gynecol 1992; 167: 675–7. Wladimiroff JW, Huisman TW, Stewart PA, Stijnen T. Normal fetal Doppler inferior vena cava, transtricuspid, and umbilical artery flow velocity waveforms between 11 and 16 weeks’ gestation. Am J Obstet Gynecol 1992; 166: 921–4. Rizzo G, Arduini D, Romanini C. Inferior vena cava flow velocity waveforms in appropriate- and small-forgestational-age fetuses. Am J Obstet Gynecol 1992; 166: 1271–80. Huisman TW, Stewart PA, Wladimiroff JW, Stijnen T. Flow velocity waveforms in the ductus venosus, umbilical

Development of fetal cardiac and extracardiac Doppler flows

131.

132.

133.

134. 135. 136.

137.

vein and inferior vena cava in normal human fetuses at 12–15 weeks of gestation. Ultrasound Med Biol 1993; 19: 441–5. Wladimiroff JW, Huisman TW, Stewart PA. Fetal cardiac flow velocities in the late 1st trimester of pregnancy: a transvaginal Doppler study. J Am Coll Cardiol 1991; 17: 1357–9. Laudy JA, Huisman TW, de Ridder MA, Wladimiroff JW. Normal fetal pulmonary venous blood flow velocity. Ultrasound Obstet Gynecol 1995; 6: 277–81. Rajagopalan B, Friend JA, Stallard T, Lee GD. Blood flow in pulmonary veins: I. Studies in dog and man. Cardiovasc Res 1979; 13: 667–76. Rudolph AM. Fetal and neonatal pulmonary circulation. Annu Rev Physiol 1979; 41: 383–95. Hong Y, Choi J. Doppler study on pulmonary venous flow in the human fetus. Fetal Diagn Ther 1999; 14: 86–91. van Splunder IP, Wladimiroff JW. Cardiac functional changes in the human fetus in the late first and early second trimesters. Ultrasound Obstet Gynecol 1996; 7: 411–15. Thornburg KL, Morton MJ. Filling and arterial pressures as determinants of RV stroke volume in the sheep fetus. Am J Physiol 1983; 244: H656–63.

171

138. Sutton MGS, Raichlen JS, Reichek N, Huff DS. Quantitative assessment of right and left ventricular growth in the human fetal heart: a pathoanatomic study. Circulation 1984; 70: 935–41. 139. Baschat AA, Gembruch U, Reiss I, Gortner L, Diedrich K. Demonstration of fetal coronary blood flow by Doppler ultrasound in relation to arterial and venous flow velocity waveforms and perinatal outcome – the ‘heart-sparing effect’. Ultrasound Obstet Gynecol 1997; 9: 162–72. 140. Baschat AA, Gembruch U, Gortner L et al. Coronary artery blood flow visualization signifies hemodynamic deterioration in growth-restricted fetuses. Ultrasound Obstet Gynecol 2000; 16: 425–31. 141. Dubiel M, Breborowicz GH, Marsal K, Gudmundsson S. Fetal adrenal and middle cerebral artery Doppler velocimetry in high-risk pregnancy. Ultrasound Obstet Gynecol 2000; 16: 413–18. 142. Wladimiroff JW, Tonge HM, Stewart PA. Doppler ultrasound assessment of cerebral blood flow in the human fetus. Br J Obstet Gynaecol 1986; 93: 471–5. 143. Teitel DF, Iwamoto HS, Rudolph AM. Effects of birthrelated events on central blood flow patterns. Pediatr Res 1987; 22: 557–66.

12 The examination of the normal fetal heart using two-dimensional echocardiography Rabih Chaoui

Ultrasound techniques used in fetal cardiology are numerous. They include different approaches such as two-dimensional examination by the transabdominal or transvaginal route (in early pregnancy), the M-mode examination, and the different methods based on Doppler (e.g. pulsed Doppler, color Doppler, power Doppler, tissue Doppler). However, the accuracy of screening for heart defects in the fetus has not increased in pace with this technical development. The basis of every fetal echocardiographic examination is still the precise assessment of heart structures using real-time ultrasound. In recent years a huge improvement has been observed in image resolution, not only at centers with high-level ultrasound machines but also at primary and secondary institutions with equipment available for performing antenatal screening. Therefore, a detailed evaluation of the different cardiac structures is no longer the monopoly of a few centers but is available at the screening level. In this chapter, the basics of a complete cardiac examination using different planes in two-dimensional ultrasound are reviewed. Different approaches for such an examination were proposed in the past, either slightly modifying the pediatric cardiologic planes or creating ‘new’ fetal planes with more flexibility.1–4 The advantage of the latter is that they are easier to learn in obstetric ultrasound. This chapter concentrates on planes we proposed some years ago3 (Figure 12.1).

Examination of the upper abdomen The assessment of the upper abdomen is an integral part of the fetal echocardiographic examination (Table 12.1). Heart malformations associated with abnormalities of the situs are known to be more severe and complex.

Furthermore, a careful analysis of the upper abdomen provides a better orientation when examining the fetal heart. The examiner begins with the assessment of the fetal position in utero, in order to distinguish the right and left sides of the fetus. The upper abdomen should then be visualized in a cross-sectional plane (Figure 12.2). A fictitious anterior–posterior line is drawn, dividing a left and a right side. On the left side is the stomach, the descending aorta, and the spleen, which is located between the stomach and diaphragm. On the right side, the liver and the inferior vena cava are found, the inferior vena cava lying anterior to the aorta. By tilting the transducer slightly, the confluence of the three liver veins toward the inferior vena cava are seen. The umbilical vein enters the liver and continues to the right side into the portal sinus. The transducer is then moved slightly cranially to visualize the next plane, i.e. the four-chamber view (Figure 12.3 and Video clip 12.1). During this movement the connection of the inferior vena cava with the right atrium is checked (venoatrial connection).

Four-chamber view The four-chamber view is the most important plane in the examination of the fetal heart. The main cardiac structures – the position, the size, the rhythm, and the contractility of the fetal heart – can all be analyzed (Table 12.2). In this slightly oblique plane, the simultaneous visualization of both atria, ventricles, atrioventricular valves, and interventricular and interatrial septa is achieved (Figure 12.4). This plane is important since it has an accuracy in detecting a wide range of fetal heart defects, and can be easily learned. Owing to the fluid-filled lungs in prenatal life, as well as to the late calcification of the ribs, different insonation views of the four-chamber plane can be obtained, reducing the disadvantage of the different fetal positions on fetal examination (Figure 12.5). The correct

174

Fetal Cardiology

LV LA

VCS WS

RV RV

Ao DA RA

Aps

Apd

1

Ao

3

RA

2

2

PV LV Ao

RV

1 VCI

3 TP

(a)

Figure 12.1

(b)

(a) The different cross-sections from the four-chamber view to the five-chamber and pulmonary views (compare with Figures 12.2, 12.3, and 12.9). (b) A cross-section of the upper thorax to obtain the three-vessel view (compare with Figures 12.13 and 12.14). From reference 3. Ao, aorta; Aoa, ascending aorta; Aod, descending aorta; Apd, right pulmonary artery; Aps, left pulmonary artery; DA, ductus arteriosus; LV, left ventricle; PV, pulmonary vein; RA, right atrium; RV, right ventricle; TP, pulmonary trunk; VCI, inferior vena cava; VCS, superior vena cava; WS, spine.

5

TP

VCS WS

Aod

Aoa

DA VCS

Acd

Acd 4

TP

4

Ao TP

PV

DA

RA

Ao

LV RV

VCI

Table 12.1 plane

Checklist for the upper abdominal

• • • •

Filled stomach is on the left side Aorta is on the left side of the spine Liver is on the right side Inferior vena cava is on the right side of the spine, ventral and lateral to the aorta • Inferior vena cava receives the hepatic veins and is connected to the right atrium

plane should include both patent atrioventricular valves connecting the atria with the corresponding ventricles (Figure 12.6). Once the four-chamber plane is visualized, an imaginary anteroposterior line is drawn, dividing the thorax into two equal left and right sides. In normal levocardia, one-third of the heart is on the right side and two-thirds are on the left, with the heart axis pointing to the left. In recent years, the assessment of the cardiac axis was

VCS 5

added as a new parameter in the analysis of heart position.5 Compared to the sagittal axis, the cardiac axis is at 45 ± 15° and is abnormal in many heart defects, especially those involving the great vessels. Analysis of the heart size is important in order to distinguish between cardiomegaly, generally due to atrioventricular valve insufficiency associated with right atrium dilatation, and the normal heart in a small thorax in growth-restricted fetuses. In unclear cases, the examiner should use cardiac measurements derived in the fourchamber view, measuring the heart length, width, area, and cardiothoracic ratio.6 The analysis of heart contractility enables the detection of hypokinesia of the myocardium. An abnormal heart rhythm is easily detected with realtime sonography, but its classification is effectively performed using M-mode. After the general examination of the heart, one would continue to inspect the heart structures. After localization of the heart’s left and right sides, as well as the atria and ventricles, the cavities are then compared to each other. Important landmarks to remember are that the lumen of

Examination of normal fetal heart using two-dimensional echocardiography

175

Table 12.2 Checklist for the four-chamber view plane • • • • • • • • •

Position of the heart in the thorax Cardiac axis Size of the heart Rhythm Contractility Size of the left and right atria Size of the left and right ventricles Size relationship of the left and right sides Position and function of the tricuspid and mitral valves • Continuity of the interventricular septum • Position and form of the interatrial septum and the valve of the foramen ovale • Connections of the pulmonary veins to the left atrium

Figure 12.2 Cross-section of the upper abdomen. The line divides the left and right sides. Visualized are the stomach (ST) and the descending aorta (AO) on the left (L) and the inferior vena cava (VCI) on the right.

Figure 12.3 The approach to obtain the different planes from the abdominal plane to the four- and five-chamber views toward the pulmonary view. See also corresponding Video clip 12.1.

the right ventricle is slightly smaller than the left, and the foramen ovale flap bulges into the left atrium. Anterolaterally to the spine, the descending aorta is recognized as a circular, pulsating structure. Just anteriorly and close to the aorta, the esophagus can often be recognized as an echogenic circular structure (Figure 12.4). During swallowing, the esophagus dilates and mimics a second vessel in front of the aorta, but then disappears after swallowing is ended. The first cardiac structure

ventrally adjacent to the aorta and esophagus is the left atrium. The left atrium is thus the cardiac structure situated most posteriorly in the chest and is recognized by the connections of the pulmonary veins and the leaflet of the foramen ovale. The foramen ovale ‘flap’ is the free part of the septum primum, which closes during embryological development of the septum primum. Owing to the rightto-left shunt at the atrial level, the flap bulges into the

176

Fetal Cardiology

Figure 12.4 Cross-section of the thorax visualizing the apical four-chamber view. In front of the spine the descending aorta (Ao) is seen and in front of it the esophagus (Es, arrow). In this plane, the following structures can be seen: right and left ventricles (RV, LV), right and left atria (RA, LA), interventricular and interatrial septa (IVS, IAS), foramen ovale, mitral (MV) and tricuspid valves (TV).

left atrium, showing a wide variation in its size and shape (Figure 12.6). This structure is semilunar, and is best seen using a left-sided approach to the heart.7 In hypoplastic left heart syndrome, paradoxical movements of this flap can be seen to the right side. The right atrium is on the right side of the left atrium and communicates with the latter via the foramen ovale (i.e. the ostium secundum). By slightly angulating the transducer cranially and/or caudally, or by tilting the transducer into a longitudinal plane, the connections of the inferior and superior venae cavae can be identified (Figure 12.7). Both atria are nearly equal in size, and are best recognized by the vein connections. Another feature is visualization of the appendages: the left atrial appendage is finger-like and has a narrow base, whereas the right atrial appendage is pyramidal in shape with a broad base. The appendages can be visualized in a plane slightly cranial to the four-chamber view, but are not identified reliably under many conditions. Directly behind the sternum, the right ventricle appears as the most anterior cardiac structure. The left ventricle is adjacent and posterior to the right ventricle, and is the most left-sided cardiac structure. Many features can be used to differentiate the right from the left ventricle. The right ventricle is trabeculated and the cavity is irregular, whereas the inner shape of the left ventricle is smooth. The lumen of the left ventricle is longer than that of the right ventricle, and reaches the apex of the heart. The right

Figure 12.5 Four-chamber view seen from the left side (left) and from the right (right). The advantage of the fetal examination is that the four-chamber view can be visualized in different fetal positions. In the lateral approaches the septum can be better estimated.

Examination of normal fetal heart using two-dimensional echocardiography

177

Figure 12.6 Apical four-chamber view in systole (left) and diastole (right). Using a cine-loop, the visualization of different phases of the heart cycle is easier. During diastole the opened valves are well recognized, as well as the bulging of the flap of the foramen ovale (FO) (spetum primum). See also corresponding Video clip 12.2.

Figure 12.7 Parasagittal right-sided longitudinal view demonstrating the connection of the superior (VCS) and inferior vena cava (VCI) to the right atrium (RA).

ventricle shows a short lumen, mainly owing to the moderator band (septomarginal trabeculum coursing from the interventricular septum to the lower free wall of the right ventricle). The ventricles can also be recognized by their corresponding atrioventricular valves: the left ventricle receives the mitral valve and the right ventricle the tricuspid valve. The tricuspid valve inserts slightly more apically than the mitral valve on the interventricular septum. Both ventricles are separated by the interventricular septum. The septum begins as a wide thickened structure at the apex of the heart, and becomes thinner as it reaches the level of the atrioventricular valves. This is due to the development and anatomic structure of the septum, with a muscular part in the lower two-thirds, and a membranous part at the junction with the atrioventricular and semilunar valves. Around 20 weeks of gestation, this thin membranous part is not correctly visualized by an apical approach. This drop-out effect sometimes leads to a falsepositive suspicion of septal defects (Figure 12.8). In these conditions, the heart should be examined using a lateral view, allowing better visualization of the septum. Thickness of the septum, ranging between 2 and 4 mm during gestation, should also be measured by the lateral approach.

178

Fetal Cardiology

Figure 12.8 Apical four chamber in systole (left) and diastole (right). This “drop-out” effect is due to the insonation angle parallel to the membranous septum. By angulating the transducer more laterally (right), the intact septum is better recognized.

The pericardium of the heart is recognized as a slight double layer around the outer cardiac wall. At the level of the atrioventricular valves, a tiny amount of pericardial fluid can be seen and should not be diagnosed as an abnormal effusion.

Visualization of left and right ventricular outflow tracts In the next planes, the five-chamber view and the pulmonary view, the arising of the aorta from the left ventricle and the pulmonary trunk from the right ventricle are visualized. Assessment of the ventriculoarterial concordance is mandatory in analyzing heart anatomy. Once the four-chamber plane is visualized, the examiner tilts the transducer slightly cranially (Figures 12.3 and 12.9) and focuses attention on the center of the heart in the left ventricle, where the mitral valve connects with the ventricular septum. The aorta arises as a vessel continuing the ventricular septum but pointing slightly to the right side. The other border of the aortic wall shows a close connection to the mitral valve. Within the aortic root, the aortic valve can be recognized as an echogenic dot. In this plane the examiner checks (Table 12.3) the continuity of the septum and aorta, the angulation of

Table 12.3 Checklist for outflow tract assessment • Normal connection of the aorta to the left ventricle and pulmonary trunk to the right ventricle • Both vessels cross over each other • Compare caliber of pulmonary trunk (PT) and aorta (PT > aorta) • Assess opening excursion of aortic and pulmonary valves • Continuity of ventricular septum to the aortic root • Normal course and caliber of great vessels and superior vena cava in the upper thorax • Assess aortic isthmus and ductus arteriosus • Rule out atypical vessels (e.g. left persisting superior vena cava)

the aorta and septum, the size of the aortic root, and the ascending aorta, as well as the opening movements of the aortic valve (most ventricular septal defects can be detected in this plane). The pulmonary trunk can be visualized by further tilting of the transducer cranially (Figure 12.9), but some authors recommend the short-axis view (Figure 12.10). This plane is easy to obtain by non-cardiologists and can be obtained by successive tilting from the four- and five-chamber planes (Figure 12.3). Once the five-chamber

Examination of normal fetal heart using two-dimensional echocardiography

179

Figure 12.9 “Five-chamber view” (left) with the aorta (Ao) arising from the left ventricle (LV) and “pulmonary view” (right) with the pulmonary artery arising from the right ventricle (RV). Compare with Figure 12.1. TP, pulmonary trunk; VCS, superior vena cava.

Figure 12.10 Short-axis view visualizing the aorta (Ao) in the center and around it the right atrium (RA), the right ventricle (RV), and the pulmonary trunk (TP) with its bifurcation into the right (APD) and left (APS) pulmonary arteries.

view is obtained, the examiner focuses attention during further tilting of the transducer on the connection of the right ventricle with the descending aorta (vessel toward the spine). The vessel then arising is the pulmonary trunk, continuing as the ductus arteriosus (Figure 12.9). The pulmonary trunk crosses perpendicularly over the ascending aorta and then becomes the vessel on the left. On its right side, two vessels in cross-section can be recognized: the ascending aorta and the superior vena cava. During the tilting movement from the five-chamber view, the examiner checks for the correct connection of the right ventricle and pulmonary trunk, as well as the crossing of the pulmonary trunk (absent in transposition of the great arteries, where both vessels show a parallel course). The size of the pulmonary trunk has a slightly larger caliber compared to that of the aorta,8 and the valve is seen as a white dot showing opening and closing movements. In some heart defects involving the great arteries, it is important to identify correctly which vessel is involved. In these conditions, the aorta and pulmonary trunk are differentiated by the arising of the stem vessels for the aorta, and the bifurcation of the right and left pulmonary arteries for the pulmonary trunk. Whereas the three arterial branches are seen by visualizing the aortic arch, the pulmonary arteries are best visualized by obtaining a short-axis view of the heart. This is achieved by visualizing the five-chamber view, and then rotating the transducer

180

Fetal Cardiology

Ascending Ao 1 2

rpa

3

Isthmus 5

Descending Ao

Figure 12.11 Longitudinal view of the aortic arch, resembling a candy cane. One can easily recognize the three vessels to the head and upper limbs. Under the aortic arch the right pulmonary artery is seen in cross-section. See also corresponding DVD clip 12.3.

Figure 12.12 in order to obtain a plane from the right hip to the left shoulder, demonstrating the ‘circle and sausage’ sign, with the aorta in the center, and the right atrium, right ventricle, pulmonary trunk, and bifurcation around it (Figure 12.10). The right pulmonary artery is seen very easily in its course under the aortic root toward the right lung.

Longitudinal views of the outflow tracts The longitudinal planes of the outflow tracts are visualized to assess the aortic and the ductus arteriosus arches. In these planes, the continuity and form of the arches are seen, as well as the brachiocephalic vessels arising from the aorta to the head and upper extremity. The examiner obtains a parasagittal plane slightly to the left, including the aortic valve and the descending aorta, and can thus visualize the aortic arch (Figure 12.11 and Video clip 12.3). In this plane the aortic arch appears to emerge from the center of the heart and shows a circular shape (‘candy cane’). Under the ascending aorta, a cross-section of the right pulmonary artery can be recognized. In the next adjacent plane to the left, the longitudinal view of the pulmonary trunk with the ductus arteriosus can be recognized (Figure 12.12). The right ventricle and the pulmonary valve are seen anteriorly, and the ductus arteriosus arch courses perpendicularly to connect with the descending aorta, recognized as having a more angular shape (‘hockey stick’). The examiner can be confronted with two problems while obtaining both of these planes.

Longitudinal view of the ductus arteriosus arch. The pulmonary artery (TP) arises from the anteriorly positioned right ventricle (RV) and courses toward the descending aorta. This arch has a nearly perpendicular shape and resembles a hockey stick.

The first is that these planes can be obtained in only a very few fetal positions: dorsoanterior or dorsoposterior. The second is that both planes are very close to one another, and inexperienced examiners can easily be confused, especially in abnormal cases. Therefore, in teaching fetal echocardiography to the non-cardiologist, we proposed an easier plane3 (Figure 12.1b) which was later called the ‘three-vessel view’, visualizing the above structures in a tangential cross-sectional approach.9

Three-vessel view From the four-chamber plane, the transducer is moved parallel in the direction of the upper thorax. In this sagittal cross-section of the upper thorax, the three vessels of the pulmonary trunk – the ductus arteriosus, aortic arch with aortic isthmus, and superior vena cava – can be seen (Figures 12.13 and 12.14; Video clip 12.1). This plane is therefore called the three-vessel view.3,9 The aortic and the ductus arches are seen in a tangential cross-section and build a V-form pointing to the posterior thorax on the left side of the spine (Figure 12.14). The trachea prior to its bifurcation is recognized as a circular structure with an echogenic wall adjacent to and on the right side of the aortic isthmus and anterior to the spine. In front of the trachea and on the right side of the aortic arch, the superior vena cava is recognized. On the left and right sides of

Examination of normal fetal heart using two-dimensional echocardiography

Figure 12.13 Cross-section of the upper thorax demonstrating the three vessels: the superior vena cava (VCS), the aorta (AO) and the pulmonary trunk (TP). In front of these vessels the thymus (THYM) is clearly recognized as an echodense structure compared to the neighboring lungs (L).

181

these vessels, lung tissue can be recognized (Figure 12.13). In front of these three vessels, the examiner can also recognize a structure with a less echogenic appearance, which is the thymus (Figure 12.13). The assessment of the threevessel view has been stressed in the past few years because, with this view, many outflow tract disorders can easily be detected.9 Furthermore, the position of the trachea in comparison to these vessels can be used as a landmark to distinguish the correct left-sided aortic arch from an abnormal right-sided aortic arch. It was then proposed to call this plane the three-vessel–trachea view.10 Visualization of the thymus in front of these vessels can also be proposed as a hint to detect defects associated with malformations of the CATCH-22 group (cardiac defects, abnormal facies, thymic hypoplasia, cleft palate, hypocalcemia, resulting from 22q11 delections).11 The advantage of this plane is that it allows the visualization of both arches in most positions of the fetus, allowing easier detection of possible abnormalities. The slightly larger size of the pulmonary trunk compared to the size of the aorta can be checked, as can the continuity of the aortic arch, or the presence of a fourth vessel on the left of the pulmonary trunk, as a left persisting superior vena cava. Using color Doppler, the visualization of antegrade flow within both outflow tracts under normal conditions can be easily distinguished from the retrograde flow in one outflow tract when severe outflow

Th

ym

TP

AO

us

L

VCS

DA

R

Trachea

Isthmus Sp

Figure 12.14 Three-vessel view in the upper thorax visualizing the aorta (AO) and isthmus, the pulmonary trunk (TP), ductus arteriosus (DA) and the superior vena cava (VCS). The trachea is recognized as a circle with an echogenic wall on the right side of the two great vessels and behind the vena cava.

182

Fetal Cardiology

obstruction on this side is present. It is expected that, in the near future, this plane will be added to the fourchamber view in the routine assessment of the fetal heart.

Recommendations and guidelines for two-dimensional examination of the fetal heart In recent years several local, regional, and international societies have published recommendations on how to examine the fetal heart during the screening ultrasound. Several authors represented in this book have recently contributed to publishing guidelines of the International Society of Obstetrics and Gynecology12 for performing the ‘basic’ and ‘extended basic’ cardiac examination. The basic cardiac screening examination relies on the four-chamber view of the fetal heart, similarly to the details explained in this chapter. The ‘extended basic’ cardiac examination includes, in addition to the four-chamber view plane, the visualization of both the left and right ventricular outflow tracts, when feasible. This increases the accuracy in

detecting major cardiac anomalies above those detectable in the four-chamber view plane.

Three- and four-dimensional ultrasound and cardiac planes Over the last few years, three-dimensional (3D) and 4D facilities have been used in fetal echocardiography, especially with the advent of the exciting acquisition technique related to a beating fetal heart, called spatiotemporal image correlation (STIC).13,14 Once a volume data set is successfully obtained it can be visualized as three orthogonal planes (Figure 12.15), as a single plane (Figure 12.16), as tomographic parallel slices (Figure 12.17), or rendered as a 3D volume (Figure 12.18).15. From a 3D/4D data set, cross-sectional views can be obtained at any desired orientation, direction, and depth. Each plane of interest (‘anyplane’) can be demonstrated by scrolling through the volume as the four-chamber (Figure 12.16), the five-chamber, the short-and long-axis views, and the three vessel–trachea view, the aortic and ductal arches, etc. These can be visualized offline both in cine loop and as still images.

Figure 12.15 Four-dimensional data set as a STIC volume. Spatial and temporal image correlation (STIC) allows the acquisition of a cardiac volume with one cardiac cycle. The examiner can thus demonstrate the data set as three orthogonal planes and chooses in the cine loop the moment of interest within the cardiac cycle.

Examination of normal fetal heart using two-dimensional echocardiography

183

Figure 12.16 From the volume data set (Figure 12.15) the examiner can display one single plane of interest and the time of interest within the cardiac cycle, here the four-chamber view in systole.

Figure 12.17 From the volume data set (Figure 12.15) the examiner can demonstrate the heart with multiple parallel slices using the tomography imaging tool. The examiner can modify the interslice distance (upper left image) to get the information of interest. In this example, the volume shows all parallel planes from the upper abdomen toward the three vessels.

184

Fetal Cardiology

References

Figure 12.18 The volume data set allows furthermore spatial demonstration of the heart, here showing a spatial view of the cardiac cavities in the four-chamber plane.

It can be imagined that in the future the planes presented in this chapter will be obtained either directly during the examination or by acquiring a volume data set and demonstrating offline the different planes. The storage of digitalized volume data instead of single images, and the usefulness of this technique in getting a second opinion or a later offline analysis, are a few of the potentials of this new promising technique. Preliminary studies reported the possibility of automated examination16 of the heart starting from such a volume data set after standardizing the positions of the planes.

Legends for the DVD Video clip 12.1 A parallel sweep from the upper abdomen to the upper thorax showing the different planes as the four-chamber, five-chamber, and the three vessel view.

Video clip 12.2 Transverse view of the thorax demonstrating an apical four-chamber plane. Both atria and ventricles are connected by the mitral and tricuspid valve, showing normal excursion.

Video clip 12.3 A sagittal view demonstrating the aortic arch with the arising three brachiocephalic vessels.

1. Allan LD. Manual of Fetal Echocardiography. Lancaster: MTP Press, 1986. 2. Abuhamad A. A Practical Guide to Fetal Echocardiography. Philadelphia: Lippincott-Raven, 1997. 3. Chaoui R, Bollmann R, Hoffmann H, Heling KS. [Sonoanatomy of the fetal heart. Proposal of simple crosssectional planes for the non-cardiologists]. Ultraschall Klin Prax 1991; 6: 59–67. [in German] 4. DeVore GR. The prenatal diagnosis of congenital heart disease—a practical approach for the fetal sonographer. J Clin Ultrasound 1985; 13: 229–45. 5. Smith RS, Comstock CH, Kirk JS, Lee W. Ultrasonographic left cardiac axis deviation: a marker for fetal anomalies. Obstet Gynecol 1995; 85: 187–91. 6. Chaoui R, Heling KS, Bollmann R. [Ultrasound measurements of the fetal heart in the 4-chamber image plane]. Geburtshilfe Frauenheilkd 1994; 54: 92–7. [in German] 7. Kachalia P, Bowie JD, Adams DB, Carroll BA. In utero sonographic appearance of the atrial septum primum and septum secundum. J Ultrasound Med 1991; 10: 423–6. 8. Chaoui R, Heling KS, Bollmann R. [Ultrasound measurements of the diameter of the aorta and pulmonary trunk of the fetus]. Gynakol Geburtshilfliche Rundsch 1994; 34: 145–51. [in German] 9. Yoo SJ, Lee YH, Kim ES et al. Abnormal three-vessel view on sonography: a clue to the diagnosis of congenital heart disease in the fetus. Ultrasound Obstet Gynecol 1997; 9: 173–82. 10. Yagel S, Arbel R, Anteby EY, Raveh D, Achiron R. The three vessels and trachea view (3VT) in fetal cardiac scanning. Ultrasound Obstet Gynecol 2002; 20: 340–5. 11. Chaoui R, Kalache KD, Heling KS et al. Absent or hypoplastic thymus on ultrasound: a marker for deletion 22q11.2 in fetal cardiac defects. Ultrasound Obstet Gynecol 2002; 20: 546–52. 12. International Society of Ultrasound in Obstetrics and Gynecology. Cardiac screening examination of the fetus: guidelines for performing the ‘basic’ and ‘extended basic’ cardiac scan. Ultrasound Obstet Gynecol 2006; 27: 107–13. 13. Devore GR, Falkensammer P, Sklansky MS, Platt LD. Spatio-temporal image correlation (STIC): new technology for evaluation of the fetal heart. Ultrasound Obstet Gynecol 2003; 22: 380–7. 14. Chaoui R, Hoffmann J, Heling KS. Three-dimensional (3D) and 4D color Doppler fetal echocardiography using spatiotemporal image correlation (STIC). Ultrasound Obstet Gynecol 2004; 23: 535–45. 15. Chaoui R, Heling KS. New developments in fetal heart scanning: three- and four-dimensional fetal echocardiography. Semin Fetal Neonatal Med 2005; 10: 567–77. 16. Abuhamad A. Automated multiplanar imaging: a novel approach to ultrasonography. J Ultrasound Med 2004; 23: 573–6.

13 First and early second trimester fetal heart screening Simcha Yagel, Sarah M Cohen, Baruch Messing, and Reuwen Achiron

Introduction Congenital heart disease (CHD) is the most common major congenital malformation, affecting approximately 6:1000 live births.1 The epidemiology of CHD is discussed in Chapter 8. It should always be borne in mind that the majority of cases of CHD do not cluster in families or populations, but rather occur in low-risk patients. In addition, a fetus diagnosed with CHD is at increased risk of chromosomal abnormalities.2–6 Major chromosomal abnormalities are found in spontaneously aborted fetuses (57%),3 midtrimester fetuses (18%), and live births affected by significant CHD (12%).4,5 When extracardiac malformations are present, this rate can be as high as 66%.6 The overall risk for aneuploidy in a fetus with CHD is estimated to be 30%.2–5 Therefore, the prenatal diagnosis of CHD is often considered as an indication for fetal karyotyping. While CHD is one of the most clinically significant malformations amenable to prenatal ultrasonographic diagnosis, at the same time it is also one of the lesions most commonly missed during fetal scans. Targeted scanning for fetal heart anomalies was usually scheduled at 20–22 weeks’ gestation and performed by the conventional transabdominal approach in specialized referral centers. Cases of heart anomalies detected before 15–16 weeks’ gestation by transabdominal sonography (TAS) have been reported since the 1980s.7 The increasing acceptance of transvaginal sonography (TVS) and the use of high-frequency (4–9 MHz), high-resolution transvaginal as well as new transabdominal (TAS) probes, along with substantial improvements in magnification imaging and signal processing, have dramatically increased our ability to visualize and examine the developing fetal heart. The characteristic changes in structural anatomy, once the

province of the embryologist and pathologist, can now be studied in great detail. The terms ‘sonoembryology’8 and ‘embryography’9 have been coined and used in this context. Indeed, the point has been reached where embryonic development, rather than technical obstacles, is the limiting factor in early imaging and detection of structural anomalies. Thus, the systematic investigation of detailed normal fetal cardiac anatomy during the first and early second trimesters of pregnancy has now shifted prenatal diagnosis of cardiac defects into that period.

Benefits of early fetal echocardiography Certain advantages are offered by the earlier diagnosis of CHD: (1) early confirmation of normal cardiac anatomy may help to relieve the anxiety of high-risk patients, or facilitate earlier and safer termination of pregnancy in cases where severe anomalies are detected; (2) there is sufficient time for fetal karyotyping and genetic counseling for parents with affected fetuses; and (3) in selected cases there is a possibility of pharmacological therapy with subsequent improvement in fetal condition. Finally, like all screening programs, early detection of CHD allows for the correct timing and place of delivery, and appropriate neonatal care facilities may be planned and arranged in advance. Intrauterine, rather than postnatal, transfer ensures the neonate an optimum condition for surgery, and may reduce early morbidity and mortality.10–12 Early screening must be repeated at mid-gestation to rule out developmental CHD.

186

Fetal Cardiology

Normal early anatomy and its visualization Organogenesis occurs over a relatively short period: by the end of the first trimester organogenesis is complete and all the major organ systems can be imaged by ultrasound. By 56 days post-fertilization the fetal heart is formed and a four-chamber structure established.13 Description, knowledge, and understanding of the anatomic landmarks of fetal cardiovascular development, with recognition of our ability to adequately visualize normal cardiac anatomy and appreciation of our limitations, are essential before the assessment of anomalous development can be attempted early in the course of pregnancy. Without this knowledge, abnormalities may be either missed or suspected where they do not exist. Several investigators have described normal early fetal heart anatomy.14–24 The cardiovascular system begins to mature during the third developmental week, and the first beats of the embryonic heart, the sine qua non of fetal viability, can be clearly discerned from the first days of the 6th gestational week, when the crown–rump length (CRL) is about 3 mm. The heart gradually attains a tubular structure that resembles a triocular cavity at the end of the 10th gestational week.13 Formation of the septae and arterial and venous connections are completed only after 8 weeks’ gestation. The aorta can be seen at the end of the 9th week, and by the end of the 12th week the brachiocephalic and carotid arteries are readily seen in every fetus.13 The defined cardiac structures visible earliest in the first trimester are probably the mitral and tricuspid valves. It has been shown that each atrioventricular valve can be individually visualized by 10 weeks’ gestation, and their normal offset position19 (with the tricuspid valve more apical than the mitral) as well as the three vessels and trachea, ventricular outflow tracts, aortic arch, and the ductus arteriosus.24 The superior and inferior venae cavae can be reliably visualized from 11 to 15 weeks’ gestation and the pulmonary veins by 12–14 weeks’ gestation.24 While complete echocardiography including all scanning parameters cannot be reliably performed at 10 gestational weeks, it is possible in 90% of cases by 12–14 weeks and in 100% by 15 weeks.24 Our experience with early cardiac visualization has been favorable: full cardiac anatomy is discernible in 95% of cases at 11–12 weeks, and in 100% at 13–15 weeks’ gestation with favorable fetal lie. Shapiro et al17 performed transvaginal measurements of the fetal heart from 14 weeks’ gestation to term. Linear correlation was found between gestational age and measurements of the right and left ventricles. The ratio between the fetal heart and the transverse diameter of the chest was almost constant at that period. The size of the embryonic heart has also been correlated with that of the CRL and abdominal diameter.17

Since the 1990s many groups have examined the feasibility of early transvaginal fetal heart scanning, whether by the transvaginal or transabdominal approach, or both, or comparing the two.24–31 Two factors critically influence the accuracy of early fetal cardiac imaging: the machine and the operator. In our ultrasound unit we use 4–9-MHz high-frequency probes, which provide a higher resolution than the 5-MHz probes that are widely used. The advantage of these highfrequency transducers is the ability to provide a clearer image because of greater axial and lateral resolution,14,15 making small structures more distinguishable. With regard to the operator, the technique of TVS requires a substantial amount of experience. Certain views may be limited by the fixed linear axis of the ultrasonographic probe during the transvaginal examination, and fetal position often dictates the views that can be imaged. Consequently, certain maneuvers and manipulations are frequently needed in order to obtain the best planes and views in a minimal examination time. It is primarily the obstetric and gynecologic sonographer who is well acquainted with the TVS technique and can perform a systematic evaluation of the heart and all other fetal organs. In our experience, the average duration of fetal examination is approximately 30 minutes, and in many cases the echocardiographic examination requires less than 5 minutes of the scanner’s time with a favorable fetal lie.32,33

The extended echocardiographic examination Already by the early 1980s,34 echocardiography had proved to be a reliable tool in the diagnosis of CHD, and incorporation of the four-chamber view of the fetal heart into screening made possible the detection of 60% of severe cardiac anomalies.35 The importance of a normal fourchamber view (FCV) was emphasized by Copel et al,36 who showed that if TAS revealed normal views, more than 90% of CHD could be ruled out, and that the FCV was abnormal in 96% of fetuses with structural heart defects. When patients were referred with a suspected diagnosis of FCV abnormalities, CHD was confirmed in over 80% of cases.35 Therefore, the American Institute of Ultrasound in Medicine (AIUM) and the American College of Radiology (ACR), in their antepartum obstetrical guidelines, recommended inclusion of the FCV in routine prenatal anomaly scanning programs.37 Although the FCV was advocated as the preferred screening view for detection of CHD, and its role as such is well established, we38 and others39,40 failed to reach the high sensitivities in the range of 90% reported by Copel et al.36 Prompted by the limitations discovered in the FCV approach, ventricular outflow tract views have

First and early second trimester fetal heart screening

been incorporated into screening programs since the early 1990s.38,39 The International Society of Ultrasound in Obstetrics and Gynecology (ISUOG) has published guidelines41 for the performance of fetal echocardiography. The examination should include general observation of normal cardiac situs, axis, and position, and the heart occupying one-third of the thorax, with the majority of the heart in the left chest. Four cardiac chambers, without observed pericardial effusion or hypertrophy, should be visualized. Atria approximately equal in size with the foramen ovale flap in the left atrium, and atrial septum primum present, should be observed. Ventricles of approximately equal size without cardiac wall hypertrophy, as well as the moderator band at the right ventricular apex, should be seen. The ventricular septum should be examined apex to crux and shown to be intact. Both atrioventricular valves should be observed to open and move freely, with the tricuspid valve leaflet inserting on the septum closer to the cardiac apex as compared to the mitral valve. The ‘extended basic’ fetal echocardiography examination proposed in the ISUOG guidelines includes all the above, and additionally the examination should demonstrate normal outflow tracts: two great vessels approximately equal in size crossing each other at right angles from their origins as they exit from their respective ventricular chambers.41 Whenever a cardiac abnormality is suspected or detected, further evaluation is performed in collaboration with the pediatric cardiologist and management team. We routinely schedule a second echocardiographic examination, using the transabdominal approach, at 20–22 weeks’ gestation for all our patients. To improve and simplify comprehensive fetal heart examination, we suggest a fetal echocardiographic examination based on five transverse planes. The first and most caudal plane is a transverse view of the upper abdomen; moving cephalad, the next is the traditional four-chamber view, which remains the key to the fetal echocardiographic examination. The third is the plane commonly termed the five-chamber view in which the aortic root is visualized; and the fourth transverse view reveals the bifurcation of the pulmonary arteries. Yoo et al42,43 suggested the three-vessel view of the fetal upper mediastinum, a crosssectional view of the major vessels, as a quick and accurate method to identify the great vessels and diagnose certain anomalies. However, the aortic arch and trachea were not included, the former being perhaps the most demanding aspect of fetal heart examination. Following Yoo’s study, we published44,45 the three-vessel and trachea (3VT) plane of insonation as the fifth short-axis view in fetal echocardiography, to complement the four traditional planes currently in use, while facilitating and expediting thorough fetal heart examination (Figure 13.1). The 3VT is the most cephalad transverse view, visualized on a plane crossing

187

the fetal upper mediastinum. It is obtained easily by moving the transducer cephalad and slightly oblique from the FCV. When properly executed, the 3VT reveals the main pulmonary trunk (MPA) in direct communication with the ductus arteriosus (DA). A transverse section of the aortic arch is seen to the right of the MPA and DA. Cross-sections of the superior vena cava (SVC), and posterior to it the trachea, are visualized (Figure 13.1).44,45 The clinical applicability of the 3VT has been demonstrated.45–51 In 99% of patients this view was quickly and easily obtained from the familiar FCV. It was shown to shorten the time required to examine the aortic arch, the most time-consuming aspect of fetal echocardiography.45 We consider one of the greatest advantages of our novel approach to be the ease with which the examiner can scan the fetal heart, beginning with the caudal upperabdomen view. By sliding the transducer cephalad in one continuous motion, all the pertinent views are readily visualized. In our center we demonstrated that five shortaxis views, including the 3VT, simplified and streamlined fetal cardiac examination, without compromising diagnostic effectiveness.45–51 Applying this methodology allows the examiner to obtain all the views necessary for the extended fetal cardiac examination as recommended by the ISUOG guidelines. It is based on the familiar segmental approach to cardiac evaluation.52,53 This approach, and its application to prenatal echocardiographic diagnosis of congenital heart defects, was recently reviewed.54

The fetal venous system Examination of the fetal venous system is important for comprehensive understanding of the fetal cardiovascular system. The venous system develops from three paired veins. At the end of the first trimester, several abnormal connections of the fetal venous system are recognized and are classified as follows: pathologies of the cardinal vein, umbilical veins, vitelline veins, and anomalous pulmonary venous connections (for detailed description see Chapter 29).

Diagnostic early echocardiography The first reports of diagnostic transvaginal fetal echocardiography date back to 1990. Gembruch et al55 used a 5-MHz transvaginal probe to diagnose complete atrioventricular canal defect and complete heart block at 11 weeks’ gestation, and subsequently performed an elective abortion

188

Fetal Cardiology

LT

RT

P (P)Ao

SVC

DA V

T

(D)Ao

LT

RT

MPA

IV LPA

T

SVC

RA

AO RPA

AO

DA

AO PA

LT

III

RT LV

RV RA LA AO

LT

II

RT

RV

LV

RA LA PV AO

I

LT

FO PV

RT

ST LI AO SP

Figure 13.1 The five short-axis views for optimal fetal heart screening. The color image shows the trachea, heart and great vessels, liver, and stomach with the five planes of insonation superimposed. Polygons show the angle of the transducer and are assigned to the relevant gray-scale images (LT, left; RT right). (I) The most caudal plane, showing the fetal stomach (ST), cross-section of the abdominal aorta (AO), spine (SP), and liver (LI). (II) The four-chamber view of the fetal heart, showing the right and left ventricles (RV, LV) and atria (RA, LA), foramen ovale (FO), and pulmonary veins (PV) to the right and left of the aorta (AO). (III) The ‘five-chamber’ view, showing the aortic root (AO), left and right ventricles (LV, RV) and atria (LA, RA) and a cross-section of the descending aorta (AO with arrow). (IV) The slightly more cephalad view showing the main pulmonary artery (MPA) and the bifurcation of left and right pulmonary arteries (LPA, RPA) and cross-sections of the ascending and descending aorta (AO and AO with arrow, respectively). (V) The three-vessel and trachea (3VT) plane of insonation, showing the pulmonary trunk (P), proximal aorta ((P)Ao), ductus arteriosus (DA), distal aorta ((D)Ao), superior vena cava (SVC), and the trachea (T) (reproduced with permission from reference 44).

First and early second trimester fetal heart screening

(a)

189

(b)

(c)

Figure 13.2 Congenital heart defects detected in the first and early second trimesters. (a) Atrioventricular canal defect in gray-scale, 13 gestational weeks, in the four-chamber view plane. Arrow indicates canal defect. (b) Ebstein’s anomaly diagnosed at 13 weeks’ gestation. Note the differential sizes of the four heart chambers. RA, right atrium; RV, right ventricle; LA, left atrium; LV, left ventricle. (c) Tetralogy of Fallot at 14 weeks’ gestation. Note the overriding aorta (Ao) in this ‘five-chamber view’ image.

on the basis of this information. Bronshtein et al56 used a high-frequency 6.5-MHz probe to diagnose a ventricular septal detect (VSD) associated with an overriding aorta, and an isolated VSD with pericardial effusion, both at 14 weeks’ gestation. The same group later reported a series of 10 fetuses with CHD diagnosed by the same technique at 12–16 weeks’ gestation.57 A further series of eight fetuses with cardiac abnormalities detected between 10 and 12 weeks’ gestation was reported by our group.58 A few examples of heart defects detected during the first and early second trimesters are shown in Figure 13.2. From the series and case reports that followed over the next

several years,55–60 we learned that transvaginal echocardiography provides accurate diagnosis of practically all classes and types of CHD. One of the largest series, including 12 793 patients over a 5-year period, originated in Israel and was reported by Bronshtein et al.56 Gembruch et al55 and our group58 reported experiences with transvaginal echocardiography in both lowand high-risk patients, reaching high sensitivities and specificities. Over the ensuing years since these seminal studies, early fetal echocardiography has been rigorously evaluated in large studies (Table 13.1).24–32,56,61,62

190

Fetal Cardiology

Table 13.1 Studies of early screening echocardiographic examination Author/year

Echocardiographic views employed

Gestational age at Sensitivity (%) Population prevalence examination (weeks) of CHD (%)

Number of patients

Bronshtein/1993 (mixed-risk)56

FCV + RLVOT

12–16

77

0.36

Achiron/1994 (low-risk)32

FCV + RLVOT

13–15

50

0.9

660

Kirk/199461

FCV + AoR

14+

78

0.85

5967

13–022

64–85

0.76

22 050

Comas Gabriel/2002 Segmental approach + (high-risk)30 D + col

12–17 (14.2)

79.2

14.4

330

McAuliffe/2005 (high-risk)26

FCV + RLVOT + AVV

11–16 (13.5)

70

12.5

160

Smrcek/2006 (mixed-risk)24

Five planes + D + col

11–13+6, 22

63, 87

7.6

2165

Becker/2006 (medium-risk)25

FCV + RLVOT + RLVIT + col + VW + AVd + GVd

11–13+6, 22

84.2, 92

1.2

3094

Yagel/1997

62

FCV + RLVOT + PA + Ao

12 793

CHD, congenital heart disease; FCV, four-chamber view; RLVOT, right/left ventricular outflow tracts; AoR, Aortic root; PA, pulmonary arteries; Ao, aorta; D, Doppler; col, color flow mapping; AVV, atrioventricular valve; RLVIT, right/left ventricular inflow tracts; VW, ventricular width; AVd, atrioventricular valve diameters; GVd, great vessel diameters.

The three-dimensional revolution in fetal echocardiography Three- and four-dimensional ultrasound (3D/4DUS) have revolutionized fetal echocardiography.63 The various techniques of volume acquisition and analysis are described in detail in this volume in Chapters 14 and 15. The application of 3D/4D to diagnosis and evaluation of the fetus with congenital heart disease as it applies to specific groups of lesions appears in Chapter 15. 3D/4DUS modalities are also applicable to screening fetal echocardiography examinations. Of particular note is the spatiotemporal image correlation (STIC) modality (see Chapter 14).64–66 Briefly, a wedge-shaped scanning volume is acquired from the fetal upper abdomen moving cephalad through the fetal thorax by tilting the transducer through approximately 15–25°. The five planes of fetal echocardiography are contained within this volume, and are acquired in a single sweep. Acquisition speed varies from 7.5 to 15 seconds, and should preferably be performed with the fetus in a quiet state. Slower scanning speed will provide higher resolution, but the fetus will have more opportunity to move or breathe. This will introduce artifacts that compromise scan quality. Therefore, younger fetal age shortens the examination time, making STIC an excellent tool for

early fetal echocardiography (see Chapters 14 and 15 for details). Once acquired, the volume contains the complete cardiac cycle (with or without color Doppler) and is available for post-processing analysis. Within a well-executed volume, all the planes necessary for complete fetal cardiac scanning are available for evaluation.44,45 Planes not accessible to direct scanning with conventional 2DUS can now be obtained and evaluated, such as the en face view of the interventricular septum and the coronal atrioventricular valves plane, showing the ‘surgical plane’ through the level of the heart valves (see Video clip 13.1 of early STIC examination.).67,68 Other examples of 3D/4DUS modalities applicable to fetal echocardiography screening examinations are the acquisition tools B-flow, 3D power Doppler (3DPD), and high definition power Doppler (Figure 13.3), and the post-processing tools tomographic ultrasound imaging (TUI) and inversion mode (IM). Various post-processing modalities and their application to fetal echocardiography have also been investigated67–73 and are all more fully described in Chapter 15. Another advantage of 3D/4D modalities is the digital archiving and sharing capabilities designed into these systems. Once the volume is acquired, it is stored to the system database and can be transferred by internet link to any connected computer for analysis. This opens unlimited possibilities for onsite and offsite multidisciplinary

First and early second trimester fetal heart screening

(a)

(b)

191

(c)

Figure 13.3 (a) Right aortic arch defect diagnosed at 15 weeks’ gestation, four-dimensional ultrasound (4DUS) B-flow image. Arrow indicates Kommerell’s diverticulum. LSVC, left superior vena cava. (b) The same patient in posterior view. Branching of the vessels from the right aortic arch. RSCA, aberrant right subclavian artery; RCA, right carotid artery; LCA, left carotid artery; RA, right aortic arch. (c) The 3VT plane of insonation, 4D high-definition color flow Doppler image. MPA, main pulmonary artery; SVC, superior vena cava; DA, ductus arteriosus.

consultations, quality review of screening programs, and teaching. Perhaps most important for screening echocardiography, this allows outlying or poorly served areas to be reached more effectively, since volumes acquired ‘in the field’ by local practitioners can be analyzed offsite by fetal cardiology specialists.74,75

Embryonic and fetal heart rate Quantitative evaluation of embryonic and fetal heart rate (HR) in cases of suspected CHD is both feasible and important. Extreme deviations of the HR from the norm convey a poor prognosis. Abnormal rates may aid not only in raising suspicion of CHD, but also in predicting the fetuses possibly destined to undergo spontaneous abortion.76 Nomograms for embryonic HR are helpful in the diagnosis of bradycardia or tachycardia in the first trimester, which signal the need for a detailed anatomic evaluation.76–79 Significant fetal tachycardia (above the upper 95% confidence interval) often induces cardiac decompensation, manifested by hydrothorax and ascites (Figure 13.4). In addition, the diagnosis of fetal tachycardia or arrhythmia early in pregnancy provides the opportunity for medical treatment for patients who desire to continue the pregnancy.76–79

Nuchal translucency and the early detection of congenital heart disease The increasing acceptance of nuchal translucency (NT) measurement screening among low-risk patients will necessarily affect fetal organ screening programs, most particularly fetal echocardiography. NT has been shown to be an effective tool not only in the identification of fetuses at high risk for chromosomal anomalies, but also those at high risk for CHD, with or without associated anomaly.80–85 NT screening will refer 3–5% of patients for comprehensive fetal echocardiography, creating a large new patient population and concomitant demand80–85 (for details see Chapter 42).

Natural course and in utero development of congenital heart disease In terms of the aforementioned, delayed or even missed diagnoses in some cardiac malformations occur despite detailed echocardiographic examination by experienced

192

Fetal Cardiology

(a)

(b)

Figure 13.4 (a) Supraventricular tachycardia (SVT) to 292 beats/min, diagnosed at 13 weeks’ gestation and associated with severe fetal hydrops. The fetus responded well to transplacental treatment of the SVT, with resolution of hydrops. (b) At presentation, signs of cardiac failure included increased scalp edema (reproduced with permission from reference 79).

Group A Group B

Cumulative % of CHD

100

90

80

70

60 TVS at 13–16 weeks

TAS at 20–22 weeks

Third trimester

Postnatal

Figure 13.5 Cumulative percentage of congenital heart defects (CHD) identified at transvaginal (TVS), transabdominal (TAS), and third trimester scans and postnatally (reproduced with permission from reference 62).

operators. Evidence has accrued that progression of cardiac disease may occur and may be observed in utero with advancing gestational age.31,62,86–95 Thus, some CHD may develop in utero, leading to a varying appearance over time. Therefore, a normal echocardiographic appearance of the heart at any gestational age does not always mean that subsequent development can be assumed to be normal, and cannot completely rule out subsequent

diagnosis of structural heart disease in late gestation, or even in the postnatal period. Our group undertook a study that attempted to further characterize the development of cardiac defects in utero, and to evaluate the effects of this newly recognized entity on the accuracy of prenatal diagnosis of CHD.62 In a retrospective study, we reviewed the medical records of 22 050 pregnant women and their newborns, of a mixed high- and low-risk population for CHD. Patients were divided into two groups: 6294 who had initial TVS at 13–16 weeks’ gestation, followed by TAS at 20–22 weeks, and 15 126 who had initial TAS at 20–22 weeks. Both groups were subsequently examined in the third trimester, and all newborns were examined by certified pediatricians. Congenital heart disease was diagnosed in 168 babies: 66 in group A and 102 in group B, for an overall rate of 7.6:1000. In group A (Figure 13.5), 42 malformations (64%) were detected at the first TVS examination, and 11 (17%) were diagnosed during the subsequent TAS. Three additional anomalies (4%) were found during the third trimester, and 10 malformations (15%) were detected postnatally. In group B, 80 malformations (78%) were detected at the initial TAS scan at mid-trimester, and an additional seven (7%) were found in the third trimester, whereas 15 (15%) were diagnosed postnatally. Overall, 85% of the affected children were diagnosed prenatally. The 10 anomalies that were diagnosed only during the third trimester included: aortic stenosis (n = 2), cardiac rhabdomyoma (n = 2), subaortic stenosis (n = 1), tetralogy of Fallot (n = 1), aortic coarctation (n = 1), sealed foramen ovale (n = 1), ventricular septal defect (n = 1), and hypertrophic cardiomyopathy (n = 1).62 Possible causes for delayed diagnosis may be classified into three major categories: limited resolution, which

First and early second trimester fetal heart screening

may be related to both instrumentation and fetal size and position (category A), progression of lesions in utero, leading to late onset of CHD (category B), and errors in diagnosis (category C). Isolated VSDs are probably the most commonly missed CHD during prenatal sonographic scanning, and this is probably the result of a combination of limited ultrasound resolution as well as erroneous diagnosis. Lesions that may evolve and progress in utero are of major interest. Examples of such lesions include major vessel stenosis and ventricular outflow tract obstruction. Abnormal pressure gradients may result in focal hypoplasia and structural remodeling, which can predominate anatomically. For example, narrowing of the outflow tract will first prompt ventricular asymmetry. Later, chamber hypoplasia, fibroelastosis, or both, may ensue. These forms of outflow tract lesions may not appear obvious during the first half of pregnancy, mainly because the process of arrested valve growth is not significant enough to be delineated so early by ultrasound. Therefore, although the heart will have completed its structural development by the end of the first trimester, an apparently normal appearance of the heart at that time does not exclude major CHD. Furthermore, physicians and patients should be aware that rarely, serious defects may develop even after mid-trimester. Consequently, follow-up examinations are of major importance and should be performed throughout pregnancy, especially in high-risk patients.

Precautions and recommendations The information outlined above clearly demonstrates that high-frequency transvaginal echocardiography provides a comprehensive assessment of the fetal heart by 13–15 weeks’ gestation. This is noteworthy, as the FCV can be imaged in 100% of cases and the extended examination can be completed in 98%, therefore providing a reliable diagnosis of the major cardiac defects at this early stage of pregnancy. In addition, in our experience, only 2–3% of the anxious patients will remain who will need a repeated scan because of inadequate visualization. However, some caveats concerning early cardiac diagnosis should be emphasized. First is the fact that pathologic confirmation of an echocardiographic diagnosis of fetal CHD is essential if the pregnancy is terminated. The small size of the specimens after pregnancy interruption during the first or mid-trimester can render this task difficult, irrespective of the technique used for termination. Postmortem confirmation of the diagnosis is almost impossible after vacuum evacuation of the uterus. By using dilatation and evacuation we found that confirmation was possible in only 62% of our cases.59 This requires certain

193

expertise and careful inspection of the products of conception, and even filtering to avoid loss of a fetal heart, which measures about 7–8 mm at this gestational age. While termination of pregnancy by prostaglandins permits a more gentle extraction of the embryo or fetus so that pathologic confirmation may be achieved in nearly 100% of cases,95 it has the disadvantage of being an inpatient procedure that carries considerable physical and psychological morbidity. Second, cardiac abnormalities diagnosed early in pregnancy tend to be more complex than those detected in the second half of pregnancy, and cause more severe hemodynamic disturbances in the small, developing fetus. For example, a common feature in seven of eight fetuses in one of our previously reported series59 was the demonstration of fluid accumulation, i.e. ascites, pleural–pericardial effusion, and a huge generalized hygroma enveloping the entire fetal body surface, as was also described by Gembruch et al.96 In comparison, only two of 23 fetuses with CHD detected during the second trimester had such fluid accumulation.38 Furthermore, because of the complexity and lethality of many of the anomalies amenable to diagnosis early in pregnancy, spontaneous miscarriage occurs frequently. Similarly, the incidence of CHD in second trimester abortions is high,3,6 as are abnormal karyotypes.2,4 This should be borne in mind while considering the management of fetuses with complex CHD, and while counseling their parents. Third, considerable experience in the techniques of TVS and fetal echocardiography is obligatory before attempting early diagnosis of CHD, since only a sonographer who is well acquainted with both of these techniques can perform transvaginal echocardiography. The FCV and great arteries are uniformly available for examination by 12 weeks’ gestation, which can shift early screening for CHD to this gestational age. We prefer to perform early fetal echocardiography at 13–15 weeks’ gestation. This is the time when the combination of the sufficient anatomic size of the heart and the higher resolution, but limited focal range of TVS, is expected to give best results. The benefits of early cardiac diagnosis may still be preserved at that period. Large studies have confirmed the feasibility and high sensitivity of early fetal echocardiography. However, it must be stressed that repeated examination at mid-trimester to rule out developmental CHD is absolutely essential.

Legend for the DVD Video clip 13.1 The video clip shows the MPR screen of a volume data set acquired with STIC in a 15 weeks’ fetus. The speed has been slowed to allow for annotation. Beginning from the four chamber view the atria and ventricles are seen, moving cephalad

194

Fetal Cardiology

the aortic valve is seen in the “A” plane (upper left quadrant) and the orthogonal “B” plane. Continuing the sweep cephalad the left and right pulmonary arteries are shown, then the main pulmonary artery and the ductus arteriosus. Next the three vessels and trachea plane is seen, which shows the main pulmonary artery, the aortic arch, ductus arteriosus, superior vena cava, the descending aorta, trachea, and the azygos vein. Also available in post-processing is the rendered image, where the foramen ovale flap is seen. Abbreviations: Lt, Rt: left and right; LV, RV: left and right ventricles; LA, RA: left and right atria; AV, AoV, aortic valve; LPA, RPA, MPA: right, left, main pulmonary arteries; DA: ductus arteriosus; AoA: aortic arch; SVC: superior vena cava; Tr: trachea; Az: azygos vein; FOF: foramen ovale flap.

References 1. Hoffman JI, Kaplan S. The incidence of congenital heart disease. J Am Coll Cardiol 2002; 39: 1890–900. 2. Ferencz C, Neill CA, Boughman JA et al. Congenital cardiovascular malformations associated with chromosome abnormalities: an epidemiologic study. J Pediatr 1989; 114: 79–86. 3. Ursell PC, Byrne JK, Strobino BA. Significance of cardiac defects in the developing fetus: a study of spontaneous abortuses. Circulation 1985; 72: 1232–6. 4. Copel JA, Cullen M, Green JJ et al. The frequency of aneuploidy in prenatally diagnosed congenital heart disease: an indication for fetal karyotyping. Am J Obstet Gynecol 1988; 158: 409–13. 5. Allan LD, Sharland GK, Chita SK, Lockhart S, Maxwell DJ. Chromosomal anomalies in fetal congenital heart disease. Ultrasound Obstet Gynecol 1991; 1: 8–11. 6. Nicolaides KH, Snijders RJM, Gosden CM, Berry C, Campbell S. Ultrasonographically detectable markers of fetal chromosomal abnormalities. Lancet 1992; 340: 704–7. 7. DeVore GR, Steiger RM, Larson EJ. Fetal echocardiography: the prenatal diagnosis of a ventricular septal defect in a 14 week fetus with pulmonary artery hypoplasia. Obstet Gynecol 1987; 69: 494–7. 8. Timor-Tritsch IE, Peisner DB, Raju S. Sonoembryology: an organ oriented approach using a high-frequency vaginal probe. J Clin Ultrasound 1990; 18: 286–98. 9. Neiman UL. Transvaginal ultrasound embryography. Semin Ultrasound CT MR 1990; 11: 22–33. 10. Bonnet D, Coltri A, Butera G et al. Detection of transposition of the great arteries in fetuses reduces neonatal morbidity and mortality. Circulation 1999; 23: 916–18. 11. Verheijen PM, Lisowski LA, Stoutenbeek P et al. Prenatal diagnosis of congenital heart disease affects preoperative acidosis in the newborn patient. J Thorac Cardiovasc Surg 2001; 121: 798–803. 12. Jaeggi ET, Sholler GF, Jones OD, Cooper SG. Comparative analysis of pattern, management and outcome of pre- versus postnatally diagnosed major congenital heart disease: a population-based study. Ultrasound Obstet Gynecol 2001; 17: 380–5.

13. Moore KL, Persaud TVN. The cardiovascular system. In: The Developing Human: Clinically Oriented Embryology, 6th edn. Philadelphia: WB Saunders, 1998: 349–404. 14. D’Amelio R, Gioriandino C, Masala L et al. Fetal echocardiography using transvaginal and transabdominal probes during the first period of pregnancy: a comparative study. Prenat Diagn 1991; 11: 69–75. 15. Rottem S, Chervenak FA. Ultrasound diagnosis of fetal anomalies. Obstet Gynecol Clin North Am 1990; 17: 17–40. 16. Timor-Tritsch IE, Farine D, Rosen MG. A close look at embryonic development with the high frequency transvaginal transducer. Am J Obstet Gynecol 1988; 159: 676–81. 17. Shapiro I, Degani S, Leibovitz Z et al. Fetal cardiac measurements derived by transvaginal and transabdominal crosssectional echocardiography from 14 weeks of gestation to term. Ultrasound Obstet Gynecol 1998; 12: 404–18. 18. Blaas HG, Eik-Nes SK, Kiserud T, Hellevik LR. Early development of the abdominal wall, stomach and heart from 7 to 12 weeks of gestation: a longitudinal ultrasound study. Ultrasound Obstet Gynecol 1995; 6: 240–9. 19. Dolkart L, Reimers FT. Transvaginal fetal echocardiography in early pregnancy: normative data. Am J Obstet Gynecol 1991; 165: 688–91. 20. Achiron R, Golan-Porat N, Gabbay U et al. In utero ultrasonographic measurements of fetal aortic and pulmonary artery diameters during the first half of gestation. Ultrasound Obstet Gynecol 1998; 11: 180–4. 21. Gembruch U, Shi C, Smrcek JM. Biometry of the fetal heart between 10 and 17 weeks of gestation. Fetal Diagn Ther 2000; 15: 20–31. 22. Achiron R, Zimand S, Hegesh J et al. Fetal aortic arch measurements between 14 and 38 weeks’ gestation: in-utero ultrasonographic study. Ultrasound Obstet Gynecol 2000; 15: 226–30. 23. Firpo C, Hoffman JI, Silverman NH. Evaluation of fetal heart dimensions from 12 weeks to term. Am J Cardiol 2001; 87: 594–600. 24. Smrcek JM, Berg C, Geipel A et al. Early fetal echocardiography: heart biometry and visualization of cardiac structures between 10 and 15 weeks’ gestation. J Ultrasound Med 2006; 25: 173–82. 25. Becker R, Wegner RD. Detailed screening for fetal anomalies and cardiac defects at the 11-13-week scan. Ultrasound Obstet Gynecol 2006; 27: 613–18. 26. McAuliffe FM, Trines J, Nield LE et al. Early fetal echocardiography—a reliable prenatal diagnosis tool. Am J Obstet Gynecol 2005; 193: 1253–9. 27. Carvalho MH, Brizot ML, Lopes LM et al. Detection of fetal structural abnormalities at the 11-14 week ultrasound scan. Prenat Diagn 2002; 22: 1–4. 28. Haak MC, Twisk JW, Van Vugt JM. How successful is fetal echocardiographic examination in the first trimester of pregnancy? Ultrasound Obstet Gynecol 2002; 20: 9–13. 29. Simpson JM, Jones A, Callaghan N, Sharland GK. Accuracy and limitations of transabdominal fetal echocardiography at 12-15 weeks of gestation in a population at high risk for congenital heart disease. BJOG 2000; 107: 1492–7. 30. Comas Gabriel C, Galindo A, Martinez JM et al. Early prenatal diagnosis of major cardiac anomalies in a high-risk population. Prenat Diagn 2002; 22: 586–93.

First and early second trimester fetal heart screening

31. Smrcek JM, Berg C, Geipel A et al. Detection rate of early fetal echocardiography and in utero development of congenital heart defects. J Ultrasound Med 2006; 25: 187–96. 32. Achiron R, Weissman A, Rotstein Z et al. Transvaginal echocardiographic examination of the fetal heart between 13 and 15 weeks’ gestation in a low-risk population. J Ultrasound Med 1994; 13: 783–9. 33. Weissman A, Achiron R. Ultrasound diagnosis of congenital anomalies in early pregnancy. In: Jurkovic D, Jauniaux E, eds. Ultrasound and Early Pregnancy. Carnforth, UK: Parthenon Publishing, 1995: 95–120. 34. Kleinman CS, Hobbins JC, Jaffe CC et al. Echocardiographic studies of the human fetus: prenatal diagnosis of congenital heart disease and cardiac dysrhythmias. Pediatrics 1980; 65: 1059–67. 35. Allan LD, Crawford DC, Chita SK, Tynan MJ. Prenatal screening for congenital heart disease. Br Med J 1986; 292: 1717–19. 36. Copel JA, Pilu G, Green J et al. Fetal echocardiographic screening for congenital heart disease: the importance of the four-chamber view. Am J Obstet Gynecol 1987; 157: 648–55. 37. Leopold G. Antepartum obstetrical guidelines. J Ultrasound Med 1986; 5: 241–2. 38. Achiron R, Glaser J, Gelernter I, Hegesh J, Yagel S. Extended fetal echocardiographic examination for detecting cardiac malformations in low risk pregnancies. BMJ 1992; 304: 671–4. 39. Bromley B, Estroff JA, Sanders SP et al. Fetal echocardiography: accuracy and limitations in a population at high and low risk for heart defects. Am J Obstet Gynecol 1992; 166: 1473–80. 40. Garmel SK, D’Afton ME. Fetal ultrasonography. West J Med 1993; 159: 273–85. 41. International Society of Ultrasound in Obstetrics and Gynecology. Cardiac screening examination of the fetus: guidelines for performing the ‘basic’ and ‘extended basic’ cardiac scan. Ultrasound Obstet Gynecol 2006; 27: 107–13. 42. Yoo SJ, Lee YH, Kim ES et al. Three-vessel view of the fetal upper mediastinum: an easy means of detecting abnormalities of the ventricular outflow tracts and great arteries during obstetric screening. Utrasound Obstet Gynecol 1997; 9: 173–82. 43. Yoo SJ, Young HL, Kyoung SC. Abnormal three-vessel view on sonography: a clue to the diagnosis of congenital heart disease in the fetus. AJR Am J Roentgenol 1999; 172: 825–30. 44. Yagel S, Cohen SM, Achiron R. Examination of the fetal heart by five short-axis views: a proposed screening method for comprehensive cardiac evaluation. Ultrasound Obstet Gynecol 2001; 17: 367–9. 45. Yagel S, Arbel R, Anteby EY, Raveh D, Achiron R. The three vessels and trachea view (3VT) in fetal cardiac scanning. Ultrasound Obstet Gynecol 2002; 20: 340–5. 46. Achiron R, Rotstein Z, Heggesh J et al. Anomalies of the fetal aortic arch: a novel sonographic approach to in-utero diagnosis. Ultrasound Obstet Gynecol 2002; 20: 553–7. 47. Vinals F, Heredia F, Giuliano A. The role of the three vessels and trachea view (3VT) in the diagnosis of congenital heart defects. Ultrasound Obstet Gynecol 2003; 22: 358–67.

195

48. Zalel Y, Wiener Y, Gamzu R et al. The three-vessel and tracheal view of the fetal heart: an in utero sonographic evaluation. Prenat Diagn 2004; 24: 174–8. 49. Del Rio M, Martinez JM, Figueras F et al. Doppler assessment of fetal aortic isthmus blood flow in two different sonographic planes during the second half of gestation. Ultrasound Obstet Gynecol 2005; 26: 170–4. 50. Vinals F, Ascenzo R, Poblete P et al. Simple approach to prenatal diagnosis of transposition of the great arteries. Ultrasound Obstet Gynecol 2006; 28: 22–5. 51. Del Bianco A, Russo S, Lacerenza N et al. Four chamber view plus three-vessel and trachea view for a complete evaluation of the fetal heart during the second trimester. J Perinat Med 2006; 34: 309–12. 52. Shinebourne EA, Macartney FJ, Anderson RH. Sequential chamber localization – logical approach to diagnosis in congenital heart disease. Br Heart J 1976; 38: 327–40. 53. Van Praagh R. Terminology of congenital heart disease. Glossary and commentary. Circulation 1977; 56: 139–43. 54. Allen LD. Development of congenital lesions in mid or late gestation. Int J Cardiol 1988; 19: 361–2. 54. Carvalho JS, Ho SY, Shinebourne EA. Sequential segmental analysis in complex fetal cardiac abnormalities: a logical approach to diagnosis. Ultrasound Obstet Gynecol 2005; 26: 105–11. 55. Gembruch U, Knopfle G, Chatterjee M, Bald R, Hansmann M. First trimester diagnosis of fetal congenital heart disease by transvaginal two-dimensional and Doppler echocardiography. Obstet Gynecol 1990; 75: 496–98. 56. Bronshtein M, Zimmer EZ, Gerlis LM, Lorber A, Drugan A. Early ultrasound diagnosis of fetal congenital heart defects in high-risk and low-risk pregnancies. Obstet Gynecol 1993; 82: 225–9. 57. Bronshtein M, Siegler E, Yoffe N, Zimmer EZ. Prenatal diagnosis of ventricular septal defect and overriding aorta at 14 weeks’ gestation, using transvaginal sonography. Prenat Diagn 1990; 10: 697–702. 58. Bronshtein M, Zimmer EZ, Milo S et al. Fetal cardiac abnormalities detected by transvaginal sonography at 12-16 weeks’ gestation. Obstet Gynecol 1991; 78: 374–8. 59. Achiron R, Rotstein Z, Lipitz S, Mashiach S, Hegesh J. Firsttrimester diagnosis of fetal congenital heart disease by transvaginal ultrasonography. Obstet Gynecol 1994; 84: 69–72. 60. Rustico MA, Benettoni A, Bussani R, Maieron A, Mandruzzato G. Early fetal endocardial fibroelastosis and critical aortic stenosis: a case report. Ultrasound Obstet Gynecol 1995; 5: 202–5. 61. Kirk JS, Riggs TW, Comstock CH et al. Prenatal screening for cardiac anomalies: the value of routine addition of the aortic root to the four-chamber view. Obstet Gynecol 1994; 84: 427–31. 62. Yagel S, Weissman A, Rotstein Z et al. Congenital heart defects: natural course and in utero development. Circulation 1997; 96: 550–5. 63. Yagel S, Cohen SM, Shapiro I, Valsky DV. 3D and 4D ultrasound in fetal cardiac scanning: a new look at the fetal heart. Ultrasound Obstet Gynecol 2007; 29: 81–95. 64. DeVore GR, Falkensammer P, Sklansky MS, Platt LD. Spatio-temporal image correlation (STIC): new technology for evaluation of the fetal heart. Ultrasound Obstet Gynecol 2003; 22: 380–7.

196

Fetal Cardiology

65. Goncalves LF, Lee W, Chaiworapongsa T et al. Fourdimensional ultrasonography of the fetal heart with spatiotemporal image correlation. Am J Obstet Gynecol 2003; 189: 1792–802. 66. Goncalves LF, Lee W, Espinoza J, Romero R. Examination of the fetal heart by four-dimensional (4D) ultrasound with spatio-temporal image correlation (STIC). Ultrasound Obstet Gynecol 2006; 27: 336–48. 67. Yagel S, Benachi A, Bonnet D et al. Rendering in fetal cardiac scanning: the intracardiac septa and the coronal atrioventricular valve planes. Ultrasound Obstet Gynecol 2006; 28: 266–74. 68. Paladini D, Russo MG, Vassallo M, Tartaglione A. The ‘inplane’ view of the inter-ventricular septum. A new approach to the characterization of ventricular septal defects in the fetus. Prenat Diagn 2003; 23: 1052–5. 69. Sciaky-Tamir Y, Cohen SM, Hochner-Celnikier D et al. Three-dimensional power Doppler (3DPD) ultrasound in the diagnosis and follow-up of fetal vascular anomalies. Am J Obstet Gynecol 2006; 194: 274–81. 70. DeVore GR, Polanco B, Sklansky MS, Platt LD. The ‘spin’ technique: a new method for examination of the fetal outflow tracts using three-dimensional ultrasound. Ultrasound Obstet Gynecol 2004; 24: 72–82. 71. Abuhamad A. Automated multiplanar imaging: a novel approach to ultrasonography. J Ultrasound Med 2004; 23: 573–6. 72. Goncalves LF, Espinoza J, Lee W et al. A new approach to fetal echocardiography: digital casts of the fetal cardiac chambers and great vessels for detection of congenital heart disease. J Ultrasound Med 2005; 24: 415–24. 73. Espinoza J, Kusanovic JP, Goncalves LF et al. A novel algorithm for comprehensive fetal echocardiography using 4-dimensional ultrasonography and tomographic imaging. J Ultrasound Med 2006; 25: 947–56. 74. Vinals F, Mandujano L, Vargas G, Giuliano A. Prenatal diagnosis of congenital heart disease using four-dimensional spatio-temporal image correlation (STIC) telemedicine via an Internet link: a pilot study. Ultrasound Obstet Gynecol 2005; 25: 25–31. 75. Michailidis GD, Simpson JM, Karidas C, Economides DL. Detailed three-dimensional fetal echocardiography facilitated by an Internet link. Ultrasound Obstet Gynecol 2001; 18: 325–8. 76. Achiron R, Tadmor O, Mashiach S. Heart rate as a predictor of first-trimester spontaneous abortion after ultrasoundproven viability. Obstet Gynecol 1991; 78: 330–4. 77. Baschat AA, Gembruch U, Knopfle G, Hansmann M. First-trimester fetal heart block: a marker for cardiac anomaly. Ultrasound Obstet Gynecol 1999; 14: 311–14. 78. Sciarrone A, Masturzo B, Botta G et al. First-trimester fetal heart block and increased nuchal translucency: an indication for early fetal echocardiography. Prenat Diagn 2005; 25: 1129–32. 79. Porat S, Anteby EY, Hamani Y, Yagel S. Fetal supraventricular tachycardia diagnosed and treated at 13 weeks of gestation: a case report. Ultrasound Obstet Gynecol 2003; 21: 302–5. 80. Hyett JA, Perdu NL, Sharland GK, Snijders RJM, Nicolaides KH. Increased nuchal translucency at

81.

82.

83.

84.

85.

86.

87. 88.

89.

90.

91.

92.

93.

94. 95. 96.

10–14 weeks of gestation as a marker for major cardiac defects. Ultrasound Obstet Gynecol 1997; 10: 242–6. Hyett JA, Perdu M, Sharland GK et al. Using fetal nuchal translucency to screen for major congenital cardiac defects at 10-14 weeks of gestation: population based cohort study. Br Med J 1999; 318: 81–5. Snijders RJM, Noble P, Sebire N, Souka A, Nicolaides KH. UK multicentre project on assessment of risk of trisomy 21 by maternal age and fetal nuchal-translucency thickness at 10-14 weeks of gestation. Lancet 1998; 352: 343–6. Haak MC, Bartelings MM, Gittenberger-De Groot AC, Van Vugt JM. Cardiac malformations in first-trimester fetuses with increased nuchal translucency: ultrasound diagnosis and postmortem morphology. Ultrasound Obstet Gynecol 2002; 20: 14–21. Makrydimas G, Sotiriadis A, Huggon IC et al. Nuchal translucency and fetal cardiac defects: a pooled analysis of major fetal echocardiography centers. Am J Obstet Gynecol 2005; 192: 89–95. Lopes LM, Brizot ML, Lopes MA et al. Structural and functional cardiac abnormalities identified prior to 16 weeks’ gestation in fetuses with increased nuchal translucency. Ultrasound Obstet Gynecol 2003; 22: 470–8. Allan LD, Sharland G, Tynan MJ. The natural history of the hypoplastic heart syndrome. Int J Cardiol 1989; 25: 341–3. Allan LD, Crawford DC, Tynan M. Evolution of the aorta in intrauterine life. Br Heart J 1984; 52: 471–3. Todors T, Presbitero P, Gaglioti P, Demarie D. Pulmonary stenosis with intact ventricular septum: documentation of development of the lesion echocardiographically during fetal life. Ant J Cardiol 1988; 19: 355–60. Marasini M, DeCaro E, Pongiglione G, Ribaidone D, Caponetto S. Left heart obstructive disease: changes in echocardiographic appearance during pregnancy. J Clin Utrasound 1993; 21: 65–8. Achiron R, Weissman A, Matitiahu A et al. Endocardial fibroelastosis secondary to critical aortic stenosis: natural course and evolution in utero. Ultrasound Obstet Gynecol 1994; 203(Suppl): 354. Rice MJ, McDonald RW, Reller MD. Progressive pulmonary stenosis in the fetus: two case reports. Am J Perinatol 1993; 10: 424–7. Yagel S, Achiron R, Ron M, Revel A, Anteby E. Transvaginal ultrasonography at early pregnancy cannot be used alone for targeted organ ultrasonographic examination in a high-risk population. Am J Obstet Gynecol 1995; 172: 971–5. Hornberger LK, Sanders SP, Sahn DJ et al. In utero pulmonary artery and aortic growth and potential for progression of pulmonary outflow tract obstruction in tetralogy of Fallot. J Am Coll Cardiol 1995; 25: 739–45. Gerlis LM. Cardiac malformations in spontaneous abortuses. Int J Cardiol 1985; 7: 29–35. Allan LD. Evolution of echocardiographic findings in the fetus. Circulation 1997; 96: 391–2. Gembruch U, Knopfle G, Bald R, Hansmann M. Early diagnosis of fetal congenital heart disease by transvaginal echocardiography. Ultrasound Obstet Gynecol 1993; 3: 310–17.

14 Four-dimensional ultrasound examination of the fetal heart by spatiotemporal image correlation Luís F Gonçalves, Jimmy Espinoza, Juan Pedro Kusanovic, Wesley Lee, and Roberto Romero Introduction Prenatal evaluation of the fetal heart is one of the most challenging components of the obstetrical ultrasound examination. Sonographers must conduct a comprehensive examination, which includes a detailed assessment of the four-chamber view (Table 14.1),1,2 the connections of the great vessels to the ventricular chambers, the venous return to the heart, and cardiac rhythm. Besides the limited amount of time that is usually available for a comprehensive ultrasound examination in clinical practice, other challenges that are particular to the examination of the fetal heart include: (1) the fact that the fetus is located within a uterus inside another body (the mother) and, thus, factors such as maternal obesity, abdominal scars, anterior placentas, and oligohydramnios may have a negative impact on image quality; (2) frequent movement or breathing during the examination; (3) less than ideal fetal position that cannot be controlled by the operator; and (4) faster heart rates than in adult and pediatric patients.3–6 Thus, extensive training is required to develop the skills necessary to examine the fetal heart effectively. Indeed, operator skill is considered by many as the most important factor affecting prenatal diagnosis of congenital heart disease.7–12 Failure to diagnose a life-threatening cardiac disorder in utero may negatively impact survival, since there is evidence that prompt intervention after delivery is associated with improved outcomes in disorders such as transposition of the great arteries, hypoplastic left heart syndrome, and coarctation of the aorta.13–16 Four-dimensional ultrasonography (4DUS) with spatiotemporal image correlation (STIC) allows examiners to acquire volume data sets of the fetal heart using gray-scale imaging only or with the addition of blood flow information from color Doppler, power Doppler, or B-flow imaging.5,17–29 Once acquired, volume data sets can be examined using the standard planes of section of two-dimensional ultrasonography (2DUS), as well as novel planes that are only possible by volumetric imaging.

These volume data sets can be thought of as ‘digital specimens’ of the fetal heart, akin to actual heart specimens that are examined by pathologists during a necropsy. The ‘digital heart’ can be oriented on the screen to be displayed in a standardized position, after which standard planes of section are obtained with the use of ‘digital scalpel’ tools. Moreover, sophisticated three-dimensional rendering techniques can be applied to display ‘digital casts’ of cardiac chambers and great vessels that look very similar to postmortem casts obtained by injecting silicone rubber into cardiovascular structures.19,30 Advantages of the ‘digital specimen’ when compared to the ‘actual specimen’ include: (1) functional information is preserved since heart beats are included in the volume data set; (2) the direction of blood flow can be analyzed in volume data sets acquired with color or power Doppler; and (3) if the examiner accidentally makes a mistake during the review of the volume data set, the ‘digital specimen’ is not damaged forever, and all that it takes to begin the examination all over again is to reset it to its original state with the click of a button. Therefore, 4DUS of the fetal heart may help to overcome the dependency on operator skills which is characteristic of fetal echocardiography performed by conventional 2DUS, and eventually improve the prenatal characterization of complex congenital cardiac defects.7–12,31 In this chapter, we will review several techniques that can be used to examine the fetal heart by 4DUS with STIC in normal fetuses as well as in those with congenital heart disease.

Technology The term 4DUS is used to describe volume data sets that incorporate information about the three spatial dimensions plus the temporal dimension.32 4DUS of the heart presents particular technical challenges, since the phases of

198

Fetal Cardiology

Table 14.1 Basic cardiac screening examination. Adapted with permission of the American Institute of Ultrasound Medicine from reference 2 General Normal cardiac situs, axis and position Heart occupies a third of thoracic area Majority of heart in left chest Four cardiac chambers present No pericardial effusion or hypertrophy Atria Atria approximately equal in size Foramen ovale flap in left atrium Atrial septum primum present Ventricles Ventricles about equal in size No cardiac wall hypertrophy Moderator band at right ventricular apex Ventricular septum intact (apex to crux) Atrioventricular valves Both atrioventricular valves open and move freely Tricuspid valve leaflet inserts on ventricular septum closer to the cardiac apex than to the mitral valve

the cardiac cycle must be not only included in the final volume data set, but also synchronized with the acquired spatial information. This synchronization is termed cardiac gating. In adults and children, a cardiac gating signal can be obtained by simultaneous recording of an electrocardiogram. Since electrocardiographic signals are difficult to obtain in the fetus, other gating methods have been proposed to overcome this problem.33–41 STIC is a gating algorithm based on the analysis of cardiac motion.35,40 The fundamental fetal heart rate is directly extracted from the volume data set (Figure 14.1).5,22,27 Although retrospective, gating is quickly performed immediately after the volume scan, while the patient is still on the scanning table.4 The end result is a volume data set containing three-dimensional plus temporal information that allows the interactive display of cardiac structures in any plane of section as well as three-dimensional reconstruction by rendering techniques.

Volume acquisition Volume acquisition is a crucial aspect of the 4DUS with STIC. The final quality of the volume data set is heavily dependent on prior adjustment of two-dimensional grayscale and color Doppler parameters.23 Fetal echocardiography presets are available in most ultrasound systems, and we favor those characterized by low persistence, high

Figure 14.1 Raw data image used to calculate the fetal heart rate. This image is generated after the spatiotemporal image correlation (STIC) acquisition is performed as a single, slow sweep. Information from these raw data is used to rearrange the two-dimensional frames. This particular image is orthogonal to the original two-dimensional frame. Because of the long acquisition time (7.5–15 seconds from the left to the right end of this image), the beating heart draws a motion pattern. This pattern is analyzed in terms of periodical changes of gray-scale information, and the fetal heart rate is calculated. Beat-to-beat changes of the heart rate would appear as shortening or elongation of the above motion pattern. This image is not visible on the system during STIC acquisition, but helps to understand the technique (reproduced from reference 22 with permission from Elsevier).

contrast, and high frame rate Magnification should be performed prior to acquisition. Harmonic imaging may improve image quality in selected cases (e.g. obese patients).42,43 Once the two-dimensional image is optimized, a region of interest is selected on the screen, and volume acquisition is performed with a single automated sweep of the transducer (Video clip 14.1). Depending on the manufacturer, volume acquisition may take from 5 to 15 seconds. The STIC algorithm detects the fundamental heart rate based on the rhythmic pattern of changes in cardiac diameter and uses this information for gating. Frames acquired during the same phase of the cardiac cycle, although from a different position in space, are merged into the same volume data set. This process is repeated for all phases of the cardiac cycle. After image rearrangement, an ordered sequence of volume data sets is displayed on the screen as a continuous cine loop containing all phases of the cardiac cycle, and the data are ready for examination.5,22 The examination can be performed either in the presence of the patient or offline, and the volumes can be saved on a hard drive for later review or transmission to a remote diagnosis center.26,27,40

4D ultrasound examination of the fetal heart by spatiotemporal image correlation

Tips to optimize volume acquisition23 Original plane of acquisition Volume acquisition using transverse sweeps across the fetal thorax are preferred when the examiner is interested in evaluation of the four-chamber, five-chamber, threevessel, and three-vessel and trachea views. Conversely, if the examiner is mainly interested in obtaining images of the aortic and ductal arches, high-quality volume data sets are acquired with sagittal sweeps through the fetal thorax.

Fetal position The ideal fetal position to examine the fetal heart is with the fetus lying on its back (i.e. with the spine oriented at approximately the 6 o’clock position on the screen). Since this is not always possible, volume data sets of sufficient quality may also be obtained when the spine is up, provided that it is not positioned between 11 and 1 o’clock. Under these circumstances, acoustic shadowing from the spine and ribs frequently compromises fetal heart imaging.

199

trachea view, superiorly, and the transverse section of the fetal abdomen containing the stomach, aorta and interior vena cava (IVC),27 inferiorly, will be available for the examination. For second trimester fetuses, acquisition angles between 25° and 30° should suffice. Adjustments may be necessary when examining smaller or larger fetuses.

Setting the acquisition time Acquisition time determines the speed that is used by the transducer to sweep through the ROI. Depending upon equipment manufacturer, acquisition time ranges from 5 to 15 seconds. Ideally, volume data sets should be acquired using the longest acquisition time possible to improve spatial resolution. Unfortunately, in the presence of intense fetal movements or breathing, examiners may be forced to select a shorter acquisition time to minimize motion artifacts. This results in a volume data set with lower spatial resolution. In practice, we preset the equipment to sweep for 10 or 12.5 seconds and adjust the acquisition time according to the presence or absence of active fetal movement or breathing. Mothers may also move during acquisition, and abdominal breathing by the mother may cause motion artifacts. Therefore, we have found it useful to ask mothers to momentarily withhold moving or breathing during acquisition.

Selecting a region of interest The region of interest (ROI) determines the width and height of the volume data set (x and y planes). When an examiner begins to use 4DUS with STIC, the tendency is to select a large ROI, including not only the heart, but also the surrounding anatomical structures such as the lungs, ribs, and amniotic fluid. The reader is reminded that the selection of wide ROIs is associated with lower than expected frame rates during acquisition and this may negatively impact the temporal resolution of the final volume data set. This is particularly important during color or power Doppler acquisitions. Thus, ROIs should be set as narrow as possible and include only the information that the examiner is interested in. This practice maximizes both the frame rate during acquisition as well as the final temporal resolution of the volume data set.

Setting the acquisition angle The acquisition angle determines the amount of information acquired in the ‘z’ or azimuth plane. For acquisitions performed in the transverse plane (e.g. the four-chamber view plane), images from the upper mediastinum all the way down to the upper abdomen should be ideally included in the volume data set.27 This practice ensures that all standard planes of section, including the three-vessel and

Examination of the fetal heart with STIC Scrolling through the volume data set The most basic approach to examine a volume data set of the fetal heart acquired with STIC is to scroll through the volume data set from top to bottom along the original plane of acquisition. For volumes acquired using transverse sweeps through the fetal chest, this approach allows examiners to visualize the transverse planes of section proposed by Yoo et al44 and Yagel et al45: (1) transverse view of the upper abdomen; (2) four-chamber view; (3) fivechamber view; (4) three-vessel view; and (5) three-vessel view and trachea view (Figure 14.2 and Video clip 14.2). Vinals et al27 evaluated this approach in 100 volume data sets acquired by sonologists with little experience in examination of the fetal heart. A specialist in fetal echocardiography, who was not involved in volume acquisition, examined the volume data sets. Visualization rates for the four-chamber view, the left and right ventricular outflow tracts, the three-vessel view, and the three-vessel and trachea view ranged from 81 to 100%. The lowest

200

Fetal Cardiology

(a)

(b)

(c)

(d)

Figure 14.2

(e)

The five transverse planes of section proposed by Yagel et al for examination of the fetal heart45: (a) the transverse view of the upper abdomen; (b) the four-chamber view; (c) the five-chamber view; (d) the three-vessel view; and (e) the three-vessel and trachea view. IVC, inferior vena cava; LV, left ventricle; RV, right ventricle; RA, right atrium; LA, left atrium; Ao, aorta; SVC, superior vena cava; PA, pulmonary artery; T, trachea.

visualization rates were observed for structures located either at the upper abdomen or upper mediastinum. This occurred because acquisition angles were not properly set, that is, the sweep angle was not wide enough to include

information from the upper mediastinum through the upper abdomen in some cases. This methodology has also been used for remote diagnosis of congenital heart disease (TELE-STIC).26,46

4D ultrasound examination of the fetal heart by spatiotemporal image correlation

Automated multiplanar slicing Several ultrasound manufacturers now provide software to automatically slice 3D and 4D volume data sets (Multislice View™; Accuvix, Medison, Seoul, Korea; Tomographic Ultrasound Imaging; GE Healthcare, Milwaukee, WI; iSlice; Philips Medical Systems, Bothell, WA; Multi-Slice View; Siemens Medical Solutions, Issaquah, WA). This technology allows an examiner to automatically obtain a series of tomographic parallel images on a single screen, akin to display methods commonly used in computed tomography and magnetic resonance imaging.47 With STIC, since motion information is preserved, multiple slices of the

201

beating heart can be visualized and examined at the same time.48,49 The user can adjust the number and position of slices with specific software controls until the desired planes of section can be visualized in the same image.48,50 Hue, brightness, and contrast can also be adjusted to optimize image quality. This approach represents an alternative to manually scrolling through the volume data sets to obtain standard cardiac views, and may decrease the time spent in the evaluation of cardiac anatomy during screening and consultative examinations by allowing examiners to simultaneously view the spatial relationships of the transverse views described by Yoo et al44 and Yagel et al45 (Figures 14.3 and 14.4 and Video clip 14.3).48

(a)

Figure 14.3 (b)

Tomographic ultrasound imaging (TUI) of a normal fetal heart in systole (a) and diastole (b). The overview image on the left upper panel of each figure shows the orthogonal sagittal plane to the sections that are being displayed. Each line represents a slice. The center slice is marked with an asterisk (*) and each subsequent plane to the right or left is marked with numbers ranging from minus −4 to +4. The plane marked by the dotted line is not displayed. In this volume data set, the five transverse planes of section proposed by Yagel et al45 for examination of the fetal heart are visualized. Please note that the fivechamber view was better visualized during systole. LPA, left pulmonary artery; FO, foramen ovale; IVS, interventricular septum; 4CH, four-chamber; 5CH: five-chamber (reproduced with permission from reference 49).

202

Fetal Cardiology

(a)

(b)

Figure 14.4 Tomographic ultrasound imaging (TUI) of a normal fetal heart in systole (a) and diastole (b). The volume data sets were acquired using B-mode and color Doppler imaging. The overview image on the left upper panel of each figure shows the orthogonal sagittal plane to the sections that are being displayed. Each line represents a slice. The center slice is marked with an asterisk (*) and each subsequent plane to the right or left is marked with numbers ranging from −4 to +4. The plane marked by the dotted line is not displayed. In this volume data set, the three planes of section proposed by Chaoui and McEwing51 for examination of the fetal heart are visualized (reproduced with permission from reference 49).

Previous studies have demonstrated that the five basic axial planes of section for examination of the fetal heart can be automatically obtained in the majority of patients.49 Manual adjustment of the interslice distance when all views are not obtained by automatically slicing the volume data set with the equipment improves the visualization rates of the transverse planes of section proposed by Yoo et al44 and Yagel et al.45 The optimal interslice distance to automatically slice the volume data set and simultaneously visualize these planes of section changes with gestational age (Table 14.2). In addition, the visualization rate for the five-chamber view improves by approximately 14% over what is possible by gray-scale

imaging alone for volumes acquired with color Doppler imaging.49

Systematic approach for visualization of the great vessels There are now several approaches to systematically demonstrate the great vessels in volume data sets acquired with STIC.21,22,52–54 In 2003, we proposed and subsequently validated a technique designed to simultaneously display the long-axis view of the left ventricular outflow tract and

4D ultrasound examination of the fetal heart by spatiotemporal image correlation

Table 14.2

Interslice distance according to gestational age. Adapted with permission from reference 50

Gestational age (weeks) 12–15

203

n

Minimum (mm)

Maximum (mm)

Mean (mm)

11

0.7

1.4

1.14

SD (mm) 0.26

16–19

10

1.4

2.6

1.78

0.34

20–24

44

2.2

3.4

2.82

0.27

25–29

15

2.2

4.6

3.63

0.65

30–34

17

3.4

5.3

4.37

0.56

35–40

6

4.3

5.3

4.75

0.37

the short-axis view of the right ventricular outflow tract on panels A and B of the multiplanar display.21,22 The technique was developed for volume data sets acquired by sweeping the fetal thorax on the transverse plane of section using the four-chamber view as the starting point, and is illustrated in detail in Figure 14.5 and Video clip 14.4. Figure 14.6 and Video clip 14.5 show the clinical application of this technique for the diagnosis of transposition of the great arteries.20 More recently, Espinoza et al54 developed an enhanced algorithm combining tomographic ultrasound imaging and standardized manipulations of the volume data set to simultaneously display the three-vessel and trachea view, the four-chamber view, and both outflow tracts on the same screen. The algorithm was developed to overcome a limitation of automatic slicing using tomographic ultrasound imaging alone, which is that the simple use of parallel planes to the four-chamber view of the heart does not allow visualization of the long-axis view of the left outflow tract and the short-axis view of the aorta, which are considered part of an integral examination of the fetal heart. Details of the algorithm are described in Figure 14.6. In a study that included 227 volume data sets of fetuses with (n = 14) and without (n = 138) congenital heart disease, simultaneous visualization of the short axis of the aorta, three-vessel and trachea view, left outflow tract, and four-chamber view was achieved in 78% (152/195) of the volume data sets from fetuses without and 40% (8/20) of those with congenital heart disease. The lower visualization rates of standard planes of section in cases of congenital heart disease may reflect abnormal spatial relationships between cardiovascular structures, and may prove helpful in the identification of specific patterns of congenital heart disease in the future. The latest approach to obtaining standard planes of section by manipulation of the volume data set has been denominated ‘automated sonography’.52,53 According to this concept, the standard planes of section for examination of the fetal heart can be automatically displayed by the equipment once the four-chamber view is selected as the reference plane. The spatial relationships of the standard planes of section to the four-chamber view52,58 as well

as the geometric changes that the fetal heart undergoes as gestational age progresses54 have been studied as a preliminary step for the development of such algorithms.

Systematic approach for visualization of the aortic and ductal arches Ideal visualization of the aortic and ductal arches requires that the volume data set be acquired with sagittal sweeps through the fetal chest. The original technique describing how to systematically visualize the aortic and ductal arches was originally published by Bega et al55 in 2001, and is illustrated in Figure 14.7 and Video clip 14.6.

Spin technique DeVore et al18 proposed in 2004 a technique to display the ascending aorta and transverse aortic arch, the main pulmonary artery and bifurcation of the right and left pulmonary arteries, the ductus arteriosus, and superior vena cava. The technique is simple, and can be applied to any structure of interest during the examination of volume data sets of the fetal heart, simply by positioning the reference dot in the center of the structure and ‘spinning’ the volume data set around the y axis until the structure is ‘opened up’ and visualized in its entirety. Figure 14.8 illustrates the application of the spin technique to identify an abnormal vessel visualized in the three-vessel and trachea view as a persistent left superior vena cava draining into a dilated coronary sinus.

Rendering techniques for visualization of intracardiac structures and valves Rendering techniques can be applied to the visualization of intracardiac structures to obtain a depth perspective of

204

Fetal Cardiology

(a)

(b)

Figure 14.5

(c)

(d)

(e)

(f)

Technique to systematically visualize the left and right outflow tracts in volume data sets acquired with STIC. Only panels A (original plane of acquisition, transverse) and B (sagittal plane) are displayed. (a) The first step in this technique consists of making sure that the left side of the heart is located on the left side of the image, and the right side of the heart is on the right side of the image; if necessary, rotate the volume around the y axis until this is achieved. The reference dot is then positioned in the crux of the heart, on both the transverse and the sagittal planes. (b) The volume data set is rotated around the z axis until a perfect apical four-chamber view is obtained. (c) The volume is then rotated around the z axis counter-clockwise until the angle between the apex and the transducer is approximately 30–40°. (d) The crucial step is to position the reference dot in the interventricular septum, midway between the crux of the heart and the apex; this will anchor the three orthogonal planes for the next rotation movement, which will display the left ventricular outflow tract. (e) The volume data set is now rotated around the y axis; this will open up the continuity between the interventricular septum and the anterior wall of the aorta; the anterior leaflet of the mitral valve is seen in continuity with the posterior wall of the aorta. (f) Once the reference dot is moved above the aortic valve, the short-axis view of the right ventricular outflow tract is displayed on the sagittal plane. LVOT, left ventricular outflow tract (reproduced with permission from reference 23).

4D ultrasound examination of the fetal heart by spatiotemporal image correlation

(a)

(b)

(c)

(d)

(e)

(f)

205

Figure 14.6 (a) An ‘overview image’ is shown in the upper left corner. The parallel lines determine the position of the eight orthogonal planes to the plane containing the ‘overview image’. (b) Volume data sets were adjusted to display the four-chamber view in panel A, where the fetal aorta was aligned with the crux of the heart in the vertical plane. The reference dot was positioned in the aorta allowing visualization of the coronal view of the descending aorta in panel C. (c). In panel C the image was rotated to display the aorta in a vertical position, when necessary. This allowed visualization of the longitudinal view of the ductal arch in panel B. (d) Only three planes were selected, using the ‘Slices’ option, including the plane that crosses the reference dot (which is labeled with an asterisk in the software), one plane to the left (‘−1’) and one to the right (‘+1’) to the plane crossing the reference dot. These images were magnified using the four-panel ‘Display format’. (e) In panel A, the image was moved until the reference dot was positioned in the center of the aorta. The ‘Adjust’ option was selected to align the ‘−1’ plane with the ductal arch and the ‘+1’ plane with the external edge of the aorta. This allowed simultaneous visualization of the ductal arch in panel A, the three-vessel and trachea view in panel B, the five-chamber view in panel C, and the four-chamber view in panel D. (f) The ‘Rotation Y’ was selected by clicking on the bar, and the five-chamber view was rotated by scrolling on the y axis until the left outflow tract was visualized in panel B. This allowed simultaneous visualization of the short axis of the aorta in panel A, the three vessel and trachea view in panel B, the long axis of the left outflow tract on panel C, and the four-chamber view on panel D (reproduced with permission of the American Institute of Ultrasound in Medicine from reference 54).

206

Fetal Cardiology

(a)

(b)

Figure 14.7

(c)

the area under examination.3,22,56 Rendering techniques can also be used to optimize contrast of myocardial borders, septa, and valves, or to obtain realistic 4D images of particular structures of interest.

Three-dimensional (3D) multiplanar slicing of the aortic and ductal arches. (a) 3D multiplanar image of the fetal thorax in the original acquisition plane (sagittal);(b) The reference dot was manipulated in the right upper panel and moved to the center of the aorta (white arrow). The resulting image in the left upper panel was a sagittal view of the aorta. Minor adjustment of this image around the y axis was required to demonstrate the aortic arch. (c) To demonstrate the ductal arch, the right upper panel image was simply rotated counter-clockwise around the z axis (curved arrow) (reproduced from reference 22 with permission from Elsevier).

In Figure 14.9 and Video clip 14.7, ‘thick slice’ rendering was applied to the atrioventricular valves to visualize the leaflets en face, as if the examiner was observing them from the ventricular chambers.17,22 This view can also be

4D ultrasound examination of the fetal heart by spatiotemporal image correlation

(a)

(b)

(c)

(d)

Figure 14.8

(e)

207

Application of the spin technique18 to the identification of the abnormal vessel marked with ‘?’. (a) Three-vessel view showing the pulmonary artery (PA), the ascending aorta (Ao), the superior vena cava (SVC), the descending aorta (DAo) and an abnormal structure that is not commonly visualized to the left of the pulmonary artery in this view. (b) The reference dot was moved from the pulmonary artery to the center of the structure under interrogation. (c) The volume was rotated around the y axis; this maneuver opens up the vessel so that it can be visualized in the sagittal plane. (d) In order to know where this vessel was draining to, the reference dot was moved to the drainage site. (e) The volume was spun once more around the y axis, revealing that the drainage site was the coronary sinus (CS). This confirmed the diagnosis of a persistent left superior vena cava (PLSVC).

208

Fetal Cardiology

(a)

Figure 14.9 (b)

obtained by selecting the direction of rendering from the atrial chambers.57 Yagel et al57 were able to consistently obtain this rendered view in 93% of 136 normal fetuses examined between 21 and 26 weeks of gestational age. In five of 35 abnormal cases examined in their study, this rendered view provided additional diagnostic information regarding the relative position of the great vessels to the atrioventricular valves as well as the appearance of the semilunar valves. Figure 14.10 presents a comparison of

‘Thick slice’ rendering of the atrioventricular valves of a normal fetus. The key to obtain the rendered image (right lower corner) is the position and size of the region of interest (ROI), encompassing only the atrioventricular valves. The green line indicates the direction of view, which is the direction that the computer software will use to convert the voxels along the projection path to pixel information to be displayed on the twodimensional screen. In this case, the position of the green line indicates that the atrioventricular valves are being visualized from the ventricular chambers. This volume data set was rendered using a combination (mix) of 60% gradient light and 40% surface mode. MV, mitral valve; TV, tricuspid valve; LV, left ventricle; RV, right ventricle (reproduced with permission from reference 23).

atrioventricular valves visualized en face in a normal case, in a case of complete atrioventricular canal, and in a case of a hypoplastic tricuspid valve associated with pulmonary atresia. In Figure 14.11 and Video clip 14.8 we demonstrate the use of ‘thick slice’ rendering with inversion mode to emphasize the abnormal insertion of the tricuspid valve in a case of Ebstein’s anomaly.19 This technique has been reported as helpful in characterizing the hypokinetic

4D ultrasound examination of the fetal heart by spatiotemporal image correlation

209

Figure 14.10 Comparison of three-dimensional surface rendering of the atrioventricular valves: normal fetus (top), atrioventricular septal defect (middle), tricuspid stenosis (bottom). The bright echogenic spot in the center of the image of the atrioventricular septal defect case (green arrowhead, diastole), corresponds to the atrial septum secundum viewed from above. In the tricuspid stenosis with ventricular septal defect case, the tricuspid valve is very small with limited excursion during diastole. Loss of continuity of the interventricular septum is observed during systole (white arrowhead) (reproduced from reference 22 with permission from Elsevier).

motion of the ventricular wall in a case of congenital left ventricular aneurysm.58 Figure 14.12 and Video clip 14.9 show another example of ‘thick slice’ rendering to visualize an overriding aorta in a case of tetralogy of Fallot associated with absent pulmonary valve syndrome. Figure 14.13 and Video clip 14.10 show the pulmonary artery, which is constricted at the level of the pulmonary valves, the poststenotic dilatation, and, as a result of the depth perspective provided by rendering, a cross-section of the dilated left pulmonary artery is also visualized. This is a common finding in absent pulmonary valve syndrome associated with tetralogy of Fallot. Figure 14.14 shows rendered views of the atrial and ventricular septum in the sagittal plane using a technique described by Yagel et al.57 This view may contribute to further elucidation of atrial and ventricular septal defects.

Figure 14.11 ‘Thick slice’ rendering of the atrioventricular valves using inversion mode in a fetus with Ebstein’s anomaly. This mode provides great contrast for visualization of the myocardium and atrioventricular valves. Note the abnormal apical insertion of the tricuspid valve and the small right ventricle (reproduced with permission from reference 23).

Rendering techniques for visualization of the great vessels Several rendering algorithms have been described for visualization of the three-dimensional structure and spatial relationships of great vessels and venous return to the heart. These images can be obtained by acquiring volume data sets with color Doppler,17,24,36,59 power Doppler17,24,60 or B-flow imaging,19,25,28 as well as by rendering gray-scale

210

Fetal Cardiology

Figure 14.12 ‘Thick slice’ rendered image of the aorta (Ao) overriding the interventricular septum (IVS) in a case of absent pulmonary valve syndrome associated with tetralogy of Fallot. The algorithm used for rendering was the ‘gradient light’ mode (reproduced with permission from reference 23).

Figure 14.15 and Video clip 14.11 show the technique that we frequently use in our unit to visualize the crisscrossing of the outflow tracts as they leave the ventricular chambers. As described in the previous paragraph, volume data sets are acquired using the fourchamber view as the starting point. For the sake of reproducibility, we reorient the heart on the screen whenever necessary by rotating the volume dataset around the y and z axes until the four-chamber view is in the apical position, and the left side of the heart is displayed on the left side of the screen. Next, the rendering box is selected and adjusted in panel B to include the whole heart within the region of interest (in this case, from the heart base touching the diaphragm to the great vessels close to the neck). In order to visualize the great vessels leaving the heart, the user must set the direction of view to project the heart from the great vessels towards its base. In our example, the direction of view was set from left to right, using panel B as a reference. The resulting rendered image is shown in panel D. Figure 14.16 shows examples of the normal pulmonary artery crisscrossing over the aorta in volume data sets acquired with color Doppler (Figure 14.16a), power Doppler (Figure 14.16b), inversion mode (Figure 14.16c), and B-flow imaging (Figure 14.16d and Video clip 14.12). In Figure 14.17, the application of this technique for the diagnosis of transposition of the great arteries is demonstrated.

Inversion mode and B-flow

Figure 14.13 ‘Thick slice’ rendered image of the right ventricle (RV) and pulmonary artery (PA) in a case of absent pulmonary valve syndrome associated with tetralogy of Fallot. The stenotic pulmonary valve annulus (PVA), the poststenotic dilatation of the pulmonary artery, and a cross-section of the dilated annulus of the left pulmonary artery (LPA) are demonstrated in a single image. The algorithm used for rendering was the ‘gradient light’ mode (reproduced with permission from reference 23).

volume data sets with inversion mode.19,40,61,62 Crisscrossing of the pulmonary artery over the ascending aorta as these vessels leave the ventricular chambers is best visualized in volume data sets acquired using transverse sweeps through the fetal thorax. Sagittal acquisitions are preferred when the objective is to visualize the aortic and ductal arches.

Inversion of gray-scale voxels to visualize blood pools from cardiac structures was originally proposed by Nelson et al,40 in 1996. This principle has recently been incorporated into commercially available ultrasound equipment and is known as inversion mode. With inversion mode, anechoic structures such as the heart chambers, vessel lumen, stomach, gallbladder, renal pelvis, and bladder appear echogenic in the rendered images, whereas structures that are normally echogenic before gray-scale inversion (e.g. bones) appear anechoic. Postprocessing adjustments are performed as necessary, including gamma curve correction to optimize contrast resolution, and gray-scale threshold and transparency to improve image quality. The technique allows examiners to obtain 4D rendered images of cardiovascular structures from volume data sets acquired with gray-scale only, without the need for color Doppler, power Doppler, or B-flow imaging (Figure 14.18 and Video clip 14.13). An example of the applicability of this technique is provided in Figure 14.19, which illustrates the three-dimensional rendering of a case of interrupted IVC with azygos continuation. B-flow technology digitally enhances weak blood reflector signals from vessels and, at the same time, suppresses strong signals from the surrounding tissues.25,63

4D ultrasound examination of the fetal heart by spatiotemporal image correlation

211

(a)

(b)

Figure 14.14 Rendered images of the interventricular septum (IVS). (a) The interventricular septum is rendered with the direction of view (green line) set from the right ventricle (RV). (b) The direction of view is set so that the volume is reconstructed with the IVS visualized from the left ventricle (LV). Flap, flap of the foramen ovale.

Since this technology does not use Doppler methods to display blood flow, it is angle independent and does not interfere with frame rate as much as color or power Doppler.63–66 Due to its high sensitivity and angle independence to blood flow, B-flow is potentially advantageous over color or power Doppler imaging for the visualization of the great vessels and venous return to the heart,19,65 as well as for visualization of small vessels such as the coronary arteries67 and aortopulmonary collateral branches in cases of pulmonary atresia with a ventricular septal defect.28 In Figure 14.20 and Video clip 14.14, the aortic and ductal arches of a normal fetus have been

rendered using a volume data set acquired with B-flow imaging and the gradient light algorithm. The principles utilized to obtain these images are the same as described for Figures 14.15–14.17.

Limitations Factors that affect image quality of conventional twodimensional ultrasonography are likely to impact the quality of STIC volume data sets. These factors include

212

Fetal Cardiology

Figure 14.15 Technique to obtain rendered images of the outflow tracts using color Doppler. The rendering box is adjusted to encompass the heart and great vessels. The direction of view (green line) is set to project the rendered image from anterior (pulmonary artery) to posterior (aorta and ventricular chambers). The same technique can be applied to volume data sets acquired with power Doppler and B-flow imaging, as well as volume data sets acquired with B-mode imaging but rendered using inversion mode (reproduced with permission from reference 23).

(a)

(b)

(c)

(d)

Figure 14.16 Crisscrossing of the outflow tracts as they exit the ventricular chambers. The pulmonary artery always crosses in front of the aorta. The rendered images were obtained with the technique described in Figure 14.11 and are explained in detail in Video clip 14.11. The volume data sets were acquired with color Doppler (a), power Doppler (b), gray-scale (then rendered with inversion mode) (c), and B-flow (d) (reproduced with permission from reference 23).

4D ultrasound examination of the fetal heart by spatiotemporal image correlation

(a)

213

(b)

Figure 14.17 Rendered images from volume data sets of the fetal heart in a normal case (a) and transposition of the great arteries (b). The volumes were acquired through a transverse sweep of the fetal chest, using power Doppler 4D ultrasonography with STIC. In the normal fetus, normal crisscrossing of the great arteries is observed, whereas in transposition of the great arteries, the vessels leave the ventricles in parallel. The technique utilized to render the volume datasets is explained in Figure 14.14 (reproduced with permission of the American Institute of Ultrasound in Medicine from reference 20).

Figure 14.18 ‘Digital casts’ of the aortic and ductal arches obtained with inversion mode (reproduced with permission from reference 62).

early gestational age, unfavorable fetal position, and maternal obesity. Fetal movement and sudden changes in fetal heart rate (e.g. fetal arrhythmias) during acquisition are additional factors that may affect the technique and cause misregistration of the information required for precise reconstruction of moving cardiac structures.22 We have observed such an artifact in a case of a fetus with transposition of the great arteries, which was misinterpreted

during analysis of the four-dimensional volume data set as double outlet right ventricle. In this case, significant fetal movement during acquisition artificially shifted the connections of the outflow tracts to appear as exiting from the right ventricle.68 Therefore, we caution that, in the case of a suspected malalignment defect (e.g. overriding ventricular septal defect, tetralogy of Fallot, double outlet right ventricle, truncus arteriosus, pulmonary atresia

214

Fetal Cardiology

(a)

Figure 14.19

(b)

Three-dimensional images of a fetal heart rendered with inversion mode in a case of interrupted inferior vena cava with azygos vein continuation associated with omphalocele. (a) An image from the right side of the heart shows that the arch of the azygos vein joins the superior vena cava (SVC) before entering the right atrium. (b) Posterior view of the fetal heart shows a dilated azygos vein located to the right of the descending aorta. The arch of this vein forms a Y image with the aortic arch before joining the SVC (reproduced with permission from reference 62).

Conclusion In this chapter we describe a practical approach for the examination of the fetal heart by 3D/4D ultrasonography with STIC. This technology allows: (1) navigation through the volume data set and examination of the fetal heart in the absence of the patient; (2) the use of techniques to systematically visualize the outflow tracts in volume data sets acquired using the four-chamber view image as the starting point; (3) examination of the fetal heart using a tomographic approach similar to that used to read computed tomography and magnetic resonance imaging examinations; and (4) 3D and 4D rendering of cardiovascular structures to visualize the relationships, size, and course of the outflow tracts in normal fetuses and those with congenital heart disease. 3D and 4D rendering of the great vessels, in particular, has been previously possible only during postmortem examination by injection of silicone rubber to produce pathological casts of the cardiovascular system.25,30,69

Legends for the DVD Figure 14.20 Rendered image of the aortic and ductal arches obtained from a volume data set acquired with B-flow imaging, using a similar acquisition and rendering technique as illustrated in Figure 14.15. Ao arch, aortic arch, DA, ductus arteriosus (reproduced with permission from reference 23).

Video clip 14.1 Two-dimensional video clip of the four-chamber view, illustrating the placement of the region of interest (ROI) and acquisition of the volume data set with STIC. The ROI defines the width and the height of the volume data set (x and y planes).

Video clip 14.2 Standard planes of section that can be obtained by scrolling from the top to the bottom of the volume data set.

with a ventricular septal defect) that is observed first by volumetric imaging, the finding should be confirmed by conventional two-dimensional ultrasonography prior to establishing the final diagnosis.

Video clip 14.3 Tomographic ultrasound imaging of a normal volume data set of the fetal heart, acquired using transverse sweeps through the fetal thorax.

4D ultrasound examination of the fetal heart by spatiotemporal image correlation

215

Video clip 14.4

Video clip 14.13

Demonstration of the technique to systematically visualize the outflow tracts in volume data sets acquired using transverse sweeps throught the fetal thorax with STIC.

Technique to render the aortic and ductal arches using volume data sets acquired with gray scale imaging only and rendered with inversion mode.

Video clip 14.5

Video clip 14.14

Demonstration of transposition of the great arteries using the technique described in Figure 14.4 and Video clip 14.4.

Four-dimensional visualization of the aortic and ductal arches in a volume data set acquired with B-flow imaging.

Video clip 14.6 Demonstration of a technique to visualize the aortic and ductal arches in 4D volume data sets acquired using sagittal sweeps through the fetal chest (described originally by Bega et al55 for volume data sets acquired by 3DUS).

Video clip 14.7 ‘Thick slice’ rendering of the atrioventricular valves of a normal fetus. The key to obtain the rendered image (right lower corner) is the position and size of the region of interest, encompassing only the atrioventricular valves. The green line indicates the direction of view, which is the direction that the computer software will use to convert the voxels along the projection path to pixel information to be displayed on the two-dimensional screen. In this case, the position of the green line indicates that the atrioventricular valves are being visualized from the ventricular chambers. This volume data set was rendered using a combination (mix) of 60% gradient light and 40% surface mode.

Video clip 14.8 ‘Thick slice’ rendering of the atrioventricular valves using inversion mode in a fetus with Ebstein anomaly. This mode provides great contrast for the visualization of the myocardium and atrioventricular valves. Note the abnormal apical insertion of the tricuspid valve.

Video clip 14.9 ‘Thick slice’ rendering in a case of absent pulmonary valve syndrome associated with tetralogy of Fallot, demonstrating the overriding aorta. The algorithm used for rendering was the ‘gradient light’ mode.

Video clip 14.10 ‘Thick slice’ rendering of the right ventricle and pulmonary artery in a case of absent pulmonary valve syndrome associated with tetralogy of Fallot. The stenotic pulmonary valve annulus, the poststenotic dilatation of the pulmonary artery and a crosssection of the dilated annulus of the left pulmonary artery are demonstrated in a single image. Labels for this image are provided in Figure 14.9. The algorithm used for rendering was the ‘gradient light’ mode.

Video clip 14.11 Technique to obtain rendered images of the outflow tracts using color Doppler. The same technique can be applied to volume data sets acquired with power Doppler and B-flow imaging, as well as for volume data sets acquired with B-mode imaging but rendered using inversion mode. PA, pulmonary artery; Ao, aorta; LV, left ventricle; RV, right ventricle.

Video clip 14.12 Crisscrossing of the outflow tracts in a volume data set acquired with B-flow imaging.

References 1. International Society of Ultrasound in Obstetrics and Gynecology. Cardiac screening examination of the fetus: guidelines for performing the ‘basic’ and ‘extended basic’ cardiac scan. Ultrasound Obstet Gynecol 2006; 27: 107–13. 2. Lee W. Performance of the basic fetal cardiac ultrasound examination. J Ultrasound Med 1998; 17: 601–7. 3. Chaoui R, Heling KS. New developments in fetal heart scanning: three- and four-dimensional fetal echocardiography. Semin Fetal Neonatal Med 2005; 10: 567–77. 4. Deng J, Rodeck CH. Current applications of fetal cardiac imaging technology. Curr Opin Obstet Gynecol 2006; 18: 177–84. 5. DeVore GR, Falkensammer P, Sklansky MS, Platt LD. Spatio-temporal image correlation (STIC): new technology for evaluation of the fetal heart. Ultrasound Obstet Gynecol 2003; 22: 380–7. 6. DeVore GR, Medearis AL, Bear MB et al. Fetal echocardiography: factors that influence imaging of the fetal heart during the second trimester of pregnancy. J Ultrasound Med 1993; 12: 659–63. 7. Allan L, Benacerraf B, Copel JA et al. Isolated major congenital heart disease. Ultrasound Obstet Gynecol 2001; 17: 370–9. 8. Carvalho JS, Mavrides E, Shinebourne EA et al. Improving the effectiveness of routine prenatal screening for major congenital heart defects. Heart 2002; 88: 387–91. 9. Hunter S, Heads A, Wyllie J, Robson S. Prenatal diagnosis of congenital heart disease in the northern region of England: benefits of a training programme for obstetric ultrasonographers. Heart 2000; 84: 294–8. 10. Rustico MA, Benettoni A, D’Ottavio G et al. Early screening for fetal cardiac anomalies by transvaginal echocardiography in an unselected population: the role of operator experience. Ultrasound Obstet Gynecol 2000; 16: 614–19. 11. Tegnander E, Eik-Nes SH. The examiner’s ultrasound experience has a significant impact on the detection rate of congenital heart defects at the second-trimester fetal examination. Ultrasound Obstet Gynecol 2006; 28: 8–14. 12. Wong SF, Chan FY, Cincotta RB et al. Factors influencing the prenatal detection of structural congenital heart diseases. Ultrasound Obstet Gynecol 2003; 21: 19–25. 13. Bonnet D, Coltri A, Butera G et al. Detection of transposition of the great arteries in fetuses reduces neonatal morbidity and mortality. Circulation 1999; 99: 916–18. 14. Franklin O, Burch M, Manning N et al. Prenatal diagnosis of coarctation of the aorta improves survival and reduces morbidity. Heart 2002; 87: 67–9.

216

Fetal Cardiology

15. Tworetzky W, McElhinney DB, Reddy VM et al. Improved surgical outcome after fetal diagnosis of hypoplastic left heart syndrome. Circulation 2001; 103: 1269–73. 16. Tzifa A, Barker C, Tibby S, Simpson J. Prenatal diagnosis of pulmonary atresia: impact on clinical presentation and early outcome. Arch Dis Child Fetal Neonatal Ed 2007; 92: F199–203. 17. Chaoui R, Hoffmann J, Heling KS. Three-dimensional (3D) and 4D color Doppler fetal echocardiography using spatiotemporal image correlation (STIC). Ultrasound Obstet Gynecol 2004; 23: 535–45. 18. DeVore GR, Polanco B, Sklansky MS, Platt LD. The ‘spin’ technique: a new method for examination of the fetal outflow tracts using three-dimensional ultrasound. Ultrasound Obstet Gynecol 2004; 24: 72–82. 19. Gonçalves LF, Espinoza J, Lee W et al. A new approach to fetal echocardiography: digital casts of the fetal cardiac chambers and great vessels for detection of congenital heart disease. J Ultrasound Med 2005; 24: 415–24. 20. Gonçalves LF, Espinoza J, Romero R et al. A systematic approach to prenatal diagnosis of transposition of the great arteries using 4-dimensional ultrasonography with spatiotemporal image correlation. J Ultrasound Med 2004; 23: 1225–31. 21. Gonçalves LF, Espinoza J, Romero R et al. Four-dimensional fetal echocardiography with spatiotemporal image correlation (STIC): a systematic study of standard cardiac views assessed by different observers. J Matern Fetal Neonatal Med 2005; 17: 323–31. 22. Gonçalves LF, Lee W, Chaiworapongsa T et al. Fourdimensional ultrasonography of the fetal heart with spatiotemporal image correlation. Am J Obstet Gynecol 2003; 189: 1792–802. 23. Gonçalves LF, Lee W, Espinoza J, Romero R. Examination of the fetal heart by four-dimensional (4D) ultrasound with spatio-temporal image correlation (STIC). Ultrasound Obstet Gynecol 2006; 27: 336–48. 24. Gonçalves LF, Romero R, Espinoza J et al. Four-dimensional ultrasonography of the fetal heart using color Doppler spatiotemporal image correlation. J Ultrasound Med 2004; 23: 473–81. 25. Pooh RK, Korai A. B-flow and B-flow spatio-temporal image correlation in visualizing fetal cardiac blood flow. Croat Med J 2005; 46: 808–11. 26. Vinals F, Mandujano L, Vargas G, Giuliano A. Prenatal diagnosis of congenital heart disease using four-dimensional spatio-temporal image correlation (STIC) telemedicine via an Internet link: a pilot study. Ultrasound Obstet Gynecol 2005; 25: 25–31. 27. Vinals F, Poblete P, Giuliano A. Spatio-temporal image correlation (STIC): a new tool for the prenatal screening of congenital heart defects. Ultrasound Obstet Gynecol 2003; 22: 388–94. 28. Volpe P, Campobasso G, Stanziano A et al. Novel application of 4D sonography with B-flow imaging and spatiotemporal image correlation (STIC) in the assessment of the anatomy of pulmonary arteries in fetuses with pulmonary atresia and ventricular septal defect. Ultrasound Obstet Gynecol 2006; 28: 40–6. 29. Yagel S, Valsky DV, Messing B. Detailed assessment of fetal ventricular septal defect with 4D color Doppler ultra-

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

sound using spatio-temporal image correlation technology. Ultrasound Obstet Gynecol 2005; 25: 97–8. Cook AC, Fagg NL, Allan LD. Use of casts in the necropsy diagnosis of fetal congenital heart disease. Br Heart J 1992; 68: 481–4. Shih JC, Chen CP. Spatio-temporal image correlation (STIC): innovative 3D/4D technique for illustrating unique and independent information and diagnosing complex congenital heart diseases. Croat Med J 2005; 46: 812–20. Deng J. Terminology of three-dimensional and fourdimensional ultrasound imaging of the fetal heart and other moving body parts. Ultrasound Obstet Gynecol 2003; 22: 336–44. Brekke S, Tegnander E, Torp HG, Eik-Nes SH. Tissue Doppler gated (TDOG) dynamic three-dimensional ultrasound imaging of the fetal heart. Ultrasound Obstet Gynecol 2004; 24: 192–8. Deng J, Birkett AG, Kalache KD et al. Conversion of umbilical arterial Doppler waveforms to cardiac cycle triggering signals: a preparatory study for online motion-gated three-dimensional fetal echocardiography. Ultrasound Med Biol 2001; 27: 51–9. Deng J, Gardener JE, Rodeck CH, Lees WR. Fetal echocardiography in three and four dimensions. Ultrasound Med Biol 1996; 22: 979–86. Deng J, Richards R. Dynamic three-dimensional gray-scale and color Doppler ultrasound of the fetal heart for dynamic diagnosis. Ultrasound Obstet Gynecol 2002; 20: 209. Deng J, Ruff CF, Linney AD et al. Simultaneous use of two ultrasound scanners for motion-gated three-dimensional fetal echocardiography. Ultrasound Med Biol 2000; 26: 1021–32. Deng J, Yates R, Birkett AG et al. Online motion-gated dynamic three-dimensional echocardiography in the fetus—preliminary results. Ultrasound Med Biol 2001; 27: 43–50. Herberg U, Goldberg H, Breuer J. Dynamic free-hand threedimensional fetal echocardiography gated by cardiotocography. Ultrasound Obstet Gynecol 2003; 22: 493–502. Nelson TR, Pretorius DH, Sklansky M, Hagen-Ansert S. Three-dimensional echocardiographic evaluation of fetal heart anatomy and function: acquisition, analysis, and display. J Ultrasound Med 1996; 15: 1–9. Sklansky MS, Nelson TR, Pretorius DH. Usefulness of gated three-dimensional fetal echocardiography to reconstruct and display structures not visualized with two-dimensional imaging. Am J Cardiol 1997; 80: 665–8. Kovalchin JP, Lewin MB, Bezold LI et al. Harmonic imaging in fetal echocardiography. J Am Soc Echocardiogr 2001; 14: 1025–9. Paladini D, Vassallo M, Tartaglione A et al. The role of tissue harmonic imaging in fetal echocardiography. Ultrasound Obstet Gynecol 2004; 23: 159–64. Yoo SJ, Lee YH, Kim ES et al. Three-vessel view of the fetal upper mediastinum: an easy means of detecting abnormalities of the ventricular outflow tracts and great arteries during obstetric screening. Ultrasound Obstet Gynecol 1997; 9: 173–82. Yagel S, Cohen SM, Achiron R. Examination of the fetal heart by five short-axis views: a proposed screening method for comprehensive cardiac evaluation. Ultrasound Obstet Gynecol 2001; 17: 367–9.

4D ultrasound examination of the fetal heart by spatiotemporal image correlation

46. Vinals F, Ascenzo R, Poblete P et al. Simple approach to prenatal diagnosis of transposition of the great arteries. Ultrasound Obstet Gynecol 2006; 28: 22–5. 47. Leung KY, Ngai CS, Chan BC et al. Three-dimensional extended imaging: a new display modality for three-dimensional ultrasound examination. Ultrasound Obstet Gynecol 2005; 26: 244–51. 48. DeVore GR, Polanko B. Tomographic ultrasound imaging of the fetal heart: a new technique for identifying normal and abnormal cardiac anatomy. J Ultrasound Med 2005; 24: 1685–96. 49. Gonçalves LF, Espinoza J, Romero R et al. Four-dimensional ultrasonography of the fetal heart using a novel Tomographic Ultrasound Imaging display. J Perinat Med 2006; 34: 39–55. 50. Paladini D, Vassallo M, Sglavo G et al. The role of spatiotemporal image correlation (STIC) with tomographic ultrasound imaging (TUI) in the sequential analysis of fetal congenital heart disease. Ultrasound Obstet Gynecol 2006; 27: 555–61. 51. Chaoui R, McEwing R. Three cross-sectional planes for fetal color Doppler echocardiography. Ultrasound Obstet Gynecol 2003; 21: 81–93. 52. Abuhamad A. Automated multiplanar imaging: a novel approach to ultrasonography. J Ultrasound Med 2004; 23: 573–6. 53. Abuhamad A, Falkensammer P. Automated sonography: defining the spatial relationships of standard diagnostic fetal cardiac planes in the second trimester of pregnancy. Ultrasound Obstet Gynecol 2006; 28: 399. 54. Espinoza J, Kusanovic JP, Gonçalves LF et al. A novel algorithm for comprehensive fetal echocardiography using 4-dimensional ultrasonography and tomographic imaging. J Ultrasound Med 2006; 25: 947–56. 55. Bega G, Kuhlman K, Lev-Toaff A et al. Application of threedimensional ultrasonography in the evaluation of the fetal heart. J Ultrasound Med 2001; 20: 307–13. 56. Espinoza J, Gonçalves LF, Lee W et al. The use of the minimum projection mode in 4-dimensional examination of the fetal heart with spatiotemporal image correlation. J Ultrasound Med 2004; 23: 1337–48. 57. Yagel S, Benachi A, Bonnet D et al. Rendering in fetal cardiac scanning: the intracardiac septa and the coronal atrioventricular valve planes. Ultrasound Obstet Gynecol 2006; 28: 266–74.

217

58. Chiang YC, Yang CK, Shih JC, Lee CN. Prenatal diagnosis of congenital left ventricular aneurysm by fourdimensional ultrasonography with spatio-temporal image correlation (STIC). Ultrasound Obstet Gynecol 2006; 28: 345–7. 59. Deng J, Yates R, Sullivan ID et al. Dynamic threedimensional color Doppler ultrasound of human fetal intracardiac flow. Ultrasound Obstet Gynecol 2002; 20: 131–6. 60. Chaoui R, Kalache KD. Three-dimensional power Doppler ultrasound of the fetal great vessels. Ultrasound Obstet Gynecol 2001; 17: 455–6. 61. Espinoza J, Gonçalves LF, Lee W et al. A novel method to improve prenatal diagnosis of abnormal systemic venous connections using three- and four-dimensional ultrasonography and ‘inversion mode’. Ultrasound Obstet Gynecol 2005; 25: 428–34. 62. Gonçalves LF, Espinoza J, Lee W et al. Three- and fourdimensional reconstruction of the aortic and ductal arches using inversion mode: a new rendering algorithm for visualization of fluid-filled anatomical structures. Ultrasound Obstet Gynecol 2004; 24: 696–8. 63. Deane C. Extended field-of-view and B-flow ultrasound: fashion or future? Ultrasound Obstet Gynecol 2000; 15: 96–7. 64. Henri P, Tranquart F. [B-flow ultrasonographic imaging of circulating blood]. J Radiol 2000; 81: 465–7. [in French] 65. Wachsberg RH. B-flow, a non-Doppler technology for flow mapping: early experience in the abdomen. Ultrasound Q 2003; 19: 114–22. 66. Weskott HP. [B-flow—a new method for detecting blood flow]. Ultraschall Med 2000; 21: 59–65. [in German] 67. Gonçalves LF, Espinoza J, Kusanovic JP et al. Early visualization of the fetal coronary arteries by four-dimensional ultrasonography with B-flow imaging and spatiotemporal image correlation (STIC). Ultrasound Obstet Gynecol 2005; 26: 366. 68. Gonçalves LF, Nien JK, Espinoza J et al. What does 2-dimensional imaging add to 3- and 4-dimensional obstetric ultrasonography? J Ultrasound Med 2006; 25: 691–9. 69. Kilner PJ, Ho SY, Anderson RH. Cardiovascular cavities cast in silicone rubber as an adjunct to post-mortem examination of the heart. Int J Cardiol 1989; 22: 99–107.

15 Three- and four-dimensional ultrasound in fetal echocardiography: a new look at the fetal heart Simcha Yagel, Sarah M Cohen, Israel Shapiro, and Dan V Valsky Background Three- and four-dimensional (3D/4D) applications in fetal ultrasound scanning have made impressive strides in the past two decades, with particularly dramatic improvement in fetal echocardiography. Recent technological developments in motion-gated scanning allow almost real-time 3D/4D cardiac examination. It appears that 3D/4D ultrasound applications will make a significant contribution to our understanding of the developing fetal heart in both normal and anomalous cases, to interdisciplinary management team consultation, to parental counseling, and to professional training. 3D/4D echocardiography may facilitate screening methods, and with its offline networking capabilities may improve healthcare delivery systems by extending the benefits of prenatal cardiac screening to poorly served areas. The introduction of ‘virtual planes’ to fetal cardiac scanning has helped sonographers obtain views of the fetal heart not generally accessible with a standard 2D approach. There is insufficient evidence to determine whether 3D/4D cardiac scanning will improve the accuracy of fetal echocardiography screening programs. However, there is no doubt that 3D/4D gives us another look at the fetal heart. In this chapter we will summarize the 3D/4D acquisition and post-processing modalities in 3D/4D fetal echocardiography, demonstrating their use through normal and anomalous case examples.

Three- and four-dimensional techniques and their application to fetal cardiac scanning Acquisition modalities Spatiotemporal image correlation Spatiotemporal image correlation (STIC) acquisition is an indirect motion-gated offline scanning mode.1–5

The methodology of STIC is comprehensively described in Chapter 14 of this text. Briefly, automated volume acquisition is made possible by the array in the transducer performing a slow single sweep, recording a single 3D data set consisting of many 2D frames one behind the other. The volume of interest (VOI) is acquired over a period of 7.5 to about 30 seconds at a sweep angle of approximately 20–40° (depending on the size of the fetus) and frame rate of about 150 frames per second. A 10-second, 25° acquisition would contain 1500 B-mode images.4 Following acquisition, mathematical algorithms are applied to the volume data to detect systolic peaks that are used to calculate the fetal heart rate. The B-mode images are arranged in order according to their spatial and temporal domain, correlated to the internal trigger, the systolic peaks that define the heart cycle4 (Figure 15.1). The result is a reconstructed complete heart cycle that displays in an endless loop. This cine-like file of a beating fetal heart can be manipulated to display any acquired scanning plane at any stage in the cardiac cycle (Figure 15.2). This reconstruction takes place directly following the scan in a matter of seconds, allowing the STIC acquisition to be reviewed with the patient still present and repeated if necessary, and saved to the scanning machine, personal computer (PC), or network. STIC acquisition can be combined with other applications by selecting the appropriate setting before acquisition (B-flow, color and power Doppler, tissue Doppler, high definition flow Doppler) or with post-processing visualization modalities (3D volume rendering, VOCAL (virtual organ computer-aided analysis), inversion mode, tomographic ultrasound imaging (TUI)).

B-flow B-flow is an old–new technology that images blood flow without relying on Doppler shift; rather, B-flow is an outgrowth of B-mode imaging. With the advent of faster frame rates and computer processing, B-flow directly depicts blood cell reflectors. It avoids some of the pitfalls of Dop-

220

Fetal Cardiology

(a) 0 sec.

1 sec.

3 sec.

2 sec.

First cycle (total 4 seconds)

Second cycle

(b)

Time from beginning of cycle Total cycle duration: 4 seconds Frames acquired in first 2D slice

1 Contracting object scanned in three consecutive slices

1 2 3 2 Frames acquired in first 2D slice

Frames acquired in first 2D slice

Combining frames of identical phase in the cycle from consecutive slices frame’s spacial position is restored 3

0 sec.

1 sec.

2 sec.

3 sec.

Figure 15.1 Schematic demonstration of spatiotemporal image correlation (STIC) technology. Cycle duration, number of slices, and number of frames per slice were chosen to simplify illustration. The scale applicable to fetal cardiac examination is discussed in the text. (a) An object is contracting in a cyclical manner (4 seconds per cycle). The shape of the object is presented at four points during the cycle. Assume that the contraction rate is too high to scan the whole object in conventional real-time 3D. (b) The object is scanned in three consecutive slices adjacent to each other (1). At least one complete cycle is recorded in real-time 2D ultrasound, thus acquiring many frames per slice. In this example four frames are recorded in each slice (2). By simultaneous analysis of the tissue movements, the software identifies the beginning of each cycle and sets the time that each frame was acquired in respect of the beginning of the cycle. Knowing the time and position of each frame the software reconstructs the 3D shape of the complete object in each phase of the cycle (3). The shape is constructed from frames arranged side by side according to their position in the object (hence spatiotemporal). Though each frame composing the object was acquired in a different cycle, their phase in respect of the beginning of the cycle is identical (hence spatiotemporal).

pler, such as aliasing and signal drop-out at orthogonal scanning angles. The resulting image is a live gray-scale depiction of blood flow and part of the surrounding lumen, creating sensitive ‘digital casts’ of blood vessels and cardiac chambers (Figure 15.3 and Video clip 15.1). B-flow is also

a sensitive acquisition tool for volume measurement. B-flow modality is a direct volume non-gated scanning method able to show blood flow in the heart and great vessels in real-time, without color Doppler flow information.6 This makes it an invaluable tool in fetal echocardiography.

Three- and four-dimensional ultrasound in fetal echocardiography

221

(c) 0 sec.

1 sec.

2 sec.

3 sec.

Reconstructed cycle

Repetition of reconstructed cycle

(d)

Figure 15.1 Continued (c) The system completes its task by creating an endless loop animation composed of the consecutive reconstructed volumes of the cycle, resulting in a moving volume resembling real-time 3D. The procedure takes only a few seconds; the stored reconstructed volumes are now available for analysis with post-processing techniques as described in the text. (d) Schematic demonstration of the multiple slices through the heart acquired during a single STIC scan. The dedicated transducer automatically changes its scanning angle, either by means of a small motor in some systems, or electronically by using a phased matrix of elements. (Reproduced with permission from Yagel S, Cohen SM, Shapiro I, Valsky DV. 3D and 4D ultrasound in fetal cardiac scanning: a new look at the fetal heart. Ultrasound Obstet Gynecol 2007; 29: 81–95. © International Society of Ultrasound in Obstetrics and Gynecology. Permission is granted by John Wiley & Sons Ltd on behalf of ISUOG.)

3D/4D with color Doppler, 3D power Doppler (3DPD), and 3D high definition power flow Doppler Color and power Doppler have been extensively applied to fetal echocardiography; scanning is incomplete today without color Doppler. Color or power Doppler, and the most recent development, high definition flow Doppler, can be combined with static 3D direct volume non-gated scanning to obtain 3D volume files with two-color Doppler information or one-color 3DPD. Color Doppler can be used more effectively in 3D/4D when combined with STIC acquisition7 in fetal echocardiography, resulting in a volume file that reconstructs the cardiac cycle, as above, with color flow information. This joins the Doppler flow to cardiac events2 and provides all the advantages of analysis (multiplanar reconstruction (MPR) rendering, TUI) with color. This combination of modalities is very sensitive for detecting intracardiac Doppler flow signals throughout the cardiac cycle, for example mild tricuspid regurgitation occurring very early in systole or very briefly.8 Extreme care must be taken when working with Doppler applications in post-processing, however, to avoid misinterpretation of flow direction as the volume is rotated. 3DPD is directionless, one-color Doppler that is most effectively joined with static 3D scanning.2 3DPD uses

Doppler shift technology to reconstruct the blood vessels in the VOI, isolated from the rest of the volume. Using the ‘glass body’ mode in post-processing, surrounding tissue is not shown, while the vascular portion of the scan is isolated for evaluation. The operator can scroll spatially to any plane in the volume (but not temporally: in this case, color Doppler with STIC is more effective, see above). In 3DPD the vascular tree of the fetal abdomen and thorax is reconstructed,9,10 obviating the necessity of reconstructing a mental picture of the idiosyncratic course of an anomalous vessel from a series of 2D planes. This has been shown to aid our understanding of the normal and anomalous anatomy and pathophysiology of vascular lesions11 (Figure 15.4). High definition power flow Doppler, the newest development in color Doppler applications, uses high resolution and a small sample volume to produce images with two-color directional information, with less ‘blooming’ of color for more realistic representation of vessel size. It depicts flow at a lower velocity than color or power Doppler, while retaining the advantage of flow directional information, thereby combining high resolution bidirectional flow Doppler with the anatomic acuity associated with power Doppler. It can be used with static 3D or 4D gated acquisition (STIC) and the glass-body mode, to produce high-resolution images of the vascular tree with bidirectional color coding (Figure 15.5 and Video clip 15.2). It is particularly sensitive for imaging small vessels. Indeed,

222

Fetal Cardiology

Figure 15.2 The four-chamber view from a STIC acquisition in a third-trimester fetus in systole (a) and diastole (b). By applying multiplanar reconstruction (MPR) the operator optimizes the four-chamber view (FCV) plane, adjusting the image both spatially along the x, y, and z axes, and to the desired stage of the cardiac cycle. The navigation point is placed on the interventricular septum in the A-plane; the B-plane shows the septum ‘en face’, and the C-plane shows a coronal plane through the ventricles. (Reproduced with permission from Yagel S, Cohen SM, Shapiro I, Valsky DV. 3D and 4D ultrasound in fetal cardiac scanning: a new look at the fetal heart. Ultrasound Obstet Gynecol 2007; 29: 81–95. © International Society of Ultrasound in Obstetrics and Gynecology. Permission is granted by John Wiley & Sons Ltd on behalf of ISUOG.)

systolic and diastolic flow are observed at the same time owing to the sensitivity of the modality. For example, when used with STIC acquisition, the ductus venosus is shown to remain filled in both systole and diastole.

inaccessible in 2D cardiac scanning.14–17 These views, once obtained, are stored to the patient file, in addition to the original volume, either as static images or 4D motion files. Any of the stored information can be shared for expert review, interdisciplinary consultation, parental counseling, or teaching.

Post-processing modalities In post-processing, various methodologies have been proposed to optimize the acquired volumes to demonstrate the classic planes of fetal echocardiography12,13 (Figure 15.6), as well as ‘virtual planes’ that are generally

MPR, 3D rendering, and TUI 3D/4D volume sets contain a ‘block’ of information; this is generally a wedge-shaped chunk of the targeted area.

Three- and four-dimensional ultrasound in fetal echocardiography

223

Figure 15.3 B-flow image of normal heart and aortic arch. Brachiocephalic trunk (BT), left common carotid (LCC), and left subclavian artery (LSA) are seen projecting from the aortic arch (AoA). Inferior vena cava (IVC) is indicated. See also corresponding Video clip 15.1. (Reproduced with permission from Yagel S, Cohen SM, Shapiro I, Valsky DV. 3D and 4D ultrasound in fetal cardiac scanning: a new look at the fetal heart. Ultrasound Obstet Gynecol 2007; 29: 81–95. © International Society of Ultrasound in Obstetrics and Gynecology. Permission is granted by John Wiley & Sons Ltd on behalf of ISUOG.)

In order to analyze this effectively, the operator displays 2D planes in either MPR mode (Figure 15.2) or 3D volume rendering. In MPR the screen is divided into four frames, referred to as A (upper left), B, and C; the fourth frame (lower right) will show either the volume model for reference, or the rendered image. Each of the three frames shows one of the three orthogonal planes of the volume. The reference dot guides the operator in navigating within the volume, as it is anchored at the point of intersection of all three planes. By moving the point the operator manipulates the volume to display any plane within the volume; if temporal information was acquired, the same plane can be displayed at any stage of the scanned cycle. From a good STIC acquisition5 the operator can scroll through the acquired volume to obtain sequentially each of the classic five planes12 of fetal echocardiography, and any plane may be viewed at any time-point throughout the reconstructed cardiac cycle loop. The cycle can be run or stopped ‘frame-by-frame’ to allow examination of all phases of the cardiac cycle, for example opening and closing of the atrioventricular valves.

Figure 15.4 3D power Doppler of the heart and major vessels. Noted are the carotid artery (CA); aorta (AO); inferior vena cava (IVC); ductus venosus (DV); and umbilical vein (UV). (Reproduced with permission from Yagel S, Cohen SM, Shapiro I, Valsky DV. 3D and 4D ultrasound in fetal cardiac scanning: a new look at the fetal heart. Ultrasound Obstet Gynecol 2007; 29: 81–95. © International Society of Ultrasound in Obstetrics and Gynecology. Permission is granted by John Wiley & Sons Ltd on behalf of ISUOG.)

By comparing the A- and B-frames of the MPR display, the operator can view complex cardiac anatomy in corresponding transverse and longitudinal planes simultaneously. So, for example, an anomalous vessel that might be disregarded in cross-section is confirmed in the longitudinal plane. 3D rendering is another analysis capability of an acquired volume. It is familiar from static 3D applications, such as imaging the fetal face in surface rendering mode. In fetal echo it is readily applied to 4D scanning. The operator places a bounding box around the region of interest within the volume (after arriving at the desired plane and time) to show a slice of the volume whose depth reflects the thickness of the slice. For example, with the A-frame showing a good four-chamber view, the operator places the bounding box tightly around the interventricular septum. The rendered image in the D-frame will show an ‘en face’ view of the septum. The operator can determine whether the plane will be displayed from the left or right, i.e. the septum from within the left or the right ventricle; the thickness of the slice will determine the depth

224

Fetal Cardiology

operator-chosen angle of rotation, 6, 9, 15, or 30%. Setting the rotation angle at 15°, for example, results in 12 planes available for measurement. The computer mouse is used to manually define the contours of the measured object (for example a heart ventricle) at each plane serially. Alternatively, the operator can opt for the system to draw the contours automatically, according to varying degrees of sensitivity. Once an outline is drawn around each plane of the target, the system reconstructs a contour model of the target. This post-processing modality has been applied to volume calculation of numerous fetal organs, including heart, lungs, and others.18–20

Inversion mode

Figure 15.5 Normal heart and great vessels: STIC acquisition with high definition power flow Doppler. See also corresponding Video clip 15.2. DV, ductus venosus; UV, umbilical vein; IVC inferior vena cava; CA, celiac artery; SMA, superior mesenteric artery; PV, pulmonary veins; dAo, descending aorta. (Reproduced with permission from Yagel S, Cohen SM, Shapiro I, Valsky DV. 3D and 4D ultrasound in fetal cardiac scanning: a new look at the fetal heart. Ultrasound Obstet Gynecol 2007; 29: 81–95. © International Society of Ultrasound in Obstetrics and Gynecology. Permission is granted by John Wiley & Sons Ltd on behalf of ISUOG.)

of the final image, for example to show texture of the trabeculations within the right ventricle (Figure 15.7). TUI is a more recent application that extends the capabilities of MPR and rendering modes. This multislice analysis mode resembles a magnetic resonance imaging or computer-assisted tomography display. Nine parallel slices are displayed simultaneously from the plane of interest (the ‘zero’ plane), giving sequential views from −4 through +4. The thickness of the slices, i.e. the distance of one plane to the next, can be adjusted by the operator. The upper left frame of the display shows the position of each plane within the region of interest, relative to the reference plane. This application has the advantage of displaying sequential parallel planes simultaneously, giving a more complete picture of the fetal heart (Figure 15.8). VOCAL mode is a semiautomated 3D measurement mode that performs rotational measurement of volume. The saved volume file is rotated 180° about a fixed central axis though a preset number of rotation steps based on the

Inversion mode (IM) is another post-processing visualization modality that can be combined with static 3D or STIC acquisition.21,22 IM analyzes the echogenicity of tissue (white) and fluid-filled (black) pixels in a volume and inverts their presentation, i.e. fluid-filled spaces such as the cardiac chambers now appear white, while the myocardium has disappeared. In fetal echocardiography it can be applied to create ‘digital casts’23 of the cardiac chambers and vessels. Or, it can produce a reconstruction of the extracardiac vascular tree, similar to 3DPD. IM has the additional advantage of showing the stomach and gallbladder as white structures, which can aid the operator in navigating within a complex anomaly scan. IM can be joined with STIC and VOCAL to quantify fetal cardiac ventricular volumes, which may prove useful in the evaluation of fetal heart function.20

Screening examination of the fetal heart with three- and four-dimensional ultrasound Guidelines Guidelines for the performance of fetal heart examination have been published by the International Society of Ultrasound in Obstetrics and Gynecology (ISUOG).24 These guidelines for ‘basic’ and ‘extended basic’ fetal cardiac scanning can incorporate 3D/4D applications; conversely, 3D/4D ultrasound can enhance ‘basic’ and ‘extended basic’ fetal cardiac scans, as well as evaluation of congenital anomalies. Many research teams have applied 3D ultrasound and STIC acquisition to fetal echocardiography, and various techniques have been put forward to optimize the use of this modality. A well-executed STIC acquisition5 contains all the necessary planes for evaluation of the five classic transverse planes of fetal echocardiography.12,13 The operator can

Three- and four-dimensional ultrasound in fetal echocardiography

LT

225

RT

P (P)Ao

SVC

DA V

T

(D)Ao

LT

RT

MPA

IV LPA

T

SVC

RA

AO RPA

AO

DA

AO PA

LT

III

RT LV

RV RA LA AO

LT

II

RT RV

LV

RA LA PV

AO

I

LT

FO PV

RT

Figure 15.6 The five planes of fetal echocardiography (reproduced with permission, from reference 6). See Chapter 13, Figure 13.1 for full explanation.

ST LI AO SP

examine the fetal upper abdomen and stomach, then scroll cephalad to obtain the familiar four-chamber view, the five-chamber view, the bifurcation of the pulmonary arteries, and finally the three-vessel and trachea view. Slight adjustment along the x or y axes may be necessary to optimize the images. Performed properly, this methodology will provide the examiner with all the necessary planes to conform to the guidelines, above. However, it must be remembered that STIC acquisition that was degraded by maternal or fetal movements, including fetal breathing movements, will contain artifacts within the scan volume.

Applications Among the most attractive facets of 3D/4D scanning are the potential for digital archiving and sharing of examination data over a network. These capabilities were applied by Vinals and colleagues to increase delivery of prenatal cardiac scanning to poorly served areas. Local practitioners in distant areas acquired and stored 3D volume sets at their centers; they were subsequently sent over an internet link and analyzed by expert examiners in central locations.25,26 This can have important implications in

226

Fetal Cardiology

Figure 15.7 Normal interventricular septum in 3D rendering mode. In frame A (a) the bounding box is placed tightly around the septum with the active side (green line) on the right. The D-frame (b) shows the septum ‘en face’: note the rough appearance of the septum from within the trabeculated right ventricle. (Reproduced with permission from Yagel S, Cohen SM, Shapiro I, Valsky DV. 3D and 4D ultrasound in fetal cardiac scanning: a new look at the fetal heart. Ultrasound Obstet Gynecol 2007; 29: 81–95. © International Society of Ultrasound in Obstetrics and Gynecology. Permission is granted by John Wiley & Sons Ltd on behalf of ISUOG.)

increasing the penetration of prenatal ultrasound services in poorly served or outlying areas of many countries. DeVore and colleagues presented the ‘spin’ technique,14 combining MPR and STIC acquisition to analyze acquired volumes and simplify demonstration of the ventricular outflow tracts. Using this technique, the operator acquires a VOI from a transverse sweep of the fetal mediastinum that includes the sequential planes of fetal echocardiography. In post-processing, the outflow tract view is imaged in the A plane, and outflow tract and adjacent vessels are then examined by placing the reference point over each vessel and rotating the image along the x and y axes until the full length of each vessel is identified.14 Abuhamad proposed an automated approach to extract the required planes from an acquired volume, coining the term ‘automated multiplanar imaging’ or AMI.15 Based on the idea that the scanned 3D volume contains all possible planes of the scanned organ, it should be possible to define the geometric planes within that volume that would be required to display each of the diagnostic planes of a given organ, for example the sequential scanning planes of fetal echocardiography. Beginning from the four-chamber view, all the other planes are in constant anatomic relationship to this plane, and a computer-automated program could present those planes once the appropriate volume block is acquired.15 Most recently, Espinoza and colleagues introduced a novel algorithm combining STIC and TUI16 to image the diagnostic planes of the fetal heart simultaneously, and

facilitate visualization of the long-axis view of the aortic arch. However, as with any post-processing technique, if the original volume was suboptimal, subsequent analysis will be prone to lower image quality and the introduction of artifacts. Nuchal translucency screening programs will refer approximately 3–5% of patients for fetal echocardiography as high-risk,27,28 increasing demand for fetal cardiac screening programs. STIC acquisition is amenable to younger gestational ages, as the smaller fetal heart can be scanned in a shorter acquisition time, thus reducing the chance of acquisition degradation from fetal movements.

Functional evaluation of the fetal heart: ventricular volumetry Evaluation of fetal heart functional parameters has long challenged fetal echocardiographers. While duplex and color Doppler flow nomograms have been quantified and are long-established in 2D fetal echo, many of the pediatric and adult measures are based on end-systolic and end-diastolic ventricular volumes: stroke volume, ejection fraction, and cardiac output. Without electrical trace or clinically applicable segmentation methods to determine the ventricular volume, these parameters have eluded practical prenatal quantification. 3D ultrasound opens new avenues for

Three- and four-dimensional ultrasound in fetal echocardiography

227

Figure 15.8

Tomographic ultrasound imaging (TUI): the −4 plane (top row, center) shows the FCV while the zero plane (asterisk, middle row, right) shows the outflow tract view and the +3 plane shows the great vessels (bottom, right). (Reproduced with permission from Yagel S, Cohen SM, Shapiro I, Valsky DV. 3D and 4D ultrasound in fetal cardiac scanning: a new look at the fetal heart. Ultrasound Obstet Gynecol 2007; 29: 81–95. © International Society of Ultrasound in Obstetrics and Gynecology. Permission is granted by John Wiley & Sons Ltd on behalf of ISUOG.)

exploration into ventricular volumetry20,29,30 and mass measurement. Bhat and colleagues used non-gated static 3D acquisition and STIC to obtain mid-diastolic scans of fetal hearts, and applied VOCAL analysis to determine cavity volume. The result was multiplied by myocardial density (1.050 g/cm3) to obtain the mass.31,32 We recently published20 a methodology combining STIC acquisition to determine the end-systolic and enddiastolic stages in the cardiac cycle, then inversion mode to isolate the fluid-filled ventricular volume, which was measured using VOCAL analysis (Figure 15.9 and Video clip 15.3). The resulting volumes allowed quantification of stroke volume and ejection fraction.20 It was found that both the inversion mode and VOCAL analysis were highly dependent on operator-determined threshold parameters, which affect the intensity of signal to be colored and

included in the volumetry. A similar study of cardiac mass is under way.

Three- and four-dimensional ultrasound in the diagnosis of congenital heart disease One of the great advantages of a 3D/4D system is its digital archiving capabilities. Examination volumes are stored for later analysis, away from the patient and the time constraints of a busy clinic. In cases of congenital heart disease (CHD), other professionals can be invited to view the examination. They can do this anywhere that an

228

Fetal Cardiology

Figure 15.9 STIC acquisition combined with inversion mode (IM) and virtual organ computer-aided analysis (VOCAL) for fetal cardiac ventricle volumetry. The resulting measurements appear in the box, bottom right. See also corresponding Video clip 15.3. (Reproduced with permission from Yagel S, Cohen SM, Shapiro I, Valsky DV. 3D and 4D ultrasound in fetal cardiac scanning: a new look at the fetal heart. Ultrasound Obstet Gynecol 2007; 29: 81–95. © International Society of Ultrasound in Obstetrics and Gynecology. Permission is granted by John Wiley & Sons Ltd on behalf of ISUOG.)

internet link is available. The first examiner can consult with the attending physician, cardiologist, surgical or other management teams, genetic counselors, and parents. Complex malformations can be elucidated through interdisciplinary discussion and made clear to laymen. In addition, stored data from cases of CHD are invaluable teaching materials for professional education. Many teams have applied 3D/4D ultrasound capabilities to the diagnosis of congenital cardiovascular malformations. Each of the modalities and applications described above lends itself to different facets of this complex endeavor.

Virtual planes As described above, a properly executed STIC acquisition results in a volume ‘block’, which is a reconstructed complete cardiac cycle. This block of spatial and temporal

image data contains and makes available many scanning planes that are not readily accessible in 2D ultrasound. The term ‘virtual planes’ was coined to refer to these rendered scanning planes. The interventricular and interatrial septa (IVS, IAS) planes, and the coronal atrioventricular (CAV) plane of the cardiac valves’ annuli, have been investigated and applied to the evaluation of CHD.17 They were shown to have added value in the diagnosis of ventricular septal defect, restrictive foramen ovale, alignment of the ventricles and great vessels, and evaluation of the atrioventricular (AV) valves.

Segmental approach The segmental approach to congenital heart disease has helped to standardize the description of cardiac lesions. In addition, it has contributed to understanding the pathophysiology of the malformed developing fetal heart,

Three- and four-dimensional ultrasound in fetal echocardiography

229

and subsequently to our conceptualization and diagnostic imaging. The sequential segmental approach essentially divides the heart into three basic segments: the atria, the ventricles, and the great arteries. These are divided and joined at the level of the atrioventricular valves, and at the ventriculoarterial junctions. The segmental approach to diagnosis of CHD is comprehensively and concisely described elsewhere;33 we will follow this sequence in describing the application and added value of 3D/4D in the diagnosis of CHD, through index cases of anomalies diagnosed in our center.

Veins and atria: total anomalous pulmonary venous connection and interrupted inferior vena cava with azygos continuation Total anomalous pulmonary venous connection (TAPVC) is a many-faceted group of malformations affecting the pulmonary veins; the variations and classification are described in detail in Chapter 29.34 Essentially, in these anomalies the pulmonary veins do not drain into the left atrium, but rather to various other locations: the right atrium, great veins, or abdominal veins. The present case was an intradiaphragmatic variation with drainage of the pulmonary veins to the portal vein. 3D/4D ultrasound can have a significant contribution to the understanding of the fetal venous system. Figure 15.10a and Video clip 15.3 show the use of MPR with the reference point to navigate this complex lesion. Placement of the reference point in the suspected anomalous blood vessel in cross-section (A-frame) showed the vessel in longitudinal plane in the B-frame. This confirmed that the finding was not an artifact, but rather the characteristic vertical vein. 3D power flow Doppler displayed the idiosyncratic vascular tree and absence of the pulmonary veins (Figure 15.10b and Video clip 15.4); rotation of the image in post-processing allowed overall examination of the lesion in 360°. Interrupted inferior vena cava (IVC) with azygos continuation is shown in Figure 15.11 and Video clip 15.5. This cardinal vein anomaly results from primary failure of the right subcardinal vein to connect to the hepatic segment of the IVC.35 Blood is shunted directly into the right supracardinal vein (which will become the superior vena cava (SVC)) blood from the lower body flows through the azygos vein to the SVC. In this instance, B-flow acquisition provided real-time representation of the anomalous course of the IVC and connection to the fetal heart. It showed the azygos vein draining into the SVC, as well as the aorta in one three-dimensional image that would be impossible to obtain with 2D color Doppler scanning. B-flow scanning provided superior imaging of the

Figure 15.10 (a) STIC acquisition in a case of total anomalous pulmonary venous connection (TAPVC). The A-plane shown raised suspicion of an anomalous vessel (caret), which is confirmed in the B-plane (arrow). (b) The heart and great vessels of this fetus: STIC acquisition and high definition power flow Doppler confirmed the characteristic vertical vein (VV). Note also the absence of pulmonary veins (compare Figure 15.7). See also corresponding Video clips 15.3 and 15.4. (Reproduced with permission from Yagel S, Cohen SM, Shapiro I, Valsky DV. 3D and 4D ultrasound in fetal cardiac scanning: a new look at the fetal heart. Ultrasound Obstet Gynecol 2007; 29: 81–95. © International Society of Ultrasound in Obstetrics and Gynecology. Permission is granted by John Wiley & Sons Ltd on behalf of ISUOG.)

slower blood flow in the azygos vein than was demonstrated with 3DPD.

Atrioventricular junction: atrioventricular septal defect and tricuspid valve stenosis Atrioventricular septal defect (AVSD) is characterized by incomplete atrial and ventricular septation, forming a common atrioventricular junction. AVSD has many forms; all involve an abnormality of the AV valves. Figure 15.12 shows the use of 3D rendering of a STIC volume acquired

230

Fetal Cardiology

Figure 15.12 Figure 15.11 B-flow image of the heart and great vessels in a fetus with interrupted inferior vena cava with azygos continuation. See also corresponding Video clip 15.5. AoA, ascending aorta; AzV, azygos vein; SVC, superior vena cava; DV, ductus venosus. (Reproduced with permission from Yagel S, Cohen SM, Shapiro I, Valsky DV. 3D and 4D ultrasound in fetal cardiac scanning: a new look at the fetal heart. Ultrasound Obstet Gynecol 2007; 29: 81–95. © International Society of Ultrasound in Obstetrics and Gynecology. Permission is granted by John Wiley & Sons Ltd on behalf of ISUOG.)

The coronal atrioventricular (CAV) plane from STIC acquisition with color Doppler mapping in a case of atriventricular septal defect (AVSD). PA, pulmonary artery; AO, aorta; AVSD, atrioventricular septal defect. (Reproduced with permission from Yagel S, Cohen SM, Shapiro I, Valsky DV. 3D and 4D ultrasound in fetal cardiac scanning: a new look at the fetal heart. Ultrasound Obstet Gynecol 2007; 29: 81–95. © International Society of Ultrasound in Obstetrics and Gynecology. Permission is granted by John Wiley & Sons Ltd on behalf of ISUOG.)

with color Doppler to demonstrate the anomalous intracardiac flow resulting from the AVSD. Another group of AV valve lesions is mitral or tricuspid valve atresia, dysplasia, or stenosis. Figure 15.13 shows the CAV plane in a case of tricuspid stenosis. This ‘virtual plane’ is obtained from a STIC volume with color Doppler, by placing the bounding box tightly around the level of the AV connection in the four-chamber view, with the superior side active (frame A); the plane is slightly adjusted along the x and y axes; the rendered image (frame D) shows the AV valves with anomalous anatomy (compare normal CAV plane, inset). This virtual plane provides a three-dimensional look at the AV and semilunar valves’ annuli, resembling the surgical plane seen when the heart is opened in surgery.

are described in Chapter 19.36 Several groups have proposed methods for evaluation of the interventricular septum.37,38 By using MPR, with the reference point placed on the septum with the four-chamber view in the A-frame, the B-frame will show the septum and defect ‘en face’ (Figure 15.14 and Video clip 15.6). We recommend, however, the use of the bounding box in 3D rendering from STIC acquisition with color Doppler. The operator places the ‘active’ side of the box to the right or left (i.e. from within the left or right ventricle) and obtains an image (in the D–frame) having more depth, for a more detailed examination of the size and nature (and number) of the VSD(s). The addition of color Doppler demonstrates blood flow across the lesion and shows at what stage in the cardiac cycle and to what degree shunting occurs.

Ventricles: ventricular septal defects

Ventriculoarterial junctions (conotruncal anomalies): transposition of the great arteries

Ventricular septal defects are perhaps the most common – and most commonly missed – congenital heart defect. The natural history and in utero development of these lesions

Transposition (or malposition or malalignment) of the great arteries (TGA) is the general name for a complex

Three- and four-dimensional ultrasound in fetal echocardiography

231

Figure 15.13 Tricuspid stenosis evaluated with 3D rendering and the CAV plane. The bounding box is placed tightly around the level of the AV valves in the A-frame (a); the D-frame (b) clearly shows the stenotic valve (arrow). Compare normal CAV plane in diastole, inset: mv, mitral valve annulus; tv, tricuspid valve annulus; ao, aortic valve; pa, pulmonary valve. (Reproduced with permission from Yagel S, Cohen SM, Shapiro I, Valsky DV. 3D and 4D ultrasound in fetal cardiac scanning: a new look at the fetal heart. Ultrasound Obstet Gynecol 2007; 29: 81–95. © International Society of Ultrasound in Obstetrics and Gynecology. Permission is granted by John Wiley & Sons Ltd on behalf of ISUOG.)

Figure 15.14 The interventricular septum (IVS) ‘virtual plane’ with color Doppler in evaluation of ventricular septal defect (VSD). The navigation point is placed on the septum in the A-plane (a); the D-frame (b) shows the rendered IVS with flow across the defect from right to left. See also corresponding Video clip 15.6. (Reproduced with permission from Yagel S, Cohen SM, Shapiro I, Valsky DV. 3D and 4D ultrasound in fetal cardiac scanning: a new look at the fetal heart. Ultrasound Obstet Gynecol 2007; 29: 81–95. © International Society of Ultrasound in Obstetrics and Gynecology. Permission is granted by John Wiley & Sons Ltd on behalf of ISUOG.)

group of anomalies with widely varying anatomic and clinical presentations. When the sequential segmental approach is applied to systematic diagnosis of congenital heart disease,33 the morphology of each successive anatomic segment is assessed in turn. The morphologic right and left atria and ventricles are established; now the examiner addresses the ventriculoarterial junction and the accordance or discordance of the great arteries and ventricles. 3D rendering with color Doppler was applied to the evaluation of suspected malalignment of the great vessels,

by examining the CAV (‘surgical plane’) at the level of the AV and semilunar valves’ annuli. We applied B-flow scanning to the evaluation of TGA and found that it was more effective than 3DPD or inversion mode in visualizing the great vessels’ structure and relationships. Figure 15.15 and Video clip 15.7 show a case of complete dextrotransposition of the great arteries. The B-flow scan clearly showed blood flow into the ventricles and out through the malaligned vessels. This demonstration of the anatomic variant of the anomaly

232

Fetal Cardiology

Figure 15.16 Figure 15.15 B-flow modality showed the parallel great vessels in a case of transposition. Application of this modality clearly shows the blood flow in the malaligned vessels. See also corresponding Video clip 15.7. PA, pulmonary artery; AO, aorta. (Reproduced with permission from Yagel S, Cohen SM, Shapiro I, valsky DV. 3D and 4D ultrasound in fetal cardiac scanning: a new look at the fetal heart. Ultrasound obstet Gynecol 2007; 29: 81–95. © International Society of Ultrasound in obstetrics and Gynecology. Permission is granted by John Wiley & Sons Ltd on behalf of ISUOG.)

aided our consultations with the parents and their attending physician.

Arterial trunks: pulmonary stenosis and right aortic arch The use of 3D rendering of a STIC acquisition with or without color Doppler to obtain virtual planes is discussed above. The CAV plane is an excellent tool for evaluation of the semilunar valves. Once the CAV plane is obtained, the 4D-cine option is initiated and blood flow across the valves evaluated through the cardiac cycle. Figure 15.16 shows a case of critical pulmonary stenosis with retrograde flow in the main pulmonary artery (MPA). Right aortic arch defect results from persistence of the right dorsal aorta and involution of the distal part of the left dorsal aorta. There are two main types, with or without a retroesophageal component.35 Figure 15.17 shows a case of right aortic arch diagnosed with B–flow; this modality showed the characteristic course of the aortic arch to the right of the trachea.

The CAV plane from STIC acquisition with color Doppler mapping in a case of transposition of the great arteries and pulmonary stenosis with retrograde flow in the main pulmonary artery. AO, aorta; PA, pulmonary artery; T, tricuspid annulus; M, mitral annulus. (Reproduced with permission from Yagel S, Cohen SM, Shapiro I, Valsky DV. 3D and 4D ultrasound in fetal cardiac scanning: a new look at the fetal heart. Ultrasound Obstet Gynecol 2007; 29: 81–95. © International Society of Ultrasound in Obstetrics and Gynecology. Permission is granted by John Wiley & Sons Ltd on behalf of ISUOG.)

Functional evaluation: ventricular volumes We recently published20 novel methodology combining STIC acquisition with post-processing application of inversion mode and VOCAL to quantify end-systolic and end-diastolic ventricular volumes. Nomograms were created from right and left ventricle end-systolic and end-diastolic volumes from 100 fetuses examined between 20 and 40 gestational weeks. The resulting measurements correlated strongly with gestational age and estimated fetal weight. The measured volumes were used to create nomograms for fetal stroke volume and cardiac ejection fraction. The methodology was applied to saved STIC volumes of cases with cardiac anomaly or dysfunction that involved changes in ventricular volume, stroke volume, or ejection fraction. These included critical pulmonary stenosis, twin-to-twin transfusion syndrome with secondary pulmonary stenosis, aortic valve stenosis with hypoplastic aortic arch, Ebstein’s anomaly, supraventricular tachycardia, and vein of Galen aneurysm.20 Our normal cases showed the effectiveness of fetal heart ventricle volumetry in cardiac evaluation and quantification; such volumetry is not readily available

Three- and four-dimensional ultrasound in fetal echocardiography

233

STIC acquisition quality The quality of a STIC acquisition may be adversely affected by fetal body or ‘breathing’ movements. To improve scan quality, the fetus should be in a quiet state and the shortest scan time possible employed. When reviewing a STIC acquisition, the B-frame reveals artifacts introduced by fetal breathing movements (Figure 15.18). If the B-frame appears sound, the volume is usually acceptable, and can be used for further investigation. The quality of the original acquisition impacts on all further stages of post– processing and evaluation.

Original angle of insonation The original angle at which a scan was performed will impact on the quality of all the planes acquired. It is important to achieve an optimal beginning 2D plane, before starting 3D or 4D acquisition.

Figure 15.17 B-flow modality in a case of right aortic arch (RAoA). MPA, main pulmonary artery; DA, ductus arteriosus. (Reproduced with permission from Yagel S, Cohen SM, Shapiro I, Valsky DV. 3D and 4D ultrasound in fetal cardiac scanning: a new look at the fetal heart. Ultrasound Obstet Gynecol 2007; 29: 81–95. © International Society of Ultrasound in Obstetrics and Gynecology. Permission is granted by John Wiley & Sons Ltd on behalf of ISUOG.)

in 2D echocardiography. The pathological cases showed the potential added value of this methodology. In the case of critical pulmonary stenosis, for example, the diagnosis was more serious than suspected by 2D echocardiography. Ventricle volumetry also provided insight into the pathophysiology of lesions such as supraventricular tachycardia (SVT) and vein of Galen aneurysm, among others.20

Acoustic shadows Shadowing artifacts pose a particular problem to 3D/4D ultrasound. When commencing scanning from the 2D plane, acoustic shadows may not be apparent. However, they may be present within the acquired volume block. It is imperative to review suspected defects with repeated 2D and 3D scanning to confirm their presence in additional scanning planes.

Three-dimensional rendering 3D rendering creates virtual images. Application of some algorithms designed to smooth the image can lead to loss of data from the original scan. 3D rendering should always be used in conjunction with the A-frame 2D image for comparison.

Flow direction Potential pitfalls of threeand four-dimensional echocardiography 3D/4D fetal echocardiography scanning is prone to artifacts similar to those encountered in 2D ultrasound, and some that are specific to 3D/4D acquisition and post-processing.

An acquired volume containing Doppler flow information is available for manipulation and may be sliced and rotated around the x, y, and z axes for analysis. However, rotation of the volume with Doppler directional flow information can mislead the operator: if the directions are reversed, flow data can be misinterpreted. The operator must confirm any suspected pathological flow patterns by confirming the original direction of scanning, whether flow was toward or away from the transducer during the acquisition scan.

234

Fetal Cardiology

Figure 15.18 Artifacts and pitfalls. STIC acquisition in a 26-week fetus: the A-frame shows left ventricular outflow tract plane. Note that the B-frame however is degraded by fetal breathing artifacts (arrows). (Reproduced with permission from Yagel S, Cohen SM, Shapiro I, Valsky DV. 3D and 4D ultrasound in fetal cardiac scanning: a new look at the fetal heart. Ultrasound Obstet Gynecol 2007; 29: 81–95. © International Society of Ultrasound in Obstetrics and Gynecology. Permission is granted by John Wiley & Sons Ltd on behalf of ISUOG.)

Accuracy While several studies have compared imaging yield between 2D and 3D/4D fetal echocardiography, and others have examined the feasibility of 3D/4D and STIC in screening programs, and still others have described the application of various 3D/4D modalities to diagnosis or evaluation of fetal cardiovascular anomalies, no large study has examined the contribution of 3D/4D to fetal echocardiography screening programs’ accuracy. Levental and co-workers compared 2D and non-gated 3D ultrasound to obtain standard cardiac views.39 MeyerWittkopf and colleagues40 evaluated 2D and Dopplergated 3D in obtaining standard echocardiography scanning planes in normal hearts. They found that 3D provided additional structural depth and allowed a dynamic 3-D

perspective of valvar morphology and ventricular wall motion.40 In evaluating CHD, Meyer-Wittkopf and co-workers41 evaluated gated 3D volume sets of 2D-diagnosed cardiac lesions, and compared key views of the heart in both modalities. They determined that 3D had added value in a small proportion of lesions.41 Wang and colleagues42 compared 3D and 2D scanning of fetuses in the spine-anterior position. This group found that only in the pulmonary outflow tract was 3D ultrasound superior to 2D. Espinoza and colleagues22 examined the added value of IM in the evaluation of anomalous venous connections. The investigators found that IM improved visualization of cases of dilated azygos or hemiazygos veins and their spatial relationships with the surrounding vascular structures.

Three- and four-dimensional ultrasound in fetal echocardiography

Most recently, Benacerraf and colleagues43 compared acquisition and analysis times for 2D and 3D fetal anatomy scanning at 17–21 gestational weeks. 3D ultrasound compared favorably with 2D in mean scanning time and accuracy of fetal biometry. The data archiving and networking capabilities of 3D/4D fetal echocardiography with STIC acquisition open new avenues for disseminating fetal echocardiography programs to outlying or poorly served areas. This can have important public health implications in these populations. Michailidis and co-workers44 and Vinals and colleagues25,26 have shown the feasibility and success of programs based on 3D/4D examination volumes acquired in one center, and reviewed by experts in a center connected by telemedicine internet link. In coming years, studies will direct 3D/4D capabilities to the evaluation of fetal cardiac functional parameters. This may provide insights into the physiological effects of fetal structural or functional cardiac defects, or maternal disease such as diabetes, on the developing fetus. To date, no large study has been performed to examine whether the addition of 3D/4D modalities to fetal echocardiography screening programs increases detection rates of cardiac defects. This technology has reached the stage when its reproducibility and added value in screening accuracy should be tested in large prospective studies, not only by teams or in centers that have made 3D/4D their specialty, but among the generality of professionals performing fetal echocardiography.

Acknowledgment This chapter originally appeared as Yagel S, Cohen SM, Shapiro I, Valsky DV. 3D and 4D ultrasound in fetal cardiac scanning: a new look at the fetal heart. Ultrasound Obstet Gynecol 2007; 29: 81–95. It is reproduced with permission.

Video clip 15.3 STIC acquisition combined with IM and VOCAL analysis for fetal cardiac ventricle volumetry. The computer mouse was used to manually define the contours of the ventricle at sequential planes. The resulting measurements appear in the box, bottom right.

Video clip 15.4 (a) STIC acquisition in a case of TAPVC. The A-plane showed raised suspicion of an anomalous vessel (caret), which is confirmed in the B-plane (arrow). (b) The heart and great vessels of this fetus: STIC acquisition and high definition power flow Doppler confirmed the characteristic vertical vein (VV; dAo, descending aorta; IVC, inferior vena cava). Note also the absence of pulmonary veins (compare Figure 15.7).

Video clip 15.5 B-flow of the heart and great vessels in a fetus with interrupted inferior vena cava with azygos continuation. (AoA, aortic arch; AzV, azygos vein; SVC, superior vena cava; DV, ductus venosus.)

Video clip 15.6 The IVS “virtuaI plane” with color Doppler in evaluation of VSD shows the rendered IVS plane, demonstrating flow across the defect from right to left.

Video clip 15.7 B-flow modality showing the parallel great vessels in a case of transposition. Application of this modality clearly shows the blood flow in the malaligned vessels. PA, pulmonary artery; AO, aorta.

References 1.

2.

3.

Legends for the DVD 4.

Video clip 15.1 B-flow of normal heart and aortic arch. Brachocephalic trunk (BT), left common carotid (LCC), and left subclavian artery (LSA) are seen projecting from the aortic arch (AoA). Inferior vena cava (IVC) is indicated.

Video clip 15.2 Normal heart and great vessels: STIC acquisition with high definition power flow Doppler. DV, ductus venosus; UV, umbilical vein; IVC, inferior vena cava; CA, celiac artery; SMA, superior mesenteric artery; PV, pulmonary veins; dAo, descending aorta.

235

5.

6.

DeVore GR, Falkensammer P, Sklansky MS, Platt LD. Spatio-temporal image correlation (STIC): new technology for evaluation of the fetal heart. Ultrasound Obstet Gynecol 2003; 22: 380–7. Deng J. Terminology of three-dimensional and fourdimensional ultrasound imaging of the fetal heart and other moving body parts. Ultrasound Obstet Gynecol 2003; 22: 336–44. Gonçalves LF, Lee W, Chaiworapongsa T et al. Fourdimensional ultrasonography of the fetal heart with spatiotemporal image correlation. Am J Obstet Gynecol 2003; 189: 1792–802. Falkensammer P. Spatio-temporal image correlation for volume ultrasound. Studies of the fetal heart. Zipf, Austria: GE Healthcare, 2005. Gonçalves LF, Lee W, Espinoza J, Romero R. Examination of the fetal heart by four-dimensional (4D) ultrasound with spatio-temporal image correlation (STIC). Ultrasound Obstet Gynecol 2006; 27: 336–48. Volpe P, Campobasso G, Stanziano A et al. Novel application of 4D sonography with B-flow imaging and spatiotemporal image correlation (STIC) in the assessment of the anatomy of pulmonary arteries in fetuses with pulmonary atresia and ventricular septal defect. Ultrasound Obstet Gynecol 2006; 28: 40–6.

236

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

Fetal Cardiology

Gonçalves LF, Romero R, Espinoza J et al. Fourdimensional ultrasonography of the fetal heart using color Doppler spatiotemporal image correlation. J Ultrasound Med 2004; 23: 473–81. Messing B, Porat S, Imbar T et al. Mild tricuspid regurgitation: a benign fetal finding at various stages of pregnancy. Ultrasound Obstet Gynecol 2005; 26: 606–9. Chaoui R, Kalache KD, Hartung J. Application of three-dimensional power Doppler ultrasound in prenatal diagnosis. Ultrasound Obstet Gynecol 2001; 17: 22–9. Chaoui R, Hoffmann J, Heling KS. Three-dimensional (3D) and 4D color Doppler fetal echocardiography using spatio-temporal image correlation (STIC). Ultrasound Obstet Gynecol 2004; 23: 535–45. Sciaky-Tamir Y, Cohen SM, Hochner-Celnikier D et al. Three-dimensional power Doppler (3DPD) ultrasound in the diagnosis and follow-up of fetal vascular anomalies. Am J Obstet Gynecol 2006; 194: 274–81. Yagel S, Cohen SM, Achiron R. Examination of the fetal heart by five short-axis views: a proposed screening method for comprehensive cardiac evaluation. Ultrasound Obstet Gynecol 2001; 17: 367–9. Yagel S, Arbel R, Anteby EY, Raveh D, Achiron R. The three vessels and trachea view (3VT) in fetal cardiac scanning. Ultrasound Obstet Gynecol 2002; 20: 340–5. DeVore GR, Polanco B, Sklansky MS, Platt LD. The ‘spin’ technique: a new method for examination of the fetal outflow tracts using three-dimensional ultrasound. Ultrasound Obstet Gynecol 2004; 24: 72–82. Abuhamad A. Automated multiplanar imaging: a novel approach to ultrasonography. J Ultrasound Med 2004; 23: 573–6. Espinoza J, Kusanovic JP, Goncalves LF et al. A novel algorithm for comprehensive fetal echocardiography using 4-dimensional ultrasonography and tomographic imaging. J Ultrasound Med 2006; 25: 947–56. Yagel S, Benachi A, Bonnet D et al. Rendering in fetal cardiac scanning: the intracardiac septa and the coronal atrioventricular valve planes. Ultrasound Obstet Gynecol 2006; 28: 266–74. Ruano R, Joubin L, Aubry MC et al. A nomogram of fetal lung volumes estimated by 3-dimensional ultrasonography using the rotational technique (virtual organ computeraided analysis). J Ultrasound Med 2006; 25: 701–9. Peralta CF, Cavoretto P, Csapo B, Falcon O, Nicolaides KH. Lung and heart volumes by three-dimensional ultrasound in normal fetuses at 12-32 weeks’ gestation. Ultrasound Obstet Gynecol 2006; 27: 128–33. Messing B, Cohen SM, Valsky DV et al. Fetal cardiac ventricle volumetry in the second half of gestation assessed by 4D ultrasound using STIC combined with inversion mode. Ultrasound Obstet Gynecol 2007; 30: 142–51. Gonçalves LF, Espinoza J, Lee W, Mazor M, Romero R. Three- and four-dimensional reconstruction of the aortic and ductal arches using inversion mode: a new rendering algorithm for visualization of fluid-filled anatomical structures. Ultrasound Obstet Gynecol 2004; 24: 696–8. Espinoza J, Gonçalves LF, Lee W, Mazor M, Romero R. A novel method to improve prenatal diagnosis of

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

abnormal systemic venous connections using three- and four-dimensional ultrasonography and ‘inversion mode’. Ultrasound Obstet Gynecol 2005; 25: 428–34. Gonçalves LF, Espinoza J, Lee W et al. A new approach to fetal echocardiography: digital casts of the fetal cardiac chambers and great vessels for detection of congenital heart disease. J Ultrasound Med 2005; 24: 415–24. International Society of Ultrasound in Obstetrics and Gynecology. Cardiac screening examination of the fetus: guidelines for performing the ‘basic’ and ‘extended basic’ cardiac scan. Ultrasound Obstet Gynecol 2006; 27: 107–13. Vinals F, Poblete P, Giuliano A. Spatio-temporal image correlation (STIC): a new tool for the prenatal screening of congenital heart defects. Ultrasound Obstet Gynecol 2003; 22: 388–94. Vinals F, Mandujano L, Vargas G, Giuliano A. Prenatal diagnosis of congenital heart disease using four-dimensional spatio-temporal image correlation (STIC) telemedicine via an Internet link: a pilot study. Ultrasound Obstet Gynecol 2005; 25: 25–31. Hyett JA, Perdu M, Sharland GK, Snijders RJM, Nicolaides KH. Using fetal nuchal translucency to screen for major congenital cardiac defects at 10-14 weeks of gestation: population based cohort study. Br Med J 1999; 318: 81–5. Hyett JA, Perdu NL, Sharland GK, Snijders RJM, Nicolaides KH. Increased nuchal translucency at 10-14 weeks of gestation as a marker for major cardiac defects. Ultrasound Obstet Gynecol 1997; 10: 242–6. Meyer-Wittkopf M, Cole A, Cooper SG, Schmidt S, Sholler GF. Three-dimensional quantitative echocardiographic assessment of ventricular volume in healthy human fetuses and in fetuses with congenital heart disease. J Ultrasound Med 2001; 20: 317–27. Esh-Broder E, Ushakov FB, Imbar T, Yagel S. Application of free-hand three-dimensional echocardiography in the evaluation of fetal cardiac ejection fraction: a preliminary study. Ultrasound Obstet Gynecol 2004; 23: 546–51. Bhat AH, Corbett V, Carpenter N et al. Fetal ventricular mass determination on three-dimensional echocardiography: studies in normal fetuses and validation experiments. Circulation 2004; 110: 1054–60. Bhat AH, Corbett VN, Liu R et al. Validation of volume and mass assessments for human fetal heart imaging by 4-dimensional spatiotemporal image correlation echocardiography: in vitro balloon model experiments. J Ultrasound Med 2004; 23: 1151–9. Carvalho JS, Ho SY, Shinebourne EA. Sequential segmental analysis in complex fetal cardiac abnormalities: a logical approach to diagnosis. Ultrasound Obstet Gynecol 2005; 26: 105–11. Yagel S, Kivilevitch Z, Achiron R. The fetal venous system: normal embryology, anatomy, and physiology and the development and appearance of anomalies. In: Yagel S, Silverman NH, Gembruch U, eds. Fetal Cardiology. London: Martin Dunitz, 2003: 321–32. Moore KL, Persaud TVN. The cardiovascular system. In: The Developing Human: Clinically Oriented Embryology, 6th edn. Philadelphia: WB Saunders, 1998: 349–404.

Three- and four-dimensional ultrasound in fetal echocardiography

36.

37.

38.

39.

40.

Birk E, Silverman NH. Intracardiac shunt malformations. In: Yagel S, Silverman NH, Gembruch U, eds. Fetal Cardiology. London: Martin Dunitz, 2003: 201–10. Paladini D, Russo MG, Vassallo M, Tartaglione A. The ‘in-plane’ view of the inter-ventricular septum. A new approach to the characterization of ventricular septal defects in the fetus. Prenat Diagn 2003; 23: 1052–5. Yagel S, Valsky DV, Messing B. Detailed assessment of fetal ventricular septal defect with 4D color Doppler ultrasound using spatio-temporal image correlation technology. Ultrasound Obstet Gynecol 2005; 25: 97–8. Levental M, Pretorius DH, Sklansky MS et al. Threedimensional ultrasonography of normal fetal heart: comparison with two-dimensional imaging. J Ultrasound Med 1998; 17: 341–8. Meyer-Wittkopf M, Rappe N, Sierra F, Barth H, Schmidt S. Three-dimensional (3-D) ultrasonography for obtaining the four and five-chamber view: comparison with

41.

42.

43.

44.

237

cross-sectional (2-D) fetal sonographic screening. Ultrasound Obstet Gynecol 2000; 15: 397–402. Meyer-Wittkopf M, Cooper S, Vaughan J, Sholler G. Three-dimensional (3D) echocardiographic analysis of congenital heart disease in the fetus: comparison with cross-sectional (2D) fetal echocardiography. Ultrasound Obstet Gynecol 2001; 17: 485–92. Wang PH, Chen GD, Lin LY. Imaging comparison of basic cardiac views between two- and three-dimensional ultrasound in normal fetuses in anterior spine positions. Int J Cardiovasc Imaging 2002; 18: 17–23. Benacerraf BR, Shipp TD, Bromley B. Three-dimensional US of the fetus: volume imaging. Radiology 2006; 238: 988–96. Michailidis GD, Simpson JM, Karidas C, Economides DL. Detailed three-dimensional fetal echocardiography facilitated by an Internet link. Ultrasound Obstet Gynecol 2001; 18: 325–8.

16 Cardiac malpositions and syndromes with right or left atrial isomerism Rabih Chaoui Isomerism and related heart malpositions belong to the most difficult chapters in pediatric cardiology. According to the Baltimore–Washington Infant Study,1 which analyzed 4390 congenital heart defects (CHDs) detected in the first year of life over a period of 10 years (1981–1989), isomerism was found in 99 cases, accounting for 2.2% of CHDs. The mortality in this small group was 51% within the first year of life. In the fetus the true prevalence is, however, not known, since some forms, especially when associated with heart block and fetal hydrops, would end as fetal death, and some other more mild forms of cardiac malposition or isolated situs inversus may be overlooked even in a child. The early prenatal detection of an abnormality of this group has a large impact on counseling the pregnant woman, especially because small details can radically change the prognosis. There are difficulties in achieving a final fetal diagnosis, as the examiner can rely only on ultrasound, where there are limited possibilities for the precise differentiation of structures. Furthermore, some basic knowledge is needed in order to understand some definitions as well as the classification of defects. In this chapter, not all aspects of these abnormalities can be covered in detail. It is tried rather to supply the reader with basic information for a practical approach to suspected abnormal conditions.

Developmental aspects and normal body configuration In contrast to other embryological organs, the thoracic and abdominal structures develop asymmetrically. There are well-defined right-sided and left-sided organs and structures. After completion of lateralization, the ‘normal’ and most common condition found is then called situs solitus for the visceral arrangement, and levocardia (heart on the left side) for the thoracic arrangement (Figure 16.1). In Chapter 12, on fetal cardiac anatomy, the focus was on

the segmental analysis of the upper abdomen and the heart, showing that under normal conditions in situs solitus the stomach and descending aorta are on the left and the liver and the inferior vena cava on the right. The umbilical vein bends to the right, continuing with the portal sinus. In levocardia the heart apex points to the left anterior thoracic cavity with normal atrial and ventricular arrangement (Figure 16.1) and, under normal conditions, the inferior vena cava is connected to the right atrium and the pulmonary veins to the left atrium.

Situs inversus, malrotations, and malpositions of the heart Compared to the most common condition of situs solitus and levocardia, there is very rarely a situation where all organs are rotated exactly to the opposite side, leading to a mirror-image arrangement. This situation is called situs inversus. When this mirror-image rotation of abdominal and intrathoracic organs is complete, the liver and inferior vena cava are on the left, whereas the stomach and the descending aorta are on the right (Figures 16.2 and 16.3). Since the inferior vena cava and the right atrium are concordant, the right atrium and right ventricle are on the left anteriorly and the left atrium and left ventricle are on the right posteriorly and the heart axis points to the right anterior thorax, a situation called dextrocardia (Figure 16.3) (mirror-image dextrocardia or situs inversus with dextrocardia, or situs inversus completus). In these conditions associated heart defects are rare. The true incidence of this situation is low, but also not exactly known, since persons with this abnormality are asymptomatic and are not identified until an X-ray, ultrasound, or medical intervention is performed. Because this condition is extremely rare, it can be recommended that, before making this diagnosis, the examiner checks the fetal position and the transducer orientation. According to our experience, we have found

240

Fetal Cardiology

Figure 16.1 The two important planes in assessing isomerism and other cardiac malpositions. The normal sonoanatomy of the upper abdomen with situs solitus (left) and the heart in levocardia (right). R, right; L, left; ST, stomach; VCI, inferior vena cava; AO, aorta; RV, right ventricle; LV, left ventricle.

Figure 16.2 Situs inversus. The fetus is in vertex position and in the upper abdomen a mirror-image rotation is found with the stomach (St.) and the aorta (Ao) on the right side and the inferior vena cava (Vci) and liver on the left side.

Figure 16.3 The heart in a fetus with complete situs inversus and mirrorimage dextrocardia. The fetus is in vertex position. The heart points to the right anterior thorax (arrow). The right ventricle (RV) (with trabeculation) is on the left side and in the anterior thorax. The descending aorta (Ao) is on the right side. LA, left atrium; RA, right atrium.

Cardiac malpositions and isomerism

241

Figure 16.4 Two examples of dextropositions: on the left, a fetus with left-sided congenital diaphragmatic hernia (ST, stomach) with a shifting of the heart into the right thorax; on the right, a fetus with the rare finding of right pulmonary agenesis. The heart fills the space of the right lung and the left lung fills the thorax.

that a high risk for situs inversus was in a family with a previous child with this condition, in consanguineous couples, and in the offspring of diabetic mothers. However, there are other conditions where the mirror-image arrangement is partial: either only affecting the heart, as in dextrocardia (with situs solitus); or affecting the visceral organs, as in situs inversus with levocardia.2 In these conditions heart defects are more common. The heart position and axis are important features to check while analyzing the heart. Compared to its normal position in the left chest with a normal base–apex axis of 45° to the left side, cardiac malpositions are divided into dextrocardia, levocardia, mesocardia, and ectopia cordis. As already stated, dextrocardia is present when the heart is in the right hemithorax with the axis pointing to the right. Mesocardia is found when the heart points to the midline of the thorax. Many children having mesocardia are incorrectly grouped in the dextrocardia group, since on X-ray the heart seems more on the right side. Levocardia (heart on the left, i.e. the normal position) is mainly considered when there is a situs inversus or ambiguus, to stress that the heart still points to the left side. Cardiac dextroposition (displacement to the right) is found when the heart is shifted into the right chest with the axis still pointing to the left, as observed in left-sided congenital diaphragmatic hernia, left-sided intrathoracic masses, or fluid accumulation, in agenesis of the right lung (Figure 16.4), or in scimitar syndrome.

Similarly, levoposition (displacement to the left) is seen when the heart is shifted into the left thoracic cavity with the axis still pointing to the left, as typically found in right-sided diaphragmatic hernia or some other right-sided thoracic lesions.

Right and left atrial isomerisms in the fetus Conversely to the above-cited very rare abnormalities with a mirror-image arrangement of the visceral and intrathoracic organs, there are more common conditions with an incomplete lateralization (heterotaxy) of these organs during embryological rotation, showing an indeterminate visceral situs. In these conditions, the arrangement is called situs ambiguus (indeterminate or uncertain situs) and is the most complex form of visceral and atrial arrangement. This group of defects has many synonyms, such as heterotaxy syndromes, cardiosplenic syndromes, or asplenia– polysplenia syndromes or isomerisms, and the conditions are regularly associated with complex cardiac defects. In these abnormal lateralizations of the visceral and intrathoracic structures, there is a tendency to symmetric development of the normally asymmetric organs, associated with either a bilateral right-sidedness

242

Fetal Cardiology

Figure 16.5 Fetus with left isomerism showing, in the upper abdomen, situs ambiguus with the stomach (ST) on the right side (R) and an interruption of the inferior vena cava with azygos (AZ) vein persistence. The double-vessel sign is typical of left isomerism. In the four-chamber view the heart shows an atrioventricular septal defect (green arrows) and behind the heart again the double-vessel aorta (Ao) and azygos (compare with Figure 16.13). These cases of atrioventricular septal defects are not associated with Down syndrome.

Figure 16.6 This fetus was referred with bradycardia. M-mode shows the presence of a complete heart block: the atria (A) have a normal sinus rhythm (yellow arrows) whereas the ventricles show bradycardia (50 beats per minute) (red arrows). Examination of the fourchamber view reveals the diagnosis: a dilated heart with pericardial effusion, and behind the heart side-by-side with the descending aorta there is a second vessel, which is the dilated azygos vein; this is a typical finding in left isomerism.

Cardiac malpositions and isomerism

243

Figure 16.7 Two examples of left isomerism with the ‘double vessel sign’ (dilated azygos and aorta side-by-side) with the stomach on the left (left case) and on the right (right case). This example shows that the diagnosis is achieved by evaluation of the vessels’ position rather than the position of the stomach (compare also Figure 16.3).

Figure 16.8 Left isomerism with atrioventricular (AV) septal defect with azygos continuation side-by-side with the aorta.

(right isomerism) or bilateral left-sidedness (left isomerism). Ivemark3 noted the association of spleen anomalies with some cardiac defects. Because the spleen develops as a left-sided organ it was in the past a sign for classification,

asplenia being the previous name of right isomerism and polysplenia of left isomerism. This group of developmental defects were therefore called cardiosplenic syndromes. However, it was found that spleen presence, position, and number do not have definitive diagnostic value, and this terminology was abandoned4 (although it is often still used in clinical pediatric cardiology). Knowing that the heart develops according to welldefined cardiac segments, Van Praagh5 proposed a classification according to the anatomy and connections of these segments, known as the segmental approach. The constant starting point for classifying these conditions is the anatomy of the atria. Therefore, diseases with isomerism are now divided into right atrial and left atrial isomerism. The identification of these anomalies and their differentiation may be easier for the pathologist during necropsy. In the fetus the diagnostic possibilities are reduced, and the reliable diagnosis or differentiation between right or left isomerism can be very difficult to achieve prenatally.6 Even for the experienced examiner this group of anomalies is considered a challenge. The diagnostic approach using ultrasound is based on the approach proposed by Huhta et al7 for neonatal echocardiography, focusing on the upper abdomen and the relationship between the venous system and the atria. The central point of diagnosis is the anatomy of the atria defined by their shape and their appendages. The left atrial appendage is finger-like and has a narrow base, whereas the right atrial appendage is pyramidal in shape and its base is rather broad. The

244

Fetal Cardiology

Figure 16.9 Thoracic cross-section (left) and longitudinal view of aorta descendens and dilated azygos vein.

Figure 16.10 Left isomerism. If persistence of an azygos vein is suspected, the examiner can confirm the diagnosis by using color Doppler. In a frontal view the descending aorta and the azygos vein can be seen side-by-side with different flow directions (left). In its further course the azygos vein crosses the diaphragm and connects with the superior vena cava to enter the right atrium (middle). Doppler findings demonstrate the venous pattern (right) (arrhythmia with the associated heart block).

appendages can be visualized in a plane slightly cranial to the four-chamber view, but are not identified reliably under many conditions. In a recent retrospective study8 on 30 fetuses with isomerism it was, however, shown that prenatally the morphology of the atrial appendages

could have been suspicious in 19 cases, with the typical bilateral sickle-shape appearance in left and the bluntshape appearance in right isomerism. Since the connecting veins are part of the atrial anatomy, the venoatrial connection is the leading diagnostic sign for

Cardiac malpositions and isomerism

diagnosis. However, there are no heart defects that are pathognomonic for the one or the other diagnostic group. In the following sections some features associated with each anomaly are enumerated, which might help in making diagnoses.

245

The types of cardiac malformation associated with left and right isomerism are complex, showing a considerable overlap. Except the azygos continuation of the inferior vena cava, considered as a typical sign of left isomerism,

Left atrial isomerism In this condition of double left-sidedness, right-sided structures such as the inferior vena cava and the right atrium with sinus node are absent or may have developed abnormally. Therefore, two leading signs in left isomerism can be expected: first, the ‘interruption’ of the inferior vena cava in its intrahepatic part and its persistence as the azygos (or hemiazygos) vein (Figures 16.5–16.10); second, arrhythmia with heart block (malformed sinus node) (Figures 16.6 and 16.10). The azygos continuation of the interrupted inferior vena cava has been shown to be present in most cases with left isomerism (> 80%). It can be recognized by the observation of the aorta and the (dilated) azygos vein on its right or left side (hemiazygos) either in the upper abdomen (Figure 16.7) or at the level of the four-chamber view (Figure 16.8).9. Sheley et al10 described it as the ‘double-vessel sign’, and found it in all eight fetuses with left isomerism that they examined, but also in one false-positive case with right isomerism. In another more recent study in 22 fetuses with left isomerism,11 21 showed this azygos continuation sign. If the examiner is aware of this sign, he can easily detect it prenatally on real-time imaging and confirm it using color Doppler (Figures 16.9 and 16.10). The azygos vein is then visualized leading into the superior vena cava or into a persisting left superior vena cava (Figures 16.10 and 16.11).

Figure 16.11 Transverse view across the upper thorax showing the so-called three-vessel view (VCS, superior vena cava; AO, aorta; TP, pulmonary trunk). In this fetus with left isomerism the (dilated) azygos vein is seen to connect to the superior vena cava. Furthermore, to the left of the pulmonary trunk, there is a fourth vessel, which is the left persistent superior vena cava (LVCS).

Figure 16.12 This fetus was referred at 14 weeks because of nuchal edema (and beginning hydrops) associated with bradycardia. We found a heart block, a stomach (St) on the right side (left) and levocardia (H) with a heart defect. The heart showed a univentricular atrioventricular connection with a small (v) and a dilated ventricle (V) connected by a ventricular septal defect (*). The liver lay centrally. After termination of pregnancy because of suspected left isomerism, the diagnosis was confirmed at autopsy (see Figure 16.13).

246

Fetal Cardiology

Figure 16.13 Necropsy of the 14-week fetus shown in Figure 16.12, demonstrating the central liver (left), the heart with two left atrial appendages (arrows), and, after removing the liver, the stomach (St) on the right side.

Figure 16.14 Right atrial isomerism. In the upper abdomen the aorta (AO) and inferior vena cava (VCI) are on the same side (here the right side (R)) and the inferior vena cava is anterior to the aorta (arrows). The stomach (ST) is nearer the midline than on the left. In a longitudinal plane (right) both the aorta (blue) and the inferior vena cava (red) are seen in one plane and are both directly anterior to the spine, a simultaneous visualization not seen under normal conditions.

there are no cardiac defects permitting a strict classification into one or other group of isomerism. The conditions of a heart with left atrial isomerism seem to be less severe than those with right isomerism, demonstrating a normal ventriculoarterial junction in almost 70% of cases.12 These hearts tend to be biventricular, and on many occasions a

ventricular septal defect or an atrioventricular septal defect is present.4,13 The association of an atrioventricular septal defect with complete heart block is considered to be pathognomonic for left atrial isomerism6 and should prompt careful examination of the venous connections. Heart block detected in the early second trimester is also

Cardiac malpositions and isomerism

247

very likely to be due to left isomerism and not to maternal autoantibodies14 (Figures 16.12 and 16.13). In the study reported above in 22 fetuses with left isomerism, a persisting bradycardia was found in 12/22 cases.11 The position of the heart can be on the left, on the right, or in the midline. The most severe complex extracardiac malformation observed in left isomerism is extrahepatic biliary atresia with absence of the gallbladder.

Right atrial isomerism

Figure 16.15 In this fetus with right isomerism the stomach was also found to be central. In late pregnancy we observed a herniation of the stomach into the thoracic cavity (arrows) through the intact diaphragm.

In this condition of double right-sidedness, left-sided structures such as the left atrium, the pulmonary veins, and the upper gastrointestinal tract are likely to be found malformed. In this group of complex malformations there are no characteristic features, such as interruption of the inferior vena cava or the heart block described for left isomerism. The inferior vena cava is present and is generally on the same side as the descending aorta (Figures 16.14, 16.16, and 16.17). The visceral heterotaxy is more common and severe in right isomerism, and anomalies of the upper abdomen are more likely to be found in right than in left isomerism; these include not only the common absence of the spleen, but also the symmetrical liver, and non-fixation of the gastrointestinal tract leading to various degrees of malrotation (Figures 16.15 and 16.16). Atresia of the esophagus or

Figure 16.16 Right isomerism with the typical sign of the juxtaposition of inferior vena cava (VCI) and aorta as in Figure 16.15 either on the left (left case) or on the right side of the spine (right case). The position of the stomach can be on the right or on the left side (the stomach is generally more central and the liver is more enlarged).

248

Fetal Cardiology

duodenum can be found as well as the herniation of a midline-positioned stomach (Figure 16.15). Heart defects in right isomerism are likely to be more severe than those with left isomerism. Among the cardiac defects, total anomalous pulmonary venous drainage is found in 70% of cases (rarely in left isomerism) (Figure 16.19). Hearts with right atrial isomerism can also show an atrioventricular septal defect, but are more likely to be associated with a univentricular atrioventricular connection and exhibit a

much higher incidence of abnormal ventriculoarterial connections (double outlet ventricle, malposition of the great arteries). These hearts show a much higher frequency of pulmonary stenosis or atresia (Figure 16.18). An absence of the coronary sinus is found in 85% of cases. A left persisting superior vena cava is found very frequently (Figure 16.11). In a recent study of 21 fetuses with right isomerism,15 20 had complex cardiac anomalies, predominantly atrioventriculary septal defects and right ventricular obstruction in 62% and 48%, respectively Only 12 fetuses in this study showed a juxtaposition of aorta and inferior vena cava, and out of six cases with anomalous pulmonary venous return four were not diagnosed prenatally. Therefore, when right isomerism is suspected in a fetus, the connections of the pulmonary veins should be examined carefully to rule out the critical infradiaphragmatic pulmonary venous return.16

Prognosis in isomerism

Figure 16.17 Right isomerism in a 13-week fetus discerned by the detection of a cardiac anomaly at routine scan. The stomach was found to be on the right side, which prompted targeted examination of the vessels showing the typical sign of the juxtaposition of VCI and aorta.

It is known that left and right isomerisms are associated with a poor prognosis.4,13 Individual differences, depending on the specific finding, should be considered when counseling parents after a prenatal diagnosis, but it is often difficult to assess the prognosis from the prenatal scan in these conditions. Fetuses with left isomerism show a poor prognosis in utero when associated with complete heart block and hydrops. Cardiac failure and hydrops may occur very early, leading to spontaneous in utero demise. Postnatally, the prognosis may be further complicated by the underlying heart defect, although this is generally less severe than in right isomerism. Among extracardiac anomalies, biliary atresia can be considered as the most severe finding, which cannot be definitively ruled out in

Figure 16.18 Fetus with right isomerism. Owing to the inferior vena cava connection the atrium on the left is recognized as a right atrium and the atrium on the right receives the (abnormal) connections of the pulmonary veins. There is a univentricular atrioventricular connection to one ventricle (V). The outflow tract evaluation revealed pulmonary atresia: color Doppler demonstrates a normal antegrade flow through the aorta (AO) and retrograde flow across the ductus arteriosus (DA) into the pulmonary trunk (TP). Univentricular hearts and right-sided obstructions are more represented in right isomerism (see text).

Cardiac malpositions and isomerism

Figure 16.19 Abnormal pulmonary venous drainage in a fetus with right isomerism.

249

utero. Under some conditions, children with left isomerism show a better survival rate than those with right isomerism. The diagnosis of right atrial isomerism is very difficult in utero and should be considered in every fetus with a complex cardiac malformation, especially when cardiac or situs malposition are suspected. The severity of the disease usually appears postnatally, and is due to the abnormal pulmonary venous connection, to the ductus-dependent pulmonary perfusion in right outflow tract obstruction, or to the complex chamber anatomy. Within the first year of life, 79–94% of all children with right isomerism were reported to die, with or without operation.13 A long-term risk in patients with right isomerism is also infection due to asplenia, which is often associated. The association with chromosomal aberrations such as trisomy 21, 13, 18, or others is extremely rare, since the diagnosis of isomerism rather rules out such chromosomal aberrations. Yates et al17 reported, however, a fetus with

Figure 16.20 Summary of the four typical findings in the upper abdomen in situs solitus (top left), situs inversus (top right), right isomerism (bottom left), and left isomerism (bottom right). The inferior vena cava (blue) and its position relative to the aorta (red) are main landmarks.

250

Fetal Cardiology

isomerism associated with a 22q11 deletion. Recent results in families with recurrences of visceral heterotaxy cases showed that the gene for these malformations is probably localized on the long arm of the X chromosome (region between Xq24 and X27.1 but probably on Xq26).18 In the near future, the prenatal diagnosis of these abnormalities may be revolutionized by this new knowledge.

Conclusion Every fetal heart defect should be analyzed in a segmental approach in order to detect (or rule out) an isomerism (Figure 16.20). This approach can be difficult to achieve in some cases and even omitted in others. Therefore, the examiner should always rule out isomerism when the following ultrasound signs are found: cardiac or stomach malpositions, a complex cardiac defect, fetal heart block, abnormal venous connections, and dilatation of the azygos vein.

References 1. Ferencz C, Rubin JD, Loffredo CA, Magee CA. The Epidemiology of Congenital Heart Disease, The Baltimore– Washington Infant Study 1981–1989. Perspectives in Pediatric Cardiology, Vol 4. Mount Kisco, NY: Futura Publishing, 1993. 2. Blaas H-G, Hals J, Bjornstad PG, Pedersen T. Situs inversus viscerum with levocardia, without associated anomalies. Fetus 1991; 1: 75–93. 3. Ivemark BI. Implications of agenesis of the spleen in the pathogenesis of conotruncus anomalies in childhood. Acta Paediatr 1955; 44 (Suppl 104): 7–110. 4. Freedom RM, Smallhorn JF. Syndromes of right or left atrial isomerism. In: Freedom R, Benson L, Smallhorn J, eds. Neonatal Heart Disease. New York: Springer, 1995: 543–60. 5. Van Praagh R. The segmental approach to diagnosis in congenital heart disease. Birth Defects 1972; 8: 4–23.

6. Hobbins JC, Drose JA. Cardiosplenic syndromes. In: Drose J, ed. Fetal Echocardiography. Philadelphia: WB Saunders, 1998: 253–62. 7. Huhta J, Smallhorn JF, Macartney FJ. Two-dimensional echocardio-graphic diagnosis of situs. Br Heart J 1982; 48: 97–108. 8. Berg C, Geipel A, Kohl T et al. Fetal echocardiographic evaluation of atrial morphology and the prediction of laterality in cases of heterotaxy syndromes. Ultrasound Obstet Gynecol 2005; 26: 538–45. 9. Phoon CK, Villegas MD, Ursell PC, Silverman NH. Left atrial isomerism detected in fetal life. Am J Cardiol 1996; 77: 1083–8. 10. Sheley RC, Nyberg DA, Kapur R. Azygous continuation of the interrupted inferior vena cava: a clue to prenatal diagnosis of the cardiosplenic syndromes. J Ultrasound Med 1995; 14: 381–7. 11. Berg C, Geipel A, Kamil D et al. The syndrome of left isomerism: sonographic findings and outcome in prenatally diagnosed cases. J Ultrasound Med 2005; 24: 921–31. 12. Peoples WM, Moller JH, Edwards JE. Polysplenia: a review of 146 cases. Pediatr Cardiol 1983; 4: 129–37. 13. Hagler DJ, O’Leary P. Cardiac malpositions and abnormalities of atrial and visceral situs. In: Emmanouilides G, Allen H, Riemenschneider T, Gutgesell H, eds. Moss and Adams’ Heart Disease in Infants, Children and Adolescents. Baltimore: Williams & Wilkins, 1995; 1307–36. 14. Baschat A, Gembruch U, Knöpfle G, Hansmann M. First trimester heart block: a marker for cardiac anomaly. Ultrasound Obstet Gynecol 1999; 14: 311–14. 15. Berg C, Geipel A, Kamil D et al. The syndrome of right isomerism – prenatal diagnosis and outcome. Ultraschall Med 2006; 27: 225–33. 16. Batukan C, Schwabe M, Heling KS, Hartung J, Chaoui R. Prenatal diagnosis of right atrial isomerism (aspleniasyndrome): case report. Ultraschall Med 2005; 26: 234–8. 17. Yates RW, Raymond FL, Cook A, Sharland GK. Isomerism of the atrial appendages associated with 22q11 deletion in a fetus. Heart 1996; 76: 548–9. 18. Casey B, Devoto M, Jones K, Ballabio A. Mapping a gene for familial situs abnormalities to human chromosome Xq24–q27.1. Nat Genet 1993; 5: 403–7.

17 Anomalies of the right heart Jean-Claude Fouron This chapter will cover significant lesions of the right side of the heart that can be observed during prenatal life. It will summarize our experience gained from fetal ultrasonographic cardiocirculatory assessments performed in our Fetal Cardiology Unit from the years 1990 to 2006. For each condition, after a brief anatomical description, the circulatory pathophysiology, echocardiographic diagnosis, clinical presentation, prognosis, and, finally, impact on obstetrical and perinatal management will be covered. For simplicity and clarity, first-line prenatal ultrasonographic studies will refer to routine screening performed on all pregnant women during the early second trimester to eliminate major malformations; second-line echocardiography will relate to ultrasonographic study focused on the cardiocirculatory system of the fetus, realized in a specialized unit by an expert sonographer.

Pathophysiology common to all prenatal right heart dysfunctions One of the major implications of the parallel disposition of the two ventricles which characterizes fetal hemodynamics is that the respective output of each ventricle could be different. Actually, during the third trimester of gestation, Doppler investigation has demonstrated that in the human fetus, right ventricular is 28% greater than left ventricular stroke volume.1 Any significant impairment to forward flow through the right-sided channels, be it at the atrial, ventricular, or arterial level, will cause an increase in the amount of blood crossing the foramen ovale to reach the left-sided cavities. In these circumstances, systemic output can be maintained if the following two conditions are met: (1) adequate size of the foramen ovale, and (2) the absence of left ventricular failure. Normal reference values for foramen ovale dimensions and Doppler velocity pattern throughout human pregnancy have been reported.2,3 It must be recognized, however, that in the presence of rightsided obstructive lesions, adequate size means that the

diameter of the foramen ovale has to be greater than normal values to avoid any restrictive effect. Combined ventricular output will then be predominantly made up of blood coming from the left side of the heart in a proportion dictated by the severity of the right-sided lesion. Flow through the lungs will comprise blood coming from either the right or the left ventricle, depending on the actual amount of blood ejected by the right ventricle. If this amount is significantly reduced, the ductus arteriosus will channel blood from the aorta towards the pulmonary circulation. Consequently, recording of flow direction through the ductus arteriosus is a reliable means of assessing the degree of severity of all lesions situated on the right side of the heart during prenatal life. Finally, any significant interference with the filling patterns of the rightsided cavities of the fetal heart should alter flow velocity waveforms in the systemic veins.4 These various hemodynamic adjustments can have major influences on secondary cardiac morphological development and on the clinical condition, both before and immediately after birth, as well as on the long-term prognosis of the disease.

Anomalies of the inlet The right atrium Idiopathic enlargement of the right atrium This is a rare entity, described predominantly in the adult population. A few cases have been reported in the pediatric literature5–8 and during prenatal life.5,9,10 On anatomic examination, the right atrium appears extremely dilated, and the tricuspid valve normally positioned. Histologically, the right atrial wall shows widespread muscular degeneration and diffuse fibrosis. Prenatal ultrasonographic diagnosis is easy since the enormous right atrium markedly alters the four-chamber view. The second-line sonographer must, however, make sure that no other malformation is present. Various

252

Fetal Cardiology

degrees of tricuspid insufficiency have been described, leading to the erroneous diagnosis of Ebstein’s anomaly.9,10 The few fetuses reported with idiopathic right atrial enlargement had no sign of circulatory failure. However, one neonate was seen in cardiac failure related to atrial tachycardia.5 A search in our prenatal database reveals only one case, identified at 39 weeks of pregnancy, delivered vaginally without problems (Figure 17.1). Postnatal echocardiographic study confirmed the diagnosis.

The tricuspid valve Tricuspid dysplasia This classification encompasses a wide spectrum of anomalies, from the simple thickening of all three leaflets of the tricuspid valve to loss of mobility related to involvement of the chordae tendinae. Tricuspid dysplasia must be distinguished from Ebstein’s anomaly on the basis of the absence of tethering of the leaflets to the posterior ventricular wall and, more frequently, to the septal surface. The dysplasia can be either isolated or, more frequently, associated with other malformations usually causing increased right ventricular pressure. In our database, 35 fetuses have been classified with this diagnosis, and it was an isolated finding in only seven cases. The isolated form of tricuspid dysplasia can be easily overlooked by the first-line sonographer, since both the four-chamber view and the relative position of the great arteries will appear normal globally. During a detailed

echocardiographic study, suspicion of tricuspid dysplasia raised by greater echogenicity of the valve will be confirmed by evidence of significant regurgitation on Doppler interrogation. This Doppler criterion is, however, plagued by its subjectivity, since some tricuspid insufficiency is commonly observed in normal fetuses. With the help of color Doppler investigation, criteria based on the extent of and surface area covered by the regurgitant jet have been proposed in an attempt to overcome this drawback.11 Somewhat disturbing is the occasional observation of tricuspid valvular dysplasia associated with right ventricular echogenic foci in fetuses with trisomy 21 (Figure 17.2). Clinically, isolated tricuspid dysplasia is usually a fortuitous finding. Significant tricuspid regurgitation causing right ventricular dilatation and dysfunction has, however, been reported.12 The prenatal prognosis is also dependent on the associated lesions, the more frequent being stenosis or atresia of the right ventricular outlet.13 Isolated tricuspid insufficiency carries a good fetal prognosis as long as the foramen ovale is wide enough to accommodate the volume overload caused by the regurgitant flow. In our group of fetuses with isolated tricuspid dysplasia, the anomaly was well tolerated throughout pregnancy in all but one case. Vaginal delivery is always possible when the regurgitation is well-tolerated. After birth, the evolution is usually uneventful, since the postnatal fall in right ventricular afterload causes a rapid decrease in the volume of regurgitation. The situation is completely different in the presence of major associated lesions, which will dictate the management and influence the outcome.

LA

Ao

LA

RA SVC

RV

FO

RA

(a)

(b)

Figure 17.1 (a) Echocardiographic real-time image recorded in a fetus with an idiopathic dilatation of the right atrium (RA). This view shows the markedly dilated right atrium and helps to rule out an Ebstein’s anomaly on the base of the normal attachment of the tricuspid valve. (b) The same fetus is scanned in a sagittal view. The two venae cavae can be seen entering the dilated RA. The color Doppler investigation confirms that the foramen ovale is not restrictive. SVC, superior vena cava, Ao, aorta, LA, left atrium; FO, foramen ovale; RV, right ventricle.

Anomalies of the right heart

TV

RA

Figure 17.2 Four-chamber view of the heart of a 34-week fetus illustrating the hyperechogenecity of the dysplastic tricuspid valve. This fetus also had a ventricular septal defect (membranous type) and the karyotype revealed a trisomy 21. TV, tricuspid valve.

Ebstein’s anomaly Although this malformation covers a wide spectrum in terms of presentation and severity, its anatomical characteristics are specific enough to allow reliable prenatal echocardiographic identification. Basically, the problem comes from abnormal attachments of the septal and posterior tricuspid valve leaflets. The leaflets are grossly deformed, tethered onto the septal surface and inferior wall, showing some mobility only at their tips. This produces two chambers within the right ventricle: one, the atrialized portion, made up of parts of the ventricular free wall and septum above the functional opening of the tricuspid valve, the other formed by the infundibulum. A ballooning anterior leaflet guards the outlet of the newly formed atrialized portion. An Ebstein’s anomaly can be observed on the left side of the heart in cases of ventricular inversion. Four of our 18 cases of Ebstein’s disease diagnosed before birth were of this form. From a hemodynamic point of view, the redundant and relatively fixed tricuspid leaflets are both stenotic and regurgitant, causing marked dilatation of the ‘new’ right atrial cavity. Flow through this dilated chamber is disturbed not only by the regurgitant tricuspid jet but also by the fact that ventricular depolarization causes myocardial contraction on both sides of the ‘functional opening’ of the tricuspid apparatus. Anterograde flow through the main pulmonary artery will depend on many factors, among them the severity of the tricuspid insufficiency, the functional capacity of the remaining infundibular portion of the right ventricle, and also the possible presence of stenosis at the subpulmonary and/or pulmonary valvular levels. In prenatal life, permeability of the outflow tract can be difficult to assess with the Doppler technique

253

since the entire right ventricular ejection fraction could go backward through the leaking tricuspid valve, which offers lower resistance than the pulmonary valve, which opens against pulmonary and systemic resistances. In severe tricuspid insufficiency, pulmonary circulation will be essentially maintained by retrograde flow coming from the ductus arteriosus, even though the right ventricular outlet could be anatomically permeable. The right-to-left shunt normally observed through the foramen ovale is, therefore, significantly increased. The foramen is usually large and non-restrictive; combined ventricular output can thus be maintained by an increase in left ventricular stroke volume. Evidence of a dilated right atrium on the four-chamber view will raise suspicion during first-line sonographic screening. However, in the incomplete form of Ebstein’s anomaly, where only the septal leaflets have a lower attachment than usual,14 the right atrial cavity may appear normal. Furthermore, even in the more complete form, the four-chamber view may not appear suspicious early in the second trimester since the right atrium may not yet be significantly enlarged (Figure 17.3). It is crucial that the first-line sonographer realizes that identification of the malformation early in gestation is essentially based on a four-chamber view which includes both atrioventricular (AV) valves; lack of mobility associated with insufficiency of the tricuspid valve might indeed be the only clue for diagnosis, at this stage of pregnancy (Figure 17.4). The second-line sonographer has the responsibility of completing the anatomic and hemodynamic assessments by obtaining the following information, which could have major pre- and/or postnatal prognostic implications: (1) the ratio of diameters of the functional tricuspid opening over the annulus; (2) the degree of displacement (ratio of the distance from the tricuspid annulus to the apex over the functional opening to the apex); (3) the surfaces of the right atrium + atrialized right ventricle over the functional right ventricle + left atrium + left ventricle;15 (4) the severity of tricuspid regurgitation;11 all these measurements can be obtained on a vertical four-chamber view (Figure 17.4); (5) the presence and type of right ventricular outflow tract obstruction better assessed on the short axis view at the base of the heart; (6) the ratio of fossa ovale diameter over the length of the atrial septum on a horizontal fourchamber view;16 (7) left ventricular output; (8) the cardiothoracic ratio;17 (9) the flow pattern through the main pulmonary artery and the ductus arteriosus (appearance of retrograde systolic and/or diastolic flows); (10) evidence of atrial or ventricular arrhythmia. Despite all this information, however, the chances of a given fetus with Ebstein’s anomaly reaching term without problems remain difficult to establish. Indeed, the criteria of severity proposed for extrauterine life do not necessarily apply to the parallel disposition of the fetal ventricles. In an effort to designate prognostic criteria specific to fetal life, we retrospectively studied 10 fetuses with Ebstein’s

254

Fetal Cardiology

LA LA

TV

RA

(a)

TR

(b)

TV LA

TV TR RA

(c)

(d)

Figure 17.3 (a, b) At 19 weeks of gestation. (a) This four-chamber view shows a slightly dilated right atrium without clear evidence of tethering of the septal leaflets. (b) Color Doppler investigation disclosed tricuspid insufficiency of moderate severity. (c, d) Same fetus at 34 weeks of gestation. (c) The right atrium is markedly dilated; the tricuspid leaflets are dysplastic, and abnormal attachments of the redundant anterolateral leaflet are clearly identified. (d) On color Doppler mode, a severe tricuspid insufficiency is now documented. TR, tricuspid regurgitation.

anomaly. This investigation suggests that the prenatal prognosis could be significantly influenced by the ability of the foramen ovale to decompress the right atrium (Figure 17.5). All six fetuses with a ratio of foramen ovale over septal atrial length greater than 0.3 went on to show normal circulatory function throughout pregnancy. The high risk of fetal death in the presence of a restrictive foramen ovale could explain why the majority of Ebstein’s anomalies seen in postnatal life are associated with large interatrial septal defects and right-to-left shunt. Arrhythmia due to extreme dilatation of the right atrium is another factor which, in all likelihood, should be a cause of sudden intrauterine death, even if flow through the foramen ovale is not restricted. Fetuses with Ebstein’s anomaly without cardiac failure or hydrops can usually tolerate vaginal delivery without

major problems. The situation is more dramatic in the presence of hydrops fetalis, which usually appears towards the end of the second or the early part of the third trimester of gestation. The attending team is then faced with the necessity of performing fetal extraction by cesarean section which will, in effect, add the risk of prematurity to that of severe circulatory failure in a fetus already crippled by a major cardiac malformation. The risk of postnatal death is thus very high in this group of infants. Irrespective of the condition of the fetus, the major challenge which will influence immediate postnatal outcome is the need to establish efficient flow through the lungs. Then, factors such as the degree of compression of the lungs by the dilated right atrium, the indices of severity of tricuspid impairment based on anatomy and the degree of regurgitation, the permeability of the right ventricular outflow

Anomalies of the right heart

255

tract, and the level of pulmonary vascular resistance will all come into play.

Tricuspid atresia

RA S and TV ''A'' RV LV

TV

Figure 17.4 Cardiac image of a 31-week fetus with an Ebstein’s anomaly. This vertical four-chamber view allows reliable assessment of (1) the tethering of the septal leaflet (S and TV); (2) the morphology of the anterior leaflet of the tricuspid valve (TV) which is redundant and loosely attached to the ventricular wall; (3) the extent of the atrialized portion of the right ventricle (‘A’ RV); (4) the degree of dilatation of the new atrium formed by the anatomical right atrium (RA) + the atrialized right ventricle. 0.7

0.6

FO / AS

0.5

0.4

0.3

0.2

0.1

0 Fetuses (n=10)

Figure 17.5 Distribution of the ratio, size of the foramen ovale (FO) over length of the interatrial septum (AS), in relation to the outcome of 10 fetuses with Ebstein’s anomaly. All fetuses with a ratio lower than 0.3 developed hydrops fetalis (adapted from reference 16). •, favorable outcome; ❒, hydropic fetus; Δ, fetal demise.

In postnatal life, tricuspid atresia is associated with either normally related (approximately 80% of cases) or transposed great arteries with, or less frequently without, ventricular septal defect.18 All but two of the 26 fetuses seen in our unit had normally related great arteries. The common feature in all these cases is the relative lack of growth of the right ventricle associated with an enlarged left ventricular cavity. This explains why tricuspid atresia is frequently classified under the hypoplastic right heart syndrome. Right ventricular hypodevelopment is, however, less marked in the presence of a ventricular septal defect; even though it is small, the cavity can then be well-identified and the tricuspid valve appears as either an imperforate membrane or fused echogenic leaflets with little or no mobility. Hypodevelopment of either pulmonary artery or aorta will also be observed, depending on which of these arteries arises from the right ventricle. Fetal survival is based on major hemodynamic adjustments. In cases of isolated tricuspid atresia and normally related great arteries, survival is possible through accommodation of the entire systemic return by the widely patent foramen ovale. Because of the parallel disposition of the two ventricles, this adjustment is compatible with normal peripheral perfusion throughout gestation. A major overload of the left-sided cavities is then observed. Secondary dilatation of the aorta associated with variable pulmonary artery hypodevelopment is also part of the classical picture. The pulmonary circulation is provided by retrograde flow through the ductus arteriosus. In tricuspid atresia with transposition of the great arteries, the widely patent foramen ovale remains an absolute necessity; overload of the left atrial and ventricular cavities also remains part of the picture. At the great arteries level, however, the pulmonary artery is dilated while the ascending aorta and the aortic arch are relatively hypoplastic, with an increased risk of coarctation of the aorta at birth; in these cases, the aortic arch and its arterial branches going to the head and arms as well as the coronary arteries are perfused by blood coming retrogradely from the aortic isthmus and the pulmonary artery. With both normally related and transposed great arteries, the presence of a large ventricular septal defect will cause passage of flow from the left to the right ventricle and some development of the right ventricular cavity. Pulmonary flow could even be normal or slightly decreased, depending on the size of the defect, and the main pulmonary artery or aorta may be within normal limits. The prenatal diagnosis of tricuspid atresia can be easily suspected during routine obstetric ultrasound. On the four-chamber view, asymmetric development of the two

256

Fetal Cardiology

ventricles is obvious (Figure 17.6a). The second-line investigation will confirm the diagnosis, not only with real-time examination of the inlet of the right ventricle showing the absence of tricuspid valve movement, but also by the lack of flow velocities across the tricuspid valve in pulsed and color Doppler investigations. The following additional points must be clarified: (1) the size of the fossa ovale and the mobility of the membrane of the foramen; (2) the size and function of the left ventricle; (3) the integrity of the ventricular septum; (4) the relative position of the main arteries; (5) the sizes of the pulmonary artery and aorta; (6) the size and direction of flow through the ductus arteriosus (Figure 17.6b and 17.6c). All these elements should help to establish the prognosis either before birth or immediately after birth. In reviewing our cases, it is surprising to see that all had good left ventricular function

LV

and no sign of circulatory failure. The most likely explanation is that cases of tricuspid atresia with relatively small foramen ovale do not survive beyond the first trimester of gestation. In the absence of hydrops, vaginal delivery is certainly possible in a fetus with tricuspid atresia. The need, however, for immediate care of the neonate in a tertiary center must be underlined. Arterial oxygen content will indeed be primarily dependent on the respective proportion of pulmonary and systemic venous return that will enter the left ventricle. In the absence of a ventricular septal defect, the survival of these neonates will rely on the patency of the ductus arteriosus, which represents the only source of pulmonary flow. The situation is not quite the same, however, in the presence of a large ventricular septal defect where forward pulmonary blood flow can be efficiently

VSD VSD RV

LA TV

(a)

TV

(b)

LV

VSD

TV Ao

(c)

Figure 17.6 (a) Four-chamber view recorded in a 20-week fetus with a tricuspid atresia, ventricular septal defect, and normally related great arteries. The relative hypoplasia of the right ventricle is obvious in this image. (b) The color Doppler investigation shows in diastole that blood is entering into the left ventricle through the mitral valve and being shunted into the right ventricle through the ventricular septal defect. (c) During the following systole, blood velocities are seen going towards the aorta from the right ventricle because of the presence of an associated pulmonic stenosis. LV, left ventricle; VSD, ventricular septal defect.

Anomalies of the right heart

maintained through the defect; such cases can even develop cardiac failure due to high cardiac output. When tricuspid atresia is associated with transposition of the great arteries, pulmonary flow is normal or increased, and the neonate may not have significant cyanosis; in these cases, however, cardiac failure, the size of the aorta, and the possibility of coarctation or arch interruption are all elements of concern.

Anomalies of the outlet: pulmonary stenosis The presence or absence of ventricular septal defect profoundly modifies the dynamics of obstructive lesions along the right ventricular outlet.

Without ventricular septal defect On the basis of morphologic appearance, clinical presentation, and prognosis, mild-to-moderate pulmonic stenosis has to be distinguished from critical stenosis or complete atresia of the valve. In the mild-to-moderate form, the thickness and mobility of the pulmonary valve are variable. The morphology of the right ventricle can be normal. In the moderate form, some right myocardial hypertrophy can be present particularly at the end of gestation. It is important to remember, however, that in prenatal life, the appearance of myocardial hypertrophy could decrease ventricular diastolic compliance and consequently end-diastolic volume. The extent of this fall in preload will vary

257

according to the gestational age at which the hypertrophy occurs. Mild-to-moderate pulmonary stenosis can be very difficult to detect on routine obstetric ultrasound because of the relatively minor changes observed on cardiac morphology. The four-chamber view can appear completely normal. Careful examination of the pulmonary valves could, however, disclose abnormal echogenicity of the leaflets. An increase in peak flow velocities through the valve can be the only clue to the diagnosis (Figure 17.7). This is the main reason why second-line sonographers should always perform a Doppler flow velocity investigation through the semilunar valves in all high-risk patients screened for cardiac malformation. The relationship between the increase in peak velocity through the stenotic valve and the severity of the lesion remains linear as long as the diastolic function of the right ventricle is not altered by myocardial hypertrophy.19 In the presence of moderate to severe stenosis, right ventricular filling characteristics change and end-diastolic pressures tend to rise, favoring an increase in the right-to-left shunting through the foramen ovale; at that point, the secondary fall in right ventricular output renders the modified Bernoulli equation based on the peak velocity measurements20 unreliable for the assessment of the severity of the stenosis. The second-line sonographer who makes the diagnosis of mildto-moderate pulmonic stenosis in prenatal life must complete the study and be sure to: (1) look at the integrity of the septal wall, on both the five-chamber and the shortaxis views at the base of the heart to eliminate ventricular septal defects (a three-dimensional (3D) ultrasonographic modality could help in this search); (2) investigate the tricuspid valve to rule out tricuspid regurgitation; (3) record

RV PV

MPA

(a)

Ao

(b)

Figure 17.7 (a) Short-axis view at the base of the heart showing the outflow tract of the right ventricle and the pulmonary artery around the aorta. The leaflets of the pulmonary valve are echogenic but the infundibulum and the pulmonary artery are of normal size. (b) The pulsed Doppler tracing recorded a few millimeters beyond the pulmonary valve reveals a peak velocity of 1.5 m/s which is above the normal limit of 0.8 m/s. After birth, a gradient of 20 mmHg was found through the pulmonary valve. PV, pulmonary valve; MPA, main pulmonary artery.

258

Fetal Cardiology

flow through the ductus arteriosus to see whether it is retrograde or fully anterograde; (4) measure the ratio of left to right ventricular end-diastolic diameters as well as the mitral and tricuspid annuli; and (5) measure the size of the pulmonary valve annulus. The prenatal prognosis of mild-to-moderate cases of valvular isolated pulmonic stenosis is excellent. A progressive increase in severity of the stenosis is rare, and an elevation of the observed gradient could well be related to the physiological rise in flow through the right side of the heart. Delivery in these cases can be achieved by the vaginal route. Usually, the neonate does not require any specific postnatal care.

In critical stenosis or complete atresia of the pulmonary valve, the anatomy of the lesions varies according to the status of the tricuspid valve.21 In the absence of tricuspid insufficiency, the right ventricle is markedly hypertrophied. The cavity itself is small and almost virtual. On histology, muscle cell disorganization has been described.22 Major anomalies are also present in the coronary arteries, characterized by abnormal origin and distribution, the absence of proximal aortic coronary connections, coronary stenosis, or interruption.23 Sinusoids are sometimes present, connecting the ventricular cavity with the coronary circulation (Figure 17.8). It has been shown experimentally that an increase in pressure within the right

Ao LV

RV

(a)

(b)

D S

(c)

Figure 17.8 (a) Two-dimensional real-time image of the four-chamber view in a 20-week fetus with pulmonary atresia and intact ventricular septum. A hypoplastic and trabeculated right ventricle (RV) is seen with very thickened free walls. The left-sided cavities are dilated. (b) In the same fetus, a color Doppler interrogation recorded in systole reveals a dilated branch of the coronary artery (arrow) which is draining blood from the right ventricular cavity towards the aorta (Ao) suggesting the persistence of sinusoids within the myocardium. (c) A pulsed Doppler recording of flow velocities within the same dilated coronary artery seen in (b) (but with the cardiac apex pointing towards the top of the picture) shows that in systole (S) the blood is circulating from the myocardium toward the aorta at the peak velocity of 3 m/s (gradient of 43 mm between right ventricle and aorta); in diastole (D), a reverse flow is noted from the aorta toward the right ventricle.

Anomalies of the right heart

ventricular cavity of fetal lambs was associated with a marked accumulation of free oxygen radicals within the myocardium.24 A link between these findings and the histopathologic anomalies seen in pulmonary atresia with intact ventricular septum is a possibility which deserves investigation. By contrast, in pulmonary atresia associated with significant tricuspid regurgitation, the blood flowing back and forth through the incompetent valve causes some growth of the ventricular cavity and prevents a significant increase in ventricular pressure above the systemic level. In these cases, the myocardium appears less hypertrophied and with almost normal organization of myocytes. Sinusoids are usually absent (Figure 17.9). In both cases, the diagnosis can be made quite easily at the first-line routine screening. The four-chamber view is indeed abnormal, and shows variable degrees of right ventricular hypoplasia. In addition, the increase in rightto-left shunting through the foramen ovale is responsible for enlargement of the left-sided cavities. The cardiologist assessing a fetus with critical pulmonic stenosis and intact ventricular septum must study the tricuspid valve morphology and rule out significant leaks; in the presence of tricuspid regurgitation, he must (1) evaluate the severity of insufficiency to establish the type of pulmonary atresia; (2) examine carefully the right ventricular myocardium, using the color Doppler mode, looking for sinusoids (Figure 17.8); (3) determine the orientation of flow through the ductus arteriosus as an important element of diagnosis, since retrograde flow should be expected; (4) measure the diameter of the pulmonary valve annulus compared to that of the aorta; the size of the main pulmonary artery should also be

259

assessed; this evaluation will influence postnatal management inasmuch as it will help to establish the indication of perforation followed by dilatation of the valve in the cardiac catheterization laboratory (Figure 17.9); (5) evaluate the size and function of the left ventricle as well as the size of and flow through the foramen ovale; the intrauterine prognosis of critical stenosis or complete atresia of the pulmonary valve depends on both the size of the foramen ovale and left ventricular function. The 18 fetuses seen in our unit with isolated pulmonary atresia or critical pulmonary stenosis with intact septum were all in good clinical condition, suggesting that they represented survivors from what was a wider spectrum from which fetuses with restrictive foramen ovale were eliminated due to intrauterine death early in gestation. The mode of delivery will be decided according to the functional condition of the left ventricle. Usually, vaginal delivery is possible. Any sign of left ventricular dysfunction, evidence of progressive flow restriction through the foramen ovale, or fluid accumulation in either the pleural or the peritoneal cavity should be an indication for cesarean section. During the immediate postnatal period, pulmonary atresia with intact septum represents a typical example where arterial oxygenation is entirely dependent on patency of the ductus arteriosus. Urgent initial treatment is, therefore, the intravenous administration of prostaglandin. Subsequent management of the neonate will be based on echocardiographic evidence of a right ventricular cavity with sufficient functional ability to warrant balloon dilatation of the valve (Figure 17.9). The presence of sinusoids establishing wide communication between the ventricular cavity and the coronary

LV RV

TR

(a)

(b)

Figure 17.9 (a) Real-time image of the heart of a fetus with pulmonary valvular atresia and severe tricuspid regurgitation. In this view, the right ventricle (RV) appears trabeculated but relatively well developed and the diameter of the tricuspid annulus is only slightly smaller than that of the mitral valve. (b) Recording of pulsed Doppler in the right atrium showing a marked tricuspid regurgitation (TR) with a peak velocity of 5.8 m/s.

260

Fetal Cardiology

circulation is an absolute contraindication to any attempt at opening the right ventricular outlet; in these cases, surgical palliation will have to be considered. Significant coronary artery stenosis is another factor which will influence survival.

With ventricular septal defect Classical tetralogy of Fallot This malformation is a consequence of anterior displacement of the infundibular septum, causing narrowing of the right ventricular outlet chamber. The ventricular septal defect is large, because of the septal malalignment. The aorta is dilated. The sizes of the pulmonary annulus and artery vary according to the severity of the stenosis. Characteristically, they are always smaller than the aortic annulus and ascending aorta. Tetralogy of Fallot can easily be missed during first-line sonographic screening unless adequate sonographic views are recorded. If the four-chamber view is obtained posteriorly, at the level of the AV valves, the septum will look perfectly intact and the view will be described as normal (Figure 17.10a). The relative position of the great arteries will, of course, be normal. Prenatal diagnosis of tetralogy of Fallot is established on a five-chamber view where the aorta appears dilated, overriding the septum (Figure 17.10b). In this view, the subaortic ventricular septal defect is usually obvious. When the diagnosis of tetralogy of Fallot is confirmed, the second-line sonographer must specify the following criteria of severity: (1) the ratio of pulmonary over aortic annulus diameters; (2) the degree of right ventricular outflow tract obstruction; Doppler investigation is unreliable for this purpose since the amount of blood which flows through the infundibulum is inversely related to the severity of the stenosis; this is due to the presence of the ventricular septal defect which allows decompression of the right ventricle towards the aorta; only visual assessment is therefore possible, preferably on the short-axis view at the base of the heart (Figure 17.10c); (3) the pattern of flow through the ductus arteriosus; this should be one of the best criteria for severity assessment: reverse flow will be observed through the ductus arteriosus only if right ventricular output is not sufficient to maintain normal flow through the lungs. The clinical condition of fetuses with tetralogy of Fallot is usually good. Combined cardiac output remains within normal limits. The only change will be the significant increase of the portion of combined output that comes from the ascending aorta compared to the pulmonary artery. Consequently, the aortic arch and isthmus are well developed and coarctation of the aorta is, for all practical purpose, never described in association with tetralogy of Fallot (Figure 17.10d). All of our 69 patients with this classical form of tetralogy of Fallot had no sign of circulatory

failure. Eight of the 69 fetuses also presented a complete AV canal. This good fetal hemodynamic picture can, however, be darkened by the relatively high incidence of chromosomal and extracardiac anomalies in this group.25 Microdeletion of chromosome 22 may be found in up to 20% of infants with tetralogy of Fallot.26 There is, however, no indication, in isolated tetralogy of Fallot, for extraction by cesarean section. The immediate postnatal condition and management will be influenced by the severity of the stenosis and associated extracardiac anomalies. Cases with decreased or diastolic retrograde flow through the ductus arteriosus in prenatal life can be expected to require specialized care immediately after birth to maintain the ductus open. Ideally, all fetuses with tetralogy of Fallot should be delivered in a tertiary center; even in less severe cases, close monitoring of the clinical condition is justified after normal closure of the ductus arteriosus.

Tetralogy of Fallot with pulmonary valve atresia This is considered an extreme form of the disease characterized by complete obstruction of the right ventricular outlet. In this condition, the main pulmonary artery is small or sometimes absent. Perfusion of the lungs comes from the ductus arteriosus, which appears tortuous, and from aortopulmonary collaterals. This form of tetralogy of Fallot is less frequent; we identified eight such cases in our files. Prenatal ultrasonographic diagnosis is based first on an abnormal five-chamber view showing the very dilated aorta overriding the septum with a large ventricular septal defect. This picture is sometimes misinterpreted as being that of a truncus arteriosus. This latter diagnosis is ruled out by identification of the blind outlet of the right ventricle followed by a small pulmonary artery. In cases of truncus arteriosus, the pulmonary artery arises directly from the trunk. The second element of diagnosis of tetralogy of Fallot with pulmonary atresia is the presence of aortopulmonary collaterals which can be seen in prenatal ultrasonographic studies coming from the thoracic descending aorta (Figure 17.11). The cardiocirculatory condition of these fetuses is usually good for the same reasons as given for the classical form of tetralogy of Fallot. These fetuses are usually able to support vaginal delivery. They must, however, undergo cardiocirculatory evaluation immediately after birth to assess the efficacy of the collaterals and of the ductus arteriosus in maintaining adequate pulmonary blood flow and sufficient oxygenation of arterial blood.

Tetralogy of Fallot with absent or dysplastic pulmonary valve This entity, although rare, needs to be identified during prenatal screening. Its anatomic features are: large ventricular septal defect with an overriding aorta, aneurysmal

Anomalies of the right heart

261

LV LEFT

Ao

Septum

RIGHT

(a)

(b)

(c)

(d)

Figure 17.10 These images were obtained from a fetus with the classical form of tetralogy of Fallot. (a) This four-chamber view taken posteriorly at the level of the atrioventricular (AV) valves could be described as normal. (b) On the five-chamber view, the subaortic ventricular septal defect is clearly observed as well as the dilated ascending aorta (Ao) and the septoaortic discontinuity. (c) A truncus arteriosus is ruled out and the diagnosis of tetralogy of Fallot established by the identification of a small pulmonary valve annulus (arrow) connected to the right ventricle, associated with pulmonary artery hypoplasia. (d) Sagittal view of the same fetus showing the aortic arch which is dilated. The reduction in diameter usually observed at the level of the isthmus is not present.

dilatation of the pulmonary trunk, appearance of the dysplastic or rudimentary pulmonary valve, absence of ductus arteriosus, and, finally, right ventricular dilatation, which is frequently observed, unlike in the classical form of tetralogy of Fallot. Studies of rat fetuses with tetralogy of Fallot and absent pulmonary valve created by maternal administration of bis-diamine led to the conclusion that neural crest cells are important in formation of the pulmonary valve and that enlargement of the pulmonary arteries and bronchial compression develop in fetal life.27 Ultrasonographic diagnosis of this form of tetralogy of Fallot is easier than in the classical form.28 During routine ultrasonographic screening, suspicion is raised by dilatation of the right ventricular cavity on four-chamber view. Here again, the diagnosis is confirmed on five-chamber

view. Close examination of the right ventricular outflow tract will show the dilated pulmonary artery with abnormal or absent pulmonary valves. Doppler investigation of the dilated pulmonary artery will disclose systolic antegrade and diastolic retrograde flows caused by severe pulmonary insufficiency (Figure 17.12). In the lungs, the dilated peripheral pulmonary artery can be clearly seen with color Doppler interrogation. Careful investigation with both real-time and color Doppler echocardiography will confirm absence of the ductus arteriosus in the majority of cases. The intrauterine prognosis of tetralogy of Fallot with absent pulmonary valves is uncertain. The two cases from our database showed no sign of cardiocirculatory failure throughout pregnancy. There are, however, in the

262

Fetal Cardiology

literature reported cases that developed polyhydramnios29 and/or hydrops fetalis.30 Tracheobronchial and esophageal compression by the dilated pulmonary artery is considered to be the etiology of the polyhydramnios. On the other hand, the to-and-fro movement of blood between the right ventricle and the dilated pulmonary artery may result in elevated right ventricular end-diastolic pressure, leading to a possible increase in venous systemic pressure and hydrops fetalis. Major respiratory problems occur after birth due to the associated abnormal development of

bronchi which, in addition, are frequently compressed by the markedly dilated pulmonary artery.

Anomalies at the myocardial level Primary anomalies Uhl’s anomaly

Desc Ao

Collat

Figure 17.11 The color Doppler investigation of this fetus with tetralogy of Fallot and pulmonary atresia discloses the presence of aortopulmonary collaterals (Collat) originating from the thoracic descending aorta (Desc Ao).

This myocardial impairment specifically involving the right ventricle has been rarely described in prenatal life. It is essentially characterized by the lack of development of myocytes.31 On macrocospic examination, the right ventricular free wall appears thin and translucid. On microscopic studies, only the pericardial and endocardial layers are present with some adipose tissue and a few myocytes. The right ventricular cavity is markedly enlarged and hypotonic. Echocardiographic diagnosis of Uhl’s anomaly is relatively easy. On first-line screening examination, the four-chamber view will be abnormal. Suspicion of the specific diagnosis will be based on the thinness of the right ventricular free wall with the absence of any apparent contraction (Figure 17.13a). Despite this major myocardial impairment and systemic pressure in the fetal pulmonary circulation, anterograde flow through the pulmonary valve has been documented in our two cases (Figure 17.13b), but in both cases the same flow pattern was observed in

PV AscAo D

Desc.Ao

S

(a)

(b)

Figure 17.12 Ultrasonographic images taken from a fetus with tetralogy of Fallot with rudimentary pulmonary valve. (a) Short-axis view at the basis of the heart showing the pulmonary valve (PV) which is echogenic and dysplastic; the ascending aorta (AscAo) is dilated; the main pulmonary artery and both branches are well visualized. Note the absence of a ductus arteriosus between the pulmonary artery and descending aorta (Desc.Ao). (b) Pulsed Doppler recording in the main pulmonary artery above the valve illustrating the systolic anterograde (S) and diastolic retrograde (D) velocities.

Anomalies of the right heart

PV

263

RV

PA

Ductus arteriosus

DA

(a)

(b)

Figure 17.13 Images of a fetus with Uhl’s anomaly. (a) The short-axis view reveals marked dilatation of the right ventricle (RV) and the thinness of the myocardium. With color Doppler interrogation, it is possible to observe some anterograde flow (blue color) through the pulmonary artery (PA); retrograde flow (red color) coming from the ductus arteriosus (DA) is however also documented. (b) Pulsed Doppler flow velocity recording in the ductus arteriosus shows reverse flow early in systole (above the zero-velocity line) followed by the normal anterograde flow in end-systole and during diastole. This pattern is compatible with a late right ventricular contraction (right bundle branch block).

the ductus arteriosus, characterized by an early systolic shunt from aorta to pulmonary artery followed by a normal anterograde flow in late systole and during diastole (Figure 17.13b). After birth, the right ventricular performance is, however, facilitated because of the fall in pulmonary vascular resistances. Another hemodynamic problem is related to frequently associated tricuspid insufficiency. To establish postnatal prognosis, the second-line sonographer will have to assess flow through the main pulmonary artery as well as the size and function of the left ventricle. A decision concerning the type of delivery will be based on the clinical condition of the fetus. Those fetuses who were able to reach term without evidence of circulatory failure should, in theory, be able to tolerate vaginal delivery. After birth, assessment of the capability of the right ventricle to maintain pulmonary flow is mandatory. Shortand long-term prognosis is reserved. One of our two cases died suddenly at home at the age of 14 months, presumably from arrhythmia. It is not known whether adult patients reported with this disease had the full pathological picture since birth.31 Morbidity and mortality later in life are frequently related to major ventricular arrhythmia, and this anomaly is alternatively named ventricular dysplasia or arrhythmogenic right ventricle.

Secondary anomalies Some fetal conditions can cause secondary myocardial changes which predominantly involve the right ventricle. These conditions deserve mention since they are not infrequent.

Restrictive ductus arteriosus During prenatal life, the normal pulmonary arch is made up of the main pulmonary artery and ductus arteriosus. The ductal segment can be absent in certain types of malformation, as mentioned in the section on tetralogy of Fallot. More frequently, it can become restrictive after maternal administration of prostaglandin synthetase inhibitors, such as indomethacin, used for suppression of premature labor or for reduction of polyhydramnios. A similar effect on the ductus arteriosus has been reported with the administration of betamethasone,32 a glucocorticoid frequently given for fetal lung maturity induction. Any decrease in the diameter of the ductus will cause a gradient between the main pulmonary artery and the aorta; suprasystemic pressure will then be present not only in the pulmonary circulation but also in the right ventricle. Progressive right ventricular hypertrophy associated with tricuspid regurgitation is then observed. Ultrasonographic diagnosis of restrictive ductus arteriosus is based on both bidimensional imaging and Doppler flow velocity signal through the ductus. The real-time picture frequently shows an hourglass appearance of the ductus (Figure 17.14a). Using the Doppler technique, peak systolic velocities > 140 cm/s, diastolic velocities > 35 cm/s, and pulsatility index (systolic velocity − diastolic velocity/ mean velocity) < 1.9 have been proposed as expressions of ductal constriction.33 An example of these changes is shown in Figure 17.14b. These fetuses must be identified quickly, since cardiocirculatory failure has been described in this condition.34 The prognosis, however, is good as long as the medication

264

Fetal Cardiology

is discontinued; disappearance of Doppler evidence of ductal constriction is usually observed within 24–48 hours after the arrest of medication.

The twin-to-twin transfusion syndrome This syndrome is observed in monochorionic twins and is characterized by arteriovenous connections within the placenta, causing passage of blood from one fetus (the donor) to the other (the recipient). The recipient twin

becomes polycythemic and plethoric. One of the features of such a condition is that the myocardium of the recipient twin tends to become thickened and echogenic. Although this myopathy involves both sides of the heart, cases of subpulmonary stenosis have been described.35 Figure 17.15 illustrates such right ventricular outlet stenosis in a case of twin–twin transfusion. Pressure rather than simple volume overload is increasingly considered as the most likely explanation for this cardiomyopathy. Elevated concentration of endothelin in the recipient,36 upregulation of the renin–angiotensin system in the donor,37 and early

DA PA

(a)

(b)

Figure 17.14 (a) Color Doppler flow imaging of the ductus arteriosus of a fetus on indomethacin for polyhydramnios. Note the hourglass appearance of the ductus arteriosus. (b) In this tracing taken in the same fetus, the pulsed Doppler sample volume encompasses both the pulmonary artery (PA) and the ductus arteriosus (DA). The flow velocity waveforms of the two arterial segments are superimposed and reveal a significant increase in peak velocities through the ductus, both in systole and in diastole.

Pulm. Artery

RV S

LV

(a)

(b)

Figure 17.15 (a) Four-chamber view of the heart of the recipient twin in the context of a twin-to-twin transfusion syndrome. Note the thickening of the septum and the free walls of the ventricle. S, septum. (b) Pulsed Doppler taken above the pulmonary valve of the same recipient twin showing an increase in peak velocity of 1.5 m/s.

Anomalies of the right heart

diastolic functional impairment in the recipient twin38 are all elements in favor of the pressure-overload pathogenic concept. Both myopathy and stenosis usually regress after birth, but a few cases have been described where progression of right ventricular outflow tract obstruction was observed.39

Diastolic overload of the right ventricle Intracerebral arteriovenous fistula. Overload of the right ventricle can be observed in all conditions affecting fetal left ventricular function because of the parallel disposition of the two ventricles. These conditions, such as hypoplastic left heart syndrome,

265

mitral disease, restrictive foramen ovale, or coarctation of the aorta, will be discussed in Chapter 20. Some prenatal circulatory anomalies, however, could cause isolated right ventricular diastolic overload while the left heart is normal. At birth, the neonates will show essentially right ventricular hypertrophy on electrocardiogram. This situation is typically observed in intracranial cerebral arteriovenous fistula. This most frequent form of cerebral fistula is characterized by aneurysmal dilatation of the vein of Galen draining a network of cerebral arteries. The degree of shunting through the fistula is variable but, for obvious reasons, the greater are the shunts, the earlier will be the time of appearance of neurologic and cardiocirculatory symptoms. All five cases of aneurysm of the vein of Galen found in our database had massive arteriovenous shunting (Figure 17.16a and 17.16b).

SVC

(a)

(b)

R

Isthmus

L

(c)

DA

(d)

Figure 17.16 (a) Color Doppler interrogation in a fetus with aneurysm of the vein of Galen. Direct connection of an anterior cerebral artery (arrow) into the dilated vein can be seen causing a jet of blood curling around into the aneurysm. (b) Because of the direct arteriovenous connection, the flow velocities through the superior vena cava (SVC) present an arterial pattern with marked forward flow during diastole. (c) In the same fetus, the right-sided cavities are mildly dilated. R, right; L, left. (d) Simultaneous recording of Doppler flow velocities in both the aortic isthmus and the ductus arteriosus illustrates the clear preponderance of the right ventricular ejection volume: blood flow is retrograde in the isthmus during the second half of systole and the entire diastole.

266

Fetal Cardiology

The prenatal hemodynamic changes are related to the marked increase in volume of flow drained by the superior vena cava going into right-sided cavities (Figure 17.16b). Besides causing dilatation of the right atrium and ventricle (Figure 17.16c), this volume overload is transmitted to the pulmonary circulation and ductus arteriosus. Retrograde diastolic flow of blood coming from the ductus arteriosus is observed through the aortic isthmus40 (Figure 17.16c and 17.16d). The prognosis of cerebral arteriovenous fistula diagnosed in prenatal life is somber. In addition to cerebral damage, intrauterine cardiac failure is a frequent compli-

cation. A team approach is essential for the management of both fetuses and neonates.

Anomalous pulmonary venous drainage: These anomalies are described in more detail in Chapter 29. The common hemodynamic feature of all variants of partial anomalous pulmonary venous connection is the diastolic overload of the right-sided cavities. Figure 17.17 is an example of an anomalous connection of the right pulmonary veins to the superior vena cava observed in a

RV LV

RA

RV

(a)

(b)

RA

SVC PV

(c)

(d)

Figure 17.17 Echocardiographic findings in a 25-week fetus with anomalous drainage of the right pulmonary vein into the superior vena cava (SVC). (a) The four-chamber view is abnormal, showing right atrial and ventricular enlargement. (b) The M-mode tracing confirms the right ventricular dilatation with a flat septum. (c) On the two-dimensional image of the SVC, a suspicious vascular imaging (arrowhead) can be noted at the junction of the SVC with the right atrium (RA). (d) Color Doppler shows clearly at least two pulmonary veins (PV) (colored in red) coming from the right lung, draining into the SVC very close to its junction with the right atrium. The aortic blood (colored in blue) is also visualized, parallel to the SVC.

Anomalies of the right heart

25-week fetus. In the presence of a four-chamber view showing right atrial and ventricular enlargement, identification and connections of the pulmonary veins are mandatory. These veins are relatively easy to follow in prenatal life because of the fluid-filled lungs. Color Doppler is very helpful in this process. Total supradiaphragmatic anomalous venous drainages are frequently associated with other cardiac malformations such as atrial isomerism with intracardiac shunts. These associated malformations are responsible for biventricular dilatation.

14.

15.

16.

17.

References 1. Kenny JF, Plappert E, Doubilet P et al. Changes in intracardiac blood flow velocities and right and left ventricular stroke volume with gestational age in the normal human fetus: a prospective Doppler echocardiographic study. Circulation 1986; 74: 1208–16. 2. Phillipos EZ, Robertson MA, Still KD. The echocardiographic assessment of the human fetal foramen ovale. J Am Soc Echocardiogr 1994; 7: 257–63. 3. Wilson AD, Syamasundar Rao P, Aeschlimann S. Normal fetal foramen flap and transatrial Doppler velocity pattern. J Am Soc Echocardiogr 1990; 3: 491–4. 4. Kiserud T. Fetal venous circulation – an update on hemodynamics. J Perinat Med 2000; 28: 90–6. 5. Blaysat G, Villain E, Marçon F et al. Prognosis and outcome of idiopathic dilatation of the right atrium in children: a cooperative study of 15 cases. Arch Mal Cœur 1997; 90: 645–8. 6. Kozlj M, Angelski R, Pavenik D, Zorman D. Idiopathic enlargement of the right atrium. Pediatr Cardiol 1998; 19: 420–1. 7. Pernot C, Hoeffel JC, Jacob F. Idiopathic dilatation of the right atrium revealed in childhood by dysrhythmias. Pediatr Radiol 1977; 6: 52–4. 8. Marin-Garcia J, Allen RG. Idiopathic dilatation of the right atrium: postoperative follow-up in a child. J Pediatr Surg 1983; 18: 196–8. 9. DaSilva AM, Witsemburg M, Elzenza N, Stewart P. Idiopathic dilatation of the right atrium diagnosed in utero. Rev Port Cardiol 1992; 11: 161–3. 10. Reinhardt-Owlya L, Sekarski N, Hurni M et al. Idiopathic dilatation of the right atrium simulating Ebstein’s anomaly. A propos of a case diagnosed in utero. Arch Mal Cœur Vaisseaux 1998; 91: 645–9. 11. Respondek ML, Kammermeier M, Ludomirsky A et al. The prevalence and clinical significance of fetal tricuspid valve regurgitation with normal heart anatomy. Am J Obstet Gynecol 1994; 171: 1265–70. 12. Hornberger LK, Sahn DJ, Kleinman CS et al. Tricuspid valve disease with significant tricuspid insufficiency in the fetus: diagnosis and outcome. J Am Coll Cardiol 1991; 17: 167–73. 13. Sharland GK, Cook AC. Pulmonary atresia with intact interventricular septum: diagnosis in the fetus. In: Redington

18. 19.

20.

21.

22.

23.

24.

25. 26.

27.

28.

29. 30.

267

AN, Brawn WJ, Deanfield JE, Anderson RH, eds. The Right Heart in Congenital Heart Disease. London: Greenwich Medical Media, 1998: 25–34. Zuberbuhler JR, Allwork SP, Anderson RH. The spectrum of Ebstein’s anomaly of the tricuspid valve. J Thorac Cardiovasc Surg 1979; 77: 202–1. Celermayer DS, Cullen S, Sullivan ID et al. Outcome in neonates with Ebstein’s anomaly. J Am Coll Cardiol 1992; 19: 1041–6. Pavlova M, Fouron JC, Drblik SP et al. Factors affecting the prognosis of Ebstein’s anomaly during fetal life. Am Heart J 1998; 135: 1081–5. Paladini D, Chita SK, Allan LD. Prenatal measurement of cardiothoracic ratio in evaluation of heart disease. Arch Dis Child 1990; 65: 20–3. Edwards JE, Burchell HB. Congenital tricuspid atresia: a classification. Med Clin North Am 1949; 33: 1117–19. Castor S, Fouron JC, Teyssier G et al. Assessment of fetal pulmonic stenosis by ultrasonography. J Am Soc Echocardiogr 1996; 9: 805–13. Lima CO, Sahn DJ, Valdès-Cruz LM et al. Noninvasive prediction of transvalvular pressure gradient in patients with pulmonary stenosis by quantitative two-dimensional echocardiographic Doppler studies. Circulation 1983; 67: 866–71. Davignon AL, Greewold WE, DuShane JW, Edwards JE. Congenital pulmonary atresia with intact ventricular septum: clinicopathologic correlation of two anatomic types. Am Heart J 1961; 62: 591–602. Freedom RM, Wilson GJ. The anatomic substrate of pulmonary atresia and intact ventricular septum. In: Tucker BL, Lindesmith GC, Takahashi M, eds. Third Clinical on Congenital Heart Disease: Obstructive Lesions of the Right Heart. Baltimore: University Park Press, 1984: 217–55. Kasznica J, Ursell PC, Blanc WA, Gersony WM. Abnormalities of the coronary circulation in pulmonary atresia and intact ventricular septum. Am Heart J 1987; 114: 1415–20. Fouron JC, Chemtob S, Chartrand C et al. Generation of reactive O2 species in the myocardium of newborn lambs following intra-uterine increase in right ventricular pressure. Pediatr Cardiol 2001; 22: 143–6. Allan LD, Sharland GK. Prognosis in fetal tetralogy of Fallot. Pediatr Cardiol 1992; 13: 1–4. Webber SA, Hatchwell E, Barber JC et al. Importance of microdeletions of chromosomal region 22q11 as a cause of selected malformations of the ventricular outflow tracts and aortic arch: a three-year prospective study. J Pediatr 1996; 129: 26–32. Momma K, Ando M, Takao A. Fetal cardiac morphology of tetralogy of Fallot with absent pulmonary valve in the rat. Circulation 1990; 82: 1343–51. Fouron JC, Sahn D, Bender R et al. Prenatal diagnosis and circulatory characteristics in tetralogy of Fallot with absent pulmonary valve. Am J Cardiol 1989; 64: 547–9. Callan NA, Kan JS. Prenatal diagnosis of tetralogy of Fallot with absent pulmonary valve. Am J Perinat 1991; 8: 15–17. Sameshima H, Nishibatake M, Ninomiya Y, Tokudome T. Antenatal diagnosis of tetralogy of Fallot with absent pulmonary valve accompanied by hydrops fetalis and polyhydramnios. Fetal Diagn Ther 1993; 8: 305–8.

268

Fetal Cardiology

31. Marcus FI, Fontaine GH, Guiraudon G et al. Right ventricular dysplasia: a report of 24 adult cases. Circulation 1982; 65: 384–98. 32. Momma K, Takao A. Increased constriction of the ductus arteriosus with combined administration of indomethacin and betamethasone in fetal rats. Pediatr Res 1989; 25: 69–75. 33. Huhta JC, Moise KJ, Fisher DJ et al. Detection and quantitation of constriction of the fetal ductus arteriosus by Doppler echocardiography. Circulation 1987; 75: 406–12. 34. Murray HG, Stone PR, Strand L, Flower J. Fetal pleural effusion following maternal indomethacin therapy. Br J Obstet Gynaecol 1993; 100: 277–82. 35. Zosmer N, Bajoria R, Weiner E et al. Clinical and echographic features of in utero cardiac dysfunction in the recipient twin in twin-twin transfusion syndrome. Br Heart J 1994; 72: 74–9.

36. Bajoria R, Sullivan M, Fisk NM. Endothelin in association with cardiac dysfunction in the recipient fetus to twin-twin transfusion syndrome. Hum Reprod 1999; 14: 1614–18. 37. Mahieu-Caputo D, Muller F, Joly D et al. Pathogenesis of twin-twin transfusion syndrome: the renin-angiotensin system hypothesis. Fetal Diagn Ther 2000; 16: 241–4. 38. Raboisson MJ, Fouron JC, Lamoureux J et al. Early intertwin differences in myocardial performance during the twin-to-twin transfusion syndrome. Circulation 2004; 110: 3043–8. 39. Lougheed J, Sinclair B, Fung KFK et al. Acquired right ventricular outflow tract obstruction in twin-twin transfusion syndrome. J Am Coll Cardiol 1999; 33: 536A. 40. Patton DJ, Fouron JC. Cerebral arteriovenous malformation: comparison of pre- and postnatal central blood flow dynamics. Pediatr Cardiol 1995; 16: 141–4.

18 Pulmonary atresia with intact ventricular septum Julene S Carvalho

Introduction Pulmonary atresia with intact ventricular septum (PAIVS) is a rare abnormality but one that has fascinated pediatric cardiologists over the years, both for the heterogeneity of the anatomical findings and for the challenges posed regarding optimal management to improve actuarial survival and to reduce morbidity. First descriptions of the anomaly date back to the late 1700s and mid-1800s.1,2 It is, however, relatively recent publications that have contributed to our understanding of the anatomical variability and the pathophysiological implications. Not infrequently, PAIVS is described as ‘hypoplastic right heart’, but this terminology is best avoided, as the right ventricle (RV) is not always hypoplastic and right ventricular hypoplasia is not pathognomonic of PAIVS. Essentially, in PAIVS there is complete obstruction to the pulmonary outflow tract (at valvar or subvalvar level) in the presence of an intact ventricular septum. This is usually associated with underdevelopment of the right ventricular cavity which is hypertrophied, muscle-bound, and hypertensive. Coronary abnormalities are common. However, variations do occur. In a minority of cases, there is important tricuspid regurgitation, the RV is dilated and may be thin-walled with low pressure. PAIVS with a very small RV is commonly diagnosed in mid-gestation due to an abnormal four-chamber view (Figures 18.1a and 18.1b). This diagnosis can also be made in the first trimester as either atresia or critical stenosis (Figure 18.2). On the other hand, it is not uncommon for the RV to be of a relatively good size at 18–22 weeks, and such examples may be overlooked in screening programs (Figure 18.3a). This is of importance, as RV growth may be subsequently impaired leading to a much smaller ventricular size at birth. Thus, fetal intervention to relieve the outflow tract obstruction may be considered in cases of either valve atresia or critical pulmonary stenosis in the hope that this may preserve RV size and function. Whatever the morphological variations and prenatal history, the neonate with PAIVS will present with a ductdependent circulation, and thus will require immediate

medical/surgical attention to maintain hemodynamic stability. Management strategies will depend on the nature of outflow tract obstruction and adequacy (or not) of the right ventricular structures to maintain (or not) a biventricular circulation in the medium to long term.

Incidence PAIVS accounts for ∼1–3% of babies with congenital heart disease,3,4 whereas the incidence at birth obtained from a 20-year population-based study in Sweden (1980–1999) is 4.2 cases per 100 000 live births, with similar sex distribution.5 An equivalent overall incidence of 4.5 per 100 000 live births (186 cases) was seen in the UK–Eire collaborative multicenter study (1991–1995), which is also population-based.6 Higher incidences of 7.1 and 8.1 per 100 000 live births were previously reported by the New England Infant Cardiac Program3 and the Baltimore–Washington Infant study,7 respectively. In the UK–Eire series, significant regional variations did occur, being lower in England and Wales and higher in Eire and Northern Ireland. As this latter study also included fetal cases of PAIVS, the influence of prenatal diagnosis on postnatal incidence can be appreciated. Pregnancy was terminated in 61% of the 86 cases diagnosed prenatally and a further 4% of fetuses died spontaneously. Live birth incidence would have been higher at 5.6 per 100 000 in England and Wales, but unchanged in Eire and Northern Ireland, where no pregnancies were terminated. This clearly shows the impact of fetal diagnosis on postnatal incidence.

Anatomical findings and ultrasound correlates In sequential segmental analysis,8 PAIVS usually occurs in the setting of normal atrial situs with concordant cardiac

270

Fetal Cardiology

Figure 18.1 Images obtained from a 21-week fetus with pulmonary atresia with intact ventricular septum (PAIVS) and small right ventricle (RV). (a, b) Four-chamber views show a very small RV cavity (indicated by the asterisk) and minute tricuspid valve (TV) (distance between calipers = 1.6 mm). (c) Abnormal pulsed wave Doppler across TV shows a monophasic pattern, of high velocity and short duration. (d, e) Note the abnormal position of the duct (arrows) which arises from underneath the aortic arch and shows retrograde flow on color Doppler. In the longitudinal view of the aortic arch (d) the duct forms an acute angle (arrow) with the aorta (Ao) and on transverse view (e) its origin is seen at a level below the aortic arch, mid-point between the proximal and distal transverse arch (arrow). The main pulmonary artery is not seen. (f) Pulsed wave Doppler across the ductus venosus shows an abnormal pattern with reversed flow during atrial contraction. post, posterior; LV, left ventricle; ant, anterior.

connections and a left-sided aortic arch. It can also be seen in association with discordant atrioventricular and ventriculoarterial connections.9 Although this is a rare occurrence, we have observed one case in our prenatal series. The morphological spectrum of PAIVS varies from the commonest form with a small right ventricular cavity to that of a dilated RV. Consequently, the cardiothoracic ratio is normal in the majority of fetuses but there may be

variable degrees of cardiomegaly. The worst cases of increased cardiothoracic ratio lead to the so-called ‘wallto-wall’ hearts, which are often associated with a poor prognosis.10 Branch pulmonary arteries are usually confluent and supplied by a left-sided arterial duct,11 although non-confluent arteries supplied by bilateral ducts are also reported. The occurrence of systemic-to-pulmonary collateral arteries as source of pulmonary blood flow is rare.12

Pulmonary atresia with intact ventricular septum

271

Figure 18.2 Images obtained from a 14-week fetus with critical pulmonary stenosis. (a, b) Four-chamber views in systole (a, right and left panels) and diastole (b). (a) Note the small RV cavity and tricuspid regurgitation of color flow mapping. (b) Color flow demarcates the LV cavity only. The asterisk indicates the RV cavity. (c) Upper mediastinal view shows an enlarged aorta and retrograde flow across the arterial duct, in keeping with severe pulmonary obstruction. DAo, descending aorta.

The right ventricle, tricuspid valve, and RV infundibulum in PAIVS Although the three components of the RV (inlet, trabecular, and outlet) are present in hearts with PAIVS,13 there is a variable degree of ventricular hypertrophy and intracavity muscular overgrowth which can lead to obliteration of one or two of these components. A nomenclature that describes this labels ventricles as being ‘unipartite’, ‘bipartite’, or ‘tripartite’.14 Unipartite ventricles are those in which only one portion (inlet) is not obliterated; bipartite ventricles have two of the ventricular components; and a tripartite chamber shows all three portions. At birth, tripartite ventricles are more common (59%) than bipartite (∼34%), whereas unipartite ventricles account for < 8% of cases.15 The tricuspid valve (TV) morphology is also variable. Its diameter is commonly expressed as ‘Z-scores’ or the number of standard deviations by which a measurement

(in this case, TV diameter) deviates from the population mean for a given patient size (in this case, fetal size). In children, Z-scores are often normalized for body surface area, whilst in the fetus, they can be calculated in relation to femur length, biparietal diameter, or gestational age.16,17 The TV can be severely stenotic and, as might be expected, unipartite and bipartite ventricles typically have a small valve. As a group, TV mean Z-score at birth is –5, but again there is considerable variability.15 A severely regurgitant valve is associated with a dilated annulus and is often dysplastic, with or without apical displacement of its leaflets, the latter being a feature of Ebstein’s malformation which occurs in ∼10% of cases.15,18 In the most severe forms with massive tricuspid regurgitation, the RV cavity may be significantly dilated with a thin myocardium and a ‘wall-to-wall’ heart. In some of these cases, the atresia may be functional rather than anatomical. The morphological substrate for the outflow tract obstruction is most commonly at valve level, when there

272

Fetal Cardiology

Figure 18.3 Images obtained from a 25-week fetus with PAIVS with good size RV. (a) Note a near-normal four-chamber view (left panel) and tricuspid regurgitation on color flow mapping (right panel). In real-time scanning, RV contractility and TV opening were impaired. (b) Pulsed wave Doppler shows high velocity tricuspid regurgitation jet. (c,d) Longitudinal views of the aortic arch (c) and RV outflow tract (d) during color flow mapping. Note reversal of flow which starts at the aortic end of the arterial duct (c, indicated by the asterisk) and reaches the level of the pulmonary valve (d, indicated by the arrowhead), compatible with pulmonary atresia. The small color jet observed in (d) originates from the tricuspid regurgitation. LA, left atrium; RA, right atrium; PA, pulmonary artery.

is fusion of the valve leaflets and a patent infundibulum, also called membranous atresia (Figures 18.3d and 18.4d). Valvar atresia accounted for 75% of cases in the UK–Eire study, the remainder showing muscular atresia with infundibular obliteration (Figure 18.1e).15 In this particular series, there were no examples of isolated infundibular obliteration. The anatomical variabity of inlet, trabecular, and outlet components of the RV is reflected by the wide spectrum of ultrasound images observed during fetal echocardiography. On the four-chamber view, the right ventricular cavity may be diminutive due to apical obliteration. This is accompanied by a small and stenotic TV with abnormal Doppler signal, which shows a monophasic waveform of short duration (Figure 18.1a–18.1c). Tricuspid regurgitation,

if present, is trivial. These cases have additional muscular overgrowth of the outflow area, often leading to infundibular atresia, thus characterizing a unipartite ventricle. Occasionally there may be hyperechogenicity of the endocardial wall, which is likely to reflect endocardial fibroelastosis. At the other end of the spectrum, the RV and right atrium are dilated (Figure 18.4). This is usually associated with an abnormality of the TV which may show dysplastic leaflets and rolled edges with or without apical displacement (Ebstein’s malformation). The valve diameter itself may be larger than normal. Color flow mapping shows severe tricuspid regurgitation. Most cases of PAIVS have an elevated RV pressure, and the pulsed or continuous wave Doppler signal is expected to show a high velocity

Pulmonary atresia with intact ventricular septum

273

Figure 18.4 Images obtained from a 22-week fetus with PAIVS and cardiomegaly. (a, b) Four-chamber views show a significantly dilated right atrium and a large RV. Note failure of coaption of the tricuspid valve leaflets (b, left panel) with significant regurgitation on color flow mapping (b, right panel). (c) Pulsed wave Doppler shows high velocity tricuspid regurgitation. (d) Note the normal alignment of pulmonary artery, aorta, and superior vena cava in the three-vessel view (indicated by the asterisks) (d, left panel). Despite the ‘near-normal’ three-vessel view, the pulmonary valve was immobile. Color flow mapping (d, right panel) shows reversal of flow across the arterial duct (red), reaching the atretic valve (arrow) from where flow bounces back to the main PA and its branches (blue). (e) Longitudinal view of the aorta shows the normal position of the arterial duct but reversed flow (indicated by the asterisk). (f) High pulsatility index in the ductus venosus but positive end-diastolic velocity.

tricuspid regurgitation jet (Figure 18.4c) The presence of low velocity regurgitation may imply that pulmonary atresia is functional rather than anatomical, i.e. when there is no effective forward flow through the pulmonary valve due to the massive tricuspid regurgitation. In between these two extremes of ventricular morphology, the RV may show near-normal dimensions on the four-chamber view with reasonably well developed inlet and apical portions (Figure 18.3a). Careful examination, however, will show increased ventricular wall thickness and decreased myocardial contractility. High velocity tricuspid regurgitation is often present, and can be promptly identified on color flow mapping (Figure 18.3b). These ventricles are usually tripartite structures, with atresia at valve level and, almost invariably, a welldeveloped infundibulum (Figure 18.3d). The pulmonary valve shows thickened and imobile leaflets.

Ventriculocoronary connections and the RV-dependent circulation Abnormal connections between the right ventricular cavity and the coronary arterial system were first described

in 1926 in a pathological specimen of PAIVS,19 and at angiography in 1964.20 It was 10 years later that Freedom and Harrington21 suggested that these communications serve as passive conduits to egress of flow from the blindend RV, and may contribute to myocardial ischemia due to the impairment of diastolic coronary flow. Since then their importance to the outcome of children with PAIVS has been increasingly recognized.22 These connections have been described as ‘coronary sinusoids’, but this terminology is best avoided as it is considered to be incorrect.22 Ventriculocoronary arterial connections (VCAC) or fistulas are better descriptions for such communications, which consist of thick-walled, distended intertrabecular myocardial spaces between a more-or-less distended capillary bed and the subepicardial coronary arteries23 (Figures 18.5–18.6). Typically, VCAC are seen in cases of PAIVS with an underdeveloped RV cavity, stenotic TV, and hypertrophied ventricular walls that often obliterate the apical and outlet portions of the ventricles, leading to suprasystemic right ventricular pressures. In pathological fetal series, coronary artery abnormalities were more common in cases with smaller ventricles.24 On the other hand, dilated, thinwalled, and non-hypertensive ventricles with abnormal

274

Fetal Cardiology

Figure 18.5 Images obtained from a 23-week fetus with PAIVS and ventriculocoronary arterial connections (VCAC). (a) Four-chamber view shows a very hypertrophied RV with minute cavity. (b, c) Color flow mapping superior to the five-chamber view level demonstrates the presence of two abnormal vessels running on the right ventricular surface (arrowheads in b and c).

TV leaflets and massive regurgitation are not associated with VCAC. As a consequence of such communications, there may be an RV-dependent coronary circulation, meaning that adequate flow to the coronary arteries depends on the hypertensive RV to drive the blood from its cavity through the VCAC and into the coronary territory. Such a circuit may be compromised when there is RV decompression, for example with surgical or catheter relief of the outflow tract obstruction. An RV-dependent circulation is diagnosed when there is a VCAC but no communication between the coronary artery itself and the aorta, or when there is interruption or significant stenosis of a major epicardial coronary artery. A review of the literature suggests that VCAC occur in about two-fifths of cases of PAIVS, and that approximately one-fifth of these have

an RV-dependent coronary circulation.22 This is similar to the 8% RV-dependent circulation encountered in a population-based study.15 In the fetus, the presence of VCAC can be evaluated systematically when scanning the aortic root and right ventricular myocardium during color flow mapping with the velocity scale set at a relatively low level (∼30 cm/s). This is often obtained in a transverse view of the fetal chest just superior to the five-chamber view (Figure 18.5). Pulsed wave Doppler interrogation shows bidirectional flow which is forward and of lower velocity in diastole, and retrograde and of higher velocity during systole.25 Such connections have been recognized as early as the first trimester of pregnancy,26 but their prenatal identification per se does not allow distinction between those that may have an RVdependent coronary circulation and those that may not.

Pulmonary atresia with intact ventricular septum

The pulmonary arteries Although the presence of confluent pulmonary arteries is the rule in cases of PAIVS, it is important that they are imaged during fetal echocardiography in order to document their size and subsequent growth as pregnancy advances. The branch pulmonary arteries can usually be demonstrated on transverse views of the fetal chest. They are more readily seen in cases of valvar atresia with a thickened pulmonary valve and anatomical continuity between the RV infundibulum and the main pulmonary artery. In such cases of membranous atresia, the main pulmonary artery is often of relatively good size and the arterial duct is in its usual position (Figures 18.3 and 18.4). Thus, the three-vessel view may seem normal on twodimensional imaging (Figure 18.4d). Color flow mapping, however, will show retrograde flow in the arterial duct (Figures 18.3d and 18.4d). On sagittal views, the normal angle (obtuse) between the duct and the descending aorta can be appreciated (Figure 18.4c). This angle is thought to reflect cases where there had been forward flow from the RV to the pulmonary artery at earlier stages of development. That is, there was severe pulmonary stenosis which progressed to atresia.27 Alternatively, the arterial duct may arise from the undersurface of the aortic arch, forming an acute angle with the aorta (Figure 18.1d). This represents the direction of flow from the aorta to the pulmonary artery in early stages of gestation.27 In such cases, the duct is usually a tortuous vessel with a sigmoid shape. This abnormal ductal take-off from the aorta is often seen in cases of muscular pulmonary atresia, and it is often more difficult to visualize the main pulmonary artery. The combination of a small or absent main pulmonary artery and the anomalous position and shape of the arterial duct means that the three-vessel view is abnormal and the arterial duct cannot be imaged in the usual way in this standard plane. Its visualization requires the use of color flow mapping while performing a careful transverse sweep from the level of the proximal ascending aorta (five-chamber view) toward the aortic arch. The duct can then be seen to arise from underneath the aortic arch, mid-point between the proximal transverse arch and the descending aorta (Figure 18.1e).

Pathophysiology and ultrasound findings In PAIVS, the fetal circulation is dependent on the patency of the foramen ovale and arterial duct. Flow from the right atrium is diverted to the left atrium, left ventricle, and aorta. On fetal echocardiography, left ventricular enddiastolic dimension and stroke volume are increased. Accordingly, Doppler velocities across the mitral and

275

aortic valves may also be increased. On real-time twodimensional imaging, left-sided volume-load is more easily observed in cases with minute TV and diminished flow circulating through the right heart than in cases with a reasonable RV size and tricuspid regurgitation. Difficulty with forward flow through the right heart can also be reflected in the ductus venosus Doppler signal, which may show increased pulsatility index or reversal of flow during atrial contraction (Figures 18.1f and 18.4f). However, this finding does not appear to correlate with outcome.28 Color flow mapping will confirm a right-to-left shunt across the atrial septum and left-to-right (reversed) flow across the arterial duct (Figures 18.1–18.4). The foramen ovale is often of adequate size, but restriction to flow at this level may lead to unusually high velocities across it. A restrictive foramen ovale may also contribute to the abnormally low forward velocities or reversed flow in the ductus venosus during atrial systole.

Prenatal natural history and associated abnormalities Progression of pulmonary stenosis in utero is well documented.6,29 In one series, six cases diagnosed as critical stenosis at a mean gestational age of 22 weeks evolved to atresia by 31 weeks.6 Pulmonary stenosis of varying severity, including critical obstruction and PAIVS, is not infrequently diagnosed in twins, especially in the recipient twin of monochorionic diamniotic pregnancies with twin-to-twin transfusion syndrome.30–32 The pathophysiological mechanisms of such an association are not fully understood, and no consensus has been reached as to the exact mechanism leading to the development of pulmonary obstruction. Spontaneous fetal death occurs in a small number of cases, accounting for 5% of ongoing pregnancies in one series.6 A grossly dilated RV with important regurgitation also appears negatively to influence perinatal outcome, it being present in a significant number of cases that died.29,33 Chromosomal and extracardiac abnormalities are uncommon, but examples of trisomy 18, trisomy 21, and 4p deletion as well as Dandy–Walker malformation have been reported.6,25,34 In our prenatal series we have also encountered one case of trisomy 18 in a fetus presenting with increased nuchal translucency and in whom PAIVS was diagnosed at 16 weeks’ gestation.

Fetal intervention Over the last few years there have been isolated reports of prenatal intervention in fetuses with severe valvar

276

Fetal Cardiology

pulmonary stenosis/PAIVS.35–37 Fetal pulmonary valvuloplasty has been carried out in an attempt to preserve RV growth in the hope that this will maximize the chances of achieving a biventricular repair postnatally. In 2006, the United Kingdom National Institute for Clinical Excellence (NICE) issued clinical guidelines on percutaneous fetal balloon valvuloplasty (available at: www.nice. org.uk/guidance). NICE recommendations state that ‘it is difficult to tell how well the procedure works because so few babies have had it and because the problems of the babies before they had the procedure varied.’ Not only are the numbers small, but there are no long-term follow-up data and no clear-cut selection criteria as to which cases may benefit from prenatal intervention. Selected cases are judged on an individual basis. NICE also states that if fetal balloon angioplasty is considered, the procedure should be carried out only in hospitals specializing in fetal interventions, involving a multidisciplinary team, and after full discussion with the family so that they understand the uncertainties about efficacy and safety. Ideally, only fetuses deemed to have postnatal univentricular surgical palliation should have fetal intervention, which is usually aimed at rescuing the RV.38 Thus, there is a need to predict the type of postnatal repair from prenatal ultrasound.38–41 Table 18.1 summarizes the various indexes that have been reported. The use of such predictors may help counseling and may also facilitate decision-making as to which fetuses may benefit from prenatal intervention. They are, however, retrospective studies, and the validity of using such measures to predict outcome can only be proven with large prospective randomized trials.

Postnatal diagnosis and management The neonate presents with cyanosis and, without prenatal diagnosis, may collapse upon constriction of the arterial duct. However, if the diagnosis is made antenatally, postnatal hemodynamic stability is maintained with the elective use of intravenous prostaglandin E, which maintains ductal patency. On auscultation, there is a single second heart sound. The chest radiograph shows oligemic lung fields, often with a normal heart size, although important cardiomegaly is seen in a minority of cases. The 12-lead electrocardiogram is abnormal, and shows dominance of left ventricular forces (S-wave in V1 and R-wave in V6) as opposed to the normal neonatal pattern of RV dominance (R-wave in V1 and S-wave in V6). Postnatal cross-sectional echocardiography quickly establishes the diagnosis if this had not already been made prenatally. The timing and role of diagnostic cardiac catheterization is contentious, but it is performed in

Figure 18.6 Right ventriculogram obtained from a child with PAIVS and VCAC. Note the dilated left coronary artery (LCA) with stenosis at its origin from the aorta.

selected cases, often to delineate coronary arterial abnormalities (Figure 18.6), whereas interventional catheterization is frequently carried out with the intent to perforate and dilate the pulmonary valve in cases of membranous atresia. If necessary, balloon atrial septostomy can be performed at the same time or under ultrasound guidance. This means enlargement of the interatrial communication to ensure an unrestricted right-to-left shunt to avoid development of right atrial hypertension. Once the neonate is stable on a prostaglandin infusion, the next step is to determine whether the baby will be suitable for a biventricular or univentricular repair. When the RV is unipartite and there is muscular infundibular obstruction, or an RV-dependent coronary circulation, a biventricular repair is not possible, and the long-term strategy is geared toward a univentricular circulation. This is when systemic and pulmonary circulations are separated, there is only one functioning ventricle (in this case, the left ventricle), and systemic venous return bypasses the heart. The initial palliation consists of a systemic to pulmonary anastomosis in the first few days of life, often a modified Blalock–Taussig (BT) shunt (i.e. surgical anastomosis with an interposition graft between the subclavian artery and the pulmonary artery), preceded or followed by balloon atrial septostomy or sometimes surgical septectomy (i.e. surgical enlargement of the atrial communication). Subsequent management consists of a bidirectional Glenn procedure (i.e. superior vena cava to right pulmonary artery end-to-side anastomosis) with later placement of an extracardiac conduit between the inferior vena cava and pulmonary artery, thus completing a total cavopulmonary connection (univentricular repair).

Pulmonary atresia with intact ventricular septum

Table 18.1

277

Fetal indicators of postnatal outcome and type of repair

Author

Predictors of outcome

Peterson et al39 (2006)

Indicators of poor outcome

Comments

TV Z-score ≤ −4 after 23 weeks TV annulus ≤ 5 mm after 30 weeks RV/LV length or width < 0.5 and/or no tricuspid regurgitation Salvin et al40 (2006)

Indicators of biventricular repair

More likely to achieve a biventricular repair

TV Z-score > −3 38

Roman et al (2007)

Indicators of univentricular repair

Before 31 weeks, if 3 of 4 criteria were met: 100% sensitivity and 75% specificity for a non-biventricular repair

RV/LV length < 0.6 TV/MV maximum diameter < 0.7 TV inflow < 31.5% cardiac cycle length Presence of coronary fistulas 41

Gardiner et al (2008)

Indicators of biventricular repair

TV Z-score was a good predictor at all gestational ages

PV Z-zcore > −1 or TV Z-score > −3.4 before 23 weeks Median TV Z-score > −3.95 before 26 weeks Median PV Z-score > −2.8 and medium TV/MV ratio > 0.7 at 26–31 weeks Median TV Z-score > − 3.9 and medium TV/MV ratio > 0.59 after 31 weeks TV, tricuspid valve; PV, pulmonary valve; MV, mitral valve; RV, right ventricle; LV, left ventricle. Reproduced with permission from Shinebourne EA, Rigby ML, Carvalho JS. Pulmonary atresia with intact ventricular septum – from fetus to adult. Heart 2008; 94: 1350–7.

In the presence of a tripartite ventricle, with tricuspid Z-score > −2.4 and valvar atresia, the management strategy is for a biventricular repair.42 Nowadays, relief of the outflow obstruction is often achieved during cardiac catheterization, using laser or radiofrequency perforation of the atretic valve followed by standard balloon dilatation.43,44 This is performed percutaneously, usually through the femoral vein. If necessary, a surgical valvotomy can be performed. The perforate anatomical continuity between the RV and the pulmonary artery may suffice to allow adequate antegrade flow. However, in a significant number of neonates, a BT shunt will still be required, as the circulation may remain duct-dependent with persistent cyanosis. The shunt can be closed at a later stage if the RV is considered adequate as the only source of pulmonary blood flow, and a biventricular circulation is then achieved. However, it is not always certain whether the RV is suitable for a biventricular repair, and various criteria have been used for decision making with regard to surgical strategies.45–47 If it is thought to be inadequate, there is the option of a ‘one-and-a-half ’ repair48 in which pulmonary

blood flow is via a combination of forward flow from the RV to the pulmonary artery and a Glenn anastomosis. In the less common form of PAIVS with significant tricuspid regurgitation and a thin-walled RV, the TV can be converted to an atretic valve, this being combined with a modified BT shunt and later univentricular palliation by a total cavopulmonary connection.49

Outcome and follow-up Overall, the outcome of babies with PAIVS has been considered relatively poor, but better understanding of the anatomy and pathophysiology of this condition including the significance of VCAC and RV-dependent circulation has led to improved surgical strategies and outcomes. In the UK–Eire study,50 actuarial survival at 1 and 5 years, excluding those who did not undergo surgery, was 71% and 64%, respectively. The risk of death was higher in the first 6 months, and decreased over the next

278

Fetal Cardiology

12 months, with no deaths occurring after 4 years up to 9 years’ follow-up. Others reported overall actuarial survival of 82%47 and 86%51 at 10 years. A breakdown of surgical results according to type of repair is given by Pawade et al,45 who reported a 10-year actuarial survival of 93% for those with a well developed infundibulum and 75% at 3 years for those without. Following interventional cardiac catheterization, Agnoletti et al52 achieved 85% survival at medium follow-up of 5.5 years, but only 35% were free from surgery. Those with a TV Z-score > −2 were more likely to achieve a biventricular repair. Guleserian et al53 reviewed the medium-term outcome of patients with an RV-dependent coronary circulation whose management was directed to a univentricular circulation. Of thirty-two patients (all had a modified BT shunt), overall mortality was 19%, all deaths occurred within 3 months of the shunt, and all three patients with aortic coronary atresia died. Actuarial survival for all patients was 81% at 5, 10, and 15 years. It is generally assumed that a biventricular repair is preferable to that of a univentricular circulation in terms of exercise tolerance. The evidence for this in PAIVS, however, is limited, as even after a biventricular repair, RV diastolic function may remain markedly abnormal,54 and this may affect exercise performance.55 Similarly, Numata et al48 found no effective difference in exercise capacity at 5 and 10 years after a one-and-a-half repair compared with a univentricular repair.

8.

9.

10.

11.

12.

13. 14.

15.

16.

17.

18.

References 1. Hunter J. Three cases of mal-conformations of the heart. Med Obs Inq 1784; 6: 294. 2. Hare CJ. Malformation of the heart. Complete closure of the orifice of the pulmonary artery. Very small foramen ovale. Cyanosis. Trans Pathol Soc London 1853; 4: 85. 3. Fyler DC. Report of the New England Regional Infant Cardiac Program. Pediatrics 1980; 65 (Suppl): 376–461. 4. Ferencz C, Rubin JD, McCarter RJ et al. Congenital heart disease: prevalence at livebirth. The Baltimore-Washington Infant Study. Am J Epidemiol 1985; 121: 31–6. 5. Ekman Joelsson BM, Sunnegardh J, Hanseus K et al. The outcome of children born with pulmonary atresia and intact ventricular septum in Sweden from 1980 to 1999. Scand Cardiovasc J 2001; 35: 192–8. 6. Daubeney PE, Sharland GK, Cook AC et al. Pulmonary atresia with intact ventricular septum: impact of fetal echocardiography on incidence at birth and postnatal outcome. UK and Eire Collaborative Study of Pulmonary Atresia with Intact Ventricular Septum. Circulation 1998; 98: 562–6. 7. Fyler DC, Neill CA, Ferencz C et al. Infants with congenital heart disease: the cases. In: Ferencz C, Rubin JD, Loffredo CA, Magee CA, eds. Epidemiology of Congenital Heart

19. 20.

21.

22.

23.

24.

Disease: The Baltimore–Washington Infant Study 1981–1989. Mount Kisco, NY: Futura, 1993: 38. Carvalho JS, Ho SY, Shinebourne EA. Sequential segmental analysis in complex fetal cardiac abnormalities: a logical approach to diagnosis. Ultrasound Obstet Gynecol 2005; 26: 105–11. Patel CR, Spector ML, Zahka KG. Congenitally corrected transposition with pulmonary atresia and intact ventricular septum. Cardiol Young 2000; 10: 268–70. Freedom RM, Jaeggi E, Perrin D et al. The ‘wall-to-wall’ heart in the patient with pulmonary atresia and intact ventricular septum. Cardiol Young. 2006; 16: 18–29. Zuberbuhler JR, Anderson RH. Morphological variations in pulmonary atresia with intact ventricular septum. Br Heart J 1979; 41: 281–8. Albanese SB, Carotti A, Toscano A et al. Pulmonary atresia with intact ventricular septum and systemic-pulmonary collateral arteries. Ann Thorac Surg 2002; 73: 1322–4. Anderson RH, Ho SY. Pathologic substrates for 1 1/2 ventricular repair. Ann Thorac Surg 1998; 66: 673–7. Bull C, de Leval MR, Mercanti C et al. Pulmonary atresia and intact ventricular septum: a revised classification. Circulation 1982; 66: 266–72. Daubeney PE, Delany DJ, Anderson RH et al. Pulmonary atresia with intact ventricular septum: range of morphology in a population-based study. J Am Coll Cardiol 2002; 39: 1670–9. Schneider C, McCrindle BW, Carvalho JS et al. Development of Z-scores for fetal cardiac dimensions from echocardiography. Ultrasound Obstet Gynecol 2005; 26: 599–605. DeVore GR. The use of Z-scores in the analysis of fetal cardiac dimensions. Ultrasound Obstet Gynecol 2005; 26: 596–8. Stellin G, Santini F, Thiene G et al. Pulmonary atresia, intact ventricular septum, and Ebstein anomaly of the tricuspid valve. Anatomic and surgical considerations. J Thorac Cardiovasc Surg 1993; 106: 255–61. Grant RT. An unusual anomaly of the coronary vessels in the malformed heart of a child. Heart 1926; 13: 273–83. Lauer RM, Fink HP, Pettry EL et al. Angiographic demonstration of intramyocardial sinusoids in pulmonary valve atresia with intact ventricular septum and hypoplastic right ventricle. N Engl J Med 1964; 271: 68–72. Freedom RM, Harrington DP. Contributions of intramyocardial sinusoids in pulmonary atresia and intact ventricular septum to a right-sided circular shunt. Br Heart J 1974; 36: 1061–5. Freedom RM, Anderson RH, Perrin D. The significance of ventriculo-coronary arterial connections in the setting of pulmonary atresia with an intact ventricular septum. Cardiol Young 2005; 15: 447–68. Gittenberger-De Groot AC, Tennstedt C, Chaoui R et al. Ventriculo coronary arterial communications (VCAC) and myocardial sinusoids in hearts with pulmonary atresia with intact ventricular septum: two different diseases. Prog Pediatr Cardiol 1991; 13: 157–64. Sandor GG, Cook AC, Sharland GK et al. Coronary arterial abnormalities in pulmonary atresia with intact ventricular septum diagnosed during fetal life. Cardiol Young 2002; 12: 436–44.

Pulmonary atresia with intact ventricular septum

25. Maeno YV, Boutin C, Hornberger LK et al. Prenatal diagnosis of right ventricular outflow tract obstruction with intact ventricular septum, and detection of ventriculocoronary connections. Heart 1999; 81: 661–8. 26. Chaoui R, Machlitt A, Tennstedt C. Prenatal diagnosis of ventriculo-coronary fistula in a late first-trimester fetus presenting with increased nuchal translucency. Ultrasound Obstet Gynecol 2000; 15: 160–2. 27. Santos MA, Moll JN, Drumond C et al. Development of the ductus arteriosus in right ventricular outflow tract obstruction. Circulation 1980; 62: 818–22. 28. Berg C, Kremer C, Geipel A et al. Ductus venosus blood flow alterations in fetuses with obstructive lesions of the right heart. Ultrasound Obstet Gynecol 2006; 28: 137–42. 29. Todros T, Paladini D, Chiappa E et al. Pulmonary stenosis and atresia with intact ventricular septum during prenatal life. Ultrasound Obstet Gynecol 2003; 21: 228–33. 30. Zosmer N, Bajoria R, Weiner E et al. Clinical and echographic features of in utero cardiac dysfunction in the recipient twin in twin-twin transfusion syndrome. Br Heart J 1994; 72: 74–9. 31. Lougheed J, Sinclair BG, Fung Kee FK et al. Acquired right ventricular outflow tract obstruction in the recipient twin in twin-twin transfusion syndrome. J Am Coll Cardiol 2001; 38: 1533–8. 32. Herberg U, Gross W, Bartmann P et al. Long term cardiac follow up of severe twin to twin transfusion syndrome after intrauterine laser coagulation. Heart 2006; 92: 95–100. 33. Allan LD, Crawford DC, Tynan MJ. Pulmonary atresia in prenatal life. J Am Coll Cardiol 1986; 8: 1131–6. 34. Sharland G. Pulmonary valve abnormalities. In: Allan L, Hornberger L, Sharland G, eds. Textbook of Fetal Cardiology. London: Greenwich Medical Media, 2000: 233–47. 35. Arzt W, Tulzer G, Aigner M et al. Invasive intrauterine treatment of pulmonary atresia/intact ventricular septum with heart failure. Ultrasound Obstet Gynecol 2003; 21: 186–8. 36. Tulzer G, Arzt W, Franklin RC et al. Fetal pulmonary valvuloplasty for critical pulmonary stenosis or atresia with intact septum. Lancet 2002; 360: 1567–8. 37. Galindo A, Gutierrez-Larraya F, Velasco JM et al. Pulmonary balloon valvuloplasty in a fetus with critical pulmonary stenosis/atresia with intact ventricular septum and heart failure. Fetal Diagn Ther 2006; 21: 100–4. 38. Roman KS, Fouron JC, Nii M et al. Determinants of outcome in fetal pulmonary valve stenosis or atresia with intact ventricular septum. Am J Cardiol 2007; 99: 699–703. 39. Peterson RE, Levi DS, Williams RJ et al. Echocardiographic predictors of outcome in fetuses with pulmonary atresia with intact ventricular septum. J Am Soc Echocardiogr 2006; 19: 1393–400. 40. Salvin JW, McElhinney DB, Colan SD et al. Fetal tricuspid valve size and growth as predictors of outcome in pulmonary atresia with intact ventricular septum. Pediatrics 2006; 118: e415–20. 41. Gardiner HM, Belmar C, Tulzer G et al. Morphological and functional predictors of eventual circulation in the fetus

42.

43.

44.

45.

46.

47.

48.

49.

50.

51.

52.

53.

54.

55.

279

with pulmonary atresia or critical pulmonary stenosis with intact septum. J Am Coll Cardiol 2008; 51: 1299–308. Bull C, Kostelka M, Sorensen K et al. Outcome measures for the neonatal management of pulmonary atresia with intact ventricular septum. J Thorac Cardiovasc Surg 1994; 107: 359–66. Rosenthal E, Qureshi SA, Chan KC et al. Radiofrequencyassisted balloon dilatation in patients with pulmonary valve atresia and an intact ventricular septum. Br Heart J 1993; 69: 347–51. Parsons JM, Rees MR, Gibbs JL. Percutaneous laser valvotomy with balloon dilatation of the pulmonary valve as primary treatment for pulmonary atresia. Br Heart J 1991; 66: 36–8. Pawade A, Capuani A, Penny DJ et al. Pulmonary atresia with intact ventricular septum: surgical management based on right ventricular infundibulum. J Card Surg 1993; 8: 371–83. de Leval M, Bull C, Stark J et al. Pulmonary atresia and intact ventricular septum: surgical management based on a revised classification. Circulation 1982; 66: 272–80. Yoshimura N, Yamaguchi M, Ohashi H et al. Pulmonary atresia with intact ventricular septum: strategy based on right ventricular morphology. J Thorac Cardiovasc Surg 2003; 126: 1417–26. Numata S, Uemura H, Yagihara T et al. Long-term functional results of the one and one half ventricular repair for the spectrum of patients with pulmonary atresia/stenosis with intact ventricular septum. Eur J Cardiothorac Surg 2003; 24: 516–20. Starnes VA, Pitlick PT, Bernstein D et al. Ebstein’s anomaly appearing in the neonate. A new surgical approach. J Thorac Cardiovasc Surg 1991; 101: 1082–7. Daubeney PE, Wang D, Delany DJ et al. Pulmonary atresia with intact ventricular septum: predictors of early and medium-term outcome in a population-based study. J Thorac Cardiovasc Surg 2005; 130: 1071. Odim J, Laks H, Plunkett MD et al. Successful management of patients with pulmonary atresia with intact ventricular septum using a three tier grading system for right ventricular hypoplasia. Ann Thorac Surg 2006; 81: 678–84. Agnoletti G, Piechaud JF, Bonhoeffer P et al. Perforation of the atretic pulmonary valve. Long-term follow-up. J Am Coll Cardiol 2003; 41: 1399–403. Guleserian KJ, Armsby LB, Thiagarajan RR et al. Natural history of pulmonary atresia with intact ventricular septum and right-ventricle-dependent coronary circulation managed by the single-ventricle approach. Ann Thorac Surg 2006; 81: 2250–7. Redington AN, Penny D, Rigby ML et al. Antegrade diastolic pulmonary arterial flow as a marker of right ventricular restriction after complete repair of pulmonary atresia with intact septum and critical pulmonary valvar stenosis. Cardiol Young 1992; 2: 382–6. Sanghavi DM, Flanagan M, Powell AJ et al. Determinants of exercise function following univentricular versus biventricular repair for pulmonary atresia/intact ventricular septum. Am J Cardiol 2006; 97: 1638–43.

19 Intracardiac shunt malformations Einat Birk and Norman H Silverman Intracardiac malformations leading to a cardiac left-to-right shunt postnatally include atrial septal defects (ASDs), atrioventricular septal defects (AVSDs), and ventricular septal defects (VSDs). These lesions constitute the largest group of cardiac defects detected during fetal life, the most common being VSD and AVSD.1,2

Defects of the atrial septum Anatomy

developmental malformation in the sinus venosus or from a primary failure in the partitioning of the true embryonic septum secundum.4 A coronary sinus ASD is a rare defect believed to occur when the atriosinus venosus fold fails to form. Therefore, instead of the normal draining of the coronary sinus into the right atrium via its usual orifice, there is a persistence of the wide communication between the sinus venosus and both atria.4 Persistent left superior vena cava terminating in the left atrium is almost always present. Unlike a large coronary sinus receiving a persistent left superior vena cava that eventually drains into the right atrium, this

An ASD is a common congenital defect seen in children, occurring in 1 in 1500 live births.3 It can present as an isolated defect or in association with complex congenital heart defects. Several mechanisms cause the formation of an atrial communication, leading to several defect types (Figure 19.1): • secundum ASD • primum ASD (also named AVSD – partial or transitional type) • sinus venosus ASD (superior and inferior types) • coronary sinus ASD. Secundum ASD is the most common atrial communication in children. It occurs when the septum primum fails to cover the oval fossa, which is patent and allows rightto-left flow during fetal life. This failure of the septum primum can result in a single defect or fenestrated defect, as well as a wide range of defect sizes. A primum ASD involves the lower part of the atrial septum and is part of the AVSD spectrum, which will be discussed separately. A sinus venosus ASD is located in the posterosuperior or posteroinferior portion of the atrial septum. The superior defect is the more common, lying at the junction of the superior vena cava, right upper pulmonary vein, and atrial septum. The superior vena cava appears to override the defect, which tends to be large. The inferior defect occurs at the junction of the inferior vena cava and the atrial septum. It is a less common defect and also tends to be of significant size. These defects result from a

SVC RAA

SSV

OS

OP CS

ISV TV

IVC

Figure 19.1 This drawing shows the different types of interatrial septal communications as seen from the right aspect of the atrial septum. CS, coronary sinus; ISV, inferior sinus venosus; IVC, inferior vena cava; OP, ostium primum; OS, ostium secundum; RAA, right atrial appendage; SSV, superior sinus venosus; SVC, superior vena cava; TV, tricuspid valve.

282

Fetal Cardiology

defect allows a communication between the two atria, and the left superior vena cava drains directly into the left atrium. During fetal life, the normal atrial communication at the oval fossa allows right-to-left atrial flow, allowing oxygen-rich blood to flow to the left heart and, consequently, to the brain and the heart. After birth, following the normal decrease in pulmonary pressure and resistance, and immediate increase of systemic vascular resistance caused by the loss of the placenta, the foramen should close and prevent intra-atrial shunting. The foraminal mechanism has been found to be substantially less efficient than was previously believed, since the advent of modern postnatal cardiac ultrasound has shown that a substantial number of infants under 6 months of age have intra-atrial left-to-right shunts. This is due to incomplete closure of the fossa ovalis by the septum primum. When the septum primum is deficient, the right atrial and ventricular pressure decreases gradually as the compliance increases, leading to a predominant left-to-right shunting across the atrial communication. The amount of left-to-right shunting is determined by the size of the atrial defect and by the relative atrial and ventricular diastolic pressure differences.

Fetal diagnosis Ostium secundum atrial septal defect During fetal life, there is normally a communication between the right and left atria, which is located at the secundum septum, namely the foramen ovale. Relatively oxygenated blood streams from the ductus venosus via the inferior vena cava and into the left atrium by diverting the septum primum flap, which lies on the left side of the septum. This flap of atrial tissue is pushed open open during most of the cardiac cycle. The foramen ovale lies in the middle third of the atrial septum and grows with gestation. The normal foramen size is similar to the aortic diameter5 and grows from approximately 3 mm at 20 weeks of gestation to 8 mm at term.6 After birth, placental flow is eliminated and two processes occur: pulmonary venous flow increases and left atrial pressure rises above right atrial pressure. The flap of the septum primum is then pushed against the foramen ovale, leading to a functional closure of the atrial communication. Only when this flap valve mechanism fails to close the foramen ovale is the communication termed a secundum defect. Since the size of the fetal foramen ovale and the primum flap varies widely in the normal fetus, it is impossible to predict the failure of the closure process during fetal life. Therefore, in contrast with postnatal life, when a secundum ASD is the commonest atrial defect, the fetal diagnosis of such an ASD possesses inherent difficulty and is rarely possible.1

Ostium secundum ASD can be a part of many forms of complex congenital heart defect, from anomalies of pulmonary venous return through to coarctation of the aorta. It is an essential part of defects such as tricuspid atresia. As in an isolated defect, it is rarely associated with extracardiac anomalies or genetic disorders.

Ostium primum atrial septal defect (partial atrioventricular septal defect) One of the most commonly diagnosed atrial communications in utero is the primum ASD.1 This lesion is a form of common atrioventricular canal defect without a ventricular component. In this defect, the lower portion of the atrial septal fusion to the underlying atrioventricular valve junction is absent; both atrioventricular valves are attached to the crest of the ventricular septum and lose their normal differential appearance. The left atrioventricular valve is referred to as being ‘cleft’, showing a commissure between the primitive anterosuperior and posteroinferior bridging leaflets. Mild degrees of valvar insufficiency from this site can be detected in some fetuses. The four-chamber view is a useful plane in detecting this lesion. With this plane, a series of scans should be looked at in a sequential fashion in order to define appropriate morphological information (Figure 19.2). The posterior cross-sectional cut reveals the coronary sinus emptying into the right atrium just above the tricuspid valve. This view may lead to the impression that both atrioventricular valves insert into the ventricular septum at the same level. However, a more anterior coronal cut will show that both atrioventricular valves open and close with their normal differential insertion. This cut is also ideal for color Doppler interrogation of both valves. A more anterior angulation will reveal the left ventricular outflow tract including the aortic valve and ascending aorta. The coronal cut where both atrioventricular valves open is probably the most useful in the diagnosis of a primum ASD (Figure 19.3a). In this view the lower atrial septum is missing and both atrioventricular valves insert into the ventricular septum at the same level. A mild degree of left atrioventricular valve insufficiency may be detected as well as occasional right atrioventricular valve insufficiency. The more posterior cut, where the coronary sinus is displayed together with the tricuspid valve, can lead to the false impression that both atrioventricular valves are at the same level, leading to the erroneous diagnosis of a primum ASD. This error is even more likely to occur when the coronary sinus is enlarged (Figure 19.3b). This is the case when a persistent left superior vena cava drains into the coronary sinus; however, this vessel can usually be detected from other views. The cleft in the left atrioventricular valve is best seen in a cross-section from the parasternal or subcostal short-axis views. When a primum ASD is detected, a complete sequential analysis of the heart is mandatory. Primum ASD is

Intracardiac shunt malformations

283

LA

RA

CS

LV RV

LA RA

LV

RV

LA RA

AO

LV RV

(a)

(b)

Figure 19.2 (a) These drawings show a series of four-chamber cuts as scanned from the back to the front of the heart. (Top) Posterior crosssectional cut reveals the coronary sinus (CS) emptying into the right atrium (RA) just above the tricuspid valve. This view may lead to a false impression that both atrioventricular valves insert into the ventricular septum at the same level (arrows). (Middle) A more anterior conventional four-chamber cut will show that both atrioventricular valves open and close with their normal differential insertion (arrows). This cut is also ideal for color Doppler interrogation of both valves. (Bottom) A more anterior cut will reveal the left ventricular (LV) outflow tract including the aortic valve (AO) and ascending aorta. The coronal cut where both atrioventricular valves open is probably the most useful in the diagnosis of a primum atrial septal defect. (b) (Top–bottom) This series of ultrasonic cuts in the four-chamber plane corresponds to the series shown in Figure 19.2a. LA, left atrium; RV, right ventricle; D Ao, descending aorta.

284

Fetal Cardiology

RA RV LV

LA

(a)

(b)

Figure 19.3 (a) The heart is imaged in the four-chamber view showing both atrioventricular valves inserting into the ventricular septum at the same level. The single arrow points to the missing lower atrial septum, indicating the presence of a primum atrial septal defect. The double arrow points to the foramen ovale, the site of a normal fetal atrial communication. (b) This frame is a four-chamber cut showing the enlarged coronary sinus (CS, arrow) entering the right atrium (RA), which reflects the increased flow from a persistent left superior vena cava entering the upper portion of the coronary sinus. The coronary sinus enlargement may give the false impression of an ostium primum type of interatrial communication.

associated with situs anomalies such as right or left atrial isomerism, when the atrial septum tends to be small, leading to the appearance of a common atrium. Left heart hypoplasia, subaortic narrowing, and coarctation of the aorta are known associated anomalies. A secundum ASD is a common associated finding. Primum ASD is associated with extracardiac anomalies, the most common being trisomy 21. It has also been rarely associated with DiGeorge syndrome and Ellis–van Creveld syndrome. Karyotyping should be performed when the diagnosis of primum ASD is made.

Sinus venosus and coronary sinus atrial septal defect To the best of our knowledge, both sinus venosus and coronary sinus ASDs have not yet been reported in the fetus. These lesions are rarely associated with cardiac, extracardiac, or chromosomal anomalies.

Natural history and outcome Small secundum ASDs usually close spontaneously during the first 2 years of life. Defects persisting beyond 2 years of age tend to stay open and lead to a left-to-right atrial shunting of variable amount.

Primum ASDs (or partial AVSDs) do not close spontaneously and usually lead to a significant left-to-right shunting as well as the risk of pulmonary hypertension and pulmonary vascular disease. Some will develop substantial insufficiency of the left atrioventricular valve. Sinus venosus ASDs never close spontaneously and almost always have a large concomitant left-to-right atrial shunt. Even patients with a large left-to-right atrial shunt may be asymptomatic for many years. Some will develop right ventricular dysfunction and atrial arrhythmias during late adult life. A serious but rare complication is the development of secondary pulmonary hypertension and pulmonary artery thrombosis. All complications can be prevented by closing the defect. For most secundum defects, transcatheter closure has become available in many centers. Larger defects as well as all primum and sinus venosus atrial septal defects have to be closed surgically. The closure of a primum ASD involves repair of the left atrioventricular valve. Some of these patients will eventually need additional surgery for left atrioventricular valve repair or replacement. A smaller group will require further surgery to alleviate progressive obstruction of the left ventricular outflow tract. While life quality and expectancy after secundum ASD repair during childhood are similar to those of the general population,7 patients following primum ASD repair have somewhat poorer results. Approximately 10% will need repeated surgery, and life

Intracardiac shunt malformations

expectancy is shorter than that of the normal population.8 When a primum defect is associated with left heart anomalies such as hypoplasia of the left atrioventricular valve, hypoplasia of the left ventricle, subaortic obstruction, or coarctation of the aorta, the overall prognosis is guarded. In rare cases the left heart hypoplasia will not allow a biventricular repair, leading to palliative solutions such as the Fontan-type repair.

Restrictive foramen ovale As discussed previously, the normal fetal foramen ovale has a wide range of what is considered to be normal size. The normal flow across it is of low velocity, ranging between 20 and 40 cm/s on pulsed Doppler.6 Restrictive flow across it corresponds to an increase in flow velocity, usually above 100 cm/s. Most case reports of restrictive flow across the foramen ovale are associated with various forms of hypoplasia of the left heart.9 In these cases the expected increase in left atrial pressure results in the opposition of the primum septum to the atrial septum and therefore a decrease in the flap valve diameter. Some authors believe that a restrictive foramen may lead to the development of hypoplastic left heart, since foraminal flow provides most flow into the left ventricle.

Complete atrioventricular septal defect Anatomy A complete AVSD is one of the more common cardiac defects detected prenatally.1,10,11 This lesion is also known by the terms endocardial cushion defect or atrioventricular canal defect. In this lesion the atrial and ventricular septation is not complete and the separation between mitral and tricuspid orifices does not occur. Instead, there is a common atrioventricular junction. This lesion can be found as a spectrum of anomalies, ranging from a complete form (when both atrial and ventricular septation is incomplete, leading to a communication at both atrial and ventricular level) to a partial, or incomplete, form (when only the atrial or ventricular communication persists). All forms involve an intrinsic abnormality of the atrioventricular valves. In most cases of AVSD, the atrioventricular junction is connected to the right and left ventricles so that the blood flows relatively evenly into each ventricle. This relationship is also described as a balanced AVSD. When the atrioventricular junction is predominantly connected to one of the ventricles, there is usually hypoplasia of the ventricle receiving the smaller portion of the atrioventricular

285

orifice. This relationship is also described as an unbalanced AVSD, and right or left dominance can be identified12 (Video clip 19.1).

Fetal diagnosis The goals of the fetal cardiac ultrasound examination are: 1. Identify the presence and extent of the AVSD; 2. Assess the relationship of the atrioventricular junction to the underlying ventricles; 3. Assess the size of both ventricles; 4. Assess the degree of atrioventricular valve regurgitation; 5. Identify associated anomalies. The apical four-chamber view is the most commonly used cut to identify an AVSD (Figure 19.4). In the normal heart, the tricuspid septal leaflet is attached to the ventricular septum, while the mitral valve has no septal attachment and inserts into the crux of the heart at a slightly more cranial position. In an AVSD the left atrioventricular valve is attached to the ventricular septum at the same level as is the right atrioventricular valve. Therefore, both atrioventricular valves are at the same level, losing the normal differential insertion. The primum septum that can be easily identified in the normal four-chamber view is absent. A ventricular communication can usually be identified in this view; in most cases this defect is large, although smaller defects can exist and are usually more difficult to identify. The apical four-chamber view is ideal

RA

RV

LA

LV

Figure 19.4 The heart is imaged in the apical equivalent four-chamber view, showing a typical picture of an atrioventricular septal defect. Both atrioventricular valves insert into the ventricular septum at the same level. Atrial and ventricular septal defects are seen above and below this insertion.

286

Fetal Cardiology

for assessing the relationship of the atrioventricular junction to the underlying ventricles as well as the size of both ventricles. In this plane the ventricles should be of similar size. The atrial septum can be malaligned with the ventricular septum: when the atrial septum is deviated to the left, the right atrium drains to both ventricles (also described as double-outlet right atrium). When the atrial septum is shifted to the right, the left atrium drains to both ventricles. Such an anomaly can be corrected by surgery. A less favorable variation is when one ventricle is significantly smaller than the other. This is also known as an unbalanced AVSD, and in extreme situations will lead to a single ventricle solution. The apical four-chamber view is also ideal for flow interrogation of the atrioventricular valves. Insufficiency of both atrioventricular valves, which is common in the newborn with an AVSD, is less commonly seen during fetal life (Figure 19.5). The shortaxis views obtained from the parasternal or subcostal equivalent angles provide a detailed picture of the atrioventricular valve anatomy. In rare cases the insertion of the right or left atrioventricular valve is on the other side of the ventricular septum. These are more difficult to repair and may lead to a worse prognosis. A more common variation is when the anterior bridging leaflet has no attachment to the interventricular septum. This anatomical form of AVSD is often associated with tetralogy of Fallot and can be diagnosed in the fetus (Video clip 19.2). To the best of our knowledge, this has not yet been reported in the fetus. As in the case of primum ASD, the presence of a large coronary sinus may be misdiagnosed as an AVSD.13

RA LA 1.1 RV LV

The coronary sinus lies behind the left atrium and is usually enlarged by additional flow from a persistent left superior vena cava draining into the coronary sinus. A posterior coronal four-chamber view can create the illusion of both tricuspid and mitral valves inserting into the ventricular septum at the same level. However, a more anterior cut (Figure 19.2) will reveal the real relationship between the valve insertions where they can be demonstrated in both open and closed position. The normal offset can then be demonstrated, avoiding the false diagnosis of an AVSD. One should keep in mind that persistent left superior vena cava and dilated coronary sinus can coexist with an AVSD.

Associated lesions When an AVSD is detected, a complete sequential analysis of the heart is mandatory. An AVSD is associated with situs anomalies such as right and left isomerism. Tetralogy of Fallot and double-outlet right ventricle are well-known associated lesions, more common in fetuses with trisomy 21. Left heart hypoplasia, subaortic narrowing, and coarctation of the aorta are all known associated cardiac anomalies, usually in fetuses with normal chromosomes. Another common association is a secundum ASD. The most common extracardiac anomaly associated with an AVSD is trisomy 21. An AVSD is also associated with other chromosomal anomalies such as trisomies 18 and 13. The fetal karyotype, therefore, should be examined whenever this diagnosis is made. It can also be a part of other syndromes such as Ellis–van Creveld, VACTERL (vertebral anomalies, anal atresia, congenital heart disease, tracheoesophageal fistula, renal dysplasia, and limb abnormalities), CHARGE (coloboma of the eye, heart defects, atresia of the choanae, retardation of growth and/or development, genital and/or urinary abnormalities, and ear abnormalities or deafness), Cornelia de Lange, and Goldenhar syndromes.14 In a recent study,15 out of 301 fetuses with an AVSD, only 51% had isolated AVSD. Right isomerism occurred in 12% and left isomerism in 20%. Extracardiac abnormalities and non-karyotypic syndromes were evident in 13%; 39% had trisomy 21 and 10% had other chromosomal abnormalities. Similar findings were found by other groups.16,17

Natural history and outcome 1.1

Figure 19.5 Color Doppler interrogation of the atrioventricular valves demonstrates a systolic jet of valve insufficiency, as shown in this four-chamber cut.

During fetal life AVSD is usually well tolerated, and most fetuses reach term and are delivered according to routine obstetric practice. A few will develop congestive heart failure and non-immune hydrops because of severe insufficiency of the atrioventricular valve or heart block, especially in left atrial isomerism.15,18,19 In these cases the odds for fetal or neonatal demise are high. According to a

Intracardiac shunt malformations

recent study, 15% of fetuses diagnosed with this lesion whose parents opted to continue with the pregnancy died in utero.15 The infant with an isolated AVSD will remain asymptomatic for a few weeks. Most will develop signs of congestive heart failure during the first 4–8 weeks of life as pulmonary vascular resistance falls. All will require surgical repair, which is usually carried out during the first 6 months of life. Survival after AVSD repair is high, exceeding 90%, although some patients (especially the group with normal chromosomes) will require additional surgery because of the development of left atrioventricular valve insufficiency or narrowing of the left ventricular outflow tract. The odds for a successful repair decline when additional cardiac anomalies are present. Again, the chance of survival to 3 years is significantly lower in the group followed since intrauterine life when compared to the surgical literature, and is quoted as low as 38%.15

Ventricular septal defect Anatomy VSD is the most common congenital heart defect diagnosed during the first year of life.20,21 VSD is also a common cardiac defect detected prenatally, at a rate lower than that of prenatal detection of an AVSD.1,2 The ventricular septum is arbitrarily divided into four sections: the inlet, membranous, trabecular, and outlet components, which have different embryological origins. When viewed from the right ventricle, the inlet septum has a lightly trabecular surface and is bounded by the tricuspid annulus and the attachments of the papillary muscles to the ventricular septum. The membranous septum is a relatively small segment lying beneath the septal leaflet of the tricuspid valve and adjacent to the aortic and mitral valves. Viewed from the left ventricle, the membranous septum lies adjacent to the right fibrous trigone just beneath the aortic valve. The membranous septum is a thin translucent structure and therefore cannot be well imaged in all planes, and is often difficult to image even after birth. This may result in the false impression of a VSD in some views, especially in the fetus. The trabecular septum, which derives its name from its heavily trabeculated appearance, extends from the inlet septum to the region of the outlet septum just proximal to the pulmonary valve, and does not lie in a single plane. It contains the moderator band (or septomarginal trabeculation, which is also called the septal band of the crista supraventricularis), which extends in a Y-shaped fashion below the pulmonary valve. Its anterior portion abuts the outlet septum, while its posterior portion is the papillary muscle of the conus (muscle of Lancisi). Inferiorly, the

287

septomarginal trabeculation extends as a broad muscle bundle. This portion of the muscle, the septomarginal trabeculation, is also referred to as the septal limb of the crista supraventricularis. The ventricular infundibular fold, which is the right ventricular muscle lying between the tricuspid and the pulmonary valves, is also called the parietal band of the crista supraventricularis. The outlet septum (also named conus septum) is a small segment extending from the septomarginal trabecula to the pulmonary valve. VSDs can occur in any of the septal locations, but also occur at the sites of fusion between them; for example, defects found around the membranous septum are termed perimembranous. They can also be named for their area of extension, such as perimembranous-inlet, perimembranous-trabecular, and perimembranous-outlet defects. Perimembranous defects comprise about 75% of all VSDs.22 When the defects are surrounded entirely by muscle, they are termed muscular-inlet, muscular-trabecular, and muscular-outlet defects. Muscular defects comprise 10–15% of all ventricular septal defects. Defects of the outlet septum most commonly are adjacent to the pulmonary and aortic valves and are termed subarterial doubly committed or supracristal defects. Such defects constitute about 5% of the VSDs, but are more common in Asian populations.22 When the different portions of the septum adjacent to the VSD are malaligned, they are termed a malalignment VSD. The VSD can be malaligned between the outlet and trabecular septum or with respect to the atrioventricular valves, associated with straddling of an atrioventricular valve. The size of ventricular septal defects varies from very small to large, involving a third or more of the ventricular septum. Since the septum does not lie in a single plane, it can be difficult to assess the VSD size. Defects may be isolated or multiple, and are commonly a part of, or associated with, other cardiac lesions.

Fetal diagnosis The goals of the ultrasound examination are: 1. Identify the presence of a VSD; 2. Define which segment of the septum is involved; 3. Identify additional anomalies. The fetal ventricular septum is easily seen in the fourchamber view, which can be visualized from an apical equivalent or from a lateral orientation. In the apical equivalent four-chamber view the ultrasound beam is parallel to the ventricular septum using lateral resolution (Figure 19.6). Since the membranous part of the septum is thin, the ventricular septum ‘disappears’ toward the internal crux of the heart. This may result in the appearance of a dropout in the septum, leading to a false diagnosis of a

288

Fetal Cardiology

RA

LA

LA RA

RV

LV

LV RV .96

.96

Figure 19.6 The ventricular septum of a 20-week fetus is visualized in a four-chamber view. A muscular septal defect is identified at the lower portion of the septum (arrow). Note the ‘T’ artifact at the defect edges.

VSD.23 In order to achieve a different angle, the transducer can be moved to a different location on the maternal abdomen so that the ultrasound beam will be perpendicular to the ventricular septum, using axial resolution. The parasternal short-axis equivalent cut is also useful, when the ultrasound beam is perpendicular to the ventricular septum; the different parts of the septum can be visualized in great detail. This view is especially useful in identifying a subarterial doubly committed VSD, also known as a supracristal VSD (Video clip 19.3). A useful physical sign of a septal defect is the so-called ‘T’ artifact: high impedance exists at the blood–tissue interface producing ballooning of echoes at the rim of a defect, creating bright spots at the defect edges.23 The ‘T’ artifact is not created when the septum is thin, leading to a simple dropout. The use of color flow Doppler imaging may augment the ability to identify a VSD in the fetus (Figures 19.7 and 19.8). Once again, when this is performed from the apical equivalent four-chamber view, the ultrasound beam is parallel to the ventricular septum so that the color tends to ‘cover’ the thin part of the septum, leading to the impression of flow across a VSD. However, when the ultrasound beam is perpendicular to the ventricular septum, color Doppler flow can be detected more accurately when flow is crossing the septum, usually in a bidirectional fashion.24,25 Since the pressure is similar in the fetal right and left ventricles, the potential pressure gradient across a VSD is small. Color flow velocity should therefore be reduced to a low Nyquist limit in order to detect low-velocity jets. Using new echocardiographic technology allowing three-dimensional reconstruction adds to the anatomical

Figure 19.7 Flow across this muscular ventricular septal defect can be detected using color flow mapping.

RA RV LA .39

LV

.39

Figure 19.8 The same ventricular septal defect as in Figure 19.7 can be shown by color flow Doppler when the ventricular septum is imaged with the ultrasound beam perpendicular to the defect.

understanding of ventricular septal defects in the fetal heart26 (Figure 19.9 and Video clip 19.4)

Associated lesions When a VSD is detected, a complete sequential analysis of the heart is mandatory. An isolated VSD is rarely associated with situs anomalies. However, it is commonly found as part of complex cardiac lesions, some of which are not obvious when the study is performed during early pregnancy. When a VSD is identified, both right and left

Intracardiac shunt malformations

a deletion in 22q11 and non-chromosomal multiply malformed fetuses.2

Chest

5 RA RV

LV

RV

289

10

Figure 19.9 This three dimensional image is taken from a fetus of 26 weeks’ gestation, with a perimembranous ventricular septal defect, shown by an unlabeled short arrow at the top of the septum. The chest, the right atrium RA, right ventricle RV, left atrium LA and left ventricle LV are labeled. Additionally, the atrial and ventricular septum can be clearly identified between their respective chambers. See also corresponding Video clip 19.4.

ventricular outflow tracts should be examined in detail. Since a VSD can be a part of tetralogy of Fallot, the size of the right ventricular outflow tract and main and branch pulmonary arteries, as well as the flow across these structures, should be examined. The usual malalignment of the outflow septum in tetralogy of Fallot is not as easy to identify during fetal life as it is in postnatal life. This diagnosis should be suspected whenever the pulmonary artery is smaller than expected or a pressure gradient is detected across the right ventricular outflow tract. The associations of other lesions with ventricular septal defects are numerous, and include left heart obstructions such as subaortic narrowing, aortic valve stenosis, coarctation of the aorta, and interrupted aortic arch. Since both right and left outflow obstructions can evolve during pregnancy and postnatal life, reassessment during late gestation is advised. A VSD can also be a part of complex lesions such as transposition of the great arteries and double-outlet right ventricle. Extracardiac anomalies associated with a VSD include chromosomal anomaly in over 40% according to some series.1,2 Such anomalies included trisomies 21, 13, and 18. This rate is significantly higher than expected from postnatal series, and may relate to the selection of patients referred for fetal echocardiography as well as to spontaneous fetal loss in chromosomally abnormal fetuses that would not be included in postnatal series. Other extracardiac anomalies associated with a VSD include

Natural history and outcome During fetal life, a VSD is well tolerated, and most fetuses reach term and are delivered according to routine obstetric practice. The odds for fetal demise are higher when extracardiac anomalies are present. Isolated perimembranous and muscular ventricular septal defects detected and followed through pregnancy show a high rate of spontaneous closure.27,28 No correlation was found between the size of the defect prenatally and its chance for spontaneous closure.27 We have data to show that smaller VSDs have a greater chance of closure than larger defects (postnatally). Infants with an isolated VSD may remain asymptomatic for a few weeks; those with relatively large defects will develop signs of congestive heart failure after a few weeks of life as pulmonary vascular resistance falls. Over 50% of VSDs located in the perimembranous or muscular septum will close spontaneously, usually during the first year of life; only a minority of such defects will require surgical repair. Defects in the inlet or outlet septum do not close spontaneously and require surgical repair. Surgery is usually carried out during the first year of life and, when isolated, carries a low mortality and complication rate.29 Multiple defects or a single large apical muscular defect carries a higher risk of surgical repair due to limited access. Such infants may need the placement of pulmonary arterial banding to decrease pulmonary flow and pressure during the first months of life, followed by surgical or transcatheter closure of the defects coupled with pulmonary debanding later in life. When associated cardiac anomalies are present, the odds for a successful repair depend on the severity and nature of the additional cardiac defects.

Legends for the DVD Video clip 19.1 Four-chamber view of an atrioventricular septal defect (AVSD) with a smaller left atrioventricular connection when compared to the right, seen both on two-dimensional image (a) and with color flow across the inlets (b).

Video clip 19.2 The superior (sup) leaflet of the AVSD does not have any attachment to the interventricular septum (type C according to the Rastelli classification). In many cases this type of AVSD is associated with tetralogy of Fallot.

Video clip 19.3 (a) In this fetus with transposition of the great arteries, a septal defect is clearly seen in the outlet portion of the interventricular

290

Fetal Cardiology

septum. Such a VSD is termed subarterial doubly committed or supracristal defect. (b) Same clip using color Doppler to demonstrate flow across the defect (in this case the flow is directed from the left ventricle via the VSD into the anterior aorta). (c) A sweep across the outlets of the right and left ventricles shows the VSD adjacent to both semilunar valves. In this fetus the aorta is the anterior vessel and the pulmonary artery is the posterior vessel.

Video clip 19.4 Three-dimensional reconstruction of a perimembranous VSD in a fetus of 26 weeks’ gestation, with a perimembranous ventricular septal defect. The chest, the right atrium RA, right ventricle RV, left atrium LA and left ventricle LV are labeled. Additionally, the atrial and ventricular septum can be clearly identified between their respective chambers.

References 1. Allan LD, Sharland GK, Milburn A et al. Prospective diagnosis of 1,006 consecutive cases of congenital heart disease in the fetus. J Am Coll Cardiol 1994; 23: 1452–8. 2. Stoll C, Garne E, Clementi M; Euroscan study group. Evaluation of prenatal diagnosis of associated congenital heart disease by fetal ultrasonographic examination in Europe. Prenat Diagn 2001; 21: 243–52. 3. Fyler DC. Report of the New England Regional Infant Cardiac Program. Pediatrics 1980; 65: 375–461. 4. Goor DA, Lillehei CW. Congenital Malformations of the Heart. New York: Grune & Stratton, 1975. 5. Wilson AD, Rao PS, Aeschlimann S. Normal fetal foramen flap and transatrial Doppler velocity pattern. J Am Soc Echocardiogr 1990; 3: 491–4. 6. Phillipos EZ, Robertson MA, Still KD. The echocardiographic assessment of the human fetal foramen ovale. J Am Soc Echocardiogr 1994; 7: 257–63. 7. Nieminen HP, Jokinen EV, Sairanen HI. Late results of pediatric cardiac surgery in Finland: a population-based study with 96% follow-up. Circulation 2001; 104: 570–5. 8. El-Najdawi EK, Driscoll DJ, Puga FJ et al. Operation for partial atrioventricular septal defect: a forty year review. J Thorac Cardiovasc Surg 2000; 119: 880–90. 9. Cohbot V, Hornberger LK, Hagen-Ansert S, Sahn DJ. Prenatal detection of restrictive foramen ovale. J Am Soc Echocardiogr 1990; 3: 15–19. 10. Allan LD. Atrioventricular septal defect in the fetus. Am J Obstet Gynecol 1999; 181: 1250–3. 11. Silverman NH. Pediatric Echocardiography. Baltimore: Williams & Wilkins, 1993. 12. Pitkanen OM, Homberger LK, Miner SE et al. Borderline left ventricle in prenatally diagnosed atrioventricular septal defect or double outlet right ventricle: echocardiographic predictors of biventricular repair. Am Heart J 2006; 152: 163.e1–7. 13. Park JR, Taylor DK, Skeels M, Towner DR. Dilated coronary sinus in the fetus misinterpretation as an atriventricular canal defect. Ultrasound Obstet Gynecol 1997; 10: 126–9.

14. Feldt RH, Porter CJ, Edwards WD, Puga FJ, Seward JB. Atrioventricular septal defects. In: Emmanouilides GC, Riemenschneider TA, Allan HD, Gutgesell HP, eds. Moss and Adams’ Heart Disease in Infants, Children and Adolescents, 5th edn. Baltimore: Williams & Wilkins, 1995: 704–24. 15. Huggon IC, Cook AC, Smeeton C, Magee AG, Sharland GK. Atrioventricular septal defects diagnosed in fetal life: associated cardiac and extra-cardiac abnormalities and outcome. J Am Coll Cardiol 2000; 36: 593–601. 16. Delisle MF, Sandor GG, Tessier F, Farquharson DF. Outcome of fetuses diagnosed with atrioventricular septal defects. Obstet Gynecol 1999; 94: 763–7. 17. Allan LD. Atrioventricular septal defect in the fetus. Am J Obstet Gynecol 1999; 181: 1250–3. 18. Silverman NH, Kleinman CS, Rudolph AM et al. Fetal atrioventricular valve insufficiency associated with nonimmune hydrops: a twodimensional echocardiographic and pulsed Doppler ultrasound study. Circulation 1985; 72: 825–32. 19. Schmidt KG, Ulmer HF, Silverman NH, Kleinman CS, Copel JA. Perinatal outcome of fetal complete atrioventricular block: a multicenter experience. J Am Coll Cardiol 1991; 17: 1360–6. 20. Anderson RH, Macartney FJ, Shinebourne EA, Tynan M. Ventricular septal defects. In: Anderson RH, ed. Paediatric Cardiology. London: McGraw-Hill, 1987: 565–90. 21. Ferencz C, Rubin DJ, Loffredo AC, Magee AC, eds. The Epidemiology of Congenital Heart Disease, The Baltimore– Washington Infant Study 1981–1989. Perspectives in Pediatric Cardiology, Vol. 4. Mount Kisco, NY: Futura Publishing, 1993: 31–3. 22. Rudolph AM. Congenital Disease of the Heart: Clinical– Physiological Considerations. Mount Kisco, NY: Futura Publishing, 2001: 198–9. 23. Canale JM, Sahn DJ, Allen HD et al. Factors affecting real-time, cross-sectional echocardiographic imaging of perimembranous ventricular septal defects. Circulation 1981; 63: 689–97. 24. Lethor JP, Maron F, de Moor M, King MEE. Physiology of ventricular septal defect shunt flow in the fetus examined by color Doppler M-mode. Circulation 2000; 101: e93. 25. Chao RC, Ho ESC, Hsieh KS. Fluctuations of interventricular shunting in a fetus with an isolated ventricular septal defect. Am Heart J 1994; 127: 955–8. 26. Yagel S, Benachi A, Bonnet D et al. Rendering in fetal cardiac scanning: the intracardiac septa and the coronal valve planes. Ultrasound Obstet Gynecol 2006; 28: 266–74. 27. Paladini D, Palmieri S, Lamberti A et al. Characterization and natural history of ventricular septal defects in the fetus. Ultrasound Obstet Gynecol 2000; 16: 118–22. 28. Axt-Fliender R, Schwarze A, Smrcek J et al. Isolated ventricular septal defects detected by color Doppler imaging: evolution during fetal and first year of postnatal life. Ultrasound Obstet Gynecol 2006; 27: 266–73. 29. Masuda M, Kado H, Kajihara N et al. Early and late results of total correction of congenital cardiac anomalies in infancy. Jpn J Thorac Cardiovasc Surg 2001; 49: 497–503.

20 Left heart malformations Lindsey D Allan Malformations of the left heart are commonly detected in fetal life. They constituted 32% of a total series of 2136 malformations reported by Allan.1 Anomalies affecting the left heart include mitral and aortic valve abnormalities, subaortic stenosis, aortic interruption, and coarctation.

Mitral valve anomalies Lesions that affect the mitral valve include mitral atresia, stenosis, or incompetence. Mitral atresia most commonly occurs in association with aortic atresia in the setting of the hypoplastic left heart syndrome, but can also occur with a ventricular septal defect and concordant great arteries or with double-outlet right ventricle. Mitral atresia is the most common mitral valve anomaly seen prenatally. Mitral stenosis as an isolated lesion is uncommon and difficult to diagnose in fetal life. It is usually found in association with aortic stenosis or atresia. Mitral incompetence can occur in a dysplastic mitral valve, which is uncommon, or in aortic stenosis. It can also be seen in association with fetal cardiac dilatation and heart failure.

Fetal diagnosis of mitral valve disease A normal mitral valve has an annulus of normal size when compared to normal ranges for the gestational age,2 with two thin leaflets that open in diastole, in addition to two normally spaced papillary muscles in the left ventricle. On the four-chamber view, both atrioventricular valves should have equal excursion in the moving image. If there is any degree of obstruction to the mitral valve or regurgitation across it, the four-chamber view will usually be abnormal.

Mitral atresia Mitral atresia in association with aortic atresia in the hypoplastic left heart syndrome is described below under

aortic atresia. It can also occur with a concordant patent aorta and ventricular septal defect or with double-outlet right ventricle.3 Mitral atresia is diagnosed in the fourchamber view by seeing no opening valve into the left ventricle and no flow across the mitral valve on color flow mapping (Figure 20.1). The left atrium is usually small and the interatrial shunt is exclusively left to right. The left ventricle is hypoplastic in association with mitral atresia but the degree of hypoplasia is variable. A ventricular septal defect may be present. If this is sizable, the left ventricle may be of near normal size and apex-forming, particularly with a concordant patent aorta. When the aorta is concordantly connected, the great arteries arise in a normal relationship to each other, although the aorta will usually be significantly smaller than the pulmonary trunk. In contrast, in mitral atresia with double-outlet connection, the great arteries either may be normally related or may arise in parallel orientation (Figure 20.2). If there is pulmonary stenosis, which is not uncommon, the pulmonary trunk will be smaller than the aorta. It is important to distinguish between mitral atresia with a ventricular septal defect and mitral atresia with double outlet, as the latter is more frequently associated with extracardiac anomalies. In addition, it is important to distinguish between mitral atresia with ventricular septal defect or double-outlet right ventricle and mitral and aortic atresia (the classic form of hypoplastic left heart syndrome, described below), as the postnatal course will be different.

Mitral stenosis In fetal life, mitral stenosis is associated with a redistribution of flow toward the right atrium through the foramen ovale. This results in increased flow to the right heart and reduced flow to the left heart. The left atrium, mitral valve orifice, left ventricle, and aorta are smaller than normal and the right atrium, right ventricle, and pulmonary artery dilated. There may be a mild increase in the velocity of mitral valve flow, although blood flow redistribution at the foramen often masks this. There may be visible dysplasia of the mitral valve (Figure 20.3) and the interatrial shunt may be reversed.

292

Fetal Cardiology

Figure 20.1 The heart is seen in a four-chamber view. The left atrium and ventricle are smaller than normal but the left ventricle reaches the apex. In the moving image, the mitral valve was not opening in diastole, mitral atresia. In the right-hand panel, on color flow mapping, there is forward flow through the tricuspid in diastole but not through the mitral valve. There is a moderate-sized ventricular septal defect (VSD). LV, left ventricle; RA, right atrium.

The discrepancy in ventricular sizes may be more obvious in the presence of additional left heart obstructive lesions including coarctation of the aorta, aortic valve disease, and more severe left ventricular outflow tract obstruction, so the subaortic area, the aortic valve, and aortic arch should be examined carefully. On the other hand, in the presence of less severe stenosis, there may be only a subtle discrepancy in ventricular sizes in the four-chamber view. Assessment of right ventricular and left ventricular size, as well as tricuspid and mitral valve diameters, may be useful, especially when their ratios are compared to the normal range. The chordae of the mitral valve may be fused or shortened, resulting in limited leaflet mobility and valve excursion. In a parachute mitral valve the chordal attachments are to a single papillary muscle, which restricts the mobility of the mitral valve leaflets. Although assessment of the mitral valve morphology is important, additional cardiovascular lesions must be excluded, as the outcome may be dictated by the other lesions.

Mitral regurgitation Mitral regurgitation is diagnosed by recognizing reverse flow from the valve during systole on color flow mapping and by confirming the finding on pulsed Doppler. The left atrium, in particular, but also the left ventricle, may appear dilated in a four-chamber view, depending on the severity of the regurgitation, although off-loading of the extra

volume across the foramen ovale may mask this. Mitral regurgitation may occur secondary to dysplasia of the mitral valve leaflets. Mitral valve anomalies may be a clue to underlying aortic valve or arch anomalies. Mitral and tricuspid regurgitation are commonly seen as secondary functional findings in cardiomyopathy (Figure 20.4) or in tachycardia in valves of normal morphology, but these are usually easily distinguishable as underlying causes.

Aortic valve anomalies Lesions that affect the aortic valve include atresia, stenosis, or incompetence, or an aortic to left ventricular tunnel. Aortic atresia in association with mitral atresia or severe stenosis (the hypoplastic left heart syndrome) is one of the most common forms of heart disease seen in the fetus.4 Aortic atresia can occur with a ventricular septal defect and a normal left ventricular size, but this is rare. Aortic stenosis is an important form of heart disease in the fetus. It tends to be critical when detected prenatally, but less severe forms have also been reported. Aortic incompetence is rarely seen in the fetus except in severe myocardial dysfunction. An aortic–left ventricular tunnel is a rare lesion where incompetence occurs around the aortic valve ring through a communication between the ascending aorta and the left ventricle.5 Subaortic stenosis is unusual prenatally but has been seen in a few cases.

Left heart malformations

293

Ao

LA RV

spine

spine

PA

(a)

(b)

Figure 20.2 (a) In this fetus with mitral atresia, there is a small left atrium and the slit-like left ventricle could barely be seen (yellow arrow). On the right-hand panel, the great arteries both arose from the right ventricle (RV) in parallel orientation with the aorta (Ao) anterior to the pulmonary artery. The aorta is significantly smaller than the pulmonary artery (PA) raising the suspicion of coarctation in addition. (b) In another fetus with mitral atresia, no left ventricle could be seen. The great arteries both arose from the right ventricle, but were normally related in position. The pulmonary artery was smaller than the aorta indicating a degree of pulmonary stenosis.

294

Fetal Cardiology

Figure 20.3 In the four-chamber view, the left ventricle is smaller than the right. The mitral orifice is smaller than normal and the valve appears dysplastic. The apparent displacement, which does not occur in the mitral valve, is artifact and is due to the dysplasia.

Figure 20.4 There is tricuspid and mitral regurgitation (MR) in this fetus with a biventricular cardiomyopathy.

Fetal diagnosis of aortic valve disease Aortic atresia The most common setting for aortic atresia is that of the hypoplastic left heart syndrome. These cases are readily detectable during routine obstetric ultrasound scanning as there will be an abnormal four-chamber view. The size

Figure 20.5 In the four-chamber view, the left ventricle is echogenic and thick-walled with a small cavity. The right ventricle forms the apex. In this fetus, there was mitral and aortic atresia, the typical form of the hypoplastic left heart syndrome.

and function of the left ventricle are abnormal. In cases associated with mitral atresia, the left ventricle will appear slit-like and is often not discernible. In some cases of aortic atresia and hypoplastic left heart syndrome, the mitral valve may be miniature but patent. In these instances, the left ventricular cavity is more easily recognized, with an echogenic, globular, and dysfunctional chamber (Figure 20.5). The ascending aorta is usually hypoplastic, often threadlike, but the degree of hypoplasia can be variable. No forward flow is detectable across the aortic valve on pulsed Doppler or color flow mapping and retrograde flow from the duct in the transverse arch is an important confirmation of the diagnosis (Figure 20.6). The foramen ovale is usually patent, but in some cases can be restrictive or, more rarely, intact. A clue to the presence of a restrictive atrial septum can be the finding of an abnormal pulmonary venous pulsed Doppler flow pattern, so this should be examined as part of the assessment of this condition (Figure 20.7). An increase in the reverse flow wave into the pulmonary veins during atrial systole correlates with restriction of the foramen.6 When the foramen ovale defect is patent, the interatrial shunt is left to right, the reverse of normal. When diagnosis of the hypoplastic left heart syndrome is made, it is essential to examine the right heart in detail to assess the suitability for Norwood palliative surgery.7 Mild degrees of tricuspid valve dysplasia are common, but if there is significant tricuspid regurgitation, this can represent a risk factor after surgery. Pulmonary stenosis can preclude Norwood repair, as would right ventricular dysfunction. Neonatal heart transplantation may be considered as an alternative form of treatment.8

Left heart malformations

295

Figure 20.6 The long-axis view of the aortic arch in the left-hand panel shows a large patent arterial duct and a small transverse aortic arch with reverse flow within it. In the right-hand panel, the arch and duct are imaged in a transverse section of the upper thorax. Flow should normally be in the same direction but here there is reverse flow in the arch due to aortic atresia. The arch is also hypoplastic when compared in size to the duct.

Figure 20.7 An abnormal pulmonary vein trace is seen with reverse flow in diastole during atrial contraction (A). ED, early diastole; S, systole. This was in the setting of a hypoplastic left heart with a restrictive atrial septum.

Aortic stenosis The appearance of the fetal heart will depend on the degree of obstruction. If aortic stenosis is moderate to severe, the left ventricle may appear normal or only mildly hypertrophied. Color flow mapping may reveal associated mitral regurgitation (Figure 20.8), which may give the first clue to the diagnosis. The aortic valve may appear abnormal, with turbulence to flow occurring at the valve orifice. An increased aortic Doppler velocity above the normal range for the gestational age will confirm that the valve is stenotic. If aortic stenosis is critical, typically the left ventricle is dilated and poorly contracting with evidence of increased echogenicity of the ventricular walls and papillary muscles

of the mitral valve (Figure 20.9). This appearance is suggestive of associated endocardial fibroelastosis. The mitral valve may be restricted in opening, owing to increased left ventricular diastolic pressure or to associated mitral stenosis. It may be difficult to demonstrate forward flow across the aortic valve in these cases, particularly flow at a high velocity, because of poor ventricular function. However, the aortic Doppler velocity, in cases with significant left ventricular compromise, may be within the normal range for gestation or only mildly elevated, usually in the range of 1–2 m/s, and often does not reflect the severity of the obstruction in this situation (Figure 20.10). In contrast, in less critical cases, with preservation of left ventricular function, velocities of up to 4 m/s have been obtained in

296

Fetal Cardiology

the fetus. If there is mitral regurgitation, the jet is at high velocity and may help to predict left ventricular pressure. Occasionally there may be evidence of mitral stenosis. The shortening fraction of the left ventricle may be severely reduced.9 The aortic valve will be thickened and doming. The aortic root and ascending aorta may be within the normal range in the mid-trimester, but become smaller than normal for the gestational age in the last trimester

of pregnancy.9–11 If aortic stenosis is critical, there will be reversed flow in the transverse aortic arch (see Figure 20.6). This can be an important discriminating feature for those cases with a poor prognosis for postnatal treatment. There is usually reversal of the interatrial shunt, but in some cases the atrial septum may be intact. In this group, there will be left atrial enlargement with an increased cardiothoracic ratio. Sequential studies have shown a reduced rate of growth of left heart structures in cases of severe left ventricular outflow tract obstruction identified in the midtrimester.9,11,12 This has also been shown by experimentation in the animal model.13 Thus, critical aortic stenosis can progress to the hypoplastic left heart syndrome by term, owing to failure of growth of the left ventricle. In some fetuses with aortic stenosis, the apex of the heart may be gradually taken over by the right ventricle as pregnancy advances, as the right ventricle grows normally and the left does not. Therefore, at term, a biventricular repair may not be considered feasible, even though there is still forward flow through the left heart. Where there is milder aortic stenosis in the mid-trimester there may be more normal growth of the left ventricle, although abnormal growth of the aortic valve may result in increasing stenosis by the time of birth.10

Figure 20.8

Aortic–left ventricular tunnel

The heart is seen in the four-chamber view. The left atrium and ventricle are dilated in a case of critical aortic stenosis with an intact atrial septum. There was significant mitral regurgitation (MR).

Aortic–left ventricular tunnel is a very rare congenital heart malformation, which comprises an abnormal communication or channel that originates in the ascending aorta,

Figure 20.9 In two further examples of aortic stenosis, note the globular shape to the left ventricle. Increased echogenicity of the mitral valve chordae is seen in the left-hand panel. In the right-hand panel, there is filling of the right ventricle on color flow mapping but none in the left during diastole. LA, left atrium.

Left heart malformations

297

Figure 20.10 The Doppler sample volume was positioned in the ascending aorta just beyond the aortic valve leaflets. The velocity measured was just over 2 m/s. This is abnormally high, but predicts a gradient of only 24 mmHg across the narrowed valve. However, this is typical of critical aortic stenosis, where a high gradient across the obstructed valve is not generated due to left ventricular dysfunction.

bypasses the aortic valve, and terminates in the left ventricle. The features of the lesion seen during fetal life which give a clue to this rare diagnosis are aortic regurgitation with a dilated ascending aorta, and a dilated left ventricle.5 A dilated left ventricle relative to the right is potentially detectable by examining the four-chamber view of the fetal heart, and thus this lesion can be detected during obstetric ultrasound examination. Color flow mapping will demonstrate turbulent regurgitant flow, which may initially appear to be across the aortic valve, but closer inspection will reveal this to be around the valve. The aortic valve can sometimes appear thickened. The dilated aortic root can potentially ‘balloon’ into the right ventricular outflow tract and cause some obstruction to right heart flow.

recognized in fetal life, whereas interrupted aortic arch is rare both in fetal and in postnatal life. Aortic arch anomalies accounted for 6.1% of all cardiovascular lesions detected antenatally in the combined series of Allan et al,4 and is detected with increasing frequency as routine obstetric scanning improves.14 Coarctation of the aorta can be identified at the time of obstetric scanning by detecting a discrepancy between the sizes of the left and right ventricles. Arch obstruction when diagnosed antenatally represents a more severe spectrum of disease than that seen postnatally, as the lesion is more likely to be associated with significant intracardiac pathology.

Fetal diagnosis of arch anomalies Other aortic valve anomalies Aortic regurgitation can occur secondary to severe left ventricular cardiomyopathy, but this is rare. Subaortic stenosis is also rare in the fetus but can be seen especially as part of Schone syndrome, which is the association of coarctation with mitral valve disease and obstruction to the left ventricular outflow tract. Endocardial fibroelastosis as a primary lesion is very rare and is usually echocardiographically indistinguishable from that occurring in association with critical aortic stenosis. This diagnosis is made at autopsy.

Arch anomalies These include coarctation and interruption of the aorta. Coarctation of the aorta is a fairly common lesion to be

The antenatal diagnosis of coarctation of the aorta is often suspected initially by the detection of ventricular size discrepancy in the four-chamber view (Figure 20.11).15 Ventricular size discrepancy, however, is not consistently present in the setting of isolated coarctation of the aorta.16 Discrepancy of great artery size is more reliable in the diagnosis of fetal coarctation of the aorta (Figure 20.12), present in one study in 75% of affected fetuses, although it is still not a consistent finding. However, the aorta to pulmonary artery ratio is a useful measurement to help discriminate true cases17,18 and also for sequential evaluation. The most sensitive diagnostic feature of fetal coarctation of the aorta which has been shown so far is the presence of transverse aortic arch and isthmic narrowing. In the series of Hornberger et al, of 20 fetuses with postnatal confirmation of coarctation of the aorta with or without intracardiac lesions, 80% had transverse aortic arch hypoplasia

298

Fetal Cardiology

Figure 20.11 There is a difference in ventricular sizes in this fetus with coarctation of the aorta. The right heart is dilated relative to the left. The color flow map sometimes accentuates this discrepancy.

Figure 20.12 In the left-hand panel, the left ventricle appears smaller than the right and the ascending aorta small. When the pulmonary artery is imaged in the same fetus at the same magnification, lying immediately above the previous section, it can be seen to measure almost twice the size of the aorta. This abnormal aorta to pulmonary artery ratio is a fairly reliable guide to the presence of coarctation.

and 100% had isthmus hypoplasia.16 Comparison of fetal aortic arch dimensions with the normal range is useful in the prenatal detection of aortic coarctation,19,20 as is direct comparison of the aortic arch size with ductal size. In a cross-sectional view of the upper thorax, to image both the ductal and the aortic arches simultaneously, significant discrepancy in the size of the arches is an important clue to the potential presence of coarctation of the aorta (Figure 20.13). Bronshtein and Zimmer described a

long-axis view of the aortic arch to image the junction of the duct and the distal arch, which they called the Y connection.21 Normally these two vessels are equal in size at this junction. However, a dilated duct and small aorta at this site was found in 13 cases, where coarctation was subsequently confirmed. I personally find the long-axis views of the aortic arch misleading as the arch can appear quite normal in this view in the presence of coarctation, or alternatively a normal arch can be made to appear

Left heart malformations

299

Figure 20.13 When the arch and duct are imaged simultaneously in the upper thorax their relative sizes can be compared. In the normal fetus, they are almost equal in size. In coarctation of the aorta, the transverse arch is significantly hypoplastic relative to the larger arterial duct, the discrepancy more severe in the right-hand than in the left-hand panel.

Figure 20.14 The aortic arch on the left-hand panel appears normal on color flow mapping. However, this fetus had coarcation of the aorta. In the right-hand panel, the distal part of the aortic arch, the isthmus (arrow), appears narrow but this appearance is within normal. This fetus had no suggestion of coarctation in other views.

abnormal (Figure 20.14). Reconstruction of the aortic arch with new technologies can be beautiful but is quite difficult and time-consuming and also not always useful (Figure 20.15). When coarctation of the aorta is suspected, further investigation for additional intracardiac lesions should be made, such as mitral valve abnormalities, ventricular septal defects, and left ventricular outflow tract obstruction in the form of subvalvular, valvular, or supravalvular

aortic stenosis. Conversely, where complex intracardiac pathology is detected, if the aorta is significantly smaller than the pulmonary artery, the aortic arch should be imaged for the possibility of coarctation of the aorta as an additional malformation. Pulsed Doppler is not useful in the diagnosis of isolated coarctation of the aorta, but color flow mapping can help to delineate the hypoplastic lumen of the distal arch for

300

Fetal Cardiology

Figure 20.15 On the left-hand panel, the distal arch appears to be normal but this fetus had coarctation. In contrast, there is a suspicious area of narrowing in the rendered image in the right-hand panel, yet this fetus had no evidence of coarctation from other findings.

comparison to the ductal arch, which is usually quite large. Blood flow through the foramen ovale is usually bidirectional in severe cases of coarctation. Color flow mapping can aid in the detection of associated ventricular septal defects. In coarctation, flow in a ventricular septal defect is usually mainly left to right, in contrast to bidirectional shunting, which is usual in an isolated ventricular septal defect. Coarctation of the aorta may develop prenatally with progression in the degree of distal arch hypoplasia and also of other left heart obstructive lesions, which may be subtle at the initial antenatal examination in the second trimester. Thus, a case of coarctation with a good-sized aortic arch in early fetal life may become one with more extensive arch hypoplasia by the time of delivery, necessitating a more difficult surgical repair. Also, in a case of coarctation with a mildly small left ventricle in early pregnancy, poor growth of the left ventricle during the rest of pregnancy can lead to a ventricle of inadequate size for a twoventricle repair. Although discrepancy in ventricular or great artery size may be present in the setting of an interrupted aortic arch, the diagnosis is confirmed by demonstrating the pathology at the level of the aortic arch (Figure 20.16). A ventricular septal defect is almost invariably associated. When this is large, there is usually no significant discrepancy in ventricular size. A significant discrepancy in great vessel size, however, is observed. In the most common form, interrupted aortic arch type B, the ascending aorta has an unusual orientation and often appears to have a straight course to the head and neck. Interruption may also be associated with progressive left ventricular outflow tract obstruction.

Associated extracardiac abnormalities Extracardiac abnormalities in fetal left heart disease can occur. Polyvalvular disease with dysplasia and redundancy of both the atrioventricular and the semilunar valves can be detected in utero in the setting of trisomy 18, often in association with a large-inlet ventricular septal defect. In a series of 108 cases of mitral atresia,22 17 had associated chromosomal anomalies (16%), most commonly trisomy 18. The hypoplastic left heart syndrome is associated with chromosomal anomalies in about 2–4% of cases,14 particularly Turner syndrome, but also trisomies 18 and 13. Valvular aortic stenosis is not commonly associated with extracardiac malformations. Coarctation of the aorta in fetal series is commonly associated with chromosomal malformations, nearly 30% in one series.14 Interrupted aortic arch has a high incidence of microdeletion of chromosome 22.23

Prenatal counseling and management Prenatal counseling regarding the presence of left heart disease is widely variable, depending on the precise anatomical lesion. In isolation, or in the context of less severe left heart obstructive lesions, the severity of mitral stenosis may be difficult to assess and therefore the postnatal implications difficult to predict. Even if the mitral valve initially appears

Left heart malformations

301

Figure 20.16 There is a large muscular ventricular septal defect in this case only seen in rather an oblique view. The ascending aorta appears to be straighter than normal and in other views was seen to branch into two, typical of an interrupted aortic arch type B. The left subclavian artery (LSA) arises from the junction of the arterial duct with the descending aorta.

normal in size, there may be progressive mitral valve hypoplasia throughout the remainder of gestation.11 Significant mitral regurgitation may result in the development of congestive heart failure.24 The immediate management of cases of mitral atresia will depend somewhat on the associated lesions, but all cases will result in a one-ventricle circulation in the long term. Aortic atresia lies at the most severe end of the spectrum of cardiac malformations Surgical options are now available for this group of abnormalities. However, in a study of 24 pregnancies which continued with an intention to treat surgically, there were only nine survivors (37.5%), with some patients excluded from surgery because of chromosomal anomalies, extracardiac malformations, or additional cardiac malformations.25 Some risk factors were not predictable prenatally, such as prematurity and evidence of neurological abnormality. In contrast, ‘ideal’ patients reaching Norwood stage 1 had a survival rate of over 80%. As yet, there is little information about the long-term survival for these children.26 It is likely that the single right ventricle will have a shorter lifespan than the single left ventricle, which usually fails after 20–30 years.27 If this happens, transplantation is all that is currently available for these patients. In addition, there have been reports of neurodevelopmental delay after Norwood palliation.28 The management of fetuses with critical aortic stenosis will depend on the severity of the lesion and the gestational age at which the diagnosis is made. Sequential studies have shown that failure of growth of the left ventricle is a frequent occurrence in fetuses with this condition. This has implications for counseling, particularly with regard to the

possibility of achieving a biventricular circulation after birth. It appears that fetuses seen in the mid-trimester with left ventricular dilatation and global dysfunction are likely to fall into the category of hypoplastic left heart syndrome at term.9,11 On the other hand, if aortic stenosis in the fetus is associated with a normal or near-normal ejection fraction, then the chances of a biventricular repair are increased. Therefore, when the diagnosis is made in the mid-trimester, all the postnatal surgical options should be discussed with the parents, including the possibility of a Norwood procedure. Prenatal balloon aortic valvuloplasty has been performed with the aim of improving left ventricular growth and function.29,30 Although prenatal intervention is an option, the selection of suitable cases poses a difficulty, as does the timing of the procedure. It should be reserved for those with the worst prognosis – for example, those with reversed flow in the arch and an intact atrial septum. Aortic–left ventricular tunnel can be a severe fetal cardiac malformation if regurgitation produces chronic volume overload and leads to the development of fetal hydrops. It is likely that the cases detected in utero represent the worst end of the spectrum of this abnormality. Coarctation of the aorta in fetal life can lead to poor left ventricular growth and an ‘unusably’ small left ventricle by term, such that a biventricular repair is impossible. This condition overlaps the spectrum of a true hypoplastic left heart. If the left heart size is adequate and coarctation of the aorta is symptomatic in the neonate, which occurs in the majority of those cases recognized prenatally, an extended arch repair is usually necessary. Although usually successful, this can have a significant mortality especially if there

302

Fetal Cardiology

are associated intracardiac lesions, such as a ventricular septal defect or mitral stenosis. Only a relatively small proportion of the cases detected in the fetus will be of the type where elective resection of the coarctation shelf can be performed at low risk in later infancy.

Fetal outcome The high mortality and morbidity associated with the treatment of most forms of left heart disease that are recognized in fetal life lead to a high incidence of pregnancy interruption. Counseling should be individualized depending on the known natural history of each type of lesion as it occurs in the fetus. Long-term potential morbidity, especially in cases likely to come to a Fontan (or one-ventricle) repair, must be included. Spontaneous intrauterine death can occur in a small proportion of cases, especially in coarctation with Turner syndrome31 or in aortic stenosis where hydrops develops. Pregnancies involving nearly all forms of left heart disease recognized prenatally, where the pregnancy continues, should be delivered in or close to a cardiac center in the hope of achieving a successful early intervention. Prostaglandin therapy should be initiated when there is retrograde arch flow or if a coarctation lesion is clearly demonstrated on the postnatal echocardiogram. Prenatal diagnosis results in improved neonatal hemodynamic condition after delivery, and this could potentially improve the outcome for all types of left heart disease.32,33 However, there is still a significant mortality for the staged Norwood repair. The long-term outlook in children requiring a one-ventricle approach must be guarded.

References 1. Allan LD. Appendix. In: Allan LD, Hornberger LK, Sharland GK, eds. Fetal Cardiology. London: Greenwich Medical Press, 2000. 2. Schneider C, McCrindle BW, Carvalho JS et al. Development of Z-scores for fetal cardiac dimensions from echocardiography. Ultrasound Obstet Gynecol 2005; 26: 599–605. 3. Allan LD, Anderson RH, Cook AC. Atresia or absence of the left-sided atrioventricular connection in the fetus: echocardiographic diagnosis and outcome. Ultrasound Obstet Gynecol 1996; 8: 295–302. 4. Allan LD, Sharland GK, Milburn A et al. Prospective diagnosis of 1, 006 consecutive cases of congenital heart disease in the fetus. J Am Coll Cardiol 1994; 23: 1452–8. 5. Cook AC, Fagg NLK, Ho SY et al. Echocardiographic– anatomical correlations in aortico-left ventricular tunnel. Br Heart J 1995; 74: 443–8. 6. Better DJ, Apfel HD, Zidere V, Allan LD. Pattern of pulmonary venous blood flow in the hypoplastic left heart syndrome in the fetus. Heart 1999; 81: 646–50.

7. O’Kelly SW, Bove EL. Hypoplastic left heart syndrome: terminal care is not the only option. Br Med J 1997; 314: 87–8. 8. Bailey LL, Assaad AN, Trimm RF et al. Orthotopic transplantation during early infancy for incurable congenital heart disease. Ann Surg 1988; 208: 279–86. 9. Sharland GK, Chita SK, Fagg N et al. Left ventricular dysfunction in the fetus: relation to aortic valve anomalies and endocardial fibroelastosis. Br Heart J 1991; 66: 219–24. 10. Simpson JM, Sharland GK. Natural history and outcome of aortic stenosis diagnosed prenatally. Heart 1997; 77: 205–10. 11. Hornberger LK, Sanders SP, Rein AJJT et al. Left heart obstructive lesions and left ventricular growth in the midtrimester fetus. A longitudinal study. Circulation 1995; 92: 1531–8. 12. Allan LD, Sharland G, Tynan MJ. The natural history of the hypoplastic left heart syndrome. Int J Cardiol 1989; 25: 341–3. 13. Fishman NH, Hof RB, Rudolph AM, Heymann MA. Models of congenital heart disease in fetal lambs. Circulation 1978; 58: 354–64. 14. Del Bianco A, Russo S, Lacerenza N et al. Four chamber view plus three-vessel and trachea view for a complete evaluation of the fetal heart during the second trimester. J Perinat Med 2006; 34: 309–12. 15. Hornberger LK. Aortic arch anomalies. In: Allan LD, Hornberger LK, Sharland GK, eds. Fetal Cardiology. London: Greenwich Medical Press, 2000: 305–21. 16. Hornberger LK, Sahn DJ, Kleinman CS et al. Antenatal diagnosis of coarctation of the aorta: a multicenter experience. J Am Coll Cardiol 1994; 23: 417–23. 17. Sharland GK, Chan K-Y, Allan LD. Coarctation of the aorta: difficulties in prenatal diagnosis. Br Heart J 1994; 71: 70–5. 18. Allan LD, Chita SK, Anderson RH et al. Coarctation of the aorta in prenatal life: an echocardiographic, anatomical, and functional study. Br Heart J 1988; 59: 356–60. 19. Brown DL, Durfee SM, Hornberger LK. Ventricular discrepancy as a sonographic sign of coarctation of the fetal aorta: how reliable is it? J Ultrasound Med 1997; 16: 95–9. 20. Hornberger LK, Weintraub RG, Pesonen E et al. Echocardiographic study of the morphology and growth of the aortic arch in the human fetus. Observations related to the prenatal diagnosis of coarctation. Circulation 1992; 86: 741–7. 21. Bronshtein M, Zimmer EZ. Sonographic diagnosis of fetal coarctation of the aorta at 14–16 weeks of gestation. Ultrasound Obstet Gynecol 1998; 11: 254–7. 22. Hornberger LK. Mitral valve anomalies. In: Allan LD, Hornberger LK, Sharland GK, eds. Fetal Cardiology. London: Greenwich Medical Press, 2000: 148–62. 23. Ryan AK, Goodship JA, Wilson DI et al. Spectrum of clinical features associated with interstitial chromosome 22q11 deletions: a European collaborative study. J Med Genet 1997; 34: 798–804. 24. Silverman NH, Kleinman CS, Rudolph AM et al. Fetal atrioventricular valve insufficiency associated with nonimmune hydrops: a two-dimensional echocardiographic and pulsed Doppler ultrasound study. Circulation 1985; 72: 825–32. 25. Allan LD, Apfel HD, Printz BF. Outcome after prenatal diagnosis of the hypoplastic left heart syndrome. Heart 1998; 79: 371–4. 26. Mahle WT, Spray TL, Wernovsky G et al. 3rd. Survival after reconstructive surgery for hypoplastic left heart

Left heart malformations

27.

28.

29.

30.

syndrome: a 15-year experience from a single institution. Circulation 2000; 102: 136–41. Fontan F, Kirklin JW, Fernandez G et al. Outcome after a ‘perfect’ Fontan operation. Circulation 1990; 81: 1520–36. Rogers BT, Msall ME, Buck GM et al. Neurodevelopmental outcome of infants with hypoplastic left heart syndrome. J Pediatr 1995; 126: 496–8. Maxwell D, Allan LD, Tynan MJ. Balloon dilatation of the aortic valve in the fetus: a report of two cases. Br Heart J 1991; 65: 256–8. Makikallio K, McElhinney DB, Levine JC et al. Fetal aortic valve stenosis and the evolution of hypoplastic left heart

303

syndrome: patient selection for fetal intervention. Circulation 2006; 113: 1401–5. 31. Surerus E, Huggon IC, Allan LD. Turner’s syndrome in fetal life. Ultrasound Obstet Gynecol 2003; 22: 264–7. 32. Kumar RK, Newburger JW, Gauvreau K, Kamenir SA, Hornberger LK. Comparison of outcome when hypoplastic left heart syndrome and transposition of the great arteries are diagnosed prenatally versus when diagnosis of these two conditions is made only postnatally. Am J Cardiol 1999; 83: 1649–53. 33. Franklin O, Burch M, Manning N et al. Prenatal diagnosis of coarctation of the aorta improves survival and reduces morbidity. Heart 2002; 87: 67–9.

21 Ventricular outflow tract anomalies: so-called conotruncal anomalies Shi-Joon Yoo, Fraser Golding, and Edgar Jaeggi The ventricular outflow tracts are the most common sites for the occurrence of congenital heart disease (CHD). The anomalies involving the ventricular outflow tracts can be categorized into two groups: those occurring as an isolated lesion in an otherwise normally formed heart, and those occurring as a complex malformation with an abnormal ventriculoarterial connection and/or an abnormal great arterial relationship. The latter group includes tetralogy of Fallot, complete and corrected transpositions of the great arteries, double-outlet ventricles, and truncus arteriosus. These lesions are also known as ‘conotruncal or truncoconal anomalies’ because they are considered to result from an errant development of the conotruncal region of the embryonic heart.1 Although interruption of the aortic arch is also considered a conotruncal anomaly, it will be discussed in Chapter 22. Conotruncal anomalies account for 25–30% of CHD in infants (Table 21.1).2–4 Tetralogy and complete transposition are the two most common conotruncal anomalies, each occurring in approximately 8–12% of infants with cardiac defects. Double-outlet ventricles, truncus arteriosus, and corrected transposition are less common. Historically, conotruncal anomalies have a lower prevalence in fetal series than in infant series (Table 21.1).5,6

This discrepancy can be explained by a low detection rate of conotruncal anomalies at the time of fetal cardiac screening, if the four-chamber view is used alone.7 Recently there has been significant improvement in the detection rates of conotruncal anomalies owing to inclusion of outflow tract examination in the screening protocol (Table 21.1).8,9 As conotruncal anomalies often require assessment and treatment immediately after birth, prenatal detection is critically important. This chapter discusses the normal anatomy of the ventricular outflow tracts, the fetal echocardiographic techniques for their evaluation, the pathology, fetal echocardiographic findings, and postnatal outcomes of conotruncal anomalies.

Normal anatomy of the ventricular outflow tracts The right and left ventricular outflow tracts exhibit significant morphological differences (Figure 21.1). The major difference is the existence of the crista supraventricularis in the right ventricle. The crista supraventricularis is the

Table 21.1 Relative distribution of the conotruncal anomalies Diagnosis

NERICP2 1969–1974 (n = 2381)

Brompton3 1973–1982 (n = 1653)

NGP4 1982–1987 (n = 4735)

Fetal diagnosis5 1980–1992 (n = 1006)

Fetal diagnosis6 1991–2001 (n = 55 (97))*

Tetralogy of Fallot

8.9%

9.9%

11.5%

3%

5% (7%)

Complete transposition

9.9%

10.4%

10.1%

2%

15% (14%)

Corrected transposition

0.7%

0.8%

0.9%

< 1%

4% (3%)

Double-outlet right ventricle

1.5%

3.0%

2.0%

3%

4% (3%)

Truncus arteriosus

1.4%

2.1%

2.0%

1.5%

2% (1%)

*Percentages in parentheses are distributions of all detected and undetected cases at fetal ultrasound. NERICP, New England Regional Infant Cardiac Program; NGP, Northern Great Plains Regional Cardiac Program.

306

Fetal Cardiology

muscular crest that separates the pulmonary and tricuspid valves in the normal right ventricle.10 The major part of the crista is a parietal structure, which is called the ventriculoinfundibular fold. Only a small part of the crista is septal structure, which is called the outlet or infundibular septum. The outlet septum is cradled between the two limbs of the trabecula septomarginalis. The crista supraventricularis completes the muscular funnel of the right ventricular outflow tract. In the left ventricle, the aortic and mitral valves are in fibrous continuity and there is no muscular crest between the two valves. Therefore, the left ventricular outflow tract is not completely muscular, and is less

conspicuous than the right ventricular outflow tract. The two ventricular outflow tracts cross each other (Figure 21.2). The right ventricular outflow tract is anterior to the left ventricular outflow tract. As they course upward toward the arterial trunks, the right ventricular outflow tract faces leftward, and the left ventricular outflow tract faces rightward. The aortic valve is deeply wedged between the tricuspid and mitral valves (Figures 21.1c and 21.1d). This unique position allows a fibrous continuity between the aortic and mitral valves. Being supported by a muscular infundibulum that faces leftward, the pulmonary valve is located to the left of and anterior to the aortic valve.

(a)

(b)

(c)

(d)

Figure 21.1 Photographs of a normal cardiac specimen. Opened right (a) and left (b) ventricles, and the base of the ventricles seen from above (c) and below (d). As the crista supraventricularis (CS) intervenes between the tricuspid (TV) and pulmonary (PV) valves, the right ventricular outflow tract is a completely muscular funnel. The crista supraventricularis consists of parietal and septal parts: the ventriculoinfundibular fold (VIF) and outlet septum (OS), respectively. The left ventricular outflow tract is partly devoid of muscular wall because of the fibrous continuity (dots in (b) and (d)) between the mitral (MV) and aortic (AV) valves. Notice the deeply wedged position of the aortic valve between the tricuspid and mitral valves in (c). ms, membranous septum; TSM, trabecula septomarginalis.

Ventricular outflow tract anomalies

307

Figure 21.2 Diagram showing the long-axis view of the normal right (RV) and left (LV) ventricles. The two outflow tracts cross each other. A, aorta; d, ductus arteriosus; LA, left atrium; P, pulmonary artery; RA, right atrium; SVC, superior vena cava.

Echocardiographic techniques for fetal ventricular outflow tract evaluation In the evaluation of the ventricular outflow tracts, the ventricles and great arteries should be identified according to morphological criteria. After morphological identification, the ventriculoarterial connections and great arterial relationship should be evaluated using the following echocardiographic views (Figures 21.3 and 21.4):11–16 • • • •

left ventricular outflow tract view right ventricular outflow tract view basal short-axis view three-vessel view.

The left and right ventricular outflow tract views are necessary for determination of the type of ventriculoarterial connection. To obtain these two views, we start from a transducer position for a four-chamber view (Figure 21.5). The transducer is then moved radially around the maternal abdomen, keeping the four-chamber plane in view, until the ventricular septum is aligned perpendicular to the sonographic beam axis. From this particular transducer position, the left ventricular outflow tract view can be obtained simply by rotating the transducer through 40–50° clockwise or counterclockwise toward the cardiac

Figure 21.3 Diagram showing the imaging planes for echocardiographic views for ventricular outflow tract examination.

apex. The transducer is then moved slightly upward until the right ventricular outflow tract view is visualized. It should be noted that the normal right and left ventricular outflow tracts cross each other and that they cannot be visualized in a single two-dimensional imaging plane (Figures 21.2 and 21.4a). The basal short-axis view is used for evaluation of the morphology of the outlet septum and the ventricular outflow tract dimensions. The short-axis plane of the heart can be found by placing the transducer to connect the right lobe of the liver and the left shoulder of the fetus (Figure 21.3). The transducer is moved upward or downward along the fetal thorax with some cranial or caudal tilt until the aortic valve is located in the center of the cardiovascular section that visualizes the right atrium, right ventricle, main pulmonary artery, and right pulmonary artery. An alternative view is the vertical long-axis view of the right ventricle (Figure 21.6). This view can be obtained from the coronal view by displacing the transducer to the left anterior or right posterior part of the chest wall. The three-vessel view is an orthogonal transverse view of the upper mediastinum, where normally the oblique section of the main pulmonary artery and cross-sections of the ascending aorta and superior vena cava are arranged in a straight line from left anterior to right posterior with a decreasing order of their size (Figure 21.4c).12–15 This view is particularly helpful in the evaluation of the spatial

308

Fetal Cardiology

(a)

(b)

Figure 21.4 Echocardiograms and corresponding diagrams for ventricular outflow tract examination. (a) Left ventricular outflow tract view. (b) Right ventricular outflow tract view.

relationship and size of the great arteries, which is abnormal in almost all cases with a ventricular outflow tract anomaly. The three-vessel view can be obtained simply by sliding the transducer upward from the four-chamber plane toward the fetal upper mediastinum.

Three-dimensional (3D) ultrasound using the so-called ‘spin technique’ enhances the above described diagnostic approach to fetal outflow tract abnormalities.17 In the spin technique, the 3D transducer acquires a volume data set from the four-chamber plane, through the three-vessel

Ventricular outflow tract anomalies

309

(c)

(d)

Figure 21.4 Continued (c) Short-axis view of the base of the ventricles. (d) Three-vessel view. A, ascending aorta; a, descending aorta; AV, aortic valve; lpa, left pulmonary artery; P, main pulmonary artery; rpa, right pulmonary artery; V, superior vena cava.

plane to the plane for the aortic and ductal arches by using static 3D or the spatiotemporal image correlation (STIC) technique. From the volume data set, the images described above can be reconstructed for ventricular outflow tract assessment. Doppler interrogation is an important adjunct to two-dimensional imaging. It demonstrates the direction

and velocity of the blood flow through the outflow tracts. It also demonstrates the flow direction and velocity through the ductus arteriosus, which is important for recognition of a possibility of so-called ‘ductus-dependent’ pulmonary or systemic circulation after birth. Power and color Doppler mapping facilitates identification of the vessels.

310

Fetal Cardiology

Figure 21.5

RV LV

RA LA

(a)

Diagram showing the maneuver for scanning the ventricular outflow tracts in a fetus in cephalic presentation and supine position. The examination starts from the position providing the four-chamber view. The transducer is moved radially around the maternal abdomen (arrow 1) until the ventricular septum is aligned perpendicular to the sonographic beam axis. Then the left ventricular outflow tract is obtained by rotating the transducer toward the cardiac apex (arrow 2). By sliding the transducer upward along the fetal thorax, the right ventricular outflow tract view is obtained. A, descending aorta (reproduced with permission from reference 14).

(b)

Figure 21.6 Vertical long-axis view of the right ventricle (a) and corresponding diagram (b). The long-axis cuts of the three vessels are aligned from the left anterior to the right posterior aspect of the fetal thorax.

Ventricular outflow tract anomalies

Individual lesions Tetralogy of Fallot The four anatomical features of tetralogy of Fallot are a ventricular septal defect, subpulmonary stenosis, overriding aorta, and right ventricular hypertrophy (Figure 21.7).18,19 These features are the consequences of anterior, superior, and leftward deviation of the outlet septum from the rest of the ventricular septum, which is considered the essential pathology of tetralogy (Figure 21.8).

Figure 21.7 Diagram showing pathology of tetralogy of Fallot in long axial oblique view. VSD, ventricular septal defect.

311

In most cases with tetralogy, the situs is solitus and there is levocardia. The four-chamber view does not show any defect in the majority of cases. On close observation, however, one may notice that the cardiac apex is deviated leftward (Figure 21.9).20–22 The diagnosis of tetralogy cannot be made unless the outflow tracts are evaluated.19,20,23,24 Deviation of the outlet septum from the rest of the septum can be best shown in a basal short-axis or vertical long-axis view of the right ventricle (Figure 21.10). The deviated outlet septum is seen as a teardrop-like structure that encroaches on the outflow tract leading to the small pulmonary valve. The size of the outlet septum varies from case to case. Uncommonly, the outlet septum is completely missing, while the pulmonary valve annulus is smaller than the aortic valve annulus. Although the outlet septum is absent, this rare form is also regarded as a variant of tetralogy (Figure 21.11). The pulmonary valve that guards the small annulus is commonly stenotic. Doppler velocities across the narrowed right ventricular outflow tract are typically normal or only mildly increased in contrast to postnatal findings, unless the pulmonary valve is atretic.19,25 The ventricular septal defect in tetralogy involves the outlet part of the ventricular septum. Most commonly, it is a perimembranous defect, in which the tricuspid valve is in direct contact with the aortic valve through the posteroinferior border of the defect.18–20 The perimembranous defect is seen immediately below the overriding aortic valve in the left ventricular outflow tract view (Figure 21.10). It is also seen in the short-axis view or vertical long-axis view of the right ventricle as a defect extending from the tricuspid annulus toward the ventricular outlet (Figure 21.10b). The ventricular septal defect is less frequently a muscular outlet type, in which the posteroinferior border is the muscular rim separating the tricuspid valve from the aortic valve. The least common type of defect is a doubly committed juxta-arterial type,

Figure 21.8 Essential pathology of tetralogy of Fallot. The outlet septum (OS) is supported by the hinges along its left anterior margin. It is pulled anteriorly into the right ventricular outflow tract in tetralogy, leaving a ventricular septal defect (VSD) of anterior malalignment type. The aortic valve overrides the ventricular septum because it is displaced anteriorly with the outlet septum. The deviated outlet septum encroaches on the subpulmonary outflow tract dimension. TV, tricuspid valve.

312

Fetal Cardiology

Figure 21.9 Tetralogy of Fallot. Four-chamber view shows no defect. The cardiac axis, however, is deviated to the left. The descending aorta is seen on the right anterior aspect of the spine. This fetus had a right aortic arch.

(a)

(b)

Figure 21.10 Tetralogy of Fallot with a perimembranous ventricular septal defect. (a) Left ventricular outflow tract view shows a ventricular septal defect (D). The aorta (A) overrides the ventricular septum. (b, c) Vertical long-axis view of the right ventricle and corresponding diagram show that the subpulmonary outflow tract (asterisk) is encroached on by the deviated outlet septum (os). The ventricular septal defect is seen behind the deviated outlet septum. The aortic and tricuspid valves are in direct contact (arrow).

(c)

Ventricular outflow tract anomalies

Figure 21.11

313

Figure 21.12

Tetralogy of Fallot with a doubly committed juxta-arterial ventricular septal defect. Basal short-axis view shows that the aortic (A) and pulmonary (P) valves are in direct contact (arrow). The outlet septum is not seen below the pulmonary and aortic valves. The pulmonary valve annulus is smaller than the aortic valve annulus. The tricuspid and aortic valves are not in direct contact because of the intervening muscle piece (asterisk).

Tetralogy of Fallot. Three-vessel view shows the abnormal alignment and size of the three vessels. The ascending aorta (A) is dilated and displaced anteriorly. The main pulmonary artery (P) is small and displaced posteriorly. The right and left pulmonary arteries (rpa and lpa) are not small. a, descending aorta.

which is characterized by the absence or extreme hypoplasia of the outlet septum. In this type, the aortic and pulmonary valves are in direct contact through the anterosuperior border of the defect. Both muscular outlet and doubly committed defects are barely visible in the left ventricular outflow tract view but can be visualized in the short-axis view of the ventricular outflow tracts (Figure 21.11). Uncommonly, tetralogy may coexist with an atrioventricular septal defect, or additional muscular ventricular septal defect(s) may be present.24 In the majority of cases, there is a disproportion between the diameters of the main pulmonary and the ascending aorta, showing a larger aorta and a smaller main pulmonary artery. The size discrepancy of the arterial trunks may be subtle and may not be apparent in the early stage of gestation but becomes evident with advancing gestation.25 As each arterial trunk may measure within the normal range, it is important to evaluate the size discrepancy by direct visual comparison or calculation of the diameter ratio.20,25 Direct visual comparison is more practical and easier than measurement, and the three-vessel view is ideally suited for this comparison (Figure 21.12). As the dilated aorta is displaced anteriorly and the main pulmonary artery more or less posteriorly, the alignment

of the three vessels is almost always abnormal. In our previous study, the clue to the diagnosis of tetralogy was found most commonly in the three-vessel view.20 In contrast to the small size of the main pulmonary artery, the sizes of the branch pulmonary arteries are usually normal (Figure 21.12).19,25 Pulmonary arterial growth during later fetal life, however, is variable and unpredictable.23,25 Progression of the outflow tract obstruction is well documented, and small branch pulmonary arteries at mid-gestation suggest severe disease. A mirror-imaged right aortic arch is common in tetralogy. When there is a right aortic arch, the descending aorta is seen on the right anterior aspect of the spine in the three-vessel view. As the trachea is filled with fluid in the fetus, the position of the aortic arch relative to the trachea can be directly visualized (Figure 21.13).15,20,26 Rarely, the right or left subclavian artery may arise aberrantly from the descending aorta. The aberrant branch always courses behind the esophagus and thus is seen behind the trachea. The ductus arteriosus is characteristically small and often difficult to identify in tetralogy.20,25 The ductus often arises from the undersurface of the aortic arch (Figure 21.14). In most cases with a right aortic arch, the

314

Fetal Cardiology

(a)

(b)

Figure 21.13

(c)

Right aortic arch in a fetus with tetralogy of Fallot. (a) Transverse view of the upper thorax shows the aortic arch to the right of the trachea. (b) Transverse view of the thorax at the level of the arterial roots shows the descending aorta located at the right anterior aspect of the spine. (c) Long axis view of the aortic arch shows a narrow arc of the arch.

patent ductus is the left ductus that arises from the left innominate or subclavian artery. The blood flow through the ductus varies considerably according to the severity of subpulmonary stenosis.20,23,27 The ductal flow may be right to left, bidirectional, or left to right. A left-to-right shunt through the ductus from the aorta to the pulmonary artery is predictive of severe disease that may require prostaglandin therapy immediately after birth and surgical correction in early life. The ductus arteriosus is typically

absent when tetralogy is complicated by absent pulmonary valve syndrome.28–31 It is interesting, however, to note that a ductus is usually patent when absent pulmonary valve syndrome occurs with an intact ventricular septum or a simple ventricular septal defect.29,32 Many other lesions may be associated with tetralogy. A secundum type of atrial septal defect may be present but cannot be reliably diagnosed in the fetus. One pulmonary artery, usually the left, may be disconnected from the main

Ventricular outflow tract anomalies

(a)

(b)

Figure 21.14 Ductus arteriosus in tetralogy of Fallot. (a) Aortic arch view with color Doppler shows that the ductus arteriosus (PDA) is small and arises from the undersurface of the aortic arch. (b) Three-vessel view shows that the patent ductus arteriosus connects to the proximal left pulmonary artery (lpa). As the PDA arises from the undersurface of the aortic arch, it is some distance from the descending aorta.

315

(a)

(b)

Figure 21.15 Absent pulmonary valve syndrome in tetralogy of Fallot. (a) Basal short-axis view of the ventricles shows narrowing of the subpulmonary outflow tract (asterisks) due to anterior cephalad deviation of the outlet septum (os). The main pulmonary artery (p) and branch pulmonary arteries (rpa and lpa) are markedly dilated, while the pulmonary valve annulus is small (arrow). The pulmonary valve leaflets ar not identifiable. (b) Color Doppler image in transverse plane shows two-way streams, both forward and retrograde flow streams (arrows) within the dilated right and left pulmonary arteries. Note that the accelerated turbulent flow starts at the level of the pulmonary valve annulus. AV, aortic valve; D, ventricular septal defect.

316

Fetal Cardiology

(a)

(b)

Figure 21.16 Major aortopulmonary collateral artery in pulmonary atresia with ventricular septal defect. Transverse views of thorax with color (a) and power (b) Doppler show a tortuous collateral artery that arises from the descending aorta and connects to the left pulmonary artery, which makes a confluence with the right pulmonary artery at the main pulmonary artery.

pulmonary artery and connected to the ipsilateral ductus arteriosus. Absent pulmonary valve syndrome accounts for 3–6% of all cases with tetralogy.28–31 Absent pulmonary valve syndrome is characterized by vestigial leaflets guarding the small pulmonary valve annulus and gross pulmonary regurgitation (Figure 21.15). As mentioned, the ductus arteriosus is absent in the majority of cases with tetralogy of Fallot and absent pulmonary syndrome. Aneurysmal dilatation of the central pulmonary arteries is a constant feature of absent pulmonary valve syndrome after 22 weeks.31 It is considered to be due to the combined effect of poststenotic dilatation, severe pulmonary regurgitation and absence of ductal pathway for run-off of blood flow through the main pulmonary artery into the systemic circulation. The dilated pulmonary arteries taper down abruptly in the lungs. A dilated pulmonary artery may be mistaken for a cystic mass in the mediastinum. As the pulmonary artery compresses the tracheobronchial tree and esophagus, polyhydramnios may develop.33 Because of severe pulmonary regurgitation, the heart is large with right ventricular dilatation from early gestation. In those cases where the ductus arteriosus is patent, the pulmonary arterial dilatation is not severe.

Pulmonary atresia with ventricular septal defect is an extreme form of tetralogy. It has been shown that tetralogy with patent pulmonary outflow tract may progress to pulmonary atresia in fetal life.19,23–25,34 Pulmonary atresia may occur at infundibular or valvar level. The main pulmonary artery is often not formed, and the branch pulmonary arteries may not be confluent. The source of blood flow to the lungs may be either ductus arteriosus or major aortopulmonary collateral arteries. The collateral arteries arise most commonly from the descending thoracic aorta, less often from the bracheocephalic branches, and rarely from a coronary artery or arteries. Color or power Doppler interrogation facilitates demonstration of the ductus or collateral arteries supplying the lungs (Figures 21.14 and 21.16). A ductus arteriosus connects with the pulmonary arteries in the mediastinum, while most collateral arteries connect to the pulmonary arteries at the hilum or within the lungs. Among the various echocardiographic findings, those highly predictive of severe postnatal disease are main pulmonary artery hypoplasia at the initial examination, little or no growth of the main and branch pulmonary arteries on follow-up, and retrograde flow through the ductus arteriosus.25

Ventricular outflow tract anomalies

317

Complete transposition of the great arteries Transposition of the great arteries refers to a condition in which the great arteries are placed across the ventricular septum, and therefore the aorta arises from the right ventricle and the pulmonary artery from the left ventricle. ‘Transposition of the great arteries’ is synonymous with the term ‘discordant ventriculoarterial connection’. When transposition occurs with normal concordant atrioventricular connection, i.e. the right atrium connects to the right ventricle and the left atrium to the left ventricle, it is called complete transposition (Figure 21.17).35 When

Figure 21.17 Diagram showing pathology of complete transposition of the great arteries.

Figure 21.18 Complete transposition of the great arteries. (a) Three-vessel view shows the classic great arterial relationship of complete transposition. The three vessels are arranged in a triangular fashion with the right anterior aorta (A) and left posterior pulmonary artery (P). (b) Ventricular outflow tract view shows discordant ventriculoarterial connection. As the right and left ventricular outflow tracts are parallel to each other, they are imaged in a single plane. (c) Sagittal view of the thorax shows the aortic arch (Aa). It forms a wide arc with the configuration of a hockey stick. It gives rise to the head and neck branches. The ductus arteriosus (d) forms a smaller arch underneath the aortic arch. (a)

(b)

(c)

318

Fetal Cardiology

Figure 21.19

Figure 21.20

Complete transposition of the great arteries with posterior malalignment type of ventricular septal defect and subpulmonary stenosis. The left ventricular outflow tract leading to the pulmonary valve is encroached upon (asterisk) by the posteriorly deviated outlet septum (os). The ventricular septal defect (D) is seen below the deviated outlet septum. The main pulmonary artery is smaller than the aorta.

Complete transposition of the great arteries with anterior malalignment type of ventricular septal defect. The aorta arises from the far anterior aspect of the right ventricle. The pulmonary valve orverrides the ventricular septum through the ventricular septal defect (D). When there is more than 50% overriding of the pulmonary valve, this condition can be called double-outlet right ventricle with subpulmonary ventricular septal defect. The subaortic outflow tract is narrow (asterisk). Note the size discrepancy between the aorta and pulmonary artery. The aortic arch was narrow.

it occurs with discordant atrioventricular connection, i.e. the right atrium connects to the left ventricle and the left atrium to the right ventricle, it is called congenitally corrected transposition. In most cases with complete transposition, the situs is solitus and there is levocardia. The right ventricle is the right-sided and anterior ventricle. The four-chamber view is usually normal when there is no associated lesion.36 The great arterial relationship is almost always abnormal, which can be easily recognized at the three-vessel view (Figure 21.18a).12–15 The great arterial relationship is closely related to the infundibular morphology. In classic transposition, the aorta is supported by a completely muscular infundibulum, while the pulmonary artery is not. The pulmonary valve is in fibrous continuity with the mitral valve. In this classic setting, the aorta is anterior to and to the right of the pulmonary artery. Rarely, the aorta may be located posterior to or to the left of the pulmonary artery.37 The ventricular outflow tracts are usually parallel, which allows visualization of both outflow tracts in a single imaging plane (Figure 21.18b). As mentioned, normal outflow tracts cross each other and cannot be visualized

in a single imaging plane. The aortic arch in complete transposition is shaped like a hockey stick and can be mistaken for a ductal arch (Figure 21.18c). In approximately 40% of cases with complete transposition, there is an associated ventricular septal defect (Figures 21.19 and 21.20). It may occupy any part of the ventricular septum. Particularly common is the outlet defect with posterior or anterior malalignment of the outlet septum. When there is posterior malalignment, the subpulmonary outflow tract is narrowed and the aortic valve may override the ventricular septum (Figure 21.19). When there is anterior malalignment, the subaortic outflow tract is narrowed and the pulmonary valve may override the ventricular septum (Figure 21.20). With a greater degree of overriding of the aortic or pulmonary valve, transposition merges into double-outlet left or right ventricle. Left ventricular outflow tract obstruction is common in complete transposition. As mentioned, it may be due to posterior malalignment of the outlet septum. Obstruction due to fibrous ridge, fibromuscular tunnel, or accessory mitral valve tissue can exist with an intact ventricular

Ventricular outflow tract anomalies

septum as well as in association with a ventricular septal defect. With a significant left ventricular outflow tract obstruction, the main pulmonary artery is smaller than the ascending aorta (Figure 21.19). The pulmonary valve may be stenotic. In contrast, right ventricular outflow tract obstruction is less common. It may be due to anterior malalignment of the outlet septum as described and is often dynamic (Figure 21.20). When there is subaortic stenosis, an obstructive lesion of the aortic arch may be present. Hypoplasia of the morphologically right ventricle is also seen in this setting. Donofrio reported an interesting fetal case of complete transposition of the great arteries which developed premature closure of the foramen ovale and ductus arteriosus.38 Although this case might be an anecdotal case rather than a common occurrence, it provides an insight into understanding the impact of a complete transposition circuit on the fetal circulation that Rudolph has recently postulated.39 In the fetus with complete transposition, well-oxygenated flow from the placenta is directed to the pulmonary artery rather than the ascending aorta. Exposure of the ductus arteriosus and pulmonary arterial bed to high oxygen content will cause constriction of the ductus arteriosus and diminished pulmonary vascular resistance. Increased pulmonary venous return to the left atrium tends to close the foramen ovale. Severe ductal constriction could result in pulmonary arterial hypertension with increased pulmonary arteriolar smooth muscle development. As this combination may result in fetal demise, immediate delivery of the baby and initiation of ECMO (extracorporeal membrane oxygenation) is suggested.38

that can be identified by the presence of a moderator band and more apical attachment of its atrioventricular valve to the ventricular septum. Occasionally, the ventricles may be related in a superior–inferior fashion. The aorta is usually supported by a completely muscular infundibulum, while the pulmonary artery is not (Figure 21.22c and 21.22d). The pulmonary valve is in fibrous continuity with the mitral valve in classic cases. In the majority of cases, the aorta is located anterior to and to the left of the pulmonary artery, which can be easily recognized on the three-vessel view (Figure 21.22b).12–15 The ventricular outflow tracts are usually parallel, which allows visualization of both outflow tracts in a single imaging plane (Figure 21.22d). In contrast to complete transposition, the majority of cases of corrected transposition are associated with additional cardiac defects, about 10% occurring as an isolated anomaly.40–42 Approximately 70% of cases are associated with a ventricular septal defect. Although any type of defect can be associated, a perimembranous defect is the most common type. It tends to extend toward the ventricular inlet. Left ventricular outflow tract obstruction is also common, occurring in 40–50% of cases. The nature of the obstruction may be a fibrous ridge, fibromuscular tunnel, aneurysm of the membranous septum, or accessory mitral valve tissue. The pulmonary valve may also be stenotic. It may infrequently be complicated by pulmonary

Congenitally corrected transposition of the great arteries Congenitally corrected transposition is a combination of discordant atrioventricular connection and discordant ventriculoarterial connection (Figure 21.21). Because of discordant connections at two levels, the physiologic abnormality intrinsic to each discordant connection is ‘congenitally corrected’. In most cases, the situs is solitus. Although it occurs most commonly with levocardia, dextrocardia and mesocardia are not uncommon.40–43 The majority of cases with an unexpected cardiac position for the given visceral situs, i.e. mesocardia or dextrocardia in situs solitus and levocardia or mesocardia in situs inversus, will have corrected transposition. Discordant atrioventricular connection can be readily recognized on the four-chamber view (Figure 21.22a). When it occurs in situs solitus, the left-sided ventricle is the morphologically right ventricle

319

Figure 21.21 Diagram showing pathology of congenitally corrected transposition of the great arteries.

320

Fetal Cardiology

(a)

(b)

(c)

(d)

Figure 21.22 Congenitally corrected transposition of the great arteries. (a) Four-chamber view shows normal cardiac position. The left-sided ventricle contains the moderator band (m), which is characteristic of the morphologically right ventricle. Therefore, there is discordant atrioventricular connection. (b) Three-vessel view shows that the aorta is located anterior and slightly left to the pulmonary artery. (c, d) Ventricular outflow tract views show that the right and left ventricular outflow tracts are parallel. The right ventricle connects to the aorta and the left ventricle connects to the pulmonary artery. There is a large ventricular septal defect (D).

atresia. Right ventricular outflow tract obstruction is less common. The left-sided tricuspid valve is abnormal in approximately 30–45% of cases. The commonest pathology is valve dysplasia with variable degree of regurgitation, while overt Ebstein’s malformation is not uncommon.

Association of corrected transposition with Ebstein’s anomaly of the tricuspid valve is often accompanied by an obstructive lesion of the aortic arch.44 Straddling of either atrioventricular valve with hypoplasia of one ventricle is not uncommon.

Ventricular outflow tract anomalies

Bradycardia with complete atrioventricular block has been reported to occur in up to 20% of fetuses with congenitally corrected transposition in the third trimester,40–42 although we have rarely seen this complication in our own cases. After birth, the incidence of arrhythmia increases with age.

Double-outlet ventricle Double-outlet right or left ventricle is a type of ventriculoarterial connection in which both great arteries arise from the morphologically right or left ventricle (Figure 21.23).45–48 In defining the ventriculoarterial connections, a great artery is considered to be connected to a ventricle when more than half of its valve is committed to that ventricle. Double-outlet right ventricle can occur with any atrial situs and any atrioventricular connection. It often occurs in heterotaxy syndrome. However, it most commonly occurs with situs solitus and concordant atrioventricular connection. In almost all cases, a large ventricular septal

321

defect is present. The hemodynamic physiology of the double-outlet right ventricle after birth is determined mainly by the location of the ventricular septal defect in relation to the great arterial valves and the concomitant presence of outflow tract obstruction. The ventricular septal defect may be subaortic, subpulmonary, doubly committed, or non-committed in location (Figure 21.23). When there is a subaortic defect, the physiology is that of an isolated ventricular septal defect. When there is a subpulmonary defect, the physiology is that of complete transposition with a ventricular septal defect. A doubly committed ventricular septal defect opens beneath both arterial valves. As the outlet septum is deficient or hypoplastic with this defect, a common right ventricular outflow tract leads to the great arteries. A non-committed ventricular septal defect is remote from both arterial valves. It involves either inlet or the trabecular part of the ventricular septum. An atrioventricular septal defect is another form of non-committed defect. When the defect is doubly committed or non-committed, the physiology depends on the intracardiac streaming of blood flow.

Figure 21.23 Diagrams showing pathology of double-outlet right ventricle. The ventricular septal defect (VSD) is classified according to its location relative to the semilunar valves. A, aorta; AL, anterior limb of trabecula septomarginalis; OS, outlet septum; P, pulmonary artery; PL, posterior limb of trabecula septomarginalis; TSM, trabecula septomarginalis; TV, tricuspid valve.

322

Fetal Cardiology

Rarely, the ventricular septal defect may be restrictive. Extremely rarely, double-outlet can be present with an intact ventricular septum. In situs solitus, the aorta is usually located to the right of the pulmonary artery. They tend to have a side-by-side relationship. However, any great arterial relationship can be found. Both arterial trunks tend to have a parallel orientation. In many cases, both great arterial valves are supported by a completely muscular infundibulum. Some consider the presence of bilateral infundibulum a hallmark of double-outlet right ventricle. Most authorities, however, have abandoned this definition.45 There has been some debate on the relationship between tetralogy of Fallot and double-outlet right ventricle. In this context, one should be reminded that the term ‘tetralogy’ describes the infundibular morphology, while the term ‘double-outlet right ventricle’ describes the type of ventriculoarterial connection. Therefore, the type of ventriculoarterial connection in a case of tetralogy with more than 50% of aortic overriding is unequivocally a double-outlet right ventricle. Double-outlet right ventricle with a subaortic ventricular septal defect (Figure 21.24) is often associated with subpulmonary obstruction. Many with this combination show an infundibular morphology that is similar or identical to that of tetralogy. In contrast, double-outlet right ventricle with a subpulmonary ventricular septal defect (Figure 21.20) is often associated with subaortic obstruction.45 When there is subaortic obstruction, an obstructive lesion of the aortic arch is common. The obstruction in either circumstance is usually due to deviated outlet septum, but there may be associated arterial valvar stenosis. Subpulmonary or subaortic obstruction can be suspected when one great artery is significantly smaller than the other at the three-vessel view. Double-outlet left ventricle is extremely rare.46 The ventricular septal defect is most commonly subaortic, less frequently subpulmonary, and least commonly doubly committed or non-committed. It was once considered that bilaterally deficient infundibulum is the hallmark of double-outlet left ventricle. A subsequent study, however, verified that any infundibular morphology is found in double-outlet left ventricle and that a subpulmonary or subaortic infundibulum is more common than bilaterally deficient infundibulum.

Truncus arteriosus Truncus arteriosus is a condition in which one arterial trunk arises from the base of the ventricles via a single arterial valve to give rise directly to the systemic, coronary, and pulmonary arteries (Figure 21.25).49 In almost all cases, there is a large ventricular septal defect immediately underneath the common arterial valve, which usually overrides the ventricular septum.49–51

(a)

(b)

Figure 21.24 Double-outlet right ventricle with subaortic ventricular septal defect. (a) Oblique coronal view of the outlet of the right ventricle shows that the main pulmonary artery arises from the right ventricle. The ascending aorta (A) is seen on the right side of the main pulmonary artery. (b) Left ventricular outflow tract view shows that the aorta arises completely from the right ventricle. The ventricular septal defect (D) is below the aorta. The only outlet of the left ventricle is the ventricular septal defect.

Ventricular outflow tract anomalies

Figure 21.25 Diagram showing pathology of truncus arteriosus. TR, truncus.

Occasionally, the common arterial trunk arises exclusively from the right or left ventricle. The ventricular septal defect can be best demonstrated in the left ventricular outflow tract view as in tetralogy. The truncal valve is almost always in fibrous continuity with the mitral valve. The truncal valve consists of two to five cusps and is often regurgitant and less commonly stenotic. As there is a single arterial trunk, only two vessels are seen at the three-vessel view (Figure 21.26a).12–15 Two vessels at three-vessel view can also be seen in pulmonary atresia with ventricular septal defect and absent or hypoplastic main pulmonary arterial trunk, and in aortic atresia with hypoplastic ascending aorta. The pulmonary arteries arise from the common arterial trunk with or without a short segment of main pulmonary artery (Figure 21.26b). As in tetralogy of Fallot, leftward deviation of the cardiac axis is common in truncus.21,22 Otherwise, the four-chamber view usually does not show any defect. The ductus arteriosus is absent in approximately half of cases.52,53 There is an inverse relationship between development of the aortic arch and ductus arteriosus. A right aortic arch is also common, occurring in approximately one-third of cases. Truncus arteriosus may be associated with interruption of the aortic arch. Unilateral absence of one pulmonary artery is also common.

Association with chromosomal and extracardiac anomalies A variety of karyotypic anomalies have been reported to occur with conotruncal anomalies, although they occur

323

(a)

(b)

Figure 21.26 Truncus arteriosus. (a) Three-vessel view shows only two vessels, the truncus arteriosus (Tr) and the superior vena cava (V). (b) Oblique coronal view shows that a single arterial trunk arises from the heart. The truncal valve (asterisks) is half committed to the left ventricle. The truncus bifurcates into the ascending aorta and pulmonary artery that gives rise to the left pulmonary artery in this view.

most commonly with microdeletion 22q11 and trisomies 21, 18, and 13.5,7,54 Chromosomal anomalies are detected more frequently in fetuses than in infants with a conotruncal anomaly (Table 21.2).5,7,24,55 They are common in

324

Fetal Cardiology

Table 21.2

Association of conotruncal anomalies with chromosomal anomalies

Diagnosis

Infant series55 (Perry et al)

Fetal series A5 (Allan et al)

Fetal series B7 (Paladini et al)

Fetal series A and B, compiled

Tetralogy of Fallot

33/297 (11.1%)

11/46 (24%)

5/25 (20%)

16/71 (23%)

Complete transposition

1/208 (0.5%)

0/20 (0%)

0/18 (0%)

0/38 (0%)

Corrected transposition

No data

0/3 (0%)

0/5 (0%)

0/8 (0%)

Double-outlet right ventricle

10/86 (11.6%)

4/33 (12%)

5/13 (38%)

9/46 (20%)

Truncus arteriosus

2/51 (3.9%)

2/14 (14%)

1/ 6 (17%)

3/20 (15%)

Table 21.3

Association of conotruncal anomalies and extracardiac anomalies

Diagnosis

Infant series2 (NERICP)

Infant series4 (NGP)

Fetal series7 (Paladini)

31%

28%

48%

Complete transposition

9%

9%

33%

Corrected transposition

6%

9%

0%

Double-outlet right ventricle

20%

16%

46%

Truncus arteriosus

48%

21%

17%

Tetralogy of Fallot

fetuses with tetralogy of Fallot, double-outlet right ventricle, or truncus arteriosus, the incidence ranging from 15 to 30% in fetal series. They are rarely, if ever, found in those with complete or corrected transposition. Deletion in chromosomal region 22q11 is the most common chromosomal abnormality.56–58 A large fetal series showed 20% incidence of chromosome 22q11 deletion in fetuses with ventricular outflow tract abnormalities or interruption of the aortic arch.58 The incidence is higher when tetralogy is associated with pulmonary atresia, an aortic arch anomaly, or absent pulmonary valve, or when truncus arteriosus is associated with aortic arch anomalies including interrupted aortic arch.34,58 This specific chromosomal abnormality usually occurs in a syndromic pattern, such as DiGeorge syndrome, velocardiofacial (or Shprintzen) syndrome, conotruncal face syndrome, and Cayler cardiofacial syndrome. As the patients share clinical and laboratory features, an acronym, CATCH-22 (cardiac defect, abnormal facies, thymic hypoplasia, cleft palate, hypocalcemia, and deletion in chromosome 22), has been used.56 Thymic hypoplasia can be appreciated at fetal ultrasound.58–60 The transverse diameter of the normal thymus in millimeters is slightly smaller than the gestational age in weeks in the second trimester, and becomes slightly larger as the pregnancy approaches term.61 The combination of associated aortic arch anomalies, thymic hypoplasia or aplasia, and intrauterine growth restriction is highly suggestive of chromosome 22q11 deletion.58 Chromosomal analysis with FISH (fluorescent in situ

hybridization) for deletion 22q11 is recommended in every fetus with a conotruncal anomaly. Although it occurs de novo in 90% of cases, parental screening is indicated in familial cases as there is a 50% recurrence risk.57 Extracardiac anomalies are common in infants and fetuses with tetralogy of Fallot, double-outlet right ventricle, and truncus arteriosus.6,7 In fact approximately 20% of cases of fetal diagnosis of tetralogy are referred for fetal echocardiography because of abnormal extracardiac findings.24 They are rare in those with transposition. They are detected more frequently in fetuses than in infants with a conotruncal anomaly (Table 21.3).

Outcome and postnatal course The outcome of fetuses with CHD may be affected by a multitude of parameters, such as the severity of the cardiac and concomitant extracardiac abnormalities, gestational age at delivery, perinatal care, and the suitability for corrective surgery, as well as the institutional experience in neonatal and cardiac critical care. There is a wide spectrum of pathology associated with conotruncal lesions that affect the morbidity and mortality of individual patients. Although antenatal diagnosis often represents a more severe spectrum of cardiovascular disease, fetal diagnosis may permit an individualized choice from the available options, which include tailored neonatal

Ventricular outflow tract anomalies

325

Table 21.4 Outcome of cases with fetal diagnosis of conotruncal anomalies. Data compiled from references 5 and 7 Diagnosis

Termination of pregnancy

Intrauterine death

Neonatal death

Survivors of continuing pregnancy

Tetralogy of Fallot (n = 71)

41%

7%

23%

36%

Complete transposition (n = 38)

34%

8%

18%

60%

Corrected transposition (n = 8)

38%

0%

38%

40%

Double-outlet right ventricle (n = 46)

70%

11%

4%

50%

Truncus arteriosus (n = 20)

50%

0%

30%

10%

Total (n = 183)

48%

7%

19%

42%

treatment and termination of pregnancy. When the diagnosis of a conotruncal anomaly is made during the first and second trimesters, approximately 30–50% of parents will choose termination of pregnancy (Table 21.4).5,7 The main reason for pregnancy termination as well as in utero and neonatal demise in most conotruncal anomalies is the frequent association with chromosomal and extracardiac anomalies. Thus, prenatal screening for non-cardiac and chromosomal anomalies is strongly recommended for those fetuses with tetralogy of Fallot, double-outlet right ventricle, truncus arteriosus, and aortic arch interruption. Fetal diagnosis allows redirection of care to tertiary centers, planning of the time and mode of delivery, and initiation of appropriate perinatal treatment, whenever indicated. Unfortunately, a significant number of cardiac lesions that require early postnatal intervention such as complete transposition of the great arteries, obstructive great arterial lesions, and interruption of the aortic arch escape prenatal diagnosis and thus may not profit from optimized periand postnatal care. There is some evidence of increased morbidity and mortality associated with the delayed postnatal diagnosis of specific critical lesions that require urgent medical and surgical intervention early after delivery. For example, complete transposition of the great arteries is a life-threatening malformation in neonates, but it is amenable to complete repair, with excellent long-term results. Newborns with isolated complete transposition typically develop severe cyanosis, metabolic acidosis, and eventually multiorgan failure unless the ductus arteriosus is maintained patent by the intravenous administration of prostaglandin and the atrial communication is enlarged by balloon atrial septostomy in the first day of life. Delay in diagnosis and treatment results in significant morbidity and mortality. In a study by Bonnet et al,36 12% of newborns referred to their institution died either before or early after surgery if the diagnosis of complete transposition was made only after birth. By contrast, there was no perioperative mortality in those cases with prenatal diagnosis, transfer to a tertiary center for delivery,

and immediate neonatal management. Moreover, the prenatally diagnosed cases benefited from a shorter time to corrective surgery with an arterial switch procedure and discharge from the hospital. Maintenance of arterial duct patency is also critical for those conotruncal anomalies that are associated with severe pulmonary or aortic stenosis or atresia. Closure of the arterial duct would inevitably result in low cardiac output, acidosis, hypoxia, and ultimately death, and thus the postnatal cardiovascular circulation is ‘ductus-dependent’.25,27 Intravenous prostaglandin infusion allows arterial duct patency for days and, if required, even weeks. Rarely ductal patency cannot be maintained, and an urgent Blalock–Taussig shunt operation or catheter intervention is required. The prenatal recognition of absent ductus arteriosus in truncus arteriosus, absent pulmonary valve syndrome, and pulmonary atresia with major aortopulmonary collateral arteries is also important, because these lesions will not respond to prostaglandin. Corrective surgery of the cardiovascular lesion is possible for the majority of conotruncal anomalies with good immediate and long-term outcomes.62 The exception is the combination of truncus arteriosus with aortic arch obstruction, which carries a high early mortality.63 Most survivors with double-outlet right ventricle, truncus arteriosus, and severe tetralogy of Fallot require repeated interventions later in life, e.g. replacement of the conduits, implantation of the valves, dilatation of the narrowed vessels and grafts, etc. Last but not least, there is a significantly increased recurrence risk of CHD for future pregnancies of the parents and the affected child, and timely fetal echocardiography is advocated.

Acknowledgments We thank Ms Jennifer Russell for reference management and Mrs Eul Kyung Kim for making diagrams.

326

Fetal Cardiology

References 1. Restivo A, Piacentini G, Placidi S, Saffirio C, Marion B. Cardiac outflow tract: a review of some embryogenetic aspects of the conotruncal region of the heart. Anat Rec A Discov Mol Cell Evol Biol 2006; 288: 936–43. 2. Fyler DC, Buckley LP, Hellenbrand WE, Cohn HE. Report of the New England Infant Cardiac Program. Pediatrics 1980; 65: 375–461. 3. Scott DJ, Rigby ML, Miller GAH, Shinebourne EA. The presentation of symptomatic heart disease in infancy based on 10 years’ experience (1973–1982): implications for the provision of services. Br Heart J 1984; 52: 248–57. 4. Moller JH, Moodie DS, Blees M, Norton JB, Nouri S. Symptomatic heart disease in infants: comparison of three studies performed during 1969–1987. Pediatr Cardiol 1995; 16: 216–22. 5. Allan LD, Sharland GK, Milburn A et al. Prospective diagnosis of 1,006 consecutive cases of congenital heart disease in the fetus. J Am Coll Cardiol 1994; 23: 1452–8. 6. Tegnander E, Williams W, Johansen OJ, Blass HGK, Eik-Nes SH. Prenatal detection of heart defects in a nonselected population of 30149 fetuses – detection rates and outcome. Ultrasound Obstet Gynecol 2006; 27: 252–65. 7. Paladini D, Rustico M, Todros T et al. Conotruncal anomalies in prenatal life. Ultrasound Obstet Gynecol 1996; 8: 241–6. 8. Tometzki AP, Suda K, Kohl T, Kovalchin JP, Silverman NH. Accuracy of prenatal echocardiographic diagnosis and prognosis of fetuses with conotruncal anomalies. J Am Coll Cardiol 1999; 33: 1696–701. 9. Sivanandam S, Gilckstein JS, Printz BF et al. Prenatal diagnosis of conotruncal malformations: diagnostic accuracy, outcome, chromosomal abnormalities, and extracardiac anomalies. Am J Perinatol 2006; 23: 241–5. 10. Anderson RH, Becker AE, Van Mierop LHS. What should we call the ‘crista’? Br Heart J 1977; 39: 856–9. 11. Achiron R, Glaser J, Gelernter I, Hegesh J, Yagel S. Extended fetal echocardiographic examination for detecting cardiac malformations in low risk pregnancies. Br Med J 1992; 304: 671–4. 12. Yoo S-J, Lee Y-H, Kim ES et al. Three-vessel view of the fetal upper mediastinum: an easy means of detecting abnormalities of the ventricular outflow tracts and great arteries during obstetric screening. Ultrasound Obstet Gynecol 1997; 9: 173–82. 13. Yoo S-J, Lee Y-H, Cho KS. Abnormal three-vessel view on sonography: a clue to the diagnosis of congenital heart disease in the fetus. AJR Am J Roentgenol 1999; 172: 825–30. 14. Yoo S-J, Lee YH, Cho KS, Kim DY. Sequential segmental approach to fetal congenital heart disease. Cardiol Young 1999; 9: 440–54. 15. Vinals F, Heredia F, Giuliano A. The role of the three vessels and trachea view (3VT) in the diagnosis of congenital heart defects. Ultrasound Obstet Gynecol 2003; 22: 358–67. 16. International Society of Ultrasound in Obstetrics and Gynecology. Cardiac screening examination of the fetus: guidelines for performing the ‘basic’ and ‘extended basic’ cardiac scan. Ultrasound Obstet Gynecol 2006; 27: 107–13.

17. DeVore GR, Polanco B, Sklansky MS, Platt LD. The ‘spin’ technique: a new method for examination of the fetal outflow tracts using three-dimensional ultrasound. Ultrasound Obstet Gynecol 2004; 24: 72–82. 18. Freedom RM, Mawson JB, Yoo S-J, Benson LN. Tetralogy of Fallot and pulmonary atresia and ventricular septal defect. In: Congenital Heart Disease: Textbook of Angiocardiography. Armonk, NY: Futura, 1997: 493–33. 19. Shinebourne EA, Babu-Narayan SV, Carvalho JS. Tetralogy of Fallot: from fetus to adult. Heart 2006; 92: 1353–9. 20. Yoo S-J, Lee Y-H, Kim ES et al. Tetralogy of Fallot in the fetus: findings at targeted sonography. Ultrasound Obstet Gynecol 1999; 14: 29–37. 21. Shipp TD, Bromley B, Hornberger LS, Nadel A, Benacerraf BR. Levorotation of the fetal cardiac axis: a clue for the presence of congenital heart disease. Obstet Gynecol 1995; 85: 97–102. 22. Smith RS, Comstock CH, Kirk JS, Lee W. Ultrasonographic left cardiac axis deviation: a marker for fetal anomalies. Obstet Gynecol 1995; 85: 187–91. 23. Pepas LP, Savis A, Jones A et al. An echocardiographic study of tetralogy of Fallot in the fetus and infant. Cardiol Young 2003; 12: 240–7. 24. Poon LCY, Huggon IC, Zidere V, Allan LD. Tetralogy of Fallot in the fetus in the current era. Ultrasound Obstet Gynecol 2007; 29: 625–7. 25. Hornberger LK, Sanders SP, Sahn DJ et al. In utero pulmonary artery and aortic growth and potential for progression of pulmonary outflow tract obstruction in tetralogy of Fallot. J Am Coll Cardiol 1995; 25: 739–45. 26. Yoo SJ, Min JY, Lee YH et al. Fetal sonographic diagnosis of aortic arch anomalies. Ultrasound Obstet Gynecol 2003; 22: 535–46. 27. Mielke G, Steil H, Kendziorra H, Goelz R. Ductus arteriosus-dependent pulmonary circulation secondary to cardiac malformations in fetal life. Ultrasound Obstet Gynecol 1997; 9: 25–9. 28. Moon-Grady AJ, Tacy TA, Brook MM, Hanley FL, Silverman NH. Value of clinical and echocardiographic features in predicting outcome in the fetus, infant, and child with tetralogy of Fallot with absent pulmonary valve complex. Am J Cardiol 2002; 89: 1280–5. 29. Razavi RS, Sharland GK, Simpson JM. Prenatal diagnosis by echocardiogram and outcome of absent pulmonary valve syndrome. Am J Cardiol 2003; 91: 429–32. 30. Volpe P, Paladini D, Marasini M et al. Characteristics, associations and outcome of absent pulmonary valve syndrome in the fetus. Ultrasound Obstet Gynecol 2004; 24: 623–8. 31. Galindo A, Gutierrez-Larraya F, Martinez JM et al. Prenatal diagnosis and outcome for fetuses with congenital absence of the pulmonary valve. Ultrasound Obstet Gynecol 2006; 28: 32–9. 32. Ettedgui JA, Sharland GK, Chita SK et al. Absent pulmonary valve syndrome with ventricular septal defect: role of the arterial duct. Am J Cardiol 1990; 66: 233–4. 33. Callan NA, Kan JS. Prenatal diagnosis of tetralogy of Fallot with absent pulmonary valve. Am J Perinatol 1991; 8: 15–17. 34. Vesel S, Rollings S, Jones A et al. Prenatally diagnosed pulmonary atresia with ventricular septal defect: echocardiography,

Ventricular outflow tract anomalies

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

genetics, associated anomalies and outcome. Heart 2006; 92: 1501–5. Freedom RM, Mawson JB, Yoo S-J, Benson LN. Complete transposition of the great arteries (atrioventricular concordance and ventriculoarterial discordance). In: Congenital Heart Disease: Textbook of Angiocardiography. Armonk, NY: Futura, 1997: 987–1070. Bonnet D, Coltri A, Butera G et al. Detection of transposition of the great arteries in fetuses reduces neonatal morbidity and mortality. Circulation 1999; 99: 916–18. Van Praagh R, Pérez-Treviño C, López-Cuellar M et al. Transposition of the great arteries with posterior aorta, anterior pulmonary artery, subpulmonary conus and fibrous continuity between aortic and atrioventricular valves. Am J Cardiol 1971; 28: 621–31. Donofrio MT. Premature closure of the foramen ovale and ductus arteriosus in a fetus with transposition of the great arteries. Circulation 2002; 105: 65–6. Rudolph AM. Aortopulmonary transposition in the fetus: speculation on pathophysiology and therapy. Pediatr Res 2007; 61: 375–80. Chiappa E, Micheletti A, Sciarrone A et al. The prenatal diagnosis of and short-term outcome for, patients with congenitally corrected transposition. Cardiol Young 2004; 14: 265–76. Sharland G, Tingay R, Jones A, Simpson J. Atriventricular and ventricularterial discordance (congenitally corrected transposition of the great arteries): echocardiographic features, associations, and outcome in 34 fetuses. Heart 2005; 91: 1453–8. Paladini D, Volpe P, Marasini M et al. Diagnosis, characterization and outcome of congenitally corrected transposition of the great arteries in the fetus: a multicenter series of 30 cases. Ultrasound Obstet Gynecol 2006; 27: 281–5. Bernasconi A, Azancot A, Simpson JM et al. Fetal dextrocardia: diagnosis and outcome in two tertiary centres. Heart 2005; 91: 1590–4. Bader R, Perrin D, Yoo SJ. Congenitally corrected transposition of the great arteries with Ebstein malformation and hypoplasia of the aortic arch in a fetus. Fetal Pediatr Pathol 2004; 23: 257–63. Freedom RM, Mawson JB, Yoo S-J, Benson LN. Double outlet right ventricle. In: Congenital Heart Disease: Textbook of Angiocardiography. Armonk, NY: Futura, 1997: 1119–61. Freedom RM, Mawson JB, Yoo S-J, Benson LN. Double outlet left ventricle. In: Congenital Heart Disease: Textbook of Angiocardiography. Armonk, NY: Futura, 1997: 1163–9. Smith RS, Comstock CH, Kirk JS et al. Double-outlet right ventricle: an antenatal diagnostic dilemma. Ultrasound Obstet Gynecol 1999; 14: 315–19. Kim N, Friedberg MK, Silverman NH. Diagnosis and prognosis of fetuses with double outlet right ventricle. Prenat Diagn 2006; 26: 740–5.

327

49. Freedom RM, Mawson JB, Yoo S-J, Benson LN. Truncus arteriosus or common arterial trunk. In: Congenital Heart Disease: Textbook of Angiocardiography. Armonk, NY: Futura, 1997: 219–41. 50. Duke C, Sharland GK, Jones AMR et al. Echocardiographic features and outcome of truncus arteriosus diagnosed during fetal life. Am J Cardiol 2001; 88: 1379–84. 51. Volpe P, Paladini D, Marasini M et al. Common arterial trunk in the fetus: characteristics, associations, and outcome in a multicentre series of 23 cases. Heart 2003; 89: 1437–41. 52. Van Praagh R, Van Praagh S. The anatomy of common aorticopulmonary trunk (truncus arteriosus communis) and its embryologic implications. Am J Cardiol 1965; 16: 406–25. 53. Calder L, Van Praagh R, Van Praagh S et al. Truncus arteriosus communis. Clinical, angiographic, and pathologic findings in 100 patients. Am Heart J 1976; 92: 23–38. 54. Allan LD, Sharland GK, Chita SK, Lockhart SL, Maxwell DJ. Chromosomal anomalies in fetal congenital heart disease. Ultrasound Obstet Gynecol 1991; 1: 8–11. 55. Perry LW, Neill CA, Ferencz C, Rubin JD, Loffredo CA. Infants with congenital heart disease: the cases. In: Ferencz C, Rubin JD, Loffredo CA, Magee CA, eds. Epidemiology of Congenital Heart Disease, the Baltimore–Washington Infant Heart Study 1981–1989. Perspectives in Pediatric Cardiology, Vol 4. Mount Kisco, NY: Futura, 1993: 33–61. 56. Wilson DI, Burn J, Scambler P, Goodship J. DiGeorge syndrome: part of CATCH 22. J Med Genet 1993; 30: 852–6. 57. Goldmuntz E, Clark BJ, Mitchell LE et al. Frequency of 22q11 deletions in patients with conotruncal defects. J Am Coll Cardiol 1998; 32: 492–8. 58. Volpe P, Marasini M, Caruso G et al. 22q11 deletions in fetuses with malformations of the outflow tracts or interruption of the aortic arch: impact of additional ultrasound signs. Prenat Diagn 2003; 23: 752–7. 59. Chaoui R, Kalache KD, Heling KS et al. Absent or hypoplastic thymus on ultrasound: a marker for deletion 22q11.2 in fetal cardiac defects. Ultrasound Obstet Gynecol 2002; 20: 546–52. 60. Barrea C, Yoo SJ, Chitayat D et al. Assessment of the thymus at echocardiograph in fetuses at risk for 22q11.2 deletion. Prenat Diagn 2003; 23: 9–15. 61. Cho JY, Min JY, Yoo SJ et al. Normal thymus diameter at fetal ultrasound. Ultrasound Obstet Gynecol 2007; 29: 634–8. 62. Sivanandam S, Glickstein JS, Printz BF et al. Prenatal diagnosis of conotruncal malformations: diagnostic accuracy, outcome, chromosomal abnormalities, and extracardiac anomalies. Am J Perinatol 2006; 23: 241–5. 63. Konstantinov IE, Karamlou T, Blackstone EH et al. Truncus arteriosus associated with interrupted aortic arch in 50 neonates: a Congenital Heart Surgeons Society study. Ann Thorac Surg 2006; 81: 214–22.

22 Aortic arch anomalies Shi-Joon Yoo, Timothy Bradley, and Edgar Jaeggi The normal left aortic arch courses from the ascending aorta upward, backward, and leftward in front of the trachea. It then crosses the proximal part of the left main bronchus, to the left of the trachea and esophagus, to join the descending aorta. It gives rise to the right innominate, left common carotid, and left subclavian arteries in sequence. The segments of the aortic arch are known as: the proximal transverse arch between the right innominate and left common carotid artery origins; the distal transverse arch between the left common carotid and left subclavian artery origins; and the aortic isthmus between the left subclavian artery origin and the insertion of the ductus arteriosus (Figure 22.1).1 The abnormalities involving the aortic arch can be divided into two distinct categories. The first category includes the obstructive lesions, such as coarctation, tubular hypoplasia, and interruption of a part or parts of the aortic arch, which are dealt with in Chapter 21. The second category includes abnormalities of position and/or branching that are commonly called the aortic arch anomalies. The aortic arch anomalies have three important clinical implications: (1) mechanical compression of the airway and/or esophagus by the vascular structures that form a vascular ring (complete encirclement) or sling (incomplete encirclement); (2) association with congenital heart defects; and (3) association with chromosomal abnormalities. The present chapter discusses the morphogenesis, morphology, fetal sonographic approach, and clinical management of various aortic arch anomalies. Some extremely rare forms of aortic arch anomalies are not discussed.

Hypothetical double aortic arch model Insight into embryogenesis and fetal circulation is tremendously helpful in understanding the prenatal and postnatal features of various malformations involving the aortic arch. It is in contrast to most cardiac defects in which embryology can be regarded as a hindrance rather than a

help in understanding the pathology and pathophysiology.2 The normal and abnormal developments of the aortic arch can be easily understood by reference to the hypothetical double aortic arch model introduced by the pioneering pathologist Dr Jesse E Edward in 1948.3,4 The model illustrates a fairly late stage of development when the aortic sac has already divided into the ascending aorta and pulmonary arterial trunk, and the descending aorta occupies a neutral position (Figure 22.2).5,6 Two aortic arches connect the ascending and descending aorta, forming a complete vascular ring around the trachea and esophagus. Each aortic arch gives rise to the ipsilateral

Figure 22.1 Nomenclature of the segments of the normal left aortic arch. The aortic arch consists of the proximal and distal transverse arch and isthmus. The aortic isthmus can be absent when the ductus arteriosus (DA) has attachment to the aortic arch proximal to the last branch or when the ductus arises from a subclavian artery. LPA, left pulmonary artery; RSA, right subclavian artery; RIA, right innominate artery; RCC, right common carotid; LCC, left common carotid; LSA, left subclavian artery.

330

Fetal Cardiology

Figure 22.2 Hypothetical double aortic arch model of Jesse E Edward (reproduced with permission from reference 6). RPA, right pulmonary artery; RAA, right aortic arch; eso, esophagus; LAA, left aortic arch; A, ascending aorta; p, main pulmonary artery.

common carotid and subclavian arteries. There are right and left ductus arteriosi, each of which connects the branch pulmonary artery to the distal part of the ipsilateral aortic arch. Normally, the left aortic arch and left ductus arteriosus persist, while the right aortic arch distal to the origin of the right subclavian artery and the right ductus arteriosus regress (Figure 22.3a). As a result, the proximal part of the embryological right aortic arch becomes the right innominate artery that bifurcates into the right common carotid and subclavian arteries. Most of the malformations showing abnormal position and/or branching of the aortic arch are assumed to result from abnormal persistence of a part or parts that should have regressed and/or abnormal regression of a part or parts that should have persisted.3–5

Aortic arch anomalies and morphogenesis The morphological description of the aortic arch anomalies should consist of the following three basic components: • the left- or right-sidedness of the aortic arch relative to the trachea • the presence or absence of an aberrant artery arising from the descending aorta • the left- or right-sidedness of the patent or ligamentous ductus arteriosus. A left aortic arch refers to an aortic arch that courses to the left side of the trachea and arches over the proximal

part of the left main bronchus. It is formed by regression of a segment of the embryological right aortic arch. Regression distal to the origin of the right subclavian artery results in a normal left aortic arch (Figure 22.3). Abnormal regression of the right arch between the origins of the right common carotid and right subclavian arteries leaves the right subclavian artery attached to the distal part of the right aortic arch (Figure 22.4).5–10 As a consequence, the distal part of the embryological right aortic arch and right subclavian artery together constitute the aberrant right subclavian artery of the developed aortic arch. With abnormal regression of the right aortic arch proximal to the origin of the right common carotid artery, the distal part of the right aortic arch will persist as the aberrant right innominate artery that gives rise to the right subclavian and right common carotid arteries. Each of these three forms of left aortic arch may have either left or right ductus arteriosus. Bilateral ductus arteriosi are exceedingly rare. When the left aortic arch has a normal branching pattern of the head and arm arteries, no vascular ring or sling is formed around the trachea and esophagus regardless of the presence of a left or right ductus arteriosus. When there is an aberrant origin of the right subclavian or innominate artery, the aberrant artery has an oblique retroesophageal course. When the left ductus arteriosus persists, a vascular sling that consists of the aortic arch and right subclavian artery is formed around the left side of the trachea and esophagus. When the right ductus is present between the aberrant artery and the right pulmonary artery, a complete vascular ring is formed. The ring consists of the ascending aorta, the left aortic arch, the descending aorta, the aberrant right subclavian or innominate artery, the right ductus arteriosus, the right pulmonary artery, and the pulmonary arterial trunk, with the underlying heart completing the ring. A right aortic arch refers to an aortic arch that courses to the right side of the trachea and arches over the proximal part of the right main bronchus. It is formed by regression of a segment of the embryological left aortic arch. Abnormal regression distal to the origin of the left subclavian artery results in a right aortic arch with a mirror-image of the normal branching pattern (Figure 22.5). Abnormal regression of the left arch between the origins of the left common carotid and left subclavian artery or proximal to the left common carotid artery results in a right aortic arch with aberrant left subclavian or left innominate artery (Figure 22.6).5–11 Each of these forms of right aortic arch may have either left or right ductus arteriosus. Bilateral ductus arteriosi are exceedingly rare. The right aortic arch with a mirror-image of normal branching pattern does not constitute a vascular ring or sling regardless of the presence of a left or right ductus arteriosus, except for a very few reported cases in which there is a left ductus arteriosus between the left pulmonary artery and a right-sided descending aorta.12–14 When there is an aberrant origin of the left subclavian or

Aortic arch anomalies

left innominate artery, the aberrant artery has an oblique retroesophageal course. When the left ductus is present between the aberrant artery and the left pulmonary artery, a complete vascular ring is formed around the trachea and esophagus (Figure 22.7). The ring consists of the ascending aorta, the right aortic arch, the descending aorta, the aberrant left subclavian or innominate artery, the left ductus arteriosus, the left pulmonary artery, and the pulmonary arterial trunk, with the underlying heart completing the ring. When the right ductus arteriosus persists instead, the circle is incomplete and a vascular sling is formed around the right side of the trachea and esophagus. A double aortic arch refers to the presence of two aortic arches, one on each side of the trachea.5–9 It is in contrast to a double-barreled or double-lumen aortic arch, in which two aortic arches are present on the same side of the trachea. In double aortic arch, the hypothetical double aortic arch model persists without regression of any segment of the right and left arches (Figure 22.8). A ductus arteriosus, much more frequently the left than the right duct, persists, although rare cases with bilateral ducts have been reported.15 Each aortic arch gives rise to the ipsilateral common carotid and subclavian arteries. In the majority of cases with double aortic arch, both arches are patent, although one may be significantly smaller. Rarely an atretic segment may exist in either arch.16 When the aortic arch is on the left, the descending aorta stays on the left side in its entire course to enter the abdomen through the aortic hiatus in the diaphragm. When there is a right aortic arch, the descending aorta starts on the right side and makes a gentle curve to the left to enter the abdomen through the normal aortic hiatus in the diaphragm on the left side. As this midline shift of the descending aorta is gradual, the trachea is not crossed by the descending aorta, while the lower part of the esophagus may be mildly compressed. So-called ‘circumflex retroesophageal aortic arch is worth discussing as a separate additional entity. In this rare malformation, the proximal part of the descending aorta is on the contralateral side of the aortic arch (Figure 22.9).17,18 This set requires the aortic arch to make an additional arc to the other side behind the trachea and esophagus to reach the descending aorta. The morphogenetic mechanism for development of this rare malformation has been explained by many, but will not be introduced in this chapter because it is hardly a help to the understanding of this rather complex and variable entity. It occurs much more frequently with a right aortic arch than with a left aortic arch. When it occurs with a right aortic arch, the aortic arch gives rise to the left common carotid, right common carotid, and right subclavian artery from its segment on the right side of the trachea (Figure 22.9). Then the aortic arch makes a sharp turn to the left side to have an oblique leftward and usually downward course to the contralateral descending aorta. The left subclavian artery arises from the top of the

331

Table 22.1 Aortic arch anomalies with or without a vascular ring or sling With a vascular ring: 1. Double aortic arch 2. Right aortic arch with aberrant left subclavian or innominate artery and left ductus arteriosus 3. Right aortic arch with mirror-image branching and left ductus arteriosus between the left pulmonary artery and a right-sided descending aorta 4. Left aortic arch with aberrant right subclavian or innominate artery and right ductus arteriosus 5. Circumflex retroesophageal aortic arch With a vascular sling: 1. Left aortic arch with aberrant right subclavian or innominate artery and left ductus arteriosus 2. Right aortic arch with aberrant left subclavian or innominate artery and right ductus arteriosus 3. Circumflex retroesophageal aortic arch Without a ring or sling: 1. Right aortic arch with mirror-image branching and either right or left ductus arteriosus

descending aorta. It can still be named as an aberrant artery in the sense that it is the last instead of the first branch of the right aortic arch, but it does not have the retroesophageal component that the other forms of aberrant arteries have. Circumflex retroesophageal aortic arch is uncommonly seen without aberrant origin of a subclavian artery.18 The aortic arch may have a very high position in the upper mediastinum, for which the term ‘cervical aortic arch’ has been entertained. Cervical aortic arch is defined as the aortic arch reaching above the level of the clavicle. It is approximately equally distributed on the right and left.5,19,20 It is commonly associated with an abnormal branching pattern of the head and arm vessels. In addition, unusual tortuosity, obstruction, and aneurysm of the aortic arch are found in the majority of cases. The aortic arch anomalies thus discussed can be categorized into three groups according to the presence or absence of a vascular ring or sling: (1) those with a vascular ring or rings, (2) those with a vascular sling, and (3) those without a ring or sling (Table 22.1).

Fetal sonographic approach to aortic arch anomalies Normal aortic arch The aortic arch and ductus arteriosus can be evaluated by using orthogonal transverse views and oblique sagittal

332

Fetal Cardiology

(a)

(b)

Figure 22.3 Normal left aortic arch with left ductus arteriosus. (a) Hypothetical models. In the left-hand diagram the red bars indicate the segments that regress. In the fetal circulation, the aortic arch (red arrow) and the left ductus arteriosus (blue arrow) make a ‘V’-shaped confluence at the descending aorta. In the postnatal circulation, the left ductus arteriosus closes and persists only as the ligamentum ductus arteriosum (reproduced with permission from reference 6). (b) Fetal echocardiograms. In the axial view (left panel), the aortic and ductal arches make a ‘V’-shaped confluence at the descending aorta on the left side of the carina. In a slightly higher axial view (middle panel), the sausage-shaped aortic arch is seen on the left side of the trachea. In the coronal view (right-hand panel), the cross-section of the aortic arch is seen on the left side of the distal trachea. The ductal arch is seen inferior and lateral to the aortic arch.

views along the aortic and ductal arches (Figure 22.3b).6,21–27 Two- and three-dimensional color and power Doppler interrogation facilitates the examination of the mediastinal vascular structures.24,28–30 The assessment of the aortic arch can be started from a three-vessel view. In this plane, the ascending aorta is located slightly to the right of the midline and the descending aorta at the left anterior aspect of the spine. In a slightly cephalad plane, the ascending and descending aortae join together through a sausageshaped aortic arch on the left side of the trachea that is filled with fluid. In the same or slightly caudal imaging plane, the ductus arteriosus connects the main pulmonary

artery to the descending aorta further laterally on the left side. The aortic arch and ductus arteriosus together make a ‘V’-shaped confluence at the descending aorta. The confluence may be seen as a ‘Y’-shaped structure in a slightly tilted position. In these views, the ductus arteriosus has a uniform diameter, while the aortic arch becomes narrower distally as it gives off the head and arm branches. The aortic isthmus is the narrowest part of the aortic arch. It is worth emphasizing that no major vascular structure crosses the midline behind the trachea. Any vessel seen behind the trachea may safely be considered as an aberrant branch of the aortic arch or the aortic arch itself that has

Aortic arch anomalies

333

(c)

Figure 22.3 Continued (c) Composite diagram showing how the aortic and ductal arch views are obtained in a fetus in supine cephalic presentation. The scan starts from any transducer position for a three-vessel view (panel a). Step I: The transducer is moved around the fetal chest until the ascending aorta and descending aorta are aligned vertically in the three-vessel view (panel b). Step II: The transducer is rotated 90° either clockwise or counterclockwise until the aortic arch is seen as a candy-cane-like structure (panel c). Step III: The transducer is moved back to a three-vessel view and moved around the fetal chest until the main pulmonary artery and the descending aorta are vertically aligned (panel a). Step IV: The transducer is rotated 90° either clockwise or counterclockwise until the ductal arch is seen as a hockey stick-like structure (panel d) (modified with permission from reference 31). aa, aortic arch; SVC, superior vena cava; RV, right ventricle; LA, left atrium; RA, right atrium.

an abnormal retroesophageal course or an aberrant left pulmonary artery in pulmonary artery sling that has an abnormal course between the trachea and esophagus. The relationship between the aortic arch and the trachea can also be appreciated in a slanted coronal plane along the trachea.6,24 In this plane, the cross-section of the aortic arch is seen above the proximal part of the left main bronchus, and the cross-section of the ductus arteriosus or an oblique cut of the left pulmonary artery is seen above the distal part of the left main bronchus (Figure 22.3b, right-hand panel). The aortic and ductal arches seen in oblique sagittal views are compared to a candy-cane and a hockey stick, respectively. Proper imaging planes for these views can be chosen from a three-vessel view (Figure 22.3c).31 For a candy-cane view of the aortic arch, the transducer should be aligned with the ascending

aorta and descending aorta in a three-vessel view, from which position the transducer is rotated 90° either clockwise or counterclockwise. The aortic arch typically arises from the space between the cranial parts of the right and left atria deep in the center of the mediastinum. It gives rise to the head and arm arteries. For a hockey stick view of the ductal arch, the transducer should be aligned with the main pulmonary artery and descending aorta in a three-vessel view, from which the transducer is rotated 90°. As the ductus arteriosus is an extension of the main pulmonary artery, the ductal arch appears to arise from the anterior mediastinum immediately behind the anterior chest wall. It should be emphasized that the aortic arch has a hockey stick appearance and the ductal arch a candy-cane appearance when there is transposition of the great arteries, because the ascending aorta

334

Fetal Cardiology

Figure 22.4 Developmental models of left aortic arch with aberrant right subclavian artery. In the hypothetical model, the red bars indicate the segments that regress. In this setting, the left ductus persists in most cases and a vascular sling is formed. Uncommonly, the right ductus persists and a vascular ring is formed around the trachea and esophagus.

is located more anterior to the main pulmonary artery in most cases. Therefore, identification of the aortic and ductal arches should not rely solely on the shape of the arches.

Vascular rings The best approach to vascular rings is to make a slow sweep through the fetal upper mediastinum in transverse planes.6,21–27 Some forms of vascular ring can be shown easily, while others require mental or physical three-dimensional reconstruction. Among the five forms of vascular rings listed in Table 22.1, the double aortic arch is the only form of vascular ring that consists solely of vessels. In other forms, the vascular ring is completed by the heart. Therefore, a classic form of double aortic arch can readily be identified as a ring at echocardiography (Figure 22.8), while other forms of vascular ring can be seen only as a loop with an open end anteriorly (Figure 22.7). One should also be aware that there are many variations in morphology and size in each form of vascular ring. In double aortic arch, one arch is larger and higher than the other in approximately 90% of cases. In the majority of cases having situs solitus, the right aortic arch is the larger arch, and there is a left ductus arteriosus and a left-sided descending aorta. The vascular ring and ductus can be imaged in a single plane, giving the appearance of a figure ‘6’ or ‘9’.6,25 Uncommonly one of the two arches is atretic. This condition is almost impossible to differentiate from a unilateral arch with an abnormal branching

pattern because the atretic segment of the vessel cannot be identified. Circumflex retroesophageal aortic arch is rare.6,17,18 In most cases, it occurs with a right aortic arch and a left ductus arteriosus, and a complete vascular ring is formed (Figure 22.9). After taking an oblique backward course on the right side of the trachea, the arch makes a rather sharp leftward and downward turn to connect to the descending aorta on the left. It is indistinguishable from a double aortic arch with an atretic segment. Commonly, the aortic arch extends to the level of the thoracic inlet, forming a cervical arch. The right aortic arch with an aberrant left subclavian or innominate artery and a left ductus arteriosus is not uncommon. In this combination, a ‘U’-shaped vascular loop is seen on three-vessel and adjacent views.6,23–27 The open end of the ‘U’-loop faces forward, and the apex of the loop is located behind the fluid-filled trachea (Figure 22.7b). The limbs of the ‘U’-loop are the aortic and ductal arches that connect to the heart through the ascending aorta and main pulmonary arterial trunk, respectively. When there is no ventricular outflow tract obstruction, the limbs are of a similar size. Additional telltale signs of a right aortic arch include a gap between the pulmonary arterial trunk and ascending aorta at three-vessel view and a right-sided or midline descending aorta on the three-vessel and four-chamber views. The left aortic arch with an aberrant right subclavian or innominate artery and a right ductus forming a vascular ring is very rare. In this setting, the limbs of the vascular loop cross one another as they arise from the heart, forming

Aortic arch anomalies

335

(a)

(b)

Figure 22.5 Right aortic arch with mirror-image branching of the heart and arm arteries. (a) In the hypothetical models, the red bars indicate the segments that regress. In this setting, the persisting ductus arteriosus is usually the left ductus as shown in the right-hand diagram. In either form, no vascular ring or sling is formed around the trachea and esophagus. (b) Fetal echocardiograms. In the axial view (left-hand panel), the aortic arch is on the right side of the trachea. In the slightly lower axial view underneath the aortic arch (middle panel) and in the sagittal view (right-hand panel), the patent ductus arteriosus connects the undersurface of the right aortic arch and the right pulmonary artery. This fetus had tetralogy of Fallot. This combination is uncommon.

a ‘γ’-shaped loop (Figure 22.4). A very rare form of vascular ring is associated with a right aortic arch with mirrorimage branching of the head and arm arteries in which a left ductus arteriosus connects the left pulmonary artery to a right-sided descending aorta, coursing behind and on the left side of the trachea and esophagus.12–14

Vascular sling Vascular sling represents an incomplete encirclement of the airway and/or esophagus by a single or composite vascular structure. A vascular sling is seen in left or right aortic arch with an aberrant subclavian or innominate

artery and with the ductus arteriosus on the same side of the aortic arch (Figures 22.4 and 22.6). The ascending aorta, aortic arch, and aberrant subclavian or innominate artery together form a vascular sling around the trachea and esophagus, open-ended on the opposite side of the aortic arch. The aberrant subclavian or innominate artery is seen as a rather small vascular structure coursing to the other side of the aortic arch behind the trachea. This combination is much more commonly seen with a left aortic arch than with a right aortic arch. This is in contrast to cases forming a vascular ring, which is much more common with a right aortic arch. Another rare but well-known form of vascular sling is the pulmonary arterial sling in which the left pulmonary

336

Fetal Cardiology

Figure 22.6 Right aortic arch with aberrant left subclavian artery. In the hypothetical models, the red bars indicate the segments that regress. In this setting, the persisting ductus arteriosus is usually the left ductus and a vascular ring is formed around the trachea and esophagus. Less commonly, the right ductus persists and the vascular encirclement of the trachea and esophagus is incomplete and a vascular sling is formed.

artery has a very distal origin from the pulmonary arterial trunk at the right anterior aspect of the trachea and makes a turn to course to the left lung through the space between the airway and esophagus (Figure 22.10).32 The pulmonary arterial sling is commonly associated with an abnormal branching and stenosis of the airway.

Aortic arch anomalies without a vascular ring or sling As is the normal left aortic arch, the right aortic arch with a mirror-image branching pattern is not associated with a vascular ring or sling (Figure 22.5b) except for the exceedingly rare cases as discussed above.5,12–14 Additionally, the ductus arteriosus is not usually mirror-imaged and the left instead of the right ductus arteriosus persists between the innominate artery and the left pulmonary artery in most cases.

Incidence and associated abnormalities The reported incidence of individual aortic arch anomalies varies according to the study population. The left aortic arch with aberrant right subclavian artery was reported as the most common aortic arch anomaly, occurring in 0.5%

of a large autopsy series.4 In prenatal low-risk populations, the right aortic arch with aberrant left subclavian artery as an isolated finding was also reported as the most common lesion, but with an incidence of only 0.1%.22,26 This large discrepancy between these autopsy and prenatal series is considered to be due to difficulty in diagnosis of the left aortic arch with aberrant right subclavian artery at fetal sonography. In the postnatal population with cardiac defects, the right aortic arch with mirror-image branching is the most common.5,12 This is because the right aortic arch with mirror-image branching pattern is associated with congenital heart disease in the majority of cases, while the right or left aortic arch with an aberrant subclavian artery usually occurs as an isolated anomaly. The right aortic arch with mirror-image branching is associated with congenital heart disease in more than 90% of cases.5,12 A right aortic arch is common in tetralogy of Fallot and truncus arteriosus with the incidence ranging 15–35% for both lesions.33 In particular when tetralogy is associated with pulmonary atresia, the incidence is as high as 30–35%. It is less frequently seen in complete transposition of the great arteries, tricuspid atresia, and isolated ventricular septal defect. In contrast to the mirror-imaged right aortic arch, the right aortic arch with aberrant left subclavian artery is associated with congenital heart disease in fewer than 20% of cases.5,33 The most commonly associated congenital heart diseases are ventricular septal defect, atrial septal defect, and complete transposition of the great arteries. The left aortic arch with aberrant right subclavian artery is uncommonly associated with congenital heart disease.

Aortic arch anomalies

337

(a)

(b)

(c)

Figure 22.7 Right aortic arch with aberrant left subclavian artery and left ductus arteriosus. (a) Hypothetical models. In the left-hand diagram, the red bars indicate the segments that regress. In the fetal circulation, the right aortic arch, the distal left aortic arch, the left ductus arteriosus, and the main pulmonary arterial trunk constitute a ‘U’-shaped vascular loop around the trachea and esophagus. As the two limbs of the ‘U’-loop are attached to the heart, a complete vascular ring is formed. In the postnatal circulation, when the ductus arteriosus closes, the proximal part of the aberrant left subclavian artery, which is embryologically the distal part of the left aortic arch, persists as the diverticulum of Kommerell. Note that the blood flow in this particular segment of the left subclavian artery is in an opposite direction in fetal life. (b) Fetal echocardiograms show a ‘U’-shaped vascular loop around the side and posterior aspect of the trachea. Note a rather wide gap between the ascending aorta and main pulmonary arterial trunk. (c) Postnatal computed tomography (CT) angiogram from a different patient shows a right aortic arch with aberrant left subclavian artery that arises from the diverticulum of Kommerell.

Chromosome 22q11 deletion is common in fetuses and patients with aortic arch anomalies.26,34,35 A fetal series showed an 8% incidence of 22q11 deletion in fetuses with a right aortic arch as an isolated abnormality, and 46% in those with right aortic arch and intracardiac abnormality.26 Postnatal series showed higher incidences of 22q11 deletion at 20–25% of cases with an isolated aortic arch anomaly.34,35 The higher incidence of 22q11 deletion in postnatal series is explained by selection bias, as these patients are more likely to be referred to a cardiologist’s

attention.26 More than 50% of patients with an intracardiac anomaly and 22q11 deletion have an aortic arch anomaly.36 In patients with so-called conotruncal malformation, anomalies of the subclavian arteries are important anatomical markers for chromosome 22q11 deletion, independent of the laterality of the aortic arch.37 The subclavian arterial anomalies encompass aberrant origin from the descending aorta, isolated origin, distal ductal origin from the pulmonary artery, and cervical origin of the subclavian artery. Chromosome 22q11 deletion

338

Fetal Cardiology

(a)

(b)

(c)

Figure 22.8 Double aortic arch. (a) Hypothetical models. The two patent arches encircle the trachea and esophagus. The left ductus is usually patent. (b) Fetal echocardiograms. The ascending aorta bifurcates into the right (R) and left (L) aortic arches. The two aortic arches, main pulmonary arterial trunk, and ductus arteriosus together form a vascular complex with a figure ‘9’ configuration around the trachea. RSA, right subclavian artery. The diagnosis of double aortic arch was entertained but was not confirmed pathologically. (c) Postnatal CT angiogram from a different patient shows a double aortic arch. The right arch is bigger than the left. Each arch gives rise to the ipsilateral common carotid and subclavian arteries.

Aortic arch anomalies

339

(a)

(b)

Figure 22.9 Circumflex retroesophageal aortic arch. (a) Volume rendered three-dimensional CT angiograms in a postnatal patient show a right aortic arch that makes an additional arc behind the trachea (and esophagus) to reach the descending aorta on the left side. The left subclavian artery arises from the top of the descending aorta through a diverticulum of Kommerell. (b) Fetal echocardiograms from a different fetus. In the transverse view (left-hand panel), the aorta forms an arch on the left side of the trachea and turns to the right to course behind the trachea. The course of the aorta and aortic arch is marked by white dots. In the coronal view (right-hand panel), the descending aorta is on the right (reproduced with permission from reference 6).

was associated in over 75% of patients with conotruncal malformation and subclavian arterial anomaly, while it was present in fewer than 30% when there was no subclavian artery anomaly. It has also been shown that there is an increased risk for Down syndrome where there is an aberrant right subclavian artery.38,39 When an aortic arch anomaly is found, fetal karyotyping is recommended, especially when it is associated with intracardiac anomalies, extracardiac malformations, or increased nuchal translucency.26

Postnatal clinical manifestations and management The aortic arch anomalies constituting a vascular ring may cause symptoms and signs of airway and/or esophageal compression that include stridor, cough, asthma, respiratory distress, apnea, recurrent episodes of pneumonia, dysphagia, and episodes of choking. Typically, the more severe is the airway or esophageal compression, the earlier

340

Fetal Cardiology

(a)

(b)

Figure 22.10 Pulmonary artery sling. (a) Volume rendered three-dimensional CT angiogram from a patient shows that the left pulmonary artery arises very distal from the main pulmonary artery and courses to the left behind the trachea and in front of the esophagus. Minimal intensity thick section shows abnormal branching pattern of the tracheobronchial tree. There is long-segment narrowing of the distal trachea between the abnormal bronchus to the right upper and middle lobes and the tracheal bifurcation. The upper tracheal bifurcation is too high for normal carina, while the lower bifurcation is too low. Some consider the upper bifurcation as the carina, the vertical narrowed part as the left main bronchus, and the bronchus to the right lower lobe as a ‘bridging bronchus’. (b) Fetal echocardiograms from a different fetus show that the left pulmonary artery has an unusual rightward sweep in front of the descending aorta (right-hand panel). It courses to the left through the space below the ductus arteriosus (courtesy of Dr Hirokazu Yorioka, Kansai Medical University Hirakata Hospital, Japan).

is the age at presentation.5,7–9 A vascular ring in double aortic arch is tight, and the affected patients manifest early in life, while a vascular ring in the right aortic arch with aberrant left subclavian artery and left ductus tends to be loose, and the clinical presentation can be later in life or even absent. An interesting case has recently been reported in which double aortic arch caused prenatal central airway obstruction with enlarged hyperechogenic lungs.40 The symptoms and signs of airway or esophageal compression

are usually mild or absent in aortic arch anomalies with a vascular sling, except for pulmonary artery sling, in which a long-segment intrinsic narrowing of the distal trachea is common. Echocardiography is of limited use for determining severity of airway compression. The definitive diagnosis is made by either contrast-enhanced computed tomography or magnetic resonance imaging. Although rigid bronchoscopy has been considered a diagnostic must, its diagnostic

Aortic arch anomalies

value as compared to computed tomography and magnetic resonance imaging should be reevaluated. Catheterization with angiography is no longer used solely for a diagnostic purpose. Management includes surgical division of the structures contributing to the vascular ring.7–9 Video-assisted thoracoscopic division of vascular rings in pediatric patients is a promising future trend.41 The aberrant subclavian artery causing significant airway or esophageal compression can be divided and connected to the ipsilateral common carotid artery. A large Kommerell diverticulum can be remodeled, because it may develop aneurysmal dilatation and continue to compress the trachea and esophagus even after the ring has been released.11,42,43 Patients with associated tracheomalacia may continue to have symptoms that can last for months, but ultimately resolve with time. Long-term results are generally excellent, with minimal morbidity and mortality.

References 1. Moulaert AJ, Bruins CC, Oppenheimer-Dekker A. Anomalies of the aortic arch and ventricular septal defects. Circulation 1976; 53: 1011–15. 2. Becker AE, Anderson RH. Cardiac embryology: a help or a hindrance in understanding congenital heart disease. In: Nora JJ, ed. Congenital Heart Disease: Causes and Processes. New York: Futura Publishing Co, 1984: 339–58. 3. Edwards JE. Anomalies of the derivatives of the aortic arch system. Med Clin North Am 1948; 32: 925–48. 4. Edwards JE. Malformation of the aortic arch system manifested as ‘vascular rings’. Lab Invest 1953; 2: 56–75. 5. Weinberg PM. Aortic arch anomalies. In: Allen HD, Clark EB, Gutgesell HP, Driscoll DJ, eds. Moss and Adams’ Congenital Heart Disease in Infants, Children and Adolescents. Philadelphia: Lippincott Williams & Wilkins, 2001: 707–35. 6. Yoo SJ, Min JY, Lee YH et al. Sonographic diagnosis of aortic arch anomalies. Ultrasound Obstet Gynecol 2003; 22: 535–46. 7. Chun K, Colombani PM, Dugeon DL, Haller JA. Diagnosis and management of congenital vascular rings: a 22-year experience. Ann Thorac Surg 1992; 53: 597–603. 8. Van Son JAM, Julsrud PR, Hagler DJ et al. Surgical treatment of vascular rings: The Mayo Clinic experience. Mayo Clin Proc 1993; 68: 1056–63. 9. Kocis KC, Midgley FM, Ruckman RN. Aortic arch complex anomalies: 20-year experience with symptoms, diagnosis, associated cardiac defects, and surgical repair. Pediatr Cardiol 1997; 18: 127–32. 10. Donnelly LF, Fleck RJ, Pacharn P et al. Aberrant subclavian arteries: cross-sectional imaging findings in infants and children referred for evaluation of extrinsic airway compression. AJR Am J Roentgenol 2002; 178: 1269–74. 11. Cinà CS, Althani H, Pasenau J, Abouzahr L. Kommerell’s diverticulum and right-sided aortic arch: a cohort study and review of the literature. J Vasc Surg 2004; 39: 131–9.

341

12. McElhinney DB, Hoydu AK, Gaynor JW et al. Patterns of right aortic arch and mirror-image branching of the brachiocephalic vessels without associated anomalies. Pediatr Cardiol 2001; 22: 285–91. 13. Han JJ, Sohn S, Kim HS, Won TH, Ahn JH. A vascular ring: right aortic arch and descending aorta with left ductus arteriosus. Ann Thorac Surg 2001; 71: 729–31. 14. Zachary CH, Myers JL, Eggli KD. Vascular ring due to right aortic arch with mirror-image branching and left ligamentum arteriosus: complete preoperative diagnosis by magnetic resonance imaging. Pediatr Cardiol 2001; 22: 71–3. 15. Peirone A, Aldulah MM, Dicke F, Freedom RM. Echocardiographic evaluation, management and outcomes of bilateral arterial ducts and complex congenital heart disease: 16 years’ experience. Cardiol Young 2002; 12: 272–7. 16. Holmes KW, Bluemke LA, Vricella LA et al. Magnetic resonance imaging of a distorted left subclavian artery course: an important clue to an unusual type of double aortic arch. Pediatr Cardiol 2006; 27: 316–20. 17. Schneeweiss A, Blieden L, Shem-Tov A, Deutsch V, Neufeld HN. Retroesophageal right aortic arch. Pediatr Cardiol 1984; 5: 191–5. 18. Philip S, Chen SJ, Wu MH, Wang JK, Lue HL. Retroesophageal aortic arch: diagnostic and therapeutic implications of a rare vascular ring. Int J Cardiol 2001; 79: 133–41. 19. McElhinney DB, Thompson LD, Weinberg PM, Jue KL, Hanley FL. Surgical approach to complicated cervical aortic arch: anatomic, developmental, and surgical considerations. Cardiol Young 2000; 10: 212–19. 20. Baravelli M, Borghi A, Rogiani S et al. Clinical, anatomicopathological and genetic pattern of 10 patients with cervical aortic arch. Int J Cardiol 2007; 114: 236–40. 21. Yagel S, Arbel R, Anteby EY et al. The three vessels and trachea view (3VT) in fetal cardiac scanning. Ultrasound Obstet Gynecol 2002; 20: 340–5. 22. Vinals F, Heredia F, Giuliano A. The role of the three vessels and trachea view (3VT) in the diagnosis of congenital heart defects. Ultrasound Obstet Gynecol 2003; 22: 358–67. 23. Achiron R, Rotstein Z, Heggesh J et al. Anomalies of the fetal aortic arch: a novel sonographic approach to in-utero diagnosis. Ultrasound Obstet Gynecol 2002; 20: 553–7. 24. Chaoui R, Schneider BES, Kalache KD. Right aortic arch with vascular ring and aberrant left subclavian artery: prenatal diagnosis assisted by three-dimensional power Doppler ultrasound. Ultrasound Obstet Gynecol 2003; 22: 661–3. 25. Patel CR, Lane JR, Spector ML, Smith PC. Fetal echocardiographic diagnosis of vascular rings. J Ultrasound Med 2006; 25: 251–7. 26. Zidere V, Tsapakis G, Huggon D, Allan LD. Right aortic arch in the fetus. Ultraound Obstet Gynecol 2006; 28: 876–81. 27. Berg G, Bender F, Soukup M et al. Right aortic arch detected in fetal life. Ultrasound Obstet Gynecol 2006; 28: 882–9. 28. Chaoui R, McEwing R. Three cross-sectional planes for fetal color Doppler echocardiography. Ultrasound Obstet Gynecol 2003; 21: 81–93. 29. Choui R, Hoffmann J, Heling KS. Three-dimensional (3D) and 4D color Doppler fetal echocardiography using spatio-temporal image correlation (STIC). Ultrasound Obstet Gynecol 2004; 23: 535–45.

342

Fetal Cardiology

30. Yagel S, Cohen SM, Shapiro I, Valsky DV. 3D and 4D ultrasound in fetal cardiac scanning: a new look at the fetal heart. Ultrasound Obstet Gynecol 2007; 29: 81–95. 31. Yoo SJ, Lee YH, Kim ES et al. Tetralogy of Fallot in the fetus: findings at targeted sonography. Ultrasound Obstet Gynecol 1999; 14: 29–37. 32. Yorioka H, Kasamatu A, Okuno S et al. Prenatal diagnosis of fetal left pulmonary arterial sling. Ultrasound Obstet Gynecol 2007; 30: 628. 33. Glew D, Hartnell GG. The right aortic arch revisited. Clin Radiol 1991; 43: 305–7. 34. Momma K, Matsuoka R, Takao A. Aortic arch anomalies associated with chromosome 22q11 deletion (CATCH 22). Pediatr Cardiol 1999; 20: 97–102. 35. McElhinney DB, Clark BJ 3rd, Weinberg PM et al. Association of chromosome 22q11 deletion with isolated anomalies of aortic arch laterality and branching. J Am Coll Cardiol 2001; 37: 2114–19. 36. Park IS, Ko JK, Kim YH et al. Cardiovascular anomalies in patients with chromosome 22q11.2 deletion: a Korean multicenter study. Int J Cardiol 2007; 114: 230–5. 37. Rauch R, Rauch A, Koch A et al. Laterality of the aortic arch anomalies of the subclavian artery – reliable indicators

38.

39.

40.

41.

42.

43.

for 22q11.2 deletion syndrome? Eur J Pediatr 2004; 163: 642–5. Goldstein WB. Aberrant right subclavian artery in mongolism. Am J Roentgenol Radium Ther Nuc Med 1965; 95: 131–4. Chaoui R, Heling KS, Sarioglu N et al. Aberrant right subclavian artery as a new cardiac sign in second- and third-trimester fetuses with Down syndrome. Am J Obstet Gynecol 2005; 192: 257–63. Shum DJ, Clifton MS, Coakley FV et al. Prenatal tracheal obstruction due to double aortic arch: a potential mimic of congenital high airway obstruction syndrome. AJR Am J Roentgenol 2007; 188: W82–5. Koontz CS, Bhatia A, Forbess J, Wulkan ML. Video-assisted thoracoscopic division of vascular rings in pediatric patients. Am Surg 2005; 71: 289–91. Ota T, Okada K, Takanashi S, Yamamoto S, Okida Y. Surgical treatment of Kommerell’s diverticulum. J Thorac Cardiovasc Surg 2006; 131: 574–8. Kouchoukos N, Masetti P. Aberrant subclavian artery and Kommerell anerysm: surgical treatment with a standard approach. J Thorac Cardiovasc Surg 2007; 133: 888–192.

23 Developments in diagnosis of transposition of the great arteries Laurent Fermont

Simple transposition of the great arteries is life-threatening, sometimes as early as the delivery theater. Fetal diagnosis resulting in preventing acute deterioration of hemodynamic imbalance should be carried out. Transposition of the great arteries is detectable prenatally within low-risk populations even in the first trimester, when such examinations of the heart have been indicated (preexisting conditions, nuchal translucency). The only way to achieve such a diagnosis would be to assess systematically the arterioventricular concordance in all fetuses at the time of sonographic examination, whatever the age of gestation or the indication. After more than 20 years of effort to obtain such care, prenatal diagnosis of transposition of the great arteries remains difficult to achieve. A wide range of diagnosis rates among surgical cardiopediatric neonates is currently observed: we note 7% in 2004 at the Children’s Hospital Boston1 compared with more than 65% in Paris.2 In contrast to most congenital heart malformations, transposition of the great arteries is considered to be isolated. It is largely accepted that neither family history nor sonographic signs pointing toward association, such as chromosomal anomalies including 22q11 deletion, are generally encountered, which excludes advice to systematically karyotype these fetuses. These assertions must be revisited. In fact, transposition of the great arteries is neither always sporadic nor always as simple as believed, based on the definition. Recurrences are rarely but certainly encountered in some families, leading to consideration of genetic3 and embryologic theories, newly described in this field.4,5 Definition of ‘transposition of the great arteries’ must include the necessary mitral–pulmonary fibrous continuities (Figure 23.1). This rules out possible indications for systematic karyotyping even in cases of ventricular septal defect or pulmonary atresia (Figure 23.2) or stenosis. On the other hand, karyotyping remains advised in cases of subpulmonary conus. Extracardiac associations must also be ruled out.

Results of switch procedures depend on many factors. The prognosis may derive from several anatomical details detectable by ultrasound. Ventricular septal defects may alter surgical indications and results when multiple or located at the posterior inlet part of the septum (Figure 23.3 and Video clip 23.1). Atrioventricular valves should be assessed to exclude atrioventricular septal defects, straddling tricuspid valves (suspected from hypoplastic right ventricles), straddling mitral valves, or other anomalies such as malformations of the papillary muscles (parachute-like apparatus) or of the chordae. Description of the anatomical pattern of the coronary arteries can be obtained preoperatively by echocardiography, but this remains the first step of the operation at the time of surgery. Many variations have been described, most of them without negative interference at the time of surgery.6 Normally, in transverse view, sigmoidal commissural alignment, with the aorta anteriorly and the pulmonary artery posteriorly, is regularly observed (Figure 23.4). A malalignment could be an indicator of coronary anomalies (Figure 23.5). Even in the case of rare occurences such as a

Figure 23.1 Simple transposition of the great arteries: left ventricle– pulmonary artery–mitral continuity.

344

Fetal Cardiology

Figure 23.2

Figure 23.4

Transposition of the great arteries with ventricular septal defect and pulmonary atresia.

Transverse view of the arterial annuli: commissural alignment.

Figure 23.3

Figure 23.5

Inlet ventricular septal defect. See also corresponding Video clip 23.1.

Figure 23.6 Intramural coronary artery between aorta and pulmonary artery.

Transverse view of the arterial annuli: commissural malalignment.

Developments in diagnosis of transposition of the great arteries

Figure 23.7

Figure 23.8

Prenatal transposition of the great arteries and coarctation: dilatation of the pulmonary artery.

Postnatal long-term dilatation of already dilated aorta.

single ostium, efficient management can be attained. Among such variations, significant risk arises from difficulties in transferring the ostia when a portion of the coronary artery lies between the aorta and the pulmonary artery.7 This condition is defined as intramural coronary artery. It seems that this specific condition could be detected prenatally (Figure 23.6). This should be confirmed by prospective studies, but as yet the prenatal consequences of such diagnoses are uncertain. The long-term consequences may be altered by prenatal discrepancy in diameters between the aorta and pulmonary artery (Figure 23.7), which may lead to the development of significant dilatation of an already dilated aorta8 (Figure 23.8), with consequences for taking part in sport, for pregnancy, and risks of reoperation and aortic valve replacement, especially when associated with aortic insufficiency. Fetal detection of simple transposition of the great arteries still seems far from easy to achieve, despite steady improvements during the past two decades. This has been demonstrated recently. The solution could be found in a greater availability of three-dimensional echocardiography in general sonography over the next few years.9,10 An ongoing concern would be the long-term consequences of altered neuropsychological effects of acute postnatal hemodynamic disturbances.11 These results may derive from altered prenatal hemodynamics,12 but equally point out the high value of prenatal diagnosis of transposition of the great arteries.

References 1. Bartlett JM, Wypij D, Bellinger DC et al. Effect of prenatal diagnosis on outcomes in D-transposition of the great arteries. Pediatrics 2004; 113: 335–40. 2. Koshnood B, De Vigan C, Vodovar V et al. [Trends in antenatal diagnosis, pregnancy termination and perinatal

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

345

mortality in infants with congenital heart disease: evaluation in the general population of Paris 1983–2000]. J Gynecol Obstet Biol Reprod (Paris) 2006; 35: 455–64. [in French] Muncke N, Jung C, Rüdiger H et al. Missense mutations and gene interruption in PROSIT240, a novel TRAP240-like gene, in patients with congenital heart defect (transposition of the great arteries). Circulation 2003; 108: 2843–50. Bajolle F, Zaffran S, Bonnet D. Bases moléculaires des cardiopathies congénitales humaines. Arch Mal Coeur Vaiss 2007; 100: 484–9. Bajolle F, Zaffran S, Kelly RG et al. Rotation of the myocardial wall of the outflow tract is implicated in the normal positioning of the great arteries. Circ Res 2006; 98: 421–8. Qamar ZA, Goldberg CS, Devaney EJ, Bove EL, Ohye RG. Current risk factors and outcomes for the arterial switch operation. Ann Thorac Surg 2007; 84: 871–8. Raisky O, Bergoend E, Agnoletti G et al. Late coronary artery lesions after neonatal arterial switch operation: results of surgical coronary revascularization. Eur J Cardiothorac Surg 2007; 31: 894–8. Mohammadi S, Serraf A, Belli E et al. Left-sided lesions after anatomic repair of transposition of the great arteries, ventricular septal defect, and coarctation: surgical factors. Thorac Cardiovasc Surg 2004; 128: 44–52. Chaoui R, Hoffmann J, Heling KS. Three-dimensional (3D) and 4D color Doppler fetal echocardiography using spatiotemporal image correlation (STIC). Ultrasound Obstet Gynecol 2004; 23: 535–45. Gonçalves LF, Espinoza J, Romero R et al. A systematic approach to prenatal diagnosis of transposition of the great arteries using 4-dimensional ultrasonography with spatiotemporal image correlation. J Ultrasound Med 2004; 23: 1225–31. McQuillen P, Amrick S, Barkovitch J et al. Balloon atrial septostomy is associated with preoperative stroke in neonates with transposition of the great arteries. Circulation 2006; 113: 281–5. Rudolph AM. Aortopulmonary transposition in the fetus: speculation on pathophysiology and therapy. Pediatr Res 2007; 61: 375–80.

24 Abnormal visceral and atrial situs and congenital heart disease Shi-Joon Yoo, Mark K Friedberg, and Edgar Jaeggi

Introduction Normally, there is conspicuous asymmetry of the forms and arrangement of the abdominal and thoracic organs (Figure 24.1, Table 24.1). Abdominal asymmetry is characterized by a specific rightward–leftward orientation of the non-paired solitary organs including the liver, stomach, and spleen. Thoracic asymmetry, on the other hand, is due to asymmetry of the forms of the paired organs including the lungs, bronchi, pulmonary arteries, and atria. The term ‘situs’ is used to define the right–left orientation or arrangement of the body organs in three anatomical levels: abdominal, bronchopulmonary, and atrial levels.1–4 The situs, however, does not define cardiac position. As abnormalities of situs are harbingers of congenital heart disease, it is very important to determine the situs as the initial step of fetal cardiac evaluation. Usually organ arrangement is concordant throughout the body, such that the organ arrangement in one anatomical location reflects that of other locations. However, less commonly, discordance may exist between organ situs of the abdomen, the bronchopulmonary system, and the atria of the heart.3,4–9 Therefore, a complete evaluation should assess the situs in all of these anatomical locations. There are three types of situs: situs solitus, situs inversus, and heterotaxy (Figure 24.2, Table 24.2). Situs solitus indicates the usual and therefore normal arrangement of the organs. Situs inversus refers to an inverted or mirrorimage arrangement of the organs. In the presence of situs solitus or inversus, discordant situs between the three different anatomical levels is exceedingly rare.3 Heterotaxy (Greek, heteros (other or different) + taxis (arrangement)) indicates that the arrangement of the organs is different from the usual arrangement of situs solitus or situs inversus.1,8 Heterotaxy has often been called ‘situs ambiguus’ (uncertain situs).2,3,5 However, ‘situs ambiguus’ is not an appropriate term, as the organ arrangement in heterotaxy is not uncertain but, rather, complex or difficult to define.10 Abdominal heterotaxy is characterized

by jumbled-up arrangement of the non-paired solitary organs.1–16 The liver and stomach may be disposed in a random fashion as far as their location on the right or left side is concerned. The intestine is commonly malrotated. In most cases, the spleen is absent (asplenia), or multiple spleens are present on the right or on the left (polysplenia). Rarely, a normal spleen is found. Bronchopulmonary heterotaxy is characterized by the symmetric forms of the lungs, bronchi, and pulmonary arteries.11–16 This symmetry is the result of duplication of either the right- or the left-sided structures, thereby dividing thoracic heterotaxy into two subtypes: right and left thoracic isomerisms. Atrial heterotaxy is characterized by symmetry of atrial appendage morphology and the relation of the pectinate muscles to the atrioventricular junction.3,6,7,11 Atrial heterotaxy can also be divided into right and left atrial isomerisms. Generally, asplenia occurs with right bronchopulmonary and atrial isomerism, while polysplenia occurs with left bronchopulmonary and atrial isomerism. However, polysplenia can occur with right isomerism, and asplenia with left isomerism. Isomerisms may even occur with a normal spleen. While splenic morphology is somewhat variable, there is tighter concordance between bronchopulmonary situs and atrial situs so that the bronchopulmonary arrangement usually reflects the atrial arrangement.6,7

Incidence and genetics The incidence of situs inversus ranges from 1 in 2500 to 1 in 25 000 living persons, with 1 per 7000–8000 being the most common estimate from mass radiographic surveys in adults.17,18 The majority of cases of situs inversus are associated with abnormal ciliary function.18 Conversely, situs inversus occurs in 50% of patients with primary ciliary dyskinesia or immotile cilia syndrome, a combination known as Kartagener syndrome.19 The majority of primary ciliary dyskinesia disorders have an autosomal recessive pattern with extensive genetic heterogeneity.

348

Fetal Cardiology

(a)

(b)

Figure 24.1 Normal asymmetry of the thoracic and abdominal organs (a) and the heart (b). Note the asymmetric forms of the paired organs of the thorax and the asymmetric arrangement of the non-paired abdominal organs. The cardiac asymmetry is characterized by asymmetric shape of the atrial appendages, presence or absence of sulcus and crista terminalis, and asymmetric extent of the atrial pectinate muscles relative to the atrioventricular junction. GB, gallbladder; PA, pulmonary artery.

Table 24.1

Features characterizing normal asymmetry of body organs

Abdominal organs

Right

Left

•

•

Larger lobe of the liver

Spleen and stomach

Lung lobation

• Three lobes

•

Two lobes

Main bronchus

•

•

Long, hyparterial

Pulmonary artery

• Transverse course in front of the right main bronchus

•

Oblique course crossing over the left main bronchus (epbronchial)

•

•

Origin of the first branch distally at the hilum

• Triangular appendage with wide junction demarcated by crista terminalis

•

Finger-like appendage with narrow junction not demarcated by crista terminalis

•

•

Pectinate muscles confined to the appendage

•

No fossa ovalis

Atrium

Short, eparterial

Origin of the first branch proximally in the mediastinum

Pectinate muscles extending to the atrioventricular junction

• Fossa ovalis with limbus

The incidence of heterotaxy is approximately 1–1.44 per 10 000 total births.9,17,18,20,21 In the New England Regional Infant Cardiac Program, heterotaxy was found in 4.2% of infants with congenital heart disease.22 In fetal series, although the relative incidences of right and left isomerism vary, left isomerism is consistently more common than right isomerism.15,16,21,23 In postnatal series. most reports

describe a higher incidence of right isomerism,11,14 while the most recent postnatal series from our institution showed mild predominance of left isomerism.21 It has been postulated that the combination of heart block and atrioventricular valve insufficiency in left isomerism leads to fetal demise and a lower postnatal incidence. There is strong evidence to suggest that heterotaxy has a genetic

Abnormal visceral and atrial situs

(a)

(b)

Figure 24.2 Classical types of visceral (a) and atrial (b) situs. A, aortic arch; a, descending aorta; az, azygos vein; IVC, inferior vena cava; Sp, spleen; St, stomach; SVC, superior vena cava.

Table 24.2

Types of visceral and atrial situs

Abdominal situs

Bronchopulmonary situs

Atrial situs

Situs solitus

Situs solitus

Situs solitus

Situs inversus

Situs inversus

Situs inversus

Heterotaxy

Heterotaxy

Heterotaxy

with asplenia

with right isomerism

with right isomerism

with polysplenia

with left isomerism

with left isomerism

with normal spleen

349

350

Fetal Cardiology

etiology.9,17,20,24–28 Clinical data on recurrence of visceral heterotaxy show that about 5% of the siblings are affected.9,20 Allan et al reported approximately 10% recurrence in their fetal series.26 Familial heterotaxy occurs with autosomal dominant, recessive, and X-linked inheritance,17,29 and it has been suggested that single gene defects result in the wide spectrum of heterotaxy phenotypes.17,25 All possible situs variants, solitus, inversus, and heterotaxy, can appear in the same family.29 There is a tendency for right isomerism to affect males more commonly, and for left isomerism to affect females more commonly. Given the

genetic complexity of these disorders, genetic counseling is indicated for affected families. Most studies have found that there is not a significanty increased risk for chromosomal abnormalities in heterotaxy.27,30 It has also been suggested that abnormal visceral situs is strongly predictive of a normal karyotype.29 However, one study from Taiwan reported a 10% incidence of trisomy 18 in heterotaxy.31 Other cases have been associated with microdeletion of chromosome 22 and other rarer translocations.20,32 Maternal diabetes is also considered a risk factor for atrial isomerism.33,34

(a)

(b)

(c)

Transverse sonograms of the upper abdomen showing the arrangement of the abdominal organs in situs solitus (a), heterotaxy with asplenia (b), and heterotaxy with polysplenia (c). The abdominal situs is determined by observing the location of the liver, stomach, spleen, abdominal aorta (a), and inferior vena cava (v). In heterotaxy, the orderly pattern of the organ arrangement is disrupted. In asplenia the aorta and inferior vena cava are seen backto-back on the same side (b). In polysplenia the aorta and dilated azygos vein are seen side-by-side in front of the spine (c). Multiple spleens (S) are seen behind the stomach. UV, umbilical vein.

Figure 24.3

Abnormal visceral and atrial situs

351

How to determine the visceral and atrial situs As discussed elsewhere in this book, congenital heart disease should be analyzed in a sequential segmental manner, especially when an abnormality of situs exists.10,35–37 Sequential segmental analysis starts with determination of the visceral and atrial situs. The fetus positions itself within the maternal uterine cavity in different ways. Accordingly, the right–left orientation of the fetus shown at fetal sonography varies. As determination of the situs requires definition of the organ arrangement relative to the midline, the right and left sides of the fetus should first be identified.38 Once the right and left sides of the fetus have been identified, the abdominal situs is determined by observing the location of the liver, stomach, spleen, abdominal aorta, and inferior vena cava in the transverse view of the upper abdomen (Figure 24.3).10 In situs solitus, the larger lobe of the liver is seen on the right, and the stomach and spleen are seen on the left (Figure 24.3a). The abdominal aorta is located posteriorly at the left anterior aspect of the spine, and the inferior vena cava is located more anteriorly on the right, almost always entering the right atrium. In situs inversus, this right–left relationship is mirror-imaged. In heterotaxy with right isomerism or asplenia, the liver is characteristically symmetrical and the stomach tends to be near the midline either on the right or on the left (Figure 24.3b). Occasionally, a part of the stomach is seen in the lower thorax.39 The absence of the spleen can be recognized by observing the posterior and lateral aspects of the stomach where the spleen, if present, would be located. In most cases with heterotaxy and asplenia, the abdominal aorta and inferior vena cava are juxtaposed on the same side of the spine.10,40 In heterotaxy with left isomerism or polysplenia, symmetry of the liver is less evident, with its major portion often lying to one side of the abdomen (Figures 24.3c and 24.4).1,4 Similarly, the stomach is seldom in the midline. The arrangement of the liver and stomach often mimics that of either situs solitus or situs inversus. As the spleen develops in the dorsal mesogastrium, multiple spleens aggregate in a single location along the greater curvature of the stomach11,13 (Figure 24.3c). Occasionally, multiple spleens may fuse together to form a single multilobulated mass. This is in contrast to accessory spleens, in which a dominant spleen is present together with multiple small splenules that can be found anywhere in the peritoneal cavity. Polysplenia is commonly associated with interruption of the suprarenal infrahepatic segment of the inferior vena cava with continuation through the azygos or hemiazygos venous system (Figure 24.5). Although it can also occur with other types of body situs, interruption of the inferior vena cava is highly suggestive of polysplenia and it can be diagnosed when the coronal or sagittal sonograms

(a)

(b)

Figure 24.4 Left isomerism with probable polysplenia. (a) Transverse sonogram of the upper abdomen shows the right-sided stomach and left-sided larger lobe of the liver. (b) Four-chamber view of the heart shows levocardia and normal cardiac anatomy. RT, right; LT, left.

do not show the inferior vena caval connection to the atrial segment (Figure 24.5a).41–44 It can also be suspected when two equally sized vessels are seen in the posterior mediastinum in either the coronal or the transverse view; one being the dilated azygos or hemiazygos vein and the other being the descending aorta (Figure 24.5b–24.5d).10,41–44 On color

352

Fetal Cardiology

(a)

(b)

(c)

(d)

Figure 24.5 Interruption of the inferior vena cava with azygos continuation. (a) Sagittal view showing the inferior vena cava that does not connect to the atrium but to the dilated azygos vein. (b) Four-chamber view shows complete atrioventricular septal defect with a common atrium. There are two vessels in the posterior mediastinum. They are the descending aorta and the dilated azygos vein draining the interrupted inferior vena cava. (c, d) Oblique coronal views of the thorax show two parallel vessels, the aorta and azygos vein, which demonstrate opposite blood flow direction. LV, left ventricle; HV, hepatic vein; mb, moderator band; RSVC, right superior vena cava, RV, right ventricle.

Doppler examination, the blood flow through these two parallel vessels is in opposite directions (Figure 24.5d). Because the fetal airway is filled with fluid, it is possible to determine the bronchopulmonary situs by ultrasound.10 In situs solitus, the right main bronchus is short, while

the left main bronchus is long. The right pulmonary artery courses horizontally in a coronal plane in front of the right main bronchus, while the left pulmonary artery courses posteriorly in an oblique direction above the left main bronchus. This asymmetric anatomy can be seen in

Abnormal visceral and atrial situs

(a)

(a)

(b)

(b)

Figure 24.6

Figure 24.7

Bronchopulmonary anatomy seen in coronal sonograms. (a) Normal bronchopulmonary anatomy. The trachea (Tr) bifurcates into the right (rb) and left (lb) main bronchi. The left aortic arch (A) and left pulmonary artery (asterisk) course above the left main bronchus. The right pulmonary artery does not course above the right main bronchus. (b) Left isomeric bronchopulmonary anatomy. Both pulmonary arteries (asterisks) are seen to course above the ipsilateral bronchi. Although it is possible to demonstrate the anatomy, the practical usefulness of this is questionable.

353

Bronchopulmonary anatomy seen in transverse sonograms of right (a) and left (b) isomerism. The pulmonary arterial branching is symmetrical in both cases. Asterisks depict the main bronchi. The difference between the two is not evident. Note the two vessels (crosses) in front of the spine in (b). One is the aorta and the other is the dilated azygos vein in this case of left isomerism. a, descending aorta; pa, pulmonary artery.

354

Fetal Cardiology

(a)

(b)

(c)

Atrial appendages. (a) Short-axis echocardiogram of a normal fetal heart showing the appendages of the right (RA) and left (LA) atria. The appendages embrace the aortic root (A) and pulmonary trunk (P) from behind. The pectinate muscles are seen as small nodular structures along the lateral wall of the right atrial appendage (RAA). The right atrial appendage is large and has a wide opening, The left atrial appendage (LAA) is small and tubular and has a narrow opening. (b, c) Symmetrical appendages (asterisks) in right (b) and left (c) isomerism. IVC, inferior vena cava.

coronal view where the trachea and both bronchi are seen (Figure 24.6a). On the left side, the left aortic arch and left pulmonary artery are seen in cross-section above the left main bronchus. On the right side, on the other hand, the right pulmonary artery is not seen above the right main bronchus, while a small azygos venous arch can be seen as a tiny dot. In situs inversus, there is a mirror-image arrangement. In right isomerism, the main bronchi are symmetrically short with neither pulmonary artery coursing backward above the main bronchus. In left isomerism, the main bronchi are symmetrically long and the

pulmonary arteries course backward above the main bronchi (Figure 24.6b). Symmetric branching of the pulmonary arteries in right or left isomerism can also be shown in the transverse view (Figure 24.7). However, sonographic determination of the bronchopulmonary situs is time-consuming and is largely impractical to perform as part of the routine examination. On examination of anatomical specimens, determination of the atrial situs is based on the morphology of the atrial appendages and, more consistently and accurately, the extent of the pectinate muscles relative

Figure 24.8

Abnormal visceral and atrial situs

(a)

355

(b)

Figure 24.9 Bilateral superior venae cavae. (a) Coronal view from a fetus with right isomerism showing bilateral superior venae cavae (SVC) connected to the roof of the common atrium. (b) Three-vessel view from another fetus with right isomerism showing bilateral superior venae cavae (asterisks). Note the symmetric pulmonary arterial branching (p). The pulmonary artery is much smaller than the aorta (A) because of severe pulmonary stenosis.

to the atrioventricular junction (Figure 24.1b).6,7,11,35 The right atrial appendage is triangular and has a wide orifice demarcated by a prominent crest termed the crista terminalis. The left atrial appendage is finger-like and has a narrow orifice that is not demarcated by a crest. The pectinate muscles of the right atrium extend from the appendage to reach the atrioventricular junction, while those of the left atrium are confined to the appendage and do not reach the atrioventricular junction, except for a small area on rare occasions.35 The atrial appendages can be visualized by fetal echocardiography (Figure 24.8).10,45,46 However, it is difficult to differentiate right atrial appendage morphology from left atrial appendage morphology based on two-dimensional echocardiograms, both prenatally and postnatally.43 Furthermore, although the pectinate muscles can be visualized, it is very difficult to evaluate the extent of these muscles. Therefore, we think it impractical and often unreliable to determine atrial situs by echocardiographic examination of atrial morphology. On the other hand, the atrial situs can reliably be predicted by determining the abdominal situs, because in most cases, atrial situs is concordant with abdominal situs. However, it should be remembered that there are occasional exceptions to the visceroatrial concordance rule.3,4–7,47,48

Situs inversus and congenital heart diseases In the presence of situs inversus, dextrocardia is the usual and appropriate position of the heart, although levocardia and mesocardia are not infrequently found. The frequency and type of congenital heart disease that occurs in situs inversus vary according to the given cardiac position. Dextrocardia in situs inversus is less commonly associated with congenital heart disease, although the true frequency is largely unknown because of problems with ascertainment of the cases. It may be as great as 50% in infants and children, but less than 10% in adolescents and adults.49,50 The type of congenital heart disease in this setting includes tetralogy of Fallot, double-outlet right ventricle, and complete and corrected transposition of the great arteries. Levocardia and mesocardia in situs inversus are rare but almost always associated with congenital heart disease. The typical pathology seen in situs inversus with levocardia or mesocardia is congenitally corrected transposition of the great arteries. This lesion is usually associated with additional defects that result in significant hemodynamic impairment requiring surgery.

356

Fetal Cardiology

Heterotaxy and congenital heart disease Congenital heart disease is exceedingly common when there is heterotaxy. We are not aware of any case with asplenia and right isomerism that is not associated with congenital heart disease. It is generally accepted that most cases of left isomerism have associated congenital heart disease, although the type of lesion is more variable than that found in association with right isomerism. It is interesting, however, to note that polysplenia and left isomerism may be found incidentally without clinical evidence of a cardiac defect.13,34 In some individuals, multiple spleens or an interrupted inferior vena cava are diagnosed by abdominal ultrasound or computed tomography performed for reasons other than to define cardiac abnormalities, while other children present with biliary atresia. Therefore, it is considered that the true incidence of congenital heart disease in polysplenia or left isomerism is largely unknown. There is considerable overlap in the cardiac malformations within the two types of heterotaxy (Table 24.3).11–19,23,33,49–54 Bilateral superior venae cavae (Figure 24.9) and complete atrioventricular septal defect (Figures 24.5b, 24.10, and 24.11) are common in both right and left isomerisms. The atrioventricular septal defect

Figure 24.10 Complete atrioventricular septal defect in right isomerism. Four-chamber view shows atrioventricular septal defect. The right ventricle (RV) is slightly smaller than the left ventricle (LV). The atrium is a common chamber with a small strand of remnant atrial septum seen as a dot.

Table 24.3 Common congenital cardiac defects in right and left isomerism (composite data from references 7, 13, 14, 21, 23, 33, 41, and 51–59) Right isomerism

Left isomerism

45%

45%

Bilateral superior venae cavae

70%

60%

Absence of coronary sinus

∼100%

∼60%

Interruption of the inferior vena cava

< 2.5%

80%

Bilateral systemic venous drainage

Juxtaposition of the aorta and inferior vena cava Extracardiac type of total anomalous pulmonary venous connection with/without obstruction

∼90% 50%, with obstruction in 50% 4%

Pulmonary venous connection to ipsilateral atria

90%

Atrioventricular septal defect

Uncommon Rare 45% 50%

Atrial septum

Functionally common atrium in 50%

Usually better formed, intact in ∼20%

Atrioventricular connection

Univentricular in 70%

Biventricular in ∼75%

Ventriculoarterial connection

Concordant only in 4%

Concordant in ∼70%

Pulmonary atresia or stenosis

80%

30%

Left-sided obstructive lesion

< 5%

∼30%

Heart block/bradycardia

Rare

25–70%

Abnormal visceral and atrial situs

(a)

357

(b)

Figure 24.11

(c)

Unbalanced atrioventricular septal defect and total anomalous pulmonary venous connection to the innominate vein in right isomerism. (a) Four-chamber view demonstrates the pulmonary venous confluence (CPV) behind the atrium. There is unbalanced complete atrioventricular septal defect with a dominant right ventricle (RV) and a small left ventricle (lv). (b) Three-vessel view shows the anteroposterior relationship of the ascending aorta (A) and pulmonary artery (PA). The confluent common pulmonary vein courses backward and leftward in front of the descending aorta (DA). The superior vena cava (V) is on the right. Note the symmetric branching pattern of the pulmonary artery. (c) Oblique axial view shows the common pulmonary vein connecting to the innominate vein (IV).

is usually characterized by an unbalanced commitment of the atrioventricular valve to the underlying ventricular mass, with resultant discrepancy in the size of the left and right ventricles.21,54 This occurs more commonly with right isomerism than with left isomerism. On the other hand, some malformations show noticeable predilection to one type of isomerism. In right isomerism the cardiac anomalies are generally more complex and multiple, and of a more primitive nature, while in general, left isomerism is associated with less complex cardiac malformations. Supracardiac or infracardiac total anomalous pulmonary venous connection, often with pulmonary venous obstruction, is particularly common in right isomerism but

uncommon in left isomerism (Figure 24.12). When the pulmonary veins are connected to the atrium or atria in the presence of right isomerism, the connection is usually through a narrow confluent channel (Figure 24.11). In left isomerism, the pulmonary veins from each lung tend to enter the posterior wall of the ipsilateral atrium separately. As previously mentioned, interruption of the inferior vena cava with azygos or hemiazygos continuation is common in left isomerism. In both conditions, the hepatic veins may have separate openings in the floor of the atrium or atria, rather than forming a confluence. In right isomerism the inferior vena cava is connected to either atrium. The atrial septation is also different in right and

358

Fetal Cardiology

Figure 24.12 Right isomerism with the pulmonary veins (PV’s) connected to the common atrium. Four-chamber view shows the close proximity of the orifices of the pulmonary veins (asterisks).

left isomerism. In 50% of cases of right isomerism, there is functionally a common atrium, in the majority of cases with just a strand of remnant atrial septum traversing the atrial cavity (Figure 24.10). In a four-chamber view, this strand is seen as a dot in the center of the common atrium (the ‘central dot sign’) and is pathognomonic for right isomerism. In left isomerism, the atrial septum is better formed or intact in two-thirds of cases. The atrioventricular connection is more often univentricular in right isomerism, while it is more often biventricular in left isomerism. Univentricular atrioventricular connection in right isomerism is commonly to a dominant chamber of right ventricular morphology, less commonly to a dominant chamber of left ventricular morphology, and rarely to a single ventricular chamber of indeterminate morphology. The morphology of the ventricular chambers in univentricular atrioventricular connection is difficult to define by trabeculation. However, the morphology of the ventricular chambers can be indirectly defined by assessing the spatial relationship between the dominant and rudimentary ventricles. When the rudimentary chamber is related to the crux cordis and is therefore located along the posterior and inferior surface of the ventricular mass, it is the morphologically left ventricle, and, thus, the dominant chamber is the morphologically right ventricle. When the rudimentary chamber is not in contact with the crux cordis but is located anteriorly and

Figure 24.13 Transposition in right isomerism. Ventricular outflow tract view shows the aorta (A) arising from the right ventricle (RV) and the pulmonary artery (P) from the left ventricle (LV). A large ventricular septal defect (d) is seen. 1-a, left-sided atrium.

Figure 24.14 Double-outlet right ventricle in right isomerism. Ventricular outflow tract view shows that both the aorta (A) and the pulmonary artery (p) arise from the right ventricle (RV). A common atrioventricular valve (CAVV) is seen in the right ventricle. The pulmonary artery is smaller because of stenosis.

Abnormal visceral and atrial situs

superiorly, either on the right or on the left, it is the morphologically right ventricle, and, thus, the dominant chamber is the morphologically left ventricle. When only one ventricle is identified, the ventricular morphology is hard to define and called indeterminate. In most cases of right isomerism, the ventriculoarterial connection is abnormal, and can be transposition of the great vessels, doubleoutlet right ventricle, single outlet, etc. (Figures 24.13 and 24.14). In contrast, in up to 70% of cases of left isomerism, the ventriculoarterial connection is concordant. In right isomerism, pulmonary stenosis or atresia is present in approximately 80% of cases (Figures 24.9 and 24.14). The pulmonary artery in right isomerism can be non-confluent, with the right and left pulmonary arteries being supplied by bilateral ductus arteriosus. In left isomerism, pulmonary stenosis or atresia occurs in only approximately 30% of cases, and left-sided obstructive lesion in about 20–30%. Bradycardia with atrioventricular dissociation is common in fetuses with left isomerism, especially when there is an atrioventricular septal defect (Figure 24.15).16,21,23,33,55–59 Left isomerism is the leading cause of fetal heart block associated with structural heart disease, followed by corrected transposition of the great arteries. The atrioventricular dissociation in left isomerism is due to discontinuity between the atrioventricular (AV) node and the conduction axis.60 Hearts with right isomerism may have an accessory sinus node and also tend to have more than one atrioventricular node and bundle, and hence are less likely to develop heart block. Two AV nodes, however, may be used for antegrade and retrograde conduction, causing reentrant tachycardia.61 The incidence of heart block in left isomerism is much higher in fetal series than in postnatal series.16,21,33,57–59 This discrepancy is likely due to the high intrauterine mortality rate of fetuses with left isomerism, structural heart disease, and heart block. Fetal diagnosis of atrioventricular dissociation can be made with the use of M-mode or pulsed Doppler interrogation. In M-mode the cursor line should be placed through the atrial and ventricular walls to demonstrate dissociation of atrial and ventricular contraction. With Doppler, the sampling cursor should be placed so that the signals from the superior vena cava and ascending aorta or the inflow and outflow tracts of the left ventricle can be obtained simultaneously.59 Alternatively, simultaneous Doppler sampling from the pulmonary arteries and veins can be used (Figure 24.15b).

Non-cardiac anomalies in situs inversus and heterotaxy4,8,11–15,19,33,52,62–67 The precise incidence of non-cardiac defects in situs inversus is unknown because a significant but unknown

359

(a)

(b)

Figure 24.15 Atrioventricular dissociation in left isomerism. (a) M-mode echocardiogram through the atrial and ventricular walls shows that the ventricular beats (V) are independent of the atrial beats (A). (b) Simultaneous Doppler tracing of the pulmonary artery and vein shows a similar pattern of atrioventricular dissociation. The letter ‘V’s indicate forward flow through the pulmonary artery, peaking during ventricular systole. The letter ‘A’s indicate reversed flow through the pulmonary vein during atrial systole.

number of situs inversus cases do not come to medical attention. The most recent series from a large referral center showed that 58% of situs inversus cases were associated with intra-abdominal pathologies such as duodenal atresia, biliary atresia, and gastroschisis with malrotation.62 Kartagener syndrome is a rare but well-recognized association of primary ciliary dyskinesia with situs inversus.19 Approximately 50% of people affected by primary ciliary dyskinesia have situs inversus. Heterotaxy is commonly associated with various degrees of intestinal malrotation.62–64 However, only a few of these patients will develop gastrointestinal symptoms related to volvulus that may require surgical intervention.64 As children with heterotaxy have a low risk of adverse outcome related to intestinal rotational abnormalities, specific investigation of intestinal rotation is indicated only in those patients who develop gastrointestinal symptoms. Annular pancreas is also commonly found in visceral heterotaxy. Left isomerism with polysplenia may be associated with biliary atresia, absence or hypoplasia of the gallbladder, and short pancreas (Table 24.4). Biliary atresia is found in up to 20% of individuals with left isomerism.33,67 Conversely, approximately 6% of biliary atresia cases are

360

Fetal Cardiology

Table 24.4 Non-cardiac anomalies in right and left isomerism (composite data from references 12, 14, 33, 52, and 62–68) Right isomerism

Left isomerism

Intestinal malrotation virtually in all Partial thoracic stomach (hiatal hernia) in ∼25%

Intestinal malrotation virtually in all Biliary atresia and/or hypoplastic or absent gallbladder in 20%

Heterogeneous anomalies encountered

Urinary anomalies in 17% Duodenal atresia in 7% Extrahepatic portosystemic shunt, infrequently

associated with left isomerism or, much less commonly, with situs inversus.67 When biliary atresia is present, pulmonary arteriovenous malformations can develop later in life. Polysplenia may also be associated with extrahepatic portosystemic shunt.68 Extrahepatic portosystemic shunt is associated with polysplenia in approximately 15% of cases. Agenesis or hypoplasia of the portal vein also occurs infrequently in polysplenia. It can be a consequence of portosystemic shunt or occur as an unrelated condition. In right isomerism with asplenia, part of the stomach is often seen in the thorax. This pathology might represent either hiatal hernia or partial ectopia of the stomach.39,52 Adrenal, genitourinary, and anal anomalies, such as horseshoe kidneys and adrenals and anal stenosis and atresia, are not uncommon in right isomerism.65

isomerism carries a significantly worse prognosis than left isomerism.12,14,21 The high mortality in right isomerism is related to the high incidence of total anomalous pulmonary venous connection, severe pulmonary outflow tract obstruction, and functionally single ventricle. Right isomerism with obstructive total anomalous pulmonary venous connection is associated with the worst prognosis. When pulmonary outflow tract obstruction is significant, administration of prostaglandin is necessary immediately after delivery to maintain patency of the ductus arteriosus. The majority of patients with heterotaxy, especially those with right isomerism, are potential candidates for single-ventricle palliation.21 However, only one-third of these cases eventually undergo the Fontan operation and a high mortality has been reported in association with surgical management. Risk factors for mortality include right isomerism, total anomalous pulmonary venous connection with obstruction, underdeveloped pulmonary arterial bed, functionally single ventricle, heart block, coarctation of the aorta, and the need for early surgical intervention. In addition, complex anatomy of the systemic as well as pulmonary venous connections, presence of bilateral superior venae cavae, unusual spatial relationship of the ventricles, and complex outflow tract pathology are also important in determining the complexity of postnatal surgical repair and long-term need for reintervention. Postnatal outcomes are also related to the extracardiac anomalies and their resultant complications including biliary atresia and intestinal obstruction. Asplenia and polysplenia with functional asplenia require lifelong antibiotic prophylaxis for prevention of infection.

Acknowledgments Fetal and neonatal outcomes The fetal death rate is higher in left isomerism than in right isomerism,15,16,21 and has been reported to be as high as 40% in left isomerism. The high incidence of fetal demise in left isomerism is thought to be due to the high incidence of sinus bradycardia and heart block that often lead to congestive heart failure and hydrops, especially in the presence of significant atrioventricular valve insufficiency.16,21,58,59 The association of left atrial isomerism with cardiomyopathy in the form of ventricular non-compaction has been recognized in the fetus and has been found to have an almost uniformally fatal outcome.69 The postnatal prognosis is largely dependent on the associated cardiac and extracardiac anomalies. Disappointingly, the most recent data from our institution showed that antenatal diagnosis did not improve overall survival in both left and right isomerism.21 Postnatally, right

We thank Ms Jennifer Russell for reference management and Mrs Eul Kyung Kim for making diagrams.

References 1. Stanger P, Rudolph AM, Edwards JE. Cardiac malpositions. An overview based on study of sixty-five necropsy specimens. Circulation 1977; 56: 159–72. 2. Van Praagh R. Terminology of congenital heart disease: glossary and terminology. Circulation 1977; 56: 139–43. 3. Tynan MJ, Becker AE, Macartney FJ et al. Nomenclature and classification of congenital heart disease. Br Heart J 1979; 41: 544–53. 4. Bartram U, Wirbelauer J, Speer CP. Heterotaxy syndrome – asplenia and polysplenia as indicators of visceral malposition and complex congenital heart disease. Biol Neonate 2005; 88: 278–90. 5. Macartney FJ, Zuberbuhler R, Anderson RH. Morphological considerations pertaining to recognition of atrial

Abnormal visceral and atrial situs

6.

7.

8.

9.

10.

11. 12.

13. 14.

15. 16.

17.

18. 19.

20.

21.

22.

23.

24.

25.

isomerism. Consequences for sequential segmental localisation. Br Heart J 1980; 44: 657–67. Uemura H, Ho SY, Devine WA, Anderson RH. Analysis of visceral heterotaxy according to splenic status, appendage morphology, or both. Am J Cardiol 1995; 76: 846–9. Uemura H, Ho SY, Devine WA, Anderson RH. Atrial appendages and venoatrial connections in hearts with visceral heterotaxy. Ann Thorac Surg 1995; 60: 561–9. Nagel BHP, Williams H, Stewart L et al. Splenic state in surviving patients with visceral heterotaxy. Cardiol Young 2005; 15: 469–73. Freedom RM, Jaeggi ET, Lim JS, Anderson RH. Hearts with isomerism of the right atrial appendages – one of the worst forms of disease in 2005. Cardiol Young 2005; 15: 554–67. Yoo S-J, Lee Y-H, Cho KS, Kim DY. Sequential segmental approach to fetal congenital heart disease. Cardiol Young 1999; 9: 430–44. Van Mierop LHS, Gessner IH, Schieber GL. Asplenia and polysplenia syndrome. Birth Defects 1972; 8: 74–82. Moller JH, Nakib A, Anderson RC, Edwards JE. Congenital cardiac disease associated with polysplenia. A developmental complex of bilateral ‘left-sidedness’. Circulation 1967; 36: 789–99. Peoples WM, Moller JH, Edwards JE. Polysplenia: a review of 146 cases. Pediatr Cardiol 1983; 4: 129–37. Phoon CK, Neill CA. Asplenia syndrome: insight into embryology through an analysis of cardiac and extracardiac anomalies. Am J Cardiol 1994; 73: 581–7. Ho SY, Cook A, Andrew RH et al. Isomerism of the atrial appendages in the fetus. Pediatr Pathol 1991; 11: 589–608. Phoon CK, Villegas MD, Ursell PC, Silverman NH. Left atrial isomerism detected in fetal life. Am J Cardiol 1996; 77: 1083–8. Bowers PN, Martina M, Yost HJ. The genes of left-right development and heterotaxia. Semin Perinatol 1996; 20: 577–88. Zhu L, Belmont JW, Ware SM. Genetics of human heterotaxias. Eur J Hum Genet 2006; 14: 17–25. Bush A, Cole P, Hariri M et al. Primary ciliary dyskinesia: diagnosis and standards of care. Eur Respir J 1998; 12: 982–8. Lin AE, Ticho BS, Westgate MN, Holmes LB. Heterotaxy: associated conditions and hospital-based prevalence in newborns. Genet Med 2000; 2: 157–72. Lim JSL, McCrindle BW, Smallhorn JF et al. Clinical features, management, and outcome of children with fetal and postnatal diagnoses of isomerism syndromes. Circulation 2005; 112: 2454–61. Fyler DC, Buckley LP, Hellenbrand WE, Cohen HE. Report of the New England Regional Infant Cardiac Programme. Pediatrics 1980; 65(Suppl): 376–461. Taketazu M, Lougheed J, Yoo SJ et al. Spectrum of cardiovascular disease, accuracy of diagnosis, and outcome in fetal heterotaxy syndrome. Am J Cardiol 2006; 97: 720–4. Kathiriya IS, Srivastava D. Left-right asymmetry and cardiac looping: implications for cardiac development and congenital heart disease. Am J Med Genet 2000; 97: 271–9. Morelli SH, Young L, Reid B et al. Clinical analysis of families with heart, midline and laterality defects. Am J Med Genet 2001; 101: 388–92.

361

26. Allan LD, Crawford DC, Chitta SK et al. The familial recurrence of congenital heart disease in a prospective series of mothers referred for fetal echocadiography. Am J Cardiol 1986; 58: 334–7. 27. Allan LD, Sharland GK, Chita SK et al. Chromosomal anomalies in fetal congenital heart disease. Ultrasound Obstet Gynecol 1991; 1: 8–11. 28. Eronen M, Kajantie E, Boldt T et al. Right isomerism in four siblings. Pediatr Cardiol 2004; 25: 141–4. 29. Casey B. Two rights make a wrong: human left-right malformations. Hum Mol Genet 1998; 7: 1565–71. 30. Brown DL, Emerson DS, Suhlman LP et al. Predicting aneuploidy in fetuses with cardiac anomalies. Significance of visceral situs and noncardiac anomalies. J Ultrasound Med 1993; 3: 153–61. 31. Lin JH, Chang CI, Wang JK et al. Intrauterine diagnosis of heterotaxy syndrome. Am Heart J 2002; 143: 1002–8. 32. Yates RW, Raymond FL, Sharland GK. Isomerism of the atrial appendages associated with 22q11 deletion in a fetus. Heart 1996; 76: 548–9. 33. Gillajm T, McCrindle BW, Smallhorn JF et al. Outcomes of left atrial isomerism over a 28-year period at a single institution. J Am Coll Cardiol 2000; 36: 908–16. 34. Slavotinek A, Hellen E, Gould S et al. Three infants of diabetic mothers with the malformation of the left-right asymmetry – further evidence for the etiological role of diabetes in this malformation spectrum. Clin Dysmorphol 1996; 5: 241–7. 35. Anderson RH. Terminology. In: Anderson RH, Baker EJ, Macartney FJ et al, eds. Paediatric Cardiology. Edinburgh: Churchill Livingstone, 2002: 19–36. 36. Edwards WD. Classification and terminology of cardiovascular anomalies. In: Emmanouilides GC, Allen HD, Riemenschneider TA, Gutgesell HP, eds. Moss and Adams’ Heart Disease in Infants, Children, and Adolescents Including the Fetus and Young Adult. Baltimore: Williams & Wilkins, 1995: 106–31. 37. Carvalho JS, Ho SY, Shinebourne EA. Sequential segmental analysis in complex fetal cardiac abnormalities: a logical approach to diagnosis. Ultrasound Obstet Gynecol 2005; 26: 105–11. 38. Cordes TM, O’Leary PW, Seward JB, Hagler DJ. Distinguishing right from left: a standardized technique for fetal echocardiography. J Am Soc Echocardiogr 1994; 7: 47–53. 39. Wang JK, Chang MH, Li YW et al. Association of hiatus hernia with asplenia syndrome. Eur J Pediatr 1993; 152: 418–20. 40. Rossi G, Corno A, Montemurro G. Prenatal diagnosis of isomerism of the right atrial appendages. Cardiol Young 1992; 2: 298–301. 41. Berg C, Geipel A, Smrcek J et al. Prenatal diagnosis of cardiosplenic syndrome. Ultrasound Obstet Gynecol 2003; 22: 451–9. 42. Celentano C, Malinger G, Rotmensch S et al. Prenatal diagnosis of interrupted inferior vena cava as an isolated finding: a benign vascular malformation. Ultrasound Obstet Gynecol 1999; 14: 215–18. 43. Atkinson DE, Drant S. Diagnosis of heterotaxy syndrome by fetal echocardiography. Am J Cardiol 1998; 82: 1147–9. 44. Espinoza J, Concalves LF, Lee W et al. A novel method to improve prenatal diasnosis of abnormal systemic venous

362

45.

46.

47.

48.

49.

50.

51.

52.

53.

54.

55. 56.

Fetal Cardiology

connections using three- and four-dimensional ultrasonography and ‘inversion mode’. Ultrasound Obstet Gynecol 2005; 25: 428–34. Colloridi V, Pizzuto F, Ventriglia F et al. Prenatal echocardiographic diagnosis of right atrial isomerism. Prenat Diagn 1994; 14: 299–302. Berg C, Geipel A, Kohl T et al. Fetal echocardiographic evaluation of atrial morphology and the prediction of laterality in cases of heterotaxy syndromes. Ultrasound Obstet Gynecol 2005; 26: 538–45. Mauser I, Deutinger J, Bernaschek G. Prenatal diagnosis of a complex fetal cardiac malformation associated with asplenia. Br Heart J 1991; 65: 293–5. Stewart PA, Becker AE, Wladimiroff JW, Essed CE. Left atrial isomerism associated with asplenia: prenatal echocardiographic detection of complex congenital cardiac malformations. J Am Coll Cardiol 1984; 4: 1015–20. Moller JH. Malposition of the heart. In: Moller JH, Neal WA, eds. Fetal, Neonatal and Infant Cardiac Disease. Norwalk: Appleton & Lange, 1990: 755–74. Hagler DJ, O’Leary PW. Cardiac malpositions and abnormalities of atrial and visceral situs. In: Emmanouilides GC, Allen HD, Riemenschneider TA, Gutgesell HP, eds. Moss and Adams’ Heart Disease in Infants, Children, and Adolescents Including the Fetus and Young Adult. Baltimore: Williams & Wilkins, 1995: 1307–36. Rubino M, Van Praagh S, Kadoba K et al. Systemic and pulmonary venous connections in visceral heterotaxy with asplenia. Diagnostic and surgical consideration based on seventy-two autopsied cases. J Thorac Cardiovasc Surg 1995; 110: 641–50. Hashmi A, Abu-Sulaiman R, McCrindle BW et al. Management and outcomes of right atrial isomerism: a 26-year experience. J Am Coll Cardiol 1998; 31: 1120–6. Patel CR, Lane JR, Muise KL. In utero diagnosis of obstructed supracardiac total anomalous pulmonary venous connection in a patient with right atrial isomerism and asplenia. Ultrasound Obstet Gynecol 2001; 17: 268–71. Huggon IC, Cook AC, Smeeton NC et al. Atrioventricular septal defects diagnosed in fetal life: associated cardiac and extra-cardiac abnormalities and outcome. J Am Coll Cardiol 2000; 36: 593–601. Wren C, Macartney FJ, Deanfield JE. Cardiac rhythm in atrial isomerism. Am J Cardiol 1987; 59: 1156–8. Shenker L, Reed KL, Anderson CF et al. Congenital heart block and cardiac anomalies in the absence of maternal

57.

58.

59.

60.

61.

62.

63.

64.

65.

66.

67.

68. 69.

connective tissue disease. Am J Obstet Gynecol 1987; 157: 248–53. Baschat AA, Gembruch U, Knopfle G, Hansmann M. First-trimester fetal heart block: a marker for cardiac anomaly. Ultrasound Obstet Gynecol 1999; 4: 311–14. Berg C, Geipel A, Kohl T et al. Atrioventricular block detected in fetal life: associated anomalies and potential prognostic markers. Ultrasound Obstet Gynecol 2005; 26: 4–15. Jaeggi ET, Hornberger LK, Smallhorn JF, Fouron JC. Prenatal diagnosis of complete atrioventricular block associated with structural heart disease: combined experience of two tertiary care centers and review of the literature. Ultrasound Obstet Gynecol 2005; 26: 16–21. Ho SY, Anderson RH, Allan L. Disposition of the atrioventricular conduction tissues in the heart with isomerism of the atrial appendages: its relation to congenital complete heart block. J Am Coll Cardiol 1992; 20: 904–10. Wu MH, Wang JK, Lin JL et al. Supraventricular tachycardia in patients with right atrial isomerism. J Am Coll Cardiol 1998; 32: 773–9. Lee SE, Kim HY, Jung SE et al. Situs anomalies and gastrointestinal abnormalities. J Pediatr Surg 2006; 41: 1237–42. Chang J, Brueckner M, Touloukian RJ. Intestinal malrotation and fixation abnormalities in heterotaxia: early detection and management. J Pediatr Surg 1993; 28: 1281–5. Choi M, Borenstein H, Hornberger L, Langer JC. Heterotaxy syndrome: the role of screening for intestinal rotation abnormalities. Arch Dis Child 2005; 90: 813–15. Ticho BS, Goldstein AM, Van Praagh R. Extracardiac anomalies in the heterotaxy syndromes with focus on anomalies of midline-associated structures. Am J Cardiol 2000; 85: 729–34. Salomon LJ, Baumann C, Delezoide AL et al. Abnormal abdominal situs: what and how should we look for? Prenat Diagn 2006; 26: 282–5. Carmi R, Magee CA, Neill CA, Karrer FM. Extrahepatic biliary atresia and associated anomalies: etiologic heterogeneity suggested by distinctive patterns of associations. Am J Med Genet 1993; 45: 683–93. Murray CP, Yoo SJ, Babyn PS. Congenital extrahepatic portosystemic shunts. Pediatr Radiol 2003; 33: 614–20. Friedberg MK, Ursell PC, Silverman NK. Isomerism of the left atrial appendage associated with ventricular noncompaction. Am J Cardiol 2005; 96: 985–90.

25 Diseases of the myocardium, endocardium, and pericardium during fetal life Paulo Zielinsky, Antonio Piccoli Jr, João Luiz Manica, and Luiz Henrique Nicoloso

Introduction Fetal cardiology is a rapidly expanding subspecialty, following the advances obtained in fetal imaging and the growing research on prenatal physiology and pathophysiology. Diseases of the fetal myocardium, as well as those affecting the endocardium and the pericardium, show a wide variety of causes and several forms of presentation. These fetal anomalies may occur as a primary disorder, with no definitive etiologic factors, but in most cases they are the consequence of other systemic abnormalities or maternal diseases. The purpose of this chapter is to review some of the present knowledge on the diagnosis and management of these disorders during fetal life.

Diseases of the myocardium Dilated cardiomyopathy Dilated cardiomyopathy is a myocardial disease not usually associated with structural or pericardial abnormalities in the fetus. Global myocardial insufficiency occurs as a result of dysfunction of isolated cardiomyocytes. For this reason, the complexity within each cell determines the site of cellular function impairment. Dilated cardiomyopathy may affect the right ventricle, the left ventricle, or both ventricles during fetal life. Dilated cardiomyopathy in the fetus may occur in situations in which there is high output heart failure as a consequence of fetal anemia or volume overload, such as in large arteriovenous shunts. Severe fetal anemia increases cardiac output in order to balance myocardial oxygen demand; with the increase in heart workload and oxygen consumption, ischemic changes occur, eventually leading to a decrease in myocardial function and consequent

dilatation of cardiac chambers. In the presence of large arteriovenous shunts, such as arteriovenous malformations, there is an important increase in the venous return to the heart, leading to volume overload. Ejection fraction and stroke volume are increased, and so is oxygen consumption. Other possible presentations of dilated cardiomyopathy are secondary to direct myocardial lesions, such as fetal infections, hypoxia, and exposure to toxins. Infections during intrauterine life represent an important group of diseases which can cause direct injury to fetal cardiomyocytes, leading to progressive dilated cardiomyopathy. Coxsackievirus, Parvovirus B19, Toxoplasma gondii, and Herpesvirus type I are known etiological agents of fetal myocarditis and consequent heart failure.1–3 Vertical transmission of human immunodeficiency virus type 1 (HIV-1) may also be a cause of prenatal cardiomyopathy, but direct cardiomyocyte infection has not yet been demonstrated in the fetus.4 Fetal hypoxia occurs in several situations, and may be responsible for direct myocardial injury and heart failure. The fetus reacts to hypoxia through vasoconstriction, bradycardia, and arterial hypertension. With progressive worsening of the hypoxic state, there is hypotension, ischemia, and myocardial cellular lesion, resulting in eventual cardiac decompensation.1 Toxic cardiomyopathy may also occur in the fetus, as a result of direct myocardial injury caused by maternal external agents, such as cocaine, antineoplastic drugs, and antiviral chemotherapy. Fetal dilated cardiomyopathy secondary to rhythm disorders may also be called arrhythmia-induced cardiomyopathy. The most common presentation is sustained supraventricular tachycardia. The rise in heart rate shortens diastole and decreases myocardial blood flow, leading to structural cardiomyocyte damage and consequent significant ventricular dysfunction.5 In the fetal sheep model, fetal hydrops usually occurs when the heart rate is over 300 beats per minute, as a result of a decrease in stroke volume and diastolic dysfunction.6 In the human

364

Fetal Cardiology

Figure 25.1 Dilated cardiomyopathy. A 33-week fetus with coxsackievirus maternal infection. The heart is enlarged and there is mitral and tricuspid regurgitation. LV, left ventricle; RV, right ventricle; LA, left atrium; RA, right atrium; MR, mitral regurgitation; TR, tricuspid regurgitation.

fetus, functional atrioventricular regurgitation, caused by the enlargement of tricuspid or mitral rings, may be a sign of tachycardia-induced cardiomyopathy, with severe ventricular function compromise.7 Significant bradycardia in fetuses with complete atrioventricular block may also lead to myocardial failure, especially when there is maternal serum positivity for anti-SSA or anti-Ro antibodies, which are related to collagen disease. These antibodies are associated with fetal conduction tissue fibrosis and also possibly with direct myocardial damage.8 Another possible structural substrate for fetal dilated cardiomyopathy is congenital aneurysm of the left ventricle. Its etiology remains unknown, but it has been associated with myocardial fibrosis, myocarditis, ischemic myocardial damage, and hereditary factors. It is usually an isolated lesion, located near the apex, but may involve the papillary muscles and the subaortic region.9 The initial echocardiographic approach to fetal dilated cardiomyopathy includes the determination of the presence of cardiomegaly. The cardiothoracic index is obtained by the ratio between transverse diameters of the heart and the thorax (normal: 0.3–0.4), or the ratio between cardiac and thoracic circumferences (normal: 0.5), or else the ratio between cardiac and thoracic areas (normal: 0.33). Other possible fetal echocardiographic features are dilatation of the cardiac chambers, especially the left atrium and the left ventricle, left ventricular contractility deficit, increased systolic and diastolic volumes, atrioventricular valve regurgitation, decreased ejection fraction, stroke volume and shortening fraction, impaired diastolic relaxation, and secondary endocardial fibroelastosis

(Figure 25.1). A shortening fraction of less than 0.25 is considered abnormal, as well as an ejection fraction lower than 0.57. A hyperdynamic left ventricle usually presents a shortening fraction above 0.35. A decrease in systolic function is the earliest echocardiographic finding in fetal dilated cardiomyopathy. Although systolic dysfunction and significant atrioventricular valve regurgitation are risk factors for mortality, diastolic dysfunction is associated with the greatest risk of mortality.3 Diastolic function (compliance and relaxation) may be assessed by transmitral flow analysis (peak E and A velocities, deceleration time, and time–velocity integral).10,11 Alternative parameters for the assessment of fetal ventricular diastolic function, proposed by our group, are the mobility of the septum primum,12,13 pulmonary vein,14,15 ductus venosus,16 and foramen ovale flow pulsatility indexes, and tissue Doppler motion analysis of the ventricular walls.17 Venous flow alterations, such as umbilical vein pulsatility, increase of reverse ‘a’ wave in hepatic veins and inferior vena cava, or reverse pre-systolic flow in the ductus venosus, may be related to right ventricular dysfunction in dilated cardiomyopathy. Recently, quantification of myocardial dysfunction and prognosis assessment in fetal heart failure have been proposed, using the Tei index (isovolumetric time/ejection time)18 and a cardiovascular score that considers fetal hydrops, venous Doppler, heart size, cardiac function, and arterial Doppler. This cardiovascular profile score gives a semiquantitative score of fetal cardiac well-being and uses known markers by ultrasound that have been correlated with poor fetal outcome.19 The prognosis is particularly poor for hydropic fetuses. Suspected dilated cardiomyopathy in fetal life can be the consequence of a heterogeneous group of disorders, and has a varied presentation. Detailed and repeated fetal echocardiography is indicated to exclude occult obstructive cardiovascular lesions, and to monitor hemodynamics. This should be combined with appropriate maternal, fetal, and postnatal investigations.

Hypertrophic cardiomyopathy An increase in right or left ventricular wall thickness in the fetus is not infrequent. Even though this finding can be associated with an increase in afterload of one or both ventricles, such as semilunar valve obstruction, ductal constriction, or fetal systemic hypertension, it is the primary form of myocardial hypertrophy that may be considered a myocardial disease, or hypertrophic cardiomyopathy. The molecular basis for cardiac hypertrophy is not fully understood, but the integrin-linked kinase, a multifunctional protein kinase that physically links β-integrins with the actin cytoskeleton, was suggested to play a mechanoreceptor role.20

Diseases of the myocardium, endocardium, and pericardium

365

Figure 25.2 Hypertrophic cardiomyopathy. A 30-week fetus whose mother had preexisting insulin-dependent diabetes. The ventricular septum and the left ventricular posterior wall show increased thickness. Left: M-mode tracing. Right: two-dimensional imaging. IVS, interventricular septum.

Hypertrophic cardiomyopathy may be a part of fetal genetic disorders and dysmorphic syndromes,21 such as Noonan syndrome.22 A severe form presenting in fetal life is the hypertrophic cardiomyopathy caused by isolated cytochrome-c oxidase deficiency.23 Another possible presentation of prenatal hypertrophic cardiomyopathy is familial disease, with autosomic dominance, in which histologic findings include severe myocardial disarray; it is characterized by asymmetrical septal hypertrophy, often showing left ventricular outflow obstruction, and is exceedingly rare during fetal life.3 Recipient twin hypertrophic cardiomyopathy observed in intertwin transfusion syndrome is a particular form of presentation, being a serious complication of monozygotic, monochorionic, diamniotic twins, and is a direct result of transplacental communication between the circulations of the twins. In this syndrome, blood is thought to be shunted from the twin-donor, who develops anemia, growth restriction, and oligohydramnios, to the twin-recipient, who becomes plethoric and macrosomic, and develops polyhydramnios. The incidence of twin–twin transfusion syndrome ranges from 5 to 15% of all twin pregnancies. If this condition develops in the second trimester, it is usually associated with spontaneous abortion and death of one or both fetuses before viability. Developing the syndrome in the third trimester has better perinatal outcome. Mortality rates range from 56 to 100%, depending on gestational age and severity of the syndrome. The ultrasound criteria for diagnosis are the presence of twins of the same sex with discordant growth, with oligohydroamnios in one twin sac and polyhydramnios in the other, one placenta, and thin membrane between the twins.

Twin–twin transfusion is an extrinsic cause of fetal cardiomyopathy. Pathogenesis of the cardiovascular manifestation remains unclear. The release of growth factors and vasoactive peptides by the placenta to the recipient twin has been implicated. The receptor fetus may show signs of dilated cardiomyopathy, with marked cardiomegaly, mitral and tricuspid regurgitation, and hypocontractility. At a later stage, hypertrophic cardiomyopathy is predominant, with important septal hypertrophy and diastolic dysfunction.24 Usually, there is postnatal regression of the disease, as observed in myocardial hypertrophy in infants of diabetic mothers. Serial amniocentesis and endoscopic laser ablation of placental vascular anastomoses may improve the outcome of affected pregnancies.3 However, the most frequent presentation in clinical practice is the myocardial hypertrophy observed in fetuses of diabetic mothers. Fetal myocardial hypertrophy is present as a complication of gestational or previous maternal diabetes in about 25–30% of cases.25 The ventricular septum is preferentially affected, but both right and left ventricular free walls may be involved, the left more than the right26 (Figure 25.2). Hypertrophy is easily detected by means of standard fetal echocardiography, usually by comparison of the septal thickness with established nomograms,27 a septal thickness above two standard deviations for gestational age being considered abnormal. Histologic features include increases in nuclear and sarcolemmal mass, as well as vacuolization and hydrops of hyperplastic myocardial cells.28,29 The etiology of myocardial hypertrophy in fetuses of diabetic mothers remains controversial. Fetal hyperinsulinism has long been suggested,30 but an actual association between hypertrophic cardiomyopathy and high levels of amniotic fluid

366

Fetal Cardiology

insulin in fetuses of diabetic mothers has only recently been demonstrated.31 Even though macrosomia is a very common finding in neonates of diabetic mothers, no association has been found with the development of fetal septal hypertrophy.26 On the other hand, it was shown that the increase of fetal ventricular septum thickness throughout pregnancy is associated with an increase in levels of insulin growth factor-1. Cardiac function in fetuses of diabetic mothers is reduced when preconceptual maternal glycated hemoglobin is increased.32 Several studies have demonstrated that left ventricular diastolic function is impaired in fetuses with myocardial hypertrophy.33,34 Classical echocardiographic assessment of fetal diastolic function uses Doppler analysis of mitral and tricuspid inflow waves. The waveforms obtained in diastole by pulsed Doppler echocardiography at the tip of both atrioventricular valves are biphasic, with an E-wave representing early ventricular filling velocity, and an A-wave related to flow velocity during atrial contraction in presystole. The normal E/A ratio during pregnancy is below 1, which means that the fetal myocardium is relatively ‘stiff ’ compared to that of newborns and older children. An increase or inversion of the E/A ratio is related to ventricular diastolic dysfunction.11,33,34 Our group reported alternative methods to assess ventricular diastolic function during fetal life using models of decreased ventricular compliance. Specifically, we showed that in fetuses of diabetic mothers: (1) the septum primum mobility is decreased13 and does not depend on the diameter of the foramen ovale;35 (2) the global shortening of the left atrium is decreased;36 (3) the pulsatility index of the pulmonary veins is increased;14 (4) the foramen ovale flow impedance is increased;37 (5) the ductus venosus pulsatility index is increased;16 and (6) the isthmus flow index is decreased. Nevertheless, in utero clinical presentation is usually not striking. Hypertrophic cardiomyopathy in neonates of diabetic pregnancies is a transient disorder, with spontaneous regression of increased septal thickness during the first 6 months of postnatal life, related to the normalization of insulin levels.38 However benign, this disorder may be a cause for neonatal cardiomegaly and respiratory distress, secondary to poor left ventricular compliance, which stresses the need for adequate prenatal assessment of diastolic function.

Restrictive cardiomyopathy and non-compacted cardiomyopathy Restrictive cardiomyopathy is the less frequent presentation of myocardial disease in the fetus.39 While in other forms of cardiomyopathy diagnosis is based on ventricular chamber morphological features, restrictive cardiomyopathy is recognized by hemodynamic alterations.40 Classically, its pathophysiological features include normal

or mildly altered ventricular size, normal contractility, and abnormal diastolic function, with rapid ventricular filling in early diastole and almost no filling in the subsequent diastolic period.41,42 In its primary form, the prognosis is somber, with fetal congestive heart failure and hydrops. In children, there is early pulmonary hypertension.43 Survival is about 50% in the first 2 years after diagnosis, and patients are readily referred to transplantation programs.39,44,45 However, a restrictive profile may be found in other forms of myocardial disease not classified as restrictive cardiomyopathy. An example is endocardial fibroelastosis, a common response of the cardiac tissue to obstructive lesions46,47 and other fetal insults,48–50 and another is ventricular non-compaction, a myocardial malformation extensively studied in recent decades, formerly known as spongiform myocardium.51 Ventricular non-compaction is considered by the World Health Organization as a distinctive form of ‘unclassified cardiomyopathy’, not included in the classical dilated, hypertrophic and restrictive forms.52 It is more frequently associated with significant systolic dysfunction, but it may appear with a severe restrictive pattern of diastolic dysfunction and preserved systolic function. In such cases, echocardiographic findings are similar to those observed in restrictive cardiomyopathy, namely dilatation of atrial chambers, minimal increase of ventricular chambers, and normal or mildly decreased myocardial contractility, but with a spongiform aspect of the cardiac muscle.51 Formerly called spongy myocardium, spongiform cardiomyopathy, hypertrabeculation, or persistent myocardial sinusoids (because of the similarity of the myocardial wall to sinusoids present in pulmonary atresia with intact interventricular septum), ventricular non-compaction is characterized by prominent, thick, and numerous trabeculations in one or more segments of the ventricular wall. These trabeculations alternate with deep intertrabecular recesses covered by endothelium of the ventricular chamber, forming an extensive trabecular layer. Different from sinusoids, the recesses in non-compacted cardiomyopathy do not connect with the coronary arteries.53 It is common knowledge that in vertebrate embryologic development, during early gestation, there is a predominance of the trabecular layer over the ventricular wall non-compacted layer, allowing blood flow and an increase in cardiac mass previous to the appearance of the coronary circulation. Based on this fact, the most accepted theory for the genesis of noncompacted cardiomyopathy is an interruption of normal endocardial and myocardial embryogenesis, with persistence of the trabecular portion and predominance over the compacted portion in fetuses at a later gestation period and after birth. In addition to the isolated form, ventricular non-compaction may also be associated with several congenital heart diseases. It is, however, important to differentiate an ischemic and hypertrophic myocardial response to valvar stenosis or atresia with intact interventricular septum, with echocardiographic aspects often

Diseases of the myocardium, endocardium, and pericardium

367

by flow from the ventricular cavity during Doppler insonation55 (Figure 25.3). The ratio between noncompacted and compacted layers is higher than 0.5. During fetal life, ventricular non-compaction may be an occasional finding at routine prenatal ultrasound screening, but the fetus may be bradycardic or hydropic. The prognosis of prenatally detected ventricular non-compaction is poor, especially when associated with fetal hydrops. In these cases, most do not survive until late gestation, and preterm interruption of pregnancy is the rule. After birth, when there is early onset of congestive heart failure, very rarely ventricular contractility improves to near normal levels.

Diseases of the endocardium Endocardial fibroelastosis Figure 25.3 Non-compacted cardiomyopathy. A 36-week fetus with ventricular non-compaction. There are several trabeculations with deep intertrabecular recesses filled by flow from the ventricular cavity during Doppler insonation.55 AO, aorta; S, septum.

similar to ventricular non-compaction.51 The prevalence of non-compacted cardiomyopathy is uncertain. Previous studies have estimated a prevalence of about 0.05–0.24% of patients referred to reference centers, but this number may be underestimated.53 Ventricular non-compaction may occur in a familial sporadic form. Familial recurrence is well documented in some case series,51,54–56 suggesting a genetic basis for this pathology. An autosomic dominant transmission form has been reported, but there are also mentions of X-linked inheritance, and gene mutations (G4.5, Cypher/ZASP, 11p15, lamin A/C) have also been described in association with ventricular non-compaction. Other associations include facial dysmorphies and syndromes, such as Barth, Noonan, Roifman, Melnick– Needles, nail–patella, and Toriello–Carey.51,53–55 Diagnosis of ventricular non-compaction by means of fetal echocardiography has been described as early as 22 weeks’ gestational age. Left ventricular involvment seems more frequent postnatally, but during fetal life the right ventricle is usually more affected, as a result of the particular fetal circulatory dynamics, with right ventricular dominance. Ventricular function might be normal, but as a rule there is diastolic dysfunction with a restrictive pattern, with ellipsoid ventricles. Systolic dysfunction may also occur, with more spheric ventricles, but, different from dilated cardiomyopathy, remaining with thick-walled non-dilated chambers. Diagnosis is made by the presence of numerous trabeculations with deep intertrabecular recesses filled

Endocardial fibroelastosis is a cardiac disorder characterized by diffuse endocardial thickening that affects mainly the left ventricle, determining the proliferation of endocardial elastic and collagen fibers, thus decreasing the ventricular compliance and stroke volume. The fibroelastic reaction, with regular layers of elastic tissue at the endocardial level, seems to occur during fetal development and growth, continuing after birth throughout early infancy.57 There are many controversies related to the causes and natural history of endocardial fibroelastosis. The etiology is probably multifactorial, histological alterations being the final stage of many conditions affecting the endocardium and myocardium. A possible non-specific endocardial response to situations of prenatal48 or immediate postnatal49 myocardial stress, such as maternal viral infections, hypoxia, vascular disease, and chromosomal abnormalities, has been suggested. The presence of endocardial fibroelastosis in growthdiscordant monozygotic twins has recently been reported.58 Secondary endocardial fibroelastosis may also be associated with fetal infections (Coxsackie virus and Parvovirus myocarditis), genetic disease with recessive autosomic transmission, X-linked cardiomyopathy, and metabolic disorders. Endocardial fibroelastosis may be consequent to many structural cardiac defects, but in these cases it is not considered a primary disorder. During fetal development, endocardial fibroelastosis may allow evolutive changes of the left ventricle, with a dilated form preceding the restrictive form.46,59 Accordingly, the echocardiographic appearance may initially demonstrate left ventricular hypocontractility and dilatation, along with a hyperechogenic thickening of the endocardial surface.60 As gestational age advances, the left ventricular cavity gradually decreases in size and there is progressive left ventricle wall hypertrophy and an increase in

368

Fetal Cardiology

hyperechogenicity of the endocardial surface. The left atrium is usually significantly enlarged. At this stage, histological examination of the left ventricular endocardium confirms endocardial thickening and the presence of subendocardial calcifications.57 The prognosis of fetal endocardial fibroelastosis is poor. This disorder, in its primary or secondary form, may be the cause of congestive heart failure and fetal hydrops.48 Moreover, 80% of children born with endocardial fibroelastosis present congestive heart failure during the first year of life.

Endocardial echogenic foci Endocardial echogenic foci (EEF), also known as ‘golf balls’, are brightly echogenic areas within the fetal heart. The etiology of EEF seems to be related to calcification within the papillary muscle,61 which has been proposed to be due to abnormal development of the microvasculature, leading to ischemic changes in the muscle. Furthermore, incomplete fenestration of the papillary muscle and the chordae tendineae due to individual variations could be the explanation for EEF in normal fetuses.62 The reported prevalence of EEF has been described at 0.5–20.3%, depending on the population and the methodology. In low-risk populations, EEF have been described in 3–5% of cases.63,64 According to a review article, in a total of 13 493 pregnancies screened, 334 (2.5%) of fetuses with EEF had been described until that time.65 EEF appear on fetal cardiac ultrasound as a structure within the fetal heart close to the papillary muscles and chordae tendineae, which moves with the valve leaflets during cardiac cycle, measures from 1 to 6 mm in diameter without distal acoustic shadowing, and may be as echogenic as the surrounding bone63–66 (Figure 25.4). Multiple foci are an infrequent finding (6–11%).66,67 The left ventricle is the most frequent site for EEF. There have been demonstrations that the foci might remain unchanged68 or have complete resolution before delivery63 or on neonatal ultrasound.69 In the last few years there has been a concern that EEF could increase the risk of trisomy 21 or trisomy 13, mainly when foci are within both ventricular chambers.70 Some studies have confirmed the high prevalence of the EEF in fetuses with aneuploidy.71,72 It was suggested that EEF could be an additional antenatal marker of chromosomal abnormality.62 However, such findings have not been described in low-risk populations. In a series of 9263 lowrisk pregnancies, none revealed fetuses with EEF associated with trisomy 21.63,64,73,74 Since EEF are non-specific findings and encountered in both normal and abnormal fetuses, their presence in isolation is nowadays not considered of any value in identifying an increased risk for chromosomal abnormalities.75,76

Figure 25.4 Endocardial echogenic focus. Routine echocardiographic four-chamber view in a 24-week fetus. There is an echogenic focus (‘golf ball’) within the left ventricle, adjacent to the mitral subvalvar apparatus (arrow).

There is little evidence relating an association between EEF and more serious defects in the fetal or neonatal heart. Some series have reported cardiac defects such as ventricular septal defect (0.4%),67 transposition of the great vessels (4%),73 poor contractility, and abnormal outflow tracts.77 On the other hand, there are several studies showing no relation between EEF and cardiac defects on prenatal and postnatal ultrasound.62,68,72,74 Our group have retrospectively assessed 14 100 fetuses studied by prenatal echocardiography during the period 1997–2004 for the presence of EEF. ‘Golf balls’ were identified in 487 (3.4%) fetuses. In 456 (93.8%), no association with structural or functional heart abnormalities was seen, and only two fetuses (0.4%) were shown to have trisomy 21.78 EEF should be distinguished from other causes of echogenic areas within the fetal heart. Diffuse echogenic areas of the endocardium in fetuses with EEF suggest a wide range of myocardial or endocardial pathological processes, such as cardiac tumors, structural heart disease, or endocardial fibroelastosis, which may be associated with metabolic disorders or viral infection.57 Our group have found a high prevalence of EEF (91.4%) in fetuses of mothers with high titers of antitoxoplasmosis antibody and other infectious agents.78 Because of these data, it remains controversial whether the finding of EEF should prompt detailed fetal echocardiography. In pregnancies at high risk for chromosomal abnormalities, the presence of EEF in association with other markers, such as increased nuchal transluscency, choroid plexus cyst, hyperechogenic bowel, short bones,

Diseases of the myocardium, endocardium, and pericardium

Figure 25.5 Pericardial effusion. A 22-week fetus with cytogenetic diagnosis of Down syndrome. There are no intracardiac defects and a small pericardial effusion (PE) is disclosed (arrows).

and pyeloectasis, might be a possible additional marker for aneuploidy. In such cases, maternal screening for chromosomal abnormalities could be considered.79–81 On the other hand, in low-risk populations, the presence of ‘golf balls’ is considered a normal finding, and does not require further investigation.82–84

Diseases of the pericardium Pericardial effusion Small amounts of pericardial fluid during routine prenatal ultrasound are not uncommon, being observed in 44% of normal fetuses.85 The presence of excessive pericardial fluid (pericardial effusion) should be considered a probable marker of systemic abnormality leading to fetal hydrops. Frequently, pericardial effusion is associated with structural heart disease, arrhythmias, viral infections and fetal hydrops, with subsequent poor prognosis. Constriction of the ductus arteriosus, as a result of maternal administration of indomethacin or related drugs, may also be a cause of pericardial effusion,86 in which case it is accompanied by an enlarged right ventricle, tricuspid regurgitation, and high velocity ductal flow during systole and diastole. Pericardial effusion and even tamponade have been described as a result of rupture of ventricular aneurysms or diverticuli.87–91 In these cases, immediate emptying of the pericardial sac may be life-saving. Other causes of pericardial effusion are pericardial teratoma,92 cystic lymphangioma,93 atrial hemangioma,94 pericardial rhabdomyoma,95 and epicardial angiofibroma.96 Pericardial effusion is easily detected by fetal echocardiography. An anechoic region larger than 2 mm separating

369

the pericardial layers may be observed (Figure 25.5). There are cases with a pericardial effusion larger than 2 mm with no anatomical or functional abnormalities. Usually, pericardial effusion is observed close to an atrioventricular valve or at the length of a ventricle, rarely around the whole heart. Difficulty in pericardial effusion characterization arises when the myocardial periphery is confused with pericardial fluid, because of the presence of circular fibers giving this region an anechoic aspect.97 In low-risk gestations and in the absence of other ultrasonic abnormalities, an isolated pericardial effusion up to 7 mm has not been associated with poor neonatal outcome. For this reason, attentive fetal monitoring throughout pregnancy after complete echocardiographic and obstetric ultrasound evaluation in order to exclude cardiac or extracardiac abnormalities is usually sufficient. On the other hand, detection of a heart defect or other serous effusions (pleural effusion, ascites) is a formal indication for fetal karyotyping by amniocentesis or cordocentesis, because of the high risk of chromosomal abnormalities.98

Therapeutic considerations The majority of the disorders addressed in this chapter will have to be treated in the neonatal period, after an assisted delivery with the presence of both the neonatologist and the pediatric cardiologist. Thus, intensive care of the neonate with congestive heart failure will include the use of inotropics, diuretics, vasodilators, assisted ventilation, and other routine measures. On the other hand, there are situations in which a prenatal therapeutic approach may be life-saving. Some examples include the reversion of lifethreatening supraventricular arrhythmias by means of antiarrhythmic drug administration, via the maternal route or through a cordocentesis, or serial intrauterine transfusions in severely hydropic fetuses with high output cardiac failure as a result of Rh isoimmunization, or else pericardial drainage in a fetus with cardiac tamponade. It is possible that emptying ascites and pericardial or pleural effusions might improve the outcome of fetuses with congestive heart failure. Other therapeutic invasive procedures that have been described are fetal dilatation of severely stenotic aortic valves,99–105 the creation of interatrial communications in restrictive atrial septa,106 and percutaneous implantation of a ventricular pacemaker.107

Conclusion As in many other fields related to fetal medicine, there is still much to be learned about the etiology, pathophysiology,

370

Fetal Cardiology

diagnosis, and management of disorders affecting fetal myocardium, endocardium, and pericardium. Advances in genetics, molecular biology, and other related areas will undoubtedly in the upcoming years shed more light on the basic science of these diseases, at the same time that improvements of ultrasound capability and device manipulation will allow progress in detecting and treating fetal cardiac diseases.

References 1. Drose JA, Dennis MA, Thickman D. Infection in utero: ultrasonographic findings in 19 cases. Radiology 1991; 178: 369–74. 2. Soulie JC. Cardiac involvement in fetal parvovirus B19 infection. Pathol Biol (Paris) 1995; 43: 416–19. 3. Pedra SR, Smallhorn JF, Ryan G et al. Fetal cardiomyopathies: pathogenic mechanisms, hemodynamic findings, and clinical outcome. Circulation 2002; 106: 585–91. 4. Rebolledo MA, Krogstad P, Chen F et al. Infection of human fetal cardiac myocytes by a human immunodeficiency virus1-derived vector. Circ Res 1998; 83: 738–42. 5. Tomita M, Spinale FG, Crawford FA, Zile MR. Changes in left ventricular volume, mass and function during the development and regression of supraventricular tachycardiainduced cardiomyopathy: disparity between recovery of systolic versus diastolic function. Circulation 1991; 83: 635–44. 6. Rudolph AM, Heyman MA. Cardiac output in the fetal lamb: the effects of spontaneous and induced changes of heart rate on right and left ventricular output. Am J Obstet Gynecol 1976; 124: 183–92. 7. Krapp M, Gembruch U, Baumann P. Venous blood flow pattern suggesting tachycardia-induced cardiomyopathy in the fetus. Ultrasound Obstet Gynecol 1997; 10: 323–40. 8. Salomonsson S, Sonesson SE, Ottosson L et al. Ro/SSA autoantibodies directly bind cardiomyocytes, disturb calcium homeostasis, and mediate congenital heart block. J Exp Med 2005; 201: 11–17. 9. El Kady D, Gerscovich EO, Moon-Grady et al. Congenital cardiac left ventricular aneurysm with pericardial effusion: early prenatal diagnosis and intervention. J Ultrasound Med 2005; 24: 1011–15. 10. Reed KL, Sahn DJ, Scagnelli S et al. Doppler echocardiographic studies of diastolic function in the human fetal heart: changes during gestation. J Am Coll Cardiol 1986; 8: 391–5. 11. Tsyvian P, Malkin K, Artemieva O, Wladimiroff JW. Assessment of left ventricular filling in normally grown fetuses, growth-restricted fetuses and fetuses of diabetic mothers. Ultrasound Obstet Gynecol 1998; 12: 33–8. 12. Firpo C, Zielinsky P. Mobility of the flap valve of the primary atrial septum in the developing human fetus. Cardiol Young 1998; 8: 67–70. 13. Firpo C, Zielinsky P. Behaviour of septum primum mobility in third-trimester fetuses with myocardial hypertrophy. Ultrasound Obstet Gynecol 2003; 21: 445–50.

14. Zielinsky P, Piccoli AL Jr, Teixeira L et al. Pulmonary vein pulsatility in fetuses of diabetic mothers: prenatal Doppler echocardiographic study. Arq Bras Cardiol 2003; 81: 604–7. 15. Zielinsky P, Piccoli A Jr, Gus E et al. Dynamics of the pulmonary venous flow in the fetus and its association with vascular diameter. Circulation 2003; 108: 2377–80. 16. Zielinsky P, Marcantonio S, Nicoloso LH et al. Ductus venosus flow and myocardial hypertrophy in fetuses of diabetic mothers. Arq Bras Cardiol 2004; 83: 51–6. 17. Zielinsky P, Nicoloso LH, Firpo C et al. Alternative parameters for echocardiographic assessment of fetal diastolic function. Braz J Med Biol Res 2004; 37: 31–6. 18. Huhta JC. Guidelines for the evaluation of heart failure in the fetus with or without hydrops. Pediatr Cardiol 2004; 25: 274–86. 19. Falkensammer CB, Paul J, Huhta JC. Fetal congestive heart failure: correlation of Tei-index and cardiovascular score. J Perinat Med 2001; 29: 390–8. 20. Lu H, Fedak PW, Dai X et al. Integrin-linked kinase expression is elevated in human cardiac hypertrophy and induces hypertrophy in transgenic mice. Circulation 2006; 114: 2271–9. 21. Chen CP, Chern SR, Lee CC et al. De novo unbalanced translocation resulting in monosomy for proximal 14q and distal 4p in a fetus with intrauterine growth retardation, Wolf-Hirshhorn syndrome, hypertrophic cardiomyopathy, and partial hemihypoplasia. J Med Genet 1998; 35: 1050–3. 22. Achiron R, Heggesh J, Grisaru D et al. Noonan syndrome: a cryptic condition in early gestation. Am J Med Genet 2000; 92: 159–65. 23. Venditti CP, Harris MC, Huff D et al. Congenital cardiomyopathy and pulmonary hypertension: another fatal variant of cytochrome-c oxidase deficiency. J Inherit Metab Dis 2004; 27: 735–9. 24. Raboisson MJ, Fouron JC, Lamoureux J et al. Early intertwin differences in myocardial performance during the twin-to-twin transfusion syndrome. Circulation 2004; 110: 3043–8. 25. Zielinsky P, Hagemann L, Daudt L et al. A pre and postnatal analysis of factors associated with fetal myocardial hypertrophy in diabetic pregnancies. J Matern Fetal Investig 1992; 2: 163–7. 26. Zielinsky P. Role of prenatal echocardiography in the study of hypertrophic cardiomyopathy in the fetus. Echocardiography 1991; 8: 661–7. 27. Allan LD. Manual of Fetal Echocardiography. Lancaster: MTP Press, 1986. 28. Miller HC, Wilson HM. Macrosomia, cardiac hypertrophy, erythroblastosis and hyperplasia of the islets of Langerhans in infants of diabetic mothers. J Pediatr 1943; 23: 251–66. 29. Menezes H, Barra M, Bello A et al. Fetal myocardial hypertrophy in an experimental model of gestational diabetes. Cardiol Young 2001; 11: 609–13. 30. Steinh J, Driscoll SG. The extractable insulin content of pancreas for fetus and infants of diabetic and control mothers. Diabetes 1965; 14: 573–7. 31. Zielinsky P, Hagemann LL. Fetal hyperinsulinism and hypertrophic cardiomyopathy: association in perinatal life. In: Imai Y, Mioma K, eds. Proceedings of the Second World

Diseases of the myocardium, endocardium, and pericardium

Congress of Pediatric Cardiology and Cardiac Surgery. New York: Futura Publishing, 1999: 112–14. 32. Gardiner HM, Pasquini L, Wolfenden J et al. Increased periconceptual maternal glycated haemoglobin in diabetic mothers reduces fetal long axis cardiac function. Heart 2006; 92: 1125–30. 33. Rizzo G, Arduini D, Romanini C. Accelerated cardiac growth and abnormal cardiac flow in fetuses of type I diabetic mothers. Obstet Gynecol 1992; 80: 369–76. 34. Pacileo G, Paladini D, Palmuri S et al. Echocardiographic assessment of diastolic function in normal human fetuses. J Perinat Med 1994; 22: 43–5. 35. Zielinsky P, Sallum M, Satler F et al. Mobility of the septum primum does not depend on the foramen ovale diameter in normal fetuses. Arq Bras Cardiol 2004; 83: 304–7. 36. Zielinsky P, Satler F, Luchese S et al. Study of global left atrial shortening in fetuses of diabetic mothers. Arq Bras Cardiol 2004; 83: 473–5; 470–2. 37. Zielinsky P, Nicoloso LH, Scheid MM et al. Behavior of the foramen ovale flow in fetuses of diabetic mothers with myocardial hypertrophy. Cardiol Young 2005; 15 (Suppl 2): 35 (abstr). 38. Zielinsky P, Costa MHL, Oliveira LT et al. Study of natural history of myocardial hypertrophy and its association of hyperinsulinism in infants of diabetic mothers. Arq Bras Cardiol 1997; 69: 1–6. 39. Russo L, Weber S. Idiopathic restrictive cardiomyopathy in children. Heart 2005; 91: 1199–202. 40. Ammash NM, Seward JB, Bailey KR et al. Clinical profile and outcome of idiopathic restrictive cardiomyopathy. Circulation 2000; 101: 2490–6. 41. Shabetai R. Pathophysiology and differential diagnosis of restrictive cardiomyopathy. Cardiovasc Clin 1988; 19: 123–32. 42. Seward JB. Restrictive cardiomyopathy: reassessment of definitions and diagnosis. Curr Opin Cardiol 1988; 3: 391–5. 43. Chen S, Balfour IC, Juroidini S. Clinical spectrum of restrictive cardiomyopathy in children. J Heart Lung Transplant 2001; 20: 90–2. 44. Fenton MJ, Chubb H, McMarron AM et al. Heart and heart-lung transplantation for idiopathic restrictive cardiomyopathy in children. Heart 2006; 92: 85–9. 45. Kimberling MT, Balzer DT, Hirsch R et al. Cardiac transplantation for pediatric restrictive cardiomyopathy: presentation, evaluation, and short-term outcome. J Heart Lung Transplant 2002; 21: 455–9. 46. Carceller A, Maroto E, Fouron JC. Dilated and contracted forms of primary fibroelastosis: a single fetal disease with two stages of development. Br Heart J 1990; 63: 311–13. 47. Rustico MA, Benettoni A, Bussani R et al. Early fetal endocardial fibroelastosis and critical aortic stenosis: a case report. Ultrasound Obstet Gynecol 1995; 5: 202–5. 48. Newbold MJ, Armstrong GR, Barson AJ. Endocardial fibroelastosis in infants with hydrops fetalis. J Clin Pathol 1991; 44: 576–9. 49. Lurie PR. Fibroelastosis is not a disease. Am J Cardiol 1988; 62: 468–70. 50. Angelov A, Kulova A, Gurdevsky M. Endocardial fibroelastosis. Clinicopathological study of 38 cases. Path Res Pract 1984; 178: 384–8.

371

51. Guntheroth W, Komarniski C, Atkinson W, Fligner C. Criterion for fetal primary spongiform cardiomyopathy: restrictive pathophysiology. Obstet Gynecol 2002; 5: 882–4. 52. Richardson P, McKeena W, Bristow M et al. Report of the 1995 World Health Organization/International Society and Federation of Cardiology task force on the Definition and Classification of Cardiomyopathies. Circulation 1996; 93: 841–2. 53. Freedom R, Yoo SJ, Perrin D et al. The morphological spectrum of ventricular noncompaction. Cardiol Young 2005; 15: 345–64. 54. Moura C, Hillion Y, Dahmane F et al. Isolated noncompaction of the myocardium diagnosed in the fetus: two sporadic and two familial cases. Cardiol Young 2002; 12: 278–83. 55. Elshershari H, Okutan V, Celiker A. Isolated noncompaction of ventricular myocardium. Cardiol Young 2001; 11: 472–5. 56. Digilio MC, Marino B, Musolino AM et al. Genetic heterogeneity of isolated noncompaction of the left ventricular myocardium. Am J Med Genet 1999; 85: 90–1. 57. Tannouri F, Rypens F, Peny MO et al. Fetal endocardial fibroelastosis: ultrasonographic findings in two cases. J ultrasound Med 1998; 17: 63–6. 58. Lazda EJ. Endocardial fibroelastosis in growth-discordant monozygotic twins. Pediatr Dev Pathol 1998; 1: 522–7. 59. Hilton DA, Variend S, Pringle JH. Demonstration of Coxsackie virus RNA in formalin-fixed tissue sections from childhood myocarditis cases by in situ hybridization and the polymerase chain reaction. J Pathol 1993; 170: 45–51. 60. Veille JC, Sivakoff M. Fetal echocardiographic signs of congenital endocardial fibroelastosis. Obstet Gynecol 1988; 72: 219–22. 61. Brown DL, Roberts DJ, Miller WA. Left ventricular echogenic focus in the fetal heart: pathologic correlation. J Ultrasound Med 1994; 13: 613–16. 62. Roberts DJ, Genest D. Cardiac histologic pathology characteristic of trisomies 13 and 21. Hum Pathol 1992; 23: 1130–40. 63. Merati R, Lovotti M, Norchi S et al. Prevalence of fetal left ventricular hyperechogenic foci in a low risk population. Br J Obstet Gynaecol 1996; 103: 1102–4. 64. Dildy GA, Judd VE, Clark SL. Prospective evaluation of the antenatal incidence and postnatal significance of the fetal echogenic cardiac focus: a case-control study. Am J Obstet Gynecol 1996; 175: 1008–12. 65. Sepulveda W, Romero D. Significance of echogenic foci in the fetal heart. Ultrasound Obstet Gynecol 1998; 12: 445–9. 66. Simpson JM, Cook A, Sharland G. The significance of echogenic foci in the fetal heart: a prospective study of 228 cases. Ultrasound Obstet Gynecol 1996; 8: 225–8. 67. Petrikovsky B, Challenger M, Gross B. Unusual appearances of echogenic foci within the fetal heart: are they benign? Ultrasound Obstet Gynecol 1996; 8: 229–31. 68. Petrikovsky BM, Challenger M, Wyse LJ. Natural history of echogenic foci within ventricles of the fetal heart. Ultrasound Obstet Gynecol 1995; 5: 92–4. 69. Dildy GA, Judd VE, Clark SL. Prospective evaluation of the antenatal incidence and postnatal significance of the fetal

372

Fetal Cardiology

echogenic cardiac focus: a case-control study. Am J Obstet Gynecol 1996; 175: 1008–12. 70. Wax JR, Philput C. Fetal intracardiac echogenic foci: does it matter which ventricle? J Ultrasound Med 1998; 17: 141–4. 71. Lehman CD, Nyberg DA, Winter TC et al. Trisomy 13 syndrome: prenatal ultrasound findings in a review of 33 cases. Radiology 1995; 194: 217–22. 72. Sepulveda W, Cullen S, Nicolaidis P et al. Echogenic foci in the fetal heart: a marker of chromosomal abnormality. Br J Obstet Gynaecol 1995; 102: 490–2. 73. How HY, Villafane J, Parihus RR, Spinnato JA. Small hyperechoic foci of the fetal cardiac ventricle: a benign sonographic finding? Ultrasound Obstet Gynecol 1994; 4: 205–7. 74. Achiron R, Lipitz S, Gabbay U et al. Prenatal ultrasonographic diagnosis of fetal heart echogenic foci: no correlation with Down syndrome. Obstet Gynecol 1997; 89: 945–8. 75. Gonçalves TR, Zamith MM, Murta CG et al. Chromosomal and cardiac anomalies in fetuses with intracardiac echogenic foci. Int J Gynecol Obstet 2006; 95: 132–7. 76. Koklanaris N, Roman AS, Perle MA, Monteagudo A. Isolated echogenic intracardiac foci in patients with low-risk triple screen results: assessing the risk of trisomy 21. J Perinat Med 2005; 33: 539–42. 77. Bronshtein M, Jakovi P, Ofir C. Multiple fetal intracardiac echogenic foci: not always a benign sonographic finding. Prenat Diagn 1996; 16: 131–5. 78. Nicoloso LH, Henke T, Zielinsky P. Prenatal study of fetal endocardial hyperrefringence and its relation to maternal toxoplasmosis. Arq Bras Cardiol 2004; 82: 32–6. 79. Beke A, Papp C, Toth-Pal E et al. Trisomies and other chromosome abnormalities detected after positive sonographic findings. J Reprod Med 2005; 50: 675–91. 80. Van den Hof MC, Wilson RD; Diagnostic Imaging Committee; Society of Obstetricians and Gynaecologists of Canada; Genetics Committee, Society of Obstetricians and Gynaecologists of Canada. Fetal soft markers in obstetric ultrasound. J Obstet Gynaecol Can 2005; 27: 592–636. 81. Benacerraf BR. The role of the second trimester genetic sonogram in screening for fetal Down syndrome. Semin Perinatol 2005; 29: 386–94. 82. Carriço A, Matias A, Areias JC. How important is a cardiac echogenic focus in a routine fetal examination? Rev Port Cardiol 2004; 23: 459–61. 83. Coco C, Jeanty P, Jeanty C. An isolated echogenic heart focus is not an indication for amniocentesis in 12,672 unselected patients. J Ultrasound Med 2004; 23: 489–96. 84. Koklanaris N, Roman AS, Perle MA, Monteagudo A. Isolated echogenic intracardiac foci in patients with lowrisk triple screen results: assessing the risk of trisomy 21. J Perinat Med 2005; 33: 539–42. 85. Brown DL, DiSalvo DN, Frates FC et al. Sonography of the fetal heart: normal variants and pitfalls. AJR Am J Roentgenol 1993; 160: 1251–5. 86. Luchese S, Manica JL, Zielinsky P. Intrauterine ductus arteriosus constriction: analysis of a historical cohort of 20 cases. Arq Bras Cardiol 2003; 81: 405–10.

87. Davidson A, Whiteman V, Gaynor JW. Images in cardiovascular medicine. Fetal cardiac diverticulum. Circulation 2006; 113: e56. 88. Gardiner HM, Wimalasundera R, Pasquini L et al. Ho SY. Images in cardiovascular medicine. Pericardiocentesis at 14 weeks: effective treatment of pericardial effusion complicating right ventricular diverticulum. Circulation 2005; 112: e120. 89. Del Rio M, Martinez JM, Bennasar M et al. Prenatal diagnosis of a right ventricular diverticulum complicated by pericardial effusion in the first trimester. Ultrasound Obstet Gynecol 2005; 25: 409–11. 90. Prefumo F, Bhide A, Thilaganathan B, Carvalho JS. Fetal congenital cardiac diverticulum with pericardial effusion: two cases with different presentations in the first trimester of pregnancy. Ultrasound Obstet Gynecol 2005; 25: 405–8. 91. McAuliffe FM, Hornberger LK, Johnson J et al. Cardiac diverticulum with pericardial effusion: report of two new cases treated by in-utero pericardiocentesis and a review of the literature. Ultrasound Obstet Gynecol 2005; 25: 401–4. 92. Mitanchez D, Grebille AG, Parat S et al. Delayed surgery in pericardial teratoma with neonatal hydrops. Eur J Pediatr Surg 2005; 15: 431–3. 93. Hayashi A, Kikuchi A, Matsumoto Y, Daenen W. Massive cystic lymphangiomas of a fetus. Congenit Anom (Kyoto) 2005; 45: 154–6. 94. Laga S, Gewillig MH, Van Schoubroeck D, Daenen W. Imminent fetal cardiac tamponade by right atrial hemangioma. Pediatr Cardiol 2006; 27: 633–5. 95. Takeuchi M, Nakayama M, Kuwae Y et al. Huge pericardial rhabdomyoma with hypoplastic left ventricle and lung. Am J Perinatol 2006; 23: 89–92. 96. Satge D, Vidalo E, Desfarges F, de Geeter B. A third case of cardiac neoplasm in a fetus with Beckwith-Wiedemann syndrome: epicardial angiofibroma. Fetal Diagn Ther 2005; 20: 44–7. 97. Brown DL, Cartier MS, Emerson DS et al. The peripheral hypoechoic rim of the fetal heart. J Ultrasound Med 1989; 8: 603–8. 98. Bromley B, Lieberman E, Laboda L, Benacerraf BR. Echogenic intracardiac focus: a sonographic signal for fetal Down syndrome. Obstet Gynecol 1995; 86: 998–1001. 99. Allan LD, Maxwell DJ, Carminati M et al. Survival after fetal aortic balloon valvoplasty. Ultrasound Obstet Gynecol 1995; 5: 90–1. 100. Chaoui R, Bollmann R, Goeldner B, Rogalsky V. Aortic balloon valvuloplasty in the human fetus under ultrasound guidance: a report of two cases. Ultrasound Obstet Gynecol 1994; 4: 162A. 101. Lopes LM, Cha SC, Kajita LJ et al. Balloon dilatation of the aortic valve in the fetus – a case report. Fetal Diagn Ther 1996; 11: 296–300. 102. Kohl T, Sharland G, Allan LD et al. World experience of percutaneous ultrasound-guided balloon valvuloplasty in human fetuses with severe aortic valve obstruction. Am J Cardiol 2000; 85: 1230–3.

Diseases of the myocardium, endocardium, and pericardium

103. Tworetzky W, Wilkins-Haug L, Jennings RW et al. Balloon dilation of severe aortic stenosis in the fetus: potential for prevention of hypoplastic left heart syndrome: candidate selection, technique and results of succsseful intervention. Circulation 2004; 110: 2125–31. 104. Wilkins-Haug LE, Tworetzky W, Benson CB et al. Factors affecting technical success of fetal aortic valve dilation. Ultrasound Obstet Gynecol 2006; 28: 47–52. 105. Makikallio K, McElhinney DB, Levine JC et al. Fetal aortic valve stenosis and the evolution of hypoplastic left heart

373

syndrome: patient selection for fetal intervention. Circulation 2006; 113: 1401–5. 106. Marshall AC, van der Velde ME, Tworetzky W et al. Creation of an atrial septal defect in utero for fetuses with hypoplastic left heart syndrome and intact or highly restrictive atrial septum. Circulation 2004; 110: 253–8. 107. Assad RS, Zielinsky P, Kalil R et al. New lead for in utero pacing for fetal congenital heart block. J Thorac Cardiovasc Surg 2003; 126: 300–2.

26 Cardiomyopathy in the fetus John M Simpson Introduction Fetal cardiomyopathy is an infrequent occurrence and accounts for 2–4% of cardiac abnormalities observed during fetal life.1,2 Cardiomyopathies are important because the reported mortality of affected fetuses is high and the cardiac findings may provide a clue to an underlying disease process. Investigation for an underlying cause of cardiomyopathy is essential because of the potential for recurrence of cardiomyopathy in subsequent pregnancies and also to provide expectant parents with an informed prognosis with regard to outcome. In most cases, fetuses with cardiomyopathy are ascertained because of abnormal sonographic views of the heart during anomaly scanning. Thus, the fetal medicine specialist or fetal cardiologist tailors appropriate investigation from the echocardiographic features. Until recently, data in the literature were relatively scarce, but two large series have been reported within the past 5 years which have provided important new information.1,2

Classification The classification of cardiomyopathies used in this chapter is the subdivision of cases into dilated cardiomyopathy and hypertrophic cardiomyopathy. Dilated cardiomyopathy is defined as a group of conditions where the left and/or right ventricle is dilated above the normal range with diminished systolic function. The term hypertrophic cardiomyopathy is used to describe fetuses in which the dominant feature is abnormal hypertrophy of the left and/or right ventricle, without there being a structural cardiac abnormality sufficient to explain such hypertrophy.

Dilated cardiomyopathy Echocardiographic features The cardiothoracic ratio is invariably increased. Dilated cardiomyopathy can affect the left ventricle, the right

ventricle, or both. Examples of affected fetuses are shown in Figures 26.1–26.3. M-mode echocardiography confirms the reduced systolic function in such fetuses, which may be so severe that there is little or no contraction of the affected ventricle (Figure 26.4). Severe dilatation of the ventricles may lead to atrioventricular valve regurgitation (Figure 26.5). In some cases ventricular function is so poor that fetal hydrops results (Figure 26.6). As well as the reduction in systolic function, diastolic ventricular function is also abnormal. A case of dilated cardiomyopathy affecting the right ventricle is shown in Figure 26.2. In this case the mitral valve Doppler inflow pattern was normal but that of the tricuspid valve was abnormal, with only a single inflow velocity peak, corresponding to atrial contraction (a-wave). The finding of such an abnormal Doppler filling pattern has been identified as an adverse prognostic factor in a recent series.2

Reported series The first fetal series of dilated cardiomyopathy reported six fetuses of whom five had abnormal indices of systolic function. In two cases, the initial echocardiographic studies performed at 20 weeks of pregnancy were normal, and the cardiomyopathy only became evident with advancing gestation. Death occurred in four of the six cases, and one of the survivors required a heart transplant.3 A recent series from this author’s institution reported 50 consecutive cases of dilated cardiomyopathy.1 Dilatation of both ventricles was noted in almost half of these cases, and in the remainder either the right ventricle (34%) or the left ventricle (18%) was affected (Table 26.1). This pattern of ventricular involvement is also supported by other series. Overall, 50% of non-hydropic fetuses survived throughout the neonatal period versus 18% of those who developed hydrops during fetal life. Thus, in our series, hydrops was an important determinant of survival. In the series of Pedra et al,2 the importance of hydrops was less marked. Other echocardiographic factors were implicated in determining prognosis such as diastolic dysfunction, atrioventricular valve regurgitation,

376

Fetal Cardiology

Figure 26.1 Dilated cardiomyopathy affecting the left ventricle (LV). The left ventricle is severely dilated and had minimal contraction. The right ventricle (RV) is relatively unaffected and contracted well. The heart virtually filled the chest and there was little lung tissue visible. There is skin edema and a pericardial effusion (PE) posterior to the left ventricle. RA, right atrium.

Figure 26.2 (a) Fetal cardiomyopathy affecting the right ventricle (RV). The right ventricle is dilated and had minimal contraction. The right ventricle also appears mildly hypertrophied. The left ventricle (LV) was normal with good systolic function. There is a pericardial effusion (PE) around the right ventricle. (b) Doppler inflow patterns of fetus with right ventricular cardiomyopathy shown in (a). There are normal e- and a-waves in the mitral inflow pattern of the unaffected left ventricle. The tricuspid valve (TV) inflow pattern is abnormal, with only a single inflow peak, corresponding to the a-wave. MV, mitral valve.

Cardiomyopathy in the fetus

377

Figure 26.3 Fetal cardiomyopathy. In this example there is marked cardiomegaly, and little lung tissue is visible. LA, left atrium.

Figure 26.4 M-mode echocardiogram from a fetus with negligible contraction of either the right or the left ventricle.

Figure 26.5 Atrioventricular valve regurgitation in a fetus with dilated cardiomyopathy. There is marked tricuspid and mitral regurgitation secondary to dilatation of the right and left ventricles. Regurgitation through the tricuspid valve is typically more than that through the mitral valve. MR, mitral regurgitation; TR, tricuspid regurgitation.

378

Fetal Cardiology

preclude the development of abnormalities later in gestation or during postnatal life.1,3–5

Causes of dilated cardiomyopathy

Figure 26.6 Ascites in a fetus with dilated cardiomyopathy affecting both right and left ventricles.

Table 26.1 Graph of outcome of fetuses with dilated cardiomyopathy in relation to the affected ventricle and presence of fetal hydrops Affected ventricle RV

LV

Both

Total

Hydropic fetuses

8

6

18

32

Alive

4

0

2

6

IUD/NND

4

6

8

18

TOP

0

0

8

8

Nonhydropic fetuses

9

3

6

18

Alive

4

2

3

9

IUD/NND

4

0

3

7

TOP

1

1

0

2

RV, right ventricle; LV, left ventricle; IUD, intra-uterine death; NND, neonatal death; TOP, termination of pregnancy (reproduced with permission from reference 1).

‘a’ wave reversal of flow in the systemic veins, and umbilical vein pulsations.2 The interpretation of such data is made more difficult by the differing exclusion criteria between series; for example, the data of Sivasankaran et al1 excluded hypertrophic cardiomyopathy and fetuses with rhythm disturbances, whereas the series of Pedra et al included such cases.2 It is extremely important to emphasize that all cardiomyopathies may progress with advancing gestational age and that normal echocardiographic findings do not

Dilated cardiomyopathy may be the end result of a number of different disease processes, including metabolic, genetic, infective, hematological, renal, and immune-mediated. In a significant minority, no cause can be identified despite extensive investigation. Table 26.2 outlines the causes of dilated cardiomyopathy that were reported in our series, correlated with the echocardiographic features. In addition to the causes listed, additional viral causes have been reported, such as cytomegalovirus infection.2 Immune-mediated cardiomyopathy due to anti-Ro/La may be observed in the context of congenital complete heart block, but this has been described in the absence of any abnormality of the cardiac rhythm.2 Fetal tachycardias may produce ventricular dysfunction which can persist beyond the time when control of the arrhythmia has been achieved. Occult intermittent fetal tachycardia should also be considered where the fetal heart is dilated and dysfunctional, without apparent cause, and a prolonged period of observation may be required. The finding of fetal anemia as a potent cause of dilated ‘cardiomyopathy’ is important, because this is amenable to intrauterine blood transfusion. Published data have confirmed increased Doppler velocities and cardiac output in anemic fetuses prior to intrauterine transfusion.6,7 Those cases presenting with severe cardiac dysfunction probably represent a minority of fetuses with anemia so severe that myocardial function is compromised, or with coexisting severe myocarditis due to underlying infection such as parvovirus.1 Therefore, fetal blood sampling should be considered to exclude fetal anemia or at least rapid assessment of maternal serum for recent parvovirus infection. Fetal renal disease as a cause of a dilated dysfunctional heart may be related in part to fetal hypertension. This is supported by increased velocity of atrioventricular valve regurgitation in affected cases. The fetus is particularly sensistive to increased afterload due to the relative paucity of contractile elements in the fetal myocardium.8 Non-compaction of the ventricular myocardium has become an entity which is increasingly identified9,10 (Figure 26.7). Although there is debate as to its precise etiology, this condition is characterized by deep crypts within the myocardium, usually toward the apex of the left and/or right ventricles. This can produce cardiac failure during fetal life due to reduced contractile function of the ventricles, and can be diagnosed echocardiographically. The standard criteria include crypts in the myocardium which are more than twice the depth of the compact layer of the myocardium. This condition

Cardiomyopathy in the fetus

Table 26.2

Underlying causes of fetal dilated cardiomyopathy, echocardiographic features and outcome Total

I

Metabolic/genetic

III

IV

V VI

Survivors

Nonsurvivors

BV DCM

LV DCM

RVDCM

11

0

11

7

2

2

sialic acid storage disease

1

0

1

0

1

0

FH of dilated cardiomyopathy in males

1

0

1

1

0

0

calcific vasculopathy

2

0

2

2

0

0

generalized myopathy

1

0

1

1

0

0

mitochondrial cytopathy

2

0

2

0

1

1

probable because of recurrence II

379

4

0

4

3

0

1

11

3

8

6

2

3

HIV

1

0

1

0

0

1

parvovirus

7

2

5

5

0

2

coxsackie

2

1

1

0

2

0

toxoplasma

1

0

1

1

0

0

Infection

Fetal anemia

5

4

1

1

0

4

anti C antibody

1

1

0

0

0

1

Possible fetomaternal transfusion

1

1

0

0

0

1

Parvovirus infection suspected not proved

3

2

1

1

0

2

Other cardiac causes

5

4

1

2

0

3

Juxta-ductal coarctation

1

1

0

1

0

0

Abdominal coarctation

1

0

1

0

0

1

Intrauterine ductal closure

2

2

0

0

0

2

Isolated noncompaction of myocardium

1

1

0

1

0

0

Renal

5

0

5

3

0

2

Probable cause identified

37

11

26

19

4

14

Idiopathic

13

4

9

5

5

3

LV, left ventricle; RV, right ventricle; BV, biventricular (both right and left ventricles affected); DCM, dilated cardiomyopathy (reproduced with permission from reference 1).

may be familial and hence investigation of parents is warranted, as the symptomatology is extremely variable so that affected parents may be entirely asymptomatic.9

Investigation of dilated cardiomyopathy in the fetus

detailed imaging of the descending aorta.1,11 Dilatation and dysfunction of the right ventricle may be related to constriction of the arterial duct.12 Anomalies of the fetal venous system should also be considered in the contexts of unexplained dilatation of the heart including cerebral arteriovenous malformations or congenital absence of the ductus venosus with direct connection of the umbilical vein to the systemic veins or heart,13,14 leading to volume overload of the circulation.

Conditions mimicking ‘cardiomyopathy’ It is important to be aware of some conditions that can cause cardiac dilatation and/or dysfunction which can be difficult to detect prenatally unless they are specifically sought. Coarctation of the aorta does not typically cause myocardial dysfunction duing fetal life, unless the narrowing is beyond the aortic arch. Abdominal coarctation has been described rarely during fetal life but should be excluded by

Investigations related to cardiomyopathy Following a diagnosis of dilated cardiomyopathy there should be a thorough investigation for possible underlying causes. This will include detailed anomaly scanning, particularly to exclude the secondary causes outlined above and also to examine the fetal kidneys. Appropriate maternal

380

Fetal Cardiology

(a)

(b)

(c)

Figure 26.7 Non-compaction of the myocardium. (a) There are crypts within the myocardium of the apex of both the left and right ventricular apices. (b) Color flow Doppler mapping in the same fetus confirms the deep crypts within the myocardium. In this author’s experience recognition of such crypts may be difficult until late in the mid-trimester. (c) Use of M-mode echocardiography confirms reduced ventricular contraction in another fetus with non-compaction of the myocardium.

samples should be taken for viral studies, as well as a TORCH screen (toxoplasmosis, rubella, cytomegalovirus, herps simplex, and human immunodeficiency virus). Anti-Ro/La antibodies should also be checked, even in the absence of heart block, particularly if there is endocardial fibroelastosis.2 Consideration should be given to fetal blood sampling to exclude anemia, but this procedure is likely to carry a relatively high risk in a fetus with poor cardiac function, compared to other indications.15 An alternative is rapid maternal testing for recent parvovirus infection. When fetal blood is obtained, we routinely send the sample for chromosome analysis, although chromosomal abnormalities are unusual in this situation, and also for DNA storage for possible future analysis. Repeat echocardiograms may be necessary to exclude

occult arrhythmia, particularly intermittent tachycardia.16 Some of the metabolic causes of dilated cardiomyopathy may be difficult to exclude prenatally, and so postnatal metabolic investigations should be instigated in cases where no cause has been identified prior to birth. In fetuses that die either pre- or postnatally it is essential that appropriate fresh unfixed tissue samples be obtained, so that these are available for metabolic investigation. This is particularly important, because some cases of dilated cardiomyopathy may recur in subsequent pregnancies. Wherever possible, DNA should be stored for possible future investigation. It will usually be essential to involve relevant subspecialities including fetal medicine, virology, cardiology, genetics, and metabolic disease.

Cardiomyopathy in the fetus

381

Intrauterine therapy There have been no controlled trials of therapy for fetal cardiomyopathy. Causes of cardiac dilatation and dysfunction which are definitely treatable include fetal anemia (intrauterine transfusion) and occult tachycardias (antiarrhythmic drug therapy). Therapy for presumed fetal viral myocarditis with immunoglobulins and steroid administration has been attempted but without success.2

Hypertrophic cardiomyopathy Echocardiographic features The unifying echocardiographic feature in this condition is severe ventricular hypertrophy that is not explained by any structural cardiac abnormality. Hypertrophy of the myocardium may affect either or both ventricles (Figures 26.8 and 26.9). Careful exclusion of obstructive lesions such as critical aortic stenosis or critical pulmonary stenosis is mandatory, to confirm that the hypertrophy is not secondary to outflow tract obstruction. This will involve Doppler interrogation of the left and right ventricular outflow tracts. In some fetuses, severe hypertrophy of the ventricle may lead to a dynamic obstruction of the outflow tract by the hypertrophied muscle (Figure 26.8b). Quantitation of the degree of ventricular hypertrophy may be helpful to follow disease progression. There are published normal ranges of septal and ventricular wall thickness using either cross-sectional17 or M-mode echocardiography.18

Causes Myocardial hypertrophy may be the end result of a number of different disease processes. The causes that have been described in series in the literature are shown in Table 26.3. Some of the different groups of diseases that may lead to hypertrophic cardiomyopathy in utero are discussed below.

Diabetic hypertrophic cardiomyopathy Maternal diabetes mellitus is an indication for detailed fetal echocardiography in view of the increased incidence of congenital heart defects in this group.19 In addition, the most frequent cause of excessive ventricular hypertrophy in the fetus is maternal diabetes mellitus.20 Such hypertrophy tends to occur late in pregnancy, typically beyond 30 weeks. Following delivery, ventricular hypertrophy resolves spontaneously, although in a minority of cases the infant may be symptomatic until the hypertrophy has

(a)

Peak velocity =1.2m/s Note late systolic accentuation

(b)

Figure 26.8 (a) Hypertrophic cardiomyopathy in Noonan syndrome. There is marked hypertrophy of the left ventricle and ventricular septum. Views of the heart were limited, owing to advanced gestational age. The left ventricular cavity is slit-like. This fetus had increased nuchal translucency early in pregnancy and a normal karyotype. Hypertrophic cardiomyopathy developed late in gestation. Noonan syndrome was confirmed postnatally. (b) Dynamic left ventricular outflow tract obstruction in the same fetus. The aortic Doppler is 1.2 m/s at 30 weeks’ gestational age. Note the increased Doppler velocity in late systole confirming dynamic obstruction.

resolved. Thus, if diabetic cardiomyopathy is to be detected prenatally, then echocardiography relatively late in pregnancy is indicated. Given that the myocardial hypertrophy resolves without therapy, and only a minority of newborn infants are symptomatic, the justification for sequential fetal echocardiography is debatable.

382

Fetal Cardiology

during fetal life.22 Some fetuses with Noonan syndrome will present initially with increased nuchal translucency and normal karyotype, which may be a useful clue to the underlying diagnosis. Chromosomal abnormalities have been reported but are rare.23

Metabolic causes Mitochondrial cytopathies may present with myocardial hypertrophy during fetal life, although the total number of reported cases is small.1,24 Therefore, detailed metabolic investigations are indicated for unexplained ventricular hypertrophy, which may include postnatal endomyocardial biopsy.25 Such investigations are extremely difficult to undertake during fetal life and may have to be deferred until after birth.

Figure 26.9 Hypertrophic cardiomyopathy. In this example there is a pericardial effusion. There is hypertrophy of both the left and right ventricles. No underlying cause was identified. LVH, left ventricular hypertrophy; RVH, right ventricular hypertrophy.

Table 26.3 Causes of hypertrophic cardiomyopathy

Renal disease In the past, one of the main causes of unexplained ventricular hypertrophy in the fetus was (fetal) renal disease. The cause of the hypertrophy is difficult to establish with certainty, but fetal hypertension may be important. In current practice, such renal abnormalities are usually identified by imaging of the renal tract, so that a diagnosis is usually made prior to fetal echocardiography.

Maternal diabetes mellitus Genetic familial Noonan syndrome chromosomal abnormality Metabolic B lipase deficiency cytochrome oxidase deficiency Fetal renal disease renal agenesis multicystic kidneys

Twin-to-twin transfusion syndrome In twin-to-twin transfusion syndrome, a variety of effects have been observed on the heart of both the donor and of the recipient twin.26,27 In the recipient twin, such effects include ventricular dysfunction, tricuspid regurgitation (Figures 26.10 and 26.11), and right ventricular outflow tract obstruction.27 The cause of these findings is not entirely clear, but volume and pressure overload of the heart of the recipient fetus is a likely explanation of the functional abnormalities.28,28 The reason why anatomic obstruction of the right ventricle occurs in some cases is not known.

congenital nephrotic syndrome Twin–twin transfusion syndrome

Outcome Genetic causes Familial hypertrophic cardiomyopathy has been detected prenatally.21 A normal fetal echocardiogram does not, however, exclude this diagnosis, because there may be progressive hypertrophy during childhood and adult life. Noonan syndrome has also been diagnosed postnatally in some fetuses presenting with hypertrophic cardiomyopathy

Pedra et al reported a series of 33 cases of hypertrophic cardiomyopathy presenting during fetal life, with an overall mortality of just over 50%. Although such data provide an overall mortality, the prognosis should be tailored for each individual fetus. Some metabolic causes may be universally fatal but, at the other extreme, diabetic cardiomyopathy resolves postnatally even if there are transient symptoms in the neonatal period.

Cardiomyopathy in the fetus

383

Figure 26.10 Appearances of the heart in twin-to-twin transfusion syndrome (recipient fetus). The right atrium is dilated secondary to marked tricuspid regurgitation.

Figure 26.11 Twin-to-twin transfusion syndrome (TTTS) (recipient fetus). In this example the left ventricle is dilated with very poor systolic function on M-mode echocardiography.

Investigation of hypertrophic cardiomyopathy The causes of hypertrophic cardiomyopathy in the fetus are diverse. In all cases a detailed fetal anomaly scan is mandatory to exclude associated malformations, particularly of the kidneys and renal tract. Exclusion of maternal diabetes mellitus is important, because myocardial hypertrophy should resolve postnatally without treatment. Although unusual, chromosomal disease may manifest as hypertrophic cardiomyopathy prenatally,23 and thus fetal karyotyping should be considered. A normal standard fetal karyotype clearly does not exclude underlying genetic disease such as Noonan syndrome. In cases where there is a family history of metabolic disease, prenatal genetic testing may be available, depending on the condition. When there is no family history, metabolic investigation may have to be deferred until postnatally, when appropriate samples may be obtained. If pregnancy results in intrauterine death or termination of pregnancy, some tissue samples should remain unfixed, so that metabolic analysis

is still possible. Detailed postmortem examination should be discussed with parents to provide additional information on causation and recurrence risks. For both dilated and hypertrophic cardiomyopathies, appropriate management is likely to involve a number of different subspecialists including a fetal cardiologist, fetal medicine specialist, clinical geneticist, specialist in metabolic disease, and pathologists. Such a combined approach has the maximum chance of providing parents with an explanation of cardiomyopathy in their fetus or child.

References 1. Sivasankaran S, Sharland GK, Simpson JM. Dilated cardiomyopathy presenting during fetal life. Cardiol Young 2005; 15: 409–16. 2. Pedra SR, Smallhorn JF, Ryan G et al. Fetal cardiomyopathies: pathogenic mechanisms, hemodynamic findings, and clinical outcome. Circulation 2002; 106: 585–91. 3. Schmidt KG, Birk E, Silverman NH, Scagnelli SA. Echocardiographic evaluation of dilated cardiomyopathy in the human fetus. Am J Cardiol 1989; 63: 599–605.

384

Fetal Cardiology

4. Yinon Y, Yagel S, Hegesh J et al. Fetal cardiomyopathy: in utero evaluation and clinical significance. Prenat Diagn 2007; 27: 23–8. 5. Pedra SR, Hornberger LK, Leal SM, Taylor GP, Smallhorn JF. Cardiac function assessment in patients with family history of nonhypertrophic cardiomyopathy: a prenatal and postnatal study. Pediatr Cardiol 2005; 26: 543–52. 6. Rizzo G, Nicolaides KH, Arduini D, Campbell S. Effects of intravascular fetal blood transfusion on fetal intracardiac Doppler velocity waveforms. Am J Obstet Gynecol 1990; 163: 1231–8. 7. Moise KJ, Mari G, Fisher DJ et al. Acute fetal hemodynamic alterations after intrauterine transfusion for treatment of severe red blood cell alloimmunization. Am J Obstet Gynecol 1990; 163: 776–84. 8. Fisher DJ, Towbin J. Maturation of the heart. Clin Perinatol 1988; 15: 421–46. 9. Moura C, Hillion Y, Daikha-Dahmane F et al. Isolated noncompaction of the myocardium diagnosed in the fetus: two sporadic and two familial cases. Cardiol Young 2002; 12: 278–83. 10. Karatza AA, Holder SE, Gardiner HM. Isolated noncompaction of the ventricular myocardium: prenatal diagnosis and natural history. Ultrasound Obstet Gynecol 2003; 21: 75–80. 11. Zeltser I, Parness IA, Ko H, Holzman IR, Kamenir SA. Midaortic syndrome in the fetus and premature newborn: a new etiology of nonimmune hydrops fetalis and reversible fetal cardiomyopathy. Pediatrics 2003; 111: 1437–42. 12. Huhta JC, Moise KJ, Fisher DJ et al. Detection and quantitation of constriction of the fetal ductus arteriosus by Doppler echocardiography. Circulation 1987; 75: 406–12. 13. Sau A, Sharland G, Simpson J. Agenesis of the ductus venosus associated with direct umbilical venous return into the heart—case series and review of literature. Prenat Diagn 2004; 24: 418–23. 14. Jaeggi ET, Fouron JC, Hornberger LK et al. Agenesis of the ductus venosus that is associated with extrahepatic umbilical vein drainage: prenatal features and clinical outcome. Am J Obstet Gynecol 2002; 187: 1031–7. 15. Maxwell DJ, Johnson P, Hurley P et al. Fetal blood sampling and pregnancy loss in relation to indication. Br J Obstet Gynaecol 1991; 98: 892–7. 16. Simpson JM, Yates RW, Milburn A, Maxwell DA, Sharland GK. Outcome of intermittent tachyarrhythmias in the fetus. Pediatr Cardiol 1997; 18: 78–83.

17. Tan J, Silverman NH, Hoffman JI, Villegas M, Schmidt KG. Cardiac dimensions determined by cross-sectional echocardiography in the normal human fetus from 18 weeks to term. Am J Cardiol 1992; 70: 1459–67. 18. Allan LD, Joseph MC, Boyd EG, Campbell S, Tynan M. M-mode echocardiography in the developing human fetus. Br Heart J 1982; 47: 573–83. 19. Meyer-Wittkopf M, Simpson JM, Sharland GK. Incidence of congenital heart defects in fetuses of diabetic mothers: a retrospective study of 326 cases. Ultrasound Obstet Gynecol 1996; 8: 8–10. 20. Veille JC, Sivakoff M, Hanson R, Fanaroff AA. Interventricular septal thickness in fetuses of diabetic mothers. Obstet Gynecol 1992; 79: 51–4. 21. Stewart PA, Buis-Liem T, Verwey RA, Wladimiroff JW. Prenatal ultrasonic diagnosis of familial asymmetric septal hypertrophy. Prenat Diagn 1986; 6: 249–56. 22. Sonesson SE, Fouron JC, Lessard M. Intrauterine diagnosis and evolution of a cardiomyopathy in a fetus with Noonan’s syndrome. Acta Paediatr 1992; 81: 368–70. 23. Chen CP, Chern SR, Lee CC et al. De novo unbalanced translocation resulting in monosomy for proximal 14q and distal 4p in a fetus with intrauterine growth retardation, Wolf–Hirschhorn syndrome, hypertrophic cardiomyopathy and partial hemihypoplasia. J Med Genet 1998; 35: 1050–3. 24. von Kleist-Retzow JC, Cormier-Daire V, Viot G et al. Antenatal manifestations of mitochondrial respiratory chain deficiency. J Pediatr 2003; 143: 208–12. 25. Rustin P, Lebidois J, Chretien D et al. Endomyocardial biopsies for early detection of mitochondrial disorders in hypertrophic cardiomyopathies. J Pediatr 1994; 124: 224–8. 26. Hecher K, Sullivan ID, Nicolaides KH. Temporary iatrogenic fetal tricuspid valve atresia in a case of twin to twin transfusion syndrome. Br Heart J 1994; 72: 457–60. 27. Zosmer N, Bajoria R, Weiner E et al. Clinical and echographic features of in utero cardiac dysfunction in the recipient twin in twin–twin transfusion syndrome. Br Heart J 1994; 72: 74–9. 28. Hecher K, Ville Y, Nicolaides KH. Fetal arterial Doppler studies in twin–twin transfusion syndrome. J Ultrasound Med 1995; 14: 101–8. 29. Wieacker P, Wilhelm C, Prompeler H et al. Pathophysiology of polyhydramnios in twin transfusion syndrome. Fetal Diagn Ther 1992; 7: 87–92.

27 Ultrasound examination of the fetal coronary circulation Ahmet A Baschat and Ulrich Gembruch Introduction The coronary circulation provides blood supply to the myocardium. For one of the most critical fetal organs, matching myocardial blood flow and demand is necessary to ensure cardiac function over a wide variety of physiologic and pathologic conditions. For this reason the examination of coronary vascular dynamics is becoming increasingly relevant in various fetal conditions. Ultrasound examination of the fetal coronary circulation has become possible through advances in ultrasound technology and a better understanding of human fetal cardiovascular physiology. Although not yet standard clinical practice, continuing trends in ultrasound technology and spreading familiarity with the examination and interpretation are likely to expand the clinical applications in the future.1 Ultrasound examination of the fetal coronary system utilizes gray-scale, zoom, and cine loop techniques, and requires optimal spatial and temporal settings of the Doppler modalities. Proper setup of the ultrasound system is therefore a necessary prerequisite. Traditional ultrasound planes used in cardiac scanning are modified to provide best visualization of the coronary vessels. A comprehensive survey of extracardiac vascular dynamics is often necessary to provide the clinical context for interpretation of intracardiac and coronary flow dynamics. This chapter will review embryology, functional anatomy, animal experiments, ultrasound technique, and clinical utility of ultrasound evaluation of the fetal coronary circulation.

Embryology and functional anatomy of the coronary circulation Oxygenated blood supply to the myocardium is delivered through the right (RCA) and left coronary arteries (LCA), arising from the ipsilateral anterior and posterior aortic

sinuses, respectively, and the left anterior descending branch (LAD) of the LCA.2,3 Left ventricular venous return drains mainly into a superficial system through the coronary sinus and anterior cardiac veins, constituting approximately two-thirds of myocardial venous return. The deep system consisting of arterioluminal vessels, arteriosinusoidal vessels, and Thebesian veins drains the remaining venous return directly into the cardiac chambers.2–4 In embryonic life, endothelial cells originating from the septum transversarium in the hepatic region form epicardial blood islands that eventually coalesce into vascular networks throughout the epicardium and myocardium.5,6 Concurrently, the RCA and LCA originate as microvessels and penetrate the aortic root, acquiring a muscular coat in this process. Connection of the main stem coronary arteries and myocardial vascular channels marks the initiation of a functional coronary circulation. Venous drainage develops independently of the arterial system and becomes fully functional when the coronary sinus, as a remnant of the left horn of the sinus venosus, becomes incorporated into the inferior wall of the right atrium and Thebesian veins gain access to the ventricular cavities. The coronary circulation is completely functional by the sixth week of embryonic life and ensures myocardial blood supply by the time the embryonic circulation is established. Coronary vascular development can be modulated by stimuli such as local oxygen tension, mechanical wall stress, and myocardial and vascular shear forces.6–9 As a result, the coronary circulation is subject to great anatomic and functional variation. Under physiologic conditions, modulation of vascular growth enables the matching of coronary vascular development to myocardial growth, ensuring a balanced relationship between ventricular mass and vascular density.10 Persistent, or progressive, tissue hypoxemia may exaggerate this physiologic process, resulting in a marked increase in vascular cross-sectional area in the coronary circulation.11–14 Under these circumstances, vascular reactivity to physiologic stimuli is also enhanced.15 Similarly, abnormal intracardiac pressure relationships such as those found in outflow tract obstructive lesions

386

Fetal Cardiology

may force the development of accessory vascular channels between the coronary vessels and the ventricular cavity (ventriculocoronary fistulae).16 The plasticity of the coronary circulation is responsible for the variation in myocardial vascular territories and blood flow found in various fetal conditions, illustrating the importance of myocardial oxygenation for proper cardiac function.

Regulation of myocardial perfusion Myocardial metabolism is almost exclusively aerobic and, in the presence of adequate oxygen, various substrates including carbohydrates, glucose, lactate, and lipids can be metabolized.17–20 In fetal life, myocardial glycogen stores and lactate oxidation constitute the major sources of energy, while fatty acid oxidation rapidly becomes the primary energy source after birth. To maintain metabolism, myocardial oxygen extraction is as high as 70–80% in the resting state. Consequently, the coronary atrioventricular O2 difference of 14 ml/dl exceeds that of most other vascular beds, and allows little further extraction of O2 unless the blood flow is significantly augmented. Therefore, coronary blood flow is closely regulated to match myocardial oxygen demands. The regulation of myocardial perfusion operates at several levels and time-frames. The unique parallel arrangement of the fetal circulation allows delivery of well-oxygenated blood through the ductus venosus to the left ventricle and thus the ascending aorta. In the fetal lamb, the coronary circulation receives approximately 8% of the left ventricular output at rest.21 This proportion may be higher in the human fetus, and may be further altered by modulating the degree of shunting through the ductus venosus.22 Once blood enters the coronary vessels from the ascending aorta, the pressure difference to the right atrium becomes the primary driving force for coronary blood flow. This perfusion pressure is further subjected to changes in vascular tone and extravascular pressure. Autonomic innervation of coronary resistance vessels regulates overall vascular tone,23,24 but ventricular contractions are the main contributor to extravascular resistance, with significant impacts on the flow velocity waveform.25–27 Myocardial perfusion predominantly occurs during diastole when the ventricles relax and pose little extravascular resistance. This diastolic timing of predominant perfusion is unique to the coronary circulation, and distinguishes it from other vascular beds in the human body. In the adult, increases above a resting heart rate of 70 beats per minute result in a disproportionate shortening of diastole. At fetal heart rates of 120–160 beats per minute, this places special demands on dynamic vascular mechanisms to regulatethe myocardial blood flow volume.

The efficiency of myocardial oxygen delivery is further enhanced by active autoregulatory control mechanisms, ensuring optimal myocardial blood flow despite fluctuations in arterial perfusion pressure.28,29 This is achieved through caliber adjustment of precapillary resistance vessels, allowing channeling of blood flow to areas of greatest oxygen demand.30,31 With maximal dilatation of these sphincters, myocardial blood flow may be elevated four times above basal flow. This increase in blood flow volume that can be achieved under these circumstances is the myocardial blood flow reserve. If myocardial oxygenation cannot be upheld long-term, adaptation with the formation of new blood vessels may be invoked, thus increasing the myocardial blood flow reserve.32–34 Such elevated myocardial blood flow reserve allows marked augmentation of blood flow during periods of acutely worsening hypoxemia or increased cardiac work, and increases as high as 12 times the basal flow have been reported.15,35,36

Coronary blood flow under various conditions in the animal model The arrangement of the fetal circulation and the role of the placenta as the primary source of oxygen, water, and nutrients have important implications for fetal vascular homeostasis. Partitioning of oxygen-rich umbilical venous blood occurs at several levels that can affect coronary perfusion. Increased diversion of umbilical venous blood through the ductus venosus increases the proportion of this oxygenated blood that reaches the left ventricle, while differential changes in right and left ventricular afterload can shift the proportion of cardiac output toward the ventricle with the lowest downstream blood flow resistance. This unique arrangement of the fetal circulation is responsible for a number of cardiovascular responses observed in fetuses with uteroplacental insufficiency. Studies of the coronary circulation in the fetal lamb indicate functional autoregulation under conditions of acute hypoxemia,19,21,25,37 increased afterload,35 and acuteon-chronic hypoxemia.25,27 The fetal heart is remarkably tolerant of a hypoxic environment, and is capable of sustaining normal cardiac function in the presence of a 50% reduction in oxygen level. Subjected to acute hypoxemia, a four- to five-fold increase in myocardial blood flow, corresponding to changes in the adult human heart, is observed. Selective elevation of right ventricular afterload is associated with a global increase in myocardial blood flow.35 This may be due to increased oxygen demands of cardiac pressure work, suggesting functional extracardiac control by humoral and/or neurogenic factors.27,35,38–41 However, the most striking changes in myocardial blood flow can be observed in fetuses subjected to acute hypoxemia after a period of chronic hypoxemia. In this setting of

Ultrasound examination of the fetal coronary circulation

acute-on-chronic hypoxemia, maximal myocardial flow reserve is increased to 12-fold the basal flow, and is amongst the highest flows observed under any condition. This increase in flow reserve most likely reflects the effects of chronic hypoxemia on coronary vascular remodeling and reactivity. The alterations in the vascular tree allow recruitment of much larger blood volume during acute hypoxemia, resulting in markedly enhanced blood flow25,27,33,34

Ultrasound examination technique Ultrasound setup The setup of the ultrasound system determines the spatial and temporal resolution of the ultrasound image, and is therefore of major importance for successful examination of the fetal coronary arteries. Gray-scale ultrasound and color and pulsed wave Doppler need to be optimized for complementary use. Although visualization of coronary vessels can be achieved using 4-MHz transducers, higher frequencies improve resolution and therefore detection. The dynamic range of the gray-scale image should be set to an intermediate level that is generally used in cardiac setups. Zoom magnification of the area of interest limits the computing power that needs to be allocated to the generation of the gray-scale image. These two maneuvers will improve the frame repetition rate and should therefore be applied before adding color Doppler imaging. When adding color Doppler imaging the filter should be set to a high degree of motion discrimination, and the color box and gate are kept as small as possible to optimize spatial and temporal resolution of this Doppler modality. The lateral dimension of the color box has the greatest impact on computing power and therefore frame rate. The color amplification gain is set to eliminate background noise on the screen. The persistence is set to a low level to minimize frame averaging. The color velocity scale is adjusted to a range that allows visualization of intra- and extracardiac flows without aliasing and suppression of wall-motion artifacts. A useful velocity range for coronary arteries is between 0.3 and 0.7 m/s for coronary arteries and between 0.1 and 0.3 m/s for the coronary sinus. Since initial detection of the coronary arteries relies on color Doppler, these aspects of the setup are essential preliminary steps. Once the coronary vessel is identified using these techniques the transducer position should be adjusted to provide an insonation angle close to 0° prior to obtaining pulsed wave measurements. The pulsed wave Doppler gate should be adjusted to exclude other cardiac and extracardiac flows and should be the only active display when measurements are taken. Concurrent activation of multiple image modalities (duplex or triplex mode) drastically increases computing requirements and

387

affects the spatial and temporal resolution of the spectral Doppler waveform.

Examination of coronary arteries Using gray-scale ultrasound the coronary ostia are discernable in late gestation. Before this time the size of the main stem arteries is below 1 mm in diameter and thus frequently below the resolution threshold of current sonographic equipment in the majority of cases.42 For this reason color and pulsed wave Doppler ultrasound are necessary to detect and verify coronary artery blood flow. Doppler examination of the fetal coronary vessels has been adopted from techniques developed for infants and neonates.43 The main stem right and left coronary arteries are best examined in a long-axis view of the left ventricular outflow tract and ascending aorta or a precordial shortaxis view of the aorta. The left anterior descending branch of the LCA is best identified in an apical short-axis view. In the standard precordial short-axis view the left coronary artery runs forward, while the right runs more parallel. This view therefore facilitates examination of the LCA. In the lateral-, or long-axis view of the left ventricular outflow tract the RCA is more readily imaged from the right side of the fetus. In this view it may also be possible to visualize both coronary arteries (Figure 27.1).44,45 The LAD may be identified from the apical four-chamber view by tilting the transducer cephalad until the level of the superior cardiac surface and interventricular groove is reached.46 Cardiac wall motion, high blood flow velocities in the ventricles and ventricular outflow tracts, and movement of pericardial fluid can all interfere with the relatively low coronary blood flow velocities on color Doppler imaging. Back and forward motion of pericardial fluid outlining the ventricular walls in particular may be mistaken for a coronary artery.47 For these reasons, identification of coronary artery blood flow by color Doppler imaging should always be followed by verification of the typical waveform pattern by pulsed wave Doppler to provide assurance that the coronary arteries have indeed been identified. Spectral Doppler measurement of coronary blood flow velocities is easiest proximally, since vessel diameter is greatest and motion during the cardiac cycle is less than distally. After coronary vessels are identified by color Doppler, the pulsed wave Doppler gate is positioned at their origin. The gate may require adjustment to achieve continuous sampling of the waveform, allowing for movement of the aortic root in the cardiac cycle. The coronary artery flow velocity waveform has a biphasic pattern, with systolic and diastolic peaks and antegrade flow throughout the cardiac cycle (Figure 27.2). The coronary waveform is unique. Due to the predominant diastolic perfusion, velocities are higher during diastole than systole. In normal fetuses coronary blood flow has been visualized from 29 weeks onwards (median gestational age of 33 + 6 weeks).

388

Fetal Cardiology

(c)

(d) (b)

P RCA (e) LCA (a) AO

TV

MV

Short-axis views

Long-axis views

Figure 27.1 A diagrammatic view of the origin of the great vessels and the atrioventricular valves shows the course of the left and right coronary arteries (RCA and LCA, respectively) in relationship to the aorta (AO), pulmonary artery (P), mitral (MV), and tricuspid valves (TV). The angle of insonation and type of cardiac axis determines the orientation of the coronary arteries on the ultrasound image. Short-axis views facilitate examination of the left coronary artery (b) and may enable visualization of both coronary arteries (c), occasionally also allowing demonstration of the origin of the left anterior descending branch (b, c). Although the right coronary artery can be examined in the short-axis view (a), it is easier to identify this vessel in a right lateral long-axis view of the left ventricular outflow tract (e). This view also allows simultaneous visualization of both coronary arteries (d).

The median systolic and diastolic peak blood flow velocities are 0.21 and 0.43 m/s, respectively, and show little change during the latter part of gestation (Figures 27.3 and 27.4.)49

Examination of the coronary sinus The larger size and location of the coronary sinus facilitate its ultrasound examination.50,51 The coronary sinus runs in the atrioventricular groove and enters the right atrium below the level of the foramen ovale just above the valve of the inferior vena cava. Because of its position, apical or basal four-chamber views provide the best opportunity for gray-scale biometry, while lateral four-chamber views provide a more favorable insonation angle for color and spectral Doppler imaging (Figures 27.5 and 27.6). Gray-scale and M-mode echocardiography have both been used to obtain normative data on the length and diameter of the coronary sinus. The caliber of the coronary sinus undergoes cyclic changes, with the cardiac cycle

being smallest at the beginning of diastole and largest in mid-systole with maximal descent of the atrioventricular (AV) ring. M-mode echocardiography allows precise documentation of caliber and dynamic changes (Figure 27.7). The coronary sinus has a maximum diameter ranging from 1 to 3 mm with advancing gestation. The method utilized to obtain these measurements does influence the reference limits.51,53 Figures 27.8 and 27.9 show gestational reference ranges for the maximal diastolic and systolic dimensions measured with M-mode. Appreciating the phenomenon of variations in coronary sinus diameter may call for verification using this M-mode technique when dilatation of the coronary sinus is suspected. Color Doppler identification of coronary sinus blood flow is successful in approximately 50% of normal fetuses. During diastole the direction of blood flow from the coronary sinus is toward the right atrium, whereas blood flow across the foramen ovale is directed toward the left atrium (Figure 27.10).52 Despite its straight course, allowing exact placement of the sample volume, specral Doppler measurements are only possible in approximately 10% of fetuses.

Ultrasound examination of the fetal coronary circulation

LCA

LAD

389

RCA

D

D

D S

S

(a)

S

(b)

(c)

Figure 27.2 Pulsed wave Doppler images of the left coronary (LCA, a), left anterior descending (LAD, b), and right coronary arteries (RCA, c) are obtained in a 29-week fetus. Of note is the predominance of blood flow during diastole (D) that is observed in all three vessels (reproduced with permission from reference 48). 0.7

0.4

0.6 0.5 m/s

m/s

0.3

0.2

0.4 0.3 0.2

0.1 PSV = 0.0002 x GA + 0.3429 R2= 0.0002 0

PDV = 0.0023 x GA + 0.3226 R2= 0.0075

0.1 0

29

31

33 35 37 39 Gestational age (weeks)

41

Figure 27.3

29

31

33 35 37 Gestational age (weeks)

39

Figure 27.4

The graph displays the median and 95% confidence interval for the peak systolic velocity (PSV) in the coronary artery of appropriately grown fetuses in relation to gestational age (GA) (reproduced with permission from reference 49).

The graph displays the median and 95% confidence interval for the peak diastolic velocity (PDV) in the coronary artery of appropriately grown fetuses in relation to gestational age (GA) (reproduced with permission from reference 49).

This low success rate is partly due to lower coronary sinus blood flow velocities and interference caused by intraatrial blood flows and/or cardiac and atrioventricular valve movement. The coronary sinus flow velocity waveform has a triphasic pattern, with systolic and diastolic antegrade

flow and occasional reversal during atrial contraction (Figure 27.11). Similar to the coronary arteries, diastolic forward velocities (median 0.38 m/s) exceed systolic velocities (median 0.18 m/s). These velocities are related to the periods of predominant myocardial blood flow.

390

Fetal Cardiology

Right ventricle

Right atrium IVC

Left ventricle

LV

FO

RV Left atrium

LA

Coronary sinus

RA

Figure 27.5 This diagrammatic view of the fetal heart shows the course of the coronary sinus in the right lateral four-chamber view. The coronary sinus runs in the atrioventricular groove and opens into the right atrium near the atrioventricular valve, in close proximity to the inferior vena cava (IVC) and foramen ovale (FO). In this imaging plane, the direction of blood flow is toward the transducer beam (reproduced with permission from reference 52).

Methods to relate coronary sinus velocities to myocardial flow reserve have been described in neonates and adults,54,55 but these are currently not practicable for validation in the human fetus.

Clinical applications in fetuses with normal cardiac anatomy There are several conditions in fetuses with normal cardiac anatomy that have prominent cardiovascular manifestations. In some of these conditions coronary blood flow dynamics may be altered to accommodate changes in myocardial oxygen requirements. Since spectral Doppler of the coronary sinus is rarely achieved, clinical observations revolve primarily around color and pulsed wave Doppler characteristics in coronary arterial vessels.

The ‘heart sparing effect’ in fetal growth restriction Severe fetal growth restriction (IUGR) can progress to decompensation of cardiovascular status documented through progressive deterioration of arterial and venous Doppler

Figure 27.6 Diagrammatic view of the fetal heart imaged in the apical four-chamber view showing the left and right ventricles (LV and RV) and the corresponding atria (LA and RA). The coronary sinus runs in the atrioventricular groove parallel to the mitral valve leaflets. The coronary sinus is visualized by tilting the transducer toward the inferior cardiac surface until the valve leaflets disappear (reproduced with permission from reference 51).

studies.56 This progression often accompanies the deterioration of acid–base status from chronic hypoxemia to acidemia.57–60 Under these circumstances the combination of elevated central venous pressure, elevated afterload, and worsening oxygenation places unique demands on myocardial oxygen balance. Elevated afterload increases myocardial oxygen demand because of an increase in cardiac work. Elevated central venous pressure and aortic pressure decrease the pressure difference across the coronary vascular bed and therefore diminish the driving force for coronary perfusion. The summation of these factors has detrimental effects on coronary perfusion at a time when myocardial oxygen balance and fetal metabolic state are drastically increased. Consequently, adaptive mechanisms need to be evoked in order to maintain myocardial oxygen balance. The necessary augmentation of coronary blood flow can be achieved in two principal ways. One is to increase the proportion of oxygenated left ventricular output available for myocardial delivery. The second is through autoregulation mediated coronary vasodilatation. Several mechanisms operate in IUGR fetuses that increase the potential delivery of oxygenated blood to the myocardium (Figure 27.12). Under conditions of elevated

Ultrasound examination of the fetal coronary circulation

LV

391

LV

(a)

(c)

(b)

Figure 27.7

The fetal heart imaged in an apical four-chamber view at 28 + 4 weeks’ gestation. The coronary sinus (arrows) can be seen running in the atrioventricular groove between the left ventricle (LV) and atrium. Using the cine-loop technique a difference in diameter between end-systolic (a) and mid-systolic (b) diameters can be appreciated. An M-mode tracing obtained from a normal coronary sinus at 29 weeks’ gestation demonstrates fluctuations during systole and diastole (c). The cursors are placed on the anterior and posterior walls of the coronary sinus (reproduced with permission from reference 51).

Figure 27.8 Scattergram showing individual measurements, the mean, and 95% confidence interval of the maximum systolic diameter of the coronary sinus with respect to gestational age (y = 0.1373x + 1.6072; R2 = 0.9849) (reproduced from reference 51).

Systolic diameter

5

mm

4

3

2

1 y = 0.1373x + 1.6072 R2= 0.9849 0 21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

Gestational age (weeks)

placental resistance, the relative proportion of left ventricular output increases60–64 (first phase). Decreases in oxygen tension may further increase the proportion of oxygenated umbilical venous blood that is delivered through the ductus venosus to the left side of the heart.65,66 Prolonged

chronic myocardial hypoxemia allows angiogenesis and increases in vascular cross-sectional area and therefore myocardial flow reserve (second phase). These responses constitute chronic heart sparing in IUGR. When acute worsening of cardiovascular status and/or oxygenation is

392

Fetal Cardiology

superimposed, the only mechanism to significantly augment myocardial blood flow is marked coronary vasodilatation, with massive recruitment of coronary vascular reserve (third phase). This vascular response is more acute, often occurring over the course of 24 hours, and is most consistently associated with severe elevation of precordial venous Doppler indices.67,68 The chronic initial phase of heart sparing can be implied by demonstrating certain Doppler abnormalities in the arterial and venous circulations. These include absent or reversed umbilical artery end-diastolic velocity and/or end-diastolic blood flow reversal in the aortic isthmus.69 In the second trimester, the magnitude of coronary blood flow may still be below the visualization threshold 3.5 Coronary sinus diastolic diameter 3

mm

2.5 2 1.5

of ultrasound equipment. Therefore, augmentation of coronary blood flow cannot be documented by spectral Doppler measurement of coronary arteries. With acute worsening of fetal cardiovascular and respiratory status, color and pulsed wave Doppler measurement of coronary artery blood flow is readily achieved as a reflection of maximal augmentation of coronary blood flow – now exceeding the visualization threshold45 (Figure 27.13 and 27.14). In IUGR both diastolic and systolic coronary artery peak blood flow velocities are significantly higher than in appropriately grown fetuses, providing additional evidence of blood flow augmentation. There are no associated changes in the coronary sinus diameter as evidence of increased coronary venous return.70 Since coronary artery blood flow may be visualized in normal and IUGR fetuses at overlapping gestational ages, concurrent examination of arterial and venous circulations and biophysical variables is mandatory to assess fetal status, and clinical management cannot be based on the evaluation of coronary vascular dynamics. In IUGR fetuses with abnormal arterial and venous Doppler and heart sparing, the prognosis is poor, with a high perinatal mortality and a high risk for acidemia and neonatal circulatory insufficiency requiring the highest level of neonatal care.

1 y = 0.0765x + 0.9242 R2= 0.9701

0.5 0

21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 Gestational age (weeks)

Figure 27.9 Scattergram showing individual measurements, the mean, and 95% confidence interval of the maximum diastolic diameter of the coronary sinus with respect to gestational age (y = 0.0765x + 0.9242; R2 = 0.9701) (reproduced with permission from reference 51).

Fetal anemia Severe fetal anemia eventually leads to a reduction of the oxygen-carrying capacity in blood, and subsequently impaired myocardial oxygenation. Fetal hydrops with tricuspid insufficiency and abnormal precordial venous flow is associated with elevated right heart pressures and a decline in coronary perfusion pressure. Under these circumstances, short-term augmentation of myocardial blood flow of 4–5 times basal flow can be achieved through

Figure 27.10

D S

(a)

(b)

The fetal heart is imaged in a lateral four-chamber view and coronary sinus blood flow toward the right atrium (RA) is identified with color Doppler imaging (a). Pulsed wave Doppler shows a triphasic flow profile with a small systolic (S) and a larger diastolic peak (D) followed by brief reversal during atrial contraction (b) (reproduced with permission from reference 48).

Ultrasound examination of the fetal coronary circulation

1.05

Case 1

Diastole Systole

0.85

Case 2

Diastole Systole Diastole Systole

Case 3 m/s

0.65 0.45 0.25 0.05 1

2

3

4 Examination

5

6

Figure 27.11 The graph displays systolic and diastolic peak blood flow measurements in three cases of severe fetal anemia. In cases 1 and 3, velocities were obtained in hydropic fetuses prior to transfusion. In case 1 a hematocrit of 9% was corrected to 39.8%, and in case 3 from 14 to 42.8%. In the second case of maternal trauma, repeat transfusions were necessary on days 1 and 5 for hematocrit levels of 21% and 24%, respectively (reproduced with permission from reference 49).

autoregulation. Coronary artery blood flow has been measured in circumstances of acute fetomaternal hemorrhage, non-immune hydrops, and hemolytic disease.49,71 Under these circumstances, peak diastolic velocities as high as 1 m/s and peak systolic velocities of 0.5 m/s may be observed, significantly exceeding those observed in any other fetal condition. Blood velocities were responsive to maternal oxygen therapy and fetal blood transfusion, and fell below the visualization threshold after normalization of the fetal hematocrit. With the development of fetal hydrops, decreases in coronary sinus dynamics can be observed. This finding is analogous to observations in adults with heart failure, where coronary sinus caliber changes are attenuated, presumably due to elevations in coronary venous pressures.51

Ductus arteriosus constriction Maternal indomethacin therapy for preterm labor carries the risk of constriction of the ductus arteriosus in the fetus. As a conduit for the right ventricle to the systemic circulation, constriction of this vessel raises afterload and therefore cardiac work and oxygen requirement. In severe constriction, tricuspid insufficiency and abnormal venous indices may develop. There is evidence for coronary blood flow augmentation, and color and pulsed wave Doppler measurement of coronary artery blood flow becomes possible. While the peak velocities are not significantly elevated, the gestational age at visualization is determined by the onset of the clinical condition. With resolution of ductus arteriosus constriction following discontinuation

393

of indomethacin, coronary blood flow could no longer be visualized.

Other fetal conditions Acute changes in fetal oxygenation and cardiac pre- and afterload also cause arterial and venous redistribution in favor of the organs essential for fetal life. These ‘heart-, brain-, and adrenal gland-sparing’ phenomena have been described in different animal models. The few observations made by Doppler ultrasound in the human fetus support the presence of the same protective mechanisms. Transient ‘brain- and heart-sparing’ phenomena were observed in a 30-week fetus following acute bradycardia after umbilical fetal blood sampling. Sudden visualization of coronary blood flow, ‘brain-sparing’, and highly pulsatile precordial venous flow persisted for a long period after the 12-minute bradycardia72 (Figure 27.15). Changes in coronary sinus dynamics have been documented in a fetus with supraventricular tachycardia.51 It is likely that more observations of alterations in coronary arterial and venous dynamics will be reported, as familiarity with the examination technique and advances in ultrasound technology facilitate examination.

Clinical applications in fetal cardiac abnormaliies Due to the vascular properties of the coronary arterial circulation, abnormalities frequently develop in cardiac lesions that are associated with disturbed intracardiac pressure/volume relationships during embryonic organogenesis. Due to the embryology of the coronary sinus, abnormalities involving this vessel frequently involve anomalous central venous drainage (both systemic and/or pulmonary). Ultrasound biometry and assessment of coronary sinus dynamics have clinical relevance, and may be the only apparent clue pointing in the direction of such anomalies.

Ventriculocoronary connections in the human fetus Ventriculocoronary connections are frequently noted in fetuses and newborns with pulmonary atresia, hypoplastic right ventricle, intact ventricular septum, or restrictive ventricular septal defect.73 In cases of hypoplastic left heart with aortic atresia, intact ventricular septum and patent mitral valve ventriculocoronary connections may also be present but are less common. The genesis of these vascular abnormalities is discussed above. The abnormal coronary

394

Fetal Cardiology

Decreasing cardiac function

Normal cardiac function AEDF/REDF

+

RV afterload

Centralization

Normalization of PI

LV afterload

Cardiac output

CO shifts towards LV

Fetoplacental perfusion

+

Right atrial pressure

Coronary perfusion pressure

Acute-on-chronic hypoxia

Chronic hypoxemia Myocardial blood flow

Autoregulation

+

Abnormal venous flow

Myocardial blood flow

Vasodilatation

Autoregulation

Coronary vasodilatation Myocardial flow reserve

Normal myocardial flow reserve

Coronary angiogenesis

Figure 27.12 This diagram represents the stages of fetal heart sparing. In the first stage the increased afterload of the right ventricle and decreased afterload of the left ventricle result in redistribution of cardiac output (CO) toward the left ventricle, thus providing increased perfusion to the myocardium. In addition, chronic intrauterine hypoxemia stimulates coronary vasodilatation and coronary angiogenesis. The magnitude of myocardial blood flow enhancement is below the visualization threshold of current ultrasound equipment. In the second stage, coronary angiogenesis results in increasing myocardial flow reserve. In the late stage there is cardiovascular decompensation associated with acute-on-chronic hypoxemia. An altered myocardial vascular bed allows recruitment of a larger myocardial flow reserve and the magnitude of coronary blood flow is now above the visualization threshold. AEDF, absent end-diastolic flow; REDF, reversed end-diastolic flow; PI, pulsatility index (reproduced with permission from reference 61).

channels may provide a conduit to release intraventricular pressures, and may partially avert hypoplasia and fibroelastosis. However, coronary blood flow dynamics may be significantly compromised, impacting on prognosis and approach to postpartum surgical management.74–76 While coronary perfusion may be well maintained in utero, the situation may change after birth. Right ventricular dependent coronary circulation may occur, and result in acute or chronic global myocardial ischemia or infarction due to coronary steal and segmental vascular obstruction. Because of these potential impacts, prenatally detected outflow tract obstructive lesions with relatively preserved ventricular architecture should prompt the search for ventriculocoronary fistula. Prenatal diagnosis of ventriculocoronary fistula is achieved by the demonstration of high velocity bidirectional flow in the coronary artery by color Doppler flow mapping, and verified by pulsed wave Doppler examination.

A severely dilated coronary artery may also be imaged by two-dimensional echocardiography. In cases of right ventricular outflow tract obstruction, diastolic flow from the aortic sinus is directed toward the hypoplastic right ventricle. Pressures are reversed during ventricular systole, and blood flows from the right ventricle to the aorta (Figure 27.16–27.18).73,77,78

Coronary arteriovenous fistula in the human fetus Congenital coronary fistulae may occur rarely if cardiac anatomy is otherwise normal. The majority of these involve a single coronary artery, less often multiple branches. Connections may involve the coronary arterial tree, right atrium, coronary sinus, caval veins, right ventricle, and the

Ultrasound examination of the fetal coronary circulation

395

Figure 27.15 In a fetus with multiple malformations, fetal bradycardia of 56 beats per minute was provoked by percutaneous ultrasounddirected puncture of the umbilical vein for blood sampling at 29 + 4 weeks’ gestation. The blood flow velocity waveform of the right coronary artery after 240 s of bradycardia showed a short systolic peak (peak blood flow velocity = 35 cm/s) and a prolonged diastolic flow (peak blood flow velocity = 69 cm/s) (reproduced with permission from reference 72).

Figure 27.13 Coronary arterial blood flow in a severely growth-restricted fetus. By color Doppler flow mapping, the origin and proximal part of the right coronary artery in a modified short-axis view is visualized at 29 weeks’ of gestation.

coronary fistula connecting to the right ventricle has been reported with demonstration of a progressive increase in size as well as tortuosity of the fistula during gestation.80 A similar case with a fistula between the LCA and the right atrium has also recently been described.81 The shunting blood caused high velocity flow in the dilated coronary sinus. In addition to the prenatal findings, a persistent left superior vena cava and a small ventricular septal defect were also identified postnatally. Following coil embolization of the coronary fistula, the further clinical course was reported as uneventful.

Idiopathic arterial calcification in the human fetus

Figure 27.14 Pulsed wave Doppler waveform of the proximal right coronary artery obtained at 27 weeks’ gestation, showing a higher peak velocity (59 cm/s) during diastole than during systole (43 cm/s). No reverse flow was observed.

pulmonary trunk. Drainage into a low pressure system can result in a large left-to-right shunt, already causing symptoms in childhood such as congestive heart failure, myocardial ischemia from coronary artery steal, right chamber enlargement, arrhythmia, thrombosis with consecutive embolization, and bacterial endocarditis.79 In the majority of cases, symptoms appear in the second and third decade of life. In a 20-week fetus, prenatal detection of an isolated

The idiopathic arterial calcification has an unknown etiology and is characterized by generalized arterial calcification and stenoses, especially of the walls in the arterial trunk of the pulmonary artery and aorta.82,83 Most commonly the coronary arteries are also affected, but peripheral arteries of gastrointestinal tract, liver, kidneys, brain, extremities, and placenta may also be involved. Severe myocardial dysfunction may cause severe fetal hydrops, tissue ischemia, and fetal death in the late second or third trimester.84 In less severe cases, especially in the absence of hydrops, palliative treatment postpartum may be started with steroids and bisphosphonates in order to stop or delay the progression of the disease.85 However, most infants with idiopathic arterial calcification die within the first year of life, complicated by cardiac and pulmonary failure, severe hypertension, renal infarction, peripheral gangrene, and bowel infarction.83

396

Fetal Cardiology

Figure 27.18

Figure 27.16 A fetus with severe pulmonary stenosis and intact interventricular septum at 21 + weeks’ gestation The fourchamber view demonstrated hypoplasia of the right ventricle.

In the same fetus as in Figure 28.15, pulsed wave Doppler examination showed a disturbed to-and-fro blood flow with a systolic flow from the right ventricle to the aortic sinus (S) and a diastolic reversal (D).

thereby decreasing the driving force across the coronary vascular bed. Concurrently, left ventricular work and therefore myocardial oxygen demand are increased. The development of acute heart sparing has been documented in a fetus presenting with severe left ventricular outflow tract obstruction and non-immune hydrops due to critical aortic stenosis. While these findings were ameliorated initially by transplacental digoxin therapy, visualization of coronary blood flow became visible at 39 weeks, coinciding with shunt reversal across the foramen ovale.86

Persistent left superior vena cava

Figure 27.17 Color Doppler imaging demonstrating a ventriculocoronary connection between the right ventricle and the aortic sinus.

Critical aortic stenosis Critical aortic stenosis in fetal life can be associated with a marked decrease in left ventricular output and reversal of shunting across the foramen ovale. Under these circumstances, coronary perfusion pressure is affected by a decrease in arterial and an elevation of right atrial pressure,

While Doppler examination of the coronary sinus has limited utility in the human fetus, substantial dilatation may result from volume overload from a persistent left superior vena cava draining into the coronary sinus.87–89 The frequency of a persistence of the left vena cava is 1–2 per 1000 but may be as high as 9% in the presence of congenital heart defects.90 The degree of dilatation is often marked, and lies appreciably above normal reference limits. This dilatation appears to be predominantly related to vascular volume changes and is independent of associated cardiac defects.68 Other causes of coronary sinus dilatation in the human fetus may be a coronary arteriovenous fistula and anomalous pulmonary vein drainage into the coronary sinus. It is important to note that because of its close proximity to the insertion of the atrioventricular valve, a dilated coronary sinus has been mistaken for an atrial septal defect of ostium primum type and/or an atrioventricular septal defect, respectively.91–93 Coronary sinus dynamics may be attenuated in fetal conditions associated with elevated right heart pressures, severe fetal cardiac dysfunction, and hydrops. These alterations in dynamics may indicate elevated coronary sinus pressures, or changes in coronary blood flow.51

Ultrasound examination of the fetal coronary circulation

References 1. Abuhammad A. Color and pulsed Doppler ultrasonography of the fetal coronary arteries: has the time come for its clinical application? Ultrasound Obstet Gynecol 2003; 21: 423–5. 2. Williams PL, Warwick R, eds. Gray’s Anatomy. New York: Churchill Livingstone, 1983. 3. McAlpine WA. Heart and Coronary Arteries. New York: Springer-Verlag, 1975. 4. Ganong WF. Review of Medical Physiology. Norwalk: Appleton & Lange, 1989. 5. Larsen WJ. Human Embryology. New York: Churchill Livingstone, 1993. 6. Tomanek RJ. Formation of the coronary vasculature: a brief review. Cardiovasc Res 1996; 31: E46–51. 7. Poole TJ, Coffin JD. Vasculogenesis and angiogenesis: two distinct morphogenetic mechanisms establish embryonic vascular pattern. J Exp Zool 1989; 251: 224–31. 8. Hudlicka O, Brown MD. Postnatal growth of the heart and its blood vessels. J Vasc Res 1996; 33: 266–87. 9. Skalak TC, Price RJ. The role of mechanical stresses in microvascular remodeling. Microcirculation 1996; 3: 143–65. 10. Engelmann GL, Dionne CA, Jaye MC. Acidic fibroblast growth factor and heart development. Role in myocyte proliferation and capillary angiogenesis. Circ Res 1993; 72: 7–19. 11. Banai S, Shweiki D, Pinson A et al. Upregulation of vascular endothelial growth factor expression induced by myocardial ischaemia: implications for coronary angiogenesis. Cardiovasc Res 1994; 28: 1176–9. 12. Levy AP, Levy NS, Loscalzo J et al. Regulation of vascular endothelial growth factor in cardiac myocytes. Circ Res 1995; 76: 758–66. 13. Ratajska A, Torry RJ, Kitten GT. Modulation of cell migration and vessel formation by vascular endothelial growth factor and basic fibroblast growth factor in cultured embryonic heart. Dev Dyn 1995; 203: 399–407. 14. Scheel KW, Eisenstein BW, Ingram LA. Coronary, collateral and perfusion territory responses to aortic banding. Am J Physiol 1984; 246: H768–75. 15. Reller MD, Morton MJ, Giraud GD. Maximal myocardial flow is enhanced by chronic hypoxaemia in late gestational fetal sheep. Am J Physiol 1992; 263: H1327–9. 16. Baschat AA, Love JC, Stewart PA, Gembruch U, Harman CR. Prenatal diagnosis of ventriculocoronary fistula. Ultrasound Obstet Gynecol 2001; 18: 39–43. 17. Ascuitto RJ, Ross-Ascuitto NT. Substrate metabolism in the developing heart. Semin Perinatol 1996; 20: 542–63. 18. Bartelds B, Knoester H, Beaufort-Krol GCM et al. Myocardial lactate metabolism in fetal and newborn lambs. Circulation 1999; 99: 1892–7. 19. Fisher DJ, Heymann MA, Rudolph AM. Fetal myocardial oxygen and carbohydrate consumption during acutely induced hypoxemia. Am J Physiol 1982; 242: H657–61. 20. Spahr R, Probst I, Piper HM. Substrate utilization of adult cardiac myocytes. Basic Res Cardiol 1985; 80 (Suppl 1): 53–6. 21. Rudolph AM. Distribution and regulation of blood flow in the fetal and neonatal lamb. Circ Res 1985; 57: 811–21. 22. Kiserud T, Rasmussen S, Skulstad S. Blood flow and the degree of shunting through the ductus venosus in the human fetus. Am J Obstet Gynecol 2000; 182: 147–53.

397

23. Bassenge E. Direct autonomic control of the coronary system. Pflügers Arch Suppl 1978; 373: R6. 24. Krajcar M, Heusch G. Local and neurohumoral control of coronary blood flow. Basic Res Cardiol 1993; 88 (Suppl 1): 25–42. 25. Cannon PJ, Sciacca RR, Fowler DL et al. Measurement of regional myocardial blood flow in man: description and critique of the method using xenon-133 and a scintillation camera. Am J Cardiol 1975; 36: 783–92. 26. Mantero S, Pietrabissa R, Fumero R. The coronary bed and its role in the cardiovascular system: a review and an introductor single-branch model. J Biomed Eng 1992; 14: 109–16. 27. Thornburg KL, Reller MD. Coronary flow regulation in the fetal sheep. Am J Physiol 1999; 277: R1249–60. 28. Guyton AC, Ross JH, Carrier OJ. Evidence for tissue oxygen demand as the major factor causing autoregulation. Circ Res 1964; 14: 60–9. 29. Mosher P, Ross J, McFate PA. Control of coronary blood flow by an autoregulatory mechanism. Circ Res 1964; 14: 250–9. 30. Barnea O, Santamore WP. Coronary autoregulation and optimal myocardial oxygen utilization. Basic Res Cardiol 1992; 87: 290–301. 31. Hoffman JIE. Maximal coronary blood flow and the concept of coronary vascular reserve. Circulation 1984; 70: 153–9. 32. Campbell SE, Kuo CJ, Hebert B et al. Development of the coronary vasculature in hypoxic fetal rats treated with a purified perfluorocarbon emulsion. Can J Cardiol 1991; 7: 234–44. 33. Holmes G, Epstein ML. Effect of growth and maturation in a hypoxic environment on maximum coronary flow rates of isolated rabbit hearts. Pediatr Res 1993; 33: 527–32. 34. Muller JM, Davis MJ, Chilian WM. Integrated regulation of pressure and flow in the coronary microcirculation. Cardiovasc Res 1996; 32: 668–78. 35. Reller MD, Morton MJ, Giraud GD. Severe right ventricular pressure loading in fetal sheep augments global myocardial blood flow to submaximal levels. Circulation 1992; 86: 581–8. 36. Thompson LP, Pinkas G, Weiner CP. Chronic hypoxia inhibits acetylcholine induced vasodilatation in constant pressure perfused fetal guinea pig hearts [Abstract]. J Soc Gynecol Investig 2000; 7 (Suppl 1): 192a. 37. Reller MD, Burson MA, Lohr JL. Nitric oxide is an important determinant of coronary flow at rest and during hypoxemic stress in fetal lambs. Am J Physiol 1995; 169: H2074–81. 38. Sheldon RE, Peeters LLH, Jones MD. Redistribution of cardiac output and oxygen delivery in the hypoxemic fetal lamb. Am J Obstet Gynecol 1979; 135: 1071–8. 39. Behrman RE, Lees MH, Peterson EN, De Lannoy CW, Seeds AE. Distribution of the circulation in the normal and asphyxiated fetal primate. Am J Obstet Gynecol 1970; 108: 956–69. 40. Block BS, Llanos AJ, Creasy RK. Responses of the growthretarded fetus to acute hypoxemia. Am J Obstet Gynecol 1984; 148: 878–85. 41. Reuss MI, Rudolph AM. Distribution and recirculation of umbilical and systemic venous blood flow in fetal lambs during hypoxia. J Dev Physiol 1980; 2: 71–84.

398

Fetal Cardiology

42. Oberhoffer R, Land D, Feilen K. The diameters of coronary arteries in infants without coronary disease. Eur J Pediatr 1989; 148: 389–92. 43. Meyer RA. Kawasaki syndrome: coronary artery disease in the young. Echocardiography 1984; 1: 75–86. 44. Gembruch U, Baschat AA. Demonstration of fetal coronary blood flow by color-coded and pulsed wave Doppler sonography: a possible indicator of severe compromise and impending demise in intrauterine growth retardation. Ultrasound Obstet Gynecol 1996; 7: 10–15. 45. Baschat AA, Gembruch U, Reiss I, Gortner L, Diedrich K. Demonstration of fetal coronary blood flow by Doppler ultrasound in relation to arterial and venous flow velocity waveforms and perinatal outcome—the ‘heart-sparing effect’. Ultrasound Obstet Gynecol 1997; 9: 162–72. 46. Mielke G, Wallwiener D. Visualization of fetal arterial and venous coronary blood flow. Ultrasound Obstet Gynecol 2001; 18: 407. 47. Yoo SJ, Min JY, Lee YH. Normal pericardial fluid in the fetus: color and spectral Doppler analysis. Ultrasound Obstet Gynecol 2001; 18: 248–52. 48. Baschat AA, Gembruch U. Evaluation of the fetal coronary circulation. Ultrasound Obstet Gynecol 2002; 20: 405–12. 49. Baschat AA, Muench MV, Gembruch U. Coronary artery blood flow velocities in various fetal conditions. Ultrasound Obstet Gynecol 2003; 21: 426–9. 50. Rein AJ, Nir A, Nadjari M. The coronary sinus in the fetus. Ultrasound Obstet Gynecol 2000; 15: 468–72. 51. Abello KC, Stewart PA, Baschat AA. Grey scale and M-Mode echocardiography of the fetal coronary sinus. Ultrasound Obstet Gynecol 2002; 20: 137–41. 52. Baschat AA, Gembruch U. Examination of fetal coronary sinus blood flow by Doppler ultrasound. Ultrasound Obstet Gynecol 1998; 11: 410–14. 53. Chaoui R. The fetal coronary system in prenatal diagnosis [Abstract]. Ultrasound Obstet Gynecol 1996; 8 (Suppl): 158. 54. Mundigler G, Zehetgruber M, Christ G. Comparison of transesophageal coronary sinus and left anterior descending coronary artery Doppler measurements for the assessment of coronary flow reserve. Clin Cardiol 1997; 20: 225–31. 55. Zehetgruber M, Mundigler G, Christ G. Estimation of coronary flow reserve by transesophageal coronary sinus Doppler measurements in patients with syndrome X and patients with significant left coronary artery disease. J Am Coll Cardiol 1995; 25: 1039–45. 56. Hecher K, Campbell S, Doyle P. Assessment of fetal compromise by Doppler ultrasound investigation of the fetal circulation. Circulation 1995; 91: 129–38. 57. Weiner CP. The relationship between the umbilical artery systolic/diastolic ratio and umbilical blood gas measurements in specimens obtained by cordocentesis. Am J Obstet Gynecol 1990; 162: 1198–202. 58. Nicolaides KH, Bilardo CM, Soothill PW. Absence of enddiastolic frequencies in umbilical artery: a sign of fetal hypoxia and acidosis. Br Med J 1988; 297: 1026–7. 59. Hecher K, Snijders R, Campbell S. Fetal venous, intracardiac and arterial blood flow measurements in intrauterine growth retardation: relationship with fetal blood gases. Am J Obstet Gynecol 1995; 173: 10–15. 60. Rizzo G, Capponi A, Arduini D. Doppler indices from inferior vena cava and ductus venosus in predicting pH and oxygen

61.

62.

63.

64.

65.

66.

67.

68.

69.

70.

71.

72.

73.

74.

75.

76.

77.

tension in umbilical blood at cordocentesis in growth-retarded fetuses. Ultrasound Obstet Gynecol 1996; 7: 401–10. Baschat AA, Gembruch U, Harman CR. Coronary blood flow in fetuses with intrauterine growth restriction. J Perinat Med 1998; 26: 143–56. Al-Ghazali W, Chita SK, Chapman MG, Allan LD. Evidence of redistribution of cardiac output in asymmetrical growth retardation. Br J Obstet Gynaecol 1989; 96: 697–704. Reed KL, Anderson CF, Shenker L. Changes in intracardiac Doppler flow velocities in fetuses with absent umbilical artery diastolic flow. Am J Obstet Gynecol 1987; 157: 774–9. Rizzo G, Arduini D. Fetal cardiac function in intrauterine growth retardation. Am J Obstet Gynecol 1991; 165: 876–82. Tchirikov M, Eisermann K, Rybakowski C, Schröder HJ. Doppler ultrasound evaluation of ductus venosus blood flow during acute hypoxemia in fetal lambs. Ultrasound Obstet Gynecol 1998; 11: 426–31. Tchirikov M, Rybakowski C, Hunecke B, Schröder HJ. Blood flow through the ductus venosus in singleton and multifetal pregnancies and in fetuses with intrauterine growth restriction. Am J Obstet Gynecol 1998; 178: 943–9. Baschat AA, Gembruch U. Triphasic umbilical venous blood flow with prolonged survival in severe intrauterine growth retardation: a case report. Ultrasound Obstet Gynecol 1996; 8: 201–5. Baschat AA, Gembruch U, Reiss I et al. Coronary blood flow visualization indicates cardiovascular compromise in IUGR fetuses. Ultrasound Obstet Gynecol 2000; 16: 425–31. Makikallio K, Jouppila P, Rasanen J. Retrograde net blood flow in the aortic isthmus in relation to human fetal arterial and venous circulations. Ultrasound Obstet Gynecol 2002; 19: 147–52. Chaoui R, Heling KS, Kalache KD. Caliber of the coronary sinus in fetuses with cardiac defects with and without left persistent superior vena cava and in growth restricted fetuses with ‘heart sparing effect’. Prenat Diagn 2003; 23: 552–7. Baschat AA, Harman CR, Alger LS, Weiner CP. Fetal coronary and cerebral blood flow in acute fetomaternal hemorrhage. Ultrasound Obstet Gynecol 1998; 12: 128–31. Gembruch U, Baschat AA. Circulatory effects of acute bradycardia in the human fetus as studied by Doppler ultrasound. Ultrasound Obstet Gynecol 2000; 15: 424–7. Maeno YV, Boutin C, Hornberger LK et al. Prenatal diagnosis of right ventricular outflow tract obstruction with intact ventricular septum and detection of ventriculocoronary connections. Heart 1999; 81: 661–8. Coles JG, Freedom RM, Lightfoot NE et al. Long-term results in neonates with pulmonary atresia and intact ventricular septum. Ann Thorac Surg 1989; 47: 213–17. Giglia TM, Mandell VS, Connor AR, Mayer JEJ, Lock JE. Diagnosis and management of right ventricle-dependent coronary circulation in pulmonary atresia with intact ventricular septum. Circulation 1992; 86: 1516–28. Hanley FL, Sade RM, Blackstone EH et al. Outcomes in neonatal pulmonary atresia with intact ventricular septum. A multiinstitutional study. J Thorac Cardiovasc Surg 1993; 105: 406–27. Arabin B, Aardenburg R, Schasfoort-van Leeuwen M, Elzenga N. Prenatal diagnosis of ventriculocoronary arterial

Ultrasound examination of the fetal coronary circulation

78.

79.

80.

81.

82.

83. 84.

85.

communications combined with pulmonary atresia. Ultrasound Obstet Gynecol 1996; 7: 461–2. Chaoui R, Tennstedt C, Göldner B, Bollmann R. Prenatal diagnosis of ventriculo-coronary communications in a second-trimester fetus using transvaginal and transabdominal color Doppler sonography. Ultrasound Obstet Gynecol 1997; 9: 194–7. Liberthson RR, Sagar K, Berkoben JP, Weintraub RM, Levine FH. Congenital coronary arteriovenous fistula: report of 13 patients, review of the literature and delineation of management. Circulation 1979; 59: 849–54. Sharland GK, Tynan M, Qureshi SA. Prenatal detection and progression of right coronary artery to right ventricle fistula. Heart 1996; 76: 79–81. Mielke G, Sieverding L, Borth-Bruns T et al. Prenatal diagnosis and perinatal management of left coronary artery to right atrium fistula. Ultrasound Obstet Gynecol 2002; 19: 612–15. Hajdu J, Marton T, Papp C, Hruby E, Papp Z. Calcification of the fetal heart—four case reports and a literature review. Prenat Diagn 1998; 18: 1186–90. Crum AK, Lenington W, Jeanty P. Idiopathic infantile arterial calcifications. Fetus 1992; 2: 74781–6. Juul S, Ledbetter D, Wright TN, Woodrum D. New insights into idiopathic infantile arterial calcinosis. Am J Dis Child 1990; 144: 229–33. Bellah RD, Zawodniak L, Librizzi RJ, Harris MC. Idiopathic arterial calcification of infancy: prenatal and

86.

87.

88.

89.

90.

91.

92.

93.

399

postnatal effects of therapy in an infant. J Pediatr 1992; 121: 930–3. Schmider A, Henrich W, Dahnert I, Dudenhausen JW. Prenatal therapy of non-immunologic hydrops fetalis caused by severe aortic stenosis. Ultrasound Obstet Gynecol 2000; 16: 275–8. Snider AR, Ports TA, Silverman NH. Venous anomalies of the coronary sinus: detection by M-mode, two-dimensional and contrast echocardiography. Circulation 1979; 60: 721–7. Rostagno C, Diricatti G, Galanti G et al. Partial anomalous venous return associated with intact atrial septum and persistent left superior vena cava: a case report and literature review. Cardiologia 1999; 44: 203–6. Papa M, Camesasca C, Santoro F et al. Fetal echocardiography in detecting anomalous pulmonary venous connection: four false positive cases. Br Heart J 1995; 73: 355–8. Nsah EN, Moore GW, Hutchins GM. Pathogenesis of persistent left superior vena cava with a coronary sinus connection. Pediatr Pathol 1991; 11: 261–9. Gembruch U, Knöpfle G, Bald R, Hansmann M. Early diagnosis of fetal congenital heart diseases by transvaginal echocardiography. Ultrasound Obstet Gynecol 1993; 3: 310–17. Allan LD, Sharland GK. The echocardiographic diagnosis of totally anomalous pulmonary venous connection in the fetus. Heart 2001; 85: 433–7. Park JK, Taylor DK, Skeels M, Towner DR. Dilated coronary sinus in the fetus: misinterpretation as an atrioventricular canal defect. Ultrasound Obstet Gynecol 1997; 10: 126–9.

28 Fetal cardiac tumors Mary T Donofrio and Gerard R Martin

Introduction Advancements in fetal ultrasound imaging have enhanced the ability to diagnose and follow fetuses with cardiac abnormalities from early in gestation. Prenatally, cardiac tumors can present as an isolated cardiac mass, or be associated with fetal dysrhythmias, pericardial effusion, or hydrops fetalis. Four histologic types of cardiac tumors have been recognized prenatally and include rhabdomyomas, teratomas, hemangiomas, and fibromas. Myxomas, lipomas, papillary tumors, sarcomas, and metastatic tumors, despite their presence in adults, are not typically found in the fetus.1 Advanced imaging technologies including three-dimensional (3D) ultrasound and cardiac magnetic resonance imaging have been utilized postnatally in an attempt to better define cardiac tumors in infants and children. These techniques offer promise for better characterization and diagnosis of fetal cardiac tumors so that as the window into the uterus becomes clearer, our understanding of these tumors will become more complete (Figure 28.1, Video clips 28.1 and 28.2).

Cardiac tumors in children and adults Cardiac tumors are rare.2–5 In autopsy studies including all age groups, the prevalence of primary cardiac tumors ranged from 0.0017 to 0.28%.2 Similarly, the occurrence in a study of infants and children revealed a prevalence of 0.027% in 11 000 autopsies.4 In a retrospective review from a single institution over 20 years, the prevalence of cardiac tumors in children referred for evaluation was found to be 0.08% among all patients evaluated.5 In another study of children, three tumor subtypes accounted for 97% of the tumors diagnosed. Rhabdomyomas occurred in 60%, followed by teratomas in 25% and fibromas in 12%.6 This differs from adults, for whom the most common tumors diagnosed are myxomas.7

Fetal cardiac tumors Prevalence The first reported case of a cardiac tumor diagnosed in a fetus was described in 1982.8 More recently, many centers have reported their experience with the diagnosis and management of cardiac tumors in utero.1,9–34 As in infants and children, the most common cardiac tumors found in fetuses are rhabdomyomas. Fibromas, teratomas, and hemangiomas have also been described. In a multicenter review, the prevalence of cardiac tumors was found to be 0.14% in pregnancies referred for fetal echocardiography.9 Rhabdomyomomas were the most common, occurring in 89%, followed by fibromas and hemangiomas. In another review of multiple studies reported in the literature, results were similar, with 64% of the 89 fetuses diagnosed having rhabdomyomas, followed by teratomas, fibromas, and hemangiomas.1

Fetal echocardiography in the diagnosis of cardiac tumors The most common reason for referral for fetal echocardiography in fetuses found to have a cardiac tumor is an abnormal obstetrical ultrasound scan or documentation of a fetal dysrhythmia.1,9,13,14,27,31 Other indications for referral include hydrops fetalis, the presence of a pericardial effusion, or a family history of tuberous sclerosis.17–24,28–30,35 A complete fetal echocardiogram should be performed in all fetuses referred for evaluation. The initial assessment should include an examination to determine whether there is congenital heart disease or an acquired heart abnormality caused by alterations in cardiac blood flow. Assessment of the tumor should include a complete description of the mass and the number of tumors present. The location of the mass or masses within the cardiac

402

Fetal Cardiology

(a)

(b)

Figure 28.1 (a) Two-dimensional (2D) fetal echocardiographic image of a cardiac tumor in the right ventricle (RV). (b) Three-dimensional (3D) fetal echocardiographic image of the same tumor. In real time the tumor appeared to be attached to the tricuspid valve apparatus. RA, right atrium; LV, left ventricle; T, tumor.

Figure 28.2 Ultrasound image of pericardial and pleural effusions, ascites, and skin edema in a fetus with a large rhabdomyoma adhered to the left ventricular free wall. A, ascites; P, pleural effusion; S, skin edema.

chambers, the septae, within the pericardial space, or external to the heart should be determined. The echogenicity of the mass should be compared to external structures. The mass should be described as homogeneous or heterogeneous, vascular, or cystic. The margins of the mass should be identified and described as smooth, irregular, or lobulated. Quantitative measurement of the size of the mass should be made so that objective serial follow-up can be done to determine tumor growth or regression. Careful assessment of the hemodynamic sequelae of the tumor should be undertaken. Cardiac size and determination of whether there is cardiomegaly by measuring the circumference of heart compared to chest circumference should be recorded. Cardiac function should be

evaluated, and the presence or absence and size of pericardial and/or pleural effusions should be documented (Figure 28.2). Doppler evaluation of blood flow in and out of the heart should be performed. The atrioventricular and semilunar valves as well as the venous return to the heart and the arteries leaving the heart should be interrogated to determine whether insufficiency or obstruction is present (Figure 28.3, Video clips 28.3–28.6). Reversal of flow in the inferior vena cava or ductus venosus may indicate impending hydrops fetalis, and dampening of the inferior vena cava waveform has been described to precede cardiac failure18 (Figure 28.4). Fetal cardiac rate and rhythm should be analyzed from inflow/outflow Doppler signals, and by M-mode echocardiography (Figure 28.5). Cardiac masses can cause dysrhythmias.13,14,27,31 If heart rate or rhythm abnormalities are found, a period of observation should be performed to estimate the significance of the dysrhythmia. Referral to a high-risk obstetrician for an assessment of fetal wellness may be helpful in determining the functional significance of the tumor. In all cases, serial imaging should be performed to track the growth or regression of the mass over time, as well as note changes in the cardiac hemodynamics including heart rate, rhythm, and function. Advanced ultrasound technology using 3D imaging of the fetal heart is currently limited in clinical practice by the time it takes to acquire data sets and the inability to gate images to the fetal heart rate. Currently there are no reports in the literature showing improved diagnostic capabilities using this technology. Three-dimensional imaging offers the advantage of better spatial resolution and assessment of structures in multiple simultaneous planes, which is very useful in evaluating the extent

Fetal cardiac tumors

RV

RV

RA

T

403

TR

T LA LV

LV

(a)

(b)

LA

RA

LA TR

RV

LV T

(c)

(d)

Figure 28.3 (a) Fetal echocardiographic image of a right ventricular tumor. (b) Color Doppler imaging in the fetus reveals severe tricuspid valve insufficiency. (c) Postnatal 2D echocardiographic image of the tumor. (d) Color Doppler imaging in the newborn reveals severe tricuspid valve insufficiency.

Figure 28.5 Figure 28.4 Inferior vena cava Doppler ultrasound in a fetus with a large right atrial cardiac tumor. Note the dampened waveform and reversed flow with atrial systole. a, atrial contraction.

M-mode ultrasound of a premature atrial contraction (PAC) leading to supraventricular tachycardia (SVT) in a fetus with a cardiac tumor in the atrium.

404

Fetal Cardiology

T LA

T

RA

RV T

LV

RA LV T T

Figure 28.6 Fetal 3D ultrasound image of multiple cardiac rhabdomyomas. Note the tumors in the right atrium, right ventricle, left ventricle, and septum (reproduced with permission from reference 36).

and hemodynamic consequences of cardiac tumors. As image acquisition becomes faster, this modality will likely become more widely utilized in the fetus (Figures 28.6 and 28.7, Video clip 28.7).

Fetal magnetic resonance imaging in the diagnosis of cardiac tumors Magnetic resonance imaging (MRI) of the fetus is now more frequently utilized in clinical practice; however, imaging of the heart is limited due to cardiac motion and the inability to gate images to the heart rate. Two studies reported in the literature show that imaging of fetal cardiac tumors is possible, though the images obtained were not superior to ultrasound.37,38 Magnetic resonance imaging has several advantages over ultrasound. It offers the ability to image through tissues with poor acoustic properties, there is the potential for 3D reconstruction, and volume calculations can be made and tracked. As the technology develops, MRI will be very useful in the assessment of the fetal heart and fetal cardiac tumors.

Additional considerations and family counseling A comprehensive obstetrical ultrasound scan should be performed in fetuses with cardiac tumors. Amniocentesis should be considered in the presence of additional abnormal findings noted in the fetus. If the tumor or tumors have characteristics of a rhabdomyoma, family members should be evaluated for signs of tuberous sclerosis. In the

Figure 28.7 Postnatal 3D ultrasound image of a cardiac rhabdomyoma in the right ventricle. The tumor is attached to the tricuspid valve apparatus (note that this is the same patient as in Figure 28.3).

fetus, MRI may be helpful in identifying cerebral tubers and other systemic findings of tuberous sclerosis.38,39 In a study of eight patients, five fetuses had hyperintense subependymal and cortical nodules on T1 weighted images of the brain39 (Figure 28.8). Families should be counseled regarding the limitations of the fetal ultrasound diagnosis of tuberous sclerosis. Several cases have been reported in which termination of pregnancy was chosen based on the possibility of tuberous sclerosis after a provisional diagnosis of cardiac rhabdomyoma was made. Necropsy in these cases revealed dystrophic calcification and no evidence of rhabdomyomas.40 In addition, tuberous sclerosis has variable expression, with some children suffering from severe neurologic and renal abnormalities, and others going relatively unaffected. Cardiac rhabdomyomas are the first identifiable marker of tuberous sclerosis in the fetus; however, their presence does not correlate with the degree of other organ involvement, which often does not become apparent until much later.9

Prenatal management Clinical follow-up without intervention is warranted as long as the fetus remains asymptomatic and there is no evidence for hydrops fetalis or fetal distress. Evaluation should include monitoring of fetal wellness including an assessment of fetal growth, amniotic fluid volume, and the development of hydrops fetalis. Fetal echocardiography should be performed to assess the mass, and the potential for progressive hemodynamic compromise from obstruction, significant dysrhythmias, or cardiac failure. In a series of cases of large tumors diagnosed prenatally,41 though the

Fetal cardiac tumors

Figure 28.8 Magnetic resonance image of the brain in a fetus with cardiac rhabdomyomas. Note the tubers demarcated by an arrowhead.

size of the tumor at initial presentation was impressive, the more important clinical finding was the position of the tumor and the effect it had on blood flow. In one case, mitral inflow and left ventricular outflow were obstructed; however, outflow from the right ventricle was maintained. Bidirectional flow at the foramen ovale likely caused by an elevation in left atrial pressure was documented. The prenatal and postnatal findings of a borderline left ventricle, small aorta, and coarctation of the aorta were thought to be due to alterations in in utero blood flow, supporting the concept that chamber, valve, and vessel growth are at least partially determined by blood flow through the fetal heart. The fetus was not compromised and was delivered near term (Figure 28.9). In a second case, the tumor compressed both right and left ventricular outflow tracts, and despite an adequate foramen ovale there was no place for egress of blood out of the heart. This led to poor cardiac output and elevated atrial and caval pressures, which then resulted in hydrops fetalis, premature delivery, and postnatal death (Figure 28.10). In a third case, the large tumor in the right atrium transiently obstructed systemic venous return, elevating caval pressures, and resulted in early signs of hydrops fetalis. Later in gestation, redirection of blow flow around the tumor was seen and the symptoms of hydrops fetalis did not progress (Figure 28.11, Video clips 28.8 and 28.9). The in utero and postnatal bradycardia was believed to be caused by compression of the sinus node by the tumor, and was not thought to be a result of in utero distress. Obstruction to fetal blood flow can occur at any point in gestation, and depends on both the size and the location

405

of the tumor in relation to all of the cardiac structures. Serial fetal assessment throughout gestation is indicated, particularly after the point of postnatal viability, to assess the hemodynamic effects that the tumor has on the heart. If obstruction to blood flow and/or early fetal compromise is noted, then the decision of whether to deliver early must be made. At birth, if obstruction to blood flow persists, surgery can be undertaken, keeping in mind that the natural history of these tumors is to shrink and become clinically less important over time. Since some fetal masses regress, there may be benefit for delaying intervention in cases where there are only mild symptoms. These fetuses should be followed very closely. If a pericardial effusion is present, particularly in cases of pericardial hemangiomas or teratomas, the fluid may be drained so that delivery can be delayed.42 Fetal tachyarrhythmias associated with cardiac masses have been reported to respond to maternal administration of digoxin.27 Other antiarrhythmic agents such as flecainide or sotalol can also be considered. In cases of clinically significant lesions causing fetal compromise, early delivery and possible surgical intervention to remove or debulk the tumor before irreversible cardiac failure and death occur may be inevitable. If delivery is unavoidable, planning should be made to deliver the baby in a tertiary care center with extracorporeal membrane oxygenation (ECMO) and a cardiac surgeon available.

Outcome Overall, the survival of fetuses diagnosed with cardiac tumors has improved; however, the clinical course depends on the type of tumor and its hemodynamic effects on the fetus. A recent analysis of 224 fetuses and neonates with cardiac tumors collected from the literature revealed that even though outcomes have improved in recent years, mortality is still high, with only 55% surviving. Of the 89 diagnosed prenatally, survival was 66%, versus only 47% in those diagnosed postnatally. Survival depended on the type of tumor in addition to the size and associated findings. Fetuses with hemangiomas did best, with an 83% survival. Fetuses with rhabdomyomas had a 67% survival. Fetuses with fibromas did worst, with only a 50% survival.1

Rhabdomyomas Rhabdomyomas are the most common of the cardiac tumors, constituting 89% of all the fetal cardiac tumors in one large, multicenter study.9 In another study that reviewed multiple reports in the literature, the prevalence was similar at 64%.1 Rhabdomyomas typically present as an incidental finding on obstetrical ultrasound, either

406

Fetal Cardiology

Figure 28.9 (a) Echocardiographic image in the tranverse plane in the same fetus shows the tumor compressing the left ventricle. (b) Image in the transverse plane shows the tumor compressing the left ventricle. (c) Fetal echocardiographic image in the same fetus showing the small aorta and likely coarctation. (d) Postnatal echocardiographic image shows the tumor compressing the left ventricle. (e) Postnatal echocardiographic image in the same patient showing the discrete coarctation.

routine or during evaluation for a fetal dysrhythmia or a family history of tuberous sclerosis. The mass or masses are usually sessile, and the majority are located in the interventricular septum or free wall of the right or left ventricles. Occasionally masses can be found near the atrioventricular groove, in a papillary muscle, or in the atria. Rhabdomyomas are usually multiple, smooth, and

lobulated. If the masses are clustered, irregular and fragmented, the diagnosis of rhabdomyosarcoma should be considered.43 On ultrasound examination, rhabdomyomas have been described as homogeneous, diffusely echogenic, and more echogenic than myocardium.9,15,32,34 Approximately 50% of rhabdomyomas are intracavitary. Inflow or outflow tract obstruction may result. Dysrhythmias and

Fetal cardiac tumors

407

protoplasmic strands which extend outward from a centrally located nucleus generating the pathognomonic ‘spider cell’ appearance.47

Tuberous sclerosis

Figure 28.10 Echocardiographic image in a fetus with a large cardiac tumor that caused outlet obstruction of the left and right ventricles. Note the pleural effusions (P) and skin edema (S) (note that this is the same patient as in Figure 28.2).

Wolff–Parkinson–White syndrome have been described in patients with rhabdomyomas.44 Rhabdomyomas tend to increase in size prenatally and then regress after birth.34,45,46 The in utero natural history and progression or regression of cardiac rhabdomyomas is not well defined. There is a report in the literature of a fetus with a normal cardiac ultrasound at 20 weeks of gestation, that later in gestation had an ultrasound demonstrating a tumor.19 In another study, nine fetuses with rhabdomyomas were followed to term. In 67% of those evaluated, growth of the smaller tumors was proportional with gestational age until 30–32 weeks and then stable until term. In the remaining 33% of fetuses, larger tumors grew disproportionately and additional smaller masses appeared, in some cases leading to partial obstruction of the outflow tracts. At birth, the rhabdomyomas regressed at least partially in all patients. Tuberous sclerosis was diagnosed in 81% of the total group, including four fetuses that were terminated.46 These data suggest that maternal hormones or other in utero environmental factors may play a role in the development and growth of fetal rhabdomyomas. Whether the tumors are present early in gestation and are too small to be detected, or whether they develop later in the pregnancy, is not clear. In one study, rhabdomyomas present at birth spontaneously regressed in over half of the children diagnosed.32

Pathology Rhabdomyomas are benign tumors that are well-circumscribed, non-encapsulated, and grayish white. When found within the heart they occur in a wide variety of sizes and shapes. On histologic examination, rhabdomyomas are composed of large, round vacuolated cells with delicate

The association of rhabdomyomas with tuberous sclerosis is well documented.17,22,28–35,48,49 Tuberous sclerosis has been found in 100% of cases of fetuses with multiple rhabdomyomas, and approximately 50% of fetuses with single tumors and no family history. Manifestations of tuberous sclerosis include mental retardation, seizures, tumor-like hamartomatous nodular malformations of the brain and visceral organs, skin hypopigmentation, and often the presence of the disease in a parent. When the prenatal diagnosis of a cardiac tumor is made and a rhabdomyoma is suspected, a careful search for additional small tumors should be performed, since this has major implications for prognosis. There is no clear association between cardiac rhabdomyomas and other manifestations of tuberous sclerosis, though in one study approximately 50% of patients with multiple cardiac tumors had other organ system involvement, particularly in the central nervous system.9 Congenital heart disease, including tetralogy of Fallot and hypoplastic left heart syndrome, has been seen in patients with tuberous sclerosis.9,25,50 Tuberous sclerosis is an autosomal dominant condition. It is caused by mutations in the TSC1 or TSC2 genes which encode the tumor suppressors hamartin and tumerin. TSC1 has been mapped to chromosome 9q, and TSC2 to chromosome 16p.51 At present, prenatal genetic testing is available for families in which the disease-causing allele has been identified in an affected family member. In these cases, DNA analysis of fetal cells obtained by chorionic villus sampling at 10–12 weeks’ gestation or amniocentesis at 15–18 weeks’ gestation can be performed. In cases without a family history, the presence of multiple intracardiac rhabdomyomas should be considered sufficient to make the diagnosis.48

Management and outcome Early reports suggest a high incidence of hemodynamic compromise and neonatal death in infants diagnosed with rhabdomyomas. One study reported a 50% mortality by 6 months of age and an 80% mortality by 1 year.47 As a result, aggressive intervention was at one time advocated.52 More recent reports document minimal, if any, hemodynamic compromise in utero, and even spontaneous resolution of rhabdomyomas over time, including in utero resolution.9,32,45,49 This may be due to increased detection of more mild cases which were previously missed. In addition, early recognition and improved surgical treatment when necessary have significantly improved the

408

Fetal Cardiology

Figure 28.11 (a) Echocardiographic image in a fetus with a large right atrial tumor. Note the dilated right atrium and pericardial effusion. (b) Postnatal echocardiographic image of the tumor. Note that multiple tumors are seen in the right ventricle making the diagnosis of multiple rhabdomyomas and tuberous sclerosis.

prognosis in more recent years. In one review of 20 fetuses with cardiac rhabdomyomas, one fetus died in utero, one was terminated, and 18 continued to term.32 In utero, no fetus had evidence of cardiovascular compromise or hydrops fetalis. At delivery, 10 of 18 had abnormal electrocardiograms, and nine had dysrhythmias. Three of 18 underwent surgery for left ventricular outflow tract obstruction and one required therapy for heart failure. In another study that analyzed multiple reports in the literature, of the 57 fetuses diagnosed prenatally with rhabdomyomas, 67% survived. Of note is that seven were stillbirths and 46 had tuberous sclerosis or a family history of the disease.1 Most centers take a conservative approach of nonintervention in fetuses diagnosed with rhabdomyomas unless hemodynamic compromise is found. Lesions which are very large, intracavitary, and/or causing dysrhythmias or obstruction are the most likely to lead to cardiovascular compromise. These are the fetuses that should be most carefully followed. Postnatally, the management of newborns with cardiac rhabdomyoma varies depending on the size, location, compression of structures within the mediastinum, and the arrhythmogenic capacity of the tumor. Dysrhythmias can typically be treated with medication or, if necessary, resection of the tumor. Persistent inflow or outflow tract obstruction is a complication that is most likely to require surgical intervention.53,54

Fibromas are likely to be diagnosed as an incidental finding of a mass on obstetrical ultrasound, or are identified as a result of a referral for a fetal dysrhythmia. Fibromas are almost always single, and are typically located in the left ventricular myocardium or interventricular septum. They may occur in the right ventricle, and occasionally in the right atrium. Ultrasound examination shows the tumor to be of uniform echogenicity and therefore it may be indistinguishable from a rhabdomyoma. These tumors may undergo central cystic degeneration and calcification, and therefore often appear heterogeneous on examination. Associated pericardial effusions have been described. Fibromas may cause inflow or outflow tract obstruction, congestive heart failure, and/or supraventricular or ventricular dysrhythmias.1,9,54,55 Fibromas are not associated with congenital heart disease or other genetic conditions.

Pathology Fibromas are round, non-encapsulated, firm, white tumors. Histology shows benign lesions composed of fibroblasts, collagen fibers, and elastic tissue. The central portion of the tumor often has multiple areas of calcium deposits.56

Outcome Fibromas Fibromas represented 5% of all fetal cardiac tumors detected in a large multicenter study.9 In a review of multiple reports, fibromas occurred in 7% of fetuses.1

The outcome for fetuses with fibromas is not good. In a study that analyzed multiple reports in the literature,1 of the 57 fetuses diagnosed prenatally with fibromas, 50% survived. In the overall group, the survival of fetuses and neonates was only 29%. Surgical intervention was

Fetal cardiac tumors

409

undertaken in 17 of the delivered neonates; four were transplanted. There were no survivors in those that did not have surgery. Death was most often due to dysrhythmias or heart failure. The strategy of surgical intervention for this group is variable. Complete resection has been successful if it is technically feasible.57

when clinically necessary to prevent tamponade has been described.42 Hemangiomas may regress after birth, so intervention should only be considered in situations where there is significant compromise that cannot be medically managed.1

Hemangiomas

Teratomas

Hemangiomas are uncommon in fetuses and children. In one study, hemangiomas accounted for only 5% of cardiac masses in a population referred for fetal echocardiography.9 In a review of multiple reports, similarly the prevalence of vascular tumors was found to be 7%.1 Typically hemangiomas can be found at the base of the heart or adjacent to the right atrium. They may also arise at any site in the pericardium, myocardium, or the ventricular cavity. These tumors often have an intracavitary component which may cause obstruction, and may also have an associated pericardial effusion. The heart is often displaced into the left hemithorax by the tumor mass. Hemangiomas may also invade the atrioventricular node, and varying degrees of heart block may be seen. Hemangiomas are of mixed echogenicity, representing endothelial cells in various stages of organization and thrombosis.

The majority of teratomas are extracardiac, intrapericardial, and attached to the aortic root or pulmonary artery.10,18,24,26 Intracardiac teratomas are rare. In a review of multiple reports in the literature, teratomas were the second most common cardiac tumor, accounting for 23% of the tumors identified in fetuses.9 These tumors typically have a mixed echogenicity on ultrasound evaluation, representing the heterogeneity of the tumor. They often have associated pericardial effusions. Most teratomas cause symptoms with either non-immune hydrops fetalis or respiratory distress in the neonatal period.

Pathology Hemangiomas are benign. There are two clinical and histologic variants. Both capillary and cavernous hemangiomas have been described in the fetal heart.11,16 Capillary hemangiomas are composed of small vascular channels which traverse within the myocardium. Histologic examination shows closely packed aggregations of thin-walled capillaries. Cavernous hemangiomas are composed of blood vessels with large vascular channels. Intravascular thrombosis, central necrosis, and calcification can occur within these tumors.56

Outcome The outcome for fetuses and neonates with hemangiomas has been reported to be better than for most tumors. Hemangiomas may cause hemodynamically significant pericardial effusions, and for this reason surgical intervention may be warranted after birth.11,16 In a review of multiple reports in the literature,1 the survival of fetuses diagnosed with vascular tumors was 83%, the most favorable for all the subtypes of benign cardiac tumors. In this review, the most common associated clinical finding was a pericardial effusion. Drainage of fetal pericardial effusions

Pathology Teratomas typically have a benign histology, but can be malignant. They are usually single, firm, and encapsulated. They contain cysts within a mucoid stroma. Intrapericardial teratomas consist of early embryonic tissue derived from three germ layers. The mesodermal tissue includes muscle, hyaline, and elastic cartilage. The endodermal tissue includes bronchial, pancreatic, intestinal, and salivary glands. The ectodermal neuroepithelial tissue includes the choroid plexus and eyes.56

Outcome In a review of reports from multiple centers, fetuses and neonates diagnosed with teratomas were reported to have the second highest survival rate of all benign cardiac tumors at 75%.1 In those diagnosed prenatally, survival was slightly worse at 65%, albeit similar to that of rhabdomyomas. Of note is that six of the 20 fetuses diagnosed with teratomas in this study were stillborn, accounting for 86% of the overall deaths. Surgical resection of the teratoma was performed in 31 of 40 infants, or 78%, and was curative in all but one. These results are similar to other reported studies in which survival is much better for teratomas diagnosed and excised postnatally, than for those diagnosed prenatally.58 Of note is that pericardial effusions are often seen with these tumors. Draining the effusion either in utero or postnatally is recommended if it is clinically significant.58

410

Fetal Cardiology

abnormal biochemical markers, and/or other ultrasound markers were present in all fetuses found to have aneuploidy. The evaluation of these parameters therefore should be used to determine the risk of aneuploidy for fetuses with echogenic foci.64

Conclusion

Figure 28.12 Fetal echocardiographic image showing a benign echogenic focus in the left ventricle (asterisk).

Echogenic foci within the fetal heart Although echogenic intracardiac foci are generally thought to be normal, fetuses with these bright spots in their heart are often referred for evaluation to determine whether they are true cardiac tumors. Benign echogenic foci are almost always found in the left ventricle within a papillary muscle. In one study, left ventricular echogenic foci were identified in 20% of second and third trimester fetuses.59 These foci may be distinguished from true tumors in that they are usually smaller and intensely echogenic (Figure 28.12). Atrial echogenic foci have also been described as a normal variant.60

Pathology The only consistent histologic finding of echogenic foci is mineralization within a papillary muscle, though pathologic studies of these lesions diagnosed in utero have been limited.40,61 Autopsies done on fetuses with abnormal chromosomes (most frequently trisomy 13 and 21) have shown papillary calcification in 15%, compared to 2% of genetically normal fetuses who have this finding.62

Outcome Echogenic foci do not increase in size and do not become clinically significant.60–63 In a large prospective study of more than 10 000 fetuses, echogenic foci were found in 1.6% of those referred for ultrasound. In this study, isolated echogenic foci were benign and did not increase the risk for fetal aneuploidy. Advanced maternal age (> 35 years),

Primary cardiac tumors are rare. When identified in the fetus they should be followed closely. Ultrasound is the most common modality currently utilized to assess the size and characteristics of fetal cardiac tumors. Newer modalities such as 3D ultrasound and magnetic resonance imaging are currently limited. Intervention should be considered for fetal cardiac tumors if hemodynamic compromise and/or fetal distress is present.

Legends for the DVD Video clip 28.1 2D fetal echocardiographic imaging of a cardiac tumor in the right ventricle.

Video clip 28.2 3D fetal echocardiographic imaging of the same tumor as in clip 1. In real time the tumor appeared to be attached to the tricuspid valve apparatus.

Video clip 28.3 2D fetal echocardiographic imaging of a right ventricular tumor.

Video clip 28.4 Color Doppler of image in clip 3 reveals severe tricuspid valve insufficiency.

Video clip 28.5 Postnatal 2D echocardiographic imaging of the tumor from clip 3.

Video clip 28.6 Color Doppler of the image from clip 5 reveals severe tricuspid valve insufficiency.

Video clip 28.7 Postnatal 3D ultrasound imaging of a cardiac rhabdomyoma in the right ventricle. The tumor is attached to the tricuspid valve apparatus.

Video clip 28.8 2D echocardiographic imaging in a fetus with a large right atrial tumor. Note the dilated right atrium and pericardial effusion.

Video clip 28.9 2D postnatal echocardiographic imaging of the tumor. Note that multiple tumors are seen in the right ventricle making the diagnosis of multiple rhabdomyomas and tuberous sclerosis.

Fetal cardiac tumors

References 1. Isaacs H. Fetal and neonatal cardiac tumors. Pediatr Cardiol 2004; 25: 252–73. 2. McAllister HA Jr. Primary tumors of the heart and pericardium. Pathol Ann 1979; 14: 325–55. 3. Silverman NA. Primary cardiac tumors. Ann Surg 1980; 191: 127–38. 4. Nadas AS, Ellison RC. Cardiac tumors in infancy. Am J Cardiol 1968; 21: 363–6. 5. Simcha A, Wells BG, Tynan MJ et al. Primary cardiac tumors in childhood. Arch Dis Child 1971; 46: 508–14. 6. Van der Hauaert LG. Cardiac tumors in infancy and childhood. Br Heart J 1971; 33: 125–32. 7. Heath D. Pathology of cardiac tumors. Am J Cardiol 1968; 21: 315–27. 8. DeVore GR, Hakin S, Kleinman CS et al. The in utero diagnosis of an interventricular septal cardiac rhabdomyoma by means of real-time-directed M-mode echocardiography. Am J Obstet Gynecol 1982; 143: 967–9. 9. Holley DG, Martin GR, Brenner JI et al. Diagnosis and management of fetal cardiac tumors: a multicenter experience and review of published reports. J Am Coll Cardiol 1995; 26: 516–20. 10. De G, etter B, Kretz JG, Nisand I et al. Intrapericardial teratoma in a newborn infant: use of fetal echocardiography. Ann Thorac Surg 1983; 6: 664–6. 11. Riggs T, Sholl JS, Ilbawi M et al. In utero diagnosis of pericardial tumor with successful surgical repair. Pediatr Cardiol 1984; 5: 23–6. 12. Dennis MA, Appareti K, Manco-Johnson ML et al. The echocardiographic diagnosis of multiple fetal cardiac tumors. J Ultrasound Med 1985; 4: 327–9. 13. Birnbaum SE, McGahan JP, Janos GG et al. Fetal tachycardia and intramyocardial tumors. J Am Coll Cardiol 1985; 6: 1358–61. 14. Hoadley SD, Wallace RL, Miller JF et al. Prenatal diagnosis of multiple cardiac tumors presenting as an arrhythmia. J Clin Ultrasound 1986; 14: 639–43. 15. Boxer RA, Seidman S, Singh S et al. Congenital intracardiac rhabdomyoma: prenatal detection by echocardiography, perinatal management, and surgical treatment. Am J Perinatol 1986; 4: 303–5. 16. Lethiser RE, Fyfe D, Weatherby E et al. Prenatal sonographic diagnosis of atrial hemangioma. Am J Roentgenol 1986; 147: 1207–8. 17. Gresser CD, Shime J, Rakowski H et al. Fetal cardiac tumor: a prenatal echocardiographic marker for tuberous sclerosis. Am J Obstet Gynecol 1987; 156: 689–90. 18. Rasmussen SL, Hwang WS, Harder J et al. Intrapericardial teratoma ultrasonic and pathologic features. J Ultrasound Med 1987; 6: 159–62. 19. Weber HS, Kleinman CS, Hellebrand WE et al. Development of a benign intrapericardial tumor between 20 and 40 weeks of gestation. Pediatr Cardiol 1988; 9: 153–6. 20. Cyr DR, Guntheroth WG, Nyberg DA et al. Prenatal diagnosis of an intrapericardial teratoma. J Ultrasound Med 1988; 7: 87–90. 21. Alegre M, Torrents M, Carreras E et al. Prenatal diagnosis of intrapericardial teratoma. Pediatr Diagn 1990; 10: 199–202.

411

22. Wallace G, Smith HC, Rimmer S et al. Tuberous sclerosis presenting with fetal and neonatal cardiac tumors. Arch Dis Child 1990; 65: 377–9. 23. Calhoun BC, Watson PT, Hegge F. Ultrasound diagnosis of an obstructive cardiac rhabdomyoma with severe hydops and hypoplastic lungs. J Reprod Med 1991; 36: 317–19. 24. Rheuban KS, McDaniel NL, Feldman PS et al. Intrapericardial teratoma causing non-immune hydrops fetalis and pericardial tamponade: a case report. Pediatr Cardiol 1991; 12: 54–6. 25. Watanabe T, Hojo Y, Kozak T et al. Hypoplastic left heart syndrome with rhabdomyoma of the left ventricle. Pediatr Cardiol 1991; 12: 121–2. 26. Todros T, Gaglioti P, Presbitero P. Management of a fetus with intrapericardial teratoma diagnosed in utero. J Ultrasound Med 1991; 10: 287–90. 27. Brand JM, Friedberg DZ. Spontaneous regression of a primary cardiac tumor presenting as fetal tachyarrhythmias. J Perinatol 1992; 12: 48–50. 28. Giacoia GP. Fetal rhabdomyoma: a prenatal echocardiographic marker of tuberous sclerosis. Am J Perinatol 1992; 9: 111–14. 29. Harding CO, Pagon RA. Incidence of tuberous sclerosis in patients with cardiac rhabdomyoma. Am J Med Genet 1990; 37: 443–6. 30. Goh RH, Lappalainnen RE, Mohide PT et al. Multiple cardiac masses in the fetus of a woman with tuberous sclerosis. Can Assoc Radiol J 1995; 46: 461–4. 31. Wu CT, Chen MR, Hou SH. Neonatal tuberous sclerosis with cardiac rhabdomyomas presenting as fetal supraventricular tachycardia. Jpn Heart J 1997; 38: 133–7. 32. Bader RS, Chitayat D, Kelly E et al. Fetal rhabdomyoma: prenatal diagnosis, clinical outcome, and incidence of associated tuberous sclerosis complex. J Pediatr 2003; 143: 620–4. 33. Lethor JP, De Moor M. Multiple cardiac tumors in the fetus. Circulation 2001; 103: e55. 34. Groves AM, Fagg NLK, Cook AC, Allan LD. Cardiac tumors in intrauterine life. Arch Dis Child 1992; 67: 1189–92. 35. Crawford DC, Garrett C, Tynan M et al. Cardiac rhabdomyomata as a marker for the antenatal detection of tuberous sclerosis. J Med Genet 1983; 20: 303–12. 36. Sklansky M et al. Images in cardiovascular medicine. Neonatal tuberous sclerosis and multiple cardiac arrhythmias. Circulation 2007; 115: e395–7. 37. Kivelitz DE, Muhler M, Rake A, Scheer I, Chaouri R. MRI of cardiac rhabdomyoma in the fetus. Eur Radiol 2003; 14: 1513–16. 38. Chen CP, Liu YP, Huang JK et al. Contribution of ultrafast magnetic resonance imaging in prenatal diagnosis of sonographically undetected cerebral tuberous sclerosis associated with cardiac rhabdomyomas. Prenat Diag 2005; 25: 523–4. 39. Sonigo P, Elmauh A, Fermond L et al. Prenatal MRI diagnosis of fetal cerebral tuberous sclerosis. Pediatr Radiol 1996; 26: 1–4. 40. Veldtman GR, Blackburn MEC, Wharton GA et al. Dystrophic calcification of the fetal myocardium. Heart 1999; 81: 92–3.

412

Fetal Cardiology

41. Lacey SR, Donofrio MT. Fetal cardiac tumors: prenatal diagnosis and outcome. Pediatr Cardiol 2007; 28: 61–7. 42. Thorpe JA, Gadt A, Gelatt M. Decompression of fetal cardiac tamponade caused by congenital capillary hemangioma of the pericardium. Obstet Gynecol 2000; 968: 16–17. 43. Schmaltz AA, Apitz J. Primary rhabdomyosarcoma of the heart. Pediatr Cardiol 1982; 2: 73–5. 44. Bosi G, Lintermans JP, Pelligrino PA et al. The natural history of cardiac rhabdomyoma with and without tuberous sclerosis. Acta Pediatr 1996; 85: 928–31. 45. Smythe JF, Dyck JD, Smallhorn JF et al. Natural history of cardiac rhabdomyoma in infancy and childhood. Am J Cardiol 1990; 66: 1247–9. 46. Fesslova V, Villa L, Rizzuti T, Mastrangelo M, Mosca F. Natural history and long-term outcome of cardiac rhabdomyomas detected prenatally. Prenat Diagn 2004; 24: 241–8. 47. Fenoglia JJ, Mcallister HA Jr, Ferran VJ. Cardiac rhabdomyoma: a clinicopathologic and electron microscopic study. Am J Cardiol 1976; 38: 241–51. 48. Roach ES, Smith M, Huttenlocher P et al. Report of the diagnostic criteria committee of the national tuberous sclerosis association. J Child Neurol 1992; 7: 221–4. 49. Watson GH. Cardiac rhabdomyomas in tuberous sclerosis. Ann NY Acad Sci 1991; 615: 50–7. 50. Russell GA, Dhasmann JP, Berry PJ et al. Coexistent cardiac tumors and malformations of the heart. Int J Cardiol 1989; 22: 89–98. 51. Wei J, Li P, Chiriboga L et al. Tuberous sclerosis in a 19 week fetus: immunohistochemical and molecular study of hamartin and tuberin. Pediatr Dev Pathol 2002; 5: 448–64. 52. Bini R, Westaby S, Bargeron LM et al. Investigation and management of primary cardiac tumors in infants and children. J Am Coll Cardiol 1983; 2: 351–7.

53. Corno A, de Simone G, Catena G et al. Cardiac rhabdomyoma: surgical treatment in the neonate. J Thorac Cardiovasc Surg 1984; 87: 725–31. 54. Bertolini P, Meisner H, Paek SU. Special considerations on primary cardiac tumors in infancy and childhood. Thorac Cardiovasc Surg 1990; 38(Suppl 2): 164–7. 55. Arciniegas E, Hakimi M, Farooki ZQ et al. Primary cardiac tumors in children. J Thorac Cardiovasc Surg 1980; 79: 582–91. 56. Marx G. Cardiac Tumors. In: Emmanouilides GC, Riemenschneider TA, Allen HD, Gutgesell HP, eds. Moss and Adams’ Heart Disease in Infants, Children, and Adolescents: Including the Fetus and Young Adult, 5th edn. Baltimore, MD: Williams & Wilkins, 1995: 1773–85. 57. Beghetti M, Gow RM, Haney I. Pediatric primary benign cardiac tumors: a 15 year review. Am Heart J 1997; 134: 1107–14. 58. Bruch SW, Adzick NS, Reiss R. Prenatal therapy for pericardial teratomas. J Pediatric Surg 1997; 32: 1113–15. 59. Levy DW, Mintz MC. The left ventricular echogenic focus: a normal finding. AJR Am J Roentgenol 1988; 150: 85. 60. Petrikovsky B, Klein V, Herrara M. Prenatal diagnosis of intra-atrial cardiac echogenic foci. Prenat Diagn 1998; 18: 968–70. 61. Brown DL, Roberts DJ, Miller WA. Left ventricular echogenic focus in the fetal heart: pathologic correlation. J Ultrasound Med 1994; 13: 613–16. 62. Roberts DJ, Genest D. Cardiac histologic pathology characteristics of trisomies 13 and 21. Hum Pathol 1992; 23: 1130–40. 63. Petrikovsky B, Challenger M, Gross B. Unusual appearances of echogenic foci within the fetal heart: are they benign? Ultrasound Obstet Gynecol 1996; 8: 229–31. 64. Bradley KE, Santulli TS, Gregory KD et al. An isolated intracardiac echogenic focus as a marker for aneuploidy. Am J Obstet Gynecol 2005; 192: 2021–6.

29 The fetal venous system: normal embryology, anatomy, and physiology and the development and appearance of anomalies Simcha Yagel, Zvi Kivilevitch, Dan V Valsky, and Reuwen Achiron Embryological development of the human venous system Three systems of paired veins are found in the early 4-week embryo: (1) the umbilical veins (UVs) from the chorion; (2) the vitelline veins from the yolk sac; and (3) the cardinal veins from the body of the embryo itself. All three sets are open to the right and left horn of the sinus venosus of the heart. From this stage the subsequent development of the fetal venous system is a harmony of changes and transformations that occur in these symmetrical venous systems (Figure 29.1a).1 The development of the fetal liver in the septum transversus plays a pivotal role in modifying the primitive vitelline and umbilical systems to their final shape. The developing hepatic sinusoids first become linked to both vitelline veins, and then tap the UV at day 32. The sinusoidal labyrinth then interrupts each vitelline vein as it enters the distal segment from the yolk sac to the liver, and proximal segment from the liver to the heart. The distal portions are converted into the portal vein (PV), the intermediate sinusoids mostly remaining as such, though in part expanding into the ductus venosus (DV), while the right proximal stem represents the hepatic vein (HV). Subsequently, the left proximal and right distal vitelline vein atrophy and disappear (Figure 29.1b). Alterations in the umbilical veins accompany the vitelline vein changes. In the 5-mm embryo, the left UV becomes the dominant conduit of blood from the placenta and empties into the left horn of the sinus venosus. As the primitive left and right lobes of the liver expand laterally, they soon come into contact with the paired UVs coursing close by. A similar anastomotic system is formed between liver sinusoids, vitelline veins, and UVs. Both UVs lose their connection with the fetal heart; their blood is tapped by liver sinusoids and from there returns to the heart. When all the umbilical blood enters the liver, as occurs in an embryo of 6 mm, the entire right umbilical

vein and proximal segment of the left umbilical vein atrophy and soon disappear. As a result, virtually all the placental blood enters the right atrium via the left distal UV, DV, and proximal right vitelline vein, which bypass the liver sinusoids. In the 9-mm embryo (6 weeks) the left and right portal veins are part of the left UV. The HV and the DV drain into the infracardiac portion of the inferior vena cava (IVC) (Figure 29.1c). The cardinal veins are the main venous drainage system of the embryo body. The anterior and posterior cardinal veins drain cranial and caudal parts of the body, respectively. Both veins empty into a common cardinal vein, which is the third venous system entering the sinus venosus of the primitive heart. During the eighth week the right anterior and right common cardinal veins become the superior vena cava (SVC), while the left anterior cardinal vein disappears through left to right anastomosis, the left brachiocephalic vein. The posterior cardinal vein, which drains the caudal part of the embryo, atrophies. Two vein systems, the sub- and supracardinal veins, gradually replace and supplement the posterior cardinal veins. The supracardinal veins break in the region of the kidneys. Above this level they unite by a cross-anastomosis and become the azygos and hemiazygos veins. Caudal to the kidneys the left supracardinal vein degenerates and the right becomes the inferior part of the IVC. The upper segments of the IVC are derived from the sub-supracardinal anastomosis at the renal region, the prerenal segment from the right subcardinal, and the hepatic segment from the proximal vitelline vein and liver sinusoids. The common pulmonary vein can be recognized in a 4-mm embryo as an invagination of the dorsal wall of the left atrium. With further development of the atrial cavity, the stem of the pulmonary vein is progressively incorporated into the wall of the left atrium. This incorporation of the pulmonary veins continues until two right and two left branches of the pulmonary stem enter the atrial cavity.

414

Fetal Cardiology

(a)

Right anterior cardinal vein

BC and TA

Right horn of SV Right common cardinal vein

PA Left horn of SV

central UV VV

VV UV

proximal

Right posterior cardinal vein

distal

Septum transversum (mesenchyme)

In summary, the left umbilical vein remains as the only vein from the placenta to the fetal heart. The development of the connection between it and the ductus venosus (‘the critical anastomosis’) is of primary importance in the development of the venous drainage system and supply of oxygenated blood to the fetus. The ductus venosus is unique to fetal life, arising from the vitelline vein within the liver (hepatic sinusoids). From this same vessel arise the portal vein, the superior mesenteric vein, and the intrathoracic inferior vena cava, which joins the ductus venosus to the heart (the hepatocardiac portion) (Figure 29.1c).1

Posterior view Anastomosis to form brachiocephalic vein

(b) BC and TA

PA

Expanded right horn of SV

Left horn of SV

Ductus venosus forms bypass

UVVV

VV UV

Posterior view

Vitelline veins fuse to form portal veins

(c) Anastomosis to form brachiocephalic vein

BC and TA

SVC PA Left horn of SV

IVC

Liver, diaphragm develop from mesenchyme

Ductus venosus

Posterior view LUV

Anatomy of the fetal venous system – from the end of the first trimester From the end of the first trimester, the umbilical vein carries blood from the placenta to the fetus. The intrafetal portion of the umbilical vein ascends by way of the falciform ligament to join the umbilical segment of the left portal vein. The latter maintains the same diameter as the umbilical vein, considerably greater than that of the right portal vein, which receives blood primarily recycled through the main portal vein. There is therefore a marked preference in both the quality and quantity of blood flowing through the left lobe of the liver, which is noticeably larger than the right. This situation is reversed in the adult, following atrophy of the umbilical vein and the ductus venosus. The left portal vein turns almost 90° to the right to join the anterior and posterior branches of the right portal vein. From this same ‘twist’, known as the pars transversa, arises the ductus venosus, which joins distally with the left hepatic vein and the inferior vena cava just proximal to the entrance into the right atrium.2,3 The ductus venosus narrows to 1–2 mm, approximately one-third of the width of the umbilical vein (Figure 29.2). There are contradictory reports of the existence of a ‘sphincter’ that regulates blood flow through the ductus venosus to the heart, and there may be oxygen concentration-dependent anatomic nervous control of the activity of this sphincter.4–6

Vessels contribute to portal system

Figure 29.1 (a) Posterior view of the paired vitelline veins (VV), umbilical veins (UV), sinus venosus (SV), truncus arteriosus (TA), and bulbus cordis (BC). PA, pulmonary artery. (b) Posterior view of the changes formed with the goal of introducing right-sided asymmetry in the adult. (c) Posterior view of the mature venous system, with atrophied vessels marked with crosses. LUV, left umbilical vein; SVC, superior vena cava; IVC, inferior vena cava (modified with permission from reference 1).

The venous circulatory system: physiology The role of the venous circulatory system in the fetus is to carry blood with the highest possible oxygen concentration to the fetal heart. The DV, HV, and IVC are the principal vessels carrying blood to the fetal heart; the DV and left UV have a critical role in this important function.

The fetal venous system

FO RA HV IVC DV

HV

RPV LPV PS EPV

Liver

LPV GB

UV

Figure 29.2 Schematic representation of the fetal umbilical, portal, and hepatic venous circulations. The arrows indicate the direction of blood flow and the color, the degree of oxygen content (red = high, purple = medium, blue = low). FO, foramen ovale; RA, right atrium; DV, ductus venosus; UV, umbilical vein; HV, hepatic veins; IVC, inferior vena cava; PS, portal sinus; LPV, left portal vein; RPV, right portal vein; EPV, extrahepatic portal vein; GB, gallbladder (reproduced with permission from reference 2).

The fetal cardioplacental system acts to facilitate smooth blood flow through the entire system to the heart. Aided by low placental resistance and improved cardiac contraction, a pressure gradient is created between the atria and ventricles which reduces the preload in the venous circulatory system and allows blood to flow toward the heart (preload index). The interdependence of the venous vascular system and the other elements of the fetal circulation is demonstrated physiologically during fetal breathing movements. Fetal breathing movements have a direct effect on venous return to the heart; this return is directly related to placental blood flow on the one hand, and blood flow to the heart on the other. The effect of fetal breathing movements on the fetal cardiovascular–placental system is an excellent example of the close physiological bond between systems. Changes in venous return to the fetal heart can influence emptying of the placental bed as well as systolic (‘upstream’) and diastolic (‘downstream’) blood flow in the arteries. Decreased venous return causes a drop in placental bed drainage and filling of the heart ventricles, resulting in a drop in the end-diastolic volume of the arteries on the one hand, and corresponding decline (through the Frank– Starling mechanism) in stroke volume of the heart on the

415

other side of the system. These in turn cause an additional negative effect on venous return and flow in the umbilical vein. Studies performed by Huisman et al and Chiba et al7,8 demonstrated changes in pressure gradient between the intra-abdominal and intrathoracic cavities resulting from fetal breathing movements that in turn caused changes in blood flow in the venous system. During inspiration the pressure gradient between the abdominal cavity (movement inward) and thoracic cavity (movement outward) rises from ∼0–3 to ∼22 mmHg, and as a result the pressure gradient between the umbilical vein and the thoracic part of the ductus venosus increases, and flow velocity in the umbilical vein rises. The opposite occurs during expiration. Indik and Reed proved that changes in intrathoracic pressure caused by fetal breathing movements affect both venous return to the heart and the arterial system.9 Decreased blood flow to the heart causes changes in placental blood flow on the fetal side of the interface. On the one hand, the placenta is prevented from emptying, causing a drop in arterial diastolic flow. On the other hand, ineffective filling of the cardiac ventricles causes a drop in arterial systolic blood flow. The most marked changes occur in the diastolic phase of the cardiac cycle. Contrarily, a rise in venous flow to the heart is accompanied at the next heartbeat by a rise in systolic and diastolic arterial blood flow. These findings prove that the placenta is not only a collection of resistant blood vessels, but also a system capable of transmitting pressure changes occurring within the heart and thorax. In addition, the existence of the Frank–Starling mechanism in the fetal heart is proved again (Figure 29.3).

Anomalies of the fetal central veins and umbilical–portal system Fetal venous evaluation has been described since the mid1980s.10,11 In recent years, high-resolution ultrasonography, combined with color Doppler, has advanced our ability to investigate the fetal venous system.12 This non-invasive technique has enhanced our understanding of fetal venous circulation in normal physiological conditions, and provided us with the ability to study circulatory changes in abnormal cases. Although our knowledge of the normal anatomy of the fetal venous system has increased enormously,13 little information is available on the mechanisms leading to in utero abnormal development. The most frequently reported abnormalities are those of the intrahepatic UV, but cases of agenesis of the DV have been reported.3,14–16 The development and widespread use of color Doppler machines aroused an increased trend toward in utero diagnosis of other anomalies of the fetal venous inlet system.17–19

416

Fetal Cardiology

Inspiration

Expiration

Increased intra-abdominal pressure

Decreased intra-abdominal pressure

Decreased intrathoracic pressure

Increased intrathoracic pressure

Increased venous flow velocity

Decreased venous flow velocity

Decreased resistance / preload

Increased resistance / preload

Increased venous return + Increased ventricular filling

Increased stroke

Increased placental emptying / decreased resistance

Increased arterial systolic flow

Increased diastolic flow

At present, the only available information regarding the cause, significance, and prognosis of these abnormalities is derived from newborns;20–22 however, fetal outcome cannot be directly extrapolated from data obtained from neonates with similar abnormalities. As the number of case reports regarding various fetal venous abnormalities increases,23–25 antenatal management of these patients may become problematic, and understanding of the embryologic basis and pathophysiology of the fetal venous inlet system abnormalities becomes increasingly important for accurate prenatal counseling. Based on the type of vein involved, embryological, precursor, and etiologic correlation (primary or secondary), we suggest classification into four major groups (Table 29.1):26 1. 2. 3. 4.

Figure 29.3

volume

Abnormal connection of the cardinal veins; Abnormalities of the umbilical vein; Anomalies of the vitelline veins; Anomalies of the pulmonary veins.

Several large series have been published recently that describe fetal venous system malformation. Hofstaetter et al16 evaluated 16 fetuses with abnormalities of the umbilical, portal, hepatic, and caval venous systems. Achiron et al26 described 19 fetuses with abnormal connection between central veins and the fetal heart. Their reports show that in a targeted fetal organ scan the course of the umbilical vein, ductus venosus, the portal and hepatic veins, and inferior vena cava, as well as the pulmonary veins, can be clearly

Venous–placental– arterial interaction during fetal breathing movements.

visualized using two-dimensional (2D) gray-scale imaging and color Doppler. Moreover, various anomalies can be accurately diagnosed. Viora and colleagues27 described 26 fetuses: in five other malformations were present, one was affected with trisomy 21, and one with intrauterine growth restriction (IUGR) and fetal intra-abdominal umbilical vein (FIUV) varix. Twenty-five were live-born, the baby with FIUV varix having died in utero. The authors underscore both the necessity of close monitoring of FIUV varix (Figure 29.4 and Video clips 29.1 and 29.2),27,28 and that isolated venous anomalies can be managed expectantly. A finding of venous anomaly is an indication for thorough fetal echocardiographic examination and, in many cases, fetal karyotyping. We speculate that the normal evolution of the fetal venous system development as described here can be disturbed mainly by two different events: (1) primary failure to transform or to form the critical anastomosis, and (2) secondary occlusion of an already transformed system. Ultrasound (US) has been shown to be the first modality in the diagnosis and evaluation of congenital venous anomalies, from 2DUS pulse, color, and power Doppler, to 3D/4D applications such as B-flow, spatiotemporal image correlation (STIC), 3D power Doppler, and 3D high definition power flow Doppler, 3D rendering, and inversion mode.29–34 For a full description of these various modalities and their applications please see Chapters 11, 12, 14, and 15.

The fetal venous system

Table 29.1 heart A.

B.

417

Proposed classification system for abnormal connection of the fetal venous system with the

Cardinal veins a. Complex malformations, heterotaxy syndromes b. Isolated malformation Umbilical veins a. Primary failure to create critical anastomosis 1. Complete – abnormal connection of UV (venous shunt) into: iliac vein, IVC, SVC, and right atrium 2. Partial – persistent right umbilical vein with or without DV

C.

b. Secondary occlusion Vitelline veins a. Primary failure to create critical anastomosis 1. Complete agenesis of portal system 2. Partial agenesis of right or left portal branch (portosystemic shunt)

D.

b. Secondary occlusion of portal branch Anomalous pulmonary venous connection a. Total b. Partial

Figure 29.4 Fetal intra-abdominal umbilical vein (FIUV) varix. The two-dimensional (2D) image with color Doppler shows the varix with its turbulent flow (a). The B-flow image isolates the lesion and shows the continuity of the affected vessel (b), while the rendered image provides the anatomic context. V, varix; UV, umbilical vein; aUV, abdominal umbilical vein. See also corresponding Video clips 29.1 and 29.2.

Cardinal vein Heterotaxy syndromes are the best example of abnormal primary development of the venous system. A mouse model exists to explain the transmission of situs abnormalities in the polysplenia–asplenia syndromes.35 Although most heterotaxy syndromes are associated with complex anomalies, particularly cardiac malformations that influence prognosis, there are some rare cases without significant cardiac defects that may survive.36 In such cases, prenatal evaluation is feasible, and prognosis may be elucidated so that the

parents may be counseled appropriately. Our study and others demonstrate that isolated anomalies of the cardinal veins are also detectable. Absence of IVC with azygos continuation (Figure 29.5 and Video clips 29.3 and 29.4) results from primary failure of the right subcardinal vein to connect with the hepatic segment of the IVC. Instead, it shunts its blood directly into the right supracardinal vein. Hence, the bloodstream from the caudal part of the body reaches the heart by way of the azygos and SVC. The persistent left SVC is also an example of primary failure to create anastomosis and to form the left brachiocephalic vein (Figure 29.6).

418

Fetal Cardiology

(a)

(b)

(c)

Figure 29.5 Interrupted inferior vena cava with azygos continuation. (a) 2D Doppler image of the azygos (Az, red) draining into the superior vena cava (SVC, blue); (b) the same plane slightly later in the cardiac cycle shows the aorta (Ao, blue) and carotid artery (CA, red). (c) B-flow image showing all these vessels together, which was not possible to demonstrate in 2D imaging (AoA, aortic arch; DV; ductus venosus). See also corresponding Video clips 29.3 and 29.4.

Figure 29.6 Persistent left superior vena cava anomaly in three-vessel–trachea view. RSVC, right superior vena cava; LSVC, left superior vena cava.

Umbilical veins Fetuses with portal and UV abnormalities form the main group of venous anomalies detected in utero. There are several examples of primary failure to form ‘critical anastomosis’, in which an aberrant vessel is shunted between the placenta and the iliac vein. Similar cases have been published in the English literature,16,22,23,37 in two of which Noonan syndrome was proved to be associated with a similar abnormal connection of the UV. One of

our cases was also associated with Noonan syndrome. In all the aforementioned cases hydrops fetalis was a common factor, the severity and generalized edema of which was out of proportion to the vascular anomaly, raising the possibility that the finding was largely attributable to some other cause, perhaps a congenital generalized lymphangiectasia as manifested in Noonan syndrome. The most convincing embryologic explanation for this rare anomaly was offered by White et al, who introduced the term ‘critical anastomosis’.38 The authors speculated that through unknown causes of embryologic maldevelopment, anastomosis between umbilical and vitelline veins failed to form, and resorption in early embryonic life of both UVs occurred. The resulting block to the flow of venous blood from the placenta led to the opening of vascular channels through anastomosis to the supracardinal components of the IVC. Smith39 showed that in a pig embryo of 7 mm in length, there is a connection between the primitive pelvic venous plexus that drains the lower limb bud and the embryonic UV. It is therefore not surprising that in a case of primary failure to form these critical anastomoses the placental venous return is rerouted to the iliac vein. The return of the placental venous blood in cases with primary failure to form critical anastomosis is not exclusive to the IVC, but may include, among the reported cases, other sites such as the SVC or direct communication with the right atrium. A common feature of all the aforementioned cases is agenesis of the DV, which is a fundamental embryologic maldevelopment. Agenesis of the DV was first described by Platauf in 1888,40 and over the next 100 years only sporadic neonates have been reported with this anomaly.

The fetal venous system

With the widespread use of prenatal ultrasound, the diagnosis of DV agenesis in utero has become feasible.41 In reviewing the literature concerning this anomaly, two subgroups may be distinguished – that in which the umbilical flow entirely bypasses the liver, connecting to the systemic venous circulation, and that in which the UV drains into the PV, causing all the umbilical blood to pass through the hepatic sinusoids. In the former group, all cases reported, including ours, had an abnormal connection of the UV to the iliac vein, or to the IVC in its infracardiac segment, and some were associated with pleural effusion, hydrops, and Noonan syndrome.22,23,37 In the latter pattern, an adequate connection between the UV and the vitelline venous system was established. The UV drains properly into the PV, but fails to establish a communication with the persistent proximal part of the right vitelline vein. In such cases, the prognosis seems to be better.23,40,42–44 Complete primary failure to form critical anastomoses in both left and right UV is an extremely rare event, supported by the small number of cases reported in the literature. However, partial failure to form critical anastomosis may be more common. If partial anastomosis between the hepatic sinusoids and UV or vitelline vein fails to develop, blood is then diverted through other channels. Two main paths have been described in such cases: the most common reported is the persistent right UV anomaly 26,45–48 (Figure 29.7 and Video clip 29.5). Persistent right UV

(a)

419

anomaly is characterized by the right UV that fails to form anastomosis with the right vitelline vein, while the left UV disappears.3 The second path is the direct entry of the UV into the right atrium, which has been reported in only a few cases.17,26,49,50 Jeanty was the first to suggest that streaming of the early flow through the right UV, primary or secondary to occlusion by thromboembolic events arising from the placenta, may cause this anomaly.3 The reason for the primary failure to form anastomosis, and consequently for the right UV to remain patent, is not clear, although it was found that teratogenic agents such as retinoic acid or deficient folate induced this anomaly in rats.51 Regarding secondary formation of such an anomaly, echogenic foci situated within the fetal liver suggest thromboembolism occluding the DV or other veins. In this case, the effect of portal–ductal occlusion is divergence of the oxygenated blood coming from the placenta into the already closed proximal left UV. This aberrant flow probably caused reopening of this part of the UV, resulting in a persistent proximal left UV.52

Vitelline veins Anomalies of the vitelline veins are extremely rare, and only a few have been reported during fetal life.16,26

(b)

Figure 29.7 Persistent right umbilical vein (PRUV) anomaly. (a) 3D high definition color flow Doppler image shows the characteristic appearance of the persistent right umbilical vein. The umbilical vein (UV) makes a ‘twist’ to the left toward the fetal stomach (S) and the portal veins (PV), as opposed to the twist toward the right in the normal fetus. Arrow indicates a small section of the fetal gallbladder, found to the left of the PRUV. (b) 3D ultrasound (US) inversion mode of the PRUV anomaly reveals the idiosyncratic course of the vessel, which is challenging to reconstruct mentally from a series of 2D images (Rt, right; Lt, left) (reproduced with permission from reference 32). See also corresponding Video clip 29.5.

420

Fetal Cardiology

Complete absence of the portal system is an extreme example of total failure of the vitelline veins to transform into the portal system, i.e. there is primary failure to form critical anastomosis with the hepatic sinusoids or UVs. As a result, the enterohepatic circulation is disturbed and the portal venous blood is shunted systemically. Mesenteric and splenic venous blood may drain into the renal veins, HVs, or directly into the IVC.53–55 Although there are reported cases in the pediatric and adult literature with congenital absence of the portal system, we were able to find only one case in whom prenatal diagnosis was achieved. In this case, the in utero ultrasonographic appearance of total agenesis of the portal system included an intrahepatic aberrant vessel, absence of the portal system, and marked dilatation of the IVC.56 Incomplete absence of the portal system or partial failure to form critical anastomosis may represent a more benign form of vitelline vein abnormality. Partial primary failure to form critical anastomosis may result in agenesis of the right portal system, with a persistent left vitelline vein connected directly to the RA (portohepatic shunt) and an absence of DV (Figure 29.8 and Video clip 29.6). Neonatal spontaneous resolution occurred later in some of the cases.26,57 In addition, we were able to find only one case with agenesis of right and left PVs, associated with a portosystemic shunt, reported prenatally.18

Figure 29.8 Agenesis of the ductus venosus. 4DUS B-flow image of the absent ductus venosus anomaly, with umbilical vein drainage to the right atrium. AO, aorta; RA, right atrium; IVC, inferior vena cava; HV, hepatic veins; UV, umbilical vein; UC, umbilical cord. See also corresponding Video clip 29.6.

Anomalous pulmonary venous connection Partial anomalous pulmonary venous connection This involves one or more, but not all, of the pulmonary veins connecting to the right atrium or a tributary (Figure 29.9).58 An atrial septal defect (ASD) is usually present. There are several variant presentations of this anomaly:33,58–61 1. The right pulmonary veins enter the superior vena cava (SVC) (Figure 29.9a). In this variant, most commonly the right upper and middle lung lobes drain into the SVC, while in some cases the upper lobe may drain into the SVC below the azygos vein, the middle lobe into the SVC at its junction with the right atrium, or the right lower lobe may drain into the left or right atrium. The lower SVC between the azygos vein and the right atrium is generally dilated, to as much as twice its normal size. Usually an ASD is present, most commonly of the sinus venosus type. Other types of ASD are found less often, and occasionally the atrial septum is intact. The right pulmonary veins may enter the right atrium. In this variant of partial anomalous pulmonary venous connection (PAPVC), veins from all lobes of the right lung drain directly into the right atrium, and an ASD is usually present, most commonly of the sinus venosus type. Occasionally a secundum ASD is found, or rarely, an ostium primum ASD. 2. The right pulmonary veins enter the inferior vena cava (IVC) (Figure 29.9b). In this variation, all or only the middle and lower lobes of the right lung are drained into the IVC, just above or below the diaphragm. The veins of the right lung form a ‘fir tree’ pattern and the atrial septum is intact. This variant of PAPVC is frequently associated with hypoplasia of the right lung, anomalies of the bronchial system, dextroposition of the heart, hypoplasia of the right pulmonary artery, anomalous arterial connection to the right lung from the aorta, and other cardiac anomalies. 3. The left pulmonary veins enter the left innominate vein (Figure 29.9c). In this anomaly, veins from only the upper lobe, or from all of the left lung, enter the left innominate vein by way of an anomalous vertical vein. An ASD is usually present, most commonly of the secundum type. Variations within this subgroup of PAPVC include left pulmonary drainage into the coronary sinus, the IVC, right SVC, right atrium, or left subclavian vein. It is often associated with cardiac defects and syndromes including polysplenia–asplenia syndrome, and Turner and Noonan syndromes. 4. Less commonly, the left pulmonary veins connect anomalously to the coronary sinus (Figure 29.9d) or to the IVC, right SVC, right atrium, or left subclavian vein. Very

The fetal venous system

(a)

421

(b) S.V.C. L.Inn.V. S.V.C.

R.P.V.

L.P.V.

L.P.V.

R.A.

R.A.

L.A.

L.A. C.S.

C.S.

I.V.C. L.V.

L.V.

R.P.V.

R.V.

(c)

(d) S.V.C.

R.V.

I.V.C.

S.V.C.

n.V.

L.In

R.P.V. L.P.V.

v.v. L.P.V.

R.P.V.

R.A.

C.S. L.A.

C.S.

L.A.

R.A.

I.V.C. L.V. R.V. I.V.C. L.V. R.V.

Figure 29.9 Common forms of partial anomalous pulmonary venous connection. (a) Anomalous connection of the right pulmonary veins (R.P.V.) to superior vena cava (S.V.C.). A high or sinus venous defect is usual in this anomaly. (b) Anomalous connections of the right pulmonary veins to the inferior vena cava (I.V.C.). The right lung commonly drains by one pulmonary vein without its usual anatomic divisions. Parenchymal abnormalities of the right lung are common and the atrial septum is usually intact. (c) Anomalous connection of the left pulmonary veins (L.P.V.) to the left innominate vein (L.Inn.V.) by way of a vertical vein (V.V.). An additional left-to-right shunt may occur through the atrial septal defect. (d) Anomalous connection of the left pulmonary veins to the coronary sinus (C.S.). R.A., right atrium; L.A., left atrium; R.V., right ventricle; L.V., left ventricle (reproduced with permission from reference 58).

rarely, the right pulmonary veins may connect to the azygos vein or coronary sinus. Echocardiographic diagnosis of PAPVC is made by visualizing pulmonary veins connecting to the right-sided cardiac structure involved: the SVC, left innominate vein, or coronary sinus. A finding

of right ventricular volume overload with an intact atrial septum may mean that a PAPVC anomaly is present. Doppler flow studies and color mapping are indispensable in identification and classification of this group of anomalies.33,58–60

422

Fetal Cardiology

Total anomalous pulmonary venous connection In total anomalous pulmonary venous connection (TAPVC) there is no connection between the pulmonary veins and the left atrium (Figure 29.10).58 Rather, the pulmonary veins drain into the right atrium or the systemic veins. One-third of patients present with an associated anomaly, while twothirds present with isolated TAPVC. TAPVC is classified into four types:58 1. 2. 3. 4.

Anomalous connection at the supracardiac level; Anomalous connection at the cardiac level; Anomalous connection at the infracardiac level; Connections occur at two or more of the above levels.

An alternative classification system divides this anomaly into two groups: supradiaphragmatic without pulmonary venous obstruction, and infradiaphragmatic with pulmonary venous obstruction. ASD or patent foramen ovale is generally present. The presence of an obstructive lesion in the pulmonary venous channel influences the hemodynamic state and clinical presentation of TAPVC. The obstruction may occur at the interatrial septum or intrinsic or extrinsic to the anomalous venous channel. One-third of patients will present with an associated major anomaly: cor bilocular, single ventricle, truncus arteriosus, transposition of the great arteries, pulmonary atresia, coarctation, hypoplastic left ventricle, and anomalies of the systemic veins. Echocardiographic evaluation of suspected TAPVC (Figure 29.11 and Video clip 29.7) must locate the site of connection of the common pulmonary vein, evaluate any obstruction of the pulmonary veins or vertical vein, and confirm or exclude associated cardiac lesions. All forms of TAPVC will present with right ventricle volume overload, an enlarged right atrium, and a bowing leftward of the interatrial septum. TAPVC should be suspected when no pulmonary veins can be visualized entering the left atrium on the short-axis scan. In addition, the common pulmonary channel may be identified posterior to the left atrium with the pulmonary veins draining into it. The channel may be followed to the site of its connection. In supracardiac forms of TAPVC, the channel will usually drain into an ascending vertical vein, which in turn may be followed to a dilated systemic venous structure, most commonly the left innominate vein or SVC. Doppler studies reveal flow in this vertical vessel to be cephalad, as opposed to the caudal flow of blood in the SVC. In infradiaphragmatic forms of TAPVC, the connection is usually to the portal venous system but may be to the hepatic veins. The pulmonary veins converge into a common channel, small and inferior to the left atrium above the diaphragm, or it may appear as a discrete chamber. If untreated, affected neonates will survive only a few days to approximately 4 months.59–61

Doppler evaluation is invaluable in the diagnosis of all forms of TAPVC. The abnormal color flow signal may be followed to identify the anomalous connection, as well as the direction and mean velocity of flow. Turbulence from any obstruction will produce a mosaic color jet that reveals its location, while pulsed or continuous wave Doppler allows for quantification of the degree of obstruction.33,58–61

Prognosis The prognosis in cases of PAPVC and TAPVC depends on the presence and size of the interatrial connection, the presence of any obstructive lesion in the anomalous venous pathways, and the overall state of the pulmonary vascular bed. Prognosis is also necessarily affected by the presence of any cardiac or extracardiac malformations. In cases of TAPVC with obstruction in the anomalous venous channel, the neonate usually dies in the first weeks of life.59–61

Comment This classification system for fetal venous anomalies is unique in that it attempts to combine embryology and etiology, i.e. whether the anomaly is a primary or secondary event. The concept of a thromboembolic mechanism as a cause of fetal hepatic echogenic lesions was proposed by Avni et al62 and confirmed by a large pathologic study of 1500 spontaneously aborted fetuses. Most of the 33 cases with hepatic calcification were vascular in type.63 Moreover, our prenatal investigation and other recent series of fetal liver hyperechoic lesions failed to find any correlation with well known factors such as infections or tumors, leaving the vascular etiology as the most probable explanation for their origin.64,65 In one of our reported cases, local obliteration of the ductal system was most likely secondary to systemic disease. In that case, a rare disease of fetal mastocytosis caused hepatic fibrosis and in utero obliteration of the ductal system. Fetal venous malformations are a complex group of rare anomalies. It may be difficult to delineate fetal and neonatal prognosis when such an anomaly is detected. However, a few conclusions may be drawn from our and others’ experience and the data accumulated in the literature. Prenatal evaluation of fetuses found to have anomalies of the venous system should include a careful search for cardiac anomalies and detailed anatomical survey of the umbilical, portal hepatic, and ductal systems, to determine aberrant communication, and if possible to delineate hints for systemic diseases, or a thromboembolic phenomenon. In cases with agenesis of the DV, careful follow-up is mandatory, since its absence may induce the appearance of hydrops fetalis, with unfavorable prognosis.16,23

The fetal venous system

(a)

(b)

S.V.C. L.Inn.V.

S.V.C.

v.v.

R.P.V.

C.P.V. L.P.V.

C.P.V.

423

R.A. L.A.

R.A. C.S.

C.S.

L.A.

I.V.C.

I.V.C.

L.V. R.V.

L.V. R.V.

(c)

(d)

S.V.C.

S.V.C.

C.P.V. L.P.V.

R.A.

R.A.

R.P.V.

L.A.

C.S.

C.S. L.A.

I.V.C. I.V.C. L.V.

L.V.

R.V.

R.V. I.V.C. R.H.

L.H.

D.V.

R.P.

L.P. P.V.

S.V.

S.M.V.

Figure 29.10 Common forms of total anomalous pulmonary venous connection. (a) To the left innominate vein (L.Inn.V.) by way of a vertical vein (V.V.). (b) To the coronary sinus (C.S.). The pulmonary veins join to form a confluence designated as the common pulmonary vein (C.P.V.), which connects to the coronary sinus. (c) To the right atrium. The left and right pulmonary veins (L.P.V. and R.P.V.) usually enter the right atrium separately, (d) To the portal vein (P.V.). The pulmonary veins form a confluence, from which an anomalous channel arises. This connects to the portal vein, which communicates with the inferior vena cava (I.V.C.) by way of the ductus venosus (D.V.) or the hepatic sinusoids. S.V., splenic vein; S.M.V, superior mesenteric vein; R.P. and L.P., right and left portal veins; R.H. and L.H., right and left hepatic veins; S.V.C., superior vena cava; R.A. and L.A., right and left atria; R.V. and L.V., right and left ventricles (reproduced with permission from reference 58).

424

Fetal Cardiology

(a)

(b)

Figure 29.11 Total anomalous pulmonary venous connection (TAPVC). This image of TAPVC was acquired with high definition power Doppler and spatiotemporal image correlation (STIC) acquisition. Note the vertical vein (VV) and absence of the pulmonary veins. Compare with normal heart and vessels (b) and with Figure 29.10d. See also corresponding Video clip 29.7. dAo, descending aorta; SMA, superior mesenteric artery.

Legends for the DVD Video clip 29.1 Fetal intra-abdominal umbilical vein (FIUV) varix. Clip 29.1a 2D image with color Doppler shows the varix in the fetal abdomen. Clip 29.1b shows the turbulent blood flow in the varix.

Video clip 29.2 The B-flow image isolates the lesion and shows the continuity of the affected vessel, while rendered image provides the anatomic context. V: varix; UV: umbilical vein; aUV: abdominal umbilical vein.

Video clip 29.3 Interrupted inferior vena cava with azygos continuation. 2D Doppler image of the azygos (Az, red) draining into the superior vena cava (SVC, blue); the same plane slightly later in the cardiac cycle, shows the aorta (Ao, blue) and carotid artery (CA, red).

Video clip 29.4 B-flow image showing all the vessels together, which was not possible to demonstrate in 2D imaging (AoA, aortic arch, CA, carotid artery, Az, azygos vein, SVC, superior vena cava, DV, ductus venosus).

Video clip 29.5 3DUS inversion mode of the PRUV anomaly reveals the idiosyncratic course of the vessel, which is challenging to mentally reconstruct from a series of 2D images.

Video clip 29.6 Agenesis of the ductus venosus. 4DUS b-flow image of the absent ductus venosus anomaly, with umbilical vein drainage to the right atrium. AO: aorta, RA: right atrium, IVC: inferior vena cava, HV: hepatic veins, UV: umbilical vein, UC: umbilical cord.

Video clip 29.7 Image of TAPVC anomaly acquired with high definition power Doppler and STIC acquisition. Note the vertical vein (VV) and absence of the pulmonary veins (dAo, descending aorta, IVC, inferior vena cava).

References 1. McGill Molson Medical Informatics Project. http://sprojects. mmi.mcgill.ca/embryology/cvs/default.html. 2. Mavrides E, Moscoso G, Carvalho JS, Campbell S, Thilaganathan B. The anatomy of the umbilical, portal and hepatic venous systems in the human fetus at 14–19 weeks of gestation. Ultrasound Obstet Gynecol 2001; 18: 598–604. 3. Jeanty P. Persistent right umbilical vein: an ominous prenatal finding? Radiology 1990; 177: 735–8. 4. Huisman TWA, Gittenberger-de Groot AC, Wladimiroff JW. Recognition of a fetal subdiaphragmatic venous vestibulum essential for fetal venous Doppler assessment. Pediatr Res 1992; 32: 338–41. 5. Pearson AA, Sauter RW. Observation on the phrenic nerves and ductus venosus in human embryos and fetuses. Am J Obstet Gynecol 1971; 110: 560–5.

The fetal venous system

6. Jensen A, Roman C, Rudolph AM. Effects of reducing uterine blood flow on fetal blood flow distribution and oxygen delivery. J Dev Physiol 1991; 15: 309–23. 7. Huisman TWA, van den Eijnde SM, Stewart PA, Wladimiroff JW. Changes in inferior vena cava blood flow velocity and diameter during breathing movements in the human fetus. Ultrasound Obstet Gynecol 1993; 3: 26–30. 8. Chiba Y, Utsu M, Kanzaki T, Hasegawa T. Changes in venous flow and intratracheal flow in fetal breathing movements. Ultrasound Med Biol 1985; 11: 43–9. 9. Indik J, Reed K. Variation and correlation in human fetus umbilical Doppler velocities with fetal breathing: evidence of the cardiac-placental connection. Am J Obstet Gynecol 1990; 163: 1792–6. 10. Chinn DK, Filly RA, Callen PW. Ultrasonic evaluation of fetal umbilical and hepatic vascular anatomy. Radiology 1982; 144: 153–7. 11. Champetier J, Yver R, Letoublon C, Vigneau B. A general review of anomalies of hepatic morphology and their clinical implications. Anat Clin 1985; 7: 285–99 12. Hecher K, Campbell S. Characteristics of fetal venous blood flow under normal circumstances and during fetal disease. Ultrasound Obstet Gynecol 1996; 7: 68–83. 13. Huisman TWA, Steward PA, Wladimiroff JW. Doppler assessment of the normal early circulation. Ultrasound Obstet Gynecol 1992; 2: 300–5. 14. Moore L, Toi A, Chitayat D. Abnormalities of the intraabdominal fetal umbilical vein: reports of four cases and a review of the literature. Ultrasound Obstet Gynecol 1996; 7: 21–5. 15. Jorgensen C, Andolf E. Four cases of absent ductus venosus: three in combination with severe hydrops fetalis. Fetal Diagn Ther 1994; 9: 395–7. 16. Hofstaetter C, Plath R, Hansmann M. Prenatal diagnosis of abnormalities of the fetal venous system. Ultrasound Obstet Gynecol 2000; 15: 231–41. 17. Greiss HB, McGahan JP. Umbilical vein entering the right atrium: significance of in utero diagnosis. J Ultrasound Med 1992; 11: 111–13. 18. Goncalves LF, Sherer DM, Romero R et al. Prenatal sonographic findings of agenesis of the right and left portal veins and associated intrahepatic portosystemic shunts. J Ultrasound Med 1995; 14: 849–52. 19. Sheley RC, Nyberg DA, Kapur R. Azygos continuation of the interrupted inferior vena cava: a clue to prenatal diagnosis of the cardiosplenic syndromes. J Ultrasound Med 1995; 14: 381–7. 20. Ricklan DE, Collett TA, Lyness SK. Umbilical vein variations: review of the literature and a case report of persistent right umbilical vein. Teratology 1988; 37: 95–100. 21. Fliegel CP, Nars PW. Aberrant umbilical vein. Pediatr Radiol 1984; 14: 55–6. 22. Currarino G, Stannard MW, Kolni H. Umbilical vein draining into the inferior vena cava via the internal iliac vein bypassing the liver. Pediatr Radiol 1991; 21: 265–6. 23. Siven M, Ley D, Hagerstrand I, Svennigsen N. Agenesis of the ductus venosus and its correlation to hydrops fetalis and the fetal hepatic circulation: case reports and review of the literature. Pediatr Pathol Lab Med 1995; 15: 39–50.

425

24. Avni EF, Chysels M, Donner C, Damis E. In-utero diagnosis of congenital absence of the ductus venosus. J Clin Ultrasound 1997; 25: 456–8. 25. Gembruch U, Baschat AA, Gailebe A, Gortner L. Prenatal diagnosis of ductus venosus agenesis: a report of two cases and review of the literature. Ultrasound Obstet Gynecol 1998; 11: 185–9. 26. Achiron R, Hegesh J, Yagel S et al. Abnormalities of the fetal central veins and umbilico-portal system: prenatal ultrasonographic diagnosis and proposed classification. Ultrasound Obstet Gynecol 2000; 16: 539–48. 27. Viora E, Sciarrone A, Bastonero S et al. Anomalies of the fetal venous system: a report of 26 cases and review of the literature. Fetal Diagn Ther 2004; 19: 440–7. 28. Valsky DV, Rosenak D, Hochner-Celnikier D, Porat S, Yagel S. Adverse outcome of isolated fetal intra-abdominal umbilical vein varix despite close monitoring. Prenat Diagn 2004; 24: 451–4. 29. Kalache K, Romero R, Goncalves LF et al. Three-dimensional color power imaging of the fetal hepatic circulation. Am J Obstet Gynecol 2003; 189: 1401–6. 30. Chaoui R, Kalache KD, Hartung J. Application of threedimensional power Doppler ultrasound in prenatal diagnosis. Ultrasound Obstet Gynecol 2001; 17: 22–9. 31. Chaoui R, Hoffmann J, Heling KS. Three-dimensional (3D) and 4D color Doppler fetal echocardiography using spatiotemporal image correlation (STIC). Ultrasound Obstet Gynecol 2004; 23: 535–45. 32. Sciaky-Tamir Y, Cohen SM, Hochner-Celnikier D et al. Three-dimensional power Doppler (3DPD) ultrasound in the diagnosis and follow-up of fetal vascular anomalies. Am J Obstet Gynecol 2006; 194: 274–81. 33. Espinoza J, Goncalves LF, Lee W, Mozor M, Romero R. A novel method to improve prenatal diagnosis of abnormal systemic venous connections using three- and four-dimensional ultrasonography and ‘inversion mode’. Ultrasound Obstet Gynecol 2005; 25: 428–34. 34. Goncalves LF, Espinoza J, Lee W, Mazor M, Romero R. Three- and four-dimensional reconstruction of the aortic and ductal arches using inversion mode: a new rendering algorithm for visualization of fluid-filled anatomical structures. Ultrasound Obstet Gynecol 2004; 24: 696–8. 35. Moore KL, Persaud TVN. The Developing Human, Clinically Oriented Embryology, 6th edn. Philadelphia: WB Saunders, 1998. 36. Langman J. Medical Embryology. Baltimore: Williams & Wilkins, 1963: 181. 37. Leonidas JC, Fellows RA. Congenital absence of the ductus venosus with direct connection between umbilical vein and the distal inferior vena cava. AJR Am J Roentgenol 1976; 126: 892–5. 38. White JJ, Brenner H, Avery ME. Umbilical vein collateral circulation: the caput medusae in a newborn infant. Pediatrics 1969; 43: 391–5. 39. Smith HW. On the development of superficial veins of the body wall in the pig. Am J Anat 1909; 9: 439. 40. Platauf R. Ein Fall von Mangel des Ductus venosus Arantii. Wien Klin Wochenschr 1888; 1: 165–7. 41. Fasouliatis S, Achiron R, Kivilevitch Z, Yagel S. The human fetal venous system normal embryology, anatomy, physiology,

426

42. 43.

44.

45.

46.

47.

48.

49.

50.

51.

52.

53. 54.

Fetal Cardiology

and developmental abnormalities. J Ultrasound Med 2002; 21: 1145–58. MacMahon HE. The congenital absence of ductus venosus: report of a case. Lab Invest 1960; 9: 127–31. Berg C, Kamil D, Geipel A et al. Absence of ductus venosus – importance of umbilical venous drainage site. Ultrasound Obstet Gynecol 2006; 28: 275–81. Sothinathan U, Pollina E, Huggon I, Patel S Greenough A. Absence of the ductus venosus. Acta Paediatr 2006; 95: 620–1. Bell AD, Gerlis LM, Variend S. Persistent right umbilical vein - case report and review of the literature. Int J Cardiol 1986; 10: 167–76. Hill LM, Mills A, Peterson C, Boyles D. Persistent right umbilical vein: sonographic detection and subsequent neonatal outcome. Obstet Gynceol 1994; 84: 923–5. Kirsch CF, Feldstein VA, Goldstein PB, Filly RA. Persistent right umbilical vein: prenatal sonographic series without significant anomalies. J Ultrasound Med 1996; 15: 371–7. Shen O, Tadmor OP, Yagel S. Prenatal diagnosis of persistent right umbilical vein. Ultrasound Obstet Gynecol 1996; 8: 31–3. Jouk PS, Champetier J. Abnormal direct entry of the umbilical vein into the right atrium: antenatal detection, embryologic aspects. Surg Radiol Anat 1991; 13: 59–62. Kinare AS, Ambadekar ST, Bhattacharya D, Pande SA. Prenatal diagnosis with ultrasound of anomalous course of the umbilical vein and its relationship to fetal outcome. J Clin Ultrasound 1996; 24: 333–8. Monie IW, Nelson MM, Evans UM. Persistent right umbilical vein as a result of vitamin deficiency during gestation. Circ Res 1957; 5: 187–90. Cohen SB, Lipitz S, Mashiach S, Heggesh J Achiron R. In-utero ultrasonographic diagnosis of abnormal fetal umbilical ductal system associated with hepatic hyperechogenicities. Prenat Diagn 1997; 17: 978–82. Marois D, Heerden JA, Carpenter HA, Sheedy PF. Congenital absence of the portal vein. Mayo Clin Proc 1979; 54: 55–9. Nakasaki K, Tanaka Y, Ohta M et al. Congenital absence of the portal vein. Ann Surg 1989; 210: 190–3.

55. Matsuoka Y, Ohotomo K, Okubo T et al. Congenital absence of the portal vein. Gastrointest Radiol 1992; 17: 31–3. 56. Laverdiere JT, Laor T, Benacerraf B. Congenital absence of the portal vein: case report and MR demonstration. Pediatr Radiol 1995; 25: 52–3. 57. Lewis A, Aquino NM. Congenital portohepatic vein fistula that resolved spontaneously in a neonate. AJR AM J Roentgenol 1992; 159: 837–8. 58. Krabill KA, Lucas RV. Abnormal pulmonary venous connections. In: Emmanouilides GC, Riemenschneider TA, Allen HD, Gutgesell HP, eds. Moss and Adams’ Heart Disease in Infants, Children, and Adolescents, 5th edn. Baltimore: Williams & Wilkins, 1995: 841–9. 59. Patel CR, Lane JR, Spector ML, Smith PC, Crane SS. Totally anomalous pulmonary venous connection and complex congenital heart disease: prenatal echocardiographic diagnosis and prognosis. J Ultrasound Med 2005; 24: 1191–8. 60. Valsangiacomo ER, Hornberger LK, Barrea C, Smallhorn JF, Yoo SJ. Partial and total anomalous pulmonary venous connection in the fetus: two-dimensional and Doppler echocardiographic findings. Ultrasound Obstet Gynecol 2003; 22: 257–63. 61. Allan LD, Sharland GK. The echocardiographic diagnosis of totally anomalous pulmonary venous connection in the fetus. Heart 2001; 85: 433–7. 62. Avni EF, Rypens F, Donnere C, Cuvefliez P, Rodesch F. Hepatic cysts and hyperechogenicities: prenatal assessment and unifying theory on their origin. Pediatr Radiol 1994; 24: 569–72. 63. Hawass HB, El-Badawi MG, Meshera BL, Makanjoula D, Edress JB. Fetal hepatic calcifications. Pediatr Radiol 1990; 20: 528–35. 64. Bronshtein M, Blazer S. Prenatal diagnosis of liver calcifications. Obstet Gynecol 1995; 86: 739–43. 65. Achiron R, Seidman D, Afek A et al. Prenatal ultrasonographic diagnosis of fetal hepatic hyperechogenicities: clinical significance and implications for management. Ultrasound Obstet Gynecol 1996; 7: 251–5.

30 Extracardiac Doppler investigation in fetuses with congenital heart disease Annegret Geipel, Ulrich Gembruch, and Christoph Berg Introduction Since the introduction of Doppler ultrasound in prenatal monitoring, numerous studies have investigated the fetal and uteroplacental circulation in low-risk and high-risk collectives. Arterial and venous Doppler sonography is currently most frequently used in the diagnosis and monitoring of fetuses with intrauterine growth restriction (IUGR). Fetal blood flow is influenced by multiple factors, including the structure and function of the heart as well as impedance of the distal vascular beds. Specific anatomical cardiac defects may therefore lead to alterations of the fetal and uteroplacental blood flow. The significance of Doppler investigations of peripheral fetal blood vessels in congenital heart disease (CHD) has so far been the subject of only a few studies, with in part contradictory results. Extracardiac Doppler has been evaluated for the purpose of screening, monitoring of the fetus with known CHD, and predicting the postnatal outcome. This chapter focuses on second and third trimester Doppler investigations of the umbilical artery (UA), the middle cerebral artery (MCA), and the ductus venosus (DV) in fetuses with congenital heart disease.

Umbilical artery Alterations in umbilical artery blood flow may be secondary to reduced volume flow, reduced cardiac contractility, or increased afterload. In a first study involving 34 fetuses with different heart defects, Al-Gazali et al described pathological results of Doppler investigation of the umbilical artery in 50% of examined fetuses, but no correlation between the severity of the heart defect and the Doppler results was found.1 Copel et al also observed pathological umbilical blood flow patterns in three of 11 fetuses with CHD, but no correlation was found between the Doppler parameters and the obstetric outcome.2 The high incidence of pathological Doppler results of both studies must be

considered in the context of associated extracardiac malformations, chromosomal anomalies, or fetal growth restriction observed in the majority of cases, as fetuses with isolated heart disease were not investigated separately 1,2 (Table 30.1). The same restriction applies to the study by Ursem et al, which investigated different parameters of umbilical artery blood flow in 13 fetuses with CHD. While no differences were found in the fetal heart rate and median pulsatility index (PI) compared to controls, reduced systolic maximum velocity and average maximum velocity among fetuses with CHD was observed. Reduced cardiac contractility as a possible cause was suggested.8 Meise et al were the first to analyze umbilical and cerebral blood flow patterns in 55 second and third trimester fetuses with isolated heart disease, i.e. after the exclusion of cases with aneuploidies, extracardiac malformations, and growth restriction.3 The investigated cases represented a wide spectrum of prenatally diagnosed cardiac malformations, including left heart obstructions, right heart obstructions, outflow-tract disorders, and septal defects. This study demonstrated the insufficiency of Doppler investigation of the UA as a sole screening parameter for CHD (Table 30.1). There were insignificant differences between cases of isolated heart disease and controls with regard to pathological results (PI > 95th centile, 7% vs 4%). As observed by previous studies, a significantly higher percentage of pathological Doppler indexes (48%) was found in the group of non-isolated heart defects (n = 60). The isolated heart defects with a UA PI > 95th centile included two cases with reverse perfusion of the pulmonary artery via the ductus arteriosus (tricuspid and pulmonary atresia, pulmonary atresia, and tricuspid dysplasia), one case with reverse perfusion of the aorta (aortic atresia with consecutive hypoplastic left heart), and one case with Ebstein’s anomaly and a distinct pulmonary valve insufficiency. The increased UA pulsatility of the first three cases was discussed with respect to the right- or left-sided outflow tract obstruction with retrograde perfusion via the ductus arteriosus, with a possible reduction of diastolic blood flow in the descending aorta and in the UA to the extent

428

Fetal Cardiology

Table 30.1 Doppler studies of the umbilical artery in fetuses with isolated congenital heart disease (CHD) Study

Type of CHD

Meise et al,3 2001 4

Mixed

Number of fetuses

Studied parameters

Results CHD vs controls

55

PI (95th centile)

NS

Jouannic et al, 2002

TGA

23

PI (median)

NS

Donofrio et al,5 2003

Mixed

36

RI (mean)

NS

HLH

28

ΔPI

NS

LHO

13

ΔPI

NS

RHO

17

ΔPI

Increased

Mixed

71

PI (mean)

NS

6

Kaltman et al, 2005

Modena et al,7 2006

TGA, transposition of the great arteries; HLH, hypoplastic left heart; LHO, left heart obstruction; RHO, right heart obstruction; PI, pulsatility index; RI, resistance index; NS, not significant.

that an increased umbilical PI resulted. In the fetus with Ebstein’s anomaly and major pulmonary insufficiency, the reduced diastolic blood flow toward the placenta may have been caused by a disturbance of the low-pressure chamber function (‘windkessel function’) of the pulmonary trunk.3 Kaltman et al likewise observed an increased pulsatility in the UA of fetuses with outflow tract obstruction compared to controls, however on a significant level only in fetuses with right heart obstruction (n = 17), whereas no significant differences were identified between the groups with left heart obstruction (n = 13) and hypoplastic left heart (HLH; n = 28)6 (Table 30.1). Donofrio et al investigated the mean resistance index (RI) of the UA in 36 fetuses with congenital heart disease, and could not identify any significant difference in the whole group or in any subgroup as compared to controls.5 A limitation of this study is due to the significant difference of gestational age between the study group and controls, and the limited numbers in some of the subgroups. Likewise Jouannic et al found no difference between the median UA PI in a group of 23 fetuses with transposition of the great arteries (TGA) and in controls4 (Table 30.1). In a recent study, Modena et al analyzed 71 fetuses with various cardiac lesions and confirmed the above data, with no differences in the UA PI, either for the mean or for the 95th centile for gestational age.7 Umbilical artery Doppler in congenital heart disease is not predictive of fetal survival. Although in the studies of Copel et al and Meise et al all fetuses with increased UA pulsatility died either pre- or postnatally, this also applied to five of nine and 19 of 51 fetuses with a normal umbilical Doppler, respectively.2,3

Middle cerebral artery Congenital heart disease may have an impact on cerebrovascular blood flow dynamics in the fetus. Several structural

congenital heart defects are accompanied by an intracardiac mixing of oxygenated and deoxygenated blood. These lesions along with obstructive lesions of the outflow tracts may interfere with normal cerebral oxygen supply and therefore modify cerebrovascular resistance. Alteration of in utero cerebral blood flow could be a critical point of later neurodevelopmental outcome. Meise et al were the first to investigate the cerebral perfusion of fetuses with isolated CHD.3 They found no difference in pulsatility of the middle cerebral artery between the study group and controls, with regard to either ΔPI or 95% reference interval (Table 30.2). In a single case analysis, a reduced PI (< 5th centile) was found in four decompensated fetuses with severe obstructive lesions (aortic atresia with HLH, critical aortic stenosis with endocardial fibroelastosis, two cases of severe pulmonary and aortic stenosis), with the latter three fetuses presenting generalized hydrops. Of the four fetuses, three died in utero and one postnatally.3 In the study by Modena et al, no differences in the mean MCA PI were found between fetuses with CHD and controls. However, significantly more values below the 5th centile were observed in fetuses with CHD (5/71 vs 0/71). These occurred exclusively in lesions with intracardiac mixing.7 In contrast to the study conducted by Meise et al, Donofrio et al found a significantly reduced pulsatility in the MCA as compared to controls in an analysis of fetuses with isolated outflow tract lesions.3,5 However, no parameters independent of gestational age (e.g. Z-score, ΔPI) were used in this study. Whereas investigation of the control group was conducted mainly during the 24th week of gestation, Doppler investigation of the study group took place on average during the 27th week of gestation. It is commonly known that the cerebral pulsatility index reaches its peak at the 24th week of gestation and then decreases. Therefore, the results could be biased – at least in part – by the higher age of gestation of the study group. Jouannic et al investigated a group of 23 fetuses with TGA and found a significantly reduced pulsatility of the

Extracardiac Doppler investigation in fetuses with congenital heart disease

Table 30.2

429

Doppler studies of the middle cerebral artery in fetuses with isolated CHD

Study

Type of CHD

Meise et al,3 2001

Mixed

Jouannic et al,4 2002 5

Donofrio et al, 2003

6

Kaltman et al, 2005

7

Modena et al, 2006

Number of fetuses

Studied parameters

Results CHD vs controls

55

PI (< 5th centile) ΔPI

NS NS

TGA

23

PI (median)

Decreased

Mixed

36

RI (mean)

Decreased

HLH

12

RI (mean)

Decreased

LHO

4

RI (mean)

NS

TGA

4

RI (mean)

Decreased

TOF

11

RI (mean)

Decreased

HRH

5

RI (mean)

NS

HLH

28

ΔPI

Decreased

LHO

13

ΔPI

NS

RHO

17

ΔPI

NS

Mixed

71

PI (mean)

NS

PI (< 5th centile)

p = 0.023

TOF, tetralogy of Fallot; HRH, hypoplastic right heart.

MCA compared to controls, suggesting that cerebral vasodilatation occurred.4 Donofrio et al obtained the same results; however, only four fetuses with TGA were investigated5 (Table 30.2). In their study, Donofrio et al conducted a separate analysis of several subgroups of heart defects: HLH (n = 12), left ventricular obstructive lesion (n = 4), tetralogy of Fallot (TOF, n = 11), and hypoplastic right heart (HRH, n = 5). A significantly reduced MCA RI was further found in fetuses with HLH and TOF. They also used cerebroplacental ratio (CPR) analysis as a measure of cerebral autoregulation and found a decreased CPR in a significant number of fetuses. Kaltman et al looked crosssectionally at 58 fetuses with CHD and observed that leftversus right-sided heart defects modified cerebrovascular impedance differently. In the group of fetuses with hypoplastic left heart (n = 28) the Z-scores of the MCA PI were significantly lower than those of the control group (Table 30.2), whereas fetuses with right heart obstruction (n = 17) showed an increased impedance. The results for the fetuses with left heart obstruction (n = 13) were between those of the controls and the fetuses with HLH.6 The observed blood flow alterations may reflect the specific hemodynamic situation in different heart defects. In normal fetal heart anatomy, the ventricles function in parallel with two distinct shunt pathways which equalize pressure differences. While the left ventricle ejects the blood into a high-resistance system, the head, upper body, and across the aortic isthmus, the right ventricle ejects blood into the lower body and the low-resistance placenta. In normal fetal circulation, oxygen saturation of the blood delivered to the heart and the brain is higher

(approximately 65%) than that delivered to the rest of the body9 (Figure 30.1a). In fetuses with transposition of the great arteries, the oxygenated blood from the left ventricle is ejected into the truncus pulmonalis. Deoxygened blood with an oxygen saturation of only 55% is ejected from the right ventricle into the aorta ascendens, the coronary arteries, and the brain/arm vessels (Figure 30.1b).9 In fetuses with hypoplastic left heart, there is intracardiac mixing of oxygenated blood with deoxygenated blood, as well as retrograde perfusion of the aorta, which results in the delivery of blood with a saturation of approximately 60% to the cerebral and coronary arteries9 (Figure 30.1c). Additional hypoplasia of the aortic arch may lead to a further reduction of blood flow toward the brain. Fetuses with aortic stenosis have blood flow restrictions of various degrees, with only a little intracardiac mixing of blood. In cases with severe restriction, retrograde perfusion of the aortic arch via the ductus arteriosus (DA) with less oxygenated blood occurs. Relatively deoxygenated blood (O2 saturation of about 63%) enters the cerebral circulation due to intracardiac mixing in fetuses with TOF, and especially in fetuses with HRH (pulmonary atresia, severe pulmonary stenosis, tricuspid atresia).9 Considering the location of the most effective fetal chemoreceptors for hypoxia or hypercapnia in the carotid arteries, the observed cerebral vasodilatation could be a fetal autoregulatory response to moderate hypoxia, as described for the abovementioned types of heart defects. Compared to fetuses with a classical blood flow redistribution (‘brainsparing effect’) in response to generalized hypoxia due to

430

Fetal Cardiology

65% O2saturation

(a)

55% O2saturation

(b)

60% O2saturation

Figure 30.1

(c)

uteroplacental dysfunction, most fetuses with isolated heart disease did not show an increased UA pulsatility, which argues in favor of a rather local response to the receipt of less oxygenated blood from the pre-isthmus aorta.4–7 Fetuses with hypoplastic left heart and transposition of the great arteries had the most pronounced alteration of flow, suggesting that they had the greatest need for autoregulation. Knowledge of a heart defect prior to delivery has been shown to improve fetal outcome. However, little is known of the consequences of congenital heart disease on fetal in utero development. In theory, blood flow disturbances may alter normal development. It is a matter of speculation whether the observed cerebral blood flow alterations are a sufficient explanation for the neurodevelopmental disorders described in those children or whether they are biased by the sequelae of surgical intervention.10–12 Nevertheless, the changes described would support the observations made by the Baltimore–Washington Infant Study, with deviations in intrauterine growth patterns in specific cardiac lesions: newborns with TGA were especially prone to a smaller head circumference and intracranial volume in proportion to weight at birth; infants with

O2 saturation in normal heart anatomy and congenital heart disease. Red arrows indicate oxygenated blood, blue arrows indicate deoxygenated blood. (a) Normal fetal blood flow. (b) Transposition of the great arteries. (c) Hypoplastic left heart.

HLH had decreased birth weight as well as a disproportionately small head circumference; newborns with TOF showed a reduced birth weight with proportional small head circumference.13

Ductus venosus The fetal ductus venosus (DV) connects the intra-abdominal umbilical vein to the inferior vena cava at its inlet to the heart. The normal biphasic flow pattern reflects the pressure gradient between the DV and the right atrium during different phases of the cardiac cycle. In conditions with increased cardiac afterload, preload, and/or myocardial dysfunction, elevation of the right atrial as well as the central venous pressure occurs, thus resulting in an increased pulsatility of the venous blood flow velocity waveforms. In fetuses with congenital heart disease, the question arises whether abnormal DV flow profiles indicate fetal compromise or rather reflect the specific hemodynamics of the heart defect itself. First investigations of the DV in fetal heart defects revealed in part significant differences in the blood flow

Extracardiac Doppler investigation in fetuses with congenital heart disease

Table 30.3

431

Doppler studies of the ductus venosus in fetuses with isolated CHD

Study

Type of CHD

Gembruch et al,16 2003

Mixed

94

ΔPVIV

NS

RHM

23

PVIV (95th centile) ΔPVIV

Increased Increased

RHM + VSD

36

PVIV (95th centile)

NS

ΔPVIV

NS

PVIV (95th centile)

Increased

ΔPVIV

Increased

Berg et al,17 2006

Number of fetuses Studied parameters

(DORV, TOF, PA)

PVIV (95th centile)

Results CHD vs controls NS

}

(vs remaining isolated CHD)

Obstructive RHM (TA + VSD, Ebstein’s, PS, PA)

47

RHM, right heart malformation; VSD, ventricular septal defect; DORV, double-outlet right ventricle; PA, pulmonary atresia; TA, tricuspid atresia; PS, pulmonary stenosis; PVIV, peak velocity index for veins.

velocity waveforms. Kiserud et al analyzed the peak systolic and end-diastolic velocities in 30 fetuses with heart defects, including cases with supraventricular tachycardia. Pathological results had a sensitivity of 64% for the diagnosis of CHD, and 81% sensitivity for malformations involving only atrioventricular valves and great arteries. However, the results were biased by the inclusion of fetuses with arrhythmias, aneuploidies, or other extracardiac defects.14 DeVore and Horenstein observed an absent forward flow during the a-wave in one fetus with pulmonary atresia and in a second with severe cardiovascular dysfunction, while one fetus with HLH showed a normal ductus venosus blood flow pattern.15 Gembruch et al were the first to investigate venous Doppler flow profiles (DV, vena cava inferior) in a large group of heterogeneous heart defects.16 In this study, isolated heart defects with hydrops (n = 7) and without hydrops (n = 89) were differentiated from non-isolated heart defects (n = 50). Differences in DV pulsatility (peak velocity index for veins, PVIV) were found between the total group of CHD including non-isolated cases and controls, but not when isolated heart defects were analyzed separately from the other cases. Although 83% of the group of fetuses with cardiogenic hydrops showed an increased DV pulsatility (> 95th centile), the results were insignificant due to the small number of cases. However, in the analysis of further subgroups, a significant increase in DV pulsatility was found in fetuses with right heart malformations (n = 23) compared to the remaining isolated cases (n = 71, Table 30.3). Venous Doppler sonography was not a reliable predictor of fetal outcome in isolated cases of CHD in this study. Although there was a higher survival (78%) for fetuses with normal DV blood flow, 58% of the fetuses survived despite abnormal Doppler examinations, while 22% of fetuses with a normal DV flow pattern died.16 The same

group observed no association between the Doppler indexes of the DV and the survival rate of fetuses with right heart defects.17 Others, however, found a significantly higher DV-PIV in non-survivors than in survivors. In their study, 10 out of 13 non-survivors (77%) demonstrated an abnormal DV PVIV as compared to 12 out of 36 survivors (33%).18 The same association was present in a subanalysis of cases with atrioventricular (AV) septal defects and abnormalities predominantly involving the right ventricle.18 The prognosis of cardiogenic hydrops was uniformly bad in all three studies, with all of them resulting in either intrauterine or postnatal demise.16–18 Building on the results of the first study conducted by Gembruch et al, Berg et al further analyzed 83 fetuses with isolated right heart lesions divided into two groups: group A with a pressure-equalizing ventricular septal defect (VSD) (double-outlet right ventricle, TOF, pulmonary atresia with VSD) and group B with inflow obstruction (tricuspid atresia with VSD), or outflow obstruction with intact septum (Ebstein’s anomaly, pulmonary atresia, pulmonary stenosis). Whereas in group A no significant alterations were found compared to controls, in group B a significantly higher PVIV was observed as well as frequent reverse flow during the a-wave (Figure 30.2). The abnormal DV flow characteristics were not significantly related to signs of heart failure (cardiomegaly, AV-valve insufficiency, hydrops), therefore more likely reflecting elevated cardiac preload rather than decompensation in those heart defects.17 The common pathophysiology of these types of heart defects consists of a relevant increase of right atrial and central venous pressure. In this case, the forward flow during atrial contraction decreases to zero or to retrograde flow, leading to an increased pulsatility of the blood flow profile (Figure 30.3). However, the development of hydrops fetalis is extremely rare in fetuses with

432

Fetal Cardiology

14 Group B 12 10

Z-score

8 6 Group A 4 2 0 –2 –4 n=

585

12

Controls

19

5

TOF DORV

13

14

Ebstein

PA + VSD

7

13

PA + IVS

TA + VSD

PS + IVS

Figure 30.2 Z-scores of ductus venosus peak velocity index for veins of 83 fetuses with right-sided cardiac lesions (group A with a large ventricular septal defect equalizing interventricular pressures, group B with obstruction of the inflow or obstruction of the outflow with intact ventricular septum) and 585 fetuses in the control group (reproduced from reference 17 with permission granted by Wiley & Sons Ltd). DORV, double-outlet right ventricle; Ebstein, Ebsteins’ anomaly; IVS, intact ventricular septum; PA, pulmonary atresia; PS, pulmonary stenosis; TA, tricuspid atresia; TOF, tetralogy of Fallot; VSD, ventricular septal defect.

(a)

Figure 30.3

right ventricular outflow obstructions, even with an intact ventricular septum, as in most cases the left ventricle compensates the abnormal hemodynamics of the right ventricle. In order to tolerate the increased transatrial right-to-left-shunt, a non-restrictive foramen ovale is mandatory. So far, none of the studies have investigated longitudinally the prognostic value of DV pulsatility with regard to the development of hydrops. Sequential examination was performed in 15 fetuses of group B by Berg et al.17 In 10 of the cases, DV pulsatility remained unaltered, in four cases the pulsatility increased, and in one the pulsatility decreased. In none of the cases with a longitudinal increase of DV pulsatility (one with tricuspid atresia, three with pulmonary atresia) was the development of hydrops observed.17 However, in some rare cardiac malformations, such as Ebstein’s anomaly, aortic atresia with restricted foramen ovale, absent pulmonary valve syndrome (APVS), or truncus arteriosus communis (TAC) with severe insufficiency of the common arterial valve, increasing pulsatilities in the flow velocity waveforms of the precordial veins are indicative of increasing venous pressure, announcing cardiac decompensation and impending hydrops fetalis. The pathophysiology of fetuses with severe Ebstein’s anomaly with tricuspid insufficiency is more complex and different from that of other right-sided cardiac lesions. The intrauterine hemodynamics depends not only on the size of the fossa ovalis and the function of the left ventricle, but also on the function of the right ventricle, the extent of tricuspid insufficiency, and the right atrial compliance. A relatively small foramen ovale and/or extreme cardiomegaly, both frequently observed in these fetuses, can lead to substantial left ventricular compression with obstruction

(b)

(a) Sonogram of a fetal heart with tricuspid atresia at 20 + 4 weeks of gestation (RV, right ventricle; LV, left ventricle; RA, right atrium; LA, left atrium). (b) The blood flow velocity waveforms of the ductus venosus demonstrate high pulsatilities with reversal of flow during atrial contraction (S, systole; D, diastole; a, atrial contraction).

Extracardiac Doppler investigation in fetuses with congenital heart disease

(a)

433

(b)

Figure 30.4

(a) Sonogram of a fetal heart with Ebstein’s anomaly at 26 + 3 weeks of gestation. (b) The blood flow velocity waveforms demonstrate higher velocities during diastole (D) than during systole (S).

Figure 30.5 Ductus venosus blood flow with presence of systolic notching (arrows) at 21 + 1 weeks of gestation in a fetus with Ebstein’s anomaly and severe tricuspid regurgitation.

of volume flow, possibly resulting in cardiac decompensation with development of hydrops fetalis. In this specific situation, an increased DV pulsatility may be a sign of impending hydrops and alert the examiner. Accordingly, Berg et al reported on significantly higher DV pulsatilities in two hydropic compared to 11 non-hydropic fetuses with Ebstein’s anomaly.17 Fetuses with severe tricuspid regurgitation of various genesis (tachycardia-induced cardiomyopathy after conversion to sinus rhythm, Ebstein’s anomaly, pulmonary atresia, rhabdomyomas) are occasionally observed to have atypical flow patterns in the ductus venosus, with decrease of the systolic forward flow (Figure 30.4) or

systolic notching (Figure 30.5), both consequences of an impairment of the venous flow during systole due to severe tricuspid insufficiency.16,19 In calculating the venous pulsatility indexes, this shifts the results toward more normal values, and should be considered when monitoring these fetuses.

Summary Fetuses with isolated structural heart defects predominantly show normal arterial and venous blood flow patterns.

434

Fetal Cardiology

Therefore, Doppler sonography of the umbilical artery, middle cerebral artery, and ductus venosus is an insufficient parameter for the purpose of screening. Abnormal blood flow patterns are frequently found in cases additionally complicated by fetal growth restriction, aneuploidies, or extracardiac malformations. Individual groups of heart defects show significant blood flow alterations compared to controls due to their own pathophysiology. This refers especially to cases with severe obstruction of the inflow and outflow. An increased pulsatility of the umbilical artery has been observed specifically in fetuses with severe outflow obstruction and reverse perfusion via the ductus arteriosus. In contrast to fetuses with intrauterine growth restriction and brain sparing following general circulatory redistribution, those with hypoplastic left heart and transposition of the great arteries may have isolated cerebral vasodilatation of the middle cerebral artery as a local adaptive response to moderate hypoxia. Ductus venosus pulsatility may be increased especially in right heart malformations with elevated rightatrial pressure, usually without being associated with cardiac compromise. Neither arterial nor venous Doppler permits reliable prediction of fetal survival; the most determining factors are the type and severity of the underlying heart defect and its influence on postnatal hemodynamics.

References 1. Al-Gazali W, Champman MG, Chita SK, Crawford DC, Allan LD. Doppler assessment of umbilical artery blood flow for the prediction of outcome in fetal cardiac abnormality. Br J Obstet Gynaecol 1987; 94: 742–5. 2. Copel JA, Hobbins JC, Kleinman CS. Can umbilical artery pulsatility index predict the outcome of fetuses with structural heart disease? J Ultrasound Med 1991; 10: 323–6. 3. Meise C, Germer U, Gembruch U. Arterial Doppler ultrasound in 115 second- and third-trimester fetuses with congenital heart disease. Ultrasound Obstet Gynecol 2001; 17: 398–402. 4. Jouannic JM, Benachi A, Bonnet D et al. Middle cerebral artery Doppler in fetuses with transposition of the great arteries. Ultrasound Obstet Gynecol 2002; 20: 122–4.

5. Donofrio MT, Bremer YA, Schieken RM et al. Autoregulation of cerebral blood flow in fetuses with congenital heart disease: the brain sparing effect. Pediatr Cardiol 2003; 24: 436–43. 6. Kaltman JR, Di H, Tian Z, Rychik J. Impact of congenital heart disease on cerebrovascular blood flow dynamics in the fetus. Ultrasound Obstet Gynecol 2005; 25: 32–6. 7. Modena A, Horan C, Visintine J et al. Fetuses with congenital heart disease demonstrate signs of decreased cerebral impedance. Am J Obstet Gynecol 2006; 195: 706–10. 8. Ursem NTC, Clark EB, Pagotto LT, Wladimiroff JW. Fetal heart rate and umbilical artery velocity variability in fetuses with congenital cardiac defects: a preliminary study. Ultrasound Obstet Gynecol 2001; 18: 135–40. 9. Rudolph AM. Congenital Diseases of the Heart. New York: Futura Publishing Company, 2001. 10. Limperopoulos C, Majnemer A, Shevell MI et al. Neurologic status of newborns with congenital heart defects before open heart surgery. Pediatrics 1999; 103: 402–8. 11. Mahle WT, Tavani F, Zimmermann RA et al. An MRI study of neurological injury before and after congenital heart surgery. Circulation 2002; 106(Suppl 1): 109–14. 12. Van Houten JP, Rothman A, Bejar R. High incidence of cranial ultrasound abnormalities in full-term infants with congenital heart disease. Am J Perinatol 2002; 13: 47–53. 13. Rosenthal GL. Patterns of prenatal growth among infants with cardiovascular malformations: possible fetal hemodynamic effects. Am J Epidemiol 1995; 143: 505–13. 14. Kiserud T, Eik-Nes SH, Hellevik LR, Blaas HG. Ductus venosus blood velocity changes in fetal cardiac diseases. J Matern Fetal Invest 1993; 3: 15–20. 15. DeVore GR, Horenstein J. Ductus venosus index: a method for evaluating right ventricular preload in the second-trimester fetus. Ultrasound Obstet Gynecol 1993; 3: 338–42. 16. Gembruch U, Meise C, Germer U, Berg C. Venous Doppler ultrasound in 146 fetuses with congenital heart disease. Ultrasound Obstet Gynecol 2003; 22: 345–50. 17. Berg C, Kremer C, Geipel A et al. Ductus venosus blood flow alterations in fetuses with obstructive lesions of the right heart. Ultrasound Obstet Gynecol 2006; 28: 137–42. 18. Baez E, Steinhard J, Huber A et al. Ductus venosus blood flow velocity waveforms as a predictor for fetal outcome in isolated congenital heart disease. Fetal Diagn Ther 2005; 20: 383–9. 19. Smrcek JM, Krapp M, Axt-Fliedner R et al. Atypical ductus venosus blood flow pattern in fetuses with severe tricuspid valve regurgitation. Ultrasound Obstet Gynecol 2005; 26: 180–2.

31 Electrophysiology for the perinatologist Edgar Jaeggi

Introduction Disturbance of the cardiac rhythm and conduction is an important aspect of fetal medicine. The outcome of the affected fetus may be influenced by a number of variables including the characteristics of the underlying arrhythmia, the potential association with major congenital heart disease, and the choice of perinatal care. The primary objectives of a newly detected rhythm abnormality are to establish its mechanism and hemodynamic fetal consequences and to determine whether the arrhythmia is serious enough to require treatment. This chapter will review relevant aspects of the normal physiology of electrical impulse generation and conduction and then address the electrophysiological mechanisms and diagnostic features of common fetal arrhythmias.

Normal impulse formation and conduction Composed of highly specialized muscle tissue, the cardiac electrical conduction system comprises the sinoatrial (SA) node, the atrioventricular (AV) node, and the His–Purkinje system, including the penetrating bundle of His, bundle branches, and the Purkinje fibers (Figure 31.1). The role of the conduction system is to generate the electrical impulse in the sinus node and to propagate it across the fibrous ring of the AV junction and throughout the ventricles, allowing the sequential electrical and mechanical activation of the atrial and ventricular myocardium with each heartbeat. Following the sequential myocardial depolarization, the conducted impulse is prevented from reactivating the myocardium by refractoriness of the tissue that just has been activated. The heart must then await a new impulse from the SA node for each subsequent activation. These electromechanical actions are made possible by cyclic changes in the transmembrane potential gradient

(membrane potential) that occur in each cardiac cell. Membrane potentials are the result of an unequal distribution of electrically charged ions across the cell membrane. K+ is the main intracellular cation, while phosphate and the conjugate bases of organic acids are the principal anions. Extracellularly, Na+ and Cl− predominate. In its resting state, the cell membrane is mainly permeable to K+. In addition, intracellular Na+ is actively exchanged for K+ by the Na–K adenosine triphosphatase (ATPase) pump, an enzyme complex that resides within the cellular membrane. The net effect is that the interior of the inactive cell exhibits a negative electrical potential with respect to the positively charged extracellular space. Stimulation above a threshold value initiates the opening of voltage-gated ion channels and an influx of positively charged ions into the cell. This triggers the depolarization characteristic of an action potential. Depolarization causes the opening of voltage-gated calcium channels and release of Ca2+ from the t-tubules. The influx of calcium induces calcium release from the sarcoplasmic reticulum, and the free Ca2+ then triggers mechanical actin–myosin interaction with shortening of the myofibers. After a delay which corresponds to the absolute refractory period, potassium channels reopen, and the resulting flow of K+ out of the cell causes repolarization and relaxation of the cardiac cell to its resting state. The cell or fiber again becomes able to elicit an action potential once it reaches its optimal resting membrane potential. If the cell is stimulated before reaching its resting state, it will either remain unresponsive (absolute refractory) or respond inadequately (relative refractory) to an outside stimulus. Cardiac action potential characteristics differ significantly among cells in different portions of the heart.1 Working atrial and ventricular muscle cells maintain the resting membrane potential until they are stimulated by external electrical stimuli. By contrast, cells of the specialized conduction tissue may exhibit automatic properties and spontaneously depolarize without any external influence, thus acting as pacemaker cells. Genetically determined differences in the ion channel population

436

Fetal Cardiology

(a)

Figure 31.1 SAN

AVN

HIS LB RB

LB

PF PF

PF

(b)

QRS T

P ECG V Aorta

MV E

A a

SVC D Diastole

IC

S Systole

IR

Diastole

and in their kinetics are believed to determine the programmed discharge rates of pacemaker cells in the diverse areas of the conduction system. Under normal conditions, it is the cells within the SA node at the upper right atrial wall that generate electrical impulses the fastest, thus overdriving the pacemaker potential of cells within the AV node or the His–Purkinje system. Depending on the degree of fetal activity and the gestational age, the normal fetal cardiac rhythm is characterized by: • regular atrial and ventricular rates between 100 and 180 beats/minute and • a timely normal 1:1 AV relationship. The average fetal heart rate declines from 140 ± 20 beats/ minute by 20 weeks’ gestation to 130 ± 20 beats/minute near term, presumably secondary to an increase in parasympathetic tone.2 The average electrical AV conduction time prolongs by about 20 ms between 15 weeks’ (90 ms) gestation and term (110 ms).3

Electrical conduction system and the correlation between electrical (ECG) myocardial activation and arterial (aorta), intracardiac (MV), and venous (SVC) blood flow. (a) The impulse propagates from the sinoatrial node (SAN) along atrial muscle fibers toward the atrioventricular (AV) junction, stimulating the atrial myocardium to contract. Within the AV node (AVN), providing the only electrical connection between the atria and ventricles across the fibrous ring of the AV junction, the electrical impulse is physiologically delayed, thus functioning as a filter against the propagation of abnormally fast atrial rates or very premature atrial beats to the ventricles. After crossing the AV node, the electrical current rapidly passes through the bundle of His, the right (RB) and left (LB) bundle branches, and the interweaving network of the Purkinje fibers (PF) to the endocardial surfaces of both ventricles. The electrical depolarization then spreads quickly from one ventricular myofiber to the next, so that both ventricles function as synchronous contractile units. (b) There is a physiological delay between the atrial (P wave) and ventricular (QRS complex) tissue depolarization, myocardial contraction, and intracardiac blood flow. The mitral valve (MV) A-wave and superior vena cava (SVC) a-wave result from atrial contraction, while forward flow in the ascending aorta (V) is caused by ventricular contraction. The onsets of the atrial and ventricular systolic flow events are used to assess the mechanical AV relationship by Doppler echocardiography and correlates with the electrical PR duration. Diastole, isovolumic contraction (IC), systole, and isovolumic relaxation (IR) are related to cyclic ventricular filling and contraction. Early diastolic flow: E-wave (MV); D-wave (SVC). Late diastolic flow: A-wave (MV); a-wave (SVC). Systolic flow: V-wave (aorta); S-wave (SVC).

Abnormal impulse generation and conduction Fetal arrhythmias may present as an irregularity of the cardiac rhythm, as an abnormally slow or fast heart rate, or as a combination of irregular rhythm and abnormal heart rate. The contributing causes can be broadly divided into abnormalities in the generation and/or conduction of electrical impulses.4–7 These disturbances result from acute or chronic critical alterations in the electrical activity of the cardiac myocyte and may occur in every region of the heart.

Abnormal impulse generation Cardiac cells in the specialized fibers of the atria, in the AV junction, and in the His–Purkinje system manifest automaticity outside the SA node. They are called latent

Electrophysiology for the perinatologist

pacemakers as they are physiologically suppressed by the more rapid rate of the SA node. Consequently, these ectopic pacemakers do not normally initiate the heartbeat. There are two mechanisms of spontaneous impulse initiation that may lead to arrhythmias, automaticity and triggered activity, each with unique cellular mechanisms in diastolic membrane depolarization. Automatic arrhythmias of the sinus node occur when the SA node fires at an abnormally fast (sinus tachycardia), or slow rate (sinus bradycardia), but is still the dominant pacemaker. Persistent fetal sinus arrhythmia is usually associated with a precipitating factor such as stress, maternal antibodymediated fetal thyroxicosis, hypoxia, and acidosis. Ectopic automatic rhythms occur when the dominant pacemaker shifts from the sinus node to a latent pacemaker, either when the intrinsic rate of the SA node decreases below the intrinsic rate of the ectopic pacemaker, the intrinsic rate of the ectopic pacemaker increases above the normal SA rate (atrial ectopic tachycardia; junctional ectopic tachycardia; ventricular tachycardia), or the normal sinus impulse is prevented from conduction throughout the heart (AV block), leaving an ectopic pacemaker free to fire at its own, slower intrinsic rate.

Abnormal impulse conduction Reentry is the propagation of an impulse through myocardial tissue already activated by the same impulse in a circular movement. Reentry is the underlying mechanism of atrial flutter and reentrant supraventricular tachycardia (SVT). Atrial flutter is sustained by a fast clockwise or counterclockwise rotating atrial macroreentrant circuit. The most common mechanism of fetal SVT is orthodromic AV reentrant tachycardia. It is usually initiated by a premature atrial contraction (PAC) and supported by a circus movement that uses the AV node to conduct from the atria to the ventricles and a rapidly conducting accessory pathway to conduct the impulse back to the atria. Reentry is also possible within the AV node that has two distinct and separate conduction pathways. On the other side, blockage of the propagating cardiac impulse occurs when it arrives in regions of the heart that are not excitable, because either the tissue is still in the refractory period after a recent depolarization (e.g. 2:1 AV conduction ratio during atrial flutter) or the tissue is functionally abnormal (e.g. replacement by scar tissue). In summary, cardiac arrhythmias and conduction abnormalities may have diverse pathologic causes. Arrhythmogenic mechanisms can be deduced with variable certainty from the clinical information and electrophysiological features, including the rate, chronology, and morphology of atrial and ventricular electrical events by postnatal electrocardiography (ECG).

437

Intrauterine investigation of fetal rhythm and atrioventricular conduction The assessment of the fetal cardiac rhythm is more challenging, as conventional ECG may not be obtained. Transmaternal fetal ECG, which has been used by a handful of fetal medicine centers, is based on signalaveraging of electrocardiographic complexes.3,8 This may provide useful information on cardiac time intervals such as PR, QRS, and QT duration during a stable cardiac rhythm, but it does not allow the analysis of individual cardiac cycles. Ultrasound imaging of the fetal heart has been used alternatively to study the beat-to-beat chronology of atrial and ventricular electrical events by their respective mechanical consequences. By way of M-mode and tissue Doppler imaging, it is possible to simultaneously record the sequence and time-relationship of atrial and ventricular systolic wall movements (Figure 31.2).3,9–12 Simultaneous recording of blood flow velocities that represent atrial and ventricular systolic events by pulsed Doppler echocardiography follows the same concept (Figure 31.3).11,13–15 By placing the Doppler cursor across the superior vena cava (SVC) and the ascending aorta, the flow velocity pattern from both vessels (SVC/aorta Doppler) may be recorded. The beginning of the retrograde flow in the SVC (a-wave) reflects the onset of atrial systole, whereas the onset of aortic forward flow marks the beginning of ventricular systole. The morphology of the systemic venous a-wave is markedly influenced by its relationship with the ventricular systole. The tallest a-waves (cannon-waves) are observed when atria and ventricles contract almost at the same time, as in orthodromic AV reentrant tachycardia. In this situation, the AV valves cannot open normally, and the contracting atria will pump backwards into the systemic and pulmonary veins. Alternatively to the SVC/aorta Doppler method, the pulmonary artery and vein or the left ventricular inflow and outflow may be simultaneously recorded by pulsed Doppler imaging. The time interval between the onsets of mitral A-wave flow and ascending aorta flow reflects the beginning of atrial and ventricular systole, respectively, and permits indirect conclusions on the electrical atrioventricular chronology.

Echocardiographic assessment of the fetal atrioventricular conduction system Electrophysiological ‘normality’ may be assumed if a regular and normocardic fetal heart rate with a timely

438

Fetal Cardiology

(a)

(b)

Figure 31.2 Simultaneous assessment of the normal sequential atrial and ventricular systolic wall motion by M-mode and tissue Doppler echocardiography. (a) M-mode. Guided by real-time two-dimensional echocardiography, the M-mode ultrasound beam is aligned simultaneously through the atrial (A) and ventricular (V) walls to record the sequence of their systolic wall motions. The preferred M-mode beam direction is through the atrial and ventricular walls immediately above and below the AV junction, because these parts show the most pronounced lateral excursion during a cardiac cycle. (b) Tissue Doppler. Normal longitudinal wall motion velocity curves of the right ventricular myocardial free wall (yellow curve) and the upper right atrial wall (green curve) are shown for two cardiac cycles. The ventricular curve is typically composed of four waves per cardiac cycle: two diastolic waves are produced by tissue motion away from the apex of the heart during early diastolic filling (Ea) and during atrial contraction (Aa), followed by two waves directed towards the apex during isovolumic contraction (IV) and ventricular systole (Sa). The time interval between the onsets of Aa and IV (interval 1) closely reflects electrical PR duration and may be particularly useful in surveillance of the fetus at risk of evolving AV block.3

normal 1:1 AV relationship is documented by echocardiography. Gestational age-matched reference values of AV time intervals have been developed for Doppler echocardiographic methods (Figure 31.4).3,11,13 Indeed, AV time measurement has gained recent popularity as it allows the serial monitoring of fetuses exposed to maternal anti-Ro/SSA and anti-La/SSB autoantibodies. These antibodies are found in about 1–2% of all pregnant women, with a similar fetal risk of developing progressive immunemediated AV nodal damage at around 20–24 weeks of gestation. There is some evidence that the evolving complete fetal AV block may be preventable if the disease can be recognized and treated at an earlier stage of AV nodal disease which is clinically characterized by a short-lived appearance of incomplete AV block.16–18 First degree heart block as the most subtle electrical conduction anomaly is characterized by a 1:1 AV relationship with a longer electrical PR interval than usual for age and heart rate. Thus, a simple, precise diagnostic tool that allows reliable detection of subtle electrical AV conduction anomalies is indispensable for the surveillance of those fetuses exposed to maternal antibodies. Nevertheless, there has been some controversy regarding the diagnostic accuracy and relevance of ultrasoundderived first degree AV block which is based on the prolongation of mechanical AV intervals. Using pulsed Doppler echocardiography and previously published

reference data,13 transient first degree AV block has been reported in up to 25% of fetuses exposed to maternal anti-Ro/La antibodies.19 While the relevance of these findings has been contested,20,21 it illustrates potential drawbacks in diagnosing electrophysiological abnormalities based on echocardiographic imaging and interpretation. Myocardial contraction or blood flow into and from the ventricles occurs after the electrical excitation, as illustrated schematically in Figure 31.1. The degree of correlation between electrophysiological PR and ultrasound-derived mechanical AV intervals is primarily determined by differences in electromechanical AV activation.3 Electromechanical AV delay may, however, be affected by numerous variables including loading condition, intrinsic myocardial properties, heart rate and size, and ultrasound interrogation sites and techniques. Moreover, to allow reproducible and accurate measurement of AV intervals, Doppler or M-mode tracings should be recorded at fast sweep speeds (at least 100 mm/s).

Echocardiographic assessment of fetal arrhythmias Arrhythmias are detected in at least 2% of all pregnancies during routine obstetrical examinations and are a common

Electrophysiology for the perinatologist

439

Figure 31.3 Simultaneous assessment of atrial and ventricular systolic flow velocities by pulsed Doppler echocardiography. (a) Simultaneous superior vena cava and ascending aorta (SVC/aorta) pulse wave Doppler. In the sagittal view of the fetal thorax, the connection of the superior vena cava (+) to the right atrium and part of the adjacent ascending aorta (∗) can be seen. Positioning of the enlarged Doppler sample volume () within both vessels allows simultaneous recording of the SVC and aortic blood flow velocities. In systole (S-wave) and early diastole (D-wave), SVC flow is directed toward the fetal heart while it is in the opposite direction during atrial contraction (a-wave). Atrioventricular (AV: from the onset to the a-wave and V-wave) and ventriculoatrial (VA: from V to a) intervals measured with this method provide detailed and precise information in cases of complex fetal arrhythmias.11,14 The technique of measuring the AV interval is illustrated by vertical bars. (b) Simultaneous mitral valve and ascending aorta (left ventricular inflow/outflow) pulse wave Doppler. Simultaneous inflow/outflow Doppler, which is easily obtained in a cranially angulated four-chamber view of the fetal heart (with the aorta seen as the fifth chamber). The Doppler sample volume () is positioned within the mitral valve and the left ventricular outlet (LVOT). This approach is mainly useful to document the AV relationship during normal rhythm. The beginning of the mitral A-wave marks the onset of atrial systole, while the beginning of aortic flow (V) marks the onset of ventricular systole. The technique of measuring the AV interval is illustrated by vertical bars.

reason for referral to a fetal cardiologist. Precise identification of the underlying etiology is essential, since management and prognosis can be quite different. M-mode and SVC/aorta Doppler are preferred imaging modalities of fetal arrhythmias, as informative, good quality tracings can be obtained in the majority of cases.10,14,17,18,22 Fetal rhythm and conduction disorders present with typical electrophysiological features (Table 31.1) that can be discerned by a stepwise interpretation of the ultrasound recording. These steps include determining: • the rate and regularity of atrial (A–A) and ventricular (V–V) events and • the relationship of atrial mechanical events to ventricular mechanical events (A:V conduction ratio; A–V and V–A chronology).

Irregular rhythm In at least 90% of an unselected pregnancy population, arrhythmias are brief, isolated, and clinically benign events,

typically presenting as occasional ‘skipped beats’ due to isolated PACs (Figure 31.5).22,23 Premature contractions are easily detectable, and present either as an isolated rhythm disorder or, in a minority of cases, in association with other arrhythmias. If PACs are conducted, they are to be distinguished from premature ventricular contractions, which are rare in the fetus (Figure 31.6). If they are blocked they need to be differentiated from second degree AV block, another fairly uncommon arrhythmia. Unlike PACs, second degree AV block denotes regular atrial rates with a failure to conduct some, but not all, atrial impulses to the ventricles. Thus, the atrial, but not the ventricular, rate is typically regular.

Abnormal rates Far less common, but potentially more dangerous, are anomalies that lead to prolonged or persistent fetal bradycardia (heart rate < 100 beats per minute) or tachycardia (heart rate > 180 beats per minute). While many of these

440

Fetal Cardiology

V/AO

LV in/out

Y = 0.7x + 94.1, R2 = 0.21, p < 0.0001

160

160

150

150

140

140

AV interval (ms)

AV interval (ms)

Y = 0.6x + 103.0, R2 = 0.19, p < 0.0001

130 120 110 100 90 10

130 120 110 100

15

20

25 30 Weeks

35

40

90 10

45

15

20

25 30 Weeks

35

40

45

Ao-Sa

Ao-IV

Y = 1.21x + 104.0, R2 = 0.30, p < 0.0001

Y = 0.9x + 69.1, R2 = 0.31, p < 0.0001 180

120

170 160 AV interval (ml)

AV interval (ms)

110 100 90 80

140 130 120 110

70 60 10

150

100 15

20

25 30 Weeks

35

40

45

90 15

20

25

30 Weeks

35

40

45

Figure 31.4 Gestational age-matched reference values of mechanical AV intervals (reproduced with permission from reference 3). Lines denote regressions and 95% confidence limits for individual observations. Aa, late diastolic wall velocity; Aa-IV, time interval from Aa to IV; Aa-Sa, time interval from Aa to Sa; AV, atrioventricular; LV in/out, left ventricular inflow/outflow pulse Doppler method; R2, coefficient of determination; Sa, systolic wall velocity during ejection phase; V/AO, superior vena cava/aorta pulse Doppler method.

heart rate abnormalities are well tolerated, at the most severe end of the spectrum they may result in low cardiac output failure, fetal hydrops, neurological damage, and even death.

Bradyarrhythmias Brief episodes of sinus bradycardia that last less than 1–2 minutes are frequent and benign findings, particularly early in gestation. A prolonged decrease in heart rate is observed during sinus bradycardia (Figure 31.7), atrial bigeminy with blocked premature beats (Figure 31.8), and high-degree AV block (Figure 31.9). During persistent sinus bradycardia, which may have serious etiologies such as fetal long-QT syndrome or fetal distress, the atrial rate is regular and slow with a normal 1:1 AV relationship.24–26

In contrast, blocked premature beats and AV block are characterized by higher atrial than ventricular rates. In non-conducted atrial bigeminy, every second atrial impulse occurs prematurely enough to fail conduction by the physiologically refractory AV junction. In high-degree AV block, the time interval between consecutive atrial impulses hardly varies, but propagation of the electrical impulse to the ventricles is intermittently (second degree AV block), alternatively (2:1 AV block), or completely (complete or third degree AV block) interrupted. The heart rate in second degree AV block is only regular if beats are ‘dropped’ in a regular 2:1 or higher AV ratio. The differentiation between non-conduction across a refractory AV node and real AV block is important, as bradycardia secondary to non-conducted PACs typically resolves spontaneously without treatment, unlike highdegree AV block that is typically associated with major

160–200 190–280 300–500 180–230 Slow–normal Normal

Sinus

AV reentry

Atrial flutter

JET, PJRT

Ventricular (+VA block)

Junctional ectopic (+VA block)

Regular

Normal Slow–normal

2:1 AV block

Third degree AVB

Regular

Regular

Regular

Regular

Regular

Regular

Regularly irregular

Normal

Atrial bigeminy, blocked

Regular

75–90

Sinus

Regular

Normal

Second degree AVB Wenckebach

Regularly irregular

Regular

Normal

Atrial bigeminy, conducted

Regularly irregular

Normal

Normal

Atrial trigeminy, blocked

Regularly irregular

Ventricular bigeminy

Normal

Atrial trigeminy, conducted

Irregular

Regular

Normal

Isolated PAC, blocked

Irregular

A–A interval

Isolated PVC (+ VA block) Normal

Normal

A rate

Isolated PAC, conducted

Arrhythmia

< 1:1

1:1

Mainly 2:1

1:1

1:1

Dissociated

2:1

2:1

160–260

180–230

150–250

190–280

160–200

35–80

60–75

65–90

75–90

Normal

> 1:1 1:1

Normal

Normal

Normal

Normal

1:2

< 1:1

1:1

3:2

Normal

Normal

> 1:1 1:1

Normal

V rate

1:1

A–V relation

Typical prenatal presentation of fetal rhythm and conduction anomalies

Regularly irregular

Regular

Mainly regular

Regular

Regular

Regular

Regular

Regular

Regular

Irregular

Regularly irregular

Irregular

Regularly irregular

Regularly irregular

Regularly irregular

Irregular

Irregular

V–V interval

Dissociated

Long VA

Short VA

Long VA

Long VA

Variable

Variable

Variable

V–A interval

Major, treatable

+ Rare

Major, treatable

Major, treatable

Major, treatable

++ Rare

Depends on cause

Rare

Major, may progress Major, irreversible

Minor, transient

+ Rare

Depends on cause

+

++

May progress

Minor, transient Rare

Rare

Minor, transient

Minor, transient

Minor, transient

Minor, transient

Minor, transient

Minor, transient

+++ +++

Relevance, outcome

Incidence

Incidence: +++, detected in 1/10–1/1000 pregnancies; ++, in 1/1000–1/10 000 pregnancies; +, in 1/10 000–1/100 000 pregnancies; rare, affects fewer than 1/100 000 pregnancies.

A, atrial; A–V, atrioventricular; JET, junctional ectopic tachycardia; PAC, premature atrial complex; PJRT, permanent junctional reciprocating tachycardia; PVC, premature ventricular complex; V, ventricular; V–A, ventriculoatrial; VA block, absence of retrograde conduction via AV node or accessory pathway.

Tachycardia

Bradycardia

Irregular rhythm

Table 31.1

Electrophysiology for the perinatologist 441

442

Fetal Cardiology

Conducted SAN

PAC

A AVN

Non-conducted SAN

PAC

A AVN

Figure 31.5 SVC/aorta Doppler recording of premature atrial contraction (PAC). Conduction of PAC to the ventricles ranges from normal conduction when the PAC occurs late after the refractory period of the AV node/His–Purkinje system (top panel) to non-conduction when the PAC arises early and falls into the refractory period of the conduction system (lower panel). When compared to the normal atrial (A–A) intervals, the atrial–PAC interval is visibly shorter. The ventricular contraction occurs prematurely if the PAC is conducted, or is missing if the PAC is non-conducted to the ventricles. In both situations, the ventricular rate is irregular unless non-conducted PACs occur in a bigeminal pattern (see Figure 31.8).

SAN

A AVN

PVC

Figure 31.6 M-mode recording of premature ventricular contraction (PVC). Guided by real-time two-dimensional fetal echocardiography, the M-mode ultrasound beam is aligned simultaneously through the ventricular (V) and atrial (A) walls to record the sequence of their systolic wall motions. In this particular case, the atrial rhythm (A–A) is regular, while the ventricular rhythm (V–V) is regularly irregular due to prematurity of every alternating ventricular beat (ventricular bigeminy). The average ventricular rate remains normal, despite the fact that only every second atrial beat is conducted (indicated with arrows).

Electrophysiology for the perinatologist

SAN

A

443

AVN

Figure 31.7 SVC/aorta Doppler recording of sinoatrial bradycardia. The tracing demonstrates abnormally slow, but regular atrial (A) and ventricular (V) rates that occur in a normal 1:1 AV relationship.

SAN

A

A2 AVN

Figure 31.8

SVC/aorta Doppler recording of atrial bigeminy. Normal SA node impulse (A) and premature atrial contraction (A2 = PAC) alternate. In this example, the PAC occurs prematurely enough to regularly fail conduction to the ventricles. Sustained atrial bigeminy with blocked premature beats may lower the average heart rate of the fetus to 60-100 beats per minute. Careful echocardiographic investigation is required to distinguish this benign and transient cause of bradyarrhythmia from potentially life-threatening high-degree AV block or sinus bradycardia.

SAN

A

AVN

Figure 31.9 SVC/aorta Doppler tracing of complete AV block. The atria (A) and ventricles (V) beat independently and regularly at their intrinsic rates because the electrical AV communication is completely interrupted. This results in ventricular rates of typically around 55–60 beats per minute (range: 35–80 beats/minute). Structural heart disease is prevalent in about 40% of fetuses with complete AV block. In the remaining cases, AV block is typically associated with maternal anti-Ro/La autoantibodies.

444

Fetal Cardiology

SAN AVN

Figure 31.10 Fetal atrial flutter. The M-mode tracing shows an atrial (A–A) rate that exceeds 300 beats per minute and is twice as fast as the ventricular (V–V) rate. This is explained with atrial flutter and 2:1 AV conduction. Fetal atrial flutter is often already suspected by 2D echocardiography, because of the very fast moving atrial walls. The diagnosis is then confirmed by M-mode or tissue Doppler imaging.

SAN AVN

Figure 31.11 Short VA tachycardia. On the SVC/aorta Doppler recording, the a-wave (A) flow reversal in the SVC occurs close to the end of aortic forward flow wave (V), suggesting that atrial systolic contraction occurs shortly after the ventricular systolic contraction. The AV and VA intervals are marked by vertical bars and indicate that the VA interval is significantly shorter when compared to the AV interval. This ‘short VA pattern’ is explained by orthodromic AV reentry or orthodromic AV node reentry (not shown). AV reentry is by far the most common mechanism of fetal tachycardia. The ventricles are excited through the normal path of the AV node, while the atria are excited through a fast, retrogradely (VA) conducting accessory pathway. As the VA conduction is faster than the AV nodal conduction, the VA interval is shorter than the AV duration on echocardiography.

structural heart disease, maternal anti-Ro/SSA and anti-La/ SSB autoantibodies, or congenital long-QT syndrome.27–32 In third degree or complete AV block, the electrical AV communication is completely interrupted and the atria and ventricles beat independently at their intrinsic pacemaker rates (Figure 31.9).

Tachyarrhythmias Fetal tachycardia typically relates to atrial flutter, supraventricular tachycardia (SVT), and sinus tachycardia as

the main etiologies.10,22,33–35 In atrial flutter (Figure 31.10) the atrial rate exceeds 300 beats per minute, which is sufficiently fast that only every second or third atrial beat is conducted to the ventricles through the AV node, resulting in ventricular response rates between 150 and 250 beats per minute. SVT encompasses three different arrhythmia mechanisms: (1) AV reentrant tachycardia related to fast retrograde accessory pathway conduction; (2) permanent junctional reciprocating tachycardia (PJRT) related to slow retrograde pathway conduction; and (3) atrial ectopic tachycardia (AET) due to enhanced atrial

Electrophysiology for the perinatologist

445

A SAN

A

SAN

SAN AVN

AVN

AVN

A

Sinus tachycardia

Atrial ectopic tachycardia

‘Slow’ VA pathway

Figure 31.12 Long VA tachycardia. The SVC/aorta Doppler tracing demonstrates fetal tachycardia with 1:1 AV conduction. The AV and VA intervals are marked by vertical bars and indicate that the VA interval is significantly longer when compared to the AV interval. This ‘long VA pattern’ may be explained by various arrhythmia mechanisms, including abnormal impulse generation in the SA node (sinus tachycardia) or atria (atrial ectopic tachycardia), and AV reentry through a slowly retrograde-conducting accessory pathway (permanent junctional reciprocating tachycardia).

6%

2%

n = 101

9% SVT 34%

4%

Atrial flutter Sinus tachycardia Complete AV block 2nd degree AV block Atrial bigeminy Sinus bradycardia Other

31% 11% 3%

Figure 31.13 Mechanisms of clinically relevant fetal arrhythmia diagnosed at the Hospital for Sick Children between the years 2000 and 2006. SVT, supraventricular tachycardia.

focal automaticity. They are distinguished based on their arrhythmia pattern and VA time relationship.10,14 AV reentry presents as a short VA tachyarrhythmia, as the retrograde atrial activation proceeds across a fast conducting accessory pathway and therefore occurs shortly after the ventricular contraction (Figure 31.11). In long VA SVT, the atrial contraction closely precedes the ventricular contraction. This activation pattern is typically seen during AET, PJRT, and sinus tachycardia. Sinus tachycardia is

usually slower than AET and PJRT and characterized by atrial rates of 180–200 beats per minute, normal 1:1 AV conduction, and some variability of the fetal heart rate (Figure 31.12). A variety of fetal and maternal conditions may be responsible for sustained sinus tachycardia including fetal distress, anemia, infections, maternal β-stimulation, and fetal thyrotoxicosis. The importance of sinus tachycardia is recognizing and treating the underlying cause. Ventricular tachycardia and junctional ectopic tachycardia are exceptional causes of fetal tachyarrhythmias.22,36 If there is no retrograde conduction across the AV node or an accessory pathway, the ventricular rate will exceed the atrial rate during ventricular or junctional tachycardia. If there is retrograde 1:1 VA conduction, these arrhythmias become difficult to discern from SVT. In summary, clinically relevant fetal rhythm disorders are not common, and even large tertiary care centers will see only a handful of affected fetuses every year (Figure 31.13). Non-invasive documentation of the underlying arrhythmia mechanism and the fetal well-being is possible by means of two-dimensional, M-mode, and Doppler ultrasound imaging. A stepwise diagnostic approach should be used to examine the regularity, rates, chronology, and conduction ratio of atrial and ventricular events, and to conclude on the most likely rhythm diagnosis. This approach will reduce the risk of irrational

446

Fetal Cardiology

drug treatment or premature delivery of fetuses with more benign findings, while it may facilitate the care and improve the outcome of those with major rhythm disturbances. Management and prognosis of the various types of fetal arrhythmias are discussed in the chapters immediately following.

References 1. Pickoff AS. Developmental electrophysiology in the fetus and neonate. In: Polin RA, Fox WW, Abman SH, eds. Fetal and Neonatal Physiology, 3rd edn. Philadelphia: Saunders, 2004; 1: 669–89. 2. Wheeler T, Murrils A. Patterns of fetal heart rate during normal pregnancy. Br J Obstet Gynaecol 1978; 85: 18–27. 3. Nii M, Hamilton RM, Fenwick L et al. Assessment of fetal atrioventricular time intervals by tissue Doppler and pulse Doppler echocardiography: normal values and correlation with fetal electrocardiography. Heart 2006; 92: 1831–7. 4. Waldo AL, Wit AL. Mechanisms of cardiac arrhythmias. Lancet 1993; 341: 1189–93. 5. Cabo C, Wit AL. Cellular electrophysiologic mechanisms of cardiac arrhythmias. Cardiol Clin 1997; 15: 517–38. 6. Boyden PA. Cellular electrophysiologic basis of cardiac arrhythmias. Am J Cardiol 1996; 78: 4–11. 7. Task Force of the Working Group on Arrhythmias of the European Society of Cardiology. The Sicilian gambit. A new approach to the classification of antiaarhythmic drugs based on their actions on arrhythmogenic mechanisms. Circulation 1991; 84: 1831–51. 8. Smith MJ, Thomas M, Green AR et al. Non-invasive fetal electrocardiography in singleton and multiple pregnancies. BJOG 2003; 110: 668–78. 9. Kleinman CS, Donnerstein RL, Jaffe CC et al. Fetal echocardiography. A tool for evaluation of in utero cardiac arrhythmias and monitoring of in utero therapy: analysis of 71 patients. Am J Cardiol 1983; 51: 237–43. 10. Jaeggi E, Fouron JC, Fournier A et al. Ventriculo-atrial time interval measured on M-mode echocardiography: a determining element in the diagnosis, treatment, and prognosis of fetal supraventricular tachycardia. Heart 1998; 79: 582–7. 11. Fouron JC, Proulx F, Miro J, Gosselin J. Doppler and M-mode ultrasonography to time fetal atrial and ventricular contractions. Obstet Gynecol 2000; 96: 732–6. 12. Rein AJJT, O’Donnell CO, Geva T et al. Use of tissue velocity imaging in the diagnosis of fetal cardiac arrhythmias. Circulation 2002; 106: 1827–33. 13. Andelfinger G, Fouron JC, Sonesson SE, Proulx F. Reference values for time intervals between atrial and ventricular contractions of the fetal heart measured by two Doppler techniques. Am J Cardiol 2001; 88: 1433–6. 14. Fouron JC, Fournier A, Proulx F et al. Management of fetal tachyarrhythmia based on superior vena cava/aorta Doppler flow recordings. Heart 2003; 89: 1211–16.

15. Carvalho JS, Prefumo F, Ciardelli V et al. Evaluation of fetal arrhythmias from simultaneous pulsed wave Doppler in pulmonary artery and vein. Heart 2007; 93: 1448–53. 16. Saleeb S, Copel J, Friedman D, Buyon JP. Comparison of treatment with fluorinated glucocorticoids to the natural history of autoantibody-associated congenital heart block: retrospective review of the research registry for neonatal lupus. Arthritis Rheum 1999; 42: 2335–45. 17. Raboisson MJ, Fouron JC, Sonesson SE et al. Fetal Doppler echocardiographic diagnosis and successful steroid therapy of Luciani-Wenckebach phenomenon and endocardial fibroelastosis related to maternal anti-Ro and anti-La antibodies. J Am Soc Echocardiogr 2005; 18: 375–80. 18. Rosenthal D, Friedman DM, Buyon J, Dubin A. Validation of the Doppler PR interval in the fetus. J Am Soc Echocardiogr 2002; 15: 1029–30. 19. Sonesson SE, Salomonsson S, Jacobsson LA, Bremme K, Wahren-Herlenius M. Signs of first-degree heart block occur in one-third of fetuses of pregnant women with anti-SSA/Ro 52-kd antibodies. Arthritis Rheum 2004; 50: 1253–61. 20. Rein AJ, Mevorach D, Perles Z, Shovali A, Elchalal LL. Fetal first-degree heart block, or where to set the confidence limit: comment on the article by Sonesson et al. Arthritis Rheum 2005; 52: 366; author reply 366–8. 21. Buyon JP, Askanase AD, Kim MY, Copel JA, Friedman DM. Identifying an early marker for congenital heart block: when is a long PR interval too long? Comment on the article by Sonesson et al. Arthritis Rheum 2005; 52: 1341–2. 22. Fouron JC. Fetal arrhythmias. The Sainte-Justine Hospital experience. Prenat Diagn 2004; 24: 1068–80. 23. Kleinman CS, Nehgme RA. Cardiac arrhythmias in the human fetus. Pediatr Cardiol 2004; 25: 234–51. 24. Maeno Y, Rikitake N, Toyoda O et al. Prenatal diagnosis of sustained bradycardia with 1: 1 atrioventricular conduction. Ultrasound Obstet Gynecol 2003; 21: 234–8. 25. Beinder E, Grancay T, Menendez T, Singer H, Hofbeck M. Fetal sinus bradycardia and the long QT syndrome. Am J Obstet Gynecol 2001; 185: 743–7. 26. Hofbeck M, Ulmer H, Beinder E, Sieber E, Singer H. Prenatal findings in patients with prolonged QT interval in the neonatal period. Heart 1997; 77: 198–204. 27. Todros T, Presbitero P, Gaglioti P, Demarie D. Conservative management of fetal bigeminy arrhythmia leading to persistent bradycardia. Eur J Obstet Gynecol Reprod Biol 1990; 34: 211–15. 28. Lin MT, Hsieh FJ, Shyu MK et al. Postnatal outcome of fetal bradycardia without significant cardiac abnormalities. Am Heart J 2004; 147: 540–4. 29. Schmidt KG, Ulmer HE, Silverman NH, Keinman CS, Copel JA. Perinatal outcome of fetal complete atrioventricular block: a multicenter experience. J Am Coll Cardiol 1991; 17: 1360–6. 30. Jaeggi ET, Hornberger LK, Smallhorn JF, Fouron JC. Prenatal diagnosis of complete atrio-ventricular block associated with structural heart disease since 1990: combined experience of two tertiary care centers and review of the literature. Ultrasound Obstet Gynecol 2005; 26: 16–21. 31. Miura M, Yamagishi H, Morikawa Y, Matsuoka R. Congenital long QT syndrome and 2:1 atrioventricular

Electrophysiology for the perinatologist

block with a mutation of the SCN5A gene. Pediatr Cardiol 2003; 24: 70–2. 32. Lupoglazoff JM, Denjoy I, Villain E et al. Long QT syndrome in neonates: conduction disorders associated with HERG mutations and sinus bradycardia with KCNQ1 mutations. J Am Coll Cardiol 2004; 43: 826–30. 33. Till J, Wren C. Atrial flutter in the fetus and young infant. Br Heart J 1992; 67: 80–3.

447

34. Jaeggi E, Fouron JC, Drblik S. Atrial flutter in the fetal period: diagnosis, clinical features, treatment and outcome. A review. J Pediatr 1998; 132: 335–9. 35. Zimmerman D. Fetal and neonatal hyperthyroidism. Thyroid 1999; 9: 727–33. 36. Villazon E, Fouron JC, Fournier A, Proulx F. Prenatal diagnosis of junctional ectopic tachycardia. Pediatr Cardiol 2001; 22: 160–2.

32 Fetal bradydysrhythmia Klaus G Schmidt

The normal fetal cardiac rhythm is regular, and the fetal heart rate ranges from 100 to 180 beats per minute during the second half of pregnancy, depending on gestational age and degree of fetal activity. Any disturbance of the fetal cardiac rhythm, therefore, may present as an irregular rhythm or as an abnormal heart rate – either abnormally slow or abnormally fast. A combination of both irregular rhythm and abnormal heart rate can also be seen. This chapter will deal with fetal bradydysrhythmia, which has the common feature of a heart rate of less than 100 beats per minute.

Diagnostic approach Postnatally, cardiac rhythm is commonly documented and analyzed through recordings of the electrical activity of the heart, the electrocardiogram (ECG). Due to the small amplitude of the fetal ECG signals recorded from the maternal abdominal wall, which may be obscured additionally by maternal ECG signals and abdominal muscular activity, conventional fetal ECG recordings have been of limited use for analysis of fetal dysrhythmias. In particular, the separation of atrial and ventricular activity has been difficult in conventional fetal ECGs. Recent advances in signal processing and filtering have improved the transabdominal recording of fetal ECGs, but limited success rates even in the most experienced hands, as well as a rather complex recording system and procedure, may preclude the widespread use of fetal ECG in prenatal cardiac rhythm analysis.1 Similarly, the use of fetal magnetocardiography has been proposed as a substitute for fetal ECG,2,3 but such recordings are even more difficult to maneuver and cannot be used for routine assessment of fetal cardiac rhythm at present. Because of the difficulties with recording fetal ECGs, it has become common practice to analyze the fetal atrial and ventricular cardiac activity by sonographic methods during the second and third trimesters of pregnancy.4–6 One has to keep in mind, however, that cardiac rhythm

analysis based on ultrasonography uses an indirect approach, because the observer draws conclusions on electrical events from the recorded mechanical events, which may limit the precision in defining fetal dysrhythmias. Both the regularity of fetal cardiac rhythm and the fetal heart rate can be assessed using pulsed Doppler flow signals obtained from sampling sites within the heart or from large vessels close to it (Figure 32.1). In the presence of a regular rhythm and a normal rate, further analysis may not be required. In the presence of an irregular rhythm or an abnormal fetal heart rate, however, further analysis of the atrial and ventricular activities and their respective association is obligatory for a complete understanding of the existing dysrhythmia. Using M-mode echocardiography, atrial and ventricular activities can be displayed simultaneously by placing the sampling line across an atrial and a ventricular wall (Figure 32.2). The onset of wall contraction reflects atrial and ventricular activation, respectively. In a similar approach the semilunar valve opening may be used as an indicator of ventricular activation (Figure 32.3). The simultaneous display of atrial and ventricular activation is also possible using Doppler ultrasound.7 Several sampling sites have been suggested for this purpose.8–10 Placing the sample volume into the left ventricular outflow tract is a frequently used technique that displays the sequence of inflow and outflow signals, thus reflecting the sequence of atrial and ventricular activation (Figure 32.4). Other sampling sites that show atrial and ventricular flow events in close proximity, such as from the ascending aorta and superior caval vein, can also be used to identify atrial and ventricular contractions.9 Particular attention has been paid recently to the assessment of the time delay between atrial and ventricular contractions using Doppler ultrasound.11 The atrioventricular contraction time interval may represent a mechanical analog to the electrical PR interval (Figures 32.4 and 32.5). Reasonable agreement was found when comparing measurements of atrioventricular contraction time interval using Doppler with the PR interval on the surface ECG in fetal lambs.12 In human fetuses, reference values

450

Fetal Cardiology

Figure 32.1 Pulsed Doppler sonographic assessment of fetal heart rate. In this fetus the sample volume has been placed above the heart in a great artery (top) as well as within the heart just distal to an atrioventricular valve (bottom). The cardiac rhythm is regular, and the beat-to-beat interval is measured readily at 424 ms which relates to a heart rate of 142 beats per minute (white bars in the top frame).

Figure 32.2 have been established for atrioventricular contraction time intervals that range from 95 to 140 ms, depending on fetal heart rate and gestational age.13,14 A comparison of Doppler measurements of atrioventricular contraction time intervals with PR intervals from fetal ECGs in human fetuses also demonstrated reasonable agreement, with slightly shorter PR intervals measured in the fetal ECG.15 Analysis and differential diagnosis of fetal cardiac dysrhythmias can be facilitated further by the use of ladder diagrams (Figure 32.5). Ladder diagrams aid in defining the sequence of atrial and ventricular events and are used in electrophysiology for delineating arrhythmias. The horizontal lines indicate the division between the atria, the atrioventricular node, and the ventricles. With normal conduction, atrial (sinus node) activity precedes atrioventricular nodal activity, which precedes ventricular activity. After assessment of the atrial and ventricular activation sequence a simple step-by-step approach may be used to analyze abnormally slow fetal cardiac rhythm (Figure 32.6).

Pathophysiology The clinical finding of a bradydysrhythmia in the fetus either results from an abnormally slow atrial pacemaker

M-mode evaluation of normal fetal cardiac rhythm. Top: within the two-dimensional image the reference line is directed through the right atrium (RA) anteriorly and the left ventricle (LV) posteriorly. Bottom: the M-mode display shows the atrial wall movements (open arrows) that precede the ventricular contractions (curved bold arrows). There is a 1:1 relationship between atrial and ventricular activities. The large bars on the time scale show an interval of 1 s; the beat-to-beat interval measures 450 ms which relates to a heart rate of 133 beats per minute.

activity with a normal 1:1 atrioventricular conduction, or results from different forms of conduction block at the atrioventricular junction. Furthermore, bradydysrhythmia may be noted in a fetus with a structurally normal heart as an isolated rhythm disorder, or it may be associated with structural fetal heart disease. Possible causes of fetal bradydysrhythmia are summarized in Table 32.1. Since a fall in fetal heart rate leads to a notable reduction in fetal cardiac output,16,17 sustained fetal bradydysrhythmia may compromise the fetal circulation considerably. This is especially important in those fetuses affected by associated structural cardiac defects, which may limit their cardiac performance further. In addition, due to the reduced compliance of fetal ventricles, diastolic ventricular filling in the fetal heart depends to

Fetal bradydysrhythmia

451

Figure 32.3 Fetal cardiac rhythm analysis by M-mode echocardiography. In this example the reference line is passed through the right atrial wall (RAW) anteriorly and the pulmonary valve (PV) posteriorly (top). The M-mode display (bottom) shows the right atrial (RA) acitvation preceding the opening of the pulmonary valve (PV) by 84 ms (white bars). The regular 1:1 relationship between atrial and ventricular activities is demonstrated (arrows).

Figure 32.4

Clinical presentation

Atrial and ventricular activities are studied simultaneously by Doppler ultrasound. Top: in this four-chamber plane the enlarged sample volume is placed across the mitral valve (MV) area and the aortic root. The right (RV) and left ventricle (LV) are also seen. LA, left atrium. Bottom: pulsed Doppler interrogation demonstrates both the inflow signals into the left ventricle (LV) that are displayed above the baseline as well as the outflow signals into the aortic root (AO) that are seen below the baseline. There is a 1:1 relationship between atrial and ventricular activities. As normally seen in the fetal heart, the a-wave of the mitral inflow signal (a) is larger than the e-wave (e; about 0.4 vs 0.3 m/s). The onset of the a-wave indicates commencing atrial contraction, while onset of the aortic flow signal indicates commencing ventricular contraction; the delay between both flow events represents the atrioventricular conduction time (vertical lines).

Sinus bradycardia may be present if the fetal heart rate is regular and slow (< 100 beats per minute), and the atrial and ventricular activities are associated in a 1:1 fashion. Sinus bradycardia may be noted quite often as short episodes of fetal bradydysrhythmia during an ultrasonographic study, and has been related to increased vagal discharge in the fetus, possibly resulting from the pressure applied to the maternal abdomen by the transducer.20 Recovery of normal heart rate is usually noted after a few seconds, and such intermittent fetal sinus bradycardia does not require particular attention.

Sustained sinus bradycardia, however, needs to be evaluated further (Figure 32.6). Sustained sinus bradycardia may be found in the seriously sick fetus and will commonly be associated with other signs of impending fetal demise such as loss of fetal movements or fetal hydrops. Other causes of fetal sinus bradycardia, however, that do not reflect fetal distress have been reported. These include sinus node dysfunction,21,22 which cannot be

a larger degree on the atrial contraction than postnatally (Figure 32.4).17,18 As a consequence, the normal atrioventricular activation sequence is particularly important for the fetal cardiac performance.17 Therefore, sustained fetal bradydysrhythmia with a very slow ventricular rate of less than 50 beats per minute and concomitant dissociation of atrial and ventricular contractions due to complete atrioventricular block is the worst scenario of a bradydysrhythmia in the fetus which commonly is not well tolerated.19

452

Fetal Cardiology

A SV

Atrium

AW

Atrial contraction (Activation) A-V Conduction Ventricular contraction (Activation)

Figure 32.5 The use of ladder diagrams in fetal cardiac rhythm analysis. Top: schematic drawing showing the atrial and ventricular activities as assessed by M-mode echocardiography (see Figure 32.3). For better clarity a ladder diagram has been added that indicates the observed mechanical events of an atrial contraction (open arrow) and subsequent opening of a semilunar valve (SV, curved bold arrow) which represents ventricular contraction (solid lines). The delay between both observed mechanical events represents the atrioventricular (A-V) conduction time (dashed line). Bottom: in this M-mode echocardiographic recording right atrial (RA) contraction (white arrows) is seen prior to the opening of the aortic (AO) valve (curved black arrows). After two normally conducted atrial beats there is a premature atrial contraction which is not followed by ventricular activation (blocked premature atrial contraction). In the ladder diagram below the block is indicated at the A-V conduction level.

diagnosed prenatally with certainty and requires postnatal work-up, as well as sinus bradycardia of moderate degree that has been noted in the presence of maternal hypothermia.23,24 The latter entity has clinical implications, since maternal hypothermia may result from the treatment of premature labor with magnesium sulfate.25

Sustained moderate fetal sinus bradycardia may also be noted in long-QT syndrome.26,27 Long-QT syndrome is a heterogeneous genetic disorder caused by mutations in several genes that encode different ion channel proteins, most of them potassium channel proteins.28 Patients with long-QT syndrome are at high risk to develop malignant ventricular tachyarrhythmias, which have been documented in prenatal life as well.29 In affected fetuses the heart rate may range from 70 to 100 beats per minute (Figure 32.7), and fetal bradycardia may be attenuated further by the presence of a functional second-degree atrioventricular block that results from an extremely delayed ventricular repolarization.30–32 Since long-QT syndrome is a genetic disorder, it should particularly be considered as a cause for fetal sinus bradycardia if a positive family history is present. The prenatal identification of long-QT syndrome has important implications for the perinatal management of these fetuses since there is a substantial risk for perinatal and infant mortality.27,30 Moderate bradycardia is also found in the fetus with frequent premature atrial contractions which are blocked at the level of the atrioventricular node.33 If every other beat – in a bigeminal fashion – is a blocked premature atrial contraction, the ventricular rate is regular, but slow at about 60–80 beats per minute (Figure 32.8). If every third beat is a blocked premature atrial contraction the ventricular rate is irregular (group beating), and not as slow as with bigeminal premature atrial contractions; heart rates commonly range from 80 to 110 beats per minute (Figure 32.9). By identifying the premature occurrence of the blocked atrial contractions, M-mode echocardiography allows a clear distinction between 2:1 atrial ectopy and second-degree atrioventricular block (Figure 32.5). Blocked premature atrial contractions are usually well tolerated by the fetus and rarely require treatment. Correct identification of bigeminal blocked premature atrial contractions is nevertheless important, because it may be mistaken for bradycardia due to fetal distress and may lead to false obstetric management decisions. Subsequent observation of a fetus with frequent blocked premature atrial contractions is reasonable, since up to 5% of these fetuses have been reported to develop tachydysrhythmia while gestation continues.33 Another very rare cause of fetal bradydysrhythmia is familial idiopathic atrial fibrillation with slow ventricular response. This entity is transmitted as an autosomal dominant trait with male predominance, and has been observed in childhood and adolescence34 as well as in prenatal life.35 Since this disorder carries a good prognosis, the main objective in prenatal diagnosis is exclusion of other causes for fetal bradydysrhythmia and follow-up. Group beating may also occur due to bradycardia with junctional or ventricular escape beats or second-degree atrioventricular block. Second-degree atrioventricular block is not diagnosed readily in the prenatal evaluation of a bradydysrhythmia. For correct identification, the

Fetal bradydysrhythmia

453

Fetal Heart Rateby Doppler-Ultrasound HR < 100 bpm :

regular

irregular

check atrial and ventricular activity

atrial & ventricular activity associated, slow atrial and slow ventricular rate

atrial & ventricular activity associated, every 2nd atrial contraction premature, blocked at AV-node

sinus bradycardia (LQTS?)

atrial & ventricular activity dissociated, normal atrial and slow ventricular rate

bigeminal PACs

complete AV-block

atrial & ventricular activity associated, every 3rd atrial contraction premature, blocked at AV-node

allorhythmic (2:1) - PACs

atrioventricular delay increasing, atrial contraction repeatedly blocked at AV-node

atrial & ventricular activity dissociated, frequent premature ventricular contractions

2.° AV block (Wenckebach)

complete AV block & PVCs

Figure 32.6 Flow chart for fetal cardiac rhythm analysis in the presence of a slow heart rate (HR). bpm, beats per minute; AV, atrioventricular; LQTS, long-QT syndrome; PACs, premature atrial contractions; PVCs, premature ventricular contractions.

Table 32.1 causes

Fetal bradydysrhythmia – possible

Sinus bradycardia short episodes: vagal tone Sustained sinus bradycardia sinus node dysfunction maternal hypothermia long-QT syndrome

plete atrioventricular block.36 Isorhythmic atrioventricular dissociation that mimics second-degree atrioventricular block was frequently identified by fetal magnetocardiography in fetuses with complete heart block when atrial and ventricular rates were in a 2:1 ratio.3 The other type of second-degree atrioventricular block (Wenckebach or Mobitz type I) is characterized by a regular atrial rate with group beating, due to the atrioventricular block that leads to the exclusion of every third, fourth, or further ventricular activation.

Frequently occurring blocked premature atrial contractions Familial idiopathic atrial fibrillation with slow ventricular response

Complete heart block

Second-degree atrioventricular block

The slowest fetal heart rate will be noted if complete atrioventricular block – or complete heart block – is present. Congenital complete heart block was diagnosed rarely in the fetus before ultrasound techniques became available; it is characterized by complete dissociation of atrial and ventricular contractions with normal atrial, but slow ventricular rates (Figure 32.11). Congenital complete heart block is an uncommon finding, reportedly occuring once in about 20 000 newborns.37 The incidence may well be higher in prenatal life, since some fetuses with complete heart block will not survive to term.19 With regard to morphology, complete heart block may result from a lack of fusion between nodal tissue and the His bundle, which initially develop separately, or it may result from

Third-degree (complete) atrioventricular block

simultaneous display of atrial and ventricular activity should be sought by M-mode echocardiography. In the presence of a regular atrial rate and 2:1 conduction to the ventricles this may be second-degree (Mobitz type II) block, but distinction from complete atrioventricular block with a ventricular escape rate that is just half the atrial rate is not easy (Figure 32.10), or at times even impossible. Apparent atrioventricular synchronization may occur for longer periods of time in patients with com-

454

Fetal Cardiology

Figure 32.7 M-mode echocardiography in a fetus with moderate bradycardia. Top: the reference line passes through the right atrium (RA) and the left ventricle (LV). Bottom: regular right atrial contractions (black arrows) are seen to precede regular ventricular contractions (white arrowheads); the fetal heart rate is 100 beats per minute, and the conduction time is normal (110 ms). Thus, this fetus has sinus bradycardia; the mother had long-QT syndrome which made the diagnosis likely in the fetus as well. After birth a very long QTc interval of 580 ms was found in the newborn infant.

secondary interruption of the atrioventricular conduction axis; evidence has been presented for both views.38–40 From a more clinical point of view, congenital complete heart block may be found as an isolated lesion in a fetus with a structurally normal heart, or it may be associated with structural heart disease. Both entities have a different etiology and natural history that may affect the management approach and outcome. Therefore, the fetal cardiac anatomy and function should be assessed thoroughly once the diagnosis of complete heart block is established in a fetus.

Figure 32.8 Bigeminal blocked premature atrial contractions. Top: the M-mode display shows regular atrial activity (white arrows) and ventricular activity (open arrows). Shortly after the conducted atrial beat there is a premature atrial contraction (arrowhead) that is blocked, thus leading to a slow ventricular rate of 75 beats per minute. Bottom: in another fetus a similar finding is noted on pulsed Doppler study; the beat-to-beat interval is 720 ms which relates to a fetal heart rate of 83 beats per minute.

Isolated complete heart block Isolated complete heart block is currently considered as an immunological disorder in the vast majority of cases. Mothers of affected fetuses often suffer from connective tissue disease (commonly Sjögren’s syndrome or lupus erythematosus), but other women may be clinically asymptomatic. Almost all of them, however, test positive for autoantibodies to the SSA/Ro or SSB/La ribonucleoproteins.41,42 These autoantibodies cross the placental barrier, and have been found in newborns with complete heart block who are born to mothers testing positive for anti-SSA/Ro or anti-SSB/La antibodies.41,42 Anti-SSA/Ro or anti-SSB/La antibodies have been shown to react with

fetal cardiac tissue, thus promoting an autoimmune inflammation and tissue injury.43 Other factors, which are partly unknown as yet, may also be involved in the disease process and subsequent destruction of the atrioventricular node. This relates to the impact of the different subclasses of Ro and La antibodies, as well as to other antibodies such as ERV-3 which are possibly involved in the disease process.44 Complete heart block has been reported in only one of human leukocyte antigen (HLA)-identical twins, both of whom had high anti-SSA/Ro antibody titers.45,46 The risk for a woman with known anti-SSA/Ro or anti-SSB/La antibodies to have a

Fetal bradydysrhythmia

455

Figure 32.9 Doppler sonographic display of group beating in a fetus with a moderate bradycardia of 96 beats per minute. Descending aortic flow signals show two regular beats with an interval of about 470 ms followed by a longer pause of 770 ms. Most likely, this is 2:1 atrial ectopic beats blocked at the atrioventricular node, but second-degree atrioventricular block (Wenckebach) would also be possible and can be excluded by M-mode echocardiography, which showed the premature atrial contractions in this case (compare Figure 32.5).

child with complete heart block is only about 2–5%, and even after having had one child with complete heart block the recurrence rate is just 15–20% in subsequent pregnancies.46,47 Commonly, isolated complete heart block develops between 18 and 24 weeks of gestation, and progression from second-degree to complete heart block has been observed in some cases.19 The pre- and perinatal course as well as outcome at the end of the neonatal period largely depends on the presence of fetal hydrops (Table 32.2). In several studies about 75% of fetuses with isolated complete heart block survived the neonatal period, and almost all of them did not exhibit fetal hydrops.19,48–52 A slow ventricular escape rate of less than 55 beats per minute appears to be another poor prognostic factor in isolated complete heart block.19,50,51 In addition, poor ventricular function or endocardial fibroelastosis may contribute to a grim prognosis for the fetus with isolated complete heart block.53

Complete heart block and structural heart disease Complete heart block associated with structural heart disease is mainly seen in fetuses with complex cardiac lesions. In several series a large majority of affected fetuses had either left atrial isomerism or discordant atrioventricular connection. Hearts with left atrial isomerism have bilateral left atrial morphology and frequently lack a normal atrioventricular node which is a right atrial structure, while in discordant atrioventricular connection

Figure 32.10 M-mode and pulsed Doppler evaluation of a fetal bradycardia noted at about 65 beats per minute. Top: the M-mode display shows regular atrial activity at a normal rate (arrowheads), but much slower regular ventricular activity (open arrows). There seems to be a 2:1 relationship between the atrial and ventricular contractions, possibly due to second-degree atrioventricular block (Mobitz I). Bottom: in the same fetus pulsed Doppler interrogation is performed with the sample volume placed in the descending aorta and the posterior left atrial wall. Left atrial wall movements (closed triangles) indicate atrial activity, aortic flow signals (open triangles) ventricular activity. Although roughly the same 2:1 pattern seems to be present, the increasing time delay between ventricular and the preceding atrial activity indicates dissociation of atrial and ventricular activities and favors the diagnosis of complete heart block.

the inversion of the ventricles often leads to disruption of the atrioventricular conduction axis. Left atrial isomerism is almost always associated with major structural heart defects that may contribute to the poor prognosis of these

456

Fetal Cardiology

period, according to the combined data of five larger series (Table 32.2); about half (10 of 19) of the survivors had atrioventricular discordance, which seems to be tolerated better regarding impending fetal demise than left atrial isomerism. The dismal prognosis for a fetus with complete block and structural heart disease is aggravated further if associated hydrops is found. None of the hydropic fetuses with complete block and structural heart disease survived in the combined experience (Table 32.2). A ventricular escape rate of less than 55 beats per minute was a poor prognostic feature, too.19,51

Prevention and prenatal treatment of complete heart block

Figure 32.11 Complete heart block in a fetus presenting with a regular, but slow heart rate. Both the simultaneous M-mode display of atrial and ventricular contractions (top: white arrows, black arrowheads) and of atrial contractions and semilunar valve movements (bottom: white arrowheads, black arrows) indicate dissociation of atrial and ventricular activities with a normal atrial rate (144 beats per minute), but a slow ventricular rate (68 beats per minute).

fetuses. In addition to such structural heart defects, left atrial isomerism and fetal complete heart block have recently been reported in association with ventricular noncompaction; this form of cardiomyopathy may worsen cardiac function further, since none of the affected fetuses survived the neonatal period.54 The fetus with complete heart block and associated structural cardiac defects has a poorer prognosis than that of the fetus with isolated complete heart block. While associated structural heart disease is seen in about 30% of newborns with congenital complete heart block,37 it occurs about twice as often in fetuses with complete heart block.19,48–52 These data suggest that there is a considerable rate of prenatal loss in fetuses with complete block and structural heart disease. In fact, only 19 out of 123 fetuses (15%) with complete block and structural heart disease were alive at the end of the neonatal

Since the majority of fetuses with isolated complete heart block will survive, and almost all with structural heart disease will succumb, one might question the necessity of prenatal treatment attempts in general. There are, however, those fetuses with early signs of cardiac failure or particularly slow ventricular escape rhythm whose condition might be improved by prenatal treatment. More recently, the search for preventive measures was stimulated by the idea that first-degree atrioventricular block may precede the development of isolated complete heart block; early recognition of first-degree block could then aid in attempts to prevent further destruction of the atrioventricular node by steroid treatment.55,56 In a recently published prospective trial, the fetuses of women with known anti-SSA/Ro antibodies were studied serially by Doppler ultrasound and mechanical atrioventricular conduction times were measured. Prolongation of this time interval was an uncommon finding; particularly, it was not seen in those three fetuses of 98 pregnancies in whom complete heart block developed eventually.57 In two other fetuses, first-degree atrioventricular block was identified and subsequently treated with dexamethasone, which led to normalization of the PR interval, but a third had proven first-degree block in postnatal ECGs while his fetal studies were within normal limits.57 These experiences demonstrate the difficulties in identifying the fetus at risk to develop complete heart block. Once complete heart block has developed, fetal drug therapy has mainly been directed towards three different features of the disease entity: cardiac failure, autoimmune inflammation, and slow heart rate. Cardiac failure as seen in the hydropic fetus with complete heart block has been treated with digoxin and furosemide administered either transplacentally58 or directly into the umbilical vein.59 Such treatment has been successful in resolving the fluid accumulations in some cases,58,59 but failures have also been reported, especially in the fetuses with heart block and associated structural heart disease.49 In a different treatment attempt, fetal effusions and the autoimmune inflammatory process that may eventually

Fetal bradydysrhythmia

Table 32.2

457

Reported series on presentation and outcome of fetal complete heart block (CHB) Of these, with:

Reference

Fetal CHB (n) CHB and structural heart disease

Machado et al48

Gembruch et al

19

Schmidt et al

49

37

21

55

21 (57%)

Left atrial isomerism 21 (100%)

Atrioventricular Isolated CHB discordance —

with hydrops 11 (52%)

with hydrops 4 (25%)

alive (ENP) 3 (14%)

alive (ENP) 12 (75%)

17 (81%)

6 (35%)

4 (24%)

with hydrops 1 (25%)

alive (ENP) 4 (24%)

alive (ENP) 3 (75%)

29 (53%)

17 (59%)

7 (24%)

—

26 (47%) Jwith hydrops 4 (15%)

alive (ENP) 4 (14%) 36

4 (19%)

with hydrops 10 (59%)

with hydrops 18 (62%) Groves et al50

16 (43%)

alive (ENP) 22 (85%) —

—

36 (100%) with hydrops 12 (33%) alive (ENP) 25 (69%)

Jaeggi et al51

52

Berg et al

59

60

24 (41%)

18 (75%)

3 (13%)

35 (59%)

with hydrops 9 (38%)

with hydrops 3 (9%)

alive (ENP) 4 (17%)

alive (ENP) 25 (71%)

32 (53%) with hydrops 23 (72%) alive (ENP) 4 (13%)

31 (97%)

1 (3%)

28 (47%) with hydrops 5 (18%) alive (ENP) 19 (68%)

ENP, end of neonatal period (1 month).

lead to disruption of the fetal conduction system have been targeted with different glucocorticoids. Several case reports have noted successful resolution of fetal effusions after transplacental administration of prednisone and betamethasone, but persistence of heart block.60,61 Other authors used dexamethasone and found regression of fetal hydrops and improvement in the degree of heart block.62 Even prophylactic glucocorticoid treatment during midgestation, with63 or without additional plasmapheresis,64 has been proposed in SSA/Ro- or SSB/La-positive women, but the reported favorable outcome might just have been due to random effects, since developing heart block was never observed in these studies. Another study suggested that fluorinated steroids (dexamethasone or betamethasone), which cross the placenta better than prednisone, should be considered for fetuses with incomplete heart block or commencing hydrops.65 Complete heart block, however, could not be reverted in this study by steroid treatment, and the outcome was similar in treated and untreated fetuses with respect to prenatal deaths, final degree of heart block, or requirement of postnatal pacing.65 More recently, the use of a standardized treatment approach at one institution reportedly led to an improved outcome in isolated fetal complete heart block,66 but other

authors did not confirm these observations.67 Prospective studies with larger numbers of patients are still required to prove the possible benefit of transplacental steroid treatment for isolated complete heart block. In addition, the possible adverse effects both in the mother and in the fetus or newborn must be considered with all steroid treatment. Because several studies have shown that ventricular rates higher than 50–55 beats per minute are related to a better outcome,19,50,51 the administration of sympathomimetic drugs has also been advocated for the treatment of fetal complete heart block. Both terbutaline68 and salbutamol69 appear to cross the placental barrier well, and may increase the fetal ventricular rate by 15–25%19,69 Maternal infusion of isoprenaline, however, did not change the fetal heart rate even at high doses, and is probably less well transferred across the placenta.69 Transplacental administration of salbutamol may be considered in the previable fetus with isolated complete heart block, slow ventricular rate, and commencing hydrops. Careful attention must be paid, however, to exclude long-QT syndrome with 2:1 atrioventricular block in the fetus prior to the introduction of transplacental sympathomimetic therapy, which may worsen the situation of such a patient.32

458

Fetal Cardiology

Theoretically, prenatal ventricular pacing would be the most logical form of treatment for the fetus with complete heart block. Experimental data from fetal lamb studies have shown that pacing is feasible in fetal lambs using either a transvenous70 or an epicardial approach;71 chronic ventricular pacing was well tolerated and did not affect fetal ventricular performance.71 Substantial increases in cardiac output have also been demonstrated with fetal ventricular pacing in experimentally induced complete heart block.17 The attempted use of ventricular pacing in human fetuses, however, has failed so far, mainly due to problems in obtaining a stable position of the pacing lead within the fetal heart.72,73 Premature delivery at 24–30 weeks of gestation and subsequent staged pacing, using an external pacemaker after birth and switching to internal pacemaker once the infants weighed > 1.5–2 kg, has been reported,74,75 and may be a better alternative for treating the hydropic fetus with isolated complete heart block. Obviously, there is no clear answer to the question of how to treat a fetus with isolated complete heart block and beginning hydrops. Prospective studies are required to answer this question.

Conclusion Fetal bradydysrhythmia may be diagnosed as a benign condition that has little or no consequence for treatment or further management, such as short episodes of sinus bradycardia or bigeminal blocked premature atrial contractions, but it may also represent a less benign condition, such as fetal long-QT syndrome or high-degree atrioventricular block, that requires further attention or treatment. Of all fetal bradydysrhythmias, complete heart block may compromise the fetus to the largest extent. The prognosis of a fetus with complete heart block depends on the presence of associated structural heart disease or hydrops, and in both circumstances the neonatal outcome is poor. In isolated complete heart block with proven maternal autoantibodies, treatment with fluorinated glucocorticoids that cross the placenta has been used with variable success. Treatment with sympathomimetic drugs has been employed in hydropic fetuses with slow heart rates, also with only variable success. Experimental treatment with prenatal pacing has also been applied, and may prove to be the best therapy in isolated fetal complete heart block in the future.

References 1. Taylor MJO, Smith MJ, Thomas M et al. Non-invasive fetal electrocardiography in singleton and multiple pregnancies. Br J Obstet Gynaecol 2003; 110: 668–78.

2. Kaehler C, Schleussner E, Grimm B et al. Fetal magnetocardiography: development of the fetal cardiac time intervals. Prenat Diagn 2002; 22: 408–14. 3. Zhao H, Cuneo BF, Strasburger JE et al. Electrophysiological characteristics of fetal atrioventricular block. J Am Coll Cardiol 2008; 51: 77–84. 4. Allan LD, Anderson RH, Sullivan ID et al. Evaluation of fetal arrhythmias by echocardiography. Br Heart J 1983; 50: 240–5. 5. Kleinman CS, Donnerstein RL, Jaffe CC et al. Fetal echocardiography: a tool for evaluation of in utero cardiac arrhythmias and monitoring of in utero therapy: analysis of 71 patients. Am J Cardiol 1983; 51: 237–43. 6. Silverman NH, Enderlein MA, Stanger P et al. Recognition of fetal arrhythmias by echocardiography. J Clin Ultrasound 1985; 13: 255–63. 7. Strasburger JF, Huhta JC, Carpenter RJ et al. Doppler echocardiography in the diagnosis and management of persistent fetal arrhythmias. J Am Coll Cardiol 1986; 7: 1386–91. 8. Chan FY, Woo SK, Ghosh A et al. Prenatal diagnosis of congenital fetal arrhythmias by simultaneous pulsed Doppler velocimetry of the fetal abdominal aorta and inferior vena cava. Obstet Gynecol 1990; 76: 200–4. 9. Fouron JC, Proulx F, Miro J et al. Doppler and M-mode ultrasonography to time fetal atrial and ventricular contractions. Obstet Gynecol 2000; 96: 732–6. 10. Carvalho JS, Prefumo F, Ciardelli V et al. Evaluation of fetal arrhythmias from simultaneous pulsed wave Dopper in pulmonary artery and vein. Heart 2007; 93: 1448–53. 11. Glickstein J, Buyon J, Kim M et al. The fetal Doppler mechanical PR interval: a validation study. Fetal Diagn Ther 2004; 19: 31–4. 12. Dancea A, Fouron JC, Miro J et al. Correlation between electrocardiographic and ultrasonographic time-interval measurements in fetal lamb heart. Pediatr Res 2000; 47: 324–8. 13. Glickstein J, Buyon J, Friedman D. Pulsed Doppler echocardiographic assessment of the fetal PR intrval. Am J Cardiol 2000; 86: 236–9. 14. Andelfinger G, Fouron JC, Sonesson SE et al. Reference values for time intervals between atrial and ventricular contractions of the fetal heart measured by two Doppler techniques. Am J Cardiol 2001; 88: 1433–6. 15. Pasquini L, Seale AN, Belmar C et al. PR interval: a comparison of electrical and mechanical methods in the fetus. Early Hum Dev 2007; 83: 231–7. 16. Rudolph AM, Heyman MA. Cardiac output in the fetal lamb: the effects of spontaneous and induced changes of heart rate on right and left ventricular output. Am J Obstet Gynecol 1976; 124: 183–92. 17. Crombleholme TM, Longaker MT, Langer JC et al. Complete heart block and AV-sequential pacing in fetal lambs: the atrial contribution to combined ventricular output in the fetus. Surg Forum 1989; 40: 268–70. 18. Reed KL, Sahn DJ, Scagnelli S et al. Doppler echocardiographic studies of diastolic function in the human fetal heart: changes during gestation. J Am Coll Cardiol 1986; 8: 391–5. 19. Schmidt KG, Ulmer HE, Silverman NH et al. Perinatal outcome of fetal complete atrioventricular block:

Fetal bradydysrhythmia

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38. 39.

a multicenter experience. J Am Coll Cardiol 1991; 17: 1360–6. Mendoza GJ, Almeida O, Steinfeld L. Intermittent fetal bradycardia induced by midpregnancy fetal ultrasonographic study. Am J Obstet Gynecol 1989; 160: 1038–40. Tyrell S, Gibbs JL. Sinuatrial disease causing persistent fetal bradycardia: intrauterine detection and management. Int J Cardiol 1988; 19: 382–4. Degani S, Abinader EG, Shapiro I, Lewinsky R, Sharf M. Fetal arrhythmia associated with sinus node dysfunction. J Perinat Med 1988; 16: 153–5. Hankins G, Leicht T, Van Hook J. Prolonged fetal bradycardia secondary to maternal hypothermia in response to urosepsis. Am J Perinatol 1997; 14: 217–19. Aboud E, Neales K. The effect of maternal hypothermia on the fetal heart rate. Int J Gynecol Obstet 1999; 66: 163–4. Cardosi RJ, Chez RA. Magnesium sulfate, maternal hypothermia, and fetal bradycardia with loss of heart rate variability. Obstet Gynecol 1998; 92: 691–3. Donofrio MT, Gullquist SD, O’Connell NG et al. Fetal presentation of congenital long QT syndrome. Pediatr Cardiol 1999; 20: 441–4. Hofbeck M, Ulmer H, Beinder E et al. Prenatal findings in patients with prolonged QT interval in the neonatal period. Heart 1997; 77: 198–204. Priori SG, Barhanin J, Hauer RNW et al. Genetic and molecular basis of cardiac arrhythmias: impact on clinical management. Circulation 1999; 99: 518–28. Yamada M, Nakazawa M, Momma K. Fetal ventricular tachycardia in long QT syndrome. Cardiol Young 1998; 8: 119–22. Presbitero P, Mangiardi L, Antolini R. Congenital long QT syndrome inducing 2:1 atrioventricular block: early detection in fetal life. Int J Cardiol 1989; 24: 109–12. Rosenbaum MB, Acunzo RS. Pseudo 2:1 atrioventricular block and T wave alternans in the long QT syndromes. J Am Coll Cardiol 1991; 18: 1363–6. Tozzi RJ, Kurer CC, Kipel G. Profound fetal bradycardia: a case report highlighting treatment options. J Perinatol 2003; 23: 509–12. Simpson JM, Yates RW, Sharland GK. Irregular heart rate in the fetus - not always benign. Cardiol Young 1996; 6: 28–31. Bertram H, Paul T, Beyer F et al. Familial idiopathic atrial fibrillation with bradyarrhythmia. Eur J Pediatr 1996; 155: 7–10. Aburawi E, Thomson J, Blackburn M. Familial idiopathic atrial fibrillation with fetal bradyarrhythmia. Acta Paediatr 2006; 95: 1700–2. Christ JE, Hoff HE. An analysis of atrioventricular synchronization in complete AV block. J Electrocardiol 1973; 6: 53–62. Michaelsson M, Engle MA. Congenital complete heart block: an international study of the natural history. Cardiovasc Clin 1972; 4: 85–101. James TN. Cardiac conduction system: fetal and postnatal development. Am J Cardiol 1970; 25: 213–26. Anderson RH, Wenick ACG, Losekoot TG et al. Congenitally complete heart block - developmental aspects. Circulation 1977; 56: 90–101.

459

40. James TN, Martin ES, Willis PW et al. Apoptosis as a possible cause of gradual development of complete heart block and fatal arrhythmias associated with absence of the AV node, sinus node, and internodal pathways. Circulation 1996; 93: 1424–38. 41. Scott JS, Maddison PJ, Taylor PV et al. Connective-tissue disease, antibodies to ribonucleoprotein, and congenital heart block. N Engl J Med 1983; 309: 209–12. 42. Buyon JP, Ben-Chetrit E, Karp S et al. Acquired congenital heart block. Pattern of maternal antibody response to biochemically defined antigens of the SSA/Ro-SSB/La system in neonatal lupus. J Clin Invest 1989; 84: 627–34. 43. Litsey SE, Noonan JA, O’Connor WN et al. Maternal connective tissue disease and congenital heart block. Demonstration of immunoglobulin in cardiac tissue. N Engl J Med 1985; 312: 98–100. 44. Horsfall AC, Neu E, Forrest G et al. Maternal autoantibodies and congenital heart block: clues from two consecutive pregnancies, one in which there was congenital complete heart block and one in which the fetus was healthy. Arthritis Rheum 1998; 41: 2079–80. 45. Kaaja R, Julkunen H, Ämmälä P et al. Congenital heart block in one of the HLA identical twins. Eur J Obstet Gynecol Reprod Biol 1993; 51: 78–80. 46. Solomon DG, Rupel A, Buyon JP. Birth order, gender and recurrence rate in autoantibody-associated congenital heart block: implications for pathogenesis and family counselling. Lupus 2003; 12: 646–7. 47. Buyon JP, Hiebert R, Copel J et al. Autoimmune-associated congenital heart block: demographics, mortality, morbidity and recurrence rates obtained from a national neonatal lupus registry. J Am Coll Cardiol 1998; 31: 1658–66. 48. Machado MVL, Tynan MJ, Curry PVL et al. Fetal complete heart block. Br Heart J 1988; 60: 512–15. 49. Gembruch U, Hansmann M, Redel DA et al. Fetal complete heart block: antenatal diagnosis, significance and management. Eur J Obstet Gynecol Reprod Biol 1989; 31: 9–22. 50. Groves AMM, Allan LD, Rosenthal E. Outcome of isolated congenital complete heart block diagnosed in utero. Heart 1996; 75: 190–4. 51. Jaeggi ET, Hornberger LK, Smallhorn JF et al. Prenatal diagnosis of complete atrioventricular block associated with structural heart disease: combined experience of two tertiary care centers and review of the literature. Ultrasound Obstet Gynecol 2005; 26: 16–21. 52. Berg C, Geipel A, Kohl T et al. Atrioventricular block detected in fetal life: associated anomalies and potential prognostic markers. Ultrasound Obstet Gynecol 2005; 26: 4–15. 53. Nield LE, Silverman ED, Taylor GP et al. Maternal anti-Ro and anti-La antibody-associated endocardial fibroelastosis. Circulation 2002; 105: 843–8. 54. Friedberg MK, Ursell PC, Silverman NH. Isomerism of the left atrial appendage associated with ventricular noncompaction. Am J Cardiol 2005; 96: 985–90. 55. Sonesson SE, Salomonsson S, Jacobsson LA et al. Signs of first-degree heart block occur in one-third of fetuses of pregnant women with anti-SSA/Ro 52-kd antibodies. Arthritis Rheum 2004; 50: 1253–61. 56. Gardiner HM, Belmar C, Pasquini L et al. Fetal ECG: a novel predictor of atrioventricular block in anti-Ro positive pregnancies. Heart 2007; 93: 1454–60.

460

Fetal Cardiology

57. Friedman DM, Kim MY, Copel JA et al. Utility of cardiac monitoring in fetuses at risk for congenital heart block. The PR Interval and Dexamethasone Evaluation (PRIDE) prospective study. Circulation 2008; 117: 485–93. 58. Harris JP, Alexson CG, Manning JA et al. Medical therapy for the hydropic fetus with congenital complete atrioventricular block. Am J Perinatol 1993; 10: 217–19. 59. Anandakumar C, Biswas A, Chew SLS et al. Direct fetal therapy for hydrops secondary to congenital atrioventricular heart block. Obstet Gynecol 1996; 87: 835–7. 60. Bierman FZ, Baxi L, Jaffe I. Fetal hydrops and congenital complete heart block: response to maternal steroid therapy. J Pediatr 1988; 112: 646–8. 61. Watson WJ, Katz VL. Steroid therapy for hydrops associated with antibody-mediated congenital heart block. Am J Obstet Gynecol 1991; 165: 553–4. 62. Copel JA, Buyon JP, Kleinman CS. Successful in utero therapy of fetal heart block. Am J Obstet Gynecol 1995; 173: 1384–90. 63. Barclay CS, French MAH, Ross LD et al. Successful pregnancy following steroid therapy and plasma exchange in a woman with anti-Ro(SS-A) antibodies. Br J Obstet Gynaecol 1987; 94: 369–71. 64. Rosenthal D, Druzin M, Chin C et al. A new therapeutic approach to the fetus with congenital complete heart block: preemptive, targeted therapy with dexamethasone. Obstet Gynecol 1998; 92: 689–91. 65. Saleeb S, Copel J, Friedman D et al. Comparison of treatment with fluorinated glucocorticoids to the natural history of autoantibody-associated congenital heart block. Arthritis Rheum 1999; 42: 2335–45.

66. Jaeggi ET, Fouron JC, Silverman ED et al. Transplacental fetal treatment improves the outcome of prenatally diagnosed complete atrioventricular block without structural heart disease. Circulation 2004; 110: 1542–8. 67. Rosenthal E, Gordon PA, Simpson JM et al. Letter regarding article by Jaeggi et al. Circulation 2005; 111: e287–8. 68. Robinson BV, Ettedgui JA, Sherman FS. Use of terbutaline in the treatment of complete heart block in the fetus. Cardiol Young 2001; 11: 683–6. 69. Groves AM, Allan LD, Rosenthal E. Therapeutic trial of sympathomimetics in three cases of complete heart block in the fetus. Circulation 1995; 92: 3394–6. 70. Kikuchi Y, Shiraishi H, Igarashi H et al. Cardiac pacing in fetal lambs: intrauterine transvenous cardiac pacing for fetal complete heart block. PACE 1995; 18: 417–23. 71. Liddicoat JR, Klein JR, Reddy M et al. Hemodynamic effects of chronic prenatal ventricular pacing for the treatment of complete atrioventricular block. Circulation 1997; 96: 1025–30. 72. Carpenter RJ, Strasburger JF, Garson A et al. Fetal ventricular pacing for hydrops secondary to complete atrioventricular block. J Am Coll Cardiol 1986; 8: 1434–6. 73. Walkinshaw SA, Welch CR, McCormack J et al. In utero pacing for fetal congenital heart block. Fetal Diagn Ther 1994; 9: 183–5. 74. Weindling SN, Saul JP, Triedman JK et al. Staged pacing therapy for congenital complete heart block in premature infants. Am J Cardiol 1994; 74: 412–13. 75. Deloof E, Devlieger H, van Hoestenberghe R et al. Management with a staged approach of the premature hydropic fetus due to complete heart block. Eur J Pediatr 1997; 156: 521–3.

33 Fetal tachyarrhythmia Ulrich Gembruch Fetal tachyarrhythmias, defined as fetal heart rates about 180–200 beats/minute, are generally subdivided into sinus tachycardia, supraventricular tachyarrhythmia, including supraventricular tachycardia (SVT) and atrial flutter, and ventricular tachyarrhythmia. On the other hand, the most common form of fetal and neonatal SVT, the atrioventricular reentry tachycardia via an accessory pathway, involves the atrium, atrioventricular (AV) node, much of the ventricles, and an accessory pathway as reentry circuit, and therefore is a ‘whole heart’ tachycardia.1 According to the three electrophysiological levels of the heart, it seems to be more accurate to divide the tachyarrhythmias into atrial tachycardia (atrial flutter, atrial ectopic tachycardia), conduction system tachycardia (atrioventricular reentry tachycardia via an apparent or ‘concealed’ accessory pathway, permanent junctional reciprocant tachycardia, and atrioventricular nodal reentry tachycardia), and ventricular tachycardia.1 In fetuses, SVT is more frequent than atrial flutter (70–75% versus 25–30%), whereas ventricular tachycardia is very rare.2 Sustained fetal tachyarrhythmia (SVT with 1:1 atrioventricular conduction, atrial flutter, and ventricular tachycardia) may cause congestive heart failure, leading to elevated right atrial and systemic venous pressure, and may be followed by non-immune hydrops, placental edema, and polyhydramnios. In addition, associated maternal complications may be complaints due to severe polyhydramnios, preterm contractions and labor, premature rupture of the membranes, and the so-called mirror syndrome or Ballantyne syndrome. This results in a maternal hyperdynamic and hypertensive state with symptoms of preeclampsia, which is sometimes observed associated with fetuses with placental edema and hydrops of various etiologies.3–5 If remission of hydrops can be achieved, preeclamptic symptoms of the mother may disappear in the ongoing pregnancy.6–8

In utero diagnosis of fetal tachyarrhythmia The majority of fetal tachyarrhythmias are detected during routine obstetric examination in the second and third

trimesters of pregnancy. Monitoring of the fetal heart rate by ultrasound, continuous wave Doppler (Doptone), or cardiotocography reveals fetal arrhythmia requiring a detailed echocardiographic examination. Polyhydramnios and fetal hydrops may also lead to the detection of an underlying tachyarrhythmia. If an intensive noninvasive and invasive search for an underlying disease is unsuccessful, paroxysmal supraventricular tachyarrhythmia should always be taken into consideration, particularly if signs of congestive heart failure such as cardiomegaly, atrioventricular valvular regurgitation, and/or increased pulsatility of venous flow velocity waveforms is present. In this situation, repeated sonographic heart rate monitoring, or long-term cardiotocography carried out several times per day, can diagnose or exclude paroxysmal supraventricular tachyarrhythmia as the cause of the hydrops. Differential diagnosis of fetal arrhythmia and also tachyarrhythmias is performed using echocardiographic techniques with a high time resolution. Tracings of the wall and valve movements and blood flow velocity waveforms of the heart, the arteries, and the veins obtained by M-mode, pulsed wave Doppler, and/or color Doppler M-mode allow indirect conclusions to be drawn from these recordings of the mechanical and flow activities against time to the electrical events (Figures 33.1 and 33.2). Thus, correct diagnosis of the type of fetal arrhythmia is possible in the vast majority of cases. For cases of fetal arrhythmia, the diagnostic approach is elaborated further in Chapter 31. New non-invasive methods for diagnosis of fetal arrhythmias are tissue velocity imaging (by pulsed wave Doppler, color-coded M-mode, or color Doppler imaging) and magnetocardiography. Tissue Doppler techniques allow simultaneous sampling of atrial and ventricular wall velocities to yield precise temporal analysis of atrial and ventricular events, similar to the M-mode and Doppler techniques.9 Significant advantages of tissue Doppler compared to the classical M-mode and Doppler techniques for the diagnosis of fetal arrhythmias have not been obvious so far, particulary as there are some limiations inheritent to tissue Doppler imaging techniques.10 Fetal magnetocardiography records the magnetic fields generated by the

462

Fetal Cardiology

Figure 33.1 M-mode echocardiogram of the heart in a fetus at 22 + 4 weeks’ gestation. The cursor is positioned across the right atrium (RA) and the left ventricle (LV), producing the M-mode tracing as shown. It reveals a supraventricular reentry tachycardia at 270 beats/ minute. Each contraction (marked by arrows) of the atrial wall (A) is followed by one of the ventricular wall (V) indicating a 1:1 atrioventricular conduction.

Figure 33.2 M-mode echocardiogram of the heart in a fetus at 31 + 2 weeks’ gestation. The cursor is positioned across the right atrium (RA) and the left ventricle (LV), producing the M-mode tracing as shown. It reveals an atrial flutter with a 2:1 atrioventricular conduction resulting in an atrial rate of around 450 beats/minute and a ventricular rate of around 225 beats/minute. Each second contraction of the atrial wall (A) is followed by one of the ventricular wall (V).

electrical activity of the fetal heart, exhibiting significantly better signal quality compared with fetal electrocardiography, because it is largely unaffected by the high electrical resistance of the fetal skin, which attenuates fetal electrocardiography. At present, sensors cooled by liquid helium are positioned several centimeters above the maternal abdomen in a magnetically shielded room. Fetal magnetocardiography offers a more precise delineation of fetal electrophysiology, or more precisely magnetophysiology, allowing measurements of different intervals such as

PR, QRS, and QT.11–16 Diagnosis of the different arrhythmias (Figure 33.3) and long-QT syndrome by fetal magnetocardiography has been reported,12–14,16 as well as electropyhisiological patterns of initiation and termination of atrioventricular reentrant tachycardia in fetuses.15 Because of its high technical prerequisites, fetal magnetocardiography is very expensive, and is available in only a few centers around the world. However, improvements of the technique may overcome these problems in the near future.14

Fetal tachyarrhythmia

2.0

B (pT)

1.5

1.0

0.5 0.0 0

200

400 Time (ms)

600

Figure 33.3 Fetal magnetocardiogram (averaged QRS complexes) in a fetus with supraventricular tachycardia (SVT) of 240 beats/minute at 34 weeks’ gestation. In a normofrequent phase a delta-wave (Δ) as a part of the R wave could be demonstrated, suggesting the presence of Wolff–Parkinson–White (WPW) syndrome. This was confirmed by neonatal electrocardiogram. B, amplitude (courtesy of Uwe Scheider, University of Jena, Germany).

Electrophysiological mechanisms of fetal tachyarrhythmia In general, the natural history and pathophysiology for many of the fetal tachyarrhythmias are documented incompletely. Using M-mode and/or Doppler echocardiography, most prenatal series separate the SVT with a 1:1 AV contraction relation from atrial flutter. An 1:1 AV conduction most probably suggests an atrioventricular reentry tachycardia via an accessory pathway. Moreover, that an atrial ectopic tachycardia and junctional reciprocating tachycardia are also associated with a 1:1 AV contraction relation cannot be excluded, although they are less common. Therefore, the type and prevalence of fetal tachyarrhythmia are inferred from more exact electrophysiological analysis in neonates. It appears that the spectrum of fetal tachyarrhythmia closely resembles that found during the neonatal period. However, the exact definition of the type of fetal tachycardia may still be difficult. Variable outcomes reported in different studies may reflect the presence of different rhythm disturbances rather than a variation in response to the arrhythmia. Therefore, the initial assessment of the type of arrhythmia may have to be revised if there is a failure of response to the common antiarrhythmic treatment. Therefore, knowledge of the mechanism of tachycardia is potentially important, as it will define both the response to treatment and the prognosis.17–21 In this context, the timing of the ventriculoatrial relation using fetal M-mode or pulsed wave spectral Doppler echocardiography may be helpful

463

in distinguishing short and long ventriculoatrial time interval tachycardias.19–21 Simultaneous Doppler recording of flow velocity waveforms in the superior vena cava and the ascending aorta reflect the ventriculoatrial (VA) and atrioventricular (AV) time intervals more accurately than M-mode. Measurements of these intervals by simultaneous recording of left ventricular inflow and outflow is impracticable at heart rates above 160 beats/minute because of mitral E- and A-wave overlap.20,21 A short ventriculoatrial time interval is characteristic of atrioventricular reentry via an accessory pathway. A long time interval, which is defined as a ratio of the ventriculoatrial to the atrioventricular time interval above 1 (long VA tachycardia) is typical of permanent junctional reciprocating tachycardia or atrial ectopic tachycardia, which are more refractory to antiarrhythmic treatment.7 Atrial ectopic tachycardia may occur in fetuses and originate from a single abnormal focus or a wandering pacemaker. The more rarely occurring permanent junctional reciprocating tachycardia belongs to the conduction system tachycardias mediated by a concealed, backward conducting accessory pathway with slow AV node-like conduction properties. In both long VA tachycardias, the P-wave is far from the QRS complex and the venous Doppler a-wave is of normal amplitude (atrial contraction during diastole against open AV valves) with a long ventricular interval (VA > AV). The atrial ectopic and the permanent junctional reciprocating tachycardia of the fetus are characterized by frequencies between 180 and 220 beats/minute, whereas for atrioventricular reentry, tachycardia frequencies between 220 and 280 beats/minute are typical. A very short ventriculoatrial interval (very short VA tachycardia) with concomitant onset of tall venous atrial and aortic Doppler waves characterizes atrioventricular nodal reentrant tachycardia and junctional ectopic tachycardia.21 Furthermore, atrial ectopic tachycardia sometimes shows a higher variability of the heart rate, termed a heating phenomenon,8 which is very atypical for the classical atrioventricular reentry tachycardia. During early human life, the frequency of the various SVTs and their underlying electrophysiological mechanisms varies depending on the age at presentation.22 Studies in newborns and infants with tachyarrhythmias22,23 suggest that around 80–90% of fetal SVTs are atrioventricular reentrant tachycardias. These are based on an accessory atrioventricular conduction pathway besides the atrioventricular node, whereas atrial ectopic tachycardia, chaotic or multifocal atrial tachycardia, atrioventricular nodal reentrant tachycardia, permanent junctional reciprocating tachycardia, and His bundle tachycardia (junctional ectopic tachycardia) are seldom the electrophysiological mechanism of perinatal SVT.1,17–22,24,25 Accordingly, a Wolff– Parkinson–White syndrome can be confirmed electrocardiographically in approximately 10% of fetuses with SVT. Much less common are other preexcitation syndromes such as the Low–Ganong–Levine syndrome and the Mahaim

464

Fetal Cardiology

syndrome.25 In the vast majority of atrioventricular reentry tachycardias there is an ‘orthodromic’ impulse conduction, meaning that the electrical impulse is conducted from the atrium to the ventricle via the atrioventricular node and subsequently back to the atrium via the accessory pathway with fast conduction properties. Therefore, this ‘orthodromic’ atrioventricular reentry tachycardia is a short VA tachycardia characterized by a narrow QRS complex and a retrograde P-wave shortly after the narrow QRS complex in the electrocardiogram; in simultaneous Doppler recording of the superior vena cava and ascending aorta, the venous a-wave in the superior vena cava is superimposed on the aortic wave and is tall because of atrial contraction against closed atrioventricular valves.20,21 This depolarizing current circuit works independently from the physiological sinoatrial pacemaker. An ‘antidromic’ reciprocating tachycardia may also occur in cases of Wolff–Parkinson–White syndrome and concealed atrioventricular reentry tachycardias with antegrade impulse conduction through a slow accessory pathway. The prerequisite for the appearance of a reentry tachycardia is different electrophysiological properties of both pathways, having a different conduction velocity and/or refractoriness. Under these circumstances the reentry tachycardia is usually triggered by an incidental supraventricular extrasystole if there is a critical relationship between conduction velocity and refractory period. The reentrant premature atrial contraction causes delayed antegrade conduction through the AV node with subseqeunt retrograde conduction through the accessory pathway, thus initiating ‘orthodromic’ atrioventricular reentry tachycardia via an accessory pathway. Atrial ectopy is common in fetuses, with an incidence of 1–3%. Approximately 0.5–1.0% of these fetuses will develop SVT during fetal and neonatal life, resulting in the general recommendation to check the fetal heart rate weekly for detection of an occurrence of SVT. Early occurrence of fetal ectopy, frequent ectopy, and the presence of multiple accessory pathways are risks for the occurrence of reentrant SVT during fetal life. Furthermore, variations in autonomic activity may influence conduction in the AV node and accessory pathway compatible with the association between fetal body movements and patterns of SVT initiation involving premature atrial contractions and sinus acceleration.15 In atrioventricular reentrant tachycardia, the conduction velocity and the length of the reentrant circuit determine the relatively fixed frequency of the reentrant tachycardia with failed heart rate variability.17,24 In fetal atrioventricular reentry tachycardia via an accessory pathway, this is between 220 and 280 beats/minute. Drug-induced alteration of conduction velocity and/or refractoriness of the reentrant circuit may stop or prevent the reentry tachycardia. Sometimes a premature ectopic beat may also interrupt a reentry tachycardia as well as parasympathetic excitation. Fetal tachycardias between 180 and 220 beats/minute suggest other electophysiologic mechanisms and types

of tachycardia, most often permanent junctional reciprocating and atrial ectopic tachycardias, respectively. The high incidence of atrioventricular reentry tachycardia during fetal and neonatal life and its decrease due to the spontaneous disappearance of SVT suggest the immaturity of the myocardium, and in particular a delayed development of the annulus fibrosus and/or prolonged persistence of accessory atrioventricular pathways as the most underlying etiology of fetal SVT.22,26 Much less common are cases due to autosomal dominant inheritance of Wolff–Parkinson–White syndrome, or SVT in fetuses with Ebstein’s anomaly, rhabdomyoma, or viral myocarditis. Fetal atrial flutter, which accounts for 25–30% of all cases of fetal tachyarrhythmias,2 and atrial fibrillation, which is extremely rare in the fetus, are most commonly generated within the atria themselves by an intra-atrial reentrant circuit. Experiments, and the observation that atrial flutter is observed only during the third trimester, support the favored hypothesis of an atrial macroreentry as the underlying mechanism of fetal atrial flutter. The atrium probably reaches a critical size for establishing a macroreentry circuit at about 27 to 30 weeks’ gestation, associated with a high vulnerability against triggering atrial extrasystoles.27 The frequency of this atrial reentrant tachycardia is between 350 and 500 beats/minute. The atrioventricular node, which is not part of the reentrant circuit, usually protects the ventricles by variably blocking atrioventricular conduction, with the result of a substantially slower fixed or varying ventricular rate, depending on the degree of AV block, which may be a fixed or varying 2:1, 3:1, or 4:1 block. In some fetuses with atrial flutter, the presence of an accessory atrioventricular pathway has been shown by postnatal transesophageal electrophysiological studies in neonates with previously documented fetal atrial flutter.25 Distinct atrial dilatation resulting from severe AV valvular regurgitation in cases of Ebstein’s anomaly and rarely atrioventricular septal defects may sometimes cause atrial flutter during fetal life, similar to the well known pathomechanism of atrial flutter as a consequence of left atrial dilatation due to mitral valve dysfunction in later life. There are only a few reports of fetal ventricular tachycardia. Prenatal diagnosis is usually performed if the ventricular tachycardia shows AV dissociation, with ventricular rates varying from 180 to 300 beats/minute in excess of the atrial rate, as documented by M-mode (Figure 33.4) and/or Doppler echocardiography.17,24,28,29 This pattern of tachyarrhythmia is highly suggestive of either a ventricular or a junctional origin of the tachycardia, whose differentiation seems to be impossible during fetal life. In some cases of ventricular tachycardia, however, retrograde AV conduction may lead to a 1:1 contraction sequence of atria and ventricles, making prenatal differentiation from SVT difficult or impossible. Ventricular tachycardia should be suspected if the fetal heart rate lies outside the normal range of 220–280 beats/minute for

Fetal tachyarrhythmia

465

Figure 33.4 In a hydropic fetus with severe dilated cardiomegaly, a ventricular tachycardia suddenly occurred at 28 + 4 weeks’ gestation. The M-mode echocardiogram – the cursor is positioned across the right atrium (RA) and the left ventricle (LV) – reveals a ventricular tachycardia with an AV dissociation (ventricular rate of 245 beats/minute and atrial rate of 145 beats/minute). Postnatally long-QT syndrome was diagnosed.

SVT, especially if transient AV dissociation is observed in addition. Similar to the more common ventricular tachycardia in coronary heart disease in later adulthood, causing focal alteration of the ventricular myocardium, a locally restricted reentrant circuit may also be the underlying electrophysiological basis of ventricular tachycardia during fetal life. In fetal life, segmental alterations of myocardial oxygen supply may occur in severe myocardial hypertrophy secondary to semilunar valve stenosis and cardiomyopathy, or from cardiac tumors resulting in ventricular tachycardia.24 Furthermore, fetuses with the long-QT syndrome predominantly show sinus bradycardia with intermittent AV block, but seldom also show transient ventricular tachycardia.30–35 The presence of ventricular tachycardia in cases with long-QT syndrome appears to be associated with a particularly poor prognosis, and also with a ‘spongy myocardium’.31,35 Therefore, sinus bradycardia in combination with episodes of AV block and periods of ventricular tachycardia are strongly suspicious for fetal long-QT syndrome. In these cases, the family history should be checked and the corrected QT interval (QTc > 0.44; QTc = QT/√RR) of both parents should be measured on the electrocardiogram, even if a positive family history may be lacking in infants with prolonged corrected QT interval.21 Inherited anomalies of myocardial structure and function may induce these arrhythmias. The long-QT syndrome (Romano–Ward syndrome) belongs to the ion channel diseases, which are clinically characterized by episodes of disturbed excitability of muscle or nerve cells. For the long-QT syndrome, various mutations in cardiac potassium and sodium channel encoding genes inducing changes in channel gating have been reported, showing a different mode of inheritance.36

Nowadays, measurement of the QT interval and prenatal diagnosis of long-QT syndrome are possible by fetal electrocardiography, and in particular by fetal magnetocardiography.13,14,16 Sinus tachycardia is usually characterized by a baseline fetal heart rate between 180 and 210 beats/minute, and is commonly a secondary manifestation of an underlying disease. These include maternal fever, chorioamnionitis, fetal distress, fetal thyrotoxicosis, and maternal medication such as β-sympathomimetics. During the late third trimester longer accelerations, which can be found in a healthy ‘jogging’ fetus in the 4F activity state, must be distinguished from sinus tachycardia. Its association with increased fetal breathing and body movements, its commonly high variability of the fetal heart rate pattern, and particularly its transient occurrence allow this differentiation to be made.

Pathophysiology Pathophysiologically, there is a substantial shortening of the diastolic period of the cardiac cycle preventing adequate early diastolic filling of the ventricles, subsequently increasing the systemic venous volume load and the central venous pressure, which is already high in the normal fetal circulation. Furthermore, results from animal studies of left atrial pacing and M-mode echocardiographic studies in human fetuses suggest that initial left atrial depolarization can be tolerated to a lesser degree by the fetus.24,25 By pre-excitation of the left atrium, partial closure of the foramen ovale may significantly disturb the interatrial right-to-left shunt. This results in a trapping of

466

Fetal Cardiology

Figure 33.5

In a non-hydropic fetus at 22 + 4 weeks’ gestation with a supraventricular tachycardia at 270 beats/minute, Doppler measurement in the left hepatic vein revealed a pulsatile reversal of blood flow occurring during diastole. The antegrade blood flow during systole is in the lower channel (away from the transducer), the reverse blood flow during diastole in the upper channel (toward the transducer).

the venous return in a, by now, volume-overloaded right atrium and right ventricle, with further elevation of central venous pressure. Premature closure of the foramen ovale has also been reported in two fetuses with atrial flutter.37 The left ventricular output is subsequently diminished. In addition, the most common fetal SVT, ‘orthodromic’ atrioventricular reentry tachycardia, is characterized by a short ventriculoatrial interval with atrial contraction against closed atrioventricular valves causing a higher increase of venous pressure than long VA tachycardias.20 Furthermore, a very high ventricular rate and a manifestation of tachycardia in earlier gestation may predispose these fetuses to the development of hydrops, because both diastolic and systolic function of the isolated myocardium and the whole heart distinctly improve with advancing gestational age. Finally, ventricular filling is predominantly dependent on the atrial systole due to the relative stiffness of the fetal myocardium. Therefore, a regular 1:1 atrioventricular conduction sequence is important, and intermittent atrial contraction against a closed atrioventricular valve may predispose some fetuses with atrial flutter to the development of hydrops. This explains the observation that some fetuses with atrial flutter develop hydrops in spite of a normal or slightly increased ventricular rate, and why, in some fetuses with atrial flutter, hydrops may persist even if the ventricular rate is normalized by druginduced higher-degree AV block, and only disappears after the establishment of normal atrial and ventricular rate with 1:1 AV conduction. Although there are important differences between atrial pacing in animal models and supraventricular reentry tachycardia, atrial flutter, and rapid sinus tachycardia in human fetus atrial pacing, studies in fetal lamb models still

give important insights into the pathogenic mechanisms of the development of fetal hydrops.38–42 In atrial pacing at rates up to 300 beats/minute there is an increase of ventricular output and a decrease of ventricular enddiastolic pressure. Prolonged left atrial pacing at rates of 300–320 beats/minute results in a decrease of cardiac output and in the development of hydrops within 4–48 hours; this confirms the suggestion that, above a critical heart rate, diastolic filling is impeded especially for the ipsilateral ventricle. In these situations cardiomegaly and hepatomegaly develop, arterial oxygen tension remains unchanged, and protein, including albumin, concentrations stay stable or slightly decrease later on in the disease process.38–41 Because no or only slight, non-significant hypoproteinemia may be observed, there is no evidence for an increase of capillary permeability for albumin. In advanced stages of congestive heart failure, however, there may be more distinct hypoproteinemia due to disturbed hepatic synthesis. The aortic pressure remains unchanged, while the mean venous pressure in the inferior vena cava increases by 75%.42 This elevation may reflect a compensatory increase of venomotor tone for maintenance of adequate cardiac output by an elevation of the preload. However, the abrupt elevation of venous pressure is associated with an immediate appearance of pulsatile reversal of blood flow occurring during diastole (Figures 33.5–33.7).42 This is observed above a ‘critical’ heart rate of 310 beats/minute. Below this heart rate, the venous flow is biphasic, with a systolic and diastolic forward surge that also occurs immediately after the pacing is stopped.42 Besides the direct impeding of diastolic filling when the diastolic interval is critically shortened, the abrupt occurrence of changes – reduction of ventricular output, immediate elevation of

Fetal tachyarrhythmia

467

Figure 33.6 In a non-hydropic fetus at 22 + 4 weeks’ gestation with a supraventricular tachycardia at 270 beats/minute, Doppler measurement in the ductus venosus showed a triphasic venous blood flow pattern with a systolic and diastolic peak during a short period of sinus rhythm. An atrial premature beat is triggering the supraventricular reentry tachycardia associated with an abrupt onset of the pulsatile reversal of blood flow occurring during diastole. The antegrade blood flow during systole is in the lower channel (away from the transducer), the reverse blood flow during diastole in the upper channel (toward the transducer).

Figure 33.7 In a non-hydropic fetus at 22 + 4 weeks’ gestation with a supraventricular reentry tachycardia at 270 beats/minute, Doppler measurement in the umbilical cord showed arterial blood flow in the lower channel (away from the transducer) and the umbilical venous blood flow pattern with monophasic pulsations as typically associated with pulsatile reversal of blood flow in the ductus venosus and other precordial veins.

venous pressure, and appearance of pulsatile venous blood flow above a ‘critical’ pacing rate – suggests ventricular dysfunction consistent with alteration of the pressure– volume relationship in association with impaired ventricular relaxation at high pacing rates. The most likely explanation is that oxygen supply to the myocardium by coronary blood flow is inadequate for the increased requirement of the myocardium during tachycardia.43,44 Myocardial blood flow is maintained primarily by the pressure gradient across the vascular bed, extravascular pressure, and local autoregulation. Since the extravascular

pressure is considerably lower in diastole than in systole, the major portion of coronary blood flow occurs in diastole. In tachycardia, however, the diastolic period is significantly shortened. Moreover, with increased atrial pressure, the myocardial pressure gradients decrease. Severe ventricular dysfunction and even injury of the myocardium may occur in prolonged tachycardia, and may cause reversible tachycardia-induced ‘cardiomyopathy’ in humans and animals.43–45 In conjunction with the enormous cardiac dilatation in fetuses with sustained SVT, functional incompetence as indicated by annular

468

Fetal Cardiology

Figure 33.8 In a non-hydropic fetus at 22 + 4 weeks’ gestation with a supraventricular reentry tachycardia at 250 beats/minute and a consequent cardiomegaly, mild bilateral atrioventricular valvular regurgitations occurred during a short period of sinus rhythm.

Figure 33.9 In a non-hydropic fetus at 22 + 4 weeks’ gestation with a supraventricular tachycardia at 270 beats/minute, the size of the heart is moderately increased as indicated by measuring the cardiothoracic ratios (CTRs). The CTR of the circumferences is 0.59 (93/157 mm) and the CTR of the areas is 0.33 (654/1969 mm2).

enlargement of both atrioventricular valves may be observed, suggesting structural remodeling of the ventricles in the presence of tachycardia-induced ‘cardiomyopathy’ (Figures 33.8 and 33.9).46 In pigs, recovery from tachycardia-induced ‘cardiomyopathy’ was accompanied by persisting chamber dilatation, significant myocardial hypertrophy, and persisting diastolic dysfunction.47 During early recovery, left ventricular function and myocardial blood flow normalize at no exertion, but stress results in marked systolic and diastolic left ventricular dysfunction

and reduced myocardial blood flow.43,44 In human fetuses after termination of SVT, cardiac dilatation, myocardial hypertrophy, atrioventricular valve incompetence, and hydrops disappear with immense interindividual differences, which could be explained by different stages of progression of tachycardia-induced ‘cardiomyopathy’ at the time of drug-induced cardioversion.46 Venous blood flow studies in the human fetus with SVT demonstrate the occurrence of monophasic forward and pulsatile reversed blood flow during diastole in the inferior vena cava, hepatic

Fetal tachyarrhythmia

veins, and ductus venosus as well as of pulsations in the umbilical venous blood flow pattern above a critical heart rate of approximately 210–220 beats/minute.48 This is in accordance with fetal lamb studies, where this change of venous blood flow pattern was associated with a considerable elevation of venous pressure.42 Furthermore, in the human fetus, in addition to the persistence of cardiomegaly and atrioventricular valve regurgitation, abnormal indices of venous blood flow during sinus rhythm indicate the existence of altered myocardial function, suggesting the presence of a reversible tachycardiainduced ‘cardiomyopathy’.49,50 Times for remission of hydrops, disappearance of atrioventricular valve regurgitation, and normalization of the venous blood flow indices show a good correlation with the immense interindividual differences. The hydrops disappears first, then the atrioventricular valve incompetence recovers, and finally the venous Doppler indices become normal.46,49,50 The time interval from the drug-induced conversion into a constant sinus rhythm may differ from 1 day to 6 weeks.46,49–51 The time needed for complete normalization of cardiac function may also vary extremely, in accordance with different stages of progression of tachycardia-induced ‘cardiomyopathy’ at the time of conversion into a sinus rhythm.46,49,50 In conclusion, the most important pathomechanism in SVT of the human fetus is an impeded ventricular filling due to an inadequately short diastolic period, which may alter the ventricular filling directly and/or by changes of diastolic ventricular function due to inadequate oxygen supply by reduced myocardial blood flow. Both mechanisms result in an elevation of venous pressure, with a subsequently increased rate of transcapillary fluid filtration into the interstitial space, and significant reduction of lymphatic flow, inadequate for drainage of the increasingly produced interstitial fluid back into the vascular space.41 Data from studies in animals and humans suggest that, in less advanced stages of disease, the elevation of the venous pressure results in a substantially increased transcapillary fluid filtration rate. Furthermore, the impaired lymphatic drainage is the most important pathomechanism in sustained tachyarrhythmia, while hypoxia-induced increase of capillary permeability to water as well as proteins, and alterations of hepatic protein synthesis, are not relevant for the development of hydrops.41 Therefore, it seems to be advisable to treat fetuses with tachyarrhythmia in utero, because the congestive heart failure and the elevation of venous pressure are reversible after drug-induced cardioversion, while tissue hypoxia damaging capillary membranes and/or liver cells seems not to be present in these fetuses. Also, the tachycardia-induced ‘cardiomyopathy’ with diastolic and systolic dysfunction will improve and disappear before birth, if constant sinus rhythm can be established. Thus, adequate prolongation of pregnancy can be achieved.

469

Fetal surveillance and assessment of cardiac function in fetal tachyarrhythmia The assessment of fetal cardiac function and also fetal surveillance appear to be very difficult in fetuses with tachyarrhythmia. The value of fetal heart rate monitoring is limited during tachyarrhythmia, whereas sonography, echocardiography, and Doppler sonography, as non-invasive and repeatable methods, are the most important tools for the diagnosis and surveillance of these fetuses: 1. Detailed sonographic and echocardiographic examination should exclude cardiac and non-cardiac malformation, in particular Ebstein’s anomaly and cardiac tumors. 2. The amount of amniotic fluid, placental structure and thickness, and, in hydropic fetuses, distribution and extent of fluid accumulation should be assessed. 3. Echocardiography and Doppler sonography of the venous system may indicate or exclude cardiac failure, which may be the primary or the secondary cause in the advanced stage of diseases other than anemia. Cardiac dilatation as well as measurement of the cardiothoracic ratio and search for AV valve incompetence may help to assess the degree of tachycardia-induced cardiomyopathy. Evaluation for the detection of cardiomegaly as a non-invasive assessment of cardiac function in nonimmune hydrops is validated by the measurement of umbilical venous pressure.52 4. The value of fetal venous Doppler velocimetry is limited by the fact that, above a critical frequency of fetal tachycardia, which is approximately 210–220 beats/minute for the human fetus, pulsatile venous blood flow with diastolic reversal appears to be independent of the cardiac function and the degree of tachycardia-induced ‘cardiomyopathy’.48 In other words, reversal of the venous a-wave in the ductus venosus and the occurrence of a pulsatile flow pattern in the umbilical vein correspond to a distinct venous pressure increase which might be sufficient to cause interstitial water accumulation and hydrops, even if the myocardial function is preserved. Therefore, in fetuses with fetal tachyarrhythmia, increased pulsation in the venous blood flow pattern is not a prognostic marker, and in particular not an ominous sign. On the other hand, recurrence of the normal biphasic forward blood flow pattern in the precordial veins seems to indicate a significant drop in the fetal venous pressure. After cardioversion to sinus rhythm, however, measurement of the venous Doppler indices is the best method to evaluate cardiac function.49,50 5. Fetal heart rate monitoring should repeatedly be performed for longer periods, for detection of paroxysmal

470

Fetal Cardiology

tachyarrhythmia and/or for monitoring of the effects of the antiarrhythmic treatment. 6. In addition, sonographic demonstration of fetal breathing and body and extremities movements indicates fetal well-being, whereas fetal heart rate monitoring and Doppler velocimetry in umbilical and fetal arteries are not useful for fetal surveillance during tachyarrhythmia. However, Doppler velocimetry in the uterine arteries may exclude an associated uteroplacental dysfunction.

Treatment of fetal tachyarrhythmia Rationale for fetal antiarrhythmic treatment In utero treatment of fetal tachyarrhythmia has to consider both the fetal and the maternal condition. Therefore, the decision for in utero antiarrhythmic treatment should only be made after a detailed risk–benefit analysis, and should depend on the presence or absence of hydrops, duration of tachycardic periods, gestational age of the fetus, and type of tachyarrhythmia, and also on the maternal condition and willingness. Detailed counseling of the parents should always be performed. On the one hand, the high mortality and morbidity of preterm newborns with hydrops due to tachyarrhythmia has to be considered; on the other hand, there are favorable results for in utero antiarrhythmic treatment. Hence, the rationale for treatment of fetal tachyarrhythmia in hydropic fetuses is to establish a constant sinus rhythm, allowing the fetus to resolve the hydrops and the cardiac dysfunction, and to prolong the pregnancy to birth near term. A special risk of antiarrhythmic medications administered for the suppression of cardiac rhythm disturbances is the provocation of new, or the exacerbation of existing, arrhythmias. This proarrhythmic effect may affect the mother and the fetus, and may occur as an early proarrhythmic complication after initiation of therapy, varying from non-serious through potentially fatal arrhythmias or manifesting late as enhanced arrhythmic death.17,24,28,53 Late proarrhythmia may occur during the administration of all antiarrhythmic drugs except for β-blocking agents.24 Severe sinus and atrioventricular node dysfunction may be initiated by digitalis glycosides and Vaughn Williams class I–IV antiarrhythmic drugs (Table 33.1), atrial tachycardia with variable atrioventricular block and with junctional tachycardia may result from digoxin toxicity, torsades de pointes polymorphic ventricular tachycardia from type Ia and III antiarrhythmic agents prolonging the QT interval, and ventricular tachycardia from Ic agents.

Congestive heart failure, hypokalemia, hypocalcemia, and hypomagnesemia may predispose to proarrhythmias.24 Furthermore, inherited and acquired anomalies and gestational age-specific features of myocardial structure and function, anomalies of ion channels, and anomalies of enzymes of the cytochrome P450 pathway may be the underlying anomalies of proarrhythmic complications, especially in interaction with other antiarrhythmic and non-antiarrhythmic drugs.24,57 Whereas maternal health and risks can be sufficiently checked by detailed examinations, fetal disposition to proarrhythmia may be only insufficiently assessed. In particular, fetuses with advanced tachycardia-induced cardiomyopathy may be at risk for the negative-inotropic and other unforeseen effects related to antiarrhythmic therapy. Sudden fetal death shortly after the initiation of an antiarrhythmic therapy was reported for flecainide as well as sotalol,34,58,59 but almost all of these fetuses were hydropic; therefore, it is still unclear whether the fetal death was the consequence of a proarrhythmic drug effect, of a negative-inotropic drug effect, or of end-stage cardiac disease in the reported cases. Fetal therapy can be best monitored by ultrasound and echocardiographic techniques, demonstrating the extent and distribution of fluid accumulations, the cardiac rhythm, the cardiac size and contractility, and the occurrence of AV valve regurgitation. Sonographic examination of fetal breathing and body and extremity movements provides an assessment of fetal condition. Fetal electrocardiography and magnetocardiography are not useful, because reliable technology to provide high-fidelity recordings of fetal electrocardiograms is lagging behind and the availability is limited, respectively. In non-hydropic fetuses with sustained or intermittent tachycardia, observation without antiarrhythmic treatment may be a safe management option close to term, because hydrops will rarely develop, presumably owing to the better intrinsic properties of the fetal heart late in gestation. On the other hand, a trial of transplacental treatment with digoxin, if successful, will facilitate vaginal delivery by allowing the interpretation of fetal heart tracing. If unsuccessful, however, delivery and direct neonatal therapy of SVT appears preferable to intrauterine therapy. Elective cesarean section is the most often recommended mode of delivery in fetuses with SVT. The trial of vaginal delivery, however, appears to be justified in selected cases, because the risk of intrapartum hypoxemia does not seem to be increased in non-hydropic fetuses with SVT in particular, and with arrhythmia in general. After the antenatal assessment of normal uteroplacental function by sonography, biophysical profile, and Doppler sonography, vaginal delivery may be performed. Sometimes increased vagal tone during labor may interrupt the SVT; in other cases, alternative techniques of intrapartum surveillance may be utilized, such as repetitive fetal scalp blood sampling for monitoring the acid–base

Fetal tachyarrhythmia

status alone or in combination with continuous PO2 monitoring.60,61 In fetuses with paroxysmal intermittent SVT with short periods of tachycardia, which can be confirmed by heart rate monitoring for 12–24 hours, treatment can be deferred, as was recently shown in fetuses with intermittent SVT.62 However, close observation is necessary for the early detection of conversion to sustained tachycardia and the occurrence of congestive heart failure and hydrops.62 If there are long tachycardic episodes, then it would be reasonable to start antiarrhythmic treatment in nonhydropic fetuses, in particular during the second and early third trimesters. The conversion to sustained tachycardia may take place and/or hydrops may develop in intermittent tachycardia,56,58,63 decreasing therapeutic options by impairing transplacental passage of drugs in the presence of hydrops. Elective delivery before 34 weeks of gestation of hydropic fetuses with sustained and intermittent tachyarrhythmia generally results in a mixture of many complications and problems in the management of premature hydropic fetuses. These are: postpartum increase of cardiac work, need for own regulation of body temperature by the neonate, mechanical ventilation, repetitive pleural drainage, congestive heart failure, simultaneous occurrence of pulmonary edema and hyaline membrane disease, reducing the effectiveness of surfactant therapy, severe degree of tachycardia-induced ‘cardiomyopathy’ with impaired diastolic and systolic cardiac function, refractory neonatal tachyarrhythmia, and side-effects of antiarrhythmic and other cardiovascular drugs on the preterm newborn. In consequence, the outcome of this approach is poor, even if the problems and side-effects of transplacental treatment are avoided. Thus, in utero antiarrhythmic treatment for adequate control of the arrhythmia and remission of hydrops is prudent in sustained fetal tachyarrhythmia with hydrops, because data from animal and human studies indicate that hydrops results from elevated venous pressure and consecutive obstruction of lymphatic drainage, but not from hypoxic damage to capillaries or other tissues. Therefore, in hydropic fetuses with tachyarrhythmia, intrauterine treatment with digoxin alone or in combination with different antiarrhythmic drugs (flecainide, sotalol, amiodarone) is the best approach for almost all fetuses. Transplacental treatment is successful in the majority of cases,2,17,56,58,64–69 whereas additional direct antiarrhythmic treatment is limited to rare, severely hydropic fetuses with tachyarrhythmia refractory to transplacental treatment.

Maternal surveillance during fetal antiarrhythmic treatment Because of its potentially hazardous and life-threatening complications, each antiarrhythmic treatment should be

471

started in an inpatient setting. Before initiation of any antiarrhythmic therapy, an accurate medical examination of the pregnant woman and a 12-lead electrocardiogram must be performed to exclude hidden maternal diseases such as Wolff–Parkinson–White syndrome, prolonged QT interval, myocarditis, or other contraindications for antiarrhythmic treatment in general or for some drugs only. Maternal serum sodium, potassium, chloride, calcium, and magnesium levels should be evaluated as well as those of blood urea nitrogen, creatinine, and albumin. Thyroid function should also be checked, especially if treatment with amiodarone is considered. Daily maternal electrocardiographic monitoring with special attention to PR, QT, and QRS prolongation and careful recording of signs and symptoms of possible maternal side-effects are the most important parameters of the therapy, whereas the measurement of drug serum concentrations is mainly helpful to document an underdosage. Dangerous complications to the mother and the fetus may further be reduced by lowering the drug dosage at the start of therapy, by incrementally increasing the dosage, and by avoiding hazardous drug combinations, especially during the phase of the change to second- or third-line therapy. On the other hand, high dosages and maternal serum levels of the antiarrhythmic agent as well as allowance of sufficient time to reach adequate drug levels even in the fetal compartments are necessary for successful cardioversion of fetal tachyarrhythmia to a constant sinus rhythm, and, in consequence, may indirectly reduce the risks by avoidance of the application of more dangerous second- or third-line agents.

Fetal antiarrhythmic treatment protocols The pharmacology, pharmacokinetics, and bioavailability of antiarrhythmic drugs may be substantially affected by physiological changes during pregnancy, such as increased extra- and intravascular fluid volume, increased glomerular filtration, delayed gastric emptying and intestinal transit, progesterone-induced enhancement of hepatic enzymatic activity, and relative decrease in plasma proteins. Furthermore, many aspects of transplacental transfer and fetal pharmacology of antiarrhythmic drug treatment related to gestational age are still unknown. Additionally, placental transfer of the drugs may vary, influenced by gestational age-dependent changes in the placenta, disorders in development of the villous placenta, mainly disorders of the fetal circulation associated with elevated venous pressure, and placental edema following tachyarrhythmia. Therefore, concentrations of the antiarrhythmic drug measured in the maternal blood may strongly differ from those in the fetal blood. The physiological and pathophysiological changes at muscular and cellular levels in the

SVT, AF, VT

AF, SVT, VT

Ic (sodium-channel blockade delaying conduction with normal repolarization)

III (+ II) (increase of repolarization by potassium-channel blockade and β-adrenoreceptor blockade)

Flecainide

Sotalol

SVT, AF

Cardiac glycoside

Digoxin

Indication

Class

Drug

Fetomaternal ratio

Renal excretion; t½: 15–17 h; (serum level: 1.5–2.5 µg/ml)

Hepatic excretion: 60% Renal excretion: 40%; t½: 12–18 h (therapeutic plasma level: 0.4–1.0 µg/ ml)

0.7–0.9

0.7–0.8

0.8–1.0; substantially Renal excretion; t½: reduced in hydropic 34–36 h Therapeutic serum levels: fetuses 2.0–2.5 ng/ml

Metabolism

Negative inotropic effect, proarrhythmia

Negative inotropic effect, proarrhythmia

Low fetal serum levels in hydropic fetuses; contradicted in WPW syndrome which is not detectable in the fetus; until now there is no report of glycosideinduced VT in the fetus Narrow therapeutic range: nausea, vomiting, anorexia, diarrhea, fatigue, colored vision, confusion, insomnia, sinus bradycardia, ES, AV block, VT. Increased toxicity in hypokalemia, hypomagnesemia, hypercalcemia. Contraindication: WPW syndrome, VT, AV block II°–III° Proarrhythmia, vertigo, nausea, disturbed vision, headache, parasthesia

Fetal

Maternal

Side-effects and precautions

Proarrhythmia: VT; PO: 80–160 mg q 12 h; bradycardia; AV block, increase to 160 mg q 8 h Dose adjustment in renal nausea failure

PO: 100 mg q (6–) 8 h

Loading dose over 2–3 days: IV: 0.3–0.5 mg q 8 h Maintenance dose: PO: 0.15–0.2 mg q 8 h Dose adjustment in renal failure

Dosage

Table 33.1 Antiarrhythmic therapy in fetuses with tachyarrhythmia – data for the common antiarrhythmic agents. This table is modified from corresponding tables and data, respectively, in the following publications: Kleinman and Copel 1991,17 Kleinman et al 1999,28 Kleinman et al 2001,24 Gembruch and Somville 1994,54 Ito et al 1994,55 Simpson 2000,56 Fouron 200421

472 Fetal Cardiology

II (β-adrenoreceptor blockade) SVT, VT

SVT, AF, VT

0.1–0.3; substantially reduced in hydropic fetuses

Fast hepatic inactivation 0.1–0.3 (‘first pass effect’); t½: 3–5 h (therapeutic plasma level: 50–1000 ng/ ml)

Hepatic metabolism to active desethylamiodarone; renal excretion of metabolites; t½: 14–100 days; (therapeutic plasma level: 1–2 µg/ ml amiodarone; desethylamiodarone 1.5–2.0-fold higher)

PO: 40–80 mg q 8–12 h

Loading dose over 5–7 days: IV: 1200 mg over 24 h permanent infusion; or PO: 200 mg q 4–5 h Maintenance dose: PO: 200 mg q 6–8 h Direct: 2.5–5 mg/kg estimated fetal body weight (deduction of hydrops) infusion into the umbilical vein over 10 min q 6–8 h Bronchospasm (contraindication: asthmatic women with increased bronchic reaction); bradycardia; AV blockade; increased hypoglycemia in diabetic women; cold hands and extremities (contraindication: Raynaud phenomenon)

Proarrhythmia: VT; malfunction of thyroid, corneal microdeposits, photosensitivity, hepatic malfunction; (under sustained treatment: lung fibrosis, neuropathy, myopathy); (anticonception at least until 12 months after treatment)

Negative inotropic effect; bradycardia; AV blockade Newborn: hypoglycemia, bradycardia, and respiratory depression; low birth weight

Malfunction of thyroid: transient hypothyroidism (control of fetal thyroid hormones); corneal microdeposits; mild negative inotropic effect (proarrhythmia)

Only maternal plasma levels of digoxin are relevant for dose adjustment during fetal treatment. Overdosage can be recognized by serial maternal electrocardigram controls PR (digoxin), QRS (flecainide), and QT (class III agents) interval, and typical clinical signs of intoxication. SVT, supraventricular tachycardia; AF, atrial flutter; VT, ventricular tachyarrhythmia, IV, intravenous; PO, oral; ES, extrasystole; AV, atrioventricular; WPW, Wolff–Parkinson–White.

Propranolol

Amiodarone III (increase of repolarization by potassium-channel blockade)

Fetal tachyarrhythmia 473

474

Fetal Cardiology

fetal heart during pregnancy are also significant. Owing to all these reasons, animal models and models of an artificial placenta can illuminate only some of the important aspects of transplacental therapy of fetal tachyarrhythmias. The same applies to data gained from single cases in human fetuses of the transplacental transfer of antiarrhythmic drugs. In vitro examinations of the human placenta showed a dose-related relaxation of arteries and veins with nearly all antiarrhythmic drugs.70 Only adenosine led to a constriction of these vessels.70 With a model of a placental lobule perfused from the fetal as well as from the maternal side, it was shown that digoxin reached the fetal compartment very well, flecainide transferred well, and amiodarone transterred only insufficiently.71,72 At a low perfusion rate the transplacental transfer of digoxin decreased.71 An increase of venous pressure with subsequent development of placental edema, a smaller villous surface capillary density, and a widened maternal–fetal diffusion in the early weeks of pregnancy probably decreases the transfer of these drugs in vivo to fetuses with tachyarrhythmias during the second and third trimesters. Choice of appropriate drug and route of administration to achieve a rapid therapeutic level in the fetal compartment, and early detection of maternal and fetal complications, determine successful intrauterine antiarrhythmic treatment of the fetus. Digoxin is generally used as agent of first choice for the intrauterine therapy of fetal tachyarrhythmias, especially if a short VA tachycardia suggesting atrioventricular reentry tachycardia is diagnosed. Serum levels in the fetus range from 70 to 100% of the maternal serum level, with normal transplacental passage in the absence of hydrops. In hydropic fetuses, however, the placental passage of digoxin is distinctly impaired, and adequate concentrations of digoxin in the fetal compartments cannot be obtained by transplacental treatment via the mother.64,73–78 Owing to the increase of glomerular filtration rate toward term, the elimination half-life of digoxin substantially decreases in the second and third trimesters of gestation, resulting in a higher dosage of digoxin for adequate loading and maintenance therapy.79 For the treatment of fetal tachyarrhythmia, high maternal digoxin levels between 2.0 and 2.5 ng/ml should be achieved. The measurement of digoxin levels should be performed at least 6–8 hours after the last dose of digoxin, using assays that do not include endogenous digoxin-like immunoreactive substances found in maternal, fetal, and neonatal blood and that are altered by digoxin therapy.80,81 The addition of amiodarone, flecainide, verapamil, and quinidine results in increased digoxin levels, and should be accompanied by a decrease in the maintenance dosage of digoxin. Rapid intravenous loading within 48–72 hours for initiation of therapy followed by oral maintenance is generally preferred, especially in hydropic fetuses.64 In non-hydropic fetuses, however, the more convenient loading of the mother by oral administration,

which is only completed after 6–7 days, can be used, especially if the admission of the pregnant woman to hospital should be avoided.56 In a case of maternal renal failure, dosage reduction is mandatory, according to the maternal creatinine clearance. For second- and third-line therapy and for first-line therapy in already hydropic fetuses, other medications including flecainide, sotalol, and amiodarone provide an alternative transplacental therapy alone or in combination with digoxin (Table 33.2). Other antiarrhythmic agents, however, such as procainamide, quinidine, disopyramide, propafenone, propranolol, and verapamil, are generally no longer used for in utero antiarrhythmic therapy, because of insufficient therapeutic effects and/or substantial sideeffects in the fetus and/or the mother.17,24,28,56 Flecainide, a class Ic antiarrhythmic agent, strongly blocks the fast sodium channel, which slows the conduction velocity in most cardiac pathways and has no influence on repolarization. Its therapeutic range lies between 200 and 1000 ng/ml of serum levels. Flecainide has an excellent bioavailability of 95% with oral therapy and a good placental passage; 80% of the maternal plasma levels are achieved also in hydropic fetuses.83 Before final cardioversion, there occurs a gradual slowing down of the frequency of the supraventricular reentry in many cases. The time interval from initiation of the flecainide therapy to fetal cardioversion into sinus rhythm may range from 1 to 14 days, mostly occurring on days 4–7.82 Paradoxical proarrhythmic effects may occur in virtually all of the antiarrhythmic drugs including provocation or exacerbation of arrhythmia after initiation of therapy;34,58,59 late arrhythmia-related death could be more common in this class than in other types of antiarrhythmic drug. This was observed in adults treated with flecainide and encainide following myocardial infarction84 and also in children treated with flecainide or encainide because of supraventricular and ventricular tachycardia, where the incidence of significantly proarrhythmic effects was 7.5%.85 Sotalol is a class III antiarrhythmic drug with an additional β-adrenergic receptor blocking effect. It substantially prolongs repolarization and the action potential and has only a mild negative inotropism. Its bioavailability and placental passage are so good that adequate fetal sotalol levels between 70 and 100% of the maternal levels may be achieved within 48–72 hours after the initiation of oral administration.86 Because of its exclusively renal elimination – the half-life is around 16 hours – dosage reduction is mandatory if maternal renal function is decreased. The class III antiarrhythmic drugs sotalol and amiodarone are also associated with a proarrhythmic risk, especially for the development of torsade de pointes/ventricular fibrillation in the mother. To minimize this risk for the mother, long-QT syndrome must be excluded before the initiation of therapy by maternal and family history of arrhythmic events and by electrocardiogram (ECG) and sometimes Holter monitoring; during treatment,

Fetal tachyarrhythmia

475

Table 33.2 Therapeutic protocol for the antiarrhythmic therapy of fetal tachyarrythmia currently used in our hospital based on data from the literature2,17,21,24,34,56,58,59 and own experience2,54,82 Tachyarrhythmia

First choice

Second choice

Third choice

PSVT (short-term and without hydrops)

Control twice a week

PSVT (long-term and/or with hydrops, especially below 30th week of gestation)

Digoxin

Digoxin + flecainide

Digoxin + amiodarone

SVT without hydrops

Digoxin

Digoxin + flecainide

Digoxin + amiodarone

SVT with hydrops, without AV valve regurgitation

Flecainide (+ digoxin)

Amiodarone

In addition: amiodarone directly in the umbilical vein

SVT with hydrops and severe AV valve regurgitation

Amiodarone (if necessary directly in addition)

In addition: amiodarone directly in the umbilical vein

Long VT tachycardia without hydrops

Amiodarone or sotalol

Control twice a week

Long VT tachycardia with hydrops

Amiodarone or sotalol

Sotalol or amiodarone, respectively

AF without hydrops

Digoxin

Control twice a week; digoxin + flecainide

AF with hydrops

Digoxin

Digoxin + flecainide; Digoxin + amiodarone alternatively: digoxin + sotalol (directly and transplacentally)

VT without hydrops

Control twice a week

VT with hydrops

Propranolol or flecainide

Flecainide or propranolol, respectively

Digoxin + sotalol

Amiodarone (+ digoxin)

1. If the frequency of tachycardia can be reduced under flecainide and other antiarrhythmic therapy below 210 beats/minute and pulsatile monophasic blood flow in the precordial veins is replaced by normal biphasic antegrade blood flow pattern, we can expect a significant reduction of fetal venous pressure and medication should be continued even if complete cardioversion is not obtained. 2. After the 34th week of gestation, fetuses with supraventricular tachycardia may be delivered, in particular if hydrops progresses despite antiarrhythmic therapy. PSVT, paroxysmal supraventricular tachycardia.

the maternal ECG should regularly be evaluated for changes in QTc interval. Meanwhile, there is profound evidence that some drug-induced arrhythmias might also represent cases of ‘forme fruste’ congenital long-QT syndrome with a corresponding molecular basis. In these cases not only cardiac drugs but also non-cardiac drugs such as erythromycin, terfenadine, haloperidol, or cisapride may cause excessive prolongation of the QTc interval and life-threatening arrhythmias.24 In the future, individuals at risk for proarrhythmic effects may be identified by screening for mutations of candidate genes.57 Proarrhythmic effects were also observed in children treated with sotalol.87 Sotalol has proved to be safe and efficacious in the treatment of fetal supraventricular tachycardia refractory to digoxin alone.88 A series of 21 fetuses treated with sotalol, however, showed a relatively high intrauterine mortality of four of the total of 21 fetuses and four of 10 hydropic fetuses. Three fetuses had supraventricular tachycardia, and one fetus had atrial flutter – with occurrence of fetal death within 1 week after the initiation of

sotalol treatment.34 It appears that the proarrhythmic impact of sotalol is more pronounced in the immature fetal and also the neonatal heart than it is in adult hearts.34 Therefore, sotalol treatment should be restricted to thirdline therapy in hydropic fetuses with supraventricular tachycardia refractory to digoxin alone and digoxin and flecainide in combination.34 In supraventricular tachycardia with long ventriculoatrial time interval (long VA tachycardia) suggestive for permanent junctional reciprocating tachycardia, or atrial ectopic tachycardia, and in atrial flutter with hydrops, the class III agents sotalol and amiodarone are indicated for fetal transplacental therapy.19–21 Low initiation dosages of sotalol and stepwise dosage increases may decrease the risk of proarrhythmia.34,86 Amiodarone, as a class III antiarrhythmic drug, substantially prolongs repolarization and the action potential. In contrast to the other antiarrhythmic drugs, it is characterized by a very long elimination half-life of 1–3 months, and has an active hepatic metabolite, desethylamiodarone. With only maternal treatment it has a low fetal

476

Fetal Cardiology

bioavailability but offers the advantage of only minimal negative inotropism. An oral or intravenous loading of the mother with 1200 mg/day for 4–6 days is followed by a maintenance dosage of oral 600–900 mg/day. Although the transplacental transfer is poor, with fetomaternal ratios for amiodarone between 10 and 40%, and seems further impaired in hydropic fetuses, there is a constant increase of fetal serum levels of amiodarone and desethylamiodarone, presumably due to the very long elimination halflife in the fetal compartments. Regarding the duration of treatment, application of amiodarone for fetal tachyarrhythmia during pregnancy is a short-term therapy, since in the majority of infants the tachyarrhythmia can be treated effectively with other drugs after birth. Because of the finite period of administration, the most serious sideeffects of long-term amiodarone therapy such as interstitial pneumonia, pulmonary fibrosis, neuropathy, and myopathy are unlikely to occur in this context. In contrast, some cases of fetal and neonatal iodine-induced transient hypothyroidism have been reported after maternal and/or fetal direct therapy,64,76,77,89–93 because 200 mg amiodarone contains 75 mg iodine, even if neonatal hypothyroidism appears to be lacking in the majority of cases with maternal amiodarone therapy during pregnancy. The frequency of transient neonatal hypothyroidism in cases of long-term maternal administration of amiodarone during pregnancy is around 20%, with favorable evolution after a few months.92,93 Therefore, before and during therapy, monitoring should be carried out of the maternal thyroid gland and thyroid hormones, as well as the neonate directly after birth and during the first weeks of life.64,76,89 Exclusion of fetal hypothyroidism by cordocentesis and subsequent fetal thyroxin therapy may be an additional option.64,76 Negative effects of fetal hypothyroidism during the second half of gestation on the later neurophysiological development are in discussion, even when hormone substitution is started immediately after birth.94,95 Postnatal long-term follow-up examination of the psychomotor development of only a few children exposed transplacentally to amiodarone during their fetal life showed favorable global intelligence quotient (IQ) scores and well developed social competence, but mild deficits in some non-verbal skills such as reading comprehension, written language, and arithmetic.96 Nevertheless, this important question has to be validated by larger follow-up studies. Only because of the possible negative effect of amiodaroneinduced transient hypothyroidism on fetal and neonatal neurodevelopment did treatment with this highly effective antiarrhythmic drug turn into second- and third-line therapy for selected severely hydropic and otherwise drugrefractory fetuses.64,75,76,97–100 Particularly if the situation forces direct fetal treatment, amiodarone seems to be the ideal drug, owing to its extremely long elimination halflife and the minimal inotropism as outlined below; the use of amiodarone is justified in advanced stages of this lifethreatening disease.46,64,75,76,97,98

Adenosine, an endogenous purine nucleoside, has an almost immediate but short-lasting effect mediated by A1-purine receptors. Intravenous application of 100–200 μg adenosine/kg estimated fetal weight (without hydrops) into the umbilical vein resulted in cardioversion within 15–30 seconds.101 This was mediated by a slowing of conduction time through the atrioventricular node, and may have been accompanied by a direct action on sinus pacemaker cells. The induced atrioventricular block immediately breaks the atrioventricular and atrioventricular nodal reentry tachycardias, but rarely atrial ectopic tachycardias and flutter.17 Because the anatomical prerequisites for reentry tachycardia, the accessory atrioventricular conduction pathway and the triggering supraventricular extrasystoles, are not influenced, adenosine has no prophylactic effect, and recurrence of reentry tachycardia is generally observed in the fetus unless effective concentrations of another antiarrhythmic drug can prevent this. Therefore, the extreme short duration of action and the absent prophylactic effect limit the use of adenosine as a therapeutic agent in fetal tachycardia, yet it may have a role to aid differential diagnosis.17 Persistent tachycardia after an adenosine bolus injection strongly suggests an atrial rather than atrioventricular reentry tachycardia, thus allowing some modification of the therapeutic approach. In severely hydropic fetuses with supraventricular tachycardia refractory to transplacental therapy, the direct application of antiarrhythmic drugs to the fetus as an ultimate method may be successful.64,75–77,97,98 Especially in the group of hydropic fetuses with substantial grade of AV valvular insufficiency during tachycardia, a most advanced stage of ‘cardiomyopathy’ may be suggested, and rapid cardioversion should be achieved. This may be possible only by direct fetal drug administration in addition to maternal administration. Injections of digoxin,64,77,102,103 amiodarone,46,64,75,76,97,98 verapamil,58,64,77 propafenone,64,77 and adenosine101 have been reported to be performed into the umbilical vein,46,56,58,64,75–77,97,98,101 the fetal muscle,78,102,103 the fetal peritoneum,64,75,76,97 the amnion,46 and the fetal heart,58 or by a combination of different routes.46,56,58,64,75,76,97 Meanwhile, intravascular application into the umbilical vein seems to be the best way, allowing direct injection into the intravascular compartment of the fetus for the most rapid loading, and in addition the monitoring of fetal therapy, measuring concentrations of the antiarrhythmic agents in the fetal blood.64,75,76,97 Amiodarone appears to be the ideal drug for direct therapy of the hydropic fetus because of its extreme long-lasting elimination half-life of 1–3 months,64,75,76,97,98 which differs from all other antiarrhythmic drugs with half-lives between 2 and 18 hours. In these rare cases with severe hydrops, high concentrations of amiodarone in the fetal compartments may rapidly be achieved by repeated injections of amiodarone into the umbilical vein, even if placental passage of the drugs is markedly impaired or absent. In order to avoid the

Fetal tachyarrhythmia

dangerous bolus injection of amiodarone, which may cause severe bradycardia and cardiac arrest, repetitive doses of 2.5–5 mg amiodarone/kg estimated fetal weight (without hydrops) over 10 minutes should be administered several times a day.46,64,76 Concurrently, the mother should receive oral digoxin and amiodarone according to the dosage outlined above, to prevent the drug crossing the placenta from the fetus to the mother and to allow transplacental transfer of drug from the mother to the fetus when fetal circulatory recompensation and remission of hydrops are accompanied by a substantial improvement in the placental passage of amiodarone and digoxin.64,75,76 Permanent junctional reciprocating tachycardia is often resistant to antiarrhyhmic therapy. On the other hand, because of its relatively low tachycardic frequency between 180 and 220 beats/minute, hydrops rarely occurs in these fetuses. Class III antiarrhythmic drugs such as sotalol and amiodarone,21 but also flecainide, may the first choice for in utero treatment.1 In fetuses with persistent ventricular tachycardia, amiodarone was successfully administered transplacentally.29 Atternatively propranolol or flecainide may be the drug of first choice in fetal ventricular tachycardia, because underlying long-QT syndrome cannot be excluded prenatally and the sodium channel blocker flecainide effectively decreases the QTc interval in some sodium channel mutations, causing long-QT syndrome.104 In contrast, class III antiarrhythmic agents such as sotalol and amiodarone may induce serious proarrhythmic side-effects and death in fetuses with long-QT syndrome. Moreover, the proarrhythmic impact of sotalol appears to be more pronounced in the immature fetal and neonatal heart than it is in the adult heart.56

Effectiveness of fetal antiarrhythmic therapy The success of fetal antiarrhythmic treatment has been reviewed in detail by Simpson.56 In addition, Krapp and colleagues performed a meta-analysis comparing SVT with atrial flutter.2 Between 50 and 75% of non-hydropic fetuses with supraventricular tachycardia will convert to sinus rhythm with transplacental digoxin monotherapy, confirmed by a recently published series.59 Flecainide as second-line agent may be successful in almost all remaining cases.2,82 Alternatively, amiodarone is also mostly successful as a second-line drug.99,100 In hydropic fetuses, however, transplacental digoxin monotherapy is rarely effective – in only about 10–15%.2,58,59 Therefore, antiarrhythmic therapy in hydropic fetuses should start directly with flecainide or amiodarone. Both drugs can be combined with digoxin using the positive-inotropic effect of this drug. In rare cases of severely advanced cardiomyopathy, direct fetal therapy with amiodarone may be indicated.

477

With this regime, successful cardioversion to constant sinus rhythm may be obtained in up to 80% of hydropic fetuses.2 Thus, the arrhythmia-related mortality rate in non-hydropic fetuses with supraventricular tachycardia is nearly zero, whereas approximately 10–20% of hydropic fetuses have a fatal outcome.2,56,58,59 In their meta-analysis, Krapp and co-workers found no difference in the success rates of digoxin in SVT and in atrial flutter.2 Measurement of the ratio between ventriculoatrial and atrioventricular time interval appears to differentiate SVT with short ventriculoatrial time interval, typical of atrioventricular reentrant tachycardia through accessory atrioventricular connection, from SVT with long ventriculoatrial time interval, typical of permanent junctional reciprocating and atrial ectopic tachycardia. The latter are usually incessant and unresponsive to most antiarrhythmic drugs, and may recur postpartum. In these rare types of tachycardia, class III agents are potential drugs of first choice.19,21 The therapeutic protocol in fetuses with atrial flutter is similar to the treatment of supraventricular tachycardia. In non-hydropic fetuses with paroxysmal atrial flutter, careful monitoring may be justified. In sustained atrial flutter and in paroxysmal atrial flutter with hydrops, however, intravenous loading with digoxin within 48–72 hours is followed by an oral maintenance dosage. Fetal atrial flutter may be suppressed successfully by digoxin alone in only 50%,2 but this therapy may be useful for its positive inotropic and negative chronotropic properties.13 However, in the absence of rhythm control and 1:1 atrioventricular conduction, hydrops does not often occur in fetuses with atrial flutter,2,27 presumably because the lower ventricular rate in fetuses with atrial flutter with 2:1 and/or higher degree AV block may protect;2 furthermore, atrial flutter seems to start mostly after 30 weeks of gestation, i.e. 2.1 weeks later than SVT2 when the intrinsic properties of the fetal myocardium are more mature. One might speculate that the fetal atrium reaches a critical size at about 27–30 weeks’ gestation, allowing atrial macroreentry, favored as the most likely mechanism of atrial flutter.27 Therefore, a second-line therapy is uncommonly indicated in fetuses with atrial flutter and should be restricted to those fetuses that develop hydrops. In these cases, sotalol and amiodarone may be used as second-line therapy,27,105 because in adults and newborns, class III agents appear to be more effective than class Ic agents in atrial flutter.34,105

Postnatal follow-up in fetuses with tachyarrhythmias In fetuses with SVT that were treated in utero, the postnatal outcome may be complicated by a recurrence of tachyarrhythmia in approximately 50% of the neonates.64,68

478

Fetal Cardiology

Therefore, prophylactic continuation of antiarrhythmic treatment during the first 6–12 months of life is recommended to prevent recurrent attacks for all newborns or at least for all the newborns with postnatal recurrence of tachyarrhythmia.56 By maturation of the infant’s conduction tissue, annulus fibrosus, and myocardium, the probability of late recurrence decreases. Only in 10–20% of infants may the tachycardia persist beyond the first year of life.25,64 Whereas in atrioventricular reentry tachycardia via an accessory pathway spontaneous resolution of tachycardia occurs in the vast majority of cases, the recurrence rate appears to be higher in other types of tachycardia.22 Atrial flutter can be relatively easily controlled after birth by direct current cardioversion, by transvenous atrial overdrive pacing, and/or by digoxin alone or in combination with other antiarrhythmic drugs such as sotalol, amiodarone, dophitilide – a new class III agent – or flecainide.27,105 Also, spontaneous termination of atrial flutter may occur directly after birth. When a sinus rhythm has been established, a relapse of atrial flutter is so rare13 that a prophylactic arrhythmic treatment beyond the neonatal period is not justified.27,105 Fetal with atrial ectopic tachycardia and junctional reciprocal tachycardia often persist after birth, and longterm antiarrhythmic treatment may be necessary. The class III antiarrhythmic agents sotalol and amiodarone, but also flecainide, are preferred under these conditions. In neonates with persistent ventricular tachycardia, even in association with long-QT syndrome, the class Ic antiarrhythmic agents flecainide and propafenone may be administered. Also, propranolol may be given. Amiodarone and sotalol may be useful in therapy-refractory ventricular tachycardias. Some neonates with long-QT syndrome, however, require a temporary pacemaker in the neonatal period or a permanent pacemaker, shortening the QT interval and thereby reducing the risk of potentially lethal ventricular tachycardias.31,35 The administration of β-blockers, in addition, may reduce the sympathetic drive to the heart of infants with long-QT syndrome.31,35 The postnatal long-term outcomes of tachyarrhythmic fetuses have not been prospectively examined. For the vast majority, normal long-term development can be expected after in utero and/or postnatal cessation of tachyarrhythmia. Only a few cases with abnormal neurological outcome have been reported in the literature. Schade et al reviewed six cases from the literature and reported three additional observations.106 All nine fetuses developed hydrops, and were born in a hydropic condition in spite of successful drug-induced cardioversion to sinus rhythm. In some cases the postnatal course was complicated by severe prematurity, perinatal asphyxia, and/or postnatal recurrence of tachyarrhythmia. On the other hand, periventricular leukomalacia and hemorrhage diagnosed in some of these cases in utero or within a few hours after birth indicated that the occurrence was due to an intrauterine hypoxic– ischemic event. In addition, Oudijk et al retrospectively

studied the neurological function of 11 infants, aged 6 months to 12 years, who were prenatally treated for fetal tachyarrhythmia complicated by hydrops.107 They found that the majority of fetuses tolerate fetal tachyarrhythmia even if hydrops results.107 Also in this collective, abnormal neurological outcome was associated with prematurity and birth of a hydropic newborn. However, it may be speculated that in fetuses with paroxysmal tachyarrhythmia, abolition of cerebral autoregulation may lead to severe impairment of the maintenance of constant cerebral perfusion.106,107 Therefore, sudden changes in fetal heart rate may cause significant fluctuations of fetal arterial blood pressure, and consequently of cerebral perfusion leading to hypoxic– ischemic brain damage,106 especially in fetuses before 32 weeks’ gestation, where the autoregulatory range of systemic blood flow pressures is narrow and the periventricular vessels are highly vulnerable. Therefore, rapid and persistent control of fetal tachyarrhythmia may prevent fetal neurological damage. This is an additional rationale for intrauterine antiarrhythmic treatment of fetal tachyarrhythmia even if it occurs only intermittently, especially in immature and/or already hydropic fetuses.

References 1. Kothari DS, Skinner JR. Neonatal tachycardias: an update. Arch Dis Child Fetal Neonatal Ed 2006; 91: 136–44. 2. Krapp M, Kohl T, Simpson JM et al. Review of diagnosis, treatment, and outcome of fetal atrial flutter compared with supraventricular tachycardia. Heart 2003; 89: 913–17. 3. van Selm M, Kanhai HH, Gravenhorst JB. Maternal hydrops syndrome: a review. Obstet Gynecol Surv 1991; 46: 785–8. 4. Carbillon L, Oury JF, Guerin JM et al. Clinical biological features of Ballantyne syndrome and the role of placental hydrops. Obstet Gynecol Surv 1997; 552: 310–14. 5. Gherman RB, Incerpi MH, Wing DA, Goodwin TM. Ballantyne syndrome: is placental ischemia the etiology? J Matern Fetal Invest 1998; 7: 227–9. 6. Duthie SJ, Walkingshaw SA. Parvovirus associated fetal hydrops: reversal of pregnancy induced proteinuric hypertension by in utero fetal transfusion. Br J Obstet Gynaecol 1995; 102: 1011–13. 7. Midgley DY, Harding K. The mirror syndrome. Eur J Obstet Gynecol Reprod Biol 2000; 88: 201–2. 8. Goeden A, Worthington D. Spontaneous resolution of mirror syndrome. Obstet Gynecol 2005; 106: 1183–6. 9. Rein AJJT, O’Donnell C, Geva T et al. Use of tissue velocity imaging in the diagnosis of fetal cardiac arrhythmias. Circulation 2002; 106: 1827–33. 10. Thomas G. Tissue Doppler echocardiography – a case of right tool, wrong use. Cardiovasc Ultrasound 2004; 2: 12. 11. Stinstra J, Goldbach E, van Leeuwen P et al. Multicentre study of fetal cardiac time intervals using magnetocardiography. BJOG 2002; 109: 1235–43. 12. Kähler C, Grimm B, Schleussner E et al. The application of fetal magnetocardiography (FMCG) to investigate fetal

Fetal tachyarrhythmia

arrhyhthmias and congenital heart defects (CHD). Prenat Diagn 2001; 21: 176–82. 13. Hosono T, Kawamata K, Chiba Y et al. Prenatal diagnosis of long QT syndrome using magnetocardiography: a case report and review of the literature. Prenat Diagn 2002; 22: 198–200. 14. Schneider U, Haueisen J, Loeff M et al. Prenatal diagnosis of a long QT syndrome by fetal magnetocardiography in an unshielded bedside environment. Prenat Diagn 2005; 25: 704–8. 15. Wakai RT, Strasburger JF, Li Z et al. Magnetocardiographic rhythm patterns at initiation and termination of fetal supraventricular tachycardia. Circulation 2003; 107: 307–12. 16. Zhao H, Strasburger JF, Cuneo BF, Wakai RT. Fetal cardiac repolarization abnormalities. Am J Cardiol 2006; 98: 491–6. 17. Kleinman CS, Copel JA. Electrophysiological principles and fetal antiarrhythmic therapy. Ultrasound Obstet Gynecol 1991; 1: 286–97. 18. Wren C. Mechanisms of fetal tachycardia. Heart 1998; 79: 536–7. 19. Jaeggi E, Fouron JC, Fournier A et al. Ventriculo-atrial time interval measured on M mode echocardiography: a determining element in diagnosis, treatment, and prognosis of fetal supraventricular tachycardia. Heart 1998; 79: 582–7. 20. Fouron JC, Fournier A, Proulx F et al. Management of fetal tachyarrhythmia based on superior vena cava/aorta Doppler flow recordings. Heart 2003; 89: 1211–16. 21. Fouron JC. Fetal arrhythmias: the Saint-Justine hospital experience. Prenat Diagn 2004; 24: 1068–80. 22. Ko JK, Deal BJ, Strasburger JF et al. Supraventricular tachycardia mechanisms and their age distribution in pediatric patients. Am J Cardiol 1992; 69: 1028–32. 23. Weindling SN, Saul JP, Walsh EP. Efficiency and risks of medical therapy for supraventricular tachycardia in neonates and infants. Am Heart J 1996; 131: 66–72. 24. Kleinman CS, Copel JA, Nehgme R. The fetus with cardiac arrhythmia. In: Harrison MR, Evans MI, Adzick NS, Holzgreve W, eds. The Unborn Patient: the Art and Science of Fetal Therapy, 3rd edn. Philadelphia: Saunders, 2001: 417–41. 25. Naheed ZJ, Strasburger JF, Deal BJ et al. Fetal tachycardia: mechanisms and predictors of hydrops fetalis. J Am Coll Cardiol 1996; 27: 1736–40. 26. James TN. Cardiac conduction system: fetal and postnatal development. Am J Cardiol 1970; 25: 213–26. 27. Jaeggi E, Fouron JC, Drblik SP. Fetal atrial flutter: diagnosis, clinical features, treatment, and outcome. J Pediatr 1998; 132: 335–9. 28. Kleinman CS, Nehgme R, Copel JA. Fetal cardiac arrythmias: diagnosis and therapy. In: Creasy RK, Resnik R, eds. Maternal–Fetal Medicine, 4th edn. Philadelphia: Saunders, 1999: 301–18. 29. Schleich JM, du Haut, Cilly FB, Laurent MC, Almange C. Early prenatal management of a fetal ventricular tachycardia treated in utero by amiodarone with long term follow-up. Prenat Diagn 2000; 20: 449–52. 30. Gembruch U, Hansmann M, Redel DA et al. Fetal complete heart block: antenatal diagnosis, significance and management. Eur J Obstet Gynecol Reprod Biol 1989; 31: 9–22.

479

31. Hofbeck M, Ulmer H, Beinder E et al. Prenatal findings in patients with prolonged QT interval in the neonatal period. Heart 1997: 198–204. 32. Lin MT, Wu MH, Hsieh FJ et al. Long QT syndrome manifested as fetal ventricular tachycardia and intermittent AV block. Am J Perinatol 1998; 15: 145–7. 33. Yamada M, Nakazawa M, Momma K. Fetal ventricular tachycardia in long QT syndrome. Cardiol Young 1998; 8: 119–22. 34. Oudijk MA, Michon MM, Kleinman CS et al. Sotalol in the treatment of fetal dysrhythmias. Circulation 2000; 101: 2721–6. 35. Manning N, Anthony JP, Östman-Smith I et al. Prenatal diagnosis and successful preterm delivery of a fetus with long QT syndrome. Br J Obstet Gynaecol 2000; 107: 1049–51. 36. Lehnart SE, Ackerman MJ, Benson DW Jr et al. Inherited arrythmias: a National Heart, Lung, Blood Institute and Office of Rare Diseases workshop consensus report about the diagnosis, phenotyping, molecular mechanisms, and therapeutic approaches for primary cardiomyopathies of gene mutations affecting ion channel function. Circulation 2007; 116: 2325–45. 37. Buis-Liem TN, Ottenkamp J, Meerman RH et al. The occurrence of fetal supraventricular tachycardia and obstruction of the foramen ovale. Prenat Diagn 1987; 7: 425–31. 38. Stevens DC, Hilliard JK, Schreiner RL et al. Supraventricular tachycardia with edema, ascites, and hydrops in fetal sheep. Am J Obstet Gynecol 1982; 142: 316–22. 39. Nimrod C, Davies D, Harder J, Iwanicki S, Kondo S. Ultrasound evaluation of tachycardia-induced hydrops in the fetal lamb. Am J Obstet Gynecol 1987; 157: 655–9. 40. Gest AL, Hansen TN, Moise AA, Hartley CJ. Atrial tachycardia causes hydrops in fetal lambs. Am J Physiol 1990; 258: H1159–63. 41. Gest AL, Bair DK, Vander Straten MC. Thoracic duct lymph flow in fetal sheep with increased venous pressure from electrically induced tachycardia. Biol Neonate 1993; 64: 325–30. 42. Gest AL, Martin CG, Moise AA, Hansen TN. Reversal of venous blood flow with atrial tachycardia and hydrops in fetal sheep. Pediatr Res 1990; 28: 223–6. 43. Spinale FG, Tanaka R, Crawford FA, Zile MR. Changes in myocardial blood flow during development of and recovery from tachycardia-induced cardiomyopathy. Circulation 1992; 85: 717–29. 44. Spinale FG, Holzgrefe HH, Mukherjee R et al. LV and myocyte structure and function after early recovery from tachycardia-induced cardiomyopathy. Am J Physiol 1995; 268: H836–7. 45. Packer DL, Bardy GH, Worley SJ et al. Tachycardia-induced cardiomyopathy: a reversible form of left ventricular dysfunction. Am J Cardiol 1986; 57: 563–70. 46. Gembruch U, Redel DA, Bald R, Hansmann M. Longitudinal study in 18 cases of fetal supraventricular tachycardia: Doppler-echocardiographic findings and pathophysiological implications. Am Heart J 1993; 125: 1290–301. 47. Tomita M, Spinale FG, Crawford FA, Zile MR. Changes in left ventricular volume, mass, and function during the development and regression of supraventricular tachycardia-induced cardiomyopathy: disparity between recovery

480

Fetal Cardiology

of systolic versus diastolic function. Circulation 1991; 83: 635–44. 48. Gembruch U, Krapp M, Baumann P. Changes of venous blood flow velocity waveforms in fetuses with supraventricular tachycardia. Ultrasound Obstet Gynecol 1995; 5: 394–9. 49. Krapp M, Gembruch U, Baumann P. Venous blood flow pattern suggesting tachycardia-induced ‘cardiomyopathy’ in the fetus. Ultrasound Obstet Gynecol 1997; 10: 32–40. 50. Gembruch U, Krapp M, Germer U, Baumann P. Venous Doppler in the sonographic surveillance of fetuses with supraventricular tachycardia. Eur J Obstet Gynecol Reprod Biol 1999; 84: 187–92. 51. Petrikovsky B, Schneider E, Ovadia M. Natural history of hydrops resolution in fetuses with tachyarrhythmias diagnosed and treated in utero. Fetal Diagn Ther 1996; 11: 292–5. 52. Johnson P, Sharland G, Allan LD et al. Umbilical venous pressure in nonimmune hydrops fetalis: correlation with cardiac size. Am J Obstet Gynecol 1992; 167: 1309–13. 53. Morganroth J. Drug-induced early and late proarrhythmia. Cardiol Clin 1992; 10: 397–401. 54. Gembruch U, Somville T. Intrauterine Diagnostik und Therapie fetaler Arrhythmien. Gynäkologe 1995; 28: 329–45. 55. Ito S, Magee L, Smallhorn J. Drug therapy for fetal arrhythmias. Clin Perinatol 1994; 21: 543–72. 56. Simpson J. Fetal arrhythmia. In: Allan L, Hornberger L, Sharland G, eds. Textbook of Fetal Cardiology. London: Greenwich Medical Media, 2000: 423–51. 57. Napolitano C, Schwartz PJ, Brown AM et al. Evidence for a cardiac ion channel mutation underlying drug-induced QT prolongation and life-threatening arrhythmias. J Cardiovasc Electrophysiol 2000; 11: 691–6. 58. Simpson JM, Sharland GK. Fetal tachycardias: management and outcome of 127 consecutive cases. Heart 1998; 79: 576–81. 59. Jouannic JM, Delahaye S, Le Bidois J et al. Résultat de la prise en charge prénatale des fœtus avec tachycardie supraventriculaire. À propos d’une série consécutive de 66 cas. J Gynecol Obstet Biol Reprod (Paris) 2003; 32: 338–44. 60. Van den Berg P, Gembruch U, Schmidt S et al. Continuous fetal intrapartum monitoring in supraventricular tachycardia by atraumatic measurement of transcutaneous carbon dioxide tension. J Perinat Med 1989; 17: 371–4. 61. Afriat R, Bercau G, Bidat L et al. The relevance of intrapartum fetal pulse oximetry in the presence of fetal cardiac arrhythmia. Acta Obstet Gynecol Scand 1999; 78: 827–8. 62. Simpson LL, Marx GR, D’Alton ME. Supraventricular tachycardia in the fetus: conservative management in the absence of hemodynamic compromise. J Ultrasound Med 1997; 16: 459–64. 63. Simpson JM, Yates RW, Milburn A et al. Outcome of intermittent tachyarrhythmias in the fetus. Pediatr Cardiol 1997; 18: 78–83. 64. Hansmann M, Gembruch U, Bald R et al. Fetal tachyarrhythmias: transplacental and direct treatment of the fetus – a report of sixty fetuses. Ultrasound Obstet Gynecol 1991; 1: 162–70. 65. Kleinman CS, Copel JA, Weinstein EM et al. Treatment of fetal supraventricular tachyarrhythmias. J Clin Ultrasound 1985; 13: 265–73.

66. Stewart PA, Wladimiroff JW. Cardiac tachyarrhythmia in the fetus: diagnosis, treatment and prognosis. Fetal Ther 1987; 2: 7–16. 67. Maxwell DJ, Crawford DC, Curry PVM et al. Obstetric importance, diagnosis, and management of fetal tachycardias. Br Med J 1988; 297: 107–10. 68. Van Engelen AD, Weijtens O, Brenner JI et al. Management, outcome and follow-up of fetal tachycardia. J Am Coll Cardiol 1994; 24: 1371–5. 69. Frohn-Mulder IM, Stewart PA, Witsenburg M et al. The efficacy of flecainide versus digoxin in the management of fetal supraventricular tachycardia. Prenat Diagn 1995; 15: 1297–302. 70. Omar HA, Rhodes LA, Ramirez R et al. Alteration of human placental vascular tone by antiarrhythmic medication in vitro. J Cardiovasc Electrophysiol 1996; 7: 1197–203. 71. Schmolling J, Jung S, Reinsberg J, Schlebusch H. Diffusion characteristics of placental preparations affect the digoxin passage across the isolated placental lobule. Ther Drug Monit 1997; 19: 11–16. 72. Schmolling J, Renke K, Plath H et al. Transplazentare Passage der Antiarrhythmika Digoxin, Flecainid und Amiodaron im isolierten humanen Plazentalobulus. Perinat Med 1998; 10: 6–11. 73. Mimura S, Suzuki C, Yamazaki T. Transplacental passage of digoxin in a case of nonimmune hydrops. Clin Cardiol 1987; 10: 63–5. 74. Younis JS, Granat M. Insufficient transplacental digoxin transfer in severe hydrops fetalis. Am J Obstet Gynecol 1987; 157: 1268–9. 75. Gembruch U, Manz M, Bald R et al. Repeated intravascular treatment with amiodarone in a fetus with refractory supraventricular tachycardia and hydrops fetalis. Am Heart J 1989; 118: 1335–8. 76. Gembruch U, Hansmann M, Bald R. Direct intrauterine treatment of fetal tachyarrhythmia with severe hydrops fetalis by antiarrhythmic drugs. Fetal Ther 1988; 3: 210–15. 77. Gembruch U, Hansmann M, Redel DA, Bald R. Intrauterine therapy of fetal tachyarrhythmias: intraperitoneal administration of antiarrhythmic drugs to the fetus in fetal tachyarrhythmias with severe hydrops fetalis. J Perinat Med 1988; 16: 39–44. 78. Hallak M, Neerhof MG, Perry R et al. Fetal supraventricular tachycardia and hydrops fetalis: combined intensive, direct, and transplacental therapy. Obstet Gynecol 1991; 78: 523–5. 79. Azancot-Benisty A, Jacqz-Aigrain E, Guirgis NM et al. Clinical and pharmacologic study of fetal supraventricular tachyarrhythmias. J Pediatr 1992; 121: 608–13. 80. Weiner CP, Landas S, Persoon TJ. Digoxin-like immunoreactive substance in fetuses with and without cardiac pathology. Am J Obstet Gynecol 1987; 157: 368–71. 81. Schlebusch H, von Mende S, Grünn U et al. Determination of digoxin in the blood of pregnant women, fetuses and neonates before and during antiarrhythmic therapy, using four immunochemical methods. Eur J Clin Chem Clin Biochem 1991; 29: 57–66. 82. Krapp M, Baschat AA, Gembruch U et al. Flecainide in the intrauterine treatment of fetal supraventricular tachycardia. Ultrasound Obstet Gynecol 2002; 19: 158–64.

Fetal tachyarrhythmia

83. Allan LD, Chita SK, Sharland GK et al. Flecainide in the treatment of fetal tachycardias. Br Heart J 1991; 65: 46–8. 84. Echt DS, Liebson PR, Mitchell LB et al. Mortality and morbidity in patients receiving encainide, flecainide, or placebo. The Cardiac Arrhythmia Suppression Trial. N Engl J Med 1991; 324: 781–8. 85. Fish FA, Gillette PC, Benson DW Jr. Proarrhythmia, cardiac arrest and death in young patients receiving encainide and flecainide. J Am Coll Cardiol 1991; 18: 356–65. 86. Oudijk MA, Ruskamp JM, Ververs FF et al. Treatment of fetal tachycardia with sotalol: transplacental pharmacokinetics and pharmacodynamics. J Am Coll Cardiol 2003; 42: 765–70. 87. Pfammatter JP, Paul T. New antiarrhythmic drug in pediatric use: sotalol. Pediatr Cardiol 1997; 18: 28–34. 88. Sonesson SE, Fouron JC, Wesslen-Eriksson E et al. Foetal supraventricular tachycardia treated with sotalol. Acta Paediatr 1998; 87: 584–7. 89. Plomp TA, Vulsma T, de Vijlder JJ. Use of amiodarone during pregnancy. Eur J Obstet Gynecol Reprod Biol 1992; 43: 201–7. 90. Foster CJ, Love HG. Amiodarone in pregnancy. Case report and review of the literature. Int J Cardiol 1988; 20: 307–16. 91. Grosso S, Berardi R, Cioni M, Morgese G. Transient neonatal hypothyroidism after gestational exposure to amiodarone: a follow-up of two cases. J Endocrinol Invest 1998; 21: 699–702. 92. Laurent M, Betremieux P, Biron Y, LeHelloco A. Neonatal hypothyroidism after treatment by amiodarone during pregnancy. Am J Cardiol 1987; 60: 942. 93. Matsumara LK, Born D, Kunii IS et al. Outcome of thyroid function in newborns from mothers treated with amiodarone. Thyroid 1992; 2: 279–81. 94. Rovet JF, Ehrlich RM, Sorbara DL. Neurodevelopment in infants and preschool children with congenital hypothyroidism: etiological and treatment factors affecting outcome. J Pediatr Psychol 1992; 117: 187–213. 95. Haddow JE, Palomaki GE, Allan WC et al. Maternal thyroid deficiency during pregnancy and subsequent neurophysiological development of the child. N Engl J Med 1999; 341: 549–55.

481

96. Magee LA, Nulman I, Rovet JF, Koren G. Neurodeveolpment after in utero amiodarone exposure. Neurotoxicol Teratol 1999; 21: 261–5. 97. Flack NJ, Zosmer N, Bennett PR et al. Amiodarone given by three routes to terminate fetal atrial flutter associated with severe hydrops. Obstet Gynecol 1993; 82: 714–16. 98. Mangione R, Guyon F, Vergnaund A et al. Successful treatment of refractory supraventricular tachycardia by repeat intravascular injection of amiodarone in a fetus with hydrops. Eur J Obstet Gynecol Reprod Biol 1999; 86: 105–7. 99. Jouannic JM, Delahaye S, Fermont L et al. Fetal supraventricular tachycardia: a role for amiodarone as second-line therapy. Prenat Diagn 2003; 23: 152–6. 100. Strasburger JF, Cuneo BF, Michon MM et al. Amiodarone therapy for drug-refractory fetal tachycardia. Circulation 2004; 109: 375–9. 101. Kohl T, Tercanli S, Kececioglu D, Holzgreve W. Direct fetal administration of adenosine for the termination of incessant supraventricular tachycardia. Obstet Gynecol 1995; 85: 873–4. 102. Weiner CP, Thompson MB. Direct treatment of fetal supraventricular tachycardia after failed transplacental therapy. Am J Obstet Gynecol 1988; 150: 670–3. 103. Parilla BV, Strasburger JF, Socol ML. Fetal supraventricular tachycardia complicated by hydrops fetalis: a role for direct fetal intramuscular therapy. Am J Perinatol 1996; 13: 483–6. 104. Abriel H, Wehrens XHT, Benhorin J et al. Molecular pharmacology of the sodium channel mutation D1790G linked to the long-QT syndrome. Circulation 2000; 102: 921–5. 105. Lisowski LA, Verheijen PM, Benatar AA et al. Atrial flutter in the perinatal age group: diagnosis, management and outcome. J Am Coll Cardiol 2000; 35: 771–7. 106. Schade RP, Stoutenbeek P, de Vries LS, Meijboom EJ. Neurological morbidity after fetal supraventricular tachyarrhythmia. Ultrasound Obstet Gynecol 1999; 13: 43–7. 107. Oudijk MA, Gooskens RHJM, Stoutenbeek P et al. Neurological outcome of children who were treated for fetal tachycardia complicated by hydrops. Ultrasound Obstet Gynecol 2004; 24: 154–4.

34 Cardiac diseases in association with hydrops fetalis Ulrich Gembruch and Wolfgang Holzgreve

Definition The Greek–Latin term ‘hydrops fetalis’ is issued for pathologically increased fluid accumulations in fetal soft tissues and serous cavities.1 ‘Immune hydrops fetalis’ refers to those cases of hydrops that are caused by alloimmune hemolytic anemia in the presence of circulating maternal antibodies against fetal erythrocytes. If there is no evidence of blood group incompatibility (isoimmunization), the hydrops is characterized as being ‘non-immune hydrops fetalis’. The prenatal diagnosis of hydrops fetalis is achieved by ultrasound which demonstrates the skin edema and/or fluid accumulations in serous cavities of the fetus (abdominal ascites, pleural and/or pericardial effusion). Fetal hydrops is defined by the demonstration of fluid accumulations in at least two of these four fluid compartments. The placenta and the amniotic sac have been included as additional ‘fetal compartments’. Placentomegaly indicating hydrops placentae can be diagnosed if the placental thickness, from chorionic plate to base, increases by 4 cm in the third trimester. Although polyhydramnios defined by an amniotic fluid index above 24 cm is associated in 30–75% of cases with non-immune hydrops,2,3 the amniotic sac should not be considered as one of the compartments for the diagnosis of hydrops, because the pathophysiological mechanisms of the occurrence of polyhydramnios can differ from those leading to an increased rate of interstitial fluid accumulation causing hydrops fetalis and placentae. Furthermore, hydrops may even be associated with oligohydramnios, for example in some preterminal fetuses with Turner syndrome or in cases with intrauterine cytomegaloviral infection. Some series of non-immune hydrops also include cases of isolated pleural effusion, abdominal ascites, or generalized skin edema, because fluid accumulation in one site may represent an early stage of a disease that may lead to fluid accumulation in several sites at a more advanced stage, especially in diseases known to result in generalized hydrops fetalis. Non-immune hydrops can occur in association with an extremely wide variety of conditions. The list of disorders

comprises far more than 150 fetal conditions, but for many of these the association is based on only one or a few case reports1–10 and the causal relationship to the occurrence of hydrops is sometimes only speculative. The causes and associations of non-immune hydrops can be differentiated into those that are either more generalized, such as in hematological, infectious, or metabolic disorders, and chromosomal and non-chromosomal syndromes, or more focal, such as in cranial, cardiac, vascular, pulmonary, gastrointestinal, and renal disorders, or tumors.1–12 The incidence of non-immune hydrops varies between 1:1500 and 1:4000 deliveries.13,14 Because of the high intrauterine loss rate among hydropic fetuses, the true occurrence rate is most likely to be higher; in some instances, however, fetal hydrops can also regress during gestation. Especially in chromosomally abnormal fetuses there is a high spontaneous abortion rate between the first and second or third trimesters of pregnancy15 following an early appearance of hydrops in the late first and early second trimesters, whereas in chromosomally normal fetuses the hydrops often develops much later.3,16–18 The large variability in incidence and distribution of causative and associated diseases in the reported series of nonimmune hydrops is strongly influenced by factors such as the presence of a general ultrasound screening program, the gestational age windows of the screening examinations, and the specific referral patterns in any region. The ethnic background of the population also has an effect; for example, homozygous α-thalassemia is the most common cause of non-immune hydrops in Southeast Asia, as opposed to cardiac diseases in Caucasians. The same influences are relevant for the reported survival rates, which vary between 10 and 60% in the literature.1–8,16,19,20 Although hydrops fetalis in its various manifestations is easily detected by ultrasound (screening), the enormous spectrum of etiologies and associations requires highly specialized knowledge in order to apply the non-invasive and invasive diagnostic and sometimes therapeutic steps in a systematic and effective way. Owing to the time pressure that is often present with progressing severity of

484

Fetal Cardiology

hydrops, non-immune hydrops is a situation in prenatal medicine where good cooperation between screening investigators and specialized centers not only allows a realistic assessment of the prognosis and subsequent proper counseling about recurrence risks, but also can often mean the difference between life and death for an affected child in utero. In this context, one of the most important diagnostic steps is a detailed and extensive anatomical and functional assessment of the cardiovascular system, because some of the underlying conditions are primary cardiac diseases and others may secondarily result in compromised fetal cardiac function, congestive heart failure, and hydrops (Table 34.1).

Table 34.1

Continued

Myocarditis Parvovirus B19 Adenovirus Coxsackievirus Chagas disease Myocardial infarction Arrhythmias Tachyarrhythmias supraventricular tachycardia atrial flutter ventricular tachycardia

Table 34.1 Cardiac and extracardiac diseases causing congestive heart failure and hydrops. Modified from references 1, 4, 5, 8–10, 21–23 Cardiac abnormalities Structural defects

Bradyarrhythmias sinus bradycardia complete heart block combined with atrial isomerism and structural defect (see above) in the presence of maternal autoimmune antibodies (anti-SSA, anti-SSB) Prolonged QT syndrome

Atrioventricular septal defect, isolated or in association with Down syndrome Atrioventricular septal defect in combination with heterotaxia syndrome (situs ambiguus, atrial isomerism) and bradyarrhythmia Tricuspid dysplasia and Ebstein’s anomaly Severe obstruction of the right ventricular outflow tract by pulmonary stenosis, pulmonary atresia, and premature obstruction of the ductus arteriosus (by indomethacin or spontaneously)

Idiopathic arterial calcification Extracardiac diseases possibly causing congestive heart failure and hydrops Tumors and vascular disorders Teratoma (sacrococcygeal, mediastinal, intracerebral, pericardial) Mediastinal fibrosarcoma Hamartoma Arteriovenous malformation of different localization

Absent pulmonary valve syndrome (mostly combined with tetralogy of Fallot and/or agenesis of the ductus arteriosus)

Vein of Galen aneurysm

Aortico–left ventricular tunnel

Diffuse neonatal hemangiomatosis

Truncus arteriosus communis with insufficiency of the truncal valve

Angio-osteohypertrophy syndrome (Klippel–Trenaunay–Weber syndrome)

Premature closure of the foramen ovale

Chorioangioma

Severe obstruction of the left ventricular outflow tract by aortic stenosis and atresia leading to interatrial left-toright shunt or premature closure of the foramen ovale

Chorioangioma as part of Wiedemann–Beckwith syndrome

Cardiac tumors

Fetal hemangioma (liver, neck, chest)

Diffuse placental chorioangiomatosis Hematological disorders causing fetal anemia

Rhabdomyoma, often as part of tuberous sclerosis Hemangioma

Excessive erythrocyte loss Intrinsic hemolysis or abnormal hemoglobins

Hamartoma

α-thalassemia

Pericardial teratoma

erythrocyte enzyme disorders: glucose-6-phosphate dehydrogenase deficiency, pyruvate kinase deficiency, glucose phosphate isomerase deficiency, congenital erythropoietic porphyria

Cardiomyopathy Isolated non-compaction of the ventricular myocardium Dilated (cardiac glycogenoses)

erythrocyte membrane disorders: abnormalities of spectrin

Restrictive (Continued )

(Continued )

Cardiac diseases in association with hydrops fetalis

Table 34.1

Continued

Extrinsic hemolysis Kasabach–Merritt sequence (arteriovenous malformations and tumors) Hemorrhage fetomaternal hemorrhage fetal closed-space hemorrhage (bowel, intracranial, tumor) Erythrocyte underproduction Liver and bone marrow replacement syndromes transient myeloproliferative disorder (e.g. in fetuses with trisomy 21) congenital leukemia Red cell aplasia and dyserythropoiesis parvovirus B19 infection Blackfan–Diamond syndrome dyserythropoietic anemia Metabolic disorders Lysosomal storage diseases mucopolysaccharidoses oligosaccharidoses lysosomal transport defects sphingolipidoses mucolipidoses Cardiac glycogen storage disease Carnitine deficiency Congenital disorders of glycosylation (CDG) Hereditary hemochromatosis Congenital myotonic dystrophy Cardiovascular disorders in twin pregnancies Twin-to-twin transfusion syndrome Twin reversed arterial perfusion (TRAP) sequence with a parasitic acardiac twin Conjoined twins

Pathophysiology of hydrops fetalis There is a constant exchange of extracellular fluids between the intravascular and interstitial compartments. The interstitial space can be subdivided into transcellular and lymphatic fluid components. Under normal conditions, the differences of the hydrostatic and colloid oncotic pressure in the intracapillary and interstitial fluid cause a fluid shift into the interstitial compartment on the arteriolar side of the capillary bed and a flux back

485

into the intravascular compartment on the venule end. The regulation of fluid movement across the capillary membrane is expressed by the Starling equation:24,25 Jv = CFC [(Pc − Pi) − σ(πc − πi)] The total flow of fluid (Jv) across the capillary membrane is influenced by the intracapillary hydrostatic pressure (Pc) and the colloid oncotic pressure of the interstitial fluid (πi), which are the two driving forces for capillary ultrafiltration into the interstitial fluid. The interstitial hydrostatic pressure (tissue turgor tension) (Pi) and the plasma colloid osmotic pressure (πc) are the opposite Starling forces. The fluid filtration coefficient (CFC) represents the net amount of fluid crossing the capillary membrane for a given imbalance of the Starling forces, and is affected by both the conductance of the capillary wall and the ease of fluid movement in the interstitial space. Changes of the permeability of a particular capillary bed change the reflection coefficient for oncotically active solute (σ), and, therefore, the influence of a given colloid osmotic pressure differs between both compartments. Thus, an increased capillary permeability allows water and protein to leak more easily into the interstitial space of the fetus and/or placenta. Abundant fluid in the interstitial compartment is returned to the vascular space via the lymphatic system. The capacity of lymphatic drainage of the interstitial space is determined by the outflow pressure for the lymphatic flow, which is the central venous pressure.24–26 Fetal interstitial fluid accumulation and hydrops generally originate from an imbalance between the higher rate of interstitial fluid formation by capillary ultrafiltration and the lower rate of interstitial fluid return through the venule side of the capillary bed and the lymphatic system back to the circulation. The six classically postulated mechanisms are: (1) primary myocardial failure with decreased cardiac output; (2) high cardiac output failure; (3) decreased colloid oncotic plasma pressure; (4) increased capillary permeability especially secondary to tissue hypoxia or sepsis; (5) obstruction of venous flow; and (6) obstruction of lymphatic flow.8,24,27 Data on fluid distribution and regulation of the intercompartmental flow are almost exclusively based on studies in the sheep, and allow only a cautious extrapolation to the human fetus. Gestational age-dependent maturation of the vascular bed and the mechanisms of its neuronal and humoral government regulating the body fluid distribution by changes of the Starling forces and the lymphatic drainage are relatively unclear in the human fetus. Considering that roughly 30% of the fetal plasma volume is outside the fetal body, the interstitial space of the fetus seems to be substantially larger in comparison to the postpartum situation, with an elevated ratio of about 4.4:1 compared to 3:1 for the adult sheep. According to fetal lamb studies, several factors facilitate transcapillary fluid filtration into the interstitial space during fetal compared to adult life.24,25 First, the permeability of the

486

Fetal Cardiology

capillary membrane for plasma proteins is 15 times higher than in adult sheep.28 This decreases the reflection coefficient for the colloids (s), impairing the effect of the colloid oncotic pressure difference between fetal vascular and interstitial space and, therefore, the driving force of fluid flux back into the vascular space. On the other hand, this lower effect of the colloid oncotic pressure difference on the transcapillary fluid flux in the fetus seems to provide a relative protection for the fetus against the negative effect of hypoproteinemia. Second, the fluid filtration coefficient (CFC) of the capillary bed is roughly five-fold higher in fetal than in adult sheep, facilitating and increasing the fluid flow for any given imbalance in the Starling forces.29 Third, the compliance of the interstitial space of the fetus is higher than in the adult sheep. Therefore, a greater accumulation of interstitial fluid is necessary in the fetus to increase the interstitial hydrostatic pressure. Expressed in another way, smaller changes in the hydrostatic pressure difference between capillary and interstitial space (ΔP = Pc − Pi) lead to a larger fluid flow across the capillary membrane.29 The increase of both the fluid filtration coefficient and the compliance of the interstitial space facilitates the transcapillary fluid flow and allows the fetus a more rapid restoration of blood volume after acute blood volume loss and expansion caused by acute hemorrhage and infusion, respectively.24,26 Changes of the transplacental fluid transfer seem to be without relevance in this context. The lymphatic system allows the return of fluid and the oncotic active proteins to the vascular space. Studies in sheep have elucidated the importance of this system for the fluid balance between the intravascular and interstitial compartments.26 Basal lymph flow rates in the fetus are substantially higher than postpartum, reflecting the enhanced transcapillary flow into the interstitial space.26,30 Very important for the pathogenesis of hydrops is the observation that the lymphatic drainage of the interstitial space is not impeded at a normal outflow pressure for the lymphatic system, which represents the central venous pressure.26,30,31 The lymphatic flow rate in the thoracic duct of fetal sheep is four- to five-fold higher than in adult sheep.26,30 If the venous pressure increases over the physiological value of 3–4 mmHg the lymphatic flow rate in the fetus is substantially reduced, and ceases at only 16 mmHg.26,30,31 In adult sheep the flow is relatively constant up to 8 mmHg and only stops at 26 mmHg. This delicate relationship between venous pressure and lymphatic flow represents one important pathophysiological mechanism for the occurrence of fetal hydrops.26,30,31 Elevation of the systemic venous pressure due to circulatory dysfunction seems to be the most important pathogenic mechanism for the development of hydrops, even if increased capillary permeability, decreased plasma oncotic pressure, and obstruction of lymphatic fluid return may also contribute to the fluid accumulation in fetal soft tissue and serous cavities.24 An increase of venous pressure may

be considered as a consequence of homeostatic mechanisms serving to preserve adequate systemic delivery of metabolic substrate when cardiocirculatory function is impaired. Decreased intravascular volume may occur due to blood loss as a result of fetofetal or fetomaternal blood transfusion or of fetal hemorrhage. It may also be due to a relative deficiency of intravascular volume as a result of an increased fluid loss into the interstitial space by increased capillary permeability or decreased plasma oncotic pressure. Tissue hypoxia may cause capillary damage. Hepatocellular injury can lead to hypoproteinemia. Furthermore, local hypoxia may lead to tissue accumulation of lactate, a powerful osmotic agent. Diminished venous return may be caused by local obstruction of venous flow, but more often by a decreased ventricular compliance and by significant shortening of the diastolic period in tachyarrhythmia. Inadequate oxygen supply during the advanced stage of various alterations may decrease the ventricular compliance, possibly also in tachyarrhythmia. This may lead to inadequate myocardial blood flow, because of the substantially shortened duration of diastole. In consequence, diastolic cardiac dysfunction and decreased venous return may result in low cardiac output due to inadequate ventricular filling. Primary myocardial dysfunction may also impair stroke volume and cardiac output, for example in cases of myocarditis, myocardial infarction, or myocardial hypoxia. Furthermore, systolic cardiac dysfunction can occur when right and left ventricular afterload are increased; this is very poorly tolerated by the fetus. In this group, obstructive lesions of the semilunar valves, constriction of the ductus arteriosus, and arterial hypertension in the recipient fetus of a fetofetal transfusion syndrome and in cases with hypoxemia-induced arterial blood flow redistribution with peripheral vasoconstriction can be included. Because of the low heart rate the cardiac output may be diminished in fetuses with complete heart block, even if the stroke volume is compensatorily increased. Furthermore, diminished supply of organs with oxygen and nutrients may be compensated for by an increase of cardiac output. The product of stroke volume and heart rate lead to high-output cardiac failure, for example in cases of decreased hemoglobin saturation and/or hemoglobin concentration, in cases of maldistribution of flow, such as in association with tumors and arteriovenous malformations, or in cases of metabolic disorders such as thyrotoxicosis.23,27 In all of these pathological conditions, local and systemic compensatory mechanisms become effective and may help the fetus to survive. On the other hand, some of these mechanisms at the same time increase the imbalance of Starling forces, resulting in enhanced interstitial fluid accumulation. In advanced disease an increase of hydrops and fetal deterioration may occur, when the compensatory mechanisms become exhausted, resulting in further elevation of venous pressure by secondary volume overload or myocardial dysfunction with

Cardiac diseases in association with hydrops fetalis

occurrence of atrioventricular valve regurgitation.23,27 Compensatory mechanisms are:23,27 1. The opening of additional capillaries, which may increase the exchange area between the vascular and interstitial space, improving the extraction of oxygen and nutrients; 2. The redistribution of arterial and venous blood flow by selective vasoconstriction, maintaining the supply of heart, brain, and adrenal glands with oxygen and nutrients at the expense of other organs. This is triggered by chemoreceptors followed by neuronal and endocrinemediated stimulation of vasoconstriction and by local autoregulation; 3. The increase of blood flow and better supply of oxygen and nutrients to organs, which may result from elevation of the cardiac output. This can be achieved by increase of the heart rate and/or systolic ventricular function as well as by an increase of intravascular blood volume and/or venous pressure. An increase of intravascular volume may be achieved by fluid retention through the kidneys and/or by transcapillary resorption, especially in cases of lowered intravascular volume and venous pressure. Elevation of venous pressure by volume increase or increase of venous tone leads to an improved ventricular filling. However, because of the diminished preload reserve and Frank–Starling mechanism of the fetal heart, which seems to operate near the top of its ventricular function curve, it seems to be difficult for the fetus to convert any increase of myocardial filling into an appropriate augmentation of cardiac output, which limits this compensatory mechanism. Above this limit, atrial distension may lead to a release of atrial natriuretic factor. In human fetuses increased concentrations of atrial natriuretic factor were demonstrated by cordocentesis in conditions associated with atrial distension, for example in anemic and acidemic fetuses, in fetuses with congestive heart failure, in the recipient fetus in twin–twin transfusion syndrome, and in fetuses after volume expansion by intrauterine blood transfusion.32,33 Regulation of fluid distribution in the fetus may also be modified by the placental circulation, receiving about 40% of the fetal combined cardiac output. Increased hydrostatic pressure or decreased plasma colloid oncotic pressure in the fetus enhance the transcapillary fluid filtration into the interstitial compartment of the fetus and the placenta; in the placenta this would also drive fluid into the maternal vascular space, at least partially counteracting the fluid retention by the fetus. On the other hand, studies in fetal sheep show that the fetal placental postcapillary resistance is quite large,34 while the somatic postcapillary resistance is very low.35 Therefore, placental capillary pressure may be protected from an elevation of the venous pressure by a small decrease in placental flow, whereas somatic capillary pressure tightly reflects the systemic venous pressure.36

487

An elevation of venous pressure cannot result in an increased compensatory fetal fluid flux into the mother.

Diagnostic approach in the hydropic fetus Hydrops is an ‘emergency’ condition in fetal life that requires fast diagnosis and therapy. Diagnostic methods and approaches are listed in Table 34.2. Non-invasive examination by maternal blood sampling includes the antibody-screening test for exclusion of alloimmune hemolytic anemia, the Kleihauer–Betke stain for hemoglobin F (HbF) cells for diagnosis of fetomaternal hemorrhage, and a search for infectious diseases, such as parvovirus B19, cytomegalovirus, syphilis, toxoplasmosis, and adeno-, coxsackie-, and herpes simplex virus infections. If family history or actual findings indicate a special etiology, more sophisticated investigations may be necessary: hemoglobin electrophoresis, determination of enzyme activity, or exclusion of autoimmune antibodies (anti-SSA, anti-SSB). Sonography, echocardiography, and Doppler sonography as non-invasive and repeatable methods are the most important tools for the diagnosis and surveillance of cases with non-immune fetal hydrops. The distribution and extent of fluid accumulation, amount of amniotic fluid, and placental structure should be assessed. Sonographic detection of malformations and hygromata colli is suggestive of specific chromosomal and non-chromosomal disorders. Specific causes of hydrops may be detected in other pathological conditions such as lung masses, tumors, heart defects, arteriovenous malformation, gastrointestinal and renal diseases, fetofetal transfusion, parasitic twins, and sustained arrhythmia. Echocardiography and Doppler sonography of arterial and venous vessels may demonstrate high cardiac output in cases of anemia and arteriovenous malformations. Echocardiography and Doppler sonography of the venous system may indicate or exclude cardiac failure, which may be the primary or the secondary cause in the advanced stage of diseases other than anemia. Cardiac dilatation, atrioventricular (AV) valve incompetence, and increased pulsatility of venous blood flow velocity waveforms are indicative of fetal congestive heart failure. Fetal heart rate monitoring should repeatedly be performed for longer periods during 24 hours to exclude paroxysmal tachyarrhythmia. For adequate clinical management, rapid karyotyping is mandatory in most situations of hydrops fetalis independent of the underlying disease. The method of sampling depends on the gestational age and the necessity for other invasive procedures. The alternative methods are chorionic villus sampling or placentesis and fetal blood sampling. The last has the advantage of an additional exclusion of

488

Fetal Cardiology

Table 34.2 Diagnostic approach in cases of non-immune hydrops fetalis. Modified from references 1, 5, 8–10, 22, 23

Table 34.2

Continued

Cardiotocography Fetal condition

Maternal history

Paroxysmal tachyarrhythmia

Ethnic background Diseases: anemia, infection, diabetes mellitus, connective tissue disease

Fetal anemia Amniocentesis*

Consanguinity

Fetal karyotype

Family history

α-Fetoprotein Antigen tests by PCR and amniotic fluid culture for syphilis, cytomegalovirus, toxoplasmosis and other infections

Obstetric history Previous affected sibling Spontaneous abortions/stillbirths

Metabolic testing of amniotic fluid

Pregnancy history

Amniotic cell morphology and culture for metabolic disorders

Gravida/para Gestational age

Fetal blood sampling*

Multiple gestations

Fetal karyotype

Infectious diseases

Fetal complete blood count with indexes

Medication

Peripheral blood smear

Maternal studies

Reticulocyte count

Blood group typing

White blood cell differentiation

Indirect Coombs’ testing Complete blood count with indexes such as mean corpuscular volume

Blood group typing Direct Coombs’ test

Peripheral blood smear for erythrocyte morphology

Fetal albumin

Kleihauer–Betke stain

Fetal liver function tests

Syphilis, parvovirus B19, cytomegalovirus, toxoplasmosis and other infections

Fetal antigen-specific IgM, IgA, and PCR for infectious causes

Additional examinations (as indicated)

Additional examinations (as indicated)

Hemoglobin electrophoresis

Hemoglobin electrophoresis

Maternal blood chemistry

Osmotic fragility

G6PD, pyruvate kinase deficiency screening

Heinz body preparation

Fetal studies

Erythrocyte enzyme determination

Ultrasound

Special testing of erythrocyte membrane skeleton proteins

Two- and three-dimensional ultrasound Fetal echocardiography

Metabolic testing

Two-dimensional ultrasound (dimension, cardiac structures, rhythm) Pulsed and color flow Doppler (intracardiac and postcardiac for detection of flow abnomalities such as valve regurgitation, stenosis, ductus arteriosus constriction, shunts, high cardiac output) M-mode (dimension, rhythm, contractility) Venous Doppler studies (cardiac function, rhythm) Arterial Doppler (high cardiac output, arteriovenous malformations) (Continued )

Chorionic villus sampling* Fetal karyotype Additional examinations (as indicated) Morphological examination for storage diseases Metabolic testing *If there is an index case or specific symptoms, molecular genetic DNA-based analysis can be performed from chorionic villus sampling, fetal blood, and amniotic cells, e.g. for diagnosing storage diseases, some inborn hematological disorders, skeletal dysplasia, or congenital myotonic dystrophy.

Cardiac diseases in association with hydrops fetalis

many other potential disorders, especially anemia and infection. Therefore, fetal blood sampling is usually the preferred method for rapid karyotyping. Moreover, its additional advantages are exclusion and differentiation of anemia (reticulocyte count, detection of hemoglobinopathies, signs of hemolytic anemia), thrombocytopenia, and leukemia, and detection of liver dysfunction and infectious diseases. In special cases some biochemical tests and DNA analysis are also possible from fetal blood for various diseases such as infections, metabolic disorders, and congenital myotonic dystrophy. Measurement of umbilical venous pressure during fetal blood sampling may assess cardiac function and differentiate between cardiac and noncardiac causes of hydrops.37,38 Nevertheless, non-invasive measurement of cardiac size and/or venous blood flow pattern shows a good correlation with cardiac function and systemic central venous pressure.37,39–41 The first diagnosis of hydrops fetalis is established sonographically. Many causes of hydrops such as sustained arrhythmias, tumors of the fetus, lung masses, skeletal dysplasia, or twin–twin transfusion syndrome can often be detected during this first scan.9 Other findings such as cystic nuchal hygroma, indicating chromosomal disorders or rare syndromes, may lead to further diagnostic procedures such as rapid karyotyping. Furthermore, primary ultrasound has to define the exact distribution of the fluid accumulation, such as skin edema, ascites, uni- or bilateral pleural effusion, pericardial effusion, hydrops placentae, and amount of amniotic fluid. A semiquantification of these fluid accumulations is the basis for further monitoring and sometimes for prognostic assessment. Classification into mild, moderate, and severe is helpful, and can additionally be specified by measurements such as the prehepatic ascites (diameter of fluid between abdominal wall and liver), the diameter between lung and lateral thoracic wall, the systolic and diastolic diameters of pericardial effusion, the parietal or frontal and abdominal skin thickness, the placental thickness, and the amniotic fluid index or maximal vertical amniotic fluid pocket. The presence and extent of mediastinal shifting and biventricular heart diameter should always be documented. The distribution of fluid accumulations may be helpful for diagnosis of the underlying cause. In hydropic fetuses with anemia, tachyarrhythmia, and complete heart block, signs such as ascites, skin edema, hydrops placentae, and polyhydramnios are usually present, but pleural effusion can be noted only in more advanced stages of the disease.16,42–45 In other underlying diseases such as alteration of lymphatic drainage and chromosomal anomalies, pleural effusion is predominant and other fluid accumulations are less often seen.16 If the first sonographic examination does not reveal the etiology of hydrops, a high-level ultrasound examination should immediately be arranged, including fetal echocardiography and Doppler examination of the arterial and venous vessels. On the one hand, this leads to the detection

489

of anomalies causing hydrops that are difficult to diagnose with the usual scan. On the other hand, anomalies such as associated cardiac defects can be detected, which can again lead to the diagnosis of a chromosomal aberration or a rare syndrome. The peripheral arterial and venous Doppler examination may reveal a high cardiac output by demonstration of increased blood flow velocities and/or a cardiac dysfunction with increased pulsatility of the venous Doppler. Hydrops without evidence for anemia or other causes can be due to a spontaneous remission of anemia subsequent to an infection with parvovirus B19 and more rarely to fetomaternal transfusion or to fetal bleeding. In these cases, mild cardiomegaly, polyhydramnios, thickened placenta, and an increased number of reticulocytes and normoblasts in fetal blood may be demonstrated. An examination should also be performed in amniotic fluid and fetal blood, including polymerase chain reaction (PCR) as well as assessment of specific immunoglobulin M (IgM). Paroxysmal supraventricular tachycardia should always be taken into consideration. Diagnosis of paroxysmal tachyarrhythmia is best established by repeated sonographic heart rate monitoring, or more easily by long-term cardiotocography several times per day. The diagnosis is confirmed by fetal echocardiography, which also allows identification of the specific type of arrhythmia. In these cases cardiomegaly, polyhydramnios, and hydrops placentae may be present. In some cases increased pulsatility of venous Doppler flow velocity waveforms during the periods of sinus rhythm can be seen, indicating tachycardia-induced ‘cardiomyopathy’ due to repeated tachycardia. The prognosis in cases of non-immune hydrops mainly depends on the underlying disease. Further prognostic indicators are: distribution as well as extent of fluid accumulation and edema, degree of cardiac decompensation, and gestational age at the time of manifestation. Therefore, correct assignment of the precise diagnosis in each case of non-immune hydrops is mandatory for adequate management of the pregnancy and counseling of the parents. It seems to be very difficult, however, to exclude all potential causes in the large group of heterogeneous disorders when investigating individual cases. Therefore, the following factors should be considered before the diagnostic technique is chosen: 1. Invasive versus non-invasive approach (non-invasive tests such as sonography and maternal blood sampling harbor no risks for the fetus and, therefore, should usually be applied first); 2. Incidence of the potential disorders at the specific gestational age; 3. Necessity for prompt exclusion of fetal anemia causing non-immune hydrops, which is life-threatening and can usually be successfully treated if diagnosis is achieved in time.

490

Fetal Cardiology

The detailed examination of the cardiovascular system is a central part of the diagnostic approach in fetuses with hydrops fetalis. Structural cardiac defects should be excluded, because they can be the primary cause of the hydrops or may be coincident in some underlying disorders such as chromosomal aberrations. Furthermore, in other underlying diseases hydrops may result from another pathomechanism, but additional congestive heart failure appears in the advanced stage of the disease. Therefore, ultrasonographic examination should be addressed to the detection of signs for fetal congestive heart failure, which means inadequate tissue perfusion. In particular, if a cardiac disease is the primary cause of hydrops, the signs of congestive heart failure may occur before the clinical state of hydrops. Besides fetal hydrops cardiomegaly, abnormal cardiac function and abnormal venous Doppler may be highly indicative of fetal congestive heart failure, resulting in the construction of a cardiovascular profile score for diagnosis of fetal cardiac failure.46–48 However, the demonstration of arterial blood flow redistribution by Doppler sonography may result from fetal congestive heart failure, but more probably suggests hypoxemia and acidosis in the advanced stage of various diseases such as intrauterine growth restriction and anemia. The demonstration of more than one of these signs makes the diagnosis of fetal congestive heart failure most likely. Calculations of the cardiac output (CO) by two-dimensional echocardiography (e.g. by Simpson’s rule, and by Doppler echocardiographic measurement following the equation CO = A × VTI × HR, where A is the area of the artery, VTI the velocity time integral, and HR the heart rate) are difficult and time-consuming and have too high a variability in clinical practice at present. Further development of ultrasound equipment in the future may eliminate these limitations. The pathophysiology of this hypoxemiainduced redistribution of cardiac output is covered in detail in Chapter 37. The more special sonographic parameters for fetal congestive heart failure are discussed below.

Cardiomegaly Cardiomegaly means the pathological increase of cardiac size by either a more symmetric enlargement of all cardiac chambers or an asymmetric widening of isolated cardiac cavities. Because of the parallel circuit of the fetal circulation, the right atrium is the most common cardiac cavity to express enlargement. This may result from increased right ventricular preload by volume overload, from a relatively restrictive foramen ovale and/or increased left atrial pressure, from a compromised right ventricular contractility and performance, and/or from an increased right ventricular afterload.24,46,47 Although right atrial distension may be an early sign of fetal congestive heart failure, right atrial enlargement may also be obvious in states of intracardiac compensatory redistribution of

blood flow, maintaining adequate combined cardiac output; this occurs in many congenital heart defects during fetal life.46,47 Therefore, diagnosis of congestive heart failure should always be supported by the demonstration of other criteria. Besides the increase of the absolute cardiac parameters compared to the reported gestational agedependent normative values,49–55 advancing enlargement of the cardiac cavities can be more easily documented by measuring the cardiothoracic ratios (Figure 34.1). These different ratios of the cardiac to the thoracic circumference or area remain relatively independent of gestational age during the second and third trimesters when the heart occupies one-third of the thorax and the cardiothoracic circumference ratio is around 0.5.53–55 Between 10 and 17 weeks of gestation, however, a more significant increase of all the cardiothoracic ratios has been reported, and the thoracic area occupied by the heart increases from one-fifth to one-quarter.52 In contrast to the absolute biometric measurements of the single cardiac cavities, the cardiothoracic ratios become abnormal only after marked chamber enlargement is manifested. Therefore, these parameters are highly specific, but not sensitive in the diagnosis of fetal congestive heart failure.46,47 By the measurement of umbilical venous pressure, evaluation of the cardiothoracic ratio showing cardiomegaly has been validated for the assessment of cardiac function in cases of non-immune hydrops in the human fetus.37 Therefore, it is not surprising that cardiomegaly is a good prognostic factor in fetuses with non-immune hydrops, as it is relatively independent of the etiology, indicating a distinct compromise of cardiac function and an unfavorable prognosis.56

Systolic and diastolic ventricular function The systolic myocardial function can best be estimated by determining the ventricular fractional shortening using M-mode echocardiography. The normal ranges for left and right ventricular fractional shortening (FS) are between 28 and 40%, calculated by the following equation: FS = (EDD − ESD)/EDD, where EDD is the enddiastolic diameter and ESD the end-systolic diameter.46,47 Myocardial compromise may result in reduced ventricular shortening and increased ventricular work in more pronounced fractional shortening. Increased afterload and increased ventricular dilatation may also decrease the ventricular fractional shortening. In addition, an unfavorable position of the fetal heart makes it impossible to obtain the standard position for exact measurement with the M-mode line perpendicular to the interventricular septum. Therefore, in prenatal diagnosis, determination of ventricular fractional shortening is not commonly used

Cardiac diseases in association with hydrops fetalis

(a)

491

(b)

(c)

Figure 34.1 (a) Cardiomegaly is shown in a fetus with Ebstein’s anomaly at 24 + 4 weeks’ gestation. The cardiothoracic circumference (0.71) and the cardiothoracic area ratio (0.46) are markedly increased. (b) Severe holosystolic tricuspid insufficiency is demonstrated in a fetus with Ebstein’s anomaly at 24 + 4 weeks’ gestation. By continuous wave Doppler, a jet velocity of 3.63 m/s was measured. Using the Bernoulli equation, a pressure gradient of 52.7 mmHg was calculated. (c) Color Doppler M-mode echocardiography shows the holosystolic character of the tricuspid regurgitation (blue). The systolic right ventricular inflow is colored red–yellow.

for the estimation of myocardial dysfunction and diagnosis of congestive heart failure. The evaluation of the ventricular ejection force is also rarely used to assess fetal systolic myocardial function,57,58 although this Doppler index appears to be less influenced by changes in ventricular preload and afterload. The velocity ejection force (VEF) is calculated using the following equation: VEF = (1.055 × valve area × VTIAT) × (PV/AT), where VTIAT is the velocity time integral during acceleration, PV is the peak velocity, and AT is the acceleration time. In growth-restricted fetuses, the ventricular ejection force of both ventricles was found to be significantly decreased, showing the low influence of ventricular afterload on this parameter of myocardial function. A further drop of ventricular ejection force was related to the

appearance of other signs of severe fetal compromise.58 In hydropic fetuses this parameter may also be useful, but has not yet been studied. Diastolic ventricular function may be assessed by evaluation of the filling pattern of the ventricles. In fetal life the peak velocities of tricuspid and mitral blood flow during the early diastole (E-wave) are lower than the second peak during atrial contraction (A-wave). The E/A ratio increases during gestation, reflecting maturation processes of the myocardium and the diastolic properties, respectively, during gestation. A decrease of the E/A ratio and, in particular, monophasic filling reflect severe diastolic dysfunction of the ventricle or cardiac tamponade. The global cardiac ventricular function may be assessed by the Tei index (myocardial performance index, MPI),

492

Fetal Cardiology

which allows the evaluation of combined systolic and diastolic ventricular myocardial performance.59 The Tei index is defined as the sum of the isovolumetric contraction time and isovolumetric relaxation time divided by the ejection time (Tei index = (ICT + IRT)/ET). This index is suggested to be independent of ventricular geometry and heart rate and not affected by mitral or tricuspid regurgitation. A gradual decrease in the Tei index represents maturation and increasing myocardial function during gestation. Abnormal Tei indexes were demonstrated in fetuses with intrauterine growth restriction and maternal diabetes.59 In chronic twin–twin transfusion syndrome, increased myocardial performance indexes of both ventricles were demonstrated in the recipient twin, partly before an increase of venous pulsatility, cardiomegaly, or hydrops occurred. These increases of the myocardial performance indexes in the recipient fetus were mainly caused by prolongation of the isovolemic relaxation time, suggesting myocardial diastolic dysfunction, whereas the indexes of the donor fetus stayed in the normal range.60,61 In hydropic fetuses, the Tei index may demonstrate global right and/or left ventricular dysfunction.61,62 In the future, besides tissue Doppler velocity, evaluation of the strain and strain rate may be successfully applied for the assessment of fetal cardiac function.63 Myocardial hypertrophy can also be measured by M-mode echocardiography. Owing to the high afterload sensitivity of the fetal heart, the obstruction of ventricular outflow usually causes myocardial hypertrophy of the ipsilateral ventricle. Biventricular myocardial hypertrophy is found in the recipient fetuses of twin–twin transfusion syndrome, in fetuses with complete heart block and other forms of sustained bradyarrhythmia, and furthermore in fetuses with long-standing tachyarrhythmia, chronic anemia, and storage diseases. Severe myocardial hypertrophy may significantly reduce systolic and diastolic ventricular function, with reduced myocardial blood flow during stress.64-67 It may also be associated with severe cardiovascular decompensation of the neonate in reaction to the increased cardiac work after birth.46,47

Incompetence of structurally normal cardiac valves Incompetence of structurally normal cardiac valves most commonly concerns the tricuspid valve. Severe insufficiency of the tricuspid valve and also the mitral valve may indicate congestive heart failure based on myocardial dysfunction, increased ventricular preload, and/or afterload.46,47 On the other hand, a trivial tricuspid regurgitation is usually transient, mostly restricted to the early and mid-systolic period, and in rare cases also holosystolic.68–70 Dependent on the sensitivity of the utilized Doppler equipment,

this phenomenon may be demonstrable in around 7% of healthy fetuses without any cardiac or extracardiac pathology.69 A holosystolic tricuspid regurgitation, however, mostly indicates a pathological significance such as cardiac defects, constriction of the arterial duct or extracardiac disorders increasing right ventricular volume or pressure, and/or compromised myocardial function. On the other hand, the severity of the regurgitation itself does not seem directly related to the occurrence of hydrops. Only additional alterations of intracardiac blood flow such as relative restriction of the fossa ovalis and inadequate left ventricular function may limit the compensatory possibilities if fetal hydrops appears. Pulmonary or aortic valve regurgitation is extremely rare in healthy fetuses.71 Significant insufficiency of the pulmonary and/or aortic valve usually occurs only in the stage of severe myocardial failure,47 when the support for the semilunar valves is decreased, for example in fetuses with complete heart block, and with tricuspid valve dysplasia and Ebstein’s anomaly (Figure 34.1), and especially in the recipient fetus of twin–twin transfusion syndrome. For assessment of the severity of AV valve regurgitation, color Doppler-derived parameters of jet morphology are used, particularly the length of the jet related to the distance from the tricuspid valve to the opposite right atrial wall, and the jet area related to the area of the right atrium. Because the color Doppler-visualized jet morphology is more dependent on the velocity than on the volume, and there is a strong influence of color Doppler equipment and operating setting on the jet morphology, most examiners use the temporal duration of the jet related to the systolic period and grade this into non-holosystolic (early and mid-systolic) regurgitation and holo- or pansystolic regurgitation.42,69,70,72 These intervals can be measured by spectral and color Doppler M-mode echocardiography.42,69,70,72 Especially for intraindividual follow-up studies, AV valve regurgitation sensitively reacts when cardiac function decreases or improves.42 Except for extremely rare cases, holosytolic regurgitation suggests a substantial alteration of the fetal circulation. The myocardial contractility can be evaluated using the spectral Doppler-derived shape of the tricuspid regurgitation, measuring the rate of ventricular pressure rise over time (dP/dt). This reacts very sensitively to changes in ventricular contractility and is relatively independent of afterload during the pre-ejection period.46,47,73 Values lower than 800 mmHg/s are abnormal.46 In a study of 20 fetuses with holosystolic tricuspid regurgitation associated either with ductal constriction or with non-immune hydrops, the Doppler-derived right ventricular dP/dt appeared to be useful for right ventricular functional assessment in fetuses with tricuspid regurgitation and for prediction of poor outcome if dP/dt was lower than 400 mmHg/s.73

Cardiac diseases in association with hydrops fetalis

Venous Doppler velocimetry An increased pulsatility of the blood flow in the pericardial veins indicates compromised cardiac function in fetuses with intrauterine growth restriction, and is commonly associated with hypoxemia and also acidemia. A substantial increase of the end-diastolic pressure of the ventricles elevates the right atrial as well as the central venous pressure, and results in decreased venous forward flow during all of diastole, including atrial systole, generating an abnormal venous flow pattern. Therefore, each increase of cardiac afterload, preload, and/or myocardial dysfunction may cause an elevation of the central venous pressure. When sufficient, this results in an increased pulsatility of the venous blood flow velocity waveforms.41,74 For fetal surveillance, recordings of the blood flow pattern of the inferior vena cava, ductus venosus, and umbilical vein are commonly used in clinical practice. An excess of normal reverse blood flow during atrial systole in the inferior vena cava indicates an increase of central venous pressure as well as reduced antegrade, absent, or reverse blood flow during atrial systole in the ductus venosus, and monophasic as well as biphasic pulsations in the umbilical vein. There are some differences between these three veins: 1. The blood flow of the inferior vena cava is directed into the right ventricle and that of the ductus venosus through the foramen ovale into the left atrium; 2. Hypoxemia increases the proportion of umbilical venous blood which is preferentially directed to the ductus venosus, bypassing the liver; 3. An active vasodilatation of the ductus venosus may reduce umbilical venous pressure and facilitates a more distal propagation of the pulse wave during atrial contraction; 4. In contrast to the vasodilatation and increased or maintained blood flow of the ductus venosus blood, flow volume in the inferior vena cava is diminished, when hypoxemia-induced arterial blood flow redistribution takes place; 5. In the inferior vena cava there may be a higher compliance than in the ductus venosus; furthermore, any stress-induced increase in sympathetic tone causes venoconstriction. However, alterations of blood flow velocity waveforms of the inferior vena cava and ductus venosus seem to be equally good in clinical practice for predicting the extent of fetal hypoxemia and acidemia in growth-restricted fetuses,75 whereas transmission of the pulsations into the portal and umbilical veins correlates with an increasing degree of compromise.76 In fetuses with elevation of the central venous pressure caused by other factors, increased pulsatility in the inferior vena cava and ductus venosus also results, and correlates with the pressure increase. Increased pusatility of venous

493

flow velocity waveforms may be observed in some fetuses with cardiac defects. It is caused by the particular hemodynamics of some cardiac defects causing increased right atrial pressure, such as tricuspid atresia and severe right ventricular outflow obstruction (pulmonary atresia and pulmonary severe stenosis) with intact interventricular septum, and Ebstein’s anomaly/tricuspid dysplasia with severe tricuspid regurgitation, presumably caused by a relative restriction of the increased transatrial flow through the fossa ovale without other indicators of compromised cardiac function.77 Therefore, venous Doppler velocimetry in the proximal veins such as the inferior vena cava seems to be sensitive to changes of fetal hemodynamics by some cardiac defects and alterations of cardiac function, and may become abnormal if cardiomegaly is still not identifiable. In advanced stages of fetal congestive heart failure, however, where hydrops is often associated, a large increase of atrial and venous pressure results in a strongly pulsatile venous blood flow and in more distal propagation of this pulsatility, especially into the umbilical venous circulation. Thus, fetuses with pulsations in the umbilical venous blood flow have the most serious compromise of cardiac function. As in fetuses with intrauterine growth restriction,76 the occurrence of umbilical venous pulsations seems to be the best predictor of intrauterine or perinatal death in fetuses with non-immune hydrops.40,78 This is the case even if compared to right and left ventricular fractional shortenings, pulmonary and aortic peak velocities, the aortic and pulmonary valve products of time velocity integral and heart rate, inferior vena caval diameter, and pulsatility of inferior venous blood flow.78 An important exception is fetal tachyarrhythmia, where pulsation in the umbilical venous blood flow is usually present during supraventricular tachycardia above a critical frequency of 210–220 beats/minute even in fetuses without hydrops and excellent prognosis.79–81 Furthermore, good fetal outcome may also occur in some cases of marked fetal bradycardia and complete heart block, showing pulsations in the umbilical vein. Although changes of ventricular preload and afterload may influence the venous flow velocity waveform, in most clinical situations a substantial increase of pulsatility in venous blood flow indicates myocardial dysfunction and congestive heart failure. In contrast to the Dopplerderived parameters of tricuspid regurgitation and the ventricular fractional shortening, which have a lower reproducibility and demand an experienced sonographer and excellent equipment, Doppler flow velocity waveforms of the ductus venosus and inferior vena cava can easily and reproducibly be recorded.80,81 Compared to measurement of the cardiothoracic ratio, venous Doppler velocimetry appears to be more sensitive to congestive heart failure. Therefore, the non-invasive evaluation of the venous Doppler flow velocity waveform to date seems to be the most valuable tool for monitoring cardiac function.

494

Fetal Cardiology

Cardiovascular anomalies With documented incidences between 20 and 40%, cardiac anomalies are the most common associations with non-immune hydrops.4,6,7,21 Sustained arrhythmia and/or structural cardiac defects can cause congestive heart failure and resultant non-immune hydrops.21 Because of specific features of the fetal circulation, with the parallel arrangement of the two ventricles and the presence of communications between the atria and the great arteries, anomalies of the right and left ventricular flow may elevate the right atrial and systemic venous pressure. Soft tissue edema and effusion into the serous cavities, but not pulmonary edema, are the consequence. On the other hand, any dysfunction affecting one side of the heart can be overcome by functional adaptation of the other side, maintaining a combined cardiac ventricular output within the normal limits, even if detailed echocardiographic studies82 have shown the presence of hemodynamic abnormalities. These include especially a reduction of combined cardiac output and stroke volume already in some fetuses with congenital heart disease, but absent hydrops fetalis. This may limit an adequate response to hemodynamic stressors such as birth, transition from fetal to neonatal circulation, and heart surgery.82 In general, however, the parallel flow circuitry protects the fetus against cardiac decompensation, congestive heart failure, and hydrops. Isolated structural cardiac defects very seldom lead to interstitial fluid accumulation and hydrops. Therefore, it is incorrect from a pathophysiological point of view that in many published reports and review articles a causative relationship between the cardiac lesion and the hydrops is postulated, for example in cases with ventricular or atrial septal defects, tetralogy of Fallot, or transposition of the great arteries. Therefore, in many cases with hydrops, the presence of a cardiac defect is coincidental, and there is no causal relationship between the cardiac abnormality and the non-immune hydrops.21,83 In these cases, different extracardiac factors may have led to hydrops, especially in fetuses with Turner syndrome and autosomal trisomies.

Structural cardiac defects Tricuspid valve dysplasia and Ebstein’s anomaly In the group of congenital heart defects with the possibility of developing congestive heart failure followed by hydrops during fetal life, tricuspid valve dysplasia and Ebstein’s anomaly are two of the most common cardiac malformations. Thickened, nodular, and often redundant tricuspid leaflets with normal attachment at the atrioventricular junction are the diagnostic criteria of tricuspid atresia,

whereas Ebstein’s anomaly is characterized by a displacement of the septal and posterior leaflets from the atrioventricular junction downward into the inlet component of the right ventricle, causing a downward displacement of the functional annulus (septal > posterior > anterior leaflet) and a functional atrialization of a variable portion of the anatomically right ventricle with variable degrees of hypertrophy and thinning of the wall.84 Because of the significant variances between the fetal and neonatal circulation, different factors determine intrauterine and postnatal survival.85 The survival of neonates with Ebstein’s anomaly and tricuspid dysplasia is essentially dependent on their ability to establish an adequate pulmonary flow. Greater impairment of pulmonary flow results from severe pulmonary obstruction, often associated with severe tricuspid insufficiency and poor right ventricular function. Severe cardiomegaly is also an unfavorable parameter, presumably as cardiac enlargement is most extensive if severe tricuspid incompetence is present (Figure 34.1). In some cases with an open pulmonary valve, the occurrence of reverse perfusion of the ductus arteriosus and the pulmonary trunk with severe pulmonary insufficiency suggests severe right ventricular dysfunction. Therefore, a right atrial (RA) area ratio (ratio between the area of the functional right atrium and the added areas of the other cardiac chambers measured in the four-chamber view) > 1 and no antegrade flow in the pulmonary trunk seem to be the best prenatal parameters predicting an unfavorable postnatal course.86 In some fetuses, significant inhibition of lung growth by long-standing compression leading to severe pulmonary hypoplasia may complicate neonatal resuscitation in addition,87,88 even if significant changes of the pulmonary artery as observed in lung hypoplasia of other etiology do not occur.89 On the other hand, if adequate pulmonary flow can be established, a decrease of pulmonary vascular resistance and right ventricular afterload may markedly improve the tricuspid incompetence after birth. Therefore, an RA area ratio > 1, fetal hydrops, no antegrade flow in the pulmonary trunk, severe pulmonary valve obstruction, severe tricuspid lesion with tethered distal attachment of the septal leaflet, severe right ventricular dysplasia, and significant cardiomegaly are the most important prognostic parameters for an unfavorable neonatal outcome.86,90 In contrast to neonatal survival, fetal survival is not dependent on adequate pulmonary flow, but is strongly linked to the ability of the fetal heart to compensatorily increase the left ventricular volume flow. Therefore, the size of the fossa ovalis allowing the required increase of transatrial right-to-left shunt and a sufficient left ventricular diastolic and systolic function are mandatory in fetuses with tricuspid atresia and Ebstein’s anomaly, to avoid the occurrence of congestive heart failure followed by hydrops and in utero fetal death.85,86 Pathoanatomical studies in infants and children with tricuspid and pulmonary atresia showed substantially

Cardiac diseases in association with hydrops fetalis

increased rates of foramen ovale over atrial septal area, suggesting that an adequate increase of transatrial flow and a wider than normal size of the fossa ovale is mandatory for in utero survival.91 In a reported series of eight fetuses with Ebstein’s anomaly, two fetuses became hydropic.85 These fetuses had the lowest ratio of fossa ovalis diameter over atrial septal length, and accordingly the lowest left ventricular output, both in the normal range. The six non-hydropic fetuses had an increased size of the fossa ovalis and elevated left ventricular output.85 There were no significant differences between the hydropic and non-hydropic fetuses concerning septal leaflet displacement, severity of tricuspid insufficiency, pulmonary valve obstruction, or cardiothoracic ratio.85 Pulmonary obstruction and a severe tricuspid lesion usually lead to severe tricuspid insufficiency during fetal life, and therefore demand a wider fossa ovalis allowing an adequate increase of left heart volume flow. This makes the occurrence of a relative restriction of the fossa ovalis more probable in these cases. Substantial compression of the left ventricle may act as an additional factor for fetal cardiac decompensation.90 Venous Doppler velocimetry in the inferior vena cava and ductus venosus can provide early detection of those fetuses with a maladapted circulation developing congestive heart failure and fetal hydrops.77 Furthermore, the onset of supraventricular tachyarrhythmia – typical is atrioventricular reentry tachycardia via an accessory pathway84 – may be an additional cause for developing hydrops in Ebstein’s anomaly, favored by the marked right atrial distension and the higher incidence of preexcitation syndrome associated with Ebstein’s anomaly. Because fetuses with severe cardiomegaly and hydrops are more easily detectable during obstetric ultrasound examination, the severe end of the spectrum is preferentially referred to the specialized center. This explains the high proportion of these fetuses in prenatal series,92 the high incidence of associated severe pulmonary valve obstruction,87,88 and the very poor outcome of prenatally diagnosed fetuses with tricuspid dysplasia and Ebstein’s anomaly, reaching an overall mortality of up to 80% as reported in the literature.86–88,92,93 Because of the poor outcome of Ebstein’s anomaly and tricuspid dysplasia in the neonatal period of hydropic fetuses (around 20–40% of all neonates diagnosed with tricuspid valve dysplasia and Ebstein’s anomaly will not survive 1 month, and fewer than 50% will survive to 5 years of age; in symptomatic neonates the prognosis is always very poor, particularly if fetal hydrops occurs, and around 90% of these will die during the perinatal period),84,86 termination of pregnancy or expectant acceptance of in utero fetal death is the preferred management for these fetuses. In individual cases, however, preterm delivery and aggressive reanimation, including mechanical ventilation, administration of prostaglandins to maintain adequate pulmonary flow, and administration of nitric oxide to reduce pulmonary resistance, may be successful86

495

if adequate pulmonary blood flow can be established, and in rare cases also different operative interventions. In special situations, transplacental treatment with digoxin may at least temporarily improve cardiac function, with remission of the hydrops, and may be indicated for avoidance of severe prematurity or, in a monochorionic twin pregnancy, for prevention of the risks of in utero death in one twin for the healthy co-twin.94 In fetuses with sustained tachyarrhythmia, transplacental treatment with antiarrhythmic drugs may be successful and should be performed in non-hydropic fetuses with Ebstein’s anomaly if long-standing reentry tachycardia is associated, in order to prevent the development of hydrops.

Atrioventricular septal defect Isolated atrioventricular septal defects very rarely cause congestive heart failure and hydrops during fetal life. Semiquantification of AV valve regurgitation by color Doppler-derived spatial and temporal parameters has shown that AV valve regurgitation can be demonstrated in almost all fetuses with atrioventricular septal defects,72 but severe AV valve insufficiency is not common in fetuses.72 Only in a small minority of fetuses with atrioventricular septal defect holosystolic regurgitation correlated with the severest degree of AV valve regurgitation was non-immune hydrops associated.72 In these cases, cardiomegaly and a pathological venous blood flow pattern are usually present, and additionally indicate congestive heart failure as the cause of hydrops. Rarely, severe dilatation of the atrium may trigger supraventricular tachyarrhythmia, resulting in hydrops fetalis.95 In the special situation of heterotaxy syndrome, however, where left atrial isomerism and atrioventricular septal defect are frequently associated with severe bradyarrhythmia, mostly as complete heart block, congestive heart failure followed by hydrops often occurs, and leads to in utero fetal death.96–99 The combination of severe bradycardia and compensatory markedly increased stroke volume and ventricular systolic pressure with a malformed and incompetent atrioventricular valve, and ischemia-induced dysfunction of the hypertrophic and structurally remodeled myocardium, may be factors for the occurrence of hydrops in these fetuses. Owing to the extremely poor prognosis of these hydropic fetuses, intrauterine treatment seems not to be justified if an atrioventricular septal defect is associated with complete heart block.

Incompetence of the semilunar valves Structural abnormalities with severe insufficiency of the semilunar valves are very rare. A few cases of truncus

496

Fetal Cardiology

(a)

(b)

(c)

(d)

Figure 34.2

(a) Absent pulmonary valve syndrome in a fetus with tetralogy of Fallot and agenesis of the ductus arteriosus at 26 + 3 weeks’ gestation. The pulmonary trunk and both pulmonary arteries are significantly dilated. (b) During systole the turbulent blood flow via the rudimentary pulmonary valve and (c) during diastole the severe pulmonary insufficiency are shown. (d) By continuous wave Doppler the systolic jet velocity of 2.61 m/s and the holodiastolic pulmonary insufficiency with a jet velocity of 2.32 m/s are demonstrated.

arteriosus communis with severe incompetence of the truncal valve and the absent pulmonary valve syndrome may cause severe atrioventricular regurgitation with consequent elevation of the right atrial and venous pressure.100–102 The absent pulmonary valve is mostly associated with tetralogy of Fallot, seldom with intact ventricular septum or discontinuous left pulmonary artery or absent left pulmonary artery. In tetralogy of Fallot with absent pulmonary valve, the regurgitant flow results in a volume overload of both ventricles, especially in fetuses with patent ductus arteriosus if blood flow from the aorta fills both ventricles during diastole. Therefore, patent ductus arteriosus in fetuses with tetralogy of Fallot and absent pulmonary valve causes a severe chronic volume overload of the fetal heart incompatible with fetal life, and results in cardiac failure, hydrops, and fetal death early in gestation.103 Besides hydrops these fetuses show a characteristic to-and-fro blood flow in the main pulmonary artery,

descending aorta, umbilical artery, middle cerebral artery, and further arteries because of the diastolic ‘steal effect’ and the absence of ‘windkessel’ function in the pulmonary trunk. Only if the ductus arteriosus is absent – agenesis of the ductus arteriosus is present in 10–20% of fetuses with tetralogy of Fallot – may the fetus reach the second trimester of gestation because the regurgitant part of the stroke volume is limited. In these fetuses, enormous dilatation and pulsation of the pulmonary trunk and the right and left pulmonary arteries are the sonographic characteristics, combined with a stenotic and regurgitant to-and-fro blood flow across the rudimentary pulmonary valve (Figure 34.2). The volume overload of both ventricles may result in hydrops, intrauterine death, or a severely ill neonate complicated by severe respiratory distress and bronchotracheomalacia resulting from long-standing compression of the trachea and primary bronchi.104–106 Fetuses with absent pulmonary valve and and intact

Cardiac diseases in association with hydrops fetalis

interventricular septum or discontinuous left pulmonary artery, however, have a much lower volume overload and may survive. In 30–40% of fetuses with tetralogy of Fallot, absent pulmonary valve, and agenesis of the ductus arteriosus, a microdeletion 22q11.2 is present.

Aortico-left ventricular tunnel and aortico-right ventricular tunnel Aortico-left ventricular tunnel is a rare malformation that comprises a communication between the aortic sinus and the left ventricle, bypassing the aortic valve. The most severe degree of this defect with a large channel can be diagnosed in utero by the finding of left ventricular dilatation and dysfunction on obstetric ultrasound examination.107,108 Exact diagnosis is obtained by demonstration of para-aortic regurgitation into the left ventricle that causes chronic left ventricular volume overload. Sometimes, significant aortic stenosis is associated.108 Direct visualization of the tunnel as an anechogenic para-aortic structure may be possible by two-dimensional echocardiography.108 A dilated and thick-walled left ventricle, dilated aortic sinus, bulging of the ascending aorta into the left ventricular outflow tract, and sometimes a thickened aortic valve may be additional findings.107,108 When, in fetuses with severe aortic regurgitation, the developing left ventricular dysfunction can no longer be balanced by the right ventricle, in utero congestive heart failure and hydrops may occur, suggesting a very poor prognosis.108 Aortico-right ventricular tunnel is diagnosed by demonstration of an abnormal connection between the ascending aorta and the right ventricle with bidirectional flow and dilatation of the right and left ventricles, pulmonary trunk, and ascending aorta.109 Volume overload heart failure and postnatal myocardial hypoperfusion may result in a rapid detoriation of the neonate. The same pathomechanism may explain hydrops and death in fetuses with isolated absence of the aortic valve.

Constriction and closure of the ductus arteriosus Significant constriction of the ductus arteriosus leads to an acute increase of right ventricular afterload, resulting in an elevation of right ventricular pressure, right ventricular systolic dysfunction, and severe tricuspid regurgitation, probably caused by papillary muscle dysfunction and/or dilatation of the tricuspid annular ring.110 Increased transatrial right-to-left shunt and/or pulmonary blood flow may also result in volume load and dilatation of the left ventricle.111 Sometimes, the occurrence of hydrops fetalis has been reported.112–115 In the vast majority of cases, ductal constriction or closure is caused by transplacental transfer of indomethacin or other

497

prostaglandin synthesis inhibitors applied to the mother for the treatment of premature contractions and/or polyhydramnios.110,116,117 After discontinuation of indomethacin therapy, ductal constriction, right ventricular systolic dysfunction, and tricuspid regurgitation usually resolve.110,111 The risk of drug-induced constriction of the arterial duct is dependent on the dosage of indomethacin and on gestational age, starting after 24 weeks and increasing in advancing pregnancy. Therefore, inhibitors of cyclo-oxygenase enzymes should not be used after 32–34 weeks’ gestation, because ductal constriction may lead to in utero hydrops and death, or may cause postnatal pulmonary hypertension. Diagnosis of ductal constriction is based on increased systolic and particularly diastolic peak velocities in the ductus arteriosus, resulting in a pulsatility index of less than 1.9, mostly accompanied by severe tricuspid regurgitation with increased peak velocities of 4–5 m/s due to the marked elevation of the right ventricular systolic pressure due to the increased afterload.118 Spontaneous constriction or complete closure of the ductus arteriosus have been reported, followed by hydrops.112–114

Obstruction of the right and left ventricular outflow tracts In the same way, afterload mismatch by structural right ventricular outflow tract obstruction (pulmonary atresia with intact interventricular septum and severe pulmonary stenosis) and, rather rarely, left ventricular outflow tract (aortic atresia and severe aortic stenosis) obstruction may cause severe insufficiency of tricuspid or mitral valves, respectively. Although central venous pressure may be increased in fetuses with right heart obstructions (tricuspid atresia, pulmonary stenosis and atresia with intact interventricular septum), as was demonstrated by increased pulsatility in the ductus venosus,77 the development of hydrops is very rare in fetuses with right ventricular outflow tract obstruction, because the left ventricle is mostly able to compensate for the alteration of right heart blood flow. Normal function of the left ventricle and a non-restrictive foramen ovale are mandatory in this situation. Therefore, significant bivalvular obstruction of both pulmonary and aortic valves appears to be a severe complication, resulting in early fetal death and, if surviving early gestation, leading to hydrops in the second trimester.119 Most fetuses with a hypoplastic left heart do not become hydropic, and the right ventricle pumps out the whole cardiac output. In some fetuses with severe aortic obstruction and hypoplastic left heart, severe mitral regurgitation may significantly elevate the left atrial pressure, leading to a substantial alteration of flow via the foramen ovale, with the appearance of an accelerated and disturbed left-toright shunt or complete closure of the foramen ovale.120–124

498

Fetal Cardiology

(a)

(b)

(c)

(d)

Figure 34.3

(a) Critical stenosis of the aortic valve is demonstrated in a fetus at 34 + 2 weeks’ gestation. The myocardium of the left ventricle shows marked hypertrophy. The size of the left atrium is significantly increased. (b) The valve of the foramen ovale shows no movement in the two-dimensional echocardiogram. By color Doppler examination a shunt across the foramen ovale cannot be demonstrated. (c) By continuous wave Doppler a maximal velocity of 4.62 m/s is measured across the stenotic aortic valve. Using the Bernoulli equation a pressure gradient of 85.47 mmHg is calculated. (d) Severe holosystolic mitral insufficiency is demonstrated by color Doppler imaging. By continuous wave Doppler a jet velocity of 4.81 m/s is measured. Using the Bernoulli equation a pressure gradient of 92.5 mmHg is calculated.

In the last group, an extreme enlargement of the left atrium with secondarily increased cardiothoracic ratio may be the consequence (Figure 34.3). In these rare variants of hypoplastic left heart syndrome with highly restrictive or closed foramen ovale, hydrops may occur despite normal systemic venous Doppler flow profiles, but the pulmonary venous waveforms demonstrate a highly increased pulmonary venous pressure. Normally, the parallel arrangement of fetal circulation, pulmonary arteriolar vasoconstriction, limited pulmonary blood flow, and reverse shunting via the foramen ovale prevent a significant increase of pulmonary venous pressure and pulmonary edema. In fetuses with aortic obstruction with highly restrictive or closed atrial septum and severe mitral

insufficiency, which is associated with a relatively high left heart volume, a markedly increased pulmonary pressure leads to lymphatic pulmonary edema – pulmonary lymphangiectasia is a common finding in such cases – and decreased oncotic pressure owing to the loss of albuminrich pulmonary fluid into the amniotic sac,124 and generalized hydrops fetalis may occur. Right ventricular overload may also result in right atrial and central venous pressure increase, in addition.120–123 In this situation, treatment with digoxin, delivery by cesarean section, and balloon dilatation of the stenotic aortic valve in the newborn resulted in a successful outcome.123 In earlier gestation, intrauterine opening and/or stenting of the foramen ovale may stabilize the fetal condition, decrease pulmonary

Cardiac diseases in association with hydrops fetalis

venous pressure, and, in addition, may allow regeneration of the altered pulmonary venous and lymphatic vessels125,126 and perhaps improve the poor prognosis of fetuses with hypoplastic left heart syndrome with an intact or highly restrictive atrial septum.127

Primary restriction or complete closure of the foramen ovale The concept of primary premature closure of the foramen ovale as an occasional cause of secondary hypoplastic left heart syndrome has now been abandoned. Restriction and smallness of the foramen ovale in hypoplastic left heart syndrome is a common secondary phenomenon in fetuses with severe left ventricular outflow tract obstruction, where a primary patent foramen ovale develops into restriction or complete obstruction in later gestation. A few case reports, however, have described an isolated restriction or premature closure of the foramen ovale in association with non-immune hydrops.128–131 In these cases, the left ventricle was normal in size or underdeveloped, whereas the right-sided chambers were substantially enlarged. Diagnosis is based on the demonstration of disturbed blood flow with increased velocities through the foramen ovale or of absent transatrial blood flow in cases of complete closure. In addition, an abnormal blood flow pattern in the pulmonary veins with substantial reversal of blood flow during atrial systole, indicative of elevated left atrial pressure,132 can be expected in this situation. Owing to the abrupt decrease in pulmonary flow, delivery with consequent disappearance of hydrops and rightsided congestive heart failure is the recommended treatment.128,129 Restriction of the foramen ovale may also occur in association with tachyarrhythmia.133 In these cases initial left atrial depolarization may result in a premature pressure increase in the left atrium, which lowers the interatrial shunt during the short diastole or may lead to premature closure of the foramen ovale. In this group, prenatal normalization of cardiac rhythm may lead to a ‘reopening’ of the foramen ovale and remission of hydrops. On the other hand, premature closure of the foramen ovale has also been reported in non-hydropic fetuses.134 In an autopsy series, this was not significantly increased among hydropic infants compared to non-hydropic controls, indicating that factors other than premature closure of the foramen ovale are operative in the pathogenesis of non-immune hydrops.83

Cardiac tumors Cardiac tumors are a rare cause of non-immune hydrops. Dependent on their localization, size, and number, cardiac

499

tumors may cause hydrops by impeding diastolic filling, by altering the function of atrioventricular valves, or by obstructing the outflow tract,135 sometimes indirectly by induction of sustained supraventricular and less often ventricular tachyarrhythmia.135–137 The vast majority of fetal intracardiac tumors are rhabdomyomata, showing a homogeneous echogenicity which is increased compared with the normal myocardium. A rhabdomyoma may prominently protrude into the ventricular or atrial cavity or may be confined as a small tumor to the cardiac wall. Spontaneous regression of the rhabdomyomata after birth commonly occurs. In around 50% of fetuses with rhabdomyomata, however, tuberous sclerosis is present, especially in fetuses with multiple cardiac tumors.136–139 In the case of a rhabdomyoma, a definitive exclusion of tuberous sclerosis may be impossible by prenatal ultrasound, because the pathognomonic giant cell astrocytomata are often undetectable by sonography of the fetal brain, even if high-resolution transvaginal scanning is used in the fetal vertex position.140 More successful in utero diagnosis of associated intracerebral lesions indicating tuberous sclerosis has been reported by magnetic resonance imaging.141 The sonographic visualization of multiple intracardiac tumors seems to be indicative of tuberous sclerosis. Successful drug-induced cardioversion of a supraventricular tachycardia into sinus rhythm leading to complete remission of hydrops in utero has been shown in a fetus with multiple rhabdomyomata.137 If congestive heart failure and hydrops resulting from tumor obstruction do occur, in utero transplacental treatment with digoxin or delivery, depending on gestational age, may be considered. However, such an aggressive treatment may not be indicated, because of the poor prognosis in such hydropic fetuses and the frequent association with tuberous sclerosis. It should be performed only in cases with a special request of the parents after detailed counseling. Furthermore, the development of fetal hydrops has been seen in a fetus with a right atrial hemangioma142 showing mixed echogenicity with echogenic and hypoechogenic parts, and in another with a hamartoma of the conduction system in association with tachyarrhythmia and a structural heart defect.143 Pericardial teratomata are very rare tumors of variable echogenicity, often containing cysts of varying size. They can grow to a considerable size, and are commonly associated with a large pericardial effusion (Figure 34.4). In these situations, the development of hydrops is a more common complication, resulting from a substantial obstruction of intracardiac blood flow due to cardiac compression.144–152 In such fetuses with cardiac tamponade caused by a massive pericardial effusion, stabilization of cardiac function may be achieved by single or repeated pericardiocenteses or by placement of an indwelling catheter as pericardioamniotic shunt to provide drainage of the pericardial effusion into the amniotic sac, resulting

500

Fetal Cardiology

Tachyarrhythmia

Figure 34.4 Four-chamber view demonstrating a pericardial teratoma with a massive pericardial effusion in a fetus at 25 + 0 weeks’ gestation. In the following gestation four pericardiocenteses were performed leading to immediate remission of ascites and skin edema and normalization of pulsatility in the ductus venosus (reproduced with permission from reference 144).

in rapid remission of fetal fluid accumulations and normalization of the pulsatility of venous flow velocity waveforms.144,149–152 In utero open fetal surgery for tumor resection may be an even more invasive alternative in very immature fetuses with hydrops caused by an intrapericardial teratoma,147,148 particularly if cardiac tamponade is predominately caused by the tumor.

Idiopathic arterial calcification Idiopathic arterial calcification is a rare disease of unknown cause, characterized by generalized calcification of large and medium-sized arteries especially in the aorta and pulmonary trunk.153 In addition, fibrous proliferation in elastic and muscular arteries has been found. Most commonly the coronary arteries are affected, but peripheral arteries of the gastrointestinal tract, kidneys, extremities, brain, and placenta may also be involved. In these fetuses, ischemia-induced myocardial dysfunction may cause severe hydrops, tissue ischemia, and fetal death in the late second or third trimester.153–156 In less severe cases, especially if no hydrops is present, palliative treatment postpartum with steroids and diphosphonates may stop the progression of the disease.153 However, most infants with idiopathic arterial calcification die within the first year of life, complicated by cardiac and pulmonary failure, renal infarction, peripheral gangrene, and bowel infarction.153

The most common cardiac disease causing fetal hydrops is fetal tachyarrhythmia. Because a separate chapter about fetal tachyarrhythmias is written in this book (Chapter 33), the pathophysiology and the clinical management are only briefly discussed in the context of fetal hydrops. Sustained fetal tachyarrhythmia may cause congestive heart failure, leading to elevated right atrial and systemic venous pressure. It may be followed by non-immune hydrops, placental edema, and polyhydramnios. In fetuses, supraventricular tachycardia is more frequent than atrial flutter, independent of the presence or absence of hydrops, whereas ventricular tachycardia is very rare.157,158 Between 80 and 90% of cases with supraventricular tachycardia are atrioventricular reentrant tachycardias based on an accessory atrioventricular conduction pathway beside the atrioventricular node, whereas atrial ectopic, atrioventricular nodal reentrant and junctional tachycardias are not often the electrophysiological mechanism of perinatal supraventricular tachycardia.159–162 An accessory atrioventricular connection seems to be present in some fetuses with atrial flutter.162 Pathophysiologically, there is a substantial shortening of the diastolic period of the cardiac cycle, which prevents adequate early diastolic filling of the ventricles. Furthermore, it is suggested that initial left atrial depolarization can be tolerated to a lesser degree by the fetus, because the left ventricular output is diminished and the interatrial right-to-left shunt is disturbed.163 Atrial pacing studies in the fetal lamb have shown an increase of ventricular output and a decrease of ventricular end-diastolic pressure at rates up to 300 beats/minute. Prolonged left atrial pacing at rates of 300–320 beats/minute results in a decrease of cardiac output and in the development of hydrops within 4–48 hours. Cardiomegaly and hepatomegaly develop, whereas arterial oxygen tension and mean aortic pressure remain unchanged.164–168 Because no or only slight hypoproteinemia may be observed, there is no evidence for an increase of capillary permeability for albumin. Above this ‘critical’ heart rate the mean venous pressure in the inferior vena cava abruptly increases by 75%.168 This abrupt elevation of venous pressure is associated with an immediate appearance of pulsatile reversal of blood flow occurring during diastole.168 Below this heart rate, the venous flow is biphasic with a systolic and diastolic forward surge, which also occurs immediately after the pacing is stopped.168 Besides the direct impedance of diastolic filling when the diastolic interval is critically shortened, the abrupt occurrence of changes – reduction of ventricular output, the immediate elevation of venous pressure, and appearance of pulsatile venous blood flow above a ‘critical’ pacing rate – suggests ventricular dysfunction consistent with an alteration of the pressure–volume relationship in association with impaired ventricular relaxation and compliance at high pacing rates. The most likely

Cardiac diseases in association with hydrops fetalis

explanation is that oxygen supply to the myocardium by coronary blood flow is inadequate for the increased requirement of the myocardium during tachycardia, in particular due to the significant shortening of the diastolic period when the major portion of coronary blood flow takes place.64 This hypothesis is supported by the observation that severe ventricular dysfunction and even injury of the myocardium may occur in prolonged tachycardia, and may cause reversible tachycardia-induced ‘cardiomyopathy’ in humans and animals.64–67 In conjunction with the enormous cardiac enlargement in fetuses with sustained supraventricular tachycardia, functional incompetence of both AV valves may be observed, suggesting structural remodeling of the ventricles in the presence of tachycardia-induced ‘cardiomyopathy’. Recovery from tachycardia-induced ‘cardiomyopathy’ was accompanied by persisting chamber dilatation, significant myocardial hypertrophy, and persisting diastolic and systolic dysfunction.67 Venous blood flow studies in the human fetus with supraventricular tachycardia have demonstrated the occurrence of monophasic forward and pulsatile reversed blood flow during diastole in the inferior vena cava and ductus venosus above a critical heart rate of approximately 210–220 beats/minute.79,80 This is in accordance with fetal lamb studies, where this change of venous blood flow pattern was associated with a considerable elevation of venous pressure.168 After the termination of supraventricular tachycardia, cardiac dilatation, myocardial hypertrophy, atrioventricular valve incompetence, hydrops, and increased pulsatility of venous blood flow disappear with immense interindividual differences. These could be explained by different stages of progression of tachycardiainduced ‘cardiomyopathy’ at the time of drug-induced cardioversion and, accordingly, varying time intervals for normalization of cardiac function.42,80,81 In conclusion, these results from animal and human studies indicate that fetal hydrops due to tachyarrhythmia is caused by congestive heart failure leading to elevated venous pressure and consecutive obstruction of lymphatic drainage, but not from hypoxic damage to capillaries or other tissues.167 Owing to the various problems postpartum in managing premature hydropic fetuses (postpartum increase of cardiac work, regulation of body temperature, mechanical ventilation, repetitive pleural drainage, simultaneous occurrence of lung edema and hyaline membrane disease, which reduce the effectiveness of surfactant therapy, severe degree of tachycardia-induced ‘cardiomyopathy’, refractory neonatal tachyarrhythmia), iatrogenic preterm delivery of hydropic fetuses for better control of arrhythmia often results in a poor outcome. Treatment in utero for adequate control of the arrhythmia and remission of hydrops is prudent in sustained tachyarrhythmia of fetuses with and also without hydrops. In hydropic fetuses with tachyarrhythmia, intrauterine treatment with digoxin alone and in combination with different antiarrhythmic drugs (flecainide, sotalol, amiodarone) is the

501

best approach for almost all fetuses, as elaborated in Chapter 33. Transplacental treatment is successful in the majority of cases.159,160,169–173 In hydropic fetuses with supraventricular tachycardia refractory to transplacental therapy, the direct intravascular application of antiarrhythmic drugs to the fetus as an ultimate method may be successful.169,174,175

Complete heart block Complete heart block may also cause fetal hydrops. Because complete heart block is discussed in a separate chapter of this book (Chapter 32), only its association with hydrops is emphasized in the following section. Especially the combination of fetal complete heart block and atrioventricular septal defect in fetuses with left atrial isomerism (Figure 34.5) seems to have a very poor prognosis, with the appearance of hydrops and intrauterine death in the majority of fetuses.96–99 The combination of a malformed and most severely incompetent AV valve with ventricular bradycardia causing high stroke volume and an increase of ventricular pressure is the pathophysiological mechanism explaining the most common appearance of hydrops in these fetuses. In contrast, the occurrence of hydrops in fetuses with AV block and ‘corrected’ transposition of the great arteries is very rare, because the AV valve apparatus is generally intact in these fetuses.96,99 This is also the reason why, in fetuses with complete heart block without congenital heart disease, the appearance of hydrops is much less frequently observed, occurring in only approximately 10–20% of fetuses.176 In these fetuses complete heart block is not caused by a malformation of the conduction system, but follows an inflammatory destruction of conduction pathways, especially in the AV nodal region, induced by transplacental transfer of maternal autoimmune antibodies. Hydrops may occur when the ventricular escape rhythm is very low; furthermore, widespread myocarditis and/or generalized extracardiac inflammation may contribute to the occurrence of hydrops in some of these fetuses. Sometimes hydrops develops only after 30 weeks of gestation, because at this time the required increase of the combined cardiac output cannot be produced by these fetuses. Another cause may be progression of a more generalized myocarditis and/or inflammation of other fetal tissues due to autoantibodies.177,178 In complete heart block without a structural cardiac malformation, intrauterine treatment of hydropic fetuses may be successful by transplacental infusion with salbutamol and isoprenaline (positive chronotropic and positive inotropic effects),179 by treatment with digoxin (positive inotropic effect),96,180,181 and by dexamethasone177,178 and/or plasmapheresis, impairing the severity of inflammation. This may be not only in the conductive tissue but also in the myocardium and other fetal organs, as elaborated

502

Fetal Cardiology

(a)

(b)

(c)

(d)

Figure 34.5 (a) Abdominal situs of a hydropic fetus with left atrial isomerism (heterotaxia syndrome), atrioventricular septal defect, and complete heart block is shown in a fetus at 23 + 6 weeks’ gestation. The stomach is on the right side of the abdomen (situs ambiguus) and moderate ascites is seen. (b) The heart is on the left side. The diastolic inflow across the common atrioventricular valve into both ventricles is colored red. (c) During diastole a regurgitant jet is shown (blue) in the middle of the common atrioventricular valve. (d) In parallel with the aorta (blue) the venous blood flow pattern in the azygos vein (red) is demonstrated in the thorax. By pulsed wave Doppler, the aortic blood flow is recorded in the upper channel showing a ventricular rate of 57 beats/minute. In the lower channel, the venous blood flow pattern demonstrated an atrial rate of 125 beats/minute independent of the ventricular systoles. The reverse flow during two of the atrial systoles is marked by arrows.

in Chapter 32. Remission of complete heart block, however, cannot be achieved by intrauterine treatment with dexamethasone or plasmapheresis.182–185 In a few reported cases, direct intrauterine pacing was technically successful, but only for some hours.186,187 Weekly echocardiographic monitoring of fetal PR interval starting at 16 weeks, and transplacental treatment by dexamethasone if prolongation of the PR interval188 is demonstrated, seem not to be effective in preventing complete heart block, because fetal complete heart block occurs abruptly without previous prolongation of the PR interval.189 Whether the anti-inflammatory effect of transplacental treatment with dexamethasone in fetuses with already manifested complete

heart block may improve secondary cardiomyopathy and the long-term outcome seems to be possible,184 but must be further evaluated because of side-effects for the mother and fetus.

Primary and secondary diseases of the myocardium The cardiomyopathies can be divided into dilated or congestive, hypertrophic, which may be obstructive in

Cardiac diseases in association with hydrops fetalis

addition, and restrictive types. Although myocardial diseases of unknown etiology were originally termed cardiomyopathy, this term is more generally used to categorize a wide range of heart muscle diseases not associated with structural cardiac anomalies or pericardial diseases. The primary form of cardiomyopathy consists of idiopathic cases.190 Non-compaction of the ventricular myocardium or spongiform cardiomyopathy results from an arrest of compaction of the loose myocardial network during fetal ontogenesis, resulting in a ‘spongy myocardium’ with ‘persisting sinusoids’ that fail to regress, and communicate to the epicardial coronary arteries. This disease may partially or totally involve the left or the right ventricle, and may be associated with other cardiac diseases such as aortic obstruction, or Ebstein’s anomaly. Isolated noncompaction of the ventricular myocardium may result in impaired ventricular function, heart failure, fetal hydrops, and arrythmias.191,192 The secondary form of cardiomyopathy consists of cases with systemic diseases resulting in significant functional alteration and structural remodeling of the myocardium. In a small number of fetuses metabolic storage diseases have been reported to significantly reduce in utero cardiac function, followed by congestive heart failure and hydrops.193–196 Cardiac glycogenosis without acid maltase deficiency and carnitine deficiency may very rarely present as non-immune hydrops fetalis.193–196 In one hydropic neonate, postnatal oral DL-carnitine supplements resulted in a dramatic improvement of cardiac function, normoglycemia, and restoration of serum carnitine levels to normal.194 Severe cardiomyopathy may also be pathophysiologically relevant in some hydropic fetuses with congenital myotonic dystrophy, an autosomal dominant disease characterized by fetal hypo- and akinesia.197,198 Myocardial infarction resulting in severe cardiac dysfunction is very rare during fetal life, and appears to be mostly induced by coronary thromboembolism.199 A chronic high-output cardiac state due to arteriovenous malformation, tumor, and parasitic twin is a more common cause of dilated cardiomyopathy in fetal life, sometimes leading to fetal hydrops. In chronic hemolysis, repetitive fetal maternal hemorrhage, and parvovirus B19 infection, significant fetal anemia causes a high-output cardiac state. In this group of diseases, the congestive heart failure appears not to be the primary cause of hydrops, but occurs in an advanced stage of fetal compromise, indicated by the appearance of cardiac enlargement, AV valve regurgitation, and increase of pulsatility of venous blood flow velocity waveforms. In earlier stages of the disease other pathomechanisms may be operating, especially hypoxia-induced capillary damage, reduced colloid oncotic pressure, and portal hypertension due to significant extramedullary erythropoiesis. Accordingly, Doppler studies in fetuses that were anemic because of alloimmunization demonstrated a hyperdynamic circulatory state with increased venous, intracardiac, and arterial blood flow velocities.200 The studies

503

suggested portal hypertension,201 but failed to show an increased pulsatility of venous blood flow representing central venous pressure and cardiac function.200 In Southeast Asia the leading cause of hydrops and perinatal death is homozygous α-thalassemia resulting in severe tissue hypoxia.22,202 Symptoms are anemia ranging from 3 to 10 g/dl, cardiomegaly, hepatosplenomegaly, hydrops fetalis and placentae, and poly- and later oligohydramnios. As in other diseases, placentomegaly is likely to be an important factor for the maternal hyperdynamic and hypertensive state, the so-called ‘mirror syndrome’.203–208 A thickened placenta seems to be the first sonographic marker, detectable from 10 to 18 weeks of gestation, followed by cardiomegaly, hepatosplenomegaly, and fetal hydrops, which may present as early as the 12th week of gestation and most commonly beyond the 20th week of gestation.209–212 Demonstration of placentomegaly, cardiomegaly resulting in an increased cardiothoracic ratio, and hydrops is used in Southeast Asia not only if DNA-based prenatal diagnosis is not readily available, but also as an alternative diagnostic means, avoiding the risks related to invasive procedures.202,210–212 Measurement of cardiothoracic ratios by transvaginal sonography can distinctly identify affected fetuses at 12–13 weeks of gestation.213 After sonographic detection, diagnosis is confirmed by fetal blood sampling showing hemolytic anemia, abnormal hemoglobin, and hypoxemia.212,213 Although some anemic fetuses show increased peak velocities at the pulmonary valve and a larger pulmonary valve diameter at 12 and 13 weeks of gestation, there is an extensive overlap between anemic and non-anemic fetuses.214 Also, nuchal translucency thickness, nuchal edema, and hydrops seem not to be useful for the prediction of anemic fetuses in early gestation.215 Cardiomegaly, AV valve regurgitation, and increased pulsatility of the venous blood flow pattern, suggesting congestive heart failure, have been reported in fetuses with transient myeloproliferative disorder and Down syndrome.216–223 Additional pathogenic mechanisms predisposing to the development of hydrops in these fetuses may be mild to moderate anemia with hemoglobin concentrations between 6 and 10 g/dl causing high cardiac output, capillary damage by hypoxia resulting from anemia and hyperviscosity, increased vascular resistance, extramedullary megakaryoblastic proliferation, and liver fibrosis. In fetuses with sacrococcygeal teratoma, large tumors with a high proportion of solid tissue demand high amounts of blood for supply of the tumor with nutrients and oxygen. Therefore, a high percentage of combined cardiac output is sacrificed exclusively for the perfusion of the teratoma, causing high cardiac output; furthermore, increased cardiac output may occur due to intratumoral arteriovenous shunting. Development of high-output cardiac failure as demonstrable by Doppler techniques may occur in about one-third of cases with sacrococcygeal

504

Fetal Cardiology

(a)

(b)

(c)

(d)

Figure 34.6

(a) Large solid sacral teratoma (dimensions 100 × 100 × 110 mm) with severe high-output cardiac failure in a fetus at 23 + 3 weeks’ gestation. (b) The fetal heart is markedly dilated and the cardiothoracic circumference (0.64) and the cardiothoracic area (0.39) ratios are increased. A thick generalized skin edema is present. (c) Severe holosystolic insufficiency of both atrioventricular valves is demonstrated by color Doppler imaging. (d) The blood flow velocity pattern of the ductus venosus already shows increased pulsatility.

teratoma (Figure 34.6). Not so common is the subsequent development of generalized hydrops fetalis and placentae being correlated with a very high spontaneous death rate.224 Hydrops has also been reported in fetuses with teratoma of other locations.144–152,225–229 Furthermore, in some fetuses, mostly mild or moderate anemia may be detected, which is caused by spontaneous hemorrhages into the tumor or by the Kasabach–Merritt sequence with signs of microangiopathic hemolytic anemia and consumption coagulopathy. In utero treatment of a fetus with a teratoma developing hydrops depends on several factors. In particular, cases with a large intracorporeal portion compressing and displacing other organs have a poorer prognosis than cases with predominantly extracorporeal components. In fetuses with a sacrococcygeal teratoma of the extracorporeal type, the presence of a completely solid tumor with its risk for malignancy

and hypervascularization seems to be an important negative prognostic factor.230 Intrauterine transfusion of packed erythrocytes for correction of anemia and transplacental digitalization may be considered. Ligation, embolization, perivascular sclerosis, and coagulation of the supplying arteries interrupting the blood supply or resection of the tumor by open fetal surgery or fetoscopy seem to constitute a causative treatment approach for correcting high-output cardiac failure in immature fetuses. Prenatal debulking of the tumor and devascularization in premature fetuses with sacrococcygeal teratoma and subsequent hydrops may significantly reduce the cardiac output and thus be the best treatment option.231 A poor prognostic sign is the development of hydrops in an elsewhere localized teratoma, such as a mediastinal225,230 or intrapericardial teratoma,144–152,229 where local compression of the heart, cardiac tamponade, lymphatic drainage,

Cardiac diseases in association with hydrops fetalis

and venous return may occur in addition, similarly to other thoracic masses, such as fibrosarcoma, with origin from the lungs or mediastinum.232 Arteriovenous malformations and vascular tumors may also cause hydrops. Examples are liver hemangiomas, cavernous hemangiomas of the chest, nuchal hemangiomas, diffuse neonatal hemangiomatosis,233,234 intracerebral hemangioma,235 umbilical cord hemangioma,236,237 placental chorioangioma,238–240 and diffuse placental chorioangiomatosis.241 The most important pathomechanism is a high-output cardiac failure due to arteriovenous shunts inside the tumor. In addition, anemia due to a Kasabach– Merritt sequence and/or hemorrhage may occur in these tumors and increase the high cardiac output state. Especially fetuses with a giant hemangioma of the liver234,242–246 or neck,247 with diffuse hemangiomatosis,233,234 and with Klippel–Trenaunay syndrome248,249 may develop hydrops and Kasabach–Merritt sequence with microangiopathic hemolytic anemia, thrombocytopenia, and consumptive coagulopathy. In these situations, treatment options for immature fetuses developing hydrops may be the intrauterine transfusion of packed erythrocytes, if anemia is present, and of thrombocytes, if severe thrombocytopenia is present, and digitalization and high doses of placentacrossing corticosteroids.250,251 Intracerebral arteriovenous malformations may involve and dilate the vein of Galen (‘aneurysm’ of the vein of Galen), resulting in a significant shunting of blood volume and high-output cardiac failure characterized by cardiomegaly, AV valve insufficiency, and development of hydrops fetalis and placentae as well as polyhydramnios.203,252 In addition, brain damage may occur if there is a relevant arteriovenous blood shunting that bypasses the brain parenchyma (‘steal’ phenomenon). Sinus venosus atrial septal defect with partially anomalous pulmonary venous return and discrete aortic coarctation are probably associated with the increased blood flow through the low resistance circuit of the vein of Galen early in gestation.253 Palliative treatment with digoxin seems to be the only sensible approach in immature non-viable fetuses with high-output cardiac failure. Other examples of secondary dilated cardiomyopathies with more direct myocardial damage may be observed in fetuses with infections, for example coxsackievirus,254 adenovirus,255,256 and parvovirus B19,257–265 or by maternal autoantibody-induced myocarditis, mostly associated with atrioventricular block.176–178,182–185 Trypanosoma cruzi, the agent of Chagas disease in Latin America, may in utero transmit to the fetus, resulting in placentitis, fetal anemia, infection of fetal organs including myocarditis, and hydrops.266,267 Significant hypertrophic cardiomyopathy has rarely been reported in utero, particularly in association with maternal diabetes, Noonan syndrome, and lysosomal and glycogen storage diseases. A secondary cardiomyopathy of dilated but also hypertrophic type can more frequently be found in the recipient twin of

505

twin–twin transfusion syndrome and in fetuses with long-standing tachy- and bradyarrhythmia. In the recipient twin of chronic fetofetal transfusion syndrome, increased pulsatility of the systemic veins suggests elevated central venous blood pressure. Elevated afterload and preload by arterial hypertension and hypervolemia, respectively, cause significant ventricular pressure and volume load. Biventricular myocardial hypertrophy, cardiomegaly, and AV valve incompetence as further signs for congestive heart failure usually appear before the development of hydrops.268–272 In contrast, generalized hydrops, AV valve incompetence, and increased pulsatility of venous blood flow are extremely rare in the donor fetus,272 which may show only occasionally isolated mild to moderate pericardial effusion and rarely hydrops if severe anemia is present. Furthermore, transient hydrops and significant increase of venous pulsatility may occur in some donor fetuses after laser coagulation for severe twin–twin transfusion syndrome and indicate a hemodynamic adaptation response following interruption of the transfusion process.273 Depending on the underlying disease, significant cardiac enlargement and/or myocardial hypertrophy may be present in fetuses with cardiomyopathy. If fetal hydrops is associated, additional signs of congestive heart failure are usually identifiable. These include reduced ventricular fractional shortening, tricuspid and/or mitral valve regurgitation, and increased pulsatility of the venous Doppler blood flow pattern. Locally circumscribed or more general hyperechogenicity of the ventricular myocardium may suggest myocardial infection. In all cases of fetal cardiomyopathy a detailed search for the underlying disease should be performed, because the exact diagnosis is the key to the prognosis, perinatal management, and adequate counseling of the parents. In some secondary cardiomyopathies, in utero treatment may be successful. This includes intrauterine blood transfusion in fetuses with anemia, transplacental treatment with corticosteroids in fetuses with myocarditis, or laser coagulation of placental anastomosis in fetuses with twin–twin transfusion syndrome. In addition, transplacental treatment with digoxin may result in a non-specific increase of myocardial inotropy causing remission of hydrops or stabilization of the fetal condition in cases with unclear etiology. Digoxin is known to decrease the catecholamine response to congestive heart failure and may improve filling and low filling pressure, if there is diastolic dysfunction in the fetus.47 If there is an elevated afterload, however, increase in oxygen consumption could result from increased inotropy without improved myocardial perfusion.47 On the other hand, it should be kept in mind that the prognosis is usually poor in fetuses with metabolic storage diseases or primary cardiomyopathy, if hydrops develops in utero. Therefore, aggressive in utero and perinatal management must be carefully considered and may only be performed with informed consent by the extensively counseled parents.

506

Fetal Cardiology

Hydrops fetalis in early gestation In the first and early second trimester, many more hydropic fetuses have chromosomal abnormalities compared to hydropic fetuses of the late second and third trimester.17,274 Almost all of these fetuses have increased nuchal translucency thickness or cystic hygromata colli; the latter represent distended fluid-filled lymphatic vessels in the nuchal region, frequently associated with generalized malformation of the lymphatic vessels resulting in hydrothorax, ascites, and generalized skin edema.275 The vast majority of hydropic fetuses die spontaneously. Some fetuses with complex cardiac malformations developing hydrops die spontaneously early in pregnancy. These include fetuses with severe bivalvular stenoses of the pulmonary and aortic valve276 and fetuses with an absent pulmonary valve and patent arterial duct. Furthermore, in early pregnancy, fetuses with trisomy 21 very often have transient nuchal edema,277 but in some cases a generalized skin edema and hydrops fetalis may occur. Both groups contain fetuses without cardiac defects, although the incidence of atrioventricular septal defects and ventricular septal defects is much higher in fetuses with increased nuchal translucency and generalized hydrops.278 Also, in the group of euploid fetuses, nuchal translucency thickness correlates with the prevalence of congenital heart disease.279 Therefore, increased nuchal translucency has been proposed as a screening method for fetal congenital heart diseases in chromosomally normal fetuses.279 In the majority of both chromosomally abnormal and normal fetuses with isolated increased nuchal translucency, the nuchal fluid accumulation disappears after 14 weeks, most probably because of more efficient drainage of the interstitial fluid by the lymphatic system, increasing myocardial performance, and/or the drop in peripheral arterial resistance.277,280 In some hydropic fetuses, but also in fetuses with isolated nuchal edema, the presence of AV valve regurgitation and abnormal venous Doppler flow velocity waveforms suggest cardiac congestive heart failure as a pathomechanism for the development of nuchal edema and hydrops.281,282 On the other hand, in some hydropic fetuses with trisomy 21, only cardiac defects seem to be associated. In these fetuses and in hydropic trisomy 21 fetuses without cardiac anomalies, other pathomechanisms have to be discussed. These include a transient myeloproliferative disorder and maldevelopment of lymphatic vessels, especially in cases of isolated hydro-/chylothorax that may be found in fetuses with trisomy 21, hypoxia-induced capillary damage, a transient myeloproliferative disorder, and/or altered collagen formation due to a dose effect from responsible genes on chromosome 21. Nuchal edema and hydrops may be associated with other autosomal trisomies, such as trisomies 18 and 13, and triploidy.276 Although the isolated nuchal edema may

be transient, hydrops already occurring in the first and early second trimester is associated with spontaneous abortion in the vast majority of these fetuses. Septated cystic hygromata colli are associated with monosomy X (Turner syndrome) in two-thirds of fetuses and with a normal karyotype in the majority of the residual cases.274 Because of a more generalized maldevelopment of the lymphatic vessels, hydrothorax and also hydrops often occur, and the fetuses die during the next few gestational weeks. In these fetuses a tubular coarctation of the aorta may frequently be present, probably resulting from early compression of the aortic arch by the distended lymphatic vessels.277

Conclusion During fetal life, fetal hydrops indicates a severe disease, which can easily be detected by obstetric ultrasound. The enormous spectrum of underlying diseases requires a high level of diagnostic knowledge. The prognosis of hydropic fetuses is crucially dependent on the etiology of hydrops, which has to be identified for adequate perinatal management, assessment of prognosis, and counseling of the parents. The examination of the fetal cardiovascular system is one of the most important steps in this context. Cardiovascular diseases may frequently be the etiological disorder in fetal hydrops. Additionally, signs of congestive heart failure may be associated with many extracardiac causes of fetal hydrops. Fetal echocardiography allows repetitive assessment to be made of fetal cardiac function, providing important information about prognosis, fetal surveillance, and monitoring of therapeutic measures. Therefore, fetal hydrops demonstrates the need for a profound knowledge of perinatal cardiology to achieve adequate management. The effective, speedy, and stepwise diagnostic approach from less to more invasive techniques is a valid indicator of the competence of any prenatal care center.

References 1.

2.

3.

Holzgreve W. The fetus with nonimmune hydrops. In: Harrison MR, Golbus MS, Filly RA, eds. The Unborn Patient. Prenatal Diagnosis and Treatment, 2nd edn. Philadelphia: WB Saunders, 1990: 228–45. Hutchinson AA, Drew JE, Yu VYH et al. Nonimmunologic hydrops fetalis: a review of 61 cases. Obstet Gynecol 1982; 59: 347–52. McCoy MC, Katz VL, Gould N, Kuller JA. Non-immune hydrops after 20 weeks’ gestation: review of 10 years’ experience with suggestions for management. Obstet Gynecol 1995; 85: 578–82.

Cardiac diseases in association with hydrops fetalis

4.

5.

6.

7.

8.

9. 10.

11. 12. 13.

14.

15.

16.

17.

18. 19.

20.

21. 22.

23.

Holzgreve W, Curry CJR, Golbus MS et al. Investigation of nonimmune hydrops fetalis. Am J Obstet Gynecol 1984; 148: 805–12. Holzgreve W, Holzgreve B, Curry CJR. Nonimmune hydrops fetalis: diagnosis and management. Semin Perinatol 1985; 9: 52–67. Hansmann M, Gembruch U, Bald R. New therapeutic aspects in nonimmune hydrops fetalis based on four hundred and two prenatally diagnosed cases. Fetal Ther 1989; 4: 29–36. Hansmann M, Gembruch U, Bald R. Management of the fetus with nonimmune hydrops. In: Harrison MR, Golbus MS, Filly RA, eds. The Unborn Patient. Prenatal Diagnosis and Treatment, 2nd edn. Philadelphia: WB Saunders, 1990: 246–8. Machin GA. Hydrops revisited: literature review of 1,414 cases published in the 1980s. Am J Med Genet 1989; 34: 366–90. Jones DC. Nonimmune fetal hydrops: diagnosis and obstetrical management. Semin Perinatol 1995; 19: 447–61. Tercanli S, Gembruch U, Holzgreve W. Nonimmune hydrops fetalis: diagnosis and management. In: Callen PW, ed. Ultrasonography in Obstetrics and Gynecology, 4th edn. Philadelphia: WB Saunders, 2000: 551–75. Forouzan I. Hydrops fetalis: recent advances. Obstet Gynecol Surv 1997; 52: 130–8. Norton ME. Nonimmune hydrops fetalis. Semin Perinatol 1994; 18: 321–32. Macafee CA, Fortune DW, Beischer NA. Nonimmunological hydrops fetalis. J Obstet Gynaecol Br Commonw 1970; 77: 226–37. Maidman JE, Yeager C, Anderson V et al. Prenatal diagnosis and management of nonimmunological hydrops fetalis. Obstet Gynecol 1980; 56: 571–6. Snijders RJM, Holzgreve W, Cuckle H, Nicolaides KH. Maternal age-specific risks for trisomies at 9–14 weeks’ gestation. Prenat Diagn 1994; 14: 543–52. Anandakumar C, Biswas A, Wong YC et al. Management of non-immune hydrops: 8 years’ experience. Ultrasound Obstet Gynecol 1996; 8: 196–200. Iskaros J, Jauniaux E, Rodeck C. Outcome of nonimmune hydrops fetalis diagnosed during the first half of pregnancy. Obstet Gynecol 1997; 90: 321–5. Jauniaux E. Diagnosis and management of early nonimmune hydrops fetalis. Prenat Diagn 1997; 17: 1261–8. Im SS, Rizos N, Joutsi P, Shime J, Benzie RL. Nonimmunologic hydrops fetalis. Am J Obstet Gynecol 1984; 148: 566–9. Wafelman LS, Pollock BH, Kreutzer J, Richards DS, Hutchinson AA. Nonimmune hydrops fetalis: fetal and neonatal outcome during 1983–1992. Biol Neonate 1999; 75: 73–81. Knilans TK. Cardiac abnormalities associated with hydrops fetalis. Semin Perinatol 1995; 19: 483–92. Arcasoy MO, Gallagher PG. Hematologic disorders and nonimmune hydrops fetalis. Semin Perinatol 1995; 19: 502–15. Gembruch U, Holzgreve W. The fetus with nonimmune hydrops. In: Harrison MR, Evans MI, Adzick NS, Holzgreve W, eds. The Unborn Patient: The Art and

24. 25.

26.

27.

28.

29.

30. 31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

507

Science of Fetal Therapy, 3rd edn. Philadelphia: WB Saunders, 2001: 525–82. Apkon M. Pathophysiology of hydrops fetalis. Semin Perinatol 1995; 19: 437–46. Wu PYK. Colloid osmotic pressure and osmoregulation in the pregnant woman, fetus, and neonate. In: Polin RA, Fox WW, eds. Fetal and Neonatal Physiology, 2nd edn. Philadelphia: WB Saunders, 1998: 1721–6. Brace RA. Fluid distribution in the fetus and newborn. In: Polin RA, Fox WW, eds. Fetal and Neonatal Physiology, 2nd edn. Philadelphia: WB Saunders, 1998: 1703–13. Anderson PAW, Kleinman CS, Lister G, Talner NS. Cardiovascular function during normal fetal and neonatal development and with hypoxic stress. In: Polin RA, Fox WW, eds. Fetal and Neonatal Physiology, 2nd edn. Philadelphia: WB Saunders, 1998: 837–90. Gold PS, Brace RA. Fetal whole-body permeability–surface area product and reflection coefficient for plasma protein. Microvasc Res 1988; 36: 262–74. Brace RA, Gold PS. Fetal whole-body interstitial compliance, vascular compliance, and capillary filtration coefficient. Am J Physiol 1984; 247: R800–5. Brace RA. Effect of outflow pressure on fetal lymph flow. Am J Obstet Gynecol 1989; 160: 494–7. Gest AL. The effect of outflow pressure upon thoracic duct lymph flow rate in fetal sheep. Pediatr Res 1992; 32: 585–8. Nageotte MP, Hurwitz SR, Kaupke CJ, Vaziri ND, Pandian MR. Atriopeptin in the twin-to-twin transfusion syndrome. Obstet Gynecol 1989; 73: 867–70. Ville Y, Proudler A, Abbas A, Nicolaides K. Atrial natriuretic factor concentration in normal, growth retarded, anemic, and hydropic fetuses. Am J Obstet Gynecol 1994; 171: 777–83. Brace RA, Christian JL. Transcapillary Starling pressure in the fetus, newborn, and pregnant adult. Am J Physiol 1981; 240: H843–7. Adamson SL, Whiteley KJ, Langille BL. Pulsatile pressure– flow relation and pulse–wave propagation in the umbilical circulation of fetal sheep. Circ Res 1992; 70: 761–72. Faber JJ, Anderson DF. Angiotensin mediated interaction of fetal kidney and placenta in the control of fetal arterial pressure and its role in hydrops fetalis. Placenta 1997; 18: 313–26. Johnson P, Sharland G, Allan LD, Tynan MJ, Maxwell DJ. Umbilical venous pressure in nonimmune hydrops fetalis: correlation with cardiac size. Am J Obstet Gynecol 1992; 167: 1309–13. Weiner CP. Umbilical pressure measurement in the evaluation of nonimmune hydrops fetalis. Am J Obstet Gynecol 1993; 168: 817–23. Hecher K, Campbell S. Characteristics of fetal venous blood flow under normal circumstances and during fetal disease. Ultrasound Obstet Gynecol 1996; 7: 68–83. Gudmundsson S, Huhta JC, Wood DC et al. Venous Doppler ultrasonography in the fetus with nonimmune hydrops. Am J Obstet Gynecol 1991; 164: 33–7. Reed KL, Chaffin DG, Anderson CF. Umbilical venous Doppler velocity pulsations and inferior vena cava pressure elevation in fetal lambs. Obstet Gynecol 1996; 87: 617–20.

508

42.

43.

44.

45. 46.

47. 48.

49.

50.

51.

52.

53.

54.

55.

56.

57.

58.

59.

Fetal Cardiology

Gembruch U, Redel DA, Bald R, Hansmann M. Longitudinal study in 18 cases of fetal supraventricular tachycardia: Doppler-echocardiographic findings and pathophysiological implications. Am Heart J 1993; 125: 1290–301. Saltzman DH, Frigoletto FD, Harlow BL, Barss VA, Benacerraf BR. Sonographic evaluation of hydrops fetalis. Obstet Gynecol 1989; 74: 106–11. Skoll MA, Sharmland GK, Allan LD. Is the ultrasound definition of fluid collections in the non-immune hydrops fetalis helpful in defining the underlying cause or predicting outcome? Ultrasound Obstet Gynecol 1991; 1: 309–12. Smoleniec J, James D. Predictive value of pleural effusions in fetal hydrops. Fetal Diagn Ther 1995; 10: 95–100. Huhta JC, Weil-Chalker S, Pagotto LT. Fetal congestive heart failure. In: Allan L, Hornberger LK, Sharland G, eds. Textbook of Fetal Cardiology. London: Greenwich Medical Media, 2000: 565–75. Huhta JC. Fetal congestive heart failure. Semin Fetal Neonatal Med 2005; 10: 542–52. Hofstaetter C, Hansmann M, Eik-Nes SH, Huhta JC, Luther SL. A cardiovascular profile score in the surveillance of fetal hydrops. J Matern Fetal Neonatal Med 2006; 19: 407–13. Sharland GK, Allan LD. Normal fetal cardiac measurements derived by cross-sectional echocardiography. Ultrasound Obstet Gynecol 1992; 2: 175–81. Tan J, Silverman NH, Hoffman JI, Villegas M, Schmidt KG. Cardiac dimensions determined by cross-sectional echocardiography in the normal fetus from 18 weeks to term. Am J Cardiol 1992; 70: 1459–67. Shapiro I, Degani S, Leibovitz Z et al. Fetal cardiac measurements derived by transvaginal and transabdominal cross-sectional echocardiography from 14 weeks of gestation to term. Ultrasound Obstet Gynecol 1998; 12: 404–18. Gembruch U, Shi CY, Smrcek JM. Biometry of the fetal heart between 10 and 17 weeks of gestation. Fetal Diagn Ther 2000; 15: 20–31. Paladini D, Calabro R, Palmieri S, D’Andrea T. Prenatal diagnosis of congenital heart disease and fetal karyotyping.. Obstet Gynecol 1993; 81: 679–82. Respondek M, Respondek A, Huhta JC, Wilczynski J. 2D echocardiographic assessment of the fetal heart size in the 2nd and 3rd trimester of uncomplicated pregnancy. Eur J Obstet Gynecol Reprod Biol 1992; 44: 185–8. Chaoui R, Bollmann R, Goldner B, Heling KS, Tennstedt C. Fetal cardiomegaly: echocardiographic findings and outcome in 19 cases. Fetal Diagn Ther 1994; 9: 92–104. Carlson DE, Platt LD, Medearis AL, Horenstein J. Prognostic indicators of the resolution of nonimmune hydrops fetalis and survival of the fetus. Am J Obstet Gynecol 1990; 163: 1785–7. Sutton MS, Gill T, Plappert T, Saltzman DH, Doubilet P. Assessment of right and left ventricular function in terms of force development with gestational age in the normal human fetus. Br Heart J 1991; 61: 285–9. Rizzo G, Capponi A, Rinaldo D, Arduini D, Romanini C. Ventricular ejection force in growth retarded fetuses. Ultrasound Obstet Gynecol 1995; 5: 247–52. Tsutsumi T, Ishii M, Eto G, Hota M, Kato H. Serial evaluation for myocardial performance in fetuses and neonates using a new Doppler index. Pediatr Int 1999; 41: 722–7.

60.

61.

62.

63.

64.

65.

66.

67.

68.

69.

70.

71.

72.

73.

74.

75.

Raboisson MJ, Fouron JC, Lamoureux J et al. Early intertwin differences in myocardial performance during the twin-to-twin transfusion syndrome. Circulation 2004; 110: 3043–8. Szwast A, Tian Z, McCann M et al. Impact of altered loading conditions on ventricular performance in fetuses with congenital cystic adenomatoid malformation and twin-twin transfusion syndrome. Ultrasound Obstet Gynecol 2007; 30: 40–6. Aoki M, Harada K, Ogawa M, Tanaka T. Quantitative assessment of right ventricular function using Doppler tissue imaging in fetuses with and without heart failure. J Am Soc Echocardiogr 2004; 17: 28–35. Perles Z, Nir A, Gavri S, Rein AJ. Assessment of fetal myocardial performance using myocardial deformation analysis. Am J Cardiol 2007; 99: 993–6. Spinale FG, Holzgrefe HH, Mukherjee R et al. LV and myocyte structure and function after early recovery from tachycardia-induced cardiomyopathy. Am J Physiol 1995; 268: H836–47. Packer DL, Bardy GH, Worley SJ et al. Tachycardiainduced cardiomyopathy: a reversible form of left ventricular dysfunction. Am J Cardiol 1986; 57: 563–70. Spinale FG, Tanaka R, Crawford FA, Zile MR. Changes in myocardial blood flow during development of and recovery from tachycardia-induced cardiomyopathy. Circulation 1992; 85: 717–29. Tomita M, Spinale FG, Crawford FA, Zile MR. Changes in left ventricular volume, mass, and function during the development and regression of supraventricular tachycardia-induced cardiomyopathy: disparity between recovery of systolic versus diastolic function. Circulation 1991; 83: 635–44. Respondek K, Kammermeier M, Ludomirsky A, Weil S, Huhta J. The prevalence and clinical significance of fetal tricuspid valve regurgitation with normal heart anatomy. Am J Obstet Gynecol 1994; 171: 1265–70. Gembruch U, Smrcek JM. The prevalence and clinical significance of tricuspid valve regurgitation in normally grown fetuses and those with intrauterine growth retardation. Ultrasound Obstet Gynecol 1997; 9: 374–82. Smrcek JM, Gembruch U. Longitudinal observations in normally grown fetuses with tricuspid valve regurgitation: report of 22 cases. Prenat Diagn 1999; 19: 197–204. Smrcek JM, Germer U, Gembruch U. Functional pulmonary valve regurgitation in the fetus. Ultrasound Obstet Gynecol 1998; 12: 254–9. Gembruch U, Knöpfle G, Chatterjee M et al. Prenatal diagnosis of atrioventricular canal malformation using up-to-date echocardiographic technology (a report of 14 cases). Am Heart J 1991; 121: 1489–97. Tulzer G, Gudmundsson S, Rotondo KM et al. Doppler in the evaluation and prognosis of fetuses with tricuspid regurgitation. J Matern Fetal Invest 1991; 1: 15–18. Reed KL, Chaffin DG, Anderson CF, Newman AT. Umbilical venous velocity pulsations are related to atrial contraction pressure waveforms in fetal lambs. Obstet Gynecol 1997; 89: 953–6. Gudmundsson S, Tulzer G, Huhta JC, Marsal K. Venous Doppler in the fetus with absent end-diastolic flow in the

Cardiac diseases in association with hydrops fetalis

76.

77.

78.

79.

80.

81.

82.

83.

84.

85.

86.

87.

88.

89.

90.

91.

92.

93.

umbilical artery. Ultrasound Obstet Gynecol 1996; 7: 262–7. Baschat AA, Güclü S, Kush ML et al. Venous Doppler in the prediction of acid-base status of growth-restricted fetuses with elevated placental blood flow resistance. Am J Obstet Gynecol 2004; 191: 277–84. Berg C, Kremer C, Geipel A et al. Ductus venosus blood flow alterations in fetuses with obstructive lesions of the right heart. Ultrasound Obstet Gynecol 2006; 28: 137–42. Tulzer G, Gudmundsson S, Wood DC et al. Doppler in non-immune hydrops fetalis. Ultrasound Obstet Gynecol 1994; 4: 279–83. Gembruch U, Krapp M, Baumann P. Changes of venous blood flow velocity waveforms in fetuses with supraventricular tachycardia. Ultrasound Obstet Gynecol 1995; 5: 394–9. Krapp M, Gembruch U, Baumann P. Venous blood flow pattern suggesting tachycardia-induced ‘cardiomyopathy’ in the fetus. Ultrasound Obstet Gynecol 1997; 10: 32–40. Gembruch U, Krapp M, Germer U, Baumann P. Venous Doppler in the sonographic surveillance of fetuses with supraventricular tachycardia. Eur J Obstet Gynecol Reprod Biol 1999; 84: 187–92. Lutin WA, Brumund MR, Jones C et al. Hemodynamic abnormalities in fetuses with congenital heart disease. Pediatr Cardiol 1999; 20: 390–5. McFadden DE, Taylor GP. Cardiac abnormalities and nonimmune hydrops fetalis: a coincidental, not causal, relationship. Pediatr Pathol 1989; 9: 11–17. Attenhofer Jost CH, Connolly HM, Edwards WD et al. Ebstein’s anomaly – review of a multifaceted congenital cardiac condition. Swiss Med Wkly 2005; 135: 269–81. Pavlova M, Fouran JC, Drblik SP et al. Factors affecting the prognosis of Ebstein’s anomaly during fetal life. Am Heart J 1998; 135: 1081–5. McElhinney DB, Salvin JW, Colan SD et al. Improving outcomes in fetuses and neonates with congenital displacement (Ebstein’s malformation) or dysplasia of the tricuspid valve. Am J Cardiol 2005; 96: 582–6. Sharland GK, Chita SK, Allan LD. Tricuspid valve dysplasia or displacement in intrauterine life. J Am Coll Cardiol 1991; 17: 944–9. Lang D, Oberhoffer R, Cook A et al. The pathological spectrum of malformations of the tricuspid valve in preand neonatal life. J Am Coll Cardiol 1991; 17: 1161–7. Tanaka T, Yammaki S, Ohno T et al. The histology of the lung in neonates with tricuspid valve disease and gross cardiomegaly due to severe regurgitation. Pediatr Cardiol 1998; 19: 133–8. Robertson DA, Silverman NH. Ebstein’s anomaly: echocardiographic and clinical features in the fetus and neonate. J Am Coll Cardiol 1989; 14: 1300–7. Atkins DL, Clark EB, Marvin WJ Jr. Foramen ovale/ atrial septum area ratio: a marker of transatrial blood flow. Circulation 1982; 66: 281–3. Allan LD, Sharland GK, Milburn A et al. Prospective diagnosis of 1,006 consecutive cases of congenital heart disease in the fetus. J Am Coll Cardiol 1994; 23: 1452–8. Hornberger LK, Sahn DJ, Kleinman CS, Copel JA, Reed KL. Tricuspid valve disease with significant tricuspid

94.

95.

96.

97. 98.

99.

100.

101.

102.

103.

104.

105.

106.

107.

108.

109.

509

insufficiency in the fetus: diagnosis and outcome. J Am Coll Cardiol 1991; 17: 167–73. Koike T, Minakami H, Shiraishi H et al. Digitalization of the mother in treating hydrops fetalis in monochorionic twin with Ebstein’s anomaly. Case report. J Perinat Med 1997; 25: 295–7. Gembruch U, Hansmann M, Bald R. Fetales Vorhofflattern bei komplettem atrioventrikulärem Kanal und Trisomie 18. Geburtshilfe Frauenheilkd 1988; 48: 445–7. Gembruch U, Hansmann M, Redel DA, Bald R, Knöpfle G. Fetal complete heart block: antenatal diagnosis, significance and management. Eur J Obstet Gynecol Reprod Biol 1989; 31: 9–22. Machado MVL, Tynan MJ, Curry PVL, Allan LD. Fetal complete heart block. Br Heart J 1988; 60: 512–15. Baschat AA, Gembruch U, Knöpfle G, Hansmann M. First trimester fetal heart block: a marker for cardiac anomaly. Ultrasound Obstet Gynecol 1999; 14: 311–14. Schmidt KG, Ulmer HE, Silverman NH, Kleinman CS, Copel JA. Perinatal outcome of fetal complete heart block: a multicenter experience. J Am Coll Cardiol 1991; 17: 1360–6. Ettedgui JA, Sharland GK, Chita SK et al. Absent pulmonary valve syndrome with ventricular septal defect: role of the arterial duct. Am J Cardiol 1990; 66: 233–4. Persutte WH, Yeasting RA, Lenke RR, Levine MM. Prenatal ultrasonographic appearance of ‘agenesis’ of the ductus arteriosus and pulmonic valve hypoplasia. A case report and review of the embryogenesis. J Ultrasound Med 1990; 9: 541–5. Liang CC, Tsai CC, Hsieh CC, Hseuh C, Soong YK. Prenatal diagnosis of tetralogy of Fallot with absent pulmonary valve accompanied by hydrops fetalis. Gynecol Obstet Invest 1997; 44: 61–3. Berg C, Thomsen Y, Geipel A, Germer U, Gembruch U. Reversed end-diastolic flow in the umbilical artery at 10–14 weeks of gestation is associated with absent pulmonary valve syndrome. Ultrasound Obstet Gynecol 2007; 30: 254–8. Razavi RS, Sharland GK, Simpson JM. Prenatal diagnosis by echocardiogram and outcome of absent pulmonary valve syndrome. Am J Cardiol 2003; 91: 429–32. Volpe P, Paladini D, Marasini M et al. Characteristics, associations and outcome of absent pulmonary valve syndrome in the fetus. Ultrasound Obstet Gynecol 2004; 24: 623–8. Galindo A, Gutierrez-Larraya F, Martinez JM et al. Prenatal diagnosis and outcome for fetuses with congenital absence of the pulmonary valve. Ultrasound Obstet Gynecol 2006; 28: 32–9. Cook AC, Fagg NLK, Ho SY et al. Echocardiographic– anatomical correlations in aorto-left ventricular tunnel. Br Heart J 1995; 74: 443–8. Sousa-Uva M, Touchot A, Fermont L et al. Aortico-left ventricular tunnel in fetuses and infants. Ann Thorac Surg 1996; 61: 1805–10. Freund MW, Stoutenbeek P, van der Laan M et al. Aortico-right ventricular tunnel: prenatal diagnosis leading to neonatal survival. Fetal Diagn Ther 2007; 2: 335–8.

510

110.

111.

112.

113.

114.

115.

116.

117.

118.

119.

120.

121.

122.

123.

124.

125.

126.

Fetal Cardiology

Huhta JC. The fetal ductus arteriosus. In: Copel JA, Reed KL, eds. Doppler Ultrasound in Obstetrics and Gynecology. New York: Raven Press, 1994: 325–31. Harada K, Rice MJ, Shiota T et al. Two-dimensional echocardiographic evaluation of ventricular systolic function in human fetuses with ductal constriction. Ultrasound Obstet Gynecol 1997; 10: 247–53. Hofstadler G, Tulzer G, Altmann R et al. Spontaneous closure of the human fetal ductus arteriosus – a cause of fetal congestive heart failure. Am J Obstet Gynecol 1996; 174: 879–83. Leal SD, Cavalle-Garrido T, Rynan G et al. Isolated ductal closure in utero diagnosed by fetal echocardiography. Am J Perinatol 1997; 14: 205–10. Harlass FE, Duff P, Brady K, Read J. Hydrops fetalis and premature closure of the ductus arteriosus: a review. Obstet Gynecol Surv 1989; 44: 541–3. Chaoui R, Hoffmann H, Bollmann R, Wauer R. Pränatale Diagnose eines vorzeitigen Verschlusses des Ductus arteriosus mit nachfolgender Entwicklung eines nichtimmunologischen Hydrops fetalis (NIHF). Geburtshilfe Frauenheilkd 1989; 49: 1096–8. Tulzer G, Gudmundsson S, Tewes G, Wood DC, Huhta JC. Incidence of indomethacin-induced human fetal ductal constriction. J Matern Fetal Invest 1992; 1: 267–9. Moise KJ. Effect of advancing gestational age on the frequency of fetal ductal constriction in association with maternal indomethacin use. Am J Obstet Gynecol 1993; 168: 1350–3. Tulzer G, Gudmundsson S, Rotondo K et al. Doppler echocardiography of fetal ductus arteriosus constriction versus increased right ventricular output. J Am Coll Cardiol 1991; 18: 532–6. Kohl T, Sharland GK, Allan LD et al. World experience of percutaneous ultrasound-guided balloon valvuloplasty in human fetus with severe aortic valve obstruction. Am J Cardiol 2000; 85: 1230–3. Olson RW, Nishibatake M, Arya S, Gilbert EF. Nonimmunologic hydrops fetalis due to intrauterine closure of fetal foramen ovale. Birth Defect 1987; 23: 433–42. Gembruch U, Chatterjee M, Bald R et al. Prenatal diagnosis of aortic atresia by colour Doppler flow mapping. Prenat Diagn 1990; 10: 211–17. Bharati S, Patel AG, Varga P, Husain AN, Lev M. In utero echocardiographic diagnosis of premature closure of the foramen ovale with mitral regurgitation and large left atrium. Am Heart J 1991; 122: 597–600. Bitar F, Byrum CJ, Kveseli DA, Lawrence DA, Smith FC. In utero management of hydrops fetalis caused by critical aortic stenosis. Am J Perinatol 1997; 14: 389–91. Williams IA, Kleinman CS. Is hydrops fetalis a manifestation of fetal pulmonary edema caused by impaired lymphatic drainage? Ultrasound Obstet Gynecol 2008; 31: 96–9. Marshall AC, van der Velde ME, Tworetzky W et al. Creation of an atrial septal defect in utero for fetuses with hypoplastic left heart syndrome and intact or highly restrictive atrial septum. Circulation 2004; 110: 253–8. Kohl T, Müller A, Tcchtcheava K, Achenbach S, Gembruch U. Fetal transesophageal echocardiography: clinical introduction as a monitoring tool during cardiac

127.

128.

129.

130.

131.

132.

133.

134.

135.

136.

137.

138.

139.

140.

141.

142.

143.

144.

intervention in a human fetus. Ultrasound Obstet Gynecol 2005; 26: 780–5. Vlahos AP, Look JE, McElhinney DB, van der Welde ME. Hypoplastic left heart syndrome with intact or highly restrictive atrial septum: outcome after neonatal transcatheter septostomy. Circulation 2004; 109: 2326–30. Redel DA, Hansmann M. Fetal obstruction of the foramen ovale detected by two-dimensional Doppler echocardiography. In: Rijsterbourgh H, ed. Echocardiography. The Hague: Martinus Nijhuis Publishers, 1981: 425–9. Pesonen E, Haavisto H, Ammala P, Teramo K. Intrauterine hydrops caused by premature closure of the foramen ovale. Arch Dis Child 1983; 58: 1015–16. Jouk PS, Rossignol AM, Bost M. L’ostium secundum restrictif: un nouveau syndrome malformatif foetal. Arch Mal Coeur 1987; 80: 538–42. Chobot V, Hornberger LK, Hagen-Ansert S, Sahn DJ. Prenatal detection of restrictive foramen ovale. J Am Soc Echocardiogr 1990; 3: 15–19. Better DJ, Apfel HD, Zidere V, Allan LD. Pulmonary venous blood flow patterns in the hypoplastic left heart syndrome. Heart 1999; 81: 646–50. Buis-Liem TN, Ottenkamp J, Meerman RH, Verwey R. The occurrence of fetal supraventricular tachycardia and obstruction of the foramen ovale. Prenat Diagn 1987; 7: 425–31. Phillipos EZ, Robertson MA, Still DK. Prenatal detection of foramen ovale obstruction without hydrops fetalis. J Am Coll Echocardiogr 1990; 3: 342–4. Calhoun BC, Watson PT, Hegge F. Ultrasound diagnosis of an obstructive cardiac rhabdomyoma with severe hydrops fetalis and hypoplastic lungs. J Reprod Med 1991; 36: 317–19. Groves AM, Fagg NL, Cook AC, Allan LD. Cardiac tumours in intrauterine life. Arch Dis Child 1992; 67: 1189–92. Geipel A, Krapp M, Germer U, Becker R, Gembruch U. Perinatal diagnosis of cardiac tumors. Ultrasound Obstet Gynecol 2001; 17: 17–21. Holley DG, Martin GR, Brenner JI et al. Diagnosis and management of fetal cardiac tumors: a multicenter experience and review of published reports. J Am Coll Cardiol 1995; 26: 516–20. Beghetti M, Gow RM, Haney I et al. Pediatric primary benign cardiac tumors. A 15-year review. Am Heart J 1997; 134: 1107–14. Sgro M, Barozzino T, Toi A et al. Prenatal detection of cerebral lesions in a fetus with tuberous sclerosis. Ultrasound Obstet Gynecol 1999; 14: 356–9. Sonigo P, Elmaleh A, Fermont L et al. Prenatal MRI diagnosis of fetal cerebral tuberous sclerosis. Pediatr Radiol 1996; 26: 1–4. Platt LD, Geierman CA, Turkel SB, Young G, Keegan KA. Atrial hemangioma and hydrops fetalis. Am J Obstet Gynecol 1981; 141: 107–9. Wedemeyer AL, Breitfeld V. Cardiac neoplasm, tachyarrhythmia, and anasarca in an infant. Am J Dis Child 1975; 129: 738–41. Kamil D, Geipel A, Dchmitz C et al. Fetal pericardial teratoma causing cardiac insufficiency: prenatal diagnosis and therapy. Ultrasound Obstet Gynecol 2006; 28: 972–3.

Cardiac diseases in association with hydrops fetalis

145.

146.

147.

148.

149.

150.

151.

152.

153. 154.

155.

156.

157.

158.

159.

160.

161.

162.

Rheuban KS, McDaniel NL, Feldman PS, Mayes DC, Rodgers BM. Intrapericardial teratoma causing nonimmune hydrops fetalis and pericardial tamponade: a case report. Pediatr Cardiol 1991; 12: 54–6. Perez-Aytes A, Sanchis N, Barbal A et al. Nonimmunological hydrops fetalis and intrapericardial teratoma: case report and review. Prenat Diagn 1995; 15: 859–63. Bruch SW, Adzik NS, Reiss R, Harrison MR. Prenatal therapy of pericardial teratomas. J Pediatr Surg 1997; 32: 1113–15. Riskin-Mashiah S, Moise KJ Jr, Wilkins I, Ayres NA, Fraser CD Jr. In utero diagnosis of intrapericardial teratoma: a case for in utero open fetal surgery. Prenat Diagn 1998; 18: 1328–30. Sklansky M, Greenberg M, Lucas V, Gruslin-Giroux A. Intrapericardial teratoma in twin fetus: diagnosis and management. Obstet Gynecol 1997; 89: 807–9. Benatar A, Vaughan J, Nicolini U et al. Prenatal pericardiocentesis: its role in the management of intrapericardial teratoma. Obstet Gynecol 1992; 79: 856–9. Czernik C, Stiller B, Hübler M, Hagen A, Henrich W. Hydrops fetalis caused by a large intrapericardial teratoma. Ultrasound Obstet Gynecol 2006; 28: 973–6. Bader R, Hornberger LK, Nijmeh LJ et al. Fetal pericardial teratoma: presentation of two cases and review of literature. Am J Perinatol 2006; 23: 53–8. Crum AK, Lenington W, Jeanty P. Idiopathic infantile arterial calcification. Fetus 1992; 2: 74781–6. Juul S, Ledbetter D, Wight TN, Woodrum D. New insights into idiopathic infantile arterial calcinosis. Am J Dis Child 1990; 144: 229–33. Nagar AM, Hanchate V, Tandon A, Thakkar H, Chaubal NG. Antenatal detection of idiopathic arterial calcification with hydrops fetalis. J Ultrasound Med 2003; 22: 553–659. Chong C, Hutchins G. Idiopathic infantile arterial calcification: the spectrum of clinical presentations. Pediatr Dev Pathol 2007 Aug 23;: 1 [Epub ahead of print] Lopes LM, Cha SC, Scanavacca MI, Tuma-Calil VML, Zugaib M. Fetal idiopathic ventricular tachycardia with nonimmune hydrops: benign course. Pediatr Cardiol 1996; 17: 192–3. Schleich JM, du Haut Chilly FB, Laurent MC, Almange C. Early prenatal management of a fetal ventricular tachycardia treated in utero by amiodarone with long term follow-up. Prenat Diagn 2000; 20: 449–52. Kleinman CS, Copel JA, Weinstein EM, Santulli TV Jr, Hobbins JC. Treatment of fetal supraventricular tachyarrhythmias. J Clin Ultrasound 1985; 13: 265–73. Kleinman CS, Copel JA. Electrophysiological principles and fetal antiarrhythmic therapy. Ultrasound Obstet Gynecol 1991; 1: 286–97. Ko JK, Deal BJ, Strasburger JF, Benson DW, Donovan M. Supraventricular tachycardia mechanisms and their age distribution in pediatric patients. Am J Cardiol 1992; 69: 1028–32. Naheed ZJ, Strasburger JF, Deal BJ, Benson DW Jr, Gidding SS. Fetal tachycardia: mechanisms and predictors of hydrops fetalis. J Am Coll Cardiol 1996; 27: 1736–40.

163.

164.

165.

166.

167.

168.

169.

170.

171.

172.

173.

174.

175.

176.

177.

178.

179.

180.

511

Kleinman CS, Dubin AM, Nehgme RA, Rosenfeld LE, Copel JA. Left atrial tachycardia in the human fetus: identifying the fetus with the greatest risk for developing nonimmune hydrops fetalis. Abstract presented at 2nd World Congress of Pediatric Cardiology and Cardiac Surgery, Hawaii, 1997: 281ES. Gest AL, Hansen TN, Moise AA, Hartley CJ. Atrial tachycardia causes hydrops in fetal lambs. Am J Physiol 1990; 258: H1159–63. Stevens DC, Hilliard JK, Schreiner RL et al. Supraventricular tachycardia with edema, ascites, and hydrops in fetal sheep. Am J Obstet Gynecol 1982; 142: 316–22. Nimrod C, Davies D, Harder J, Iwanicki S, Kondo S. Ultrasound evaluation of tachycardia-induced hydrops in the fetal lamb. Am J Obstet Gynecol 1987; 157: 655–9. Gest AL, Bair DK, Vander Straten MC. Thoracic duct lymph flow in fetal sheep with increased venous pressure from electrically induced tachycardia. Biol Neonate 1993; 64: 325–30. Gest AL, Martin CG, Moise AA, Hansen TN. Reversal of venous blood flow with atrial tachycardia and hydrops in fetal sheep. Pediatr Res 1990; 28: 223–6. Hansmann M, Gembruch U, Bald R, Manz M, Redel DA. Fetal tachyarrhythmias: transplacental and direct treatment of the fetus – a report of sixty fetuses. Ultrasound Obstet Gynecol 1991; 1: 162–70. Van Engelen AD, Weijtens O, Brenner JI et al. Management, outcome and follow-up of fetal tachycardia. J Am Coll Cardiol 1994; 24: 1371–5. Simpson JM, Sharland GK. Fetal tachycardias: management and outcome of 127 consecutive cases. Heart 1998; 79: 576–81. Jaeggi E, Fouron JC, Drblik SP. Fetal atrial flutter: diagnosis, clinical features, treatment, and outcome. J Pediatr 1998; 132: 336–9. Lisowski LA, Verheijen PM, Benatar AA et al. Atrial flutter in the perinatal age group: diagnosis, management and outcome. J Am Coll Cardiol 2000; 35: 771–7. Gembruch U, Hansmann M, Redel DA, Bald R. Intrauterine therapy of fetal tachyarrhythmias: intraperitoneal administration of antiarrhythmic drugs to the fetus in fetal tachyarrhythmias with severe hydrops fetalis. J Perinat Med 1988; 16: 39–44. Gembruch U, Manz M, Bald R et al. Repeated intravascular treatment with amiodarone in a fetus with refractory supraventricular tachycardia and hydrops fetalis. Am Heart J 1989; 118: 1335–8. Groves AMM, Allan LD, Rosenthal E. Outcome of isolated congenital complete heart block diagnosed in utero. Heart 1996; 75: 190–4. Richards DS, Wagman AJ, Cabaniss ML. Ascites not due to congestive heart failure in a fetus with lupus-induced heart block. Obstet Gynecol 1990; 76: 957–9. Copel JA, Buyon JP, Kleinman CS. Successful in utero therapy of fetal heart block. Am J Obstet Gynecol 1995; 173: 1384–90. Groves AM, Allan LD, Rosenthal E. Therapeutical trial of sympathomimetics in three cases of complete heart block in the fetus. Circulation 1995; 15: 3394–6. Harris JP, Alexson CG, Manning JA, Thompson HO. Medical therapy for the hydropic fetus with congenital

512

181.

182.

183.

184.

185.

186.

187.

188.

189.

190.

191.

192.

193.

194.

195.

196.

Fetal Cardiology

complete atrioventricular block. Am J Perinatol 1993; 10: 217–19. Anandakumar C, Biswas A, Chew SSL et al. Direct fetal therapy for hydrops secondary to congenital atrioventricular heart block. Obstet Gynecol 1996; 87: 835–7. Breur JM, Visser GH, Kruize AA, Stootenbeek P, Meijboom EJ. Treatment of fetal heart block with maternal steroid therapy: case report and review of the literature. Ultrasound Obstet Gynecol 2004; 24: 467–72. Berg C, Geipel A, Kohl T et al. Atrioventricular block detected in fetal life: associated anomalies and potential prognostic markers. Ultrasound Obstet Gynecol 2005; 26: 4–15. Jaeggi ET, Fouron JC, Silverman ED et al. Transplacental fetal treatment improves the outcome of prenatally diagnosed complete atrioventricular block without structural heart disease. Circulation 2004; 110: 1542–8. Maeno Y, Himeno W, Saito A et al. Clinical course of fetal congenital atrioventricular block in the Japanese population: a multicentre experience. Heart 2005; 91: 1075–9. Carpenter RJ, Strasburger JF, Garson A et al. Fetal ventricular pacing for hydrops secondary to complete heart block. J Am Coll Cardiol 1986; 8: 1434–6. Walkinshaw SA, Welch CR, McCormack J, Walsh K. In utero pacing for fetal congenital heart block. Fetal Diagn Ther 1994; 9: 183–5. Nii M, Hamilton RM, Fenwick L et al. Assessment of fetal atrioventricular time intervals by tissue Doppler and pulse Doppler echocardiography: normal values and correlation with fetal electrocardiography. Heart 2006; 92: 1831–7. Friedman DM, Kim MY, Copel JA et al; for the PRIDE investigators. Utility of cardiac monitoring in fetuses at risk for congenital heart block. The PR interval and dexamethasone evaluation (PRIDE) prospective study. Circulation 2008; 117: 485–93. Schmidt KG, Birk E, Silverman NH, Scagnelli SA. Echocardiographic evaluation of dilated cardiomyopathy in the human fetus. Am J Cardiol 1989; 63: 599–605. Grebe S, Ichida F, Grabitz R et al. Reversed pulmonary artery flow in isolated noncompaction of the ventricular myocardium. Fetal Diagn Ther 2007; 22: 29–32. Hidaka N, Tsukimori K, Hojo S et al. Transplacental digitalization for nonimmune hydrops fetalis caused by isolated noncompaction of the ventricular myocardium. J Ultrasound Med 2007; 26: 519–24. Atkin J, Snow JW, Zellweger H, Rhead WJ. Fatal infantile cardiac glycogenosis without acid maltase deficiency presenting as congenital hydrops. Eur J Pediatr 1984; 142: 150. Steenhout P, Elmer C, Clercx A et al. Carnitine deficiency with cardiomyopathy presenting as neonatal hydrops: successful response to carnitine therapy. J Inherit Metab Dis 1990; 13: 69–75. Kooper AJA, Janssens PMW, de Groot ANJA et al. Lysosomal storage diseases in non-immune hydrops fetalis pregnancies. Clin Chim Acta 2006; 371: 176–82. Rodríguez MM, Chaves F, Romaguera RL et al. Value of autopsy in nonimmune hydrops fetalis: series of 51 stillborn fetuses. Pediatr Dev Pathol 2002; 5: 365–74.

197.

198.

199.

200.

201.

202.

203.

204.

205.

206.

207.

208.

209.

210.

211.

212.

213.

214.

Igarashi H, Momoi MY, Yamagata T, Shiraishi H, Eguchi I. Hypertrophic cardiomyopathy in congenital myotonic dystrophy. Pediatr Neurol 1998; 18: 366–9. Joseph JT, Richards CS, Anthony DC et al. Congenital myotonic dystrophy pathology and somatic mosaicism. Neurology 1997; 49: 1457–60. Patel CR, Judge NE, Muise KL, Levine MM. Prenatal myocardial infarction suspected by fetal echocardiography. J Am Soc Echocardiogr 1996; 9: 721–3. Hecher K, Snijders R, Campbell S, Nicolaides K. Fetal venous, arterial, and intracardiac blood flows in red blood cell isoimmunization. Obstet Gynecol 1995; 85: 122–8. D’Ancona RL, Rahman F, Ozcan T, Copel J, Mari G. The effect of intravascular blood transfusion on the flow velocity waveform of the portal venous system of the anemic fetus. Ultrasound Obstet Gynecol 1997; 10: 333–7. Chui D, Waye S. Hydrops fetalis caused by alphathalassemia: an emerging health care problem. Blood 1998; 91: 2213–22. Ordorica SA, Marks F, Frieden FJ, Hoskins IA, Young BK. Aneurysm of the vein of Galen: a new cause for Ballantyne syndrome. Am J Obstet Gynecol 1990; 162: 1166–7. Van Selm M, Kanhai HH, Gravenhorst JB. Maternal hydrops syndrome: a review. Obstet Gynecol Surv 1991; 46: 785–8. Dorman SL, Cardwell MS. Ballantyne syndrome caused by a large placental chorangioma. Am J Obstet Gynecol 1995; 173: 1632–3. Carbillon L, Oury JF, Guerin JM, Azancot A, Blot P. Clinical biological features of Ballantyne syndrome and the role of placental hydrops. Obstet Gynecol Surv 1997; 552: 310–14. Gherman RB, Incerpi MH, Wing DA, Goodwin TM. Ballantyne syndrome: is placental ischemia the etiology? J Matern Fetal Med 1998; 7: 227–9. Duthie SJ, Walkingshaw SA. Parvovirus associated fetal hydrops: reversal of pregnancy induced proteinuric hypertension by in utero fetal transfusion. Br J Obstet Gynaecol 1995; 102: 1011–13. Ko TM, Tseng LH, Hsu PM et al. Ultrasonographic scanning of placental thickness and the prenatal diagnosis of homozygous alpha-thalassemia 1 in the second trimester. Prenat Diagn 1995; 15: 7–10. Tongsong T, Wanapirak C, Srisomboon J, Piyamongkol W, Sirichotiyakul S. Antenatal sonographic features of 100 alpha-thalassemia hydrops fetalis fetuses. J Clin Ultrasound 1996; 24: 73–7. Lam YH, Ghosh A, Tang M, Lee CP, Sin SY. Second trimester hydrops fetalis in pregnancies affected by homozygous alpha-thalassaemia-1. Prenat Diagn 1997; 17: 267–9. Lam YH, Tang M. Prenatal diagnosis of haemoglobin Bart’s disease by cordocentesis at 12–14 weeks’ gestation. Prenat Diagn 1997; 17: 501–4. Lam YH, Tang MHY, Lee CP, Tse HY. Prenatal ultrasonographic prediction of homozygous type 1 alpha-thalassemia at 12 to 13 weeks of gestation. Am J Obstet Gynecol 1999; 180: 148–50. Lam YH, Tang MH, Lee CP, Tse HY. Cardiac blood flow studies in fetuses with homozygous alpha-thalassemia-1 at 12–13 weeks of gestation. Ultrasound Obstet Gynecol 1999; 13: 48–51.

Cardiac diseases in association with hydrops fetalis

215.

216.

217.

218.

219.

220.

221.

222.

223.

224.

225.

226.

227.

228. 229.

230.

231.

Lam YH, Tang MH, Lee CP, Tse HY. Nuchal translucency in fetuses affected by homozygous alpha-thalassemia-1 at 12–13 weeks of gestation. Ultrasound Obstet Gynecol 1999; 13: 238–40. Creutzig U, Baumann M. Spontaneous remission in neonates with Down’s syndrome and acute myelogenous leukemia – transient myeloproliferative disease. Onkologie 1998; 21: 184–8. Baschat AA, Wagner T, Malisius R, Gembruch U. Prenatal diagnosis of a transient myeloproliferative disorder in trisomy 21. Prenat Diagn 1998; 18: 731–6. Zerres K, Schwanitz G, Niesen M et al. Prenatal diagnosis of acute non-lymphoblastic leukaemia in Down syndrome. Lancet 335; 1990: 117. Hendricks SK, Sorensen TK, Baker ER. Trisomy 21, fetal hydrops, and anemia: prenatal diagnosis of transient myeloproliferative disorder. Obstet Gynecol 1993; 82: 703–5. Donnenfeld AE, Scott SC, Hensefelder-Kimmel M, Dampier D. Prenatally diagnosed non-immune hydrops caused by congenital transient leukaemia. Prenat Diagn 1994; 14: 721–4. Macones GA, Johnson A, Tilley D, Wade R, Wapner R. Foetal hepatosplenomegaly associated with transient myeloproliferative disorder in trisomy 21. Fetal Diagn Ther 1995; 10: 131–3. Smrcek JM, Baschat AA, Germer U, Gloeckner-Hofmann K, Gembruch U. Fetal hydrops and hepatosplenomegaly in the second half of pregnancy: a sign of myeloproliferative disorder in fetuses with trisomy 21. Ultrasound Obstet Gynecol 2001; 17: 403–9. Hojo S, Tsukimori K, Kitade S et al. Prenatal sonographic findings and hematological abnormalities in fetuses with transient abnormal myelopoiesis with Down syndrome. Prenat Diagn 2007; 27: 507–11. Holzgreve W, Mahoney BS, Glick PL et al. Sonographic demonstration of sacrococcygeal teratoma. Prenat Diagn 1985; 5: 245–57. Weinraub Z, Gembruch U, Födisch HJ, Hansmann M. Intrauterine mediastinal teratoma associated with nonimmune hydrops fetalis. Prenat Diagn 1989; 9: 368–72. Sherer DM, Abramowicz JS, Eggers PC et al. Prenatal ultrasonographic diagnosis of intracranial teratoma and massive craniomegaly with associated high-output cardiac failure. Am J Obstet Gynecol 1993; 168: 97–9. Sherer DM, Onyeije CI. Prenatal ultrasonographic diagnosis of fetal intracranial tumors: a review. Am J Perinatol 1998; 15: 319–28. Kuller JA, Laifer SA, Martin JG et al. Unusual presentation of fetal teratoma. J Perinatol 1991; 11: 294–6. Cyr DR, Guntheroth WG, Nyberg DA et al. Prenatal diagnosis of an intrapericardial teratoma: a cause for nonimmune hydrops. J Ultrasound Med 1988; 7: 87–90. Holterman A-X, Filiatrault D, Lallier M, Youssef S. The natural history of sacrococcygeal teratomas diagnosed through routine obstetric sonogram: a single institution experience. J Pediatr Surg 1998; 33: 899–903. Graf JL, Housley T, Albanese CT, Adzik S, Harrison MR. A surprising histological evolution of preterm sacrococcygeal teratoma. J Pediatr Surg 1998; 33: 177–9.

232.

233.

234.

235.

236.

237.

238.

239.

240.

241.

242.

243.

244.

245.

246.

247.

248.

249.

250.

513

Navajas A, Astigarraga I, Fernandez-Teijeiro A et al. Hydrops fetalis and fibrosarcoma: case report of an uncommon association. Eur J Pediatr 1997; 156: 62–4. Wu TJ, Teng RJ. Diffuse neonatal haemangiomatosis with intra-uterine haemorrhage and hydrops fetalis. A case report. Eur J Pediatr 1994; 153: 759–61. Gembruch U, Baschat AA, Gloeckner-Hoffmann K, Gortner L, Germer U. Prenatal diagnosis and management of fetuses with liver hemangiomata. Ultrasound Obstet Gynecol 2002; 19: 454–60. Drut R, Sapia S, Gril D, Velasco JC, Drut RM. Nonimmune hydrops fetalis, hydramnios, microcephaly, and intracranial meningeal hemangioendothelioma. Pediatr Pathol 1993; 13: 9–13. Seifer DB, Ferguson JE 2nd, Behrens CM et al. Nonimmune hydrops fetalis in association with hemangioma of the umbilical cord. Obstet Gynecol 1985; 66: 283–6. Carles D, Maugey-Laulom B, Roux D et al. Lethal hydrops fetalis secondary to an umbilical cord hemangioma. Ann Pathol 1994; 14: 244–7. Jones CE, Rivers RP, Taghizadeh A. Disseminated intravascular coagulation and fetal hydrops in a newborn infant in association with a chorangioma of placenta. Pediatrics 1972; 50: 901–7. D’Ercole C, Cravello L, Boubli L et al. Large chorioangioma associated with hydrops fetalis: prenatal diagnosis and management. Fetal Diagn Ther 1996; 11: 357–60. Horigome H, Hamada H, Sohda S et al. Large placental chorioangiomas as a cause of cardiac failure in two fetuses. Fetal Diagn Ther 1997; 12: 241–3. Russell RT, Carlin A, Ashworth M, Welch CR. Diffuse placental chorioangiomatosis and fetal hydrops. Fetal Diagn Ther 2007; 22: 183–5. Anai T, Miyakawa I, Ohki H, Ogawa T. Hydrops fetalis caused by fetal Kasabach-Merritt syndrome. Acta Paediatr Jpn 1992; 34: 324–7. Gonen R, Fong K, Chiasson DA. Prenatal sonographic diagnosis of hepatic hemangioendothelioma with secondary nonimmune hydrops fetalis. Obstet Gynecol 1989; 73: 485–7. Skopec LL, Lakatua DJ. Non-immune fetal hydrops with hepatic hemangioendothelioma and Kasabach-Merritt syndrome: a case report. Pediatr Pathol 1989; 9: 87–93. Albano G, Pugliese A, Stabile M, Sirimarco F, Arsieri R. Hydrops foetalis caused by hepatic haemangioma. Acta Paediatr 1998; 87: 1307–9. Sharara FI, Khoury AN. Prenatal diagnosis of a giant cavernous hemangioma in association with nonimmune hydrops. A case report. J Reprod Med 1994; 39: 547–9. Nakamura Y, Komatsu Y, Yano H et al. Nonimmunologic hydrops fetalis: a clinicopathological study of 50 autopsy cases. Pediatr Pathol 1989; 7: 19–30. Mor Z, Schreyer P, Weinraub Z, Hayman E, Caspi E. Nonimmune hydrops fetalis associated with angioosteohypertrophy (Klippel–Trenaunay) syndrome. Am J Obstet Gynecol 1988; 159: 1185–6. Peng HH, Wang TH, Chao AS et al. Klippel-TrenaunayWeber syndrome involving fetal thigh: prenatal presentations and outcomes. Prenat Diagn 2006; 26: 825–30. Sheu BC, Shyu MK, Lin YF et al. Prenatal diagnosis and corticosteroid treatment of diffuse neonatal

514

251.

252.

253.

254.

255.

256.

257. 258.

259.

260. 261.

262.

263.

264. 265.

266.

267.

Fetal Cardiology

hemangiomatosis; case report. J Ultrasound Med 1994; 13: 495–9. Mejides AA, Adra AM, O’Sullivan MJ, Nicholas MC. Prenatal diagnosis and therapy for a fetal hepatic vascular malformation. Obstet Gynecol 1995; 85: 850–3. Sepulveda W, Platt CC, Fisk NM. Prenatal diagnosis of cerebral arteriovenous malformation using color Doppler ultrasonography: case report and review of the literature. Ultrasound Obstet Gynecol 1995; 6: 282–6. McElhinney DB, Halbach VV, Silverman NH, Dowd CF, Hanley FL. Congenital cardiac anomalies with vein of Galen malformations in infants. Arch Dis Child 1998; 78: 548–51. Benirschke K, Swartz WH, Leopold G, Sahn D. Hydrops due to myocarditis in a fetus. Am J Cardiovasc Pathol 1987; 1: 131–3. Towbin JA, Griffin LD, Martin AB et al. Intrauterine adenoviral myocarditis presenting as nonimmune hydrops fetalis: diagnosis by polymerase chain reaction. Pediatr Infect Dis J 1994; 13: 144–50. Ranucci-Weiss D, Uerpairojkit B, Bowles N, Towbin JA, Chan L. Intrauterine adenoviral infection associated with fetal non-immune hydrops. Prenat Diagn 1998; 18: 182–5. Porter HJ, Quantrill AM, Fleming KA. B19 parvovirus infection of myocardial cells. Lancet 1988; 5: 535–6. Naides SJ, Weiner CP. Antenatal diagnosis and palliative treatment of non-immune hydrops fetalis secondary to fetal parvovirus B19 infection. Prenat Diagn 1989; 9: 105–14. Van Elsacker-Niele AM, Salimans MM, Weiland HT et al. Fetal pathology in human parvovirus B19 infection. Br J Obstet Gynaecol 1989; 96: 768–75. Katz VL, Chescheir NC, Bethea M. Hydrops fetalis from B19 parvovirus infection. J Perinatol 1990; 10: 366–9. Morey AL, Keeling JW, Porter HJ, Fleming KA. Clinical and histopathological features of parvovirus B19 infection in the human fetus. Br J Obstet Gynaecol 1992; 99: 566–74. Brandenburg H, Los FJ, Cohen-Overbeek TE. A case of early intrauterine parvovirus B19 infection. Prenat Diagn 1996; 16: 75–7. von Kaisenberg CS, Bender G, Scheewe J et al. A case of fetal parvovirus B19 myocarditis, terminal cardiac heart failure, and perinatal heart transplantation. Fetal Diagn Ther 2001; 16: 427–32. von Kaisenberg CS, Jonat W. Fetal parvovirus B19 infection. Ultrasound Obstet Gynecol 2001; 18: 280–8. Ramirez MM, Mastrobattista JM. Diagnosis and management of human parvovirus B19 infection. Clin Perinatol 2005; 32: 697–704. Okumura M, Aparecida dos Santos V, Camargo ME, Schultz R, Zugaib M. Prenatal diagnosis of congenital Chagas’ disease (American trypanosomiasis). Prenat Diagn 2004; 24: 179–81. Coura JR. Chagas disease: what is known and what is needed – a background article. Mem Inst Oswaldo Cruz 2007; 102(Suppl 1): 113–22.

268.

269.

270.

271.

272.

273.

274.

275.

276.

277.

278.

279.

280.

281.

282.

Gembruch U, Bald R, Fahnenstich H, Hansmann M. Chronic twin–twin transfusion syndrome: a Dopplersonographic and Doppler-echocardiographic longitudinal study. J Matern Fetal Investig 1993; 3: 201 (abstr). Zosmer N, Bajoria R, Weiner E et al. Clinical and echographic features of in utero cardiac dysfunction in the recipient twin in twin–twin transfusion syndrome. Br Heart J 1994; 72: 74–9. Hecher K, Ville Y, Snijders R, Nicolaides K. Doppler studies of the fetal circulation in twin–twin transfusion syndrome. Ultrasound Obstet Gynecol 1995; 5: 318–24. Fesslova’ V, Villa L, Nava S, Mosca F, Nicolini U. Fetal and neonatal echocardiographic findings in twin–twin transfusion syndrome. Am J Obstet Gynecol 1998; 179: 1056–62. Zikulnig L, Hecher K, Bregenzer T, Bäz E, Hackelöer BJ. Prognostic factors in severe twin–twin transfusion syndrome treated by endoscopic laser surgery. Ultrasound Obstet Gynecol 1999; 14: 380–7. Gratacós E, Van Schoubroeck D, Carreras E et al. Transient hydropic signs in the donor fetus after fetoscopic laser coagulation in severe twin–twin transfusion syndrome: incidence and clinical relevance. Ultrasound Obstet Gynecol 2002; 19: 449–53. Has R. Non-immune hydrops fetalis in the first trimester: a review of 30 cases. Clin Exp Obstet Gynecol 2001; 28: 187–90. Malone FD, Ball RH, Nyberg DA et al; FASTER Trial Research Consortium. First-trimester septated cystic hygroma. Obstet Gynecol 2005; 106: 288–94. Gembruch U, Germer U, Denzel S, Gloeckner-Hofmann K. Prenatal demonstration of polyvalvular disease in a fetus with trisomy 18. Ultrasound Obstet Gynecol 2000; 16(Suppl 1): 92 (abstr). Snijders RJM, Noble P, Sebire N, Souka A, Nicolaides KH. UK multicentre project on assessment of risk of trisomy 21 by maternal age and fetal nuchal-translucency thickness at 10–14 weeks of gestation. Lancet 1998; 352: 343–6. Hyett JA, Moscoso G, Nicolaides KH. Abnormalities of the heart and great arteries in first trimester chromosomally abnormal fetuses. Am J Med Genet 1997; 69: 207–16. Hyett J, Perdu M, Sharland G, Snijders R, Nicolaides KH. Using fetal nuchal translucency to screen for major congenital cardiac defects at 10–14 weeks of gestation: population based cohort study. Br Med J 1999; 318: 81–5. Gembruch U, Knöpfle G, Bald R, Hansmann M. Early diagnosis of fetal congenital heart diseases by transvaginal echocardiography. Ultrasound Obstet Gynecol 1993; 3: 310–17. Matias A, Gomes C, Flack N, Montenegro N, Nicolaides KH. Screening for chromosomal abnormalities at 10–14 weeks: the role of ductus venosus blood flow. Ultrasound Obstet Gynecol 1998; 12: 380–4. Matias A, Huggon I, Areias JC, Montenegro N, Nicolaides KH. Cardiac defects in chromosomally normal fetuses with abnormal ductus venosus blood flow at 10–14 weeks. Ultrasound Obstet Gynecol 1999; 14: 307–10.

35 Mending the tiniest hearts: an overview Thomas Kohl Introduction Over the past three decades, revolutionary improvements in ultrasound imaging technology have allowed study of the evolution of cardiac malformations in the womb. It became apparent that, over the course of gestation, altered pressure and flow relationships within the fetal heart may result in varying degrees of underdevelopment of cardiovascular structures, often accompanied by severe structural damage.1–9 Due to the parallel arrangement of the fetal circulation, these pathological changes are in most instances compatible with normal fetal growth and development but may have an important impact on postnatal morbidity and mortality, as well as on treatment options. The detrimental prenatal disease course of severe semilunar valve obstructions, but far more often the development of life-threatening fetal cardiac failure from both cardiac and extracardiac origins, has prompted interest in the development of prenatal interventions (Figure 35.1).10–25 This chapter aims to provide a short overview of these generally experimental procedures.

Percutaneous ultrasound-guided fetal cardiac interventions Percutaneous ultrasound-guided fetal cardiac interventions require a needle to be inserted through the maternal abdomen into the uterus and then through the fetal chest wall into the heart (Figure 35.2). Following intracardiac positioning of the needle, interventional materials (e.g. guide wire, valvuloplasty catheter, pacing wire, electrode catheter) can be delivered through the needle shaft toward the cardiac region of interest. Although the approach has been used most often for fetal balloon valvuloplasties, in a few cases fetal cardiac access by percutaneous ultrasound-guided direct puncture has been chosen in order to perform potentially life-saving

treatment attempts by cardiac pacing or overdrive stimulation in fetuses with therapy-refractory arrhythmias. Until now, these latter procedures have all been in vain, due to early dislodgment of the electrode catheters or loss of capture. Hence, the success of future attempts depends on the availability of specialized equipment and using more reliable methods for lead implantation. Most recently, percutaneous ultrasound-guided direct cardiac punctures have also been introduced, with encouraging results, in fetuses with hypoplastic left heart syndrome and an intact atrial septum. In this group, the goal is to alleviate chronic pulmonary venous congestion as an important risk factor for peripartum cardiopulmonary adaptation and subsequent palliative surgery.26–28

Fetal balloon valvuloplasty Percutaneous ultrasound-guided direct punctures of the fetal heart have been employed for almost two decades in order to perform balloon valvuloplasties in fetuses with severe semilunar valve obstructions. These lesions result in pressure overload of the affected left or right ventricle. As a result, marked degrees of dilatation, hypertrophy, and endocardial fibroelastosis as well as growth failure of the obstructed ventricle and maldevelopment of associated cardiovascular structures can be observed.29–32 These pathological changes can be so profound that, by the end of gestation, the affected side of the heart cannot participate in the normal postnatal biventricular circulation, and only palliative singleventricular surgical procedures can be offered. Doppler ultrasound examination of flow directions and/or patterns across the atrial septum, mitral valve, and aortic arch permits definition of which fetuses with severe aortic valve stenosis are most likely to develop dysfunctional left ventricles before term.33,34 In these fetuses, percutaneous ultrasound-guided balloon valvuloplasties aim to restore growth and preservation of function of the affected left ventricle as well as the associated valvar and vascular structures (Figure 35.3).

516

Fetal Cardiology

(a) (a)

(b)

(b)

Figure 35.1

Figure 35.2

Fetal cardiac interventions may be performed in fetuses with severe aortic valve stenosis (a) in order to preserve left ventricular function, as well as in fetuses with severe cardiac failure from twin-to-twin transfusion syndrome (TTTS) (b). Whereas in the former group fetal cardiac intervention aims primarily at improving the postnatal prognosis, in the latter group it is performed as a life-saving procedure. LA, left atrium; RA, right atrium; LV, left ventricle; RV, right ventricle.

Percutaneous ultrasound-guided direct punctures of the fetal heart have been employed for almost two decades in order to perform balloon valvuloplasties in fetuses with severe aortic valve obstructions. (a) Following cleansing and sterile draping of the maternal abdomen, a needle is inserted into the fetal left ventricle under ultrasound control. (b) Once the needle shaft has been inserted into the left ventricle underneath the aortic valve, a small wire and balloon catheter can be advanced via the needle shaft for valve dilatation. Note the poor image quality from scatter originating from the needle shaft. In this case, the scatter is oriented away from the transducer. In even less favorable conditions, the scatter may be oriented toward the transducer, further precluding adequate visualization of the procedure.

Although the maternal morbidity of percutaneous ultrasound-guided fetal balloon valvuloplasty has been low, the procedure has been associated with significant fetal morbidity and mortality.35,36 In particular, it is poorly tolerated by hydropic fetuses. Yet, also, fetuses that are hemodynamically stable at the onset of the procedure may exhibit sudden deterioration of cardiac function due to arrhythmias, bleeding events, or unknown causes. Management strategies that have evolved over the past few years addressing these problems focus on the use of resuscitation drugs and blood products as well as technical issues.37 To date, the largest clinical experience with this procedure has been gathered by the interdisciplinary team at Boston University.

As the percutaneous ultrasound-guided procedures allow only simple maneuvers through small-dimensioned needle shafts, their technical success rate is largely dependent on fetal lie, satisfactory alignment of the needle with the region of interest (e.g. outflow tract of an obstructed ventricle), good imaging quality, and availability of sufficient acoustic windows. If satisfactory alignment cannot be achieved, the procedure is usually

Mending the tiniest hearts

(a)

517

percutaneous ultrasound-guided fetal balloon valvuloplasties for severe aortic valve stenosis will be rewarded with a biventricular circulation at delivery. This unsatisfactory outcome seems to originate from the difficulty of defining which fetus at the time of presentation is still likely to benefit from the procedure, because in the middle of the second trimester the secondary damage to the left side of the heart has often advanced to stages beyond a chance for recovery. Therefore, further research into improving the current selection criteria for fetal cardiac intervention in fetuses with severe aortic stenosis is desired. As for fetuses with left heart obstructions, a growing body of ultrasound criteria has become available in recent years in order to guide the selection process for fetal cardiac intervention in fetuses with pulmonary atresia–intact septum.15,39,40

Fetoscopy-assisted fetal cardiac interventions

(b)

Figure 35.3 Four-chamber views taken before (a) and 1 week after (b) percutaneous ultrasound-guided balloon valvuloplasty at 28+5 weeks of gestation in a fetus with severe aortic valve stenosis. Note reversal of hypertrophy and normalization of left and right ventricular proportions following the successful intervention. In this case, a normal biventricular circulation was achieved.

performed in vain; repeated cardiac punctures in this situation may then result in fetal hemopericardium and demise.35 At our center a fetus subjected to this procedure is given as much time as is necessary to attain a satisfactory position, but other colleagues have favored performing a maternal laparotomy followed by manual positioning for the same purpose.38 Apart from optimizing fetal position, using a catheter with more compliant balloon material that retains a low profile after deflation, in combination with trocar needles, has helped to prevent balloon rupture, or parts of the catheters being torn or cut off from the main body of the catheter during fetal balloon valvuloplasties.35 Despite the described improvements in management, only about 25% of fetuses surviving technically successful

Minimally invasive fetoscopic techniques that may allow alleviation of severe semilunar valve obstructions prior to 20 weeks of gestation are being developed in animal models.41–44 It is hoped that such an early intervention might yield better results than a later procedure in restoring ventricular growth and function in a larger number of fetuses with severe semilunar valve obstructions. The following sections describe their encouraging stepwise clinical introduction.41 When, during a percutaneous ultrasound-guided cardiac intervention, the needle shaft is advanced into the ultrasound beam, strong scatter originating from its surface will markedly impair image quality. Poor maternal transabdominal imaging of the intervention for this and less mundane reasons can nowadays be overcome by fetoscopically assisted fetal transesophageal echocardiography.42,43 At the start of the intervention, a small-calibered multimodal ultrasound catheter is inserted into the fetal esophagus via a single trocar, aided by fetoscopy (Figure 35.4). From its intraesophageal position immediately adjacent to the fetal heart, the catheter allows highfrequency images, unsurpassed by conventional imaging, to be obtained by the maternal transabdominal route, as sound attenuation from the maternal abdominal and uterine walls as well as by the fetal chest and spine becomes irrelevant. Due to the intrathoracic imaging position, transducer frequencies above 10 MHz can be employed. As the catheter images the heart from behind, any scatter from the interventional needle is projected anteriorly and cannot interfere with monitoring of the intervention.42 The high depth penetration of the phased-array catheter of more than 10 cm permits clear definition of even the anterior

518

Fetal Cardiology

(a)

(b)

(c)

(d)

(e)

(f)

Figure 35.4 Fetal transesophageal echocardiography (TEE) as a monitoring tool for fetal cardiac intervention in a fetus with aortic atresia and intact atrial septum. (a) Using fetoscopy, a catheter sheath is inserted into the fetal pharynx (b). (c) The ultrasound catheter is inserted into the catheter sheath and placed into the esophagus behind the heart. (d) From its intraesophageal position immediately adjacent to the fetal heart, the catheter allows high-frequency imaging of the left (LA) and right atria (RA) as well as the septum (arrow). (e) During the intervention, the jaws of a tiny grasper (arrow) are guided by fetal TEE. As the catheter images the heart from behind, any scatter from the device is projected anteriorly and cannot interfere with monitoring of the intervention. (f) Color Doppler imaging confirms that despite various attempts no lasting opening could be achieved within the atrial septum in this case. The procedure was then aborted because of the development of hemopericardium (arrow).

Mending the tiniest hearts

margins of the fetal heart and chest from the intraesophageal position. This ability may facilitate selection of the appropriate incision site in order to achieve direct fetal cardiac access (see below) for future attempts at fetal cardiac pacing or interventional catheterization. The multimodal Doppler capabilities of the phasedarray intravascular ultrasound catheter open the door for detailed hemodynamic assessment during fetoscopic and open fetal cardiac interventions. By simple adjustments of the intraesophageal catheter position, low incidence angles can be achieved for most flow sites, providing the basis for accurate estimation of transvalvar pressure gradients and quantitation of valve regurgitation.42 In fetuses too small for intraesophageal catheter insertion, the phased-array ultrasound catheter may also be used as an intra-amniotic imaging device.44 Over the past 5 years, the clinical experience with fetoscopic techniques previously developed in ovine models has matured to a stage whereby direct fetal cardiac access in a human fetus would now be possible.24 In fetal sheep and human fetal postmortem studies, transdiaphragmatic access following fetal laparotomy has proved to be a more reliable, safer, and less traumatic approach to accessing to the heart than a fetal sternotomy (Figure 35.5).44 The highly magnified fetoscopic images, in concert with fetal transesophageal echocardiography, permit accurate placement of interventional devices within hearts so small that they seem not amenable to the current maternal transabdominal percutaneous ultrasound-guided approach. The major limitation to clinical introduction of the fetoscopic procedure remains the lack of specialized intervention material. The ultimate goal of this approach is to permit fetal semilunar valve interventions from 16 weeks of gestation onward, as well as pacemaker insertion in fetuses beyond 20 weeks of gestation. In addition, the fetoscopic direct fetal cardiac access approach lends itself to the removal of large pericardial and lung tumors.

Maternofetal hyperoxygenation A wide spectrum of fetal cardiac malformations is accompanied by hypoplasia of right- or left-sided cardiovascular structures (hypoplastic left heart complex, coarctation, tetralogy of Fallot, pulmonary atresia with ventricular septal defect, etc.). Maternofetal hyperoxygenation in late gestation, by promoting lung blood flow, results in marked increases of both pulmonary and systemic venous return to the fetal heart. Exploiting this physiological principle for therapeutic purposes, I observed that these flow changes can result in marked catch-up growth of hypoplastic cardiovascular structures, avoiding the need for, or facilitating the performance of, postnatal surgical procedures or interventions.45 In addition, the approach may be life-saving in fetuses with some rare

519

forms of cardiac malformations when critically impaired lung flow is expected after delivery (Figure 35.6). Intensive basic research and clinical studies are now required in order to define the full potential of this promising new treatment approach. I expect that, due to its universal availability, efficacy, and simplicity, maternofetal hyperoxygenation will in some years from now be applied to an incomparably larger number of fetuses than any other prenatal cardiac intervention approach.

Twin-to-twin-transfusion syndrome Twin-to-twin-transfusion syndrome (TTTS) occurs in about 10–15% of monochorionic twin gestations. In sharing a single placenta, the presence of vascular anastomoses on its surface permits exchanges of blood between both twins that commonly result in severe fetal cardiac failure and death. The monitoring and treatment of TTTS has been a domain of prenatal medicine specialists and neonatologists. As a fetal hemodynamic disturbance that only rarely results in more severe cardiac problems in survivors after delivery, TTTS has been neglected by both pediatric cardiologists and cardiac surgeons. This lack of attention has deprived these specialists of fascinating and fruitful opportunities to learn about the fetal cardiovascular system functioning in extremis. Fetoscopic laser ablation of the pathological placental vessels has become the procedure of choice in advanced cases of monochorionic twin gestations complicated by TTTS (Figure 35.7).46,47 If untreated, most pregnancies are lost, or surviving fetuses suffer cerebral damage. In contrast, fetoscopic laser ablation results in the survival of at least one twin in about 80% of cases and has proved to be superior to serial drainage of amniotic fluid in a landmark European trial.47 Given the high mortality of the untreated disease course, this rate is already impressive. Yet, further research efforts seem worthwhile, in order to improve the survival rate for both twins, which currently stands at only about 40%. A recent technical advance in the fetoscopic treatment of this condition is the use of partial carbon dioxide insufflation of the amniotic cavity (PACI).48 This novel strategy results in markedly improved visualization in the presence of amniotic fluid stained by blood or detritus. Hence, laser ablations can be performed and fetuses salvaged under conditions previously thought not to be amenable to the conventional fetoscopic approach.

Fetal laryngeal atresia Fetal laryngeal atresia is a grave but fortunately rare condition that, in the second trimester, usually results in

520

Fetal Cardiology

(a)

(b)

Figure 35.5

(c)

In fetal sheep and human fetal postmortem studies, transdiaphragmatic access following fetal laparotomy has proved to be a more reliable, safer, and less traumatic approach to the heart than fetal sternotomy. (a) Postmortem operation in a human fetus. Following suspension of the fetal chest wall, a 2-cm laparotomy is performed below the xiphoid process using laser energy. (b) Following a 1-cm incision in the diaphragm, the base of the heart becomes visible. (c) Fetoscopy then permits high-power visualization of the fetal heart. In this case, the myocardial insertion of a pacing lead (arrow) followed by fetoscopic implantation of a pacemaker was simulated.

severe fetal cardiac failure and demise. The atretic larynx results in chronic congestion of lung fluid within the bronchial tree and trachea, leading to massive dilatation and hyperplasia of the lung (Figure 35.8).49 The increased intrathoracic pressure impairs cardiac filling, favoring the development of low output cardiac failure, hydrops, and death. This fatal sequence has been given the telling acronym ‘CHAOS’ (congenital high airway obstruction syndrome). In about half of cases, fetal laryngeal atresia is associated with Fraser syndrome (cryptophthalmos– syndactyly syndrome).

In order to decompress the lungs and improve fetal hemodynamics, the atretic region can be perforated and stented employing a percutaneous fetoscopic approach.50,51 As a causative treatment approach, the minimally invasive procedure results in immediate improvements of fetal cardiac filling and normalization of heart–lung size relationships. Alternatively, iatrogenic preterm premature delivery by ex utero intrapartum treatment (EXIT) has been chosen in some less severe cases.52–56 This complex delivery approach permits the establishment of a fetal tracheotomy, during which fetal gas exchange is maintained by the placenta.

Mending the tiniest hearts

521

(a)

(b)

(c)

(d)

Figure 35.6 Fetus with tetralogy of Fallot, severe subvalvar and valvar pulmonary stenosis, and an underdeveloped ductus arteriosus. Retrograde flow within the hypoplastic duct (diameter 1 mm) indicated postnatal ductal dependency. Because of the largely diminished pulmonary blood flow in this lesion (a), maternofetal hyperoxygenation (HO) was performed as a potentially life-saving treatment attempt. By promoting pulmonary vasodilatation, HO results in marked increases of both pulmonary and systemic venous returns to the fetal heart. (b) After 16 days of HO, markedly improved pulmonary blood flows were observed in this patient. In addition, HO resulted in an increase of right ventricular outflow dimensions (c, d). After delivery, the neonate still had critically low pulmonary blood flow and required continuous ventilation until surgery. Following repair, the baby survived to discharge from hospital. (RVOT, right ventricular outflow tract; PV, pulmonary valve; MPA, main pulmonary artery.

522

Fetal Cardiology

(a)

(b)

(c)

(d)

(e)

(f)

Figure 35.7 Twin-to-twin-transfusion syndrome (TTTS). In sharing a single placenta, the presence of vascular anastomoses on its surface permits exchanges of blood between twins that commonly result in severe fetal cardiac failure and death. (a) Polyhydramnios and discrepancy in fetal size are strong indicators for the presence of TTTS in a monochorionic twin gestation. (b) Suffering hypotension and hypovolemia, renal failure occurs in the donor fetus. As a consequence of anuria, the donor can no longer fill its amniotic sac and becomes totally entrapped within amniotic membranes (stuck twin). (c) Fetoscopy permits excellent visualization of the membranous division (arrows) of both amniotic sacs as well as crossing pathological placental anastomoses. (d) Employing laser energy, the pathological anastomoses are ablated. (e) Low velocity umbilical artery signal with absent end-diastolic flow indicates a more severe TTTS stage in this donor fetus. (f) After fetoscopic laser ablation, the umbilical artery flow velocity has increased and flow is again present at the end of diastole. However, the acutely improved loading puts strain on the donor heart and umbilical venous pulsations occur.

Mending the tiniest hearts

(a)

(b)

(c)

(d)

(e)

(f)

523

Figure 35.8 Fetal laryngeal atresia is a grave but fortunately rare condition that in the second trimester usually results in severe fetal cardiac failure and demise. (a) The atretic larynx results in chronic congestion of lung fluid within the bronchial tree and trachea leading to massive lung enlargement and cardiac compression. (b) Increased pulsatility in the ductus venosus indicates increased resistance to diastolic filling of the heart. (c) Following access to the amniotic cavity by a single trocar and partial carbon dioxide insufflation, the fetal mouth (arrow) is found. (d) Fetoscopic image of the atretic region. Below the vocal cords, a membrane is detected (arrow). Following perforation of the membrane with a 0.4-mm laser fiber (e), normalization of the heart–lung size relationship is observed (f).

524

Fetal Cardiology

Hydrothorax The accumulation of a large pleural effusion (i.e. hydrothorax) may lead to profound fetal cardiac failure and demise.57 Although the mediastinal shift from unilateral effusions is usually well tolerated, the intrathoracic pressure increase from large bilateral effusions interferes not only with normal lung development but also with adequate cardiac filling. As a consequence, pulmonary hypoplasia becomes the main cause of postnatal demise while cardiac failure is the most common prenatal cause of death in untreated fetuses. In the majority of fetuses, the effusion appears in isolation. However, it may be associated with a spectrum of other serious conditions (chromosomal anomalies, metabolic disease, cardiac tumors, lung tumors, diaphragmatic hernia, etc.). Percutaneous ultrasound-guided insertion of a thoracoamniotic shunt has been performed for nearly three decades, and permits uni- or bilateral drainage of the effusion(s) into the amniotic fluid.57 Particularly in fetuses with bilateral hydrothorax, the concomitant drop in intrathoracic pressure results in immediate improvements of cardiac filling (Figure 35.9). To the frustration of parents and caregivers alike, the post-interventional course more often than not is complicated by repeated shunt dislodgment or obstruction, requiring the need for a series of reinterventions. Hence, the success of shunting is difficult to predict in an individual fetus, and the ultimate prognosis relies on gestational age at presentation, sufficient prenatal cardiac and postnatal lung functions, reversal of hydrops, and the underlying pathology.

Vascular steal Vascular steal from large tumors or atrioventricular malformations may result in severe fetal cardiac failure and demise. In order to save the life of a small number of affected fetuses, open fetal surgery and tumor removal has been performed in cases with large vascularized sacrococcygeal teratomas.58 Unfortunately, the invasiveness of the open operative approach in concert with its anesthetic and tocolytic requirements may further impair fetal hemodynamics and contribute to fetal demise. In order to avoid the detrimental effects of the open operative approach, a minimally invasive experimental treatment has recently been devised that employs percutaneous ultrasoundguided radiofrequency ablation of the tumor-feeding vessels (Figure 35.10).59

Complications of minimally invasive approaches The known complications of any of the above-described minimally invasive approaches for fetal cardiac intervention

are preterm premature rupture of the membranes (15%), preterm delivery prior to 30 weeks of gestation (20%), and infection (5%). Significant maternal bleeding events from abdominal or uterine vessels or the placenta, as well as complications from maternal bowel injury, are rare. A particularly high mortality rate of up to 30% must be accepted by both the expectant mother and the interventional team when performing direct fetal cardiac punctures for balloon valvuloplasty. Needle insertion directly into the fetal heart commonly prompts bleeding complications, refractory bradycardias, or sudden cardiac pump failure. Clearly, the specific and general risks of minimally invasive procedures need to be conveyed to the mother during the consent process. Alternatively, the feasibility of fetal cardiac interventions by open operative approaches has been under investigation. Although the experimental development of these procedures lags far behind that of minimally invasive fetal cardiac interventions, they might hold promise for the future treatment of cardiac malformations or hemodynamic problems that are barely or not at all amenable to the latter approaches.14,60–64 In order to avoid falsely high expectations of the mother and caregivers, the interventional goals, benefits, and drawbacks of the available procedures must be defined for each case by an interdisciplinary care team. As most fetuses with severe semilunar valve obstructions do not develop cardiac failure and survive gestation when termination of pregnancy is not an option, fetal cardiac intervention primarily aims at improving postnatal quality of life and overall prognosis. These potential benefits are accompanied by high risk for an adverse outcome. From a different perspective, fetal cardiac interventions may, even in these lesions, acquire a life-saving character when disease progression would prompt a mother’s decision for termination. This situation is less complex in immature fetuses when cardiac failure is present; the majority of them will not survive gestation or benefit from desperate attempts at iatrogenic preterm delivery in order to undergo postnatal treatment. Interventions in these fetuses will clearly be undertaken as potentially life-saving procedures, avoiding some of the moral and ethical dilemmas surrounding the treatment of fetuses with severe semilunar valve obstructions. The often very difficult ethical, moral, and patient safety issues, as well as the attitudes of other specialists concerning the development and clinical introduction of novel experimental treatment approaches like those described in this chapter, commonly overwhelm small interventional teams or a single investigator.65–67 Therefore, the development and clinical introduction of these innovative therapies ought to be supervised and supported by committees of human research. After an adequate development and learning curve, the overall outcome and quality of life of the prenatally treated patients will ultimately determine whether fetal cardiac interventions will become beneficial

Mending the tiniest hearts

525

(a)

(b)

(c)

Figure 35.9 (a) Bilateral hydrothorax may result in profound fetal cardiac failure and pulmonary hypoplasia when the increased intrathoracic pressure severely impairs cardiac filling and compresses the lungs. Also note the marked skin edema in this fetus. In concert with ascites, these findings indicated the presence of an advanced disease stage. (b) Image taken following successful insertion of a thoracoamniotic shunt via a needle shaft employing a percutaneous ultrasound-guided approach. (c) Insertion of the shunt into the right pleural space results in immediate lung expansion and improved cardiac filling.

526

Fetal Cardiology

(a)

(b)

(c)

(d)

(e)

(f)

(g)

Figure 35.10 (a) In this fetus at 25 weeks of gestation, vascular steal from a huge sacrococcygeal teratoma has led to profound cardiac failure with massive cardiac enlargement (b). (c) Note the large vein on the surface of the tumor (arrow). The vein diameter was larger than the umbilical cord diameter. (d) Color Doppler imaging clearly demonstrates two large feeding arteries at the tumor base. (e) External view of the set-up for percutaneous ultrasound-guided radiofrequency ablation of the tumor-feeding vessels. (f) Hyperechogenicity at the site of the needle tip indicates successful tissue ablation. (g) After the procedure, color Doppler imaging at low Nyquist limits demonstrates successful closure of the feeding vessels. This was followed by immediate improvements in fetoplacental flows.

Mending the tiniest hearts

therapeutic alternatives to currently available postnatal procedures.

References 1. Allan LD, Crawford DC, Tynan MJ. Pulmonary atresia in prenatal life. J Am Coll Cardiol 1986; 8: 1131–6. 2. Allan LD, Sharland G, Tynan MJ. The natural history of the hypoplastic left heart syndrome. Int J Cardiol 1989; 25: 341–3. 3. Danford DA, Cronican P. Hypoplastic left heart syndrome: progression of left ventricular dilation and dysfunction to left ventricular hypoplasia in utero. Am Heart J 1992; 123: 1712–13. 4. Hornberger LK, Sanders SP, Rein AJ et al. Left heart obstructive lesions and left ventricular growth in the midtrimester fetus. A longitudinal study. Circulation 1995; 92: 1531–8. 5. Hornberger LK, Sanders SP, Sahn DJ et al. In utero pulmonary artery and aortic growth and potential for progression of pulmonary outflow tract obstruction in tetralogy of Fallot. J Am Coll Cardiol 1995; 25: 739–45. 6. Hornberger LK, Need L, Benacerraf BR. Development of significant left and right ventricular hypoplasia in the second and third trimester fetus. J Ultrasound Med 1996; 15: 655–9. 7. Hornberger LK, Benacerraf BR, Bromley BS, Spevak PJ, Sanders SP. Prenatal detection of severe right ventricular outflow tract obstruction: pulmonary stenosis and pulmonary atresia. J Ultrasound Med 1994; 13: 743–50. 8. Rice MJ, McDonald RW, Reller MD. Progressive pulmonary stenosis in the fetus: two case reports. Am J Perinatol 1993; 10: 424–7. 9. Todros T, Presbitero P, Gaglioti P, Demarie P. Pulmonary stenosis with intact ventricular septum: documentation of development of the lesion echocardiographically during fetal life. Int J Cardiol 1988; 8: 1131–6. 10. Maxwell D, Allan L, Tynan MJ. Balloon dilatation of the aortic valve in the fetus: a report of two cases. Br Heart J 1991; 65: 256–8. 11. Chaoui R, Bollmann R, Göldner B, Rogalsky V. Aortic balloon valvuloplasty in the human fetus under ultrasound guidance: a report of two cases. Ultrasound Obstet Gynecol 1994; 4: 162A. 12. Lopes LM, Cha SC, Kajita LJ et al. Balloon dilatation of the aortic valve in the fetus. A case report. Fetal Diagn Ther 1996; 11: 296–300. 13. Huhta J. Fetal echocardiography in the detection and management of fetal heart disease. In: Spitzer AR, ed. Intensive Care of the Fetus and Neonate. St Louis: Mosby, 1996: 772–86. 14. Hanley FL. Fetal cardiac surgery. Adv Card Surg 1994; 5: 47–74. 15. Wright JGC, Skinner JR, Stumper O. Radiofrequency assisted pulmonary valvotomy in a fetus with pulmonary atresia and intact ventricular septum. Presented at 13th Meeting of the International Fetal Medicine and Surgery Society, Antwerp, Belgium, 1994.

527

16. Tulzer G, Arzt W, Franklin RC et al. Fetal pulmonary valvuloplasty for critical pulmonary stenosis or atresia with intact septum. Lancet 2002; 360: 1567–8. 17. Walkinshaw SA, Welch CR, McCormack J, Walsh K. In utero pacing for fetal congenital heart block. Fetal Diagn Ther 1994; 9: 183–5. 18. Carpenter R Jr, Strasburger JF, Garson A Jr et al. Fetal ventricular pacing for hydrops secondary to complete atrioventricular block. J Am Coll Cardiol 1986; 8: 1434–6. 19. Dunn JM, Weil SR, Russo P. Management of fetal cardiac anomalies. In: Hanson MA, Spencer JAD, Rodeck CH, eds. Fetus and Neonate. Physiology and Clinical Applications – The Circulation. Cambridge: Cambridge University Press, 1993: 377–95. 20. Bical O, Gallix P, Donzeau-Gouge P et al. [Antenatal cardiac surgery. Creation of an experimental model of pulmonary stenosis in the fetus and repair in utero]. Arch Mal Coeur Vaiss 1985; 78: 445–9. [in French] 21. Kohl T, Witteler R, Strümper D et al. Operative techniques and strategies for minimally invasive fetoscopic fetal cardiac interventions in sheep. Surg Endosc 2000; 14: 424–30. 22. Kohl T, Westphal M, Strümper D et al. Multimodal fetal transesophageal echocardiography for fetal cardiac intervention in sheep. Circulation 2001; 114: 1757–60. 23. Kohl T, Szabo Z, Suda K et al. Fetoscopic and open transumbilical fetal cardiac catheterization in sheep. Potential approaches for human fetal cardiac intervention. Circulation 1997; 95: 1048–53. 24. Kohl T, Strumper D, Witteler R et al. Fetoscopic direct fetal cardiac access in sheep: an important experimental milestone along the route to human fetal cardiac intervention. Circulation 2000; 102: 1602–4. 25. Assad RS, Zielinsky P, Kalil R et al. New lead for in utero pacing for congenital heart block. J Thorac Cardiovasc Surg 2003; 126: 300–2. 26. Marshall AC, van der Velde ME, Tworetzky W et al. Creation of an atrial septal defect in utero for fetuses with hypoplastic left heart syndrome and intact or highly restrictive atrial septum. Circulation 2004; 110: 253–8. 27. Vida VL, Bacha EA, Larrazabal A et al. Hypoplastic left heart syndrome with intact or highly restrictive atrial septum: surgical experience from a single center. Ann Thorac Surg 2007; 84: 581–5. 28. Quintero RA, Huhta J, Suh E et al. In utero cardiac fetal surgery: laser atrial septotomy in the treatment of hypoplastic left heart syndrome with intact atrial septum. Am J Obstet Gynecol 2005; 193: 1424–8. 29. Rudolph AM. The fetal circulation. In: Congenital Diseases of the Heart. Chicago: Year Book Medical Publishers, 1974: 1–16. 30. Sharland GK, Chita SK, Fagg NL et al. Left ventricular dysfunction in the fetus: relation to aortic valve anomalies and endocardial fibroelastosis. Br Heart J 1991; 66: 419–24. 31. Fishman NH, Hof RB, Rudolph AM, Heymann MA. Models of congenital heart disease in fetal lambs. Circulation 1978; 58: 354–64. 32. Bical O, Gallix P, Toussaint M et al. Intrauterine versus postnatal repair of created pulmonary artery stenosis in the lamb. Morphologic comparison. J Thorac Cardiovasc Surg 1990; 99: 685–90.

528

Fetal Cardiology

33. Hornberger LK, Sanders SP, Rein AJ et al. Left heart obstructive lesions and left ventricular growth in the midtrimester fetus. A longitudinal study. Circulation 1995; 92: 1531–8. 34. Mäkikallio K, McElhinney DB, Levine JC et al. Fetal aortic valve stenosis and the evolution of hypoplastic left heart syndrome: patient selection for fetal intervention. Circulation 2006; 113: 1401–5. 35. Kohl T, Sharland G, Allan LD et al. World experience of percutaneous ultrasound-guided balloon valvuloplasty in human fetuses with severe aortic valve obstruction. Am J Cardiol 2000; 85: 1230–3. 36. Tworetzky W, Wilkins-Haug L, Jennings RW et al. Balloon dilation of severe aortic stenosis in the fetus: potential for prevention of hypoplastic left heart syndrome: candidate selection, technique, and results of successful intervention. Circulation 2004; 110: 2125–31. 37. Mizrahi-Arnaud A, Tworetzky W, Bulich LA et al. Pathophysiology, management, and outcomes of fetal hemodynamic instability during prenatal cardiac intervention. Pediatr Res 2007; 62: 325–30. 38. Wilkins-Haug LE, Tworetzky W, Benson CB et al. Factors affecting technical success of fetal aortic valve dilation. Ultrasound Obstet Gynecol 2006; 28: 47–52. 39. Roman KS, Fouron JC, Nii M et al. Determinants of outcome in fetal pulmonary valve stenosis or atresia with intact ventricular septum. Am J Cardiol 2007; 99: 699–703. 40. Salvin JW, McElhinney DB, Colan SD et al. Fetal tricuspid valve size and growth as predictors of outcome in pulmonary atresia with intact ventricular septum. Pediatrics 2006; 118: e415–20. 41. Kohl T, Hering R, Van de Vondel P et al. Analysis of the step-wise clinical introduction of experimental percutaneous fetoscopic surgical techniques for upcoming minimally-invasive fetal cardiac interventions. Surg Endosc 2006; 20: 1134–43. 42. Kohl T, Müller A, Tchatcheva K, Achenbach S, Gembruch U. Fetal transesophageal echocardiography – clinical introduction as a monitoring tool during fetal cardiac intervention in a human fetus. Ultrasound Obstet Gynecol 2005; 26: 780–5. 43. Kohl T, Breuer J, Heep A et al. Fetal transesophageal echocardiography during balloon valvuloplasty for severe aortic valve stenosis at 28+6 weeks of gestation. J Thorac Cardiovasc Surg 2007; 134: 256–7. 44. Kohl T, Tchatcheva K, Van de Vondel P, Gembruch U. Intraamniotic fetal echocardiography – a new fetal cardiovascular monitoring approach during human fetoscopic surgery. Circulation 2006; 114: e594–6. 45. Kohl T, Tchatcheva K, Stressig R, Geipel A, Heitzer S, Gembruch U. Maternal hyperoxygenation in late gestation promotes rapid increase of cardiac dimensions in fetuses with hypoplastic left hearts with intrinsically normal or slightly abnormal aortic and mitral valves. Ultraschall in Med 2008; 29: 92. 46. De Lia JE, Cruikshank DP, Keye WR Jr. Fetoscopic neodymium: YAG laser occlusion of placental vessels in severe twin-twin transfusion syndrome. Obstet Gynecol 1990; 75: 1046–53. 47. Senat MV, Deprest J, Boulvain M et al. Endoscopic laser surgery versus serial amnioreduction for severe

48.

49.

50.

51.

52.

53.

54.

55.

56.

57.

58.

59.

60.

61.

twin-to-twin transfusion syndrome. N Engl J Med 2004; 351: 136–44. Kohl T, Tchatcheva K, Berg C et al. Partial amniotic carbon dioxide insufflation (PACI) facilitates fetoscopic interventions in complicated monochorionic twin pregnancies. Surg Endosc 2007; 21: 1428–33. Crombleholme TM, Albanese CT. The fetus with airway obstruction. In: Harrison MR, Evans MI, Adzick NS et al, eds. The Unborn Patient – The Art and Science of Fetal Therapy, 3rd edn. Philadelphia: Saunders, 2001: 357–71. Kohl T, Hering R, Bauriedel G et al. Percutaneous fetoscopic and ultrasound-guided decompression of the fetal trachea permits normalization of fetal hemodynamics in a human fetus with Fraser syndrome and congenital high airway obstruction syndrome (CHAOS) from laryngeal atresia. Ultrasound Obstet Gynecol 2005; 27: 84–8. Kohl T, Van de Vondel P, Stressig R et al. Percutaneous fetoscopic laser decompression of congenital high airway obstruction syndrome (CHAOS) from laryngeal atresia via a single trocar – current technical constraints and potential solutions for future interventions. Fetal Diagn Ther 2008; in press. Lim FY, Crombleholme TM Hedrick HL et al. Congenital high airway obstruction syndrome: natural history and management. J Pediatr Surg 2003; 38: 940–5. DeCou JM, Jones DC, Jacobs HD, Touloukian RJ. Successful ex utero intrapartum treatment (EXIT) procedure for congenital high airway obstruction syndrome (CHAOS) owing to laryngeal atresia. J Pediatr Surg 1998; 33: 1563–5. Bui TH, Grunewald C, Frenckner B et al. Successful EXIT (ex utero intrapartum treatment) procedure in a fetus diagnosed prenatally with congenital high-airway obstruction syndrome due to laryngeal atresia. Eur J Pediatr Surg 2000; 10: 328–33. Oepkes D, Teunissen AKK, Van de Velde M et al. Congenital high airway obstruction syndrome successfully managed with ex-utero intrapartum treatment. Ultrasound Obstet Gynecol 2003; 22: 437–9. Kanamori Y, Kitano Y, Hashizume K et al. A case of laryngeal atresia (congenital high airway obstruction syndrome) with chromosome 5p deletion syndrome rescued by ex utero intrapartum treatment. J Pediatr Surg 2004; 39: E25–8. Farmer DL, Albanese CT. Fetal hydrothorax. In: Harrison MR, Evans MI, Adzick NS et al, eds. The Unborn Patient – The Art and Science of Fetal Therapy, 3rd edn. Philadelphia: Saunders, 2001: 373–7. Hedrick HL, Flake AW, Crombleholme TM et al. Sacrococcygeal teratoma: prenatal assessment, fetal intervention, and outcome. J Pediatr Surg 2004; 39: 430–8. Paek BW, Jennings RW, Harrison MR et al. Radiofrequency ablation of human fetal sacrococcygeal teratoma. Am J Obstet Gynecol 2001; 184: 503–7. Fenton KN, Zinn HE, Heinemann MK, Liddicoat JR, Hanley FL. Long-term survivors of fetal cardiac bypass in lambs. J Thorac Cardiovasc Surg 1994; 107: 1423–7. Reddy VM, Liddicoat JR, Klein JR, Wampler RK, Hanley FL. Long-term outcome after fetal cardiac bypass: fetal survival to full term and organ abnormalities. J Thorac Cardiovasc Surg 1996; 111: 536–44.

Mending the tiniest hearts

62. Scagliotti D, Shimokochi DD, Pringle KC. Permanent cardiac pacemaker implant in the fetal lamb. Pacing Clin Electrophysiol 1987; 10: 1253–61. 63. Assad RS, Jatene MB, Moreira LF et al. Fetal heart block: a new experimental model to assess fetal pacing. Pacing Clin Electrophysiol 1994; 17: 1256–63. 64. Murotsuki J, Okamura K, Watanabe T, Kimura Y, Yajima A. Production of complete heart block and utero cardiac pacing in fetal lambs. J Obstet Gynaecol 1995; 21: 233–9.

529

65. Chervenak FA, McCullough LB, Kurjak A. An essential clinical ethical concept. In: Chervenak FA, Kurjak A, ed. Current Perspectives on the Fetus as a Patient. New York: Parthenon Publishing, 1996: 1–9. 66. Lyerly AD, Gates EA, Cefalo RC, Sugarman J. Toward the ethical evaluation and use of maternal-fetal surgery. Obstet Gynecol 2001; 98: 689–97. 67. Lyerly AD, Cefalo RC, Socol M, Fogarty L, Sugarman J. Attitudes of maternal-fetal specialists concerning maternalfetal surgery. Am J Obstet Gynecol 2001; 185: 1052–8.

36 Fetal cardiac function in normal and growth-restricted fetuses Giuseppe Rizzo, Alessandra Capponi, and Domenico Arduini

In adults, blood circulates sequentially through the systemic and pulmonary vasculature and there is essentially no mixture of blood oxygenated in the lung and that in the systemic circulation. Fetal cardiac hemodynamics differs from that seen postnatally. During fetal life, blood is oxygenated in the placenta and returns to the fetal body via the umbilical vein. Studies in chronically instrumented fetal lambs have shown that in physiological conditions, about 55% of umbilical vein blood bypasses the hepatic circulation, entering the inferior vena cava directly via the ductus venosus.1 From the inferior vena cava this highly oxygenated blood preferentially streams through the foramen ovale into the left atrium, left ventricle, and descending aorta.2 On the other hand, poorly oxygenated blood from the hepatic and superior vena cava circulations enters the right atrium and is almost completely directed through the tricuspid valve into the right ventricle and pulmonary artery.2 Because fetal blood is not oxygenated by the lungs, an additional shunt (i.e. the ductus arteriosus) operates to bypass the pulmonary circulation, preferentially directing the right ventricle output to the descending aorta. As a consequence, both ventricles eject into the systemic circulation in parallel. The output of the left ventricle is directed through the ascending aorta to the upper body organs, thus making the most highly oxygenated blood available to the heart and brain. The right ventricle ejects through the patent ductus arteriosus and the descending aorta to the lower body and placenta. These features of the fetal circulation raise interesting and important questions concerning mechanisms to provide oxygen and substrates to the different organs during the evolution of normal and pathological pregnancies. However, in the past the intrauterine environment has limited the possibility of studying these mechanisms in the human fetus, and much of the understanding and present knowledge of fetal hemodynamics are derived from animal studies. Nevertheless, during the last few years, technological advances in ultrasound have made it possible to

study the human fetal heart. In particular, the advent of pulsed and color Doppler techniques has allowed noninvasive examination of fetal cardiovascular pathophysiology, thus enabling hemodynamics studies in fetuses under both normal and abnormal conditions. Because no therapy at present has been shown to significantly improve placental function, the objective of Doppler studies has been to optimize the timing of delivery, early enough to avoid fetal death or neonatal sequelae and late enough to avoid the effects of iatrogenic severe prematurity. This chapter outlines the principles of fetal Doppler echocardiography and its practical uses, and discusses its current and possible future applications.

Technique General principles The parameters used to describe fetal cardiac velocity waveforms differ from those used in fetal peripheral vessels. In this latter situation, indexes such as pulsatility index, resistance index, or systolic/diastolic (S/D) ratio are used. These values are derived from relative ratios between systolic, diastolic, and mean velocity, and are therefore independent from the absolute velocity values and from the angle of insonation between the Doppler beam and direction of blood flow.3 At the cardiac level, all measurements represent absolute values. Measurements of absolute flow velocities require knowledge of the angle of insonation, which may be difficult to obtain with accuracy. The error in the estimation of absolute velocity resulting from the uncertainty of angle measurement is strongly dependent on the magnitude of the angle itself. For angles less than about 20°, the error will be reduced to practical insignificance. For larger angles, the cosine term in the Doppler equation

532

Fetal Cardiology

changes the small uncertainty in measurement of the angle to a large error in the velocity equations.3 As a consequence, recordings should be obtained always keeping the Doppler beam as parallel as possible to the bloodstream, and all recordings with an estimated angle greater than 20° should be rejected. Color Doppler may solve many of these problems by showing the flow direction in real time, allowing proper alignment of the Doppler beam in the direction of blood flow. To record velocity waveforms, pulsed Doppler is generally preferred to continuous wave Doppler because of its range of resolution. During recordings, the sample volume is placed immediately distal to the location to be investigated (e.g. distal to the aortic semilunar valves to record the left ventricle outflow). However, in conditions of particularly high velocity (e.g. in the ductus arteriosus), continuous Doppler may be useful because it avoids the aliasing effect.

Parameters measured The parameters most commonly used to describe cardiac velocity waveforms are the following:4 • the peak velocity (PV) expressed as the maximum velocity at a given moment (e.g. systole, diastole) on the Doppler spectrum • the time to peak velocity (TPV) or acceleration time expressed by the time interval between the onset of the waveform and its peak • the time velocity integral (TVI) calculated by planimetering the area underneath the Doppler spectrum. It is also possible to calculate absolute cardiac flow from both the atrioventricular valve and outflow tracts by multiplying the TVI by the valve area by the fetal heart rate (HR). These measurements are particularly prone to errors, mainly due to inaccuracies in the valve area. The area is derived from the valve diameter, which is near the limit of ultrasound resolution, and is then halved and squared in the calculation, thus amplifying potential errors. However, such measurements can be used in longitudinal studies over short time intervals, in which the valve dimensions are assumed to remain constant. Furthermore, it is also possible to accurately calculate the relative ratio between the right and left cardiac output (RCO/LCO) avoiding measurement of the cardiac valve, as the relative dimensions of the aortic and pulmonary valves remain constant throughout gestation in the absence of cardiac structural diseases.5 However, these recordings are time-consuming, and require a favorable position of the fetus and an experienced operator. These factors have limited its application in clinical practice. Recent advances in ultrasonography such as fourdimensional (4D) ultrasound with spatiotemporal image

correlation (STIC) allow acquisition of fetal cardiac volume and reconstruction of a cine-loop cardiac cycle from which a specific cardiac phase can be identified and analyzed.6–9 This approach has been used to acquire fetal cardiac volumes from a standard four-chamber view and offline visualization of the outflow tract,10 thus obtaining in a reproducible way measurements from the aortic and pulmonary valves11 (Figure 36.1). A further application of 4D STIC in the human fetus is to calculate stroke volume by subtracting the volumes of ventricles measured during systole and diastole using offline analysis, as shown in Figure 36.1.12 Evaluation of the ventricular ejection force (VEF) has been also used to assess fetal cardiac function.13,14 VEF is a Doppler index based on Newton’s law, which estimates the energy transferred from right and left ventricular myocardial shortening to work done by accelerating blood into the pulmonary and systemic circulations, respectively.15 This index appears to be less influenced by changes in preload and afterload than other Doppler indexes,16 and results seem to be more accurate than other Doppler variables such as peak velocities for the assessment of ventricular function in adults with chronic congestive heart failure. VEF is calculated according to Newton’s second law of motion. Indeed, the force developed by ventricular contraction accelerates a column of blood into the aorta or pulmonary artery and represents transfer of energy of myocardial shortening to work done on the pulmonary and systemic circulation. Newton’s second law estimates the force as the product of mass and acceleration. The mass component in this model is the mass of blood accelerated into the outflow tract over a time interval, and may be calculated as the product of the density of blood (1.055), the valve area, and the flow velocity time integral during acceleration (FVIAT), which is the area under the Doppler spectrum envelope up to the time of peak velocity. The acceleration component of the equation is estimated as the PV divided by the TPV. Thus, VEF can be calculated using the following equation:13 VEF = (1.055 × valve area × FVIAT) × (PV/TPV)

Doppler depiction of fetal cardiac circulation In the human fetus, blood flow velocity waveforms might be recorded at all cardiac levels, including venous return, foramen ovales, atrioventricular valves, outflow tracts, pulmonary arteries, and ductus arteriosus. Various factors affect the morphology of the velocity waveforms from different districts. Among these are preload,17,18 afterload,19,20 myocardial contractility,21 ventricular compliance,22 and fetal heart rate.23 The impossibility of obtaining simultaneous recordings of pressure and volume does not allow full differentiation between these factors in the human

Cardiac function in normal and growth-restricted fetuses

533

(a)

(b)

Figure 36.1 Example of cardiac volume acquired with four-dimensional spatiotemporal image correlation (4D STIC). (a) The volume is displayed in multiplanar mode showing the acquisition plane with the four-chamber view. After manipulation of the volume10 both the aorta and pulmonary artery are shown (b).

fetus. However, as each parameter and site of recording is more specifically affected by one of these factors, it is possible indirectly to elucidate the underlying pathophysiology by performing measurements at various cardiac levels.

Venous circulation Blood flow velocity waveforms may be recorded from superior and inferior venae cavae, the ductus venosus, hepatic veins, pulmonary veins, and the umbilical vein. The vascular districts more intensively studied are the inferior vena cava (IVC) and the ductus venosus (DV). The IVC velocity waveforms, recorded from the segment of the vessel just distal to the entrance of the ductus venosus,24 are characterized by a triphasic profile with a first forward wave concomitant with ventricular systole, a second forward wave of smaller dimensions occurring with early diastole, and a third wave with reverse flow during

atrial contraction25,26 (Figure 36.2). Several indexes have been suggested to analyze IVC waveforms, but we demonstrated that the preload index (PLI) is more efficient than others described in the literature in predicting fetal compromise.27 This index, expressed by the ratio between the peak velocity during atrial contraction and the peak velocity during systole (PLI = A/S),28 is considered to be related to the pressure gradient present between the right atrium and the right ventricle during end-diastole, which is a function of both ventricular compliance and ventricular end-diastolic pressure.29 The ductus venosus (DV) may be evidenced in a transverse section of the upper fetal abdomen at the level of its origin from the umbilical vein. Color is then superimposed, and the pulsed Doppler sample volume placed just above its inlet (close to the umbilical vein) at the point of maximum flow velocity as expressed by color brightness. DV flow velocity waveforms exhibit a biphasic pattern with a first peak concomitant with systole (S), a second

534

Fetal Cardiology

peak concomitant with diastole (D), and a nadir during atrial contraction (A) (Figure 36.3). Among the indexes suggested to quantify velocity waveforms from the DV, the ratio between S peak velocity and A peak velocity (S/A) has proved to be an angle-independent parameter that efficiently describes DV hemodynamics.28,30 The morphology of the velocity waveforms from hepatic veins is similar to that of the IVC. There is a scarcity of reports on the use of these vessels in the human fetus, and from the data available it may be argued that the analysis of these vessels may have the same significance as that of the IVC. Pulmonary vein velocity waveforms may be recorded at the level of their entrance to the right atrium. Their

Figure 36.2 Velocity waveforms from the inferior vena cava in a normal fetus at 32 weeks of gestation depicting the systolic (S) and diastolic (D) waves (bottom) and reverse flow during atrial contraction (top).

morphology is characterized by positive velocities also during atrial contraction31 (Figure 36.4). The striking variation in velocity waveform morphology between the IVC and the pulmonary vein is of interest and may reflect the different hemodynamic conditions occurring in the systemic and pulmonary venous circulation during intrauterine life.29 Umbilical venous blood flow is usually continuous (Figure 36.5). However, in the presence of a relevant amount of reverse flow during atrial contraction in the inferior vena cava, pulsations with heart rate in umbilical venous flow occur. In normal pregnancies, these pulsations occur only before the 12th week of gestation, secondary to

Figure 36.3 Blood flow velocity waveforms from the ductus venosus in a normal fetus at 34 weeks of gestation.

Figure 36.4 Velocity waveform from a pulmonary vein in a normal fetus at 30 weeks of gestation. Note the presence of forward flow during atrial contraction.

Cardiac function in normal and growth-restricted fetuses

the stiffness of the ventricles present at this gestational age, causing a high percentage of reverse flow in the inferior vena cava.32 Later in gestation the presence of pulsations in the umbilical vein denotes severe cardiac compromise. The improvement in ultrasound resolution allows reliable measurements of the intra-abdominal umbilical vein diameter, thus making it possible to quantify absolute umbilical flow33 (Figure 36.6). Its value, and in particular the ratio with continuous cardiac output (CCO) or DV flow,

535

seem to be promising early markers of fetal compromise in growth restriction.34,35

Atrioventricular valves Flow velocity waveforms at the level of the mitral and tricuspid valves are recorded from the apical four-chamber view of the fetal heart, and are characterized by two diastolic peaks corresponding to early ventricular filling (E-wave) and active ventricular filling during atrial contraction (A-wave) (Figure 36.7). The ratio between E- and A-waves (E/A) is a widely accepted index of ventricular diastolic function, and it is an expression of both the cardiac compliance and preload conditions.4,15,36

Outflow tracts Flow velocity waveforms from the aorta and pulmonary artery are recorded, respectively, from the five-chamber and short-axis views of the fetal heart (Figure 36.8). PV and TPV are the most commonly used indexes. The former is influenced by several factors including valve size, myocardial contractility, and afterload,4,17,18 while the latter is believed to be secondary to the mean arterial pressure.32

Coronary blood flow

Figure 36.5 Blood flow velocity waveforms from the umbilical artery (top) and vein (bottom). Note the continous flow pattern in the umbilical vein.

(a)

Figure 36.6

Coronary blood flow may be visualized with the use of high resolution ultrasound equipment and color Doppler echocardiography. In normal fetuses, both right and left coronary arteries may be identified after 31 weeks of gestation under optimal conditions of fetal imaging.33 In compromised fetuses these vessels may be identified at an earlier gestational age, probably due to an increased coronary blood flow.33

(b)

Visualization of the intra-abdominal portion of the umbilical vein insonated at 90° (a) and Doppler recording of umbilical flow at 0° in a 20-week fetus (b).

536

Fetal Cardiology

Pulmonary vessels Velocity waveforms may be recorded from the right and left pulmonary arteries or from peripheral vessels within the lung.34,35,37,38 The waveform morphology differs

according to the site of sampling, and there is a progressive increase in the diastolic component in the more distal vessels37,38 (Figure 36.9). Their analysis may be used to study of the normal development of lung circulation.

Reproducibility

Velocity waveform from the mitral valve at 26 weeks. The ratio between E- and A-waves (E/A ratio) is 0.78.

A major concern in obtaining absolute measurements of velocities or flow is their reproducibility. In order to obtain reliable recordings it is particularly important to minimize the angle of insonation as mentioned above, to verify in real-time and color flow imaging the correct position of the sample volume before and after each Doppler recording, and to limit the recordings to periods of fetal rest and apnea, as behavioral states greatly influence the recordings.39–46 In these conditions it is necessary to select a series (at least five) of consecutive velocity waveforms characterized by uniform morphology and high signal/noise ratio before performing the measurements. Using this technique of

(a)

(b)

(c)

(d)

Figure 36.7

Figure 36.8

Visualization of aortic (a) and pulmonary (c) valves insonated at 90° and Doppler recordings from both outflow tracts (b, d) in the same fetus at 20 weeks.

Cardiac function in normal and growth-restricted fetuses

537

Figure 36.9 Velocity waveform from the fetal lung circulation. Note the different morphology of blood flow velocity waveforms from the outflow tract to the peripheral vessels.

recording and analysis, we managed to obtain a coefficient of variation below 10% for all the echocardiographic indexes with the exception of those needing valve dimensions. These results are in agreement with those reported by other centers (Groenenberg et al, coefficient of variation < 7%,46 Al-Ghazali et al, coefficient of variation < 7.6%,47 Reed et al, maximal variation < 10%.48 In addition, the use of angle-independent indexes from the venous circulation, or the use of offline analysis with 4D STIC, may further improve the reproducibility of Doppler echocardiographic recordings.11,12

Normal ranges of Doppler echocardiographic indexes The advent of the transvaginal color Doppler technique has allowed cardiac flow velocity waveforms to be recorded from 8 weeks of gestation onwards.49,50 Particularly evident changes occur at all cardiac levels from this gestational age up to 20 weeks. Namely, the PLI in the IVC and the S/A in the DV decrease significantly50,51 (Figures 36.10 and 36.11), the E/A ratios at both atrioventricular levels increase49,50 (Figure 36.10), and the PV and TVI values in

538

Fetal Cardiology

Inferior vena cava (PLI)

1.2 1 0.8 0.6

11 wks

0.4 0.2 0 16

20

24

28

32

36

40

età gestazionale (sett)

(a)

40 wks

(b)

Figure 36.10 Change of blood flow velocity waveforms from the inferior vena cava during gestation and reference limits for gestation of the preload index (PLI) (a). Color visualization. (b) Blood flow velocity waveforms.

the outflow tracts increase.49 These changes suggest a rapid development of ventricular compliance (which may explain the decrease of IVC PLI and increase of E/A and a shift of cardiac output toward the right ventricle probably secondary to the decreased right ventricle afterload due to the fall of placental resistance). After 20 weeks of gestation there is a further but less evident decrease of IVC PLI22 associated with a significant decrease of the S/A ratio from the DV25 (Figures 36.10 and 36.11). At the levels of the atrioventricular valves the E/A ratios increase,52,53 while PV values increase linearly at the levels of both the pulmonary and aortic valves.54 Small changes are present in TPV values during gestation.55 TPV values at the level of the pulmonary valve are lower than at the aortic level, suggesting slightly higher blood pressure in the pulmonary artery than in the ascending aorta.56 Quantitative measurements have shown that the right cardiac output (RCO) is higher than the left cardiac output (LCO), and that from 20 weeks onward the RCO/LCO ratio remains constant, with a mean value of 1.3.57,58 This value is lower than that reported in the fetal sheep (RCO/ LCO = 1.8), and this difference may be explained by the higher brain weight in humans which increases the left cardiac output.59

In normal fetuses, the VEF increases exponentially with advancing gestation at the levels of both right and left ventricles.13 No significant differences are present between right and left VEF values, and their ratio remains stable during gestation (mean value = 1.09).14

Hemodynamic modifications in growth-restricted fetuses Fetuses with intrauterine growth restriction (IUGR) secondary to uteroplacental insufficiency are characterized by selective changes of peripheral vascular resistances (i.e. the so-called brain-sparing effect) (Figure 36.12–36.15). Indeed there is an increase in the blood supply to the brain, myocardium, and adrenal glands and a reduction in perfusion of the kidneys, gastrointestinal tract, and lower extremities (Table 36.1). Although knowledge of the factors governing circulatory readjustments and their mechanisms of action is incomplete, it appears that partial pressures of oxygen and carbon dioxide play a role, presumably through their action on chemoreceptors.60 This mechanism allows

Cardiac function in normal and growth-restricted fetuses

539

Ductus venosus (S/A)

4.50 4.00 3.50 3.00

11 wks

2.50 2.00 1.50 1.00 0.50 16

20

24

28

32

36

40

età gestazionale (sett)

(a)

34 wks

(b)

Figure 36.11 Change of blood flow velocity waveforms from the ductus venosus during gestation and reference limits for gestation of the ratio between systolic peak velocity and atrial contraction peak velocity (S/A ratio). (a) Color visualization. (b) Blood flow velocity waveforms.

preferential delivery of nutrients and oxygen to vital organs, thereby compensating for diminished placental resources. However, compensation through cerebral vasodilatation is limited, and a plateau corresponding to a nadir of pulsatility index in cerebral vessels is reached at least 2 weeks before the development of fetal jeopardy. Similar findings may be found in the other arterial vessels. As a consequence, arterial vessels are unsuitable for longitudinal monitoring of growthrestricted fetuses. Indeed, we examined IUGR fetuses longitudinally and described a curvilinear relationship between impedance in cerebral vessels and the state of fetal oxygenation; the progressive fall in impedance reached a nadir 2 weeks before the onset of late fetal heart rate decelerations.61 This suggests that the maximum degree of vascular adaptation to hypoxemia precedes the critical degree of impairment of fetal oxygenation. Secondary to the brain-sparing condition, selective modifications occur in cardiac afterload, with a decreased left ventricle afterload due to cerebral vasodilatation and an increased right ventricle afterload due to systemic and

pulmonary vasoconstriction.4,40,62 Furthermore, hypoxemia might impair myocardial contractility while the polycythemia usually present might alter blood viscosity and therefore preload.4 As a consequence, IUGR fetuses show impaired ventricular filling properties, with a lower E/A ratio at the level of the atrioventricular valves,53 lower PV in the aorta and pulmonary arteries54 (Figures 36.12 and 36.13), increased aortic and decreased pulmonary TPV,55 and a relative increase of LCO associated with decreased RCO.47 These hemodynamic intracardiac changes are compatible with a preferential shift of cardiac output in favor of the left ventricle, leading to improved perfusion to the brain. Thus, in the first stages of the disease, the supply of substrates and oxygen can be maintained at near normal levels despite any absolute reduction of placental transfer. Longitudinal studies of progressively deteriorating IUGR fetuses have allowed elucidation of the natural history of these hemodynamic modifications during uteroplacental insufficiency.63–65 Such studies have shown that both TPV in the aorta and pulmonary arteries and the ratio between right and left ventricle outputs remain stable

540

Fetal Cardiology

(a)

Figure 36.12

(b)

Velocity waveforms from the umbilical artery in a normal fetus (a) and in a severely growth-restricted fetus (b). Note the reverse end-diastolic flow in the latter situation.

during serial recordings. These findings are consistent with the absence of other significant changes in outflow resistance (a parameter inversely related to TPV value) and with cardiac output redistribution after establishment of the brain-sparing mechanism.63 However, in deteriorating IUGR fetuses, PV and cardiac output gradually decline,

suggesting a progressive deterioration of cardiac function.63 As a consequence also the cardiac filling is impaired. Studies in the fetal venous circulation64,65 have demonstrated that an increase of IVC reverse flow during atrial contraction occurs with progressive fetal deterioration, suggesting a higher pressure gradient in the right atrium

Cardiac function in normal and growth-restricted fetuses

(a)

541

(b)

Figure 36.13 Velocity waveforms from the descending thoracic aorta in a normal fetus (a) and in a severely growth-restricted fetus (b). Note the reverse end-diastolic flow in the latter situation.

(a)

Figure 36.14

(b)

Velocity waveforms from the peripheral pulmonary artery in a normal fetus (a) and in a severely growth-restricted fetus (b). Note the absent end-diastolic flow in the latter situation.

542

Fetal Cardiology

(a)

(b)

Figure 36.15 Velocity waveforms from the middle cerebral artery in a normal fetus (a) and in a severely growth-restricted fetus (b). Note the increased diastolic flow in the latter situation.

Table 36.1 Hemodynamic changes occurring in fetal arterial vessels during hypoxemia and acidemia induced by uteroplacental insufficiency Vessel

Impedance to flow

Descending aorta Renal artery Femoral artery Peripheral pulmonary artery Mesenteric arteries Cerebral arteries Adrenal artery Splenic artery Coronary arteries

Increased Increased Increased Increased Increased Decreased Decreased Decreased Decreased

(Figure 36.16). The next step of the disease is the extension of the abnormal reversal of blood velocities in the IVC to the ductus venosus, inducing an increase of the S/A ratio mainly due to a reduction of the A component of the velocity waveform21,55 (Figure 36.17). Finally, the high venous pressure induces a reduction of velocity at enddiastole in the umbilical vein, causing typical end-diastolic pulsations66 (Figure 36.18). The development of these pulsations is close to the onset of fetal heart rate anomalies and is frequently associated with acidemia and fetal endocrine changes.67–69 At this stage, reduced or reversed enddiastolic velocity may also be present in pulmonary veins,68 and coronary blood flow may be visualized with higher velocity than in normally grown third trimester fetuses.

Figure 36.16 Inferior vena cava velocity waveforms in a severely growthrestricted fetus. Note the marked increase of the A-wave.

If the fetuses are not delivered, intrauterine death may occur after a median of 3.5 days.37 Studies of VEF in IUGR have demonstrated that it is significantly and symmetrically decreased at the level of both ventricles.14 The presence of a symmetrical decrease of VEF from both ventricles, despite the dramatically different hemodynamic conditions present in the vascular district of ejection of the two ventricles (i.e. reduced cerebral resistance for the left ventricle and increased splanchnic and placental resistances for the right ventricle),

Cardiac function in normal and growth-restricted fetuses

Figure 36.18

Figure 36.17

Umbilical artery and vein velocity waveforms in a severely growth-restricted fetus at 27 weeks of gestation. Note the absence of end-diastolic velocities in the artery and the presence of pulsations in the vein.

Ductus venosus velocity waveforms in a severely growthrestricted fetus. The A-wave is reversed.

Reduced cardiac contractility

543

Increased venous pressure

Preload

Abnormal venous waveforms

IVC

DV

Figure 36.19

UV

supports a pivotal role of the intrinsic myocardial function in the compensatory mechanism of the IUGR fetus following establishment of the brain-sparing effect. Indeed it was demonstrated in fetuses followed longitudinally until either intrauterine death or the onset of abnormal fetal heart rate patterns requiring early deliverythat VEF dramatically decreases in a short time interval (i.e. 1 week), showing an impairment of ventricular force close to fetal distress.14 Furthermore, a significant relationship between the severity of fetal acidosis at cordocentesis and VEF values14 validates the pivotal role of the fetal heart in progressive fetal compromise.

PV

Suggested pathophysiological steps in deteriorating fetuses with intrauterine growth restriction. IVC, inferior vena cava; DV, ductus venosus; UV, umbilical vein; PV, pulmonary vein.

A schematic description of these hemodynamic changes is presented in Figure 36.19. We speculate that the fall in cardiac output and VEF terminally may reflect decompensation of a normally protective mechanism responsible for the brain-sparing effect. According to this model, the fetal heart adapts to placental insufficiency in order to maximize substrates and oxygen supply to the brain. With the progressive deterioration of fetal conditions, this protective mechanism is overwhelmed by the fall of cardiac output, and fetal distress occurs. Although several studies have demonstrated that abnormal Doppler findings, particularly at the level of the ductus

544

Fetal Cardiology

venosus, can predict fetal acidosis, perinatal death, and major neonatal morbidities,25,69 there are still some doubts about their application in clinical practice.70 Indeed, other factors irrespective of the hemodynamic condition such as gestational age and birth weight play a major role in predicting neonatal outcome. As a consequence, Doppler studies of the ductus venosus seem to be a useful adjunct to standard monitoring of fetal well-being only after 29 weeks of gestation and for fetal weight > 600 g.70 Earlier predictors would therefore be desirable, and preliminary results suggest that the absolute quantification of umbilical flow and cardiac output may be useful in early identification of fetuses on the verge of compromise.32,34

References 1. Edelstone DI, Rudolph AM. Preferential streaming of ductus venosus blood to the brain and heart in fetal lambs. Am J Physiol 1979; 237: H724–9. 2. Rudolph AM. Distribution and regulation of blood flow in the fetal and neonatal lamb. Circ Res 1985; 57: 811–21. 3. Burns PN. Doppler flow estimations in the fetal and maternal circulations: principles, techniques and some limitations. In: Maulik D, McNellis D, eds. Doppler Ultrasound Measurement of Maternal–Fetal Hemodynamics. Ithaca, NY: Perinatology Press, 1987: 43–78. 4. Rizzo G, Arduini D, Romanini C. Doppler echocardiographic assessment of fetal cardiac function. Ultrasound Obstet Gynecol 1992; 2: 434–45. 5. Comstock CH, Riggs T, Lee W, Kirk J. Pulmonary to aorta diameter ratio in the normal and abnormal fetal heart. Am J Obstet Gynecol 1991; 165: 1038–43. 6. Gonçalves LF, Lee W, Chaiworapongsa T et al. Four-dimensional ultrasonography of the fetal heart with spatiotemporal image correlation. Am J Obstet Gynecol 2003; 189: 1792–802. 7. DeVore GR, Falkensammer P, Sklansky MS, Platt LD. Spatiotemporal image correlation (STIC): new technology for evaluation of the fetal heart. Ultrasound Obstet Gynecol 2003; 22: 380–7. 8. Chaoui R, Hoffmann J, Heling KS. Three-dimensional (3D) and 4D color Doppler fetal echocardiography using spatiotemporal image correlation (STIC). Ultrasound Obstet Gynecol 2004; 23: 535–45. 9. Yagel S, Cohen SM, Shapiro I, Valsky DV. 3D and 4D ultrasound in fetal cardiac scanning: a new look at the fetal heart. Ultrasound Obstet Gynecol 2007; 29: 81–95. 10. Rizzo G, Capponi A, Cavicchioni O et al. Examination of the fetal heart by four-dimensional ultrasound with spatiotemporal image correlation during routine second trimester examination: the ‘three steps technique’. Fetal Diagn Ther 2008; 24: 126–31. 11. Rizzo G, Capponi A, Cavicchioni O et al. Fetal cardiac stroke volume determination by four-dimensional ultrasound with spatio-temporal image correlation compared with twodimensional and Doppler ultrasonography. Prenat Diagn 2007; 1147–50.

12. Messing B, Cohen SM, Valsky DV et al. Fetal cardiac ventricle volumetry in the second half of gestation assessed by 4D ultrasound using STIC combined with inversion mode. Ultrasound Obstet Gynecol 2007; 30: 142–51. 13. Sutton MS, Gill T, Plappert T et al. Assessment of right and left ventricular function in terms of force development with gestational age in the normal human fetus. Br Heart J 1991; 66: 285–9. 14. Rizzo G, Capponi A, Rinaldo D et al. Ventricular ejection force in growth retarded fetuses. Ultrasound Obstet Gynecol 1995; 5: 247–52. 15. Isaaz K, Ethevenot G, Admant P et al. A new Doppler method of assessing left ventricular ejection force in chronic congestive heart failure. Am J Cardiol 1989; 64: 81–7. 16. Stottard MF, Pearson AC, Kern MJ et al. Influence of alteration in preload of left ventricular diastolic filling as assessed by Doppler echocardiography in humans. Circulation 1989; 79: 1226–36. 17. Gardin JM. Doppler measurements of aortic blood velocity and acceleration: load-independent indexes of left ventricular performance. Am J Cardiol 1989; 64: 935–6. 18. Bedotto JB, Eichorn EJ, Grayburn PA. Effects of left ventricular preload and afterload on ascendic aortic blood velocity and acceleration in coronaric artery disease. Am J Cardiol 1989: 64: 856–9. 19. Brownwall E, Ross J, Sonnenblick EH. Mechanism of Contraction in the Normal and Failing Heart, 2nd edn. Boston: Little Brown, 1976: 92–129. 20. Takaneka K, Dabestani A, Gardin JM et al. Left ventricular filling in hypertrophic cardiomyopathy: a pulsed Doppler echocardiographic study. J Am Coll Cardiol 1986; 7: 1263–71. 21. Kenny J, Plappert T, Doubilet P et al. Effects of heart rate on ventricular size, stroke volume and output in the normal human fetus: a prospective Doppler echocardiographic study. Circulation 1987; 76: 52–8. 22. Rizzo G, Caforio L, Arduini D, Romanini C. Effects of sampling site on inferior vena cava flow velocity waveforms. J Matern Fetal Investig 1992; 2: 153–6. 23. Reed KL, Appleton CP, Anderson CF et al. Doppler studies of vena cava flows in human fetuses - insights into normal and abnormal cardiac physiology. Circulation 1990; 81: 498–505. 24. Appleton CP, Hatle LK, Popp RL. Superior vena cava and hepatic vein Doppler echocardiography in healthy adults. J Am Coll Cardiol 1987; 10: 1032–9. 25. Rizzo G, Capponi A, Talone PE et al. Doppler indices from inferior vena cava and ductus venosus in predicting pH and oxygen tension in umbilical blood at cordocentesis in growth retarded fetuses. Ultrasound Obstet Gynecol 1996; 7: 401–10. 26. Kanzaki T, Chiba Y. Evaluation of preload condition of the fetus by inferior vena cava blood flow pattern. Fetal Diagn Ther 1990; 5: 168–74. 27. Okamura K, Murotsuki J, Kobajashi M et al. Umbilical venous pressure and Doppler flow patterns of inferior vena cava in the fetus. Am J Perinatol 1994; 11: 255–9. 28. Rizzo G, Capponi A, Arduini D, Romanini C. Ductus venosus velocity waveforms in appropriate and small for gestational age fetuses. Early Hum Dev 1994; 39: 15–26.

Cardiac function in normal and growth-restricted fetuses

29. Better DJ, Kaufman S, Allan LD. The normal pattern of pulmonary venous flow on pulsed Doppler examination of the human fetus. J Am Soc Echocardiogr 1996; 9: 281–5. 30. Rizzo G, Capponi A, Pasquini L et al. Abnormal fetal pulmonary venous blood flow velocity waveforms in the presence of complete transposition of the great arteries. Ultrasound Obstet Gynecol 1996; 7: 299–300. 31. Rizzo G, Arduini D, Romanini C. Pulsations in umbilical vein: a physiological finding in early pregnancy. Am J Obstet Gynecol 1992; 167: 675–7. 32. Kitabatake A, Inoue M, Asao M et al. Noninvasive evaluation of pulmonary hypertension by a pulsed Doppler technique. Circulation 1983; 68: 302–9. 33. Kiserud T, Ebbing C, Kessler J, Rasmussen S. Fetal cardiac output, distribution to the placenta and impact of placental compromise. Ultrasound Obstet Gynecol 2006; 28: 126–36. 34. Rizzo G, Capponi A, Cavicchioni O et al. Low cardiac output to the placenta: an early hemodynamic adaptive mechanism in growth restriction. Ultrasound Obstet Gynecol 2008; 32: 155–9. 35. Kiserud T, Ebbing C, Kessler J, Rasmussen S. Ductus venosus shunting in growth restricted fetuses and the effect of umbilical circulatory compromise. Ultrasound Obstet Gynecol 2006; 28: 143–9. 36. Labovitz AJ, Pearson C. Evaluation of left ventricular diastolic function: clinical relevance and recent Doppler echocardiographic insights. Am Heart J 1987; 114: 836–51. 37. Baschat AA, Gembruch U, Reiss I et al. Demonstration of fetal coronary blood flow by Doppler ultrasound in relation to arterial and venous flow velocity waveforms and perinatal outcome. The heart sparing effect. Ultrasound Obstet Gynecol 1997; 9: 162–72. 38. Laudy JAM, De Riddler MA, Wladimiroff J. Doppler velocimetry in branch pulmonary arteries of normal human fetuses during the second half of pregnancy. Pediatr Res 1997; 41: 897–901. 39. Rasanen J, Hutha JC, Weiner S et al. Fetal branch pulmonary arterial vascular impedance during the second half of pregnancy. Am J Obstet Gynecol 1996; 174: 1441–9. 40. Rizzo G, Capponi A, Chaoui R et al. Blood flow velocity waveforms from peripheral pulmonary arteries in normally grown and growth-retarded fetuses. Ultrasound Obstet Gynecol 1996; 8: 87–92. 41. Chaoui R, Taddei F, Rizzo G et al. Doppler echocardiography of the main stems of the pulmonary arteries in the normal human fetus. Ultrasound Obstet Gynecol 1998; 11: 173–9. 42. Huhta JC, Moise KJ, Fisher DJ et al. Detection and quantitation of constriction of the fetal ductus arteriousus by Doppler echocardiography. Circulation 1987; 75: 406–12. 43. Van de Mooren K, Barendregt LG, Waladimiroff J. Flow velocity waveforms in the human fetal ductus arteriosus during the normal second trimester of pregnancy. Pediatr Res 1991; 30: 487–90. 44. Rizzo G, Arduini D, Valensise H, Romanini C. Effects of behavioural states on cardiac output in the healthy human fetus at 36-38 weeks of gestation. Early Hum Dev 1990; 23: 109–15. 45. Huisman TW, Brezinka C, Stewart PA et al. Ductus venous flow velocity waveforms in relation to fetal behavioural states. Br J Obstet Gynaecol 1994; 101: 220–4.

545

46. Groenenberg IAL, Hop WCJ, Wladimiroff JW. Doppler flow velocity waveforms in the fetal cardiac outflow tract; reproducibility of waveform recording and analysis. Ultrasound Med Biol 1991; 17: 583–7. 47. Al-Ghazali W, Chita SK, Chapman MG, Allan LD. Evidence of redistribution of cardiac output in asymmetrical growth retardation. Br J Obstet Gynaecol 1989; 96: 697–704. 48. Reed KL, Meijboom EJ, Sahn DJ et al. Cardiac Doppler flow velocities in human fetuses. Circulation 1986; 73: 41–56. 49. Rizzo G, Arduini D, Romanini C. Fetal cardiac and extracardiac circulation in early gestation. J Matern Fetal Investig 1991; 1: 73–8. 50. van Splunder P, Stijnen T, Wladimiroff JW. Fetal atrioventricular flow-velocity waveforms and their relationship to arterial and venous flow velocity waveforms at 8 to 20 weeks of gestation. Circulation 1996; 94: 1372–8. 51. Wladimiroff JW, Huisman TWA, Stewart PA, Stijnen TH. Normal fetal Doppler inferior vena cava, transtricuspid and umbilical artery flow velocity waveforms between 11 and 16 weeks’ gestation. Am J Obstet Gynecol 1992; 166: 46–9. 52. Reed KL, Sahn DJ, Scagnelli S et al. Doppler echocardiographic studies of diastolic function in the human fetal heart: changes during gestation. J Am Coll Cardiol 1986; 8: 391–5. 53. Rizzo G, Arduini D, Romanini C, Mancuso S. Doppler echocardiographic assessment of atrioventricular velocity waveforms in normal and small for gestational age fetuses. Br J Obstet Gynaecol 1988; 95: 65–9. 54. Kenny JF, Plappert T, Saltzman DH et al. Changes in intracardiac blood flow velocities and right and left ventricular stroke volumes with gestational age in the normal human fetus: a prospective Doppler echocardiographic study. Circulation 1986; 74: 1208–16. 55. Rizzo G, Arduini D, Romanini C, Mancuso S. Doppler echocardiographic evaluation of time to peak velocity in the aorta and pulmonary artery of small for gestational age fetuses. Br J Obstet Gynaecol 1990; 97: 603–7. 56. Machado MVL, Chita SC, Allan LD. Acceleration time in the aorta and pulmonary artery measured by Doppler echocardiography in the midtrimester normal human fetus. Br Heart J 1987; 58: 15–18. 57. Allan LD, Chita SK, Al-Ghazali W et al. Doppler echocardiographic evaluation of the normal human fetal heart. Br Heart J 1987; 57: 528–33. 58. De Smedt MCH, Visser GHA, Meijboom EJ. Fetal cardiac output estimated by Doppler echocardiography during mid- and late gestation. Am J Cardiol 1987; 60: 338–42. 59. Rizzo G, Arduini D. Cardiac output in anencephalic fetuses. Gynecol Obstet Invest 1991; 32: 33–5. 60. Peeters LLH, Sheldon RF, Jones MD et al. Blood flow to fetal organ as a function of arterial oxygen content. Am J Obstet Gynecol 1979; 135: 637–46. 61. Arduini D, Rizzo G, Romanini C. Changes of Pulsatility Index preceding the onset of late decelerations in growth retarded fetuses. Obstet Gynecol 1992; 79: 605–10. 62. Rizzo G, Arduini D. Fetal cardiac function in intrauterine growth retardation. Am J Obstet Gynecol 1991; 165: 876–82. 63. Rizzo G, Arduini D, Romanini C. Inferior vena cava flow velocity waveforms in appropriate and small for gestational age fetuses. Am J Obstet Gynecol 1992; 166: 1271–80.

546

Fetal Cardiology

64. Hecker K, Hackeloer BJ. Cardiotocogram compared to Doppler investigation of the fetal circulation in the premature growth-retarded fetus: longitudinal observations. Ultrasound Obstet Gynecol 1997; 9: 152–61. 65. Rizzo G, Capponi A, Soregaroli M et al. Umbilical vein pulsations and acid base status at cordocentesis in growth retarded fetuses with absent end diastolic velocity in umbilical artery. Biol Neonate 1995; 68: 163–8. 66. Gudmundusson S, Tulzer G, Hutha JC, Marsal K. Venous Doppler in the fetus with absent end diastolic flow in umbilical artery. Ultrasound Obstet Gynecol 1996; 7: 262–7.

67. Capponi A, Rizzo G, De Angelis C et al. Atrial natriuretic peptide levels in fetal blood in relation to inferior vena cava velocity waveforms. Obstet Gynecol 1997; 89: 242–7. 68. Rizzo G, Capponi A, Angelini E et al. Pulmonary arterial and venous flow velocity waveforms in growth restricted fetuses. Ultrasound Obstet Gynecol 2000; 16 (Suppl 1): 7. 69. Baschat AA, Cosmi E, Bilardo CM et al. Predictors of neonatal outcome in early onset placental dysfunction. Obstet Gynecol 2007; 109: 253–61. 70. Ghidini A. Doppler of the ductus venosus in severe preterm fetal growth restriction: a test in search of a purpose? Obstet Gynecol 2007; 109: 250–2.

37 Venous flow in intrauterine growth restriction and cardiac decompensation Torvid Kiserud

Introduction Doppler examination of venous volume flow was introduced 25 years ago;1,2 it has since provided valuable physiological information, but so far not progressed to become a clinical tool. On the other hand, the assessment of venous pulsations3–5 has become part of the fetal hemodynamic evaluation. Both the chronic challenge of intrauterine growth restriction (IUGR) due to placental compromise and its final stages with cardiac decompensation are examples of how changes in fetal hemodynamics are reflected in the fetal veins. With reason, we believe that the venous velocity pattern is a more instantaneous indicator for hemodynamic performance than is the umbilical artery velocity pattern. In the following, mainly the ductus venosus and umbilical vein will be used to illustrate the general mechanisms operating in the precordial and peripheral fetal veins.

Physiological background We shall focus on the common clinical problem of IUGR due to placental hemodynamic compromise, but also keep in mind that a variety of other causes of altered growth may modify circulatory development and hemodynamic responses (e.g. malformations, chromosomal aberrations, metabolic disorders, infections, radiation, teratogens, drugs, smoking, alcohol, and malnutrition). Although the dimension of the umbilical vein is on average smaller in IUGR than in the appropriately grown fetus2,6,7 and flow is low,2,6,7 the umbilical venous pressure is maintained within normal ranges.8 There is an increased incidence of hypoxemia and acidosis in this group of fetuses.9 Based on the experience of animal studies, such fetuses are expected to have a reduced flow in the inferior compared to the superior vena cava, increased ductus venosus shunting,

and reduced pulmonary flow.10 In line with such experiments, it was actually found that IUGR fetuses had an augmented ductus venosus shunting,11–13 increased foramen ovale flow, and reduced flow through the fetal lungs,14 with canceled difference between the left and right cardiac output.14 In recent studies this concept has been somewhat modified, as follows. In the human version of placental compromise, it seems that mild and moderate degrees of compromise provoke this well-known pattern of redistribution, which includes brain-sparing. However, when progressing to further deterioration, which includes a reduction of umbilical venous return compared to the combined cardiac output,15 these compensatory mechanisms seem to break down.14–16 Lowoxygenated blood from the portal vein is increasingly blended into the ductus venosus shunt,11,17,18 the foramen ovale is relatively smaller19 with less blood shunted,14 less blood through the ductus arteriosus,14 and more pulmonary venous return to the left atrium,14 and the right ventricle takes a relatively larger proportion of the combined cardiac output.15 Another consequence of this deterioration is the reduced umbilical flow distributed to the fetal liver,11 a major determinant for fetal growth.20,21 Within the liver there is a sparing of the left lobe that continues to receive oxygenated umbilical blood, while the right lobe is down-prioritized, increasingly receiving deoxygenated portal blood.22,23 Hematocrit (and thus viscosity) tends to be higher in IUGR fetuses, compared to normal fetuses,24 with a higher concentration of cathecolamines,25 higher concentration of atrial natriuretic peptide,26 an augmented endothelin-1 response to cordocentesis,27 and an augmented cortisol response to hypoxemia.28 Any increase in hematocrit and viscosity induced by acute hypoxemia favors a redistribution of umbilical blood from the liver to the ductus venosus.29 The reduced umbilical venous return is expected to cause a lower pressure in the intra-abdominal umbilical vein, further reducing fetal

548

Fetal Cardiology

liver perfusion.29 An active distension of the inlet of the ductus venosus (mediated by nitroxide and prostaglandins) enhances this redistribution, particularly during hypoxemia.30–34 Interestingly, the distension is not confined to the isthmus (or sphincter) of the ductus venosus alone but to the entire length of the vessel.32,34 A more pronounced constrictive response in the portal branches than in the ductus venosus adds to the effect of deviating umbilical blood from the fetal liver.35 Experimentally, growthrestricted fetuses have on average a higher heart rate and a lower blood pressure, and in addition respond with less bradycardia to acute hypoxia, compared to appropriately grown fetuses.28 Gestational age is an important determinant. The fetus approaching term has more developed endocrine and neural regulation mechanisms to respond to hypovolemia and hypoxemia by an initial increase in blood pressure and peripheral resistance.36 During the early weeks of fetal life, dimensions are small, pressures low, and endocrine mechanisms poorly developed. Reduction in maternal PO2 is readily reflected in the amniotic fluid PO2 also before midgestation,37 but the circulatory responses at this early stage reflect rather direct effects on the heart and vasculature (e.g. bradycardia and reduced flow).38 Development and maturation of endocrine functions are still incomplete during the second trimester, which is reflected in the responses.39

Determinants of pulsatile flow in veins Apart from transporting blood towards the heart Figure 37.1), the fetal veins act as transmission lines for oppositely directed pressure waves.40–42 The inferior vena cava (IVC), ductus venosus, and umbilical vein form a clinically important transmission line for atrial waves toward the umbilical cord (Figure 37.2). Waves transmitted in veins obey the same fluid dynamic laws as in arteries, and are modified by several factors.41,43 One such factor is the direction of the pressure wave compared to the blood velocity.44 If the direction is the same for both, the pressure wave will cause a velocity increment, and conversely with opposite directions there will be a velocity deflection, which is common in the precordial veins (Figures 37.3 and 37.4).45 Augmented atrial contraction is a documented cause of increased pressure amplitude,46–48 and thus pulsatility of the blood velocity. Increased afterload and hypoxic stress lead to increased atrial contraction, mediated by an increased adrenergic drive, a common mechanism in severe IUGR and congestive heart failure. Other determinants are vessel wall stiffness and intravascular pressure.49 An increased intravascular and transmural pressure in the venous system reduces compliance,

Figure 37.1 Distributional mechanism of venous flow at the inferior caval inlet to the heart. Oxygenated umbilical blood is accelerated in the ductus venosus (DV) and directed toward the foramen ovale to enter the left atrium as preferential streaming. Deoxygenated blood in the abdominal portion of the inferior vena cava is predominantly directed to the right atrium (reproduced with permission from reference 42).

increases the speed of the wave, and promotes transport of pulsation further to the periphery. These factors are prominent in congestive heart failure. However, the single most important determinant for wave propagation is the phenomenon of wave reflection.49 Similar to the reflection and transmission of light at an interface, the pressure wave emitted from the atria is partially reflected and partially transmitted at the junction of veins with different impedance (Z), i.e. between the IVC and ductus venosus outlet, the ductus venosus inlet, and umbilical vein. The degree of reflection at the ductus venosus–umbilical vein junction, for example, is expressed in the reflection coefficient (RC): RC =

Reflected wave ZUV − Z DV = Incident wave ZUV + Z DV

where ZUV and ZDV represent the impedance of the umbilical vein and the ductus venosus, respectively. The crosssection (or diameter) is the main determinant for impedance of the vessel. Normally the diameter of the umbilical vein is four times larger than that of the isthmus of the ductus venosus (95% prediction limits: 2, 6).41 This implies a large step in impedance and thus an extensive reflection of waves at the junction between these venous sections, and correspondingly little is transmitted into the

Venous flow in intrauterine growth restriction

(a)

549

(b)

Figure 37.2 The pressure wave generated in the atria is emitted into the precordial veins. The inferior vena cava, ductus venosus, and umbilical vein form one clinically important transmission line for this wave. The intensity of the wave travelling down the line is reduced by reflections at each junction. Particularly the junction between the ductus venosus inlet and the umbilical vein represents a large step in impedance (due to the difference in cross-section) and causes most of the wave to be reflected, leaving little energy for transmission into the umbilical vein (a). An increase in the diameter of the ductus venosus (e.g. during hypoxemia) reduces the difference in impedance at the junction, leading to reduced wave reflection, increased transmission (b), and subsequently an increased probability of velocity pulsation in the umbilical vein (reproduced with permission from reference 42).

Pressure wave

Blood velocity

Flow direction (a)

(b)

Figure 37.3 A pressure wave that travels in the same direction as blood flow causes a blood velocity increment (a). An example is found at the umbilical ring in the abdominal wall (b). The pressure wave in the umbilical artery is transmitted to the intra-abdominal umbilical vein. The result is a velocity increment (arrow) as both pressure wave and velocity have the same direction.

550

Fetal Cardiology

Pressure wave

Blood velocity

Flow direction (a)

(b)

Figure 37.4 A pressure wave that travels in the opposite direction to flow causes a corresponding reduction in blood flow velocity (a). The umbilical venous velocity inflection (arrow) due to an augmented atrial contraction represents an example of a pressure wave traveling against the flow direction (b).

umbilical vein (Figure 37.2a). This is the main reason why no umbilical venous pulsation is seen normally during the second half of pregnancy. The other extreme would be that the diameter, and Z, were the same above and below the junction, RC = 0, leading to full transmission of the pressure wave and no reflection. Wall stiffness (and compliance) varies along the transmission line,50 and is associated with differences in impedance, and is therefore a source for wave reflection, but to a much lesser extent compared to the effect of diameter variation alone. Other prominent fluid dynamic determinants for umbilical venous pulsation are the physical properties of the umbilical vein itself.49,51 The large dimension of the vein allows it to function as a compliant reservoir, which requires a high amount of pulse energy to produce visible changes in blood velocity.49 However, increased stiffness of the wall (e.g. increased vascular tone), increased intravascular pressure (e.g. congestion), and a small diameter (e.g. early pregnancy)52 promote transformation of the pressure wave into visible velocity waves in the reservoir.53

Umbilical venous flow For fetuses without chromosomal or anatomical abnormalities and with a confirmed birth weight below the 2.5 centile, the umbilical blood flow is almost uniformly low before 32 weeks of gestation, while later in pregnancy, small fetuses seem to represent a greater variation in hemodynamics (Figure 37.5).4 Umbilical venous flow, and particularly normalized flow based on estimated fetal weight (mL kg−1 min−1) was early on suggested as a clinical assessment of growth-restricted fetuses.2 The low flow is in part due to the small dimension of the umbilical vein, and in part to a

Figure 37.5 Umbilical venous flow is generally low in cases of severe growth restriction, particularly in cases with compromised placental circulation reflected in an increased pulsatility index of the umbilical artery blood velocity (closed circles). Mean (thin rule) and 95% prediction limits (thick rules), are shown (reproduced with permission from reference 4).

lower blood velocity than in a normally grown fetus. The concept of assessing umbilical blood flow is based on sound physiological principles, as fetal development depends on the 30%15,54 of the combined cardiac output circulating the placenta (≤ 20% near term15). The volume flow can be calculated from the umbilical diameter (DUV) and either the weighted mean blood velocity1,55,56 (VUVwmean) or the maximum velocity4,56 (VUVmax) (Figure 37.5): 2

⎛D ⎞ 1. p ⎜ UV ⎟ ⋅VUVwmean ⎝ 2 ⎠ 2

⎛D ⎞ 2. p ⎜ UV ⎟ ⋅ 0.5VUV max ⎝ 2 ⎠

Venous flow in intrauterine growth restriction

Equation 1 carries the risk of overestimating flow as the lowest velocities in the vein are lost in the filter, particularly at higher angles of insonation, and is susceptible to interference or incomplete representation of the spatial distribution of the velocities. On the other hand, method 2 has the disadvantage of assuming that flow is parabolic, which may not always be the case.57 However, when comparing the two methods in low-risk pregnancies, they produce literally identical results.56 A more important source of error is the diameter measurement, as this error is expanded by a power of two in calculating flow. This is a likely reason why the method never gained broader acceptance. However, the recent refinements of ultrasound equipment with memory buffer, adjustable focusing, and high frequency transducers have revived volume flow assessment. Based on repeat measurements of the diameter, the error can be controlled,58,59 and a new set of reference ranges for human umbilical flow have been established based on both cross-sectional55,60,61 and longitudinal data.56,62 At 20 weeks of gestation the mean flow is 35 ml min−1 and at 40 weeks 240 ml min−1 in our hands.55 The corresponding normalized umbilical venous flow drops from 115 to 64 ml kg−1 min−1 during the same period. These results vary slightly with center and study56,60–62 and tend to be slightly lower than previously reported, which may in part be due to the fact that many of these studies are based on outer–inner measurements (leading edge) of vessel diameter rather than inner–inner measurements used today. The method can be advocated for physiological studies, and is a promising method for clinical use, particularly in serial observations (Figure 37.6). Combining the flow

551

measurement with the fetal abdominal circumference rather than with the estimated fetal weight is another suggested improvement of the method.60 Applying such methods, it has recently been shown that the umbilical venous return normally constitutes onethird of the combined cardiac output at mid-gestation (Figure 37.7a),15 which is in line with previous invasive63 and non-invasive studies.54 However, near term, the fraction is down to one-fifth, which implies a considerable degree of recirculation within the fetal body.15 Such normal fetuses maintain the same normalized cardiac output, 400 ml min−1 kg,−1 during the second half of pregnancy. Although the IUGR fetuses maintain the same normalized cardiac output, they cannot keep up the umbilical circulation correspondingly. The result is a lower fraction of the combined cardiac output returning as umbilical venous flow, those with extreme degrees of abnormal umbilical artery pulsatility having the lowest values (Figure 37.7b).15 It seems that this hemodynamic pattern precedes fetal growth impairment.64

Umbilical venous blood velocity Conventionally, blood velocity (and diameter) is measured distally in the straight portion of the intra-abdominal umbilical vein, but the cord has also been used. The sample volume should include the entire vessel cross-section and the angle of insonation should be zero or close to zero.65 As a major proportion of severe IUGR fetuses have a fetoplacental circulation with increased impedance, it is of no surprise that these fetuses have reduced umbilical flow reflected in reduced blood velocity.6,7 Fetuses with signs of intrauterine asphyxia seem to have the lowest velocities.5–7,9 However, umbilical venous velocities in IUGR show a substantial overlap with normal ranges, which makes the method unsuitable for discriminating small fetuses at risk of asphyxia or cases of impaired cardiac performance.

Pulsatile umbilical venous flow velocity

Figure 37.6 Serial measurements show how the umbilical venous flow improves in a case of congestive heart failure due to atrial flutter, when the heart rate gradually returns to sinus rhythm and normal myocardial contractility is restored. Mean (thin rule) and 95% prediction limits (thick rules) are shown.

Since Lingman et al introduced the pulsatile umbilical venous flow velocity as a sign of circulatory compromise in fetuses with imminent asphyxia,5 its simplicity has made it a commonly used marker. Gudmundsson et al expanded on this concept and showed that hydropic fetuses with congestive heart failure and pulsatile umbilical venous flow had a higher mortality than fetuses without pulsation.3 Pulsations in the umbilical vein are associated with hypoxemia and impaired acid–base status.66–68 On the other hand,

Fetal Cardiology

Placenta/CCO flow fraction (%)

552

80

80

70

70

60

60

50

50

40

40

30

30

20

20

10

10

0 17

37 22 27 32 Gestational age (weeks)

42

(a)

0 17

22

27

32

37

42

Gestational age (weeks) (b)

Figure 37.7 In normally growing fetuses, one-third of the combined cardiac output (CCO) is directed to the placenta at mid-gestation (a); after 32 weeks the average is one-fifth, with the lowest fraction near term. The implication is an increased degree of recirculation of deoxygenated blood within the fetal body near term. In general, growth-restricted fetuses tend to augment this effect as they direct a lower fraction of the CCO to the placenta (b), particularly those with absent or reversed diastolic flow in the umbilical artery (black circles). Open circles, normal pulsatility index (PI) in the umbilical artery; gray circles, umbilical artery PI > 97.5 centile (Fetal caradiac output, distribution to the placenta, and impact of placental compromise, Kiserud T, Ebbing C, Kessler J, Rasmussen S, Ultrasound Obstet Gynecol 2006; 28: 126–36, © International Society of Ultrasound in Obstetrics and Gynecology. Reproduced with permission. Permission is granted by John Wilen & Sons Ltd on behalf)15 of the ISUOG.

pulsation is a normal phenomenon in early pregnancy (< 13 weeks)52 and may also occur in the intra-abdominal portion of the vein in normal late pregnancy.53,69–71 To make the umbilical venous pulsation a useful diagnostic method, we need to distinguish the pulsations caused by augmented atrial contractions from those caused by other factors.72 Once fetal respiratory movements are excluded, the blood velocity in the umbilical vein can be regarded as a steady flow, and the two principal pressure waves that can influence the velocity come from the atria or the neighboring arterial system. As mentioned previously, the pressure wave of interest (e.g. the a-wave) is emitted from the atria to travel along the transmission line formed by the connecting veins, but in the opposite direction to the flow velocity, the result being a velocity deflection (Figure 37.4).45 In contrast, a pressure wave transmitted from the neighboring artery into the vein at the abdominal wall may end up traveling in the same direction as flow, causing a synchronized velocity increment (Figure 37.3). In most cases the two sources can be distinguished; the pressure pulse from the artery causes a smooth velocity increment (Figure 37.3) while the atrial

contraction wave is a shorter and sharper velocity deflection (Figure 37.4b). Factors promoting umbilical venous pulsation of atrial origin are: • augmented atrial contraction (increased after- and preload, adrenergic drive, hypoxia) • bradycardia (Frank–Starling mechanism on the atria) • active distension of the ductus venosus (e.g. hypoxia) • increased umbilical venous vascular tone (adrenergic responses) • reduced umbilical compliance (e.g. small dimension in IUGR, venous congestion) • low gestational age (i.e. small dimensions of veins). For such reasons, the impaired cardiac function may change the steady umbilical velocity pattern into a pulsatile pattern, and, with further deterioration, transmit the entire cardiac pressure variation into the umbilical vein velocity pattern (Figure 37.8). However, even a triphasic umbilical venous pulsation may be associated with prolonged intrauterine survival.73 These signs are more likely to appear in early gestation and in conjunction with a preexisting increased afterload or low heart rate.4

Venous flow in intrauterine growth restriction

553

Figure 37.8 Umbilical venous flow velocity is generally non-pulsatile during the second half of pregnancy (upper panel). An augmented atrial contraction may lead to a short deflection of the velocity (a) in the umbilical vein (middle panel). A further deterioration (lower panel) causes the entire pressure variation during the cardiac cycle to be reflected in the umbilical venous velocity. (s, ventricular systole; d, passive diastolic filling).

Ductus venosus flow During experimental hypoxia and hypovolemia, an increased fraction of the umbilical blood is shunted through the ductus venosus.74–78 Recent studies have confirmed that this mechanism, first described in animal experiments, also exists in the human fetus, but on a slightly different scale. The fraction normally shunted through the human ductus venosus is estimated to be 30% at 20 weeks and 20% after 30 weeks,55,62,79 small fetuses shunting relatively more blood through the ductus venosus than their larger counterparts.13,55,80 Recent studies have shown that the more severe is the placental compromise (assessed by umbilical artery pulsatility index (PI)), and the earlier in pregnancy it comes, the larger is the fraction of umbilical blood shunted through the ductus venosus.11,12 Volume flow assessment in the ductus venosus is a rather uncertain undertaking due to the small dimension of its isthmic portion, and should thus be limited to studying groups rather than individuals.58,59

Ductus venosus blood velocity The recommended method for recording blood velocity of the ductus venosus uses a large sample volume at the isthmic portion (the inlet) in a near sagittal scan (at a minimum angle of insonation), or an oblique transection of the fetal abdomen.81,82 In early pregnancy, the sample volume needs to be restricted in order to discriminate reversed a-wave and the velocities found in the umbilical and

hepatic veins.83,84 The blood velocity in the ductus venosus reflects essentially the same cardiac events as the recording in other precordial veins (Figure 37.9). However, its direct connection to the umbilical vein combined with its active control makes this a special vessel, whose diagnostic potential is not yet fully explored. The cardiac function can be evaluated by determining the absolute velocity at the various phases of the cardiac cycle,81,82,85–88 or by a pulsatility index.85,86,88,89 As such indexes are independent of angle correction, they have gained popularity. Normal ranges are established for several indexes.85,86,88 Zero or reversed velocity during atrial contraction, rather than the pulsatility index, has been advocated and used as a simple clinical sign of affected cardiac function,90 particularly in early pregnancy.83,91 In general, the severely growth-restricted fetus manages to maintain systolic blood flow velocity in the ductus venosus within normal ranges in spite of compromised cardiac function (Figure 37.10a).4 However, the a-wave is commonly augmented in precordial veins (including the ductus venosus), reflecting altered cardiac performance (Figures 37.10b and 37.11).4,81,92–97 Increased afterload and preload, and thus increased end-diastolic pressure, may be aggravated by direct effects of hypoxia and increased adrenergic drive, all leading to an augmented atrial contraction and a pronounced deflection during the a-wave in the ductus venosus Doppler recording (Figure 37.12). The amplification of the a-wave is likely to carry the wave further distally in the venous system than is seen in uncompromised fetuses. An increased venous pressure, increased vascular tone, and active distension of the ductus venosus greatly

554

Fetal Cardiology

Figure 37.9 Normal ductus venosus blood velocity reflecting the various components of the cardiac cycle: ventricular systole (s), passive diastolic filling (d), and the minimum reached during atrial contraction (a).

1

0.5 Velocity (m/s)

Velocity (m/s)

0.75 0.75 0.5 0.25

0.25 0 –0.25

0

–0.5 16

20

24

28

32

36

40

16

Gestational age (weeks) (a)

20

24

28

32

36

40

Gestational age (weeks) (b)

Figure 37.10 Severely growth-restricted fetuses manage to maintain peak systolic velocity within normal ranges in the ductus venosus in spite of the compromised placental circulation reflected in the increased pulsatility index of the umbilical artery blood velocity (filled circles in a). Rather, a circulatory decompensation is reflected in the reduced or negative velocity in the ductus venosus during atrial contraction (b). The sign seems to be more prominent before 32 weeks of gestation and is associated with perinatal death (filled circles). Mean (thin rule) and 95% prediction limits (thick rules), are shown (reproduced with permission from reference 4).

promote propagation along this transmission link to the umbilical vein. Hypoxemia causes distension of the ductus venosus by itself, thus reducing the difference in crosssection (and impedance) between the ductus venosus and the umbilical vein. This results in reduced wave reflection at the junction and an increased wave transmission and umbilical venous pulsation (Figure 37.2b). Advanced cardiac decompensation is characterized by greatly reduced or reversed velocity during the a-wave (Figure 37.12). A reduced velocity in the period between

the systolic and diastolic peak probably signifies a further deterioration into a preterminal state (Figure 37.12).98 In early pregnancy, a reversed a-wave in the ductus venosus is associated with chromosomal aberrations and cardiac malformations in fetuses with increased nuchal translucency.83,99,100 It is speculated that this may be due to suboptimal cardiac function rather than the cardiac decompensation seen in late pregnancy. Second trimester fetuses are also more prone to respond with an augmented a-wave than fetuses near term.4,41,90 Hypoxemia and other

Venous flow in intrauterine growth restriction

555

Velocimetry of the caval, hepatic, portal, and other fetal veins

Figure 37.11 Impaired cardiac function is commonly seen in severe cases of intrauterine growth restriction as increased pulsatility of the ductus venosus blood velocity, the main reason being reduced blood velocity during an augmented atrial contraction phase (a). s, systolic peak; d, diastolic peak.

Figure 37.12 A further deterioration of cardiac function leads to zero or reversed velocity during atrial contraction (a) and reduced velocity in the interval between the systolic (s) and diastolic peaks (d), probably due to impaired myocardial function.

insults tend to have a more direct and pronounced effect on the immature heart,38,101 while during the third trimester it is essentially the matured endocrine responses and reflexes that modulate hemodynamics,39 and a reversed a-wave in the ductus venosus is rarely seen (Figure 37.10). Studies of IUGR fetuses and otherwise compromised pregnancies have shown that fetuses with an increased a-wave, or increased pulsatility, have a higher risk of hypoxemia, acidosis, perinatal morbidity, and demise.4,67,92,93,102–104 However, as shown in Figure 37.10b, the changes we have described in the ductus venosus velocity pattern tend to be more pronounced in the second trimester, and signify a serious prognosis at an age when prematurity in itself carries a high risk of complications and demise.

As compared to the ductus venosus the IVC has lower velocities and a negative a-wave is a normal finding in most of the pregnancy,86,94,97,105,106 but otherwise essentially the same changes are seen in the decompensating IUGR fetus.86,94,97,105 Although ductus venosus Doppler velocimetry in the neonatal period represents exciting new possibilities,107,108 the pediatrician is more acquainted with examination of the IVC, sometimes making this vessel the preferred one for measurements also before birth. In the fetus, the velocity measurement is preferably taken in the abdominal portion of the IVC, below the hepatic confluence and ductus venosus outlet, due to the high risk of interference and increased variability.109 One study has shown that Doppler of the IVC gives a better prediction of hypoxemia and acidosis than does examination of the ductus venosus.104 This has not been reproduced.92,93 These vessels may tell different stories, particularly since the ductus venosus is specifically controlled and not connected to a capillary system but directly to the umbilical vein. Velocimetry of the hepatic veins is not commonly used, but the method is reproducible since these veins are easily accessed in most positions of the fetus.86,110,111 Again, the changes seen in the ductus venosus and caval veins during hemodynamic compromise are much the same in the hepatic veins. The magnitude and direction of blood velocity in the left portal vein has recently been suggested as a simple marker of circulatory decompensation.17,18,22 The blood velocity is recorded in the left portal branch between the ductus venosus inlet and the junction with the main portal stem. A low or reversed velocity in this section would reflect a down-prioritization of the oxygenation of the right liver lobe (i.e. left liver lobe sparing).23 Intracranial veins have recently attracted attention, with studies suggesting that increased pulsation may be used as another marker of circulatory decompensation.112,113

How to use venous Doppler in the evaluation Venous Doppler is increasingly being used in a battery of Doppler measurements to evaluate the fetal circulation. While the waveform of the umbilical artery commonly reflects longstanding changes in the placental circuit, the reduced pulsatility of the middle cerebral artery is interpreted as a response to impaired placental perfusion. The Doppler recordings of the ductus venosus and umbilical vein reflect instantaneous changes in cardiac function, and

556

Fetal Cardiology

are commonly added to further differentiate circulatory status and determine time of delivery.114 Such a battery has been shown to be useful in describing the sequence of changes that follow a deteriorating placental function before 32 weeks of gestation.102,103,115 The pattern is less clear after this stage of pregnancy. By adding the Doppler recording of the umbilical vein and ductus venosus, the prediction of fetal acidosis and intrauterine death can be refined,67 neonatal outcome better predicted,116 and the group at risk of developing necrotizing enterocolitis better defined.117 In order to have a more systematic approach, a ‘fetal cardiovascular profile score’ has been suggested118 and tried out.119

14.

15.

16.

17.

18.

References 1.

2.

3.

4.

5.

6.

7.

8.

9.

10. 11.

12.

13.

Gill RW. Pulsed Doppler with B-mode imaging for quantitative blood flow measurement. Ultrasound Med Biol 1979; 5: 223–35. Gill RW, Kossoff G, Warren PS, Garrett WJ. Umbilical venous flow in normal and complicated pregnancies. Ultrasound Med Biol 1984; 10: 349–63. Gudmundsson S, Huhta JC, Wood DC et al. Venous Doppler ultrasonography in the fetus with nonimmune hydrops. Am J Obstet Gynecol 1991; 164: 33–7. Kiserud T, Eik-Nes SH, Blaas H-G et al. Ductus venosus blood velocity and the umbilical circulation in the seriously growth retarded fetus. Ultrasound Obstet Gynecol 1994; 4: 109–14. Lingman G, Laurin J, Marsál K, Persson P-H. Circulatory changes in fetuses with imminent asphyxia. Biol Neonate 1986; 49: 66–73. Jouppila P, Kirkinen P. Umbilical vein blood flow as an indicator of fetal hypoxia. Br J Obstet Gynaecol 1984; 91: 107–10. Laurin J, Lingman G, Mars l K, Persson P-H. Fetal blood flow in pregnancies complicated by intrauterine growth retardation. Obstet Gynecol 1987; 69: 895–902. Ville Y, Sideris I, Hecher K et al. Umbilical venous pressure in normal, growth-retarded, and anemic fetuses. Am J Obstet Gynecol 1994; 170: 487–94. Laurin J, Marsál K, Persson P-H, Lingman G. Ultrasound measurement of fetal blood flow in predicting fetal outcome. Br J Obstet Gynaecol 1987; 94: 940–8. Rudolph AM. Distribution and regulation of blood flow in the fetal and neonatal lamb. Circ Res 1985; 57: 811–21. Bellotti M, Pennati G, de Gasperi C et al. Simultaneous measurements of umbilical venous, fetal hepatic, and ductus venosus blood flow in growth-restricted human fetuses. Am J Obstet Gynecol 2004; 190: 1347–58. Kiserud T, Kessler J, Ebbing C, Rasmussen S. Ductus venosus shunting in growth-restricted fetuses and the impact of umbilical circulatory compromise. Ultrasound Obstet Gynecol 2006; 28: 143–9. Tchirikov M, Rybakowski C, Hünecke B, Schröder HJ. Blood flow through the ductus venosus in singleton and

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

multifetal pregnancies and in fetuses with intrauterine growth retardation. Am J Obstet Gynecol 1998; 178: 943–9. Mäkikallio K, Jouppila P, Räsänen J. Retrograde aortic isthmus net blood flow and human fetal cardiac function in placental insufficiency. Ultrasound Obstet Gynecol 2003; 22: 351–7. Kiserud T, Ebbing C, Kessler J, Rasmussen S. Fetal cardiac output, distribution to the placenta, and impact of placental compromise. Ultrasound Obstet Gynecol 2006; 28: 126–36. Weiner Z, Farmakides G, Schulman H, Penny B. Central and peripheral hemodynamic changes in fetuses with absent end-diastolic velocity in umbilical artery: correlation with computerized fetal heart rate pattern. Am J Obstet Gynecol 1994; 170: 509–15. Kilavuz Ö, Vetter K, Kiserud T, Vetter P. The left portal vein is the watershed of the fetal venous system. J Perinat Med 2003; 31: 184–7. Kiserud T, Kilavuz Ö, Hellevik LR. Venous pulsation in the left portal branch - the effect of pulse and flow direction. Ultrasound Obstet Gynecol 2003; 21: 359–64. Kiserud T, Chedid G, Rasmussen S. Foramen ovale changes in growth-restricted fetuses. Ultrasound Obstet Gynecol 2004; 24: 141–6. Tchirikov M, Kertschanska S, Schroder HJ. Obstruction of ductus venosus stimulates cell proliferation in organs of fetal sheep. Placenta 2001; 22: 24–31. Tchirikov M, Kertschanska S, Sturenberg HJ, Schroder HJ. Liver blood perfusion as a possible instrument for fetal growth regulation. Placenta 2002; 23(Suppl A): S153–8. Kessler J, Rasmussen S, Kiserud T. The left portal vein as a watershed in the fetal circulation: development during the second half of pregnancy and a suggested method of evaluation. Ultrasound Obstet Gynecol 2007; 30: 757–64. Kessler J, Rasmussen S, Kiserud T. The shift of umbilicalportal watershed in growth-restricted fetuses assessed by velocity measurement in the left portal vein. Ultrasound Obstet Gynecol 2007; 30: 527. Jouppila P, Kirkinen P, Puukka R. Correlation between umbilical vein blood flow and umbilical blood viscosity in normal and complicated pregnancies. Arch Gynecol 1986; 237: 191–7. Jones CT, Robinson JS. Studies on experimental growth retardation in sheep. Plasma catecholamines in fetuses with small placentae. J Dev Physiol 1983; 5: 77–87. Capponi A, Rizzo G, De Angelis C et al. Atrial natriuretic peptide levels in fetal blood in relation to inferior vena cava velocity waveforms. Obstet Gynecol 1997; 89: 242–7. Rizzo G, Capponi A, Rinaldo D et al. Release of vasoactive agents during cordocenteseis: difference between normally grown and growth-restricted fetuses. Am J Obstet Gynecol 1996; 175: 563–70. Robinson JS, Jones CT, Kingston EJ. Studies on experimental growth retardation in sheep. The effect of maternal hypoxaemia. J Dev Physiol 1983; 5: 89–100. Kiserud T, Stratford L, Hanson MA. Umbilical flow distribution to the liver and ductus venosus: an in vitro investigation of the fluid dynamic mechanisms in the fetal sheep. Am J Obstet Gynecol 1997; 177: 86–90.

Venous flow in intrauterine growth restriction

30. Coceani F, Olley PM. The control of cardiovascular shunts in the fetal and perinatal period. Can J Pharmacol 1988; 66: 1129–34. 31. Coceani F. The control of the ductus venosus: an update. Eur J Pediatr 1993; 152: 976–7. 32. Kiserud T, Ozaki T, Nishina H et al. Effect of NO, phenylephrine and hypoxemia on the ductus venosus diameter in the fetal sheep. Am J Physiol 2000; 279: H1166–71. 33. Momma K, Takeuchi H, Hagiwara H. Pharmacological constriction of the ductus arteriosus and ductus venosus in fetal rats. In: Nora J, Takao A, eds. Congenital Heart Disease: Causes and Processes. Mount Kisco: Futura, 1984: 313–27. 34. Momma K, Ito T, Ando M. In situ morphology of the ductus venosus and related vessels in the fetal and neonatal rat. Pediatr Res 1992; 32: 386–9. 35. Tchirikov M, Kertschanska S, Schroder HJ. Differential effects of catecholamines on vascular rings from ductus venosus and intrahepatic veins of fetal sheep. J Physiol 2003; 548: 519–26. 36. Jensen A, Berger R eds. Regional Distribution of Cardiac Output. Cambridge: Cambridge University Press, 1993. 37. Jauniaux E, Kiserud T, Ozturk O et al. Amniotic gas values and acid-base status during acute maternal hyperoxemia and hypoxemia in the early fetal sheep. Am J Obstet Gynecol 2000; 182: 661–5. 38. Kiserud T, Jauniaux E, West D et al. Circulatory responses to acute maternal hyperoxaemia and hypoxaemia assessed non-invasively by ultrasound in fetal sheep at 0.3–05 gestation. Br J Obstet Gynaecol 2001; 108: 359–64 39. Hanson MA. Do we now understand the control of the fetal circulation? Eur J Obstet Gynecol Reprod Biol 1997; 75: 55–61. 40. Kiserud T, Crowe C, Hanson M. Ductus venosus agenesis prevents transmission of central venous pulsations to the umbilical vein in the fetal sheep. Ultrasound Obstet Gynecol 1998; 11: 190–4. 41. Kiserud T. Hemodynamics of the ductus venosus. Eur J Obstet Gynecol Reprod Biol 1999; 84: 139–47. 42. Kiserud T. Fetal venous circulation – an update on hemodynamics. J Perinat Med 2000; 28: 90–6. 43. Acharya G, Kiserud T. Ductus venosus blood velocity and diameter pulsations are more prominent at the outlet than at the inlet. Eur J Obstet Gynecol Reprod Biol 1999; 84: 149–54. 44. Nichols WW, O’Rourke MF. McDonald’s Blood Flow in Arteries. Theoretical, Experimental and Clinical Principles, 4th edn. London: Arnold, 1998. 45. Kiserud T. The fetal venous circulation. Fetal Matern Med Rev 2003; 14: 57–97 46. Hasaart TH, de Haan J. Phasic blood flow patterns in the common umbilical vein of fetal sheep during umbilical cord occlusion and the influence of autonomic nervous system blockade. J Perinat Med 1986; 14: 19–26. 47. Reed KL, Chaffin DG, Anderson CF. Umbilical venous Doppler velocity pulsations and inferior vena cava pressure elevations in fetal lambs. Obstet Gynecol 1996; 87: 617–20. 48. Reuss ML, Rudolph AM, Dae MW. Phasic blood flow patterns in the superior and inferior venae cavae and

49.

50.

51.

52.

53.

54.

55.

56.

57.

58.

59.

60.

61.

62.

63.

64.

65.

557

umbilical vein of fetal sheep. Am J Obstet Gynecol 1983; 145: 70–6. Hellevik LR, Stergiopulos N, Kiserud T et al. A mathematical model of umbilical venous pulsation. J Biomech 2000; 33: 1123–30. Hellevik LR, Kiserud T, Irgens F et al. Mechanical properties of the fetal ductus venosus and umbilical vein. Heart Vessels 1998; 13: 175–80. Nakai Y, Imanaka M, Nishio J, Ogita S. Umbilical venous pulsation associated with hypercoiled cord in growthretarded fetuses. Gynecol Obstet Invest 1997; 43: 6–7. Rizzo G, Arduini D, Romanini C. Umbilical vein pulsation: a physiological finding in early gestation. Am J Obstet Gynecol 1992; 167: 675–7. Skulstad SM, Kiserud T, Rasmussen S. The effect of vascular constriction on umbilical venous pulsation. Ultrasound Obstet Gynecol 2004; 23: 126–30. Sutton MSJ, Plappert T, Doubilet P. Relationship between placental blood flow and combined ventricular output with gestational age in normal fetuses. Cardiovasc Res 1991; 25: 603–8. Kiserud T, Rasmussen S, Skulstad SM. Blood flow and degree of shunting through the ductus venosus in the human fetus. Am J Obstet Gynecol 2000; 182: 147–53. Acharya G, Wilsgaard T, Berntsen GKR et al. Reference ranges for umbilical vein blood flow in the second half of pregnancy based on longitudinal data. Prenat Diagn 2005; 25: 99–111. Pennati G, Bellotti M, de Gasperi C, Rognoni G. Spatial velocity profile changes along the cord in normal human fetuses: can these affect Doppler measurements of venous umbilical blood flow? Ultrasound Obstet Gynecol 2004; 23: 131–7. Kiserud T, Rasmussen S. How repeat measurements affect mean diameter of the umbilical vein and the ductus venosus. Ultrasound Obstet Gynecol 1998; 11: 419–25. Kiserud T, Saito T, Ozaki T et al. Validation of diameter measurements by ultrasound. Intra-observer and interobserver variation assessed in vitro and in the fetal sheep. Ultrasound Obstet Gynecol 1999; 13: 52–7. Barbera A, Galan HL, Ferrazzi E et al. Relationship of umbilical vein blood flow to growth parameters in the human fetus. Am J Obstet Gynecol 1999; 181: 174–9. Boito S, Struijk PC, Ursem NTC et al. Umbilical venous volume flow in the normally developing and growthrestricted human fetus. Ultrasound Obstet Gynecol 2002; 19: 344–9. Kessler J, Rasmussen S, Godfrey K et al. Longitudinal study of umbilical and portal venous flow to the fetal liver: low pregnancy weight gain is associated with preferential supply to the left fetal liver lobe. Pediatr Res 2008; 63: 315–20. Rudolph AM, Heymann MA, Teramo K et al. Studies on the circulation of the previable human fetus. Pediatr Res 1971; 5: 452–65. Rizzo G, Capponi A, Cavicchioni O et al. Evidence of early changes in the distribution of cardiac output to the placenta in fetal growth restriction. Ultrasound Obstet Gynecol 2007; 30: 444. Gill RW, Trudinger BJ, Garrett WJ et al. Fetal umbilical venous flow measured in utero by pulsed Doppler and

558

66.

67.

68.

69.

70.

71.

72.

73.

74.

75.

76.

77.

78.

79.

80.

81.

Fetal Cardiology

B-mode ultrasound. Am J Obstet Gynecol 1981; 139: 720–5. Baschat AA, Gembruch U, Gorther L et al. Neonatal nucleated red blood cell counts in growth-restricted fetuses: relationship to arterial and venous Doppler studies. Am J Obstet Gynecol 1999; 181: 190–5. Baschat AA, Guclu S, Kush ML et al. Venous Doppler in the prediction of acid-base status of growth-restricted fetuses with elevated placental blood flow resistance. Am J Obstet Gynecol 2004; 1991: 277–84. Rizzo G, Capponi A, Soregaroli M et al. Umbilical vein pulsation and acid-base status at cordocentesis in growthretarded fetuses with abscent end-diastolic velocity in the umbilical artery. Biol Neonate 1995; 68: 163–8. Nakai Y, Miyazaki Y, Matsuoka Y et al. Pulsatile umbilical venous flow and its clinical significance. Br J Obstet Gynaecol 1992; 99: 977–80. Nakai Y, Imanaka M, Nishio J, Ogita S. Umbilical cord venous pulsation in normal fetuses and its incidence after 13 weeks gestation. Ultrasound Med Biol 1994; 21: 443–6. van Splunder IP, Huisman TWA, Stijnen T, Wladimiroff JW. Presence of pulsations and reproducibility of waveform recording in the umbilical and left portal vein in normal pregnancies. Ultrasound Obstet Gynecol 1994; 4: 49–53. Huhta JC. Deciphering the hieroglyphics of venous Doppler velocities. Ultrasound Obstet Gynecol 1997; 9: 300–1. Baschat AA, Gembruch U. Triphasic umbilical venous blood flow with prolonged survival in severe intrauterine growth retardation: a case report. Ultrasound Obstet Gynecol 1996; 8: 201–5. Behrman RE, Lees MH, Peterson EN et al. Distribution of the circulation in the normal and asphyxiated fetal primate. Am J Obstet Gynecol 1970; 108: 956–69 Edelstone DI, Rudolph AM, Heymann MA. Effect of hypoxemia and decreasing umbilical flow on liver and ductus venosus blood flows in fetal lambs. Am J Physiol 1980; 238: H656–63. Itskovitz J, LaGamma EF, Rudolph AM. Effects of cord compression on fetal blood flow distribution and O2 delivery. Am J Physiol 1987; 252: H100–9. Meyers RL, Paulick RP, Rudolph CD, Rudolph AM. Cardiovascular responses to acute, severe haemorrhage in fetal sheep. J Dev Physiol 1991; 15: 189–97. Paulick RP, Meyers RL, Rudolph CD, Rudolph AM. Venous and hepatic vascular responses to indomethacin and prostaglandin E1 in the fetal lamb. Am J Obstet Gynecol 1990; 163: 1357–63. Bellotti M, Pennati G, De Gasperi C et al. Role of ductus venosus in distribution of umbilical flow in human fetuses during second half of pregnancy. Am J Physiol 2000; 279: H1256–63. Emerson DS, Cartier MS, Felker RE et al. Increased shunting of umbilical venous blood via the ductus venosus in fetuses with IUGR: a Doppler study. J Ultrasound Med 1991; 10: S55. Kiserud T, Eik-Nes SH, Blaas H-G, Hellevik LR. Ultrasonographic velocimetry of the fetal ductus venosus. Lancet 1991; 338: 1412–14

82. Kiserud T, Eik-Nes SH, Hellevik LR, Blaas H-G. Ductus venosus – a longitudinal doppler velocimetric study of the human fetus. J Matern Fetal Investig 1992; 2: 5–11. 83. Borrell A, Antolin E, Costa D et al. Abnormal ductus venosus blood flow in trisomy 21 fetuses during early pregnancy. Am J Obstet Gynecol 1998; 179: 1612–17. 84. Matias A, Montenegro N, Areias JC. Anomalous venous return associated with major chromosomopathies in the late first trimester of pregnancy. Ultrasound Obstet Gynecol 1998; 11: 209–13. 85. Bahlmann F, Wellek S, Reinhardt I et al. Reference values of ductus venosus flow velocities and calculated waveform indices. Prenat Diagn 2000; 20: 623–34. 86. Hecher K, Campbell S, Snijders R, Nicolaides K. Reference ranges for fetal venous and atrioventricular blood flow parameters. Ultrasound Obstet Gynecol 1994; 4: 381–90. 87. Huisman TWA, Stewart PA, Wladimiroff JW. Ductus venosus blood flow velocity waveforms in the human fetus - a doppler study. Ultrasound Med Biol 1992; 18: 33–7. 88. Kessler J, Rasmussen S, Kiserud T. Longitudinal reference ranges for ductus venosus flow velocity and waveform indices. Ultrasound Obstet Gynecol 2006; 28: 890–8. 89. DeVore GR, Horenstein J. Ductus venosus index: a method for evaluating right ventricular preload in the second-trimester fetus. Ultrasound Obstet Gynecol 1993; 3: 338–42. 90. Kiserud T, Eik-Nes SH, Hellevik LR, Blaas H-G. Ductus venosus blood velocity changes in fetal cardiac diseases. J Matern Fetal Investig 1993; 3: 15–20. 91. Matias A, Huggon I, Areias JC et al. Cardiac defects in chromosomally normal fetuses with abnormal ductus venosus blood flow at 10-14 weeks. Ultrasound Obstet Gynecol 1999; 14: 307–10. 92. Hecher K, Campbell S, Doyle P et al. Assessment of fetal compromise by Doppler ultrasound investigation of the fetal circulation. Ciculation 1995; 91: 129–38. 93. Hecher K, Snijders R, Campbell S, Nicolaides K. Fetal venous, intracardiac, and arterial blood flow measurements in intrauterine growth retardation: relationship with fetal blood gases. Am J Obstet Gynecol 1995; 173: 10–15. 94. Kanzaki T, Chiba Y. Evaluation of the preload condition of the fetus by inferior vena caval blood flow pattern. Fetal Diagn Ther 1990; 5: 168–74. 95. Nakata M. Doppler-velocity waveforms in ductus venosus in normal and small-for-gestational-age fetuses. J Obstet Gynaecol Res 1996; 22: 489–96. 96. Reed KL, Appleton CP, Anderson CF et al. Doppler studies of vena cava flows in human fetuses; insights into normal and abnormal cardiac physiology. Circulation 1990; 81: 498–505. 97. Rizzo G, Arduini D, Romanini C. Inferior vena cava flow velocity waveforms in appropriate- and small-for-gestational-age fetuses. Am J Obstet Gynecol 1992; 166: 1271–80. 98. Szunyogh N, Thuring A, González R et al. Ductus venosus systolic and early diastolic wave indices: new markers of pre-terminal changes in cardic function. Ultrasound Obstet Gynecol 2006; 28: 392.

Venous flow in intrauterine growth restriction

99. Matias A, Gomes C, Flack N et al. Screening for chromosomal defects at 11-14 weeks: the role of ductus venosus blood flow. Ultrasound Obstet Gynecol 1998; 12: 380–4. 100. Murta CG, Moron AF, Avila MA, Weiner CP. Application of ductus venosus Doppler velocimetry for the detection of fetal aneuploidy in the first trimester of pregnancy. Fetal Diagn Ther 2002; 17: 308–14. 101. Iwamoto HS, Kaufman T, Keil LC, Rudolph AM. Responses to acute hypoxemia in fetal sheep at 0.6–0.7 gestation. Am J Physiol 1989; 256: H613–20. 102. Ferrazzi E, Bozzo M, Rigano S et al. Temporal sequence of abnormal Doppler changes in peripheral and central circulatory systems of the severely growth-restricted fetus. Ultrasound Obstet Gynecol 2002; 19: 140–6. 103. Hecher K, Bilardo CM, Stigter RH et al. Monitoring of fetuses with intrauterine growth restriction: a longitudinal study. Ultrasound Obstet Gynecol 2001; 18: 564–70. 104. Rizzo G, Capponi A, Talone P et al. Doppler indices from inferior vena cava and ductus venosus in predicting pH and oxygen tension in umbilical blood at cordocentesis in growth-retarded fetuses. Ultrasound Obstet Gynecol 1996; 7: 401–10. 105. Huisman TWA, Stewart PA, Wladimiroff JW, Stijnen T. Flow velocity waveforms in the ductus venosus, umbilical vein and inferior vena cava in normal human fetuses at 12-15 weeks of gestation. Ultrasound Med Biol 1993; 19: 441–5. 106. Huisman TWA, Stewart PA, Wladimiroff JW. Flow velocity waveforms in the fetal inferior vena cava during the second half of normal pregnancy. Ultrasound Med Biol 1991; 17: 679–82. 107. Fugelseth D, Lindemann R, Liestøl K et al. Ultrasonographic study of ductus venosus in healthy neonates. Arch Dis Child 1997; 77: F131–4. 108. Fugelseth D, Kiserud T, Liestøl K et al. Ductus venosus blood velocity in persistent pulmonary hypertension of the newborn. Arch Dis Child 1999; 81: F35–9.

559

109. Rizzo G, Caforio L, Arduini D, Romanini C. Effect of sampling site on inferior vena cava flow velocity waveforms. J Matern Fetal Investig 1992; 2: 153–6. 110. Hofstaetter C, Gudmundsson S, Dubiel M, Marsál K. Fetal right hepatic venous blood velocity in normal and high-risk pregnancies. Eur J Ultrasound 1996; 4: 153–60. 111. Mari G, Uerpairojkit B, Copel JA. Abdominal venous system in the normal fetus. Obstet Gynecol 1995; 86: 729–33. 112. Cheema R, Dubiel M, Breborowicz G, Gudmundsson S. Fetal cerebral venous Doppler velocitmetry in normal and high-risk pregnancy. Ultrasound Obstet Gynecol 2004; 24: 147–53. 113. Li H, Gudmundsson S, Olofsson P. Acute changes of cerebral venous blood flow velocity waveforms during provoked uterine contractions in high-risk pregnancy. Ultrasound Obstet Gynecol 2004; 24: 516–21. 114. Baschat AA. Arterial and venous Doppler in the diagnosis and management of early onset growth restriction. Early Hum Dev 2005; 81: 877–87. 115. Baschat AA, Gembruch U, Harman CR. The sequence of changes in Doppler and biophysical parameters as severe fetal growth restriction worsens. Ultrasound Obstet Gynecol 2001; 18: 571–7. 116. Baschat AA, Cosmi E, Bilardo CM et al. Predictors of neonatal outcome in early-onset placental dysfunction. Obstet Gynecol 2007; 109: 253–61. 117. Manogura AC, Turan O, Kush ML et al. Predictors of necrotizing enterocolitis in preterm growth-restricted neonates. Am J Obstet Gyncol 2008; 198: 638.e1–5. 118. Huhta J. Guidlines for the evaluation of heart failure in the fetus with or without hydrops. Pediatr Cardiol 2004; 25: 274–86. 119. Mäkikallio K, Räsänen J, Mäkikallio T et al. Human fetal cardiovascular profile score and neonatal outcome in intrauterine growth restriction. Ultrasound Obstet Gynecol 2008; 31: 48–54.

38 Congestive heart failure in the fetus James C Huhta Introduction Fetal echocardiography has progressed to be able to diagnose many forms of congenital heart disease and to assess the prognosis of cardiac lesions based on their anatomy and presentation in utero. However, the presence of signs of fetal heart failure such as hydrops or valvular regurgitation makes the assessment of prognosis more difficult. Congestive heart failure, defined as inadequate tissue perfusion for normal organ development and function, is a final common pathway for all fetal disease states that lead to demise in utero. A tool for this assessment is the Cardiovascular Profile Score, which combines ultrasonic markers of fetal cardiovascular function based on univariate parameters that have been correlated with perinatal mortality documented in the literature (see the Appendix following the References section of this chapter).1 This profile could then become the ‘heart failure score’ and could potentially be used in the same way as, and in combination with, the Biophysical Profile Score (a non-invasive tool for assessment of brain function). This chapter will review the pathophysiology of congestive heart failure prenatally and present a straightforward method for the rapid evaluation of the fetus that may have fetal congestive heart failure. The progression from normal to the prehydropic state, to hydrops, will be illustrated. The use of the score in the hydropic fetus to ascertain whether or not there are signs of congestive heart failure as the cause will be discussed. Mechanisms of congestive heart failure such as increased afterload (such as seen in the twin–twin transfusion recipient), valvular or myocardial failure (such as seen in Ebstein’s malformation of the tricuspid valve), or high output failure (such as seen with anemia or sacrococcygeal teratoma) will be illustrated.

The fetal circulation After birth, the circulation consists of desaturated venous blood entering the right heart, and passing to the lungs with little or no admixture with oxygenated blood. The pulmonary venous blood is highly oxygenated and passes

through the left atrium and left ventricle to the aorta. Therefore, the postnatal circulation is a series circuit. The fetal circulation is unique and significantly different from that in the newborn, infant, or child. Knowledge of these morphologic and physiologic differences is crucial for the understanding and assessment of fetal cardiovascular function. Most of this knowledge is derived from experimental animal data and, more recently, ultrasonography, which has allowed direct observation of the circulation in normal and abnormal human fetuses. In the fetus, the ventricles pump blood in parallel rather than in series, with the left ventricle pumping to the aorta and upper body, and the right ventricle pumping to the ductus arteriosus and the lower body and placenta.2 The lungs have a high resistance in utero, and the placenta fulfills the role of oxygenating the blood and ridding the body of wastes, requiring a significant proportion of the combined cardiac output. The highly oxygenated blood from the placenta passes to the ductus venosus where a portion bypasses the liver and passes predominantly to the left atrium. The relatively deoxygenated blood from the upper body passes to the tricuspid valve and then to the ductus arteriosus and lungs. The deoxygenated blood from the inferior vena cava and the right hepatic veins is directed to the right atrium and predominantly to the tricuspid valve. This distribution of lower body flow is accomplished by the posterior portion of the inferior vena cava connecting directly to the foramen ovale and the superior portion of the atrial septum, the crista dividens, which overlies the inferior vena cava, effectively dividing it into two streams. Therefore, the presence of three shunts (ductus venosus, foramen ovale, and ductus arteriosus) allows the fetal heart to work with two parallel rather than one series circulation. Right and left atrial pressures are almost equal because of the presence of the foramen ovale, and right and left ventricular pressures are equal due to the ductus arteriosus. The left ventricle ejects into the upper body and cerebral circulation, and the right ventricle ejects into the pulmonary arteries and through the ductus arteriosus into the lower body and the placental circulation. The vascular beds of the upper and lower body are connected via the aortic isthmus. As a further consequence of the parallel

562

Fetal Cardiology

arrangement of these circulations, ventricular outputs can be different and, in the case of obstruction on one side of the heart, the other side is able to increase its work or even completely supply the whole circulation alone.

Factors affecting perinatal cardiac output The cardiac ventricular output is the product of the heart rate and the stroke volume of both the right and left ventricles. The stroke volume is determined by the preload, afterload, and myocardial contractility of each ventricle. An increase in the stretch of the ventricular chamber will result in an increased contractility and stroke volume according to the Frank–Starling mechanism. This mechanism is now known to be present as early as 8 weeks’ gestation in the human. This allows the cardiac output to be unchanged during periods of heart rate change between 50 and 200 beats per minute. Distribution of cardiac output prenatally is different from that after birth and changes throughout gestation.3 The fetal right ventricle always ejects more blood than the left ventricle (in the fetal lamb almost twice as much); the lungs are perfused only by a small (8–25%) proportion of the combined ventricular output (CVO), the aortic isthmus 10%, and the placenta receives up to 40–45% of the CVO. Umbilical vein flow is divided at the entrance to the liver, where approximately 50% of its highly oxygenated blood is shunted through the ductus venosus directly to the inferior vena cava (IVC), right atrium, and through the foramen ovale to the left atrium. The other 50% passes through the liver before reaching the IVC. Streaming within the IVC is such that the higher oxygenated blood, coming from the ductus venosus, is directed to the left atrium, whereas lower oxygenated blood from the hepatic circulation and superior vena cava (SVC) preferentially enters the right ventricle. As a consequence, the coronary arteries and brain are perfused with higher oxygenated blood. In the case of circulatory compromise, the fetus will take advantage of these shunts to shift more oxygenated blood to the left side of the heart, thus ensuring adequate perfusion of vital organs. Throughout gestation, the distribution of the CVO changes such that, with increasing gestational age, the lower body, lungs, gut, and brain receive a higher, and placenta and kidneys a lower percentage.4 Because the pulmonary and systemic circulations are separate in the fetus, each ventricle has a stroke volume determined by the individual preload, afterload, and contractility of that chamber. Both ventricles are linked by a common heart rate and humoral environment. They are also linked by the atrial pressures which are similar due to the presence of the foramen ovale. They are also linked by

the ventricular septum which is shared by each ventricle, and by the common arterial pressure which is the result of the widely patent ductus arteriosus. The unique feature of the parallel nature of the ventricular ejection is that if there is increased afterload of one ventricle, the output of that ventricle will fall and the output of the contralateral ventricle will increase in a compensatory manner. This leads to the commonly observed feature associated with congenital heart disease of disproportionate growth of the normal side of the heart. A functional separation of the ventricles at the level of the aortic isthmus has been observed such that any blood pressure drop in the lower body causes increased right ventricular output without a change of ascending aortic pressure. The metabolic source of energy for the fetal myocardium is glucose almost exclusively. In adults, fatty acids are the major source of energy for the myocardium. Growth or increased workload in the fetus results in hyperplasia of the myocardium with an increased number of cells, whereas growth of the myocardium after birth is increased only by cell size or hypertrophy (increased protein content of each cell). Besides the difference in contractility, the fetal heart reacts differently in response to pre- and afterload changes. Several studies in isolated myocardium and intact hearts have demonstrated a reduced compliance of the fetal myocardium.5 Studies of the effect of preload changes on cardiac output in the fetus have shown that, while a reduction of preload results in a decrease of cardiac output, cardiac output rises only when filling pressures increase 2–4 mmHg above resting pressures, but a further increase of atrial pressure does not result in greater ventricular output.4 This is in contrast to postnatal hearts, where a progressive increase in ventricular output is observed with an atrial pressure rise to 15–20 mmHg. However, there is usually an interaction with afterload. Increased afterload in the fetus is followed by a reduction in myocardial shortening and in stroke volume. So, if arterial pressures are kept constant, ventricular stroke volume increases even with atrial pressures as high as 10–15 mmHg.6 Thus, the Frank–Starling mechanism is present in the fetus although it operates at the upper limit. In the fetus, one has to consider that right ventricular afterload is mainly determined by the vascular bed of the placenta, while left ventricular output is determined by the cerebral circulation. The effect of heart rate on fetal CVO is much more pronounced than postnatally. The fetus has a range of heart rates between 50 and 200 beats per minute, at which the stroke volume of the ventricular chambers can adapt to maintain adequate CVO and tissue perfusion. Outside of this range, heart failure will result. In summary, the major determinant of cardiac output is the afterload of the fetal ventricle. Any influence which raises the impedance of ejection will inversely lower the ventricular stroke volume by the effect on both the systolic and diastolic function of the heart. For example, in growth restriction in

Congestive heart failure in the fetus

the fetus due to placental dysfunction, the combined cardiac output drops due to increased placental resistance.

The transitional circulation After birth, the function of gas exchange is transferred from the placenta to the lungs. The major changes in the circulation after birth are the decrease in pulmonary vascular resistance and the closure of the ductus arteriosus and foramen ovale. The ductus closes within 2–3 days in the term neonate, and the patency of the ductus during this time results in a significant left-to-right shunt. This raises the left atrial pressure and effectively restricts the right-to-left shunting at the atrial level. Shunting through the ductus venosus ceases normally within 2–3 days after birth.

The etiology of hydrops fetalis Cardiac failure in the fetus End-stage fetal heart failure results in hydrops fetalis. The reduced ability of the fetal heart to contract and to generate force, the lower myocardial compliance and the diminished Frank–Starling mechanism, the higher dependence of cardiac output on heart rate, the lack of adrenoceptors – all these contribute to decreased cardiac reserve in response to stress and to a higher susceptibility of the fetus for the development of cardiac failure.

Factors contributing to hydrops Several features are responsible for fluid accumulation in fetal tissue. The final common pathway of many different conditions compromising the cardiovascular system is elevation of ventricular end-diastolic pressure, atrial pressure, and central venous pressure. In the fetus even small increases in venous pressure have been shown to have great effects.7 The younger is the fetus, the higher is its extracellular water content and the lower its tissue pressure. Fluid movement between intravascular and extravascular space is dependent on intra- and extravascular hydrostatic and oncotic pressure and the fluid filtration coefficient, which is determined by the capillary membrane (in the fetus more permeable for fluid and protein). Albumin concentration, largely responsible for oncotic pressure, is lower in the fetus and increases with gestational age. All these factors favor fluid movement out of the capillary into tissue. Thus, lymphatic drainage of tissue seems to be much more important in the fetus. An elevated venous pressure may reduce lymphatic flow, further favoring the development

563

of hydrops. A decrease in arterial blood pressure and an elevation of filling pressures additionally trigger hormonal responses including production of plasma arginine vasopressin (decreases urinary production), angiotensin II (increases fluid accumulation), and atrial natriuretic peptide (increases capillary permeability). Pulmonary edema as part of congestive heart failure does not usually occur in the fetus. The reason for this is that in the presence of a patent foramen ovale, left atrial hypertension does not develop, the pulmonary arterioles are constricted, and there is a fluid-filled lung, where the positive intra-amniotic pressure is transmitted. However, in the case of total anomalous pulmonary venous drainage with stenosis of the draining vessel, or premature constriction or occlusion of the foramen ovale associated with leftsided obstructions, pulmonary venous hypertension can occur with secondary damage to the fetal lungs, such as pulmonary vascular disease and pulmonary lymphangiectasia. Faced with the fetus with hydrops fetalis, one must first determine whether the hydrops is cardiac, inflammatory, or metabolic. Many cases of hydrops are now being attributed to fetal systemic infection. New markers are identifying etiologic agents such as parvovirus or adenovirus. The associated hepatitis with these infections can compromise the protein-producing capability of the fetus, thereby decreasing the fetal oncotic pressure in the vascular space and resulting in fluid loss out of the circulation. Immune hydrops must always be considered in the differential diagnosis, but other causes of anemia can cause hydrops, such as hemoglobinopathies. Infections can cause hemolytic anemia, which can be treated by fetal transfusion. High central venous pressure may exceed the oncotic pressure of the interstitial space, causing fluid to pass into spaces such as the abdominal cavity (ascites), pleural or pericardial spaces (effusions), or any of the vital organs. Multiple mechanisms of hydrops may coexist, and the primary cause may not be immediately obvious. Of more importance is determination of the prognosis of hydrops. This task would be aided by a semiquantitative measure of fetal heart failure. In other words, is this hydrops from heart failure? This chapter will present such an assessment tool. The challenge of hydrops assessment and the diagnosis of heart failure can be summarized as the difficulty in knowing how well the fetal myocardium is performing under changing loading conditions. By combining information from the obstetrical and cardiological evaluations, the perinatal cardiologist can assess whether it is likely that the function abnormality is transient or permanent. The etiology cannot always be known, but the differential is often between infectious, inherited, congenital, or toxinrelated. After birth, the prognosis will depend on the diagnosis and the evolution of the functional abnormality over time. The long-term outcome will be dependent on whether or

564

Fetal Cardiology

not the insult is reversible, and whether there were periods of ischemia and/or brain injury. There are several possibilities for the cause of heart failure in the fetus after ruling out fetal infection (Table 38.1).

Prognosis of fetal heart failure – markers of fetal mortality The cardiovascular system provides a large volume of information about the well-being of the fetus. It is accessible because of rapid developments in the technology of non-invasive techniques, particularly ultrasound. The fetus has become the new patient of the decade, due to the rapid changes in ultrasonic technologies and other fetal assessment techniques. For example, fetal heart rate monitoring by non-focused ultrasound can detect abnormal rate changes, and a lack of normal variability may be related to ischemia. New techniques examine the signal averaged electrocardiogram (ECG) to detect ST–T wave changes.8 The fetal biophysical profile is useful to detect changes in fetal well-being, because it assesses brain function indirectly.9 The decision to deliver a fetus prematurely due to cardiac changes must be made in the context of the risks both pre- and postnatally. Most associations of how cardiovascular changes correlate with other organ function in the fetus have yet to be defined. Therefore, any assessment demands a coordinated team approach between perinatalogists, cardiologists, and neonatologists.10 Table 38.2 lists factors to be considered in assessing fetal cardiac function. The definition of fetal congestive heart failure is similar to after birth – inadequate tissue perfusion. Inadequate cardiac output results in a series of complex reflexes and adaptations to improve forward flow or direct it to vital organs. This state can be described as a deficiency of flow of blood to the tissues such that certain reflexes are triggered for the survival of the fetus. One is the secretion of an excess of circulating catacholamines, which are

Table 38.1 failure

Causes of fetal congestive heart

• Fetal arrhythmias (Figure 38.1) • Anemia • Congenital heart disease with valvular regurgitation (Figure 38.2) • Non-cardiac malformations such as diaphragmatic hernia or cystic hygroma • Twin–twin transfusion recipient volume and pressure overload • Arteriovenous fistula with high cardiac output (Figure 38.3)

Table 38.2 Factors which should be considered in the assessment of fetal heart function • Ventricular shortening less than 0.28 • Valvular regurgitation Tricuspid regurgitation (non-holosystolic but greater than 70 ms duration) Significant tricuspid regurgitation (holosystolic) Mitral regurgitation Pulmonary or aortic valve regurgitation Abnormal atrioventricular valve regurgitation dP/dt normal > 1000 mmHg/s mortality < 400 mmHg/s • Ventricular hypertrophy • Pattern of ventricular filling by pulsed Doppler

produced in response to peripheral vascular detection of abnormal perfusion. Powerful hormonal reflexes are triggered, including those that control salt and water retention, in an attempt to increase myocardial preload, and adrenocorticoid excess, which mobilizes additional calories for the increased metabolic demand that is present. It is now known that the fetus is capable of producing cytokine activation, as well as the secretion of endothelin, troponin T,11 tumor necrosis factor, and brain natriuretic factor (BNP).12 The maturational changes of the systemic vascular bed with gestational age are not known, but at some point in the pregnancy, it is thought that vasoconstriction of the fetal systemic resistance vessels occurs in reponse to stress. How this effects compensation in the fetal circulation is currently being investigated. The most useful predictor of perinatal death in fetal hydrops is the presence of umbilical venous pulsations.13 This is true because the most common pathway of perinatal demise is compromised fetal cardiac output – fetal congestive heart failure. What follows is a method to detect this entity, and to attempt to decide which fetus should be referred to a fetal center. Initially, data are collected during fetal echocardiography: • cardiac size/thoracic size (C/T): cardiac divided by thoracic area ratio (normal 0.25–0.35) or C/T circumference ratio (normal < 0.5) (Figure 38.1) • venous Doppler: inferior caval (or hepatic venous) (increased atrial reversal) and umbilical cord vein (pulsations) (Figure 38.2) • four-valve Doppler: any leak of the valve should be evaluated further. If there are abnormalities in any of these measurements, then a cardiac cause or associated physiological problem may be present, and detailed study is indicated to rule out serious cardiovascular involvement.

Congestive heart failure in the fetus

(a)

565

(b)

Figure 38.1 Heart area to chest area (HA/CA) ratio in a normal fetus showing the area ratio to be less than 0.3 (a). Marked heart enlargement with the HA/CA ratio greater than 0.5 (b).

Figure 38.2 Ductus venosus Doppler with increased atrial contractions (increased pulsatility index) but no atrial reversal. Scale is in meters per second.

Ventricular function in the fetus Most attention has been paid to the right ventricular function in the human fetus, because this ventricle is most likely to show an abnormality during clinical situations associated with increased workload. The right ventricle in the fetus is now understood to be a large contributor to the work output of the fetal myocardium because of the large volume and pressure work required of it. It is observed that the earliest signs of altered function in the fetal heart are a reflection of right heart hemodynamics. For example, isolated right atrial enlargement is a sign of many abnormalities, especially early in gestation. This may be because the right atrium is at the center of the fetal circulation. Flow from the superior vena cava is passing to the tricuspid

valve, directed by the right venous valve inside the right atrium. Flow from left hepatic veins and the ductus venosus is crossing inferiorly across the right atrium to reach the foramen ovale and the left atrium. The normal flow from the right atrium to the left atrium reflects the normal right atrial to left atrial pressure gradient. Any increase in flow to the heart, such as with anemia or arteriovenous fistula, will translate into enlargement of the right atrium. The right ventricle is pumping primarily to the lower body and placenta, and right atrial pressure elevation will result from any increase in the resistance that is seen by this ventricle. For example, placental dysfunction later in gestation will cause right ventricular dysfunction and secondary right atrial enlargement, while the left ventricle will show no signs of dysfunction. This chapter will focus on the right ventricle (RV) and the abnormalities that affect it.

566

Fetal Cardiology

Normal right ventricle growth and function The geometry of the RV in utero is different from that of the left ventricle. The RV is tripartite with inflow, apical, and outflow portions. Calculating the volume of this chamber is difficult because of the complex shape of the chamber. In utero, it is more spherical and more like the LV but still must be measured with non-geometric techniques. After birth the RV atrophies, and the shape becomes more flattened and thinned. RV volume ejected is greater than LV by echocardiography measurements throughout gestation.14 The RV supplies the pulmonary arteries, and the descending aorta and placenta via the ductus arteriosus. The high resistance of the pulmonary arteries is in parallel with the lower resistance of the lower body. The workload of the RV is determined by the volume of blood pumped and the afterload. In general, the work of the RV can be described as the product of the afterload (blood pressure) and the stroke volume of ejection. The afterload of the RV is unique because there are pressure reflections form the downstream resistance. These reflections from the proximal arterioles and, more important, from the proximal ductus arteriosus result in increased power expenditure for a given cardiac output. This is illustrated by the shortened acceleration time in the pulmonary valve compared to the aortic valve. In other words, the RV must perform the work of overcoming inertia, moving the mass of blood forward, and the oscillatory work resulting from the downstream pressure reflections that return to the pulmonary valve shortly after the onset of ejection. When the volume of blood pumped by a ventricle is reduced, the overall size of the ventricle becomes smaller than the contralateral side due to redistribution of flow at the atrial level. This finding of disproportion at the ventricular level is an excellent screening marker of ventricular dysfunction or congenital defect. Rarely does pressure overload alone result in ventricular dilatation. Another indication that the RV is performing more work than the LV is given by data collected in fetal lambs measuring the coronary blood flow of both ventricles. The RV coronary flow was consistently one-third greater than that of the LV.15 RV size limits the volume of blood which can be pumped with each systolic ejection. In situations with reduced RV size with an intrinsically normal RV, the RV volume can be reduced by variables such as pericardial fluid, or diastolic stiffness. A hypoplastic RV chamber may limit ejection volume if a step increase in stroke volume is demanded of that chamber. The heart rate also limits the volume of blood which can be ejected by the right or left ventricle by limiting the filling time of the ventricle. With fetal tachycardia, the cardiac output will drop with increasing heart rate. It is known from clinical experience that the fetus with a heart rate greater that 240 beats per minute can become hydropic in 3–5 days during incessant tachycardia.

Changes with increasing gestational age The inflow pattern of blood velocity can be useful to determine the diastolic characteristics of the RV. The normal pattern is biphasic, with an early filling wave (E-wave) immediately after opening of the tricuspid valve followed by a later atrial contraction wave (A-wave). The atrial contraction augments RV filling and the filling waves can sometimes be used to assess the filling properties of the RV. Tulzer and others analyzed longitudinally the filling patterns of the RV16,17 and showed that the fetal circulation is RV dominant. With increasing gestational age, the early filling wave increases in area and the peak velocities show increasing E- and decreasing A-waves. The right ventricular chamber increases in size, similar to the left during gestation. A normal shortening fraction (diastolic dimension minus systolic dimension divided by diastolic) is greater than 28%. Calculations of the distribution of cardiac output have shown that the right ventricle pumps more than the left ventricle. This distribution is maintained in spite of significant changes in the pulmonary vascular resistance and flow between 20 and 37 weeks’ gestation.18 Foramen ovale flow changes inversely with pulmonary blood flow, thereby maintaining the net RV output greater than the left.17 Because the RV is complex in shape, techniques such as calculation of the RV function by non-geometric techniques using inflow and outflow time intervals are employed. The myocardial performance index, or so-called Tei index, of the right ventricle is not dependent on geometrical assumptions of the RV, and can be used as a parameter to follow changes in RV function over time. The Tei index is the sum of the isovolumic time divided by the ejection time. The isovolemic time is the sum of two intervals: the isovolemic relaxation time, and the isovolemic contraction time. This term can be measured by subtracting the ejection time from the time of inflow of the right ventricle (the time from tricuspid valve opening to closure). Changes in loading of the RV will result in hypertrophy of the chamber walls, and severe thickening can result in diastolic dysfunction (poor relaxation). Description of the RV function requires that the loading conditions of the ventricle be known, and this is one of the most important goals in RV assessment in the human fetus. The pulmonary vascular resistance, for example, has been estimated in the fetus, and appears to decrease between 20 and 30 weeks’ gestation and then increase again until term.

Tricuspid valve The tricuspid valve is intimately associated with the RV function. This valve is normally never regurgitant, and is

Congestive heart failure in the fetus

maintained competent by a complex interaction with the RV endocardium from which it arises early in fetal development. Any acute increase in RV afterload, such as with acute constriction of the ductus arteriosus, will result in a small amount of tricuspid valve regurgitation within two heartbeats.19 Sudden relief of this pressure work causes immediate cessation of the regurgitation. This implies that the shape and micromorphology of the RV chamber are intimately associated with tricuspid valve function and that there are constant instantaneous adjustments going on in the fetal RV to maintain the workload. After examining over 1000 normal fetuses, Respondek et al20 concluded that there is rarely any normal tricuspid valve regurgitation (TR) in the human fetus at any time. Monitoring Tei function in the first trimester, even before the development of the mature tricuspid valvular apparatus, shows the same result – not even a trace of leak. Therefore, this marker of TR is a useful one clinically in detecting the RV that may have dysfunction. The time frame of the TR is also useful in deciding whether the TR could be a sign of disordered myocardial function or is simply a result of normal adjustments to workload. For example, a sudden onset of increased workload does not allow time for the ventricle to adapt, and TR results. The rapidity of the upstroke of the TR jet along with its peak velocity gives information about the RV function. A calculation of the parameter dP/dt (where P is pressure and t is time) can be performed with a good-quality TR jet.21 A value of less than 400 mmHg/second was associated with death in hydropic fetuses. The peak velocity can be used to assess the peak pressure in the RV. Using continuous wave Doppler, the peak velocity is measured, and four times this velocity squared equals the pressure gradient from RV to right atrium (RA) in millimeters of mercury (torr). Therefore, with progressive RV dysfunction, the RV pressure will decrease while the RA pressure during ventricular systole will increase. The difference between the RV and RA then decreases, and is useful in monitoring the interaction of fetal blood pressure and central venous pressure. Systolic function can also be assessed by analyzing the ejection force of the fetal ventricles. This is done at the ouflow valves. The right and left ventricular force development is estimated by Newton’s equation in which force is defined as the product of mass and acceleration and the ejection force can be assessed using the shape of the ejection curve.22,23 Such calculations show that the ejection forces of both ventricles are similar. Altered loading conditions can significantly alter the force of ejection. The most severe forms of RV diastolic dysfunction are manifested by the finding of monophasic inflow velocity. This sign has been associated with compromised prognosis. The ability of the right ventricle to relax is dramatically demonstrated in the rare fetus delivered at term with ductal occlusion. The RV usually stays thick, small, and

567

contracted prior to birth, but immediately after the first breath, the afterload reduces and the ventricle relaxes and fills. Within days the RV may be able to change from a small thick ventricle to one which is supporting the majority of the circulation.

Normal pulmonary valve The blood velocity pattern of RV ejection through the pulmonary valve can be used to learn whether or not the RV afterload is increased. With decreasing RV output, the time–velocity integral decreases while the acceleration time shortens. The pulmonary valve function is also affected in relation to the underlying RV function. Normally, the pulmonary valve in the fetus is perfectly competent. With severe and end-stage RV dysfunction, the valve may exhibit regurgitation, and this finding can have significant prognostic implications.

Normal ductus arteriosus The ductus arteriosus is a site where much can be learned about the left and right heart circulations.24 Flow is normally from the pulmonary artery to the descending aorta in systole and diastole. The peak velocity is between 0.8 and 2.0 m/s and increases with increasing gestational age.

Abnormal right ventricular function Effects of increased afterload Growth restriction RV dysfunction has been recognized more commonly in the most severe cases of intrauterine growth restriction.25 Both systemic and pulmonary vascular resistances are increased and the placental resistance can be dramatically elevated. The marked decreases in oxygenation in this disease may also play a role in the compromise of RV shortening due to the limitation of RV coronary arterial oxygen extraction. Dilatation of the RV and later decreased RV shortening demand prompt attention because they can signal acidosis in the fetus or impending difficulty. These signs of heart failure are best assessed using venous Doppler and other cardiac signs. Tricuspid valve regurgitation in this disease also signals compromised RV function and the need for more detailed study.

568

Fetal Cardiology

Twin–twin circulation The recipient, larger twin in twin–twin transfusion syndrome can experience marked elevations in cardiac output and blood pressure. Typically, the larger twin maintains a high cardiac output due to the volume transfusion plus a state of vasoconstriction due to vasoactive substances produced by the smaller twin. With slow increases in the workload of the RV, there is compensatory hypertrophy and systolic function and minimal signs of hemodynamic compensation. With more rapid onset of volume and pressure overload, the RV stretches and the tricuspid valve begins to leak. The coronary perfusion and subendocardial blood flow are progressively compromised by the hypertrophy and increased early systolic workload. These signs are probably associated with increases in the end-diastolic pressure of the RV. This is reflected in the end-diastolic pressure of the right atrium, which is elevated. These A-contractions against an elevated pressure resistance produce retrograde flow during atrial systole in the hepatic and inferior caval veins. The ductus venosus is the earliest site to see altered flow patterns. With the onset of atrial reversal in this site or umbilical venous pulsations, the onset of metabolic acidosis is imminent. Although many factors influence this outcome, one of the most important is the RV myocardial reserve.

Ductal constriction/occlusion Ductal constriction is defined by the shape of the curve and the estimated mean velocity. Spontaneous or medication-induced constriction of the ductus arteriosus can be detected by the velocity in it, and the severity of ductal constriction can be estimated.24 One of the most common clinical situations in which to observe RV function changes is during fetal ductal changes. Either drug-induced or naturally occurring constriction or occlusion of the ductus arteriosus will affect RV shortening. In this situation, the overall RV cross-sectional area change with systole and diastole decreases dramatically, so that although the RV moves with contraction there is little wall movement and the ejection volume is greatly decreased. The extent of this decrease was shown by Tulzer and others, and the total combined cardiac output may not be normal. One of the most important observations from these data is that decreased shortening may not indicate abnormal contractility of the ventricle. This was shown by calculating dP/dt, the first derivative of the pressure change in the RV using the TR jet, and in most cases of acute ductal constriction, this parameter is maintained normal. Therefore, there is no reported case in the literature at this point of hydrops resulting from ductal occlusion or constriction without concomitant restriction of the foramen ovale. This is a reflection of the redistribution of cardiac output to the left

ventricle. Monitoring of venous Doppler is most useful for assessing the impact of altered RV function in this setting. In most cases, the pregnancy progresses without alteration in growth of the fetus and the RV becomes smaller and thicker.

Increased preload: arteriovenous fistula A shunt from the arterial to the venous circulation in the human fetus will result in a large volume of blood returning to the right atrium. This results in dilated right heart structures and a high cardiac output. The fetal heart is well adapted to handle increasing volume loads if they are imposed gradually. If rapidly increasing severity shunts occur, as is sometimes seen in sacrococcygeal teratoma, then cardiac decompensation and the development of hydrops fetalis can occur quickly. An important sign of RV decompensation in this scenario is valvular regurgitation. Early signs of tricuspid regurgitation, if only very mild, should signal an increase in fetal surveillance. The subsequent development of mitral regurgitation is ominous and is a clear sign of a pre-hydropic state.

Congenital heart disease Pulmonary valve abnormalities The most common right-sided heart abnormality is a bicuspid pulmonary valve. This usually results in trivial narrowing of the right ventricular outflow tract and no significant hemodynamic alteration. Less common is severe pulmonary stenosis with significant pressure increase in the RV. In this lesion, the RV pressure will only increase to suprasystemic levels when it is most severe. At this point, a gradient will be detected by an increased velocity across the pulmonary valve. One can estimate the degree to which the RV pressure is greater than systemic by converting the peak velocity to a gradient using the four times velocity squared equation. This is associated with hypertrophy of the RV and a decrease in the shortening fraction. More severe degrees of RV outflow obstruction occur with pulmonary atresia. In this lesion, the pulmonary valve annulus is smaller than normal and there are usually signs of endocardial change in the RV such as echogenicity. In congenital heart defects with pulmonary stenosis or atresia and a ventricular septal defect such as tetralogy of Fallot, there are no RV changes and the RV size is maintained normal and equal to the LV because the pressures are equal. There is an uncommon form of pulmonary atresia with intact ventricular septum where infarction of the RV occurs and there is tricuspid valve regurgitation which is severe. Then, the RV

Congestive heart failure in the fetus

is larger than normal and the RA may be massively enlarged. In the rare tetralogy of Fallot with absent pulmonary valve syndrome there is no ductus arteriosus and the pulmonary arteries are enlarged. Pulmonary valve regurgitation is moderate or severe because there is no pulmonary valve and it is replaced by a fibrous ring. Both the RV and left ventricle experience the effects of the pulmonary regurgitant volume and are usually seen to be normal. Rarely, there is dysfunction of the RV, which is indicated by disproportion of the ventricular sizes with RV enlargement and segmental RV wall motion abnormalities.

Tricuspid valve abnormalities Tricuspid valve congenital abnormalities which result in significant amounts of tricuspid valve regurgitation are often fatal. The most common defect noted in fetal studies in the last decade is known as tricuspid valve dysplasia. In this lesion, the RV and RA are massively dilated and color Doppler shows that regurgitation is the cause. The key issue in this lesion is the nature of the RV outflow tract. If there is no pulmonary stenosis then the lesion may have occurred later in gestation and the prognosis is better. The degree of RV cardiomyopathy can be assessed by measurement of the peak velocity of the TR jet. The lower is the velocity, the worse are the RV dysfunction and the prognosis. Ebstein’s malformation of the tricuspid valve with various severities of pulmonary stenosis or atresia in the fetus is the ‘cancer’ of congenital heart disease. The heart becomes enlarged to occupy more than 60–70% of the chest, and this is due to RA and atrialized RV enlargement. The pulmonary development can be altered irreparably by compression from the heart if it is severe, is associated with pleural effusions, and occurs in the critical time frame of 20–30 weeks when the lungs are developing. Peak TR velocities of less than 2 m/s are associated with a poor prognosis due to the development of RV myopathy. Primary cardiomyopathy of the right ventricle such as Uhl’s anomaly or arrhythmogenic right ventricle have not been reported in the fetus. However, any intractable ventricular arrhythmia in the fetus should prompt consideration of one of these diagnoses.

Cardiovascular profile score development The diagnosis of fetal congestive heart failure, therefore, must be addressed in a clinical fashion similar to that after birth. The classical clinical tetrad of cardiomegaly, tachycardia, tachypnea, and hepatomegaly has been used in neonates and children. This clinical state in the fetus can be characterized by findings in at least five categories,

569

which are obtained during the ultrasonographic examination. The following five categories are each worth 2 points in a 10-point scoring system to assess the cardiovascular system. Abnormalities in the Cardiovascular Profile Score may occur prior to the clinical state of hydrops fetalis. The five categories are: • • • • •

hydrops umbilical venous Doppler heart size abnormal myocardial function arterial Doppler.

Within specific disease entities, more emphasis is placed on certain areas by the attending physician to predict the prognosis. As always, this information can only constitute a portion of the total picture and must be integrated by the attending physician into the diagnostic and treatment plan for the patient. The Cardiovascular Profile Score gives a semiquantitative score of the fetal cardiac well-being, and uses known markers from ultrasound which have been correlated with poor fetal outcome. This profile is normal if the score is 10 and signs of cardiac abnormalities result in a decrease of the score from normal. For example, if there is hydrops with ascites and no other abnormalities, there would be a deduction of 1 point for hydrops (ascites but no skin edema) and no deductions for the other categories, for a score of 9 out of 10.

Hydrops Hydrops fetalis may present with ascites, pleural effusion, pericardial effusion, or a combination of these findings. In advanced hydrops, there is generalized skin edema seen easily over the scalp and abdominal wall. In scoring hydrops for the Cardiovascular Profile Score, 1 point is deducted for early hydrops and 2 points for skin edema.

Umbilical and ductus venosus Doppler Fetal venous blood velocities have been examined and investigations have been clinically promising.26 Several studies have confirmed that normal flow in the inferior vena cava of the fetus has a pulsatile, triphasic pattern. The first forward wave begins to increase with atrial relaxation, reaches its peak during ventricular systole, and then falls to reach its nadir at the end of ventricular systole. The second forward wave occurs during early diastole, and a reverse flow is usually present in late diastole with atrial contraction. In normal pregnancy, the peak velocities obtained during the first wave of systole are greater than the early diastolic values. The systolic-to-diastolic ratios do not

570

Fetal Cardiology

appear to change with advancing gestational age, but a significant decrease in flow reversal with atrial contraction is evident. Studies in the fetal lamb have shown that this decrease in percentage of reversed flow seen in normal pregnancy is related to the pressure gradient between the right atrium and the right ventricle during end-diastole. It appears to be related to both ventricular compliance and ventricular end-diastolic pressure, and therefore is a reflection of central venous pressure. Recording venous blood velocity might thus give important information on fetal cardiac pump function. Previous studies in humans have shown that alterations in central venous blood velocity patterns accurately reflect abnormalities in cardiac hemodynamics. The abnormal pulsatility pattern consists of increased velocity of blood flow reversal away from the heart during atrial contraction, and has been reported in the fetus with congestive heart failure and may be a sign of increased end-diastolic pressure in the ventricles of the failing heart. Abnormal IVC flow velocity patterns have been described in several fetal pathologic conditions including anemia, non-immune hydrops, and arrhythmias, and in severely growth-restricted fetuses characterized by the absence of end-diastolic flow in the umbilical artery. The compromised fetus with acidosis is known to manifest abnormalities of venous Doppler, including increased atrial reversal in excess of normal in the inferior vena cava at the junction with the right atrium27 and increased pulsatility in the ductus venosus. The prognostic importance of these abnormalities has been confirmed in the fetus with intrauterine growth restriction and in those with hydrops. An increased A/S ratio in the ductus venosus (peak atrial reversal divided by the peak filling wave during ventricular systole) appears to be the most useful sign in quantifying the increase of atria contraction in fetuses with growth restriction. Normally, the ratio of the area of the atrial reversal to the entire forward flow area should be less than 7%. Transmission of the venous pulsations into the portal and umbilical circulation correlates with increasing degrees of cardiac compromise. Tulzer et al studied the cardiac factors related to prognosis in hydrops and noted that umbilical venous pulsations could stand in for a number of cardiac variables in predicting prognosis, including ventricular shortening fraction, ejection velocities, and percentage IVC atrial reversal.28 In some disease states, abnormal venous Doppler progresses retrograde from the heart in the following order: (1) increased atrial reversal in the IVC, (2) ductus venosus atrial reversal, (3) portal venous atrial pulsations, (4) umbilical venous atrial pulsations. The end-stage finding of abnormal venous Doppler is atrial pulsations in the umbilical cord vein. This finding of so-called ‘diastolic block’ predicts perinatal mortality. Double umbilical venous pulsations, or the pattern of the normal IVC in the umbilical vein, is an ominous clinical finding.28 Venous pulsations are not normal in the portal

vein, and such a finding may precede the progression to umbilical venous pulsations.

Venous Doppler in congenital heart disease Pagatto et al, working with us, studied a group of 41 fetuses diagnosed with congenital cardiac defects in utero and confirmed postnatally. The gestational age ranged from 18 to 38 weeks with a mean gestational age of 27.5 weeks. The fetuses were grouped into those with ventricular septal defects (n = 11), tricuspid atresia or hypoplasia (n = 4), hypoplastic left heart syndrome (n = 19), and others(n = 7), and we analyzed the venous Doppler patterns. Abnormal inferior vena caval waveforms were present only in fetuses with tricuspid atresia or other right heart lesions where the returning flow to the heart all passed through the foramen ovale to reach the heart. Umbilical waveforms were non-pulsatile in all patients. We concluded that venous Doppler in fetuses with congenital heart disease without heart failure is normal, with the exception of tricuspid atresia and pulmonary atresia with intact septum, such that the combined cardiac output must pass through the foramen ovale. When abnormal central venous flow patterns occurred in association with cardiac defects in utero, they were usually secondary to another process occurring simultaneously which affected ventricular compliance (such as endocardial fibroelastosis) or rhythm-related hemodynamics (such as complete heart block). To assess the venous system consistently, pulsed Doppler sampling is obtained in the inferior vena cava, the ductus venosus, the umbilical vein in the abdomen, and the umbilical cord vein as part of each serial examination. Transmission of the atrial reversal into the ductus venosus and later into the portal and cord vein sites over time suggests progression of heart failure. The Cardiovascular Profile Score has deductions for abnormal venous Doppler as shown below: • ductus venosus atrial reversal, −1 point • umbilical venous atrial pulsations, −2 points. Maximum deduction in any category is 2 points.

Cardiomegaly Enlargement of the cardiac chambers is a universal sign of heart failure. This is true in the fetus as well, but few of the mechanisms are understood. It is likely that neural humoral reflexes are triggered, resulting in retention of extracellular volume leading to increased end-diastolic volume of the ventricles. At some point, this increased ventricular size indicates increased end-diastolic pressure. However, unlike the postnatal human it is uncommon to encounter

Congestive heart failure in the fetus

persistent tachycardia with signs of catacholamine excess. It is possible that the levels of humoral agents are modified by the fetal–maternal exchange mechanisms, which exist when the placenta is functioning normally. The most common cardiac chamber to express enlargement as a sign of impending cardiac failure is the right atrium. The reasons for this relate to the many causes for heart failure, but the right atrium is a final pathway for blood flow returning to the heart and will manifest enlargement in situations of relative foramen obstruction, volume overload, tricuspid valve regurgitation, and increased afterload. Increased RA size may be due to increased RV enddiastolic pressure, which may be due to increased afterload or coronary insufficiency. The right ventricle may be more susceptible to increased work because of the nature of the afterload and the resultant increased demands for oxygen in the face of increased chamber wall stress. It is generally believed that increased atrial wall stress without increased ventricular work does not lead to clinical difficulties in the fetus. Such a situation could be an early marker of cardiac decompensation and may predispose to supraventricular arrhythmias. Secretion of atrial natriuretic peptide (ANP) could be a marker of this finding. Slower than normal heart rate or persistent rapid heart rate leads to cardiomegaly. The time frame of the onset of arrhythmia may therefore be estimated by the effect on the cardiac size. For example, an intermittent arrhythmia that has appeared recently would not be expected to cause cardiac enlargement. Small heart size with external compression has been correlated with hydrops and poor outcome in fetuses with cystic adenomatoid malformation.29 When heart size was less than 20% of the chest area, fetal outcome was affected. Cardiomegaly is a heart to chest area ratio greater than 0.35 at any time in gestation. Cardiac size calculations are: • C/T area ratio = cardiac area/chest area (normal 0.2–0.35) • C/T circumference ratio = cardiac circumference/chest circumference (normal < 0.5)

571

Both of the right and left ventricles should shorten their diameters more than 28% in systole compared to diastole. Measurements of cardiac dimensions with time are performed using M-mode echocardiography. The shortening fraction of a ventricle is calculated by taking the difference between the diastolic (DD) and systolic dimensions (SD) and dividing by the diastolic dimension. Fractional shortening = (DD − SD)/DD, normal > 0.28 An abnormal shortening fraction could reflect myocardial compromise or an increase in the fetal ventricular workload. Regardless, an increase in diastolic dimension is often related to a decrease in shortening fraction and should be regarded as an indication for more intensive monitoring. The atrioventricular and semilunar valves are competent in the normal fetus, and if regurgitation is detected it is usually a sign of altered cardiovascular physiology. Respondek et al showed that 7% of fetuses having a fetal echocardiogram display trivial or significant tricuspid valve regurgitation.20 Most had some reason for this, such as constriction of the ductus arteriosus from indomethacin treatment of preterm labor, but 93% had no trace of regurgitation with state-of-the-art equipment and careful examination. Since tricuspid valve regurgitation is common after birth, we may speculate that the fetal right ventricle is well adapted to systemic pressure work. Therefore, valvular competence is normal, and only in disturbances of cardiovascular physiology where there is increased ventricular wall stress is tricuspid valve regurgitation present. Trace tricuspid regurgitation defined as non-holosystolic regurgitation lasting at least 70 ms is not normal. This may be the first sign of a problem, but has little prognostic importance. Holosystolic tricuspid regurgitation is abnormal and always indicates the need for further investigation.19 When regurgitation is detected by color Doppler, it must be confirmed and graded by pulsed Doppler. With congenital diseases of the tricuspid valve, hydrops and fetal death can occur (Figure 38.3).

therefore: • • • •

normal heart/chest: area ratio ≤ 0.35 and > 0.20 mild cardiomegaly: area ratio > 0.35 and ≤ 0.5, −1 point severe cardiomegaly: area ratio > 0.50, −2 points small heart: ratio > 0.2, –2 points

Maximum deduction is 2 points.

Abnormal myocardial function The cardiac function is assessed indirectly by the global shortening (and thickening) of the walls of the ventricles, and by the function of the atrioventricular and semilunar valves.

Figure 38.3 Severe tricuspid valve (TV) regurgitation on color Doppler with fetal Ebstein’s malformation of the TV.

572

Fetal Cardiology

Figure 38.5 Severe biventricular hypertrophy in the larger recipient twin in twin–twin transfusion syndrome.

Figure 38.4 Tricuspid (TR) and mitral (MR) valve regurgitation on color Doppler after fetal intervention for fetal aortic stenosis. LV, left ventricle; RV, right ventricle.

Regurgitation of other valves is usually a sign of more advanced congestive heart failure, and may occur in the moribund fetus with acidosis and severe heart failure as a sign of myocardial compromise. Tricuspid valve regurgitation can be a reversible sign of heart failure, as one observes in fetuses with successful in utero therapy for anemia or tachycardia. Progression to mitral valve regurgitation is always a sign of fetal congestive heart failure and usually means that a significant increase in left ventricular wall stress is present (Figure 38.4). With severe myocardial failure, the support for the semilunar valves is compromised, and pulmonary or aortic valve regurgitation can occur. The nature of the velocity waveform of valvular regurgitation has prognostic value in the calculation of fetal ventricular dP/dt. With holosystolic tricuspid valve regurgitation, the time interval from one RV–RA pressure difference to another can be used to calculate the change in pressure over time, or the dP/dt. A value of less than 800 mmHg/s is abnormal, and a value less than 400 predicts a poor fetal outcome.20 This measurement requires continuous wave Doppler, and the peak velocity may be from 2.5 to 4.5 m/s. We have found that the most useful range for dP/dt measurement in fetal TR is 0.5–2.5 m/s. In other words, an RV–RA gradient of 1 mmHg to 25 mmHg or a 24-mmHg difference. The fetal ventricles are at equal and systemic pressure throughout gestation, and therefore the blood pressure of the fetus is being estimated by this technique. The filling pattern of the ventricles in diastole is an indicator of the diastolic function of the heart. Normal values show that the proportion of atrial filling during the atrial contraction is constant from 14 to 40 weeks’ gestation.15

Monophasic filling of the ventricles is a sign of compromised diastolic function and is a sign of fetal heart failure.29 Several disease states are now being identified where thickening of the ventricular chambers (myocardial hypertrophy) occurs in the absence of congenital ventricular outflow obstruction. This is assessed by measuring the end-diastolic wall thickness of the left ventricle and comparing it with the normal values for age. Any left ventricle posterior wall (LVPW) thickness greater than or equal to 4 mm is abnormal. The most severe cases of fetal hypertension that have been detected are in the larger of twins in the syndrome known as twin–twin transfusion, where a mortality of over 70% for the fetuses is common (Figure 38.5). It appears that early identification of hypertrophy in the larger twin can be useful in patient management. Treatment strategies to prevent hydrops in the larger fetus result in improved survival. Postnatally, neonatal hypertension can be severe and life-threatening. In utero interventions are currently being explored, and appear to be indicated in selected fetuses for this cause of congestive heart failure, including serial amniocentesis and laser ablation of the vascular communications.30 Regardless of the etiology, thickening of the fetal ventricles could restrict the cardiac reserve before or after birth. Because cardiac hypertrophy can occur rapidly but takes weeks or months to resolve, its identification is an important marker of a cardiovascular system at risk. Abnormalities of diastolic function could be expected, and should be excluded by comparing the filling patterns of the ventricles using pulsed Doppler to standardized normal values. One rule of thumb is that the A-wave of the ventricular filling is always greater than the E-wave, and, if it is higher or is indistinguishable, then a detailed cardiac study should be performed. Monophasic filling of the ventricles occurs in severe diastolic dysfunction and with external cardiac compression which is severe. Studies using the myocardial performance index may be

Congestive heart failure in the fetus

useful to reflect an abnormality of systolic and diastolic function, and deserve more research.31 For cardiac function: • • • • • • •

RV/LV shortening fraction < 0.28, − 1 point tricuspid valve regurgitation (holosystolic), − 1 point mitral regurgitation, − 1 point monophasic ventricular filling, − 2 points pulmonary or aortic valve regurgitation, −1 point valve regurgitation dP/dt < 400 mmHg/s, −2 points ventricular hypertrophy, −1 point. Maximum deduction is 2 points.

Arterial Doppler: redistribution of fetal cardiac output It is now well established that the blood velocities measured by Doppler echocardiography in the umbilical artery and in other peripheral vascular beds can be used as an indirect indicator of the relative vascular impedances. Findings of an increased pulsatility index in the umbilical artery (UA) and descending aorta (DAo) and a decreased index in the middle cerebral artery (MCA) are non-invasive signs of redistribution of flow. It is important to recognize that a pulsed Doppler finding in one portion of the circulation is affected by changes in the rest of the circulation. For example, if there is significant aortic valvular regurgitation in the fetus, the diastolic reversal in the descending aorta and the increased pulsatility index in the umbilical artery are secondary to this change in the heart and do not reflect peripheral resistance only. The most common cause of elevated vascular resistance in the fetus is placental dysfunction secondary to vasculopathy leading to asymmetrical growth restriction. This complex pathophysiololgical state is poorly understood, but there is evidence that there is hypoxemia resulting from placental dysfunction and additional compromise of nutrition severe enough to impair growth. Once the normal pattern of growth is disturbed (usually asymmetrical such that the brain continues growing but the body does not), the fetus is at risk of organ damage from hypoxemic/ ischemic injury. The umbilical artery manifests this problem with a loss or reversal of diastolic blood flow (Figure 38.9). There is a redistribution of flow to the brain (so-called brainsparing) due to reflex vasodilatation of the cerebral vessels. This is manifested by a decrease in the pulsatility index in the middle cerebral artery such that diastolic flow is relatively increased (PI less than 2 SD below the mean).32 In the fetus with hypoxemia, the peripheral fetal vessels are vasoconstricted, and the larger arteries are suspected to be non-compliant compared to normals with increased blood pressure.33 In other words, this is a physiological state characterized by increased vascular resistances and, at end-stage, decreased cardiac output. Right ventricular enlargement occurs in some cases.

573

There is evidence that reversal of diastolic flow in the umbilical artery, if confirmed, may be a significant risk factor for abnormal outcomes. Further work is needed in this area. As a sign of fetal heart failure, the vasoconstriction resulting from decreased cardiac output and the compensatory sign of vasodilatation in the brain can be included in the cardiovascular profile score shown below: • absent end-diastolic flow in the umbilical artery + brainsparing (increased MCA diastolic velocity), − 1 point • reversed end-diastolic flow in the umbilical artery, − 2 points.

Cardiovascular profile score The Cardiovascular Profile Score comprises 2 points in each of the five categories used in serial studies to provide a method of uniform physiological assessment. By taking a multivariate approach, this type of multifactorial score can combine assessment of direct and indirect markers of cardiovascular function. Initial validation of the Cardiovascular Profile Score in hydrops was shown by Falkensammer et al.34 Seven fetuses with hydrops, including three with congenital heart disease, had correlation of the Cardiovascular Profile Score with the myocardial performance index (Tei index). RV and LV Tei indexes were assessed in normals and showed no change with gestational age. Hofstaetter et al35,36 measured the Cardiovascular Profile Score in 59 hydrops fetuses. Mortality was 21/59. The average score in those who died pre- or postnatally was 5. The average score in the survivors was 6.31 The score was evaluated in fetuses with congenital heart disease37 and in fetal growth restriction38 in recent studies.

Treatment of fetal heart failure Treatment of fetal cardiovascular problems can be classified into five of the most common subgroups based on the etiology of congestive heart failure: (1) abnormal peripheral impedances causing redistribution of flow and growth failure, (2) high output due to anemia or arteriovenous fistula, (3) primary or secondary valvular regurgitation, (4) heart failure due to myocardial dysfunction, and (5) tachycardia/ bradycardia. Interventions aimed at improving the effective cardiac output are also aimed at prolonging the pregnancy and preventing prematurity and prenatal asphyxia. The rapidity with which a disease progresses determines the urgency with which treatment should occur. This is due to the fact that the myocardial response to increased wall stress will be either adequate or inadequate depending on the severity and timing and duration of the insult, the coronary perfusion, the nutritional state of the fetus, and the other problems in the pregnancy.

574

Fetal Cardiology

The usual treatment of placental dysfunction is designed to improve the vascular impedance of the placenta and to increase the flow of oxygenated blood to the fetus. With bedrest, improved nutrition, or maternal oxygen there may be improvement in placental function. Tocolytic medications may relax the placenta and improve its function. Myocardial support for advanced growth restriction has not been proposed, partly because the validation of diagnostic methods is lacking. Studies of ventricular ejection force in growth restriction have shown that both ventricles have decreased ejection force.17 Advanced heart failure in this setting with severely decreased arterial PaO2 and poor nutrition are manifested by non-specific signs of increased RV and RA size, atrial reversal in the venous Doppler pattern, and altered forward flow velocities. Treatment with digoxin for evidence of decreased ventricular shortening is controversial. Digoxin is known to decrease the catecholamine response to congestive heart failure, and if there is diastolic dysfunction in the fetus, then this may improve filling and lower filling pressures. If the afterload is high, then an increase in oxygen consumption could result from increased inotropy without improved myocardial perfusion. Terbutaline appears to have promise as an inotropic and chronotropic agent,39 but studies of the possible negative effects on the fetal myocardium are needed. At the present time, we use digoxin for fetal cardiac failure due to arrhythmias and high output states such as fistula and anemia. In a recent case of acardiac twinning where the normal fetus was supporting two circulations, digoxin appeared to improve cardiac function and result in a prolonged and successful gestation for the normal twin. In our experience we consider the use of digoxin for the fetus in sinus rhythm with signs of congestive heart failure with a cardiovascular profile score of 7 or less. We use Lanoxin® (Lanoxicaps®) 0.2 mg orally two or three times per day based on maternal serum levels and clinical signs. We use a trough level of 1.0–2.0 mg to avoid any maternal side-effects. Laser treatment of the twin–twin communications or cord ligation with acardiac twins can be applied to improve cardiac failure. With anemia, it is possible to transfuse the fetus via the umbilical vein. The diagnosis of fetal anemia can be made using the middle cerebral artery peak velocity. With anemia, the cardiac output is increased with a reduced oxygencarrying capacity. When fetal valvular regurgitation is present on a congenital basis, it could be useful to decrease the afterload of the fetal ventricles as is done in infants with a similar problem. However, medications which reduce the afterload such as angiotensin converting enzyme (ACE) inhibitors are known to be dangerous to the fetus in pregnancy. Reduction of catacholamine levels could have a similar effect, and digoxin could be useful in this situation. In pregnancies where the mother has significant levels of anti-Ro and anti-La antibodies, we recommend dexamethasone 4 mg daily if there are signs of valvular regurgitation,

heart block, valvulitis, myocardial dysfunction, myocardial echogenicity, or effusion. Early use of this medication may prevent progression of heart block and myocardial injury. When myocardial dysfunction is seen without obvious reasons, and fetal infection has been excluded, we consider that an inherited form of cardiomyopathy of either the left or right ventricle can present in utero. We use digoxin for these patients as long as there is no sign of ventricular ectopy or tachycardia.

Fetal cardiac interventions Several postnatal forms of congenital heart disease may lead to fetal congestive heart failure or irreversible secondary damage to the fetal heart and lungs, for example right heart obstructive lesions such as tricuspid atresia, critical pulmonary stenosis, or pulmonary atresia with intact ventricular septum with restrictive interatrial shunting, severe atrioventricular valve insufficiency, or premature constriction or occlusion of the foramen ovale associated with left heart obstructions. It was speculated that correcting or improving these anatomical problems in utero could prevent heart failure and/or fatal secondary damage. The first intracardiac fetal interventions were reported in 1991.40 Two fetuses with critical valvular aortic stenosis underwent an in utero attempt at percutaneous valvuloplasty with the goal to prevent irreversible myocardial damage of the left ventricle. In one of these fetuses the aortic valve could be dilated successfully, but it died postnatally due to persistent left ventricular dysfunction. Due to enormous technical difficulties, poor results (out of 12 human fetuses only one remained alive) and with increasingly better results for stage 1 Norwood palliation, this method was abandoned for several years. More recently this method has been successfully used again for fetuses with pulmonary atresia with intact septum and signs of heart failure. In these fetuses, relief of right ventricular outflow obstruction led to improvement of echocardiographic signs of heart failure. However, as most fetuses with this anatomy do not develop heart failure in utero unless there is restrictive interatrial shunting or severe tricuspid regurgitation, this method should be limited only to a small subset. Whether prenatal decompression of small hypertensive right ventricles is able to establish growth, to prevent coronary artery fistulae, and improve the chances of a biventricular repair postnatally has yet to be determined. It also remains unclear whether early dilatation of critical stenotic aortic valves can prevent the development of hypoplastic left heart syndrome. There is evidence that an intact atrial septum in hypoplastic left heart syndrome leads to morphologic changes of the pulmonary vasculature and pulmonary lymphangiectasia.41 Creating an atrial communication and decompressing the left atrium should enable normal pulmonary

Congestive heart failure in the fetus

venous drainage and prevent further damage to pulmonary vessels and parenchyma. This could become an additional goal for cardiac interventions. To date, many technical problems regarding equipment, imaging, and access have not yet been solved. Clinical indications for therapy have yet to be defined, and there is still a need for more natural history studies regarding in utero progression of congenital heart disease and/or congestive heart failure. Therefore, these treatment options should still be considered innovative.

Conclusions Fetal cardiac findings must be integrated into the clinical management of the fetus by the perinatologist. The Cardiovascular Profile Score can be used to communicate between visits and specialists to assess the urgency of abnormalities and the prognosis. Focused centers of excellence in fetal cardiac assessment are needed to investigate and achieve effective fetal treatment. However, fetal diagnosis is only a dream unless providers of ultrasound screening detect abnormalities during otherwise normal gestations. Therefore, each center of excellence in perinatal cardiology must accept the responsibility of education in the surrounding region. Good communication between heart screening sites and perinatal cardiology centers will benefit all involved, and will allow progress to occur in this new field of cardiology.

References 1. Huhta JC. Fetal congestive heart failure. Semin Fetal Neonatal Med 2005; 10: 542–52. 2. Kiserud T. Physiology of the fetal circulation. Semin Fetal Neonatal Med 2005; 10: 493–503. 3. Angelini A, Allan LD, Anderson RH et al. Measurements of the dimensions of the aortic and pulmonary pathways in the human fetus: a correlative echocardiographic and morphometric study. Br Heart J 1988; 60: 221–6. 4. Rudolph AM, Heymann MA. Cardiac output in the fetal lamb: the effects of spontaneous and induced changes of heart rate on right and left ventricular output. Am J Obstet Gynecol 1976; 124: 183–92. 5. Friedman WF. The intrinsic properties of the developing heart. Prog Cardiovasc Dis 1972; 15: 87–111. 6. Thornburg KL, Morton MJ. Filling and arterial pressures as determinants of RV stroke volume in the sheep fetus. Am J Physiol 1983; 244: H656–63. 7. Johnson P, Sharland G, Allan LD, Tynan MJ, Maxwell DJ. Umbilical venous pressure in nonimmune hydrops fetalis: correlation with cardiac size. Am J Obstet Gynecol 1992; 167: 1309–13. 8. Westgate J, Harris M, Curnow JS, Greene KR. Randomised trial of cardiotocography alone or with ST waveform analysis for intrapartum monitoring. Lancet 1992; 340: 194–8.

575

9. Manning FA, Morrison I, Harman CR, Menticoglu SM. The abnormal fetal biophysical profile score. V. Predictive accuracy according to score composition. Am J Obstet Gynecol 1990; 162: 918–24; discussion 924–7. 10. Huhta JC. What is a perinatal cardiologist? Ultrasound Obstet Gynecol 1995; 5: 145–7. 11. Makikallio K, Vuolteenaho O, Jouppila P, Rasanen J. Association of severe placental insufficiency and systemic venous pressure rise in the fetus with increased neonatal cardiac troponin T levels. Am J Obstet Gynecol 2000; 183: 726–31. 12. Troughton RW, Frampton CM, Yandle TG et al. Treatment of heart failure guided by plasma aminoterminal brain natriuretic peptide (N-BNP) concentrations. Lancet 2000; 355: 1126–30. 13. Gudmundsson S, Huhta JC, Wood DC et al. Venous Doppler ultrasonography in the fetus with nonimmune hydrops. Am J Obstet Gynecol 1991; 164: 33–7. 14. Chaoui R, Heling KS, Taddei F, Bollmann R. [Doppler echocardiographic analysis of blood flow through the fetal aorta and pulmonary valve in the second half of pregnancy]. Geburtshilfe Frauenheilkd 1995; 55: 207–17. [in German] 15. Thornburg KL, Reller MD. Coronary flow regulation in the fetal sheep. Am J Physiol 1999; 277: R1249–60. 16. Tulzer G, Khowsathit P, Gudmundsson S et al. Diastolic function of the fetal heart during second and third trimester: a prospective longitudinal Doppler-echocardiographic study. Eur J Pediatr 1994; 153: 151–4. 17. Veille JC, Smith N, Zaccaro D. Ventricular filling patterns of the right and left ventricles in normally grown fetuses: a longitudinal follow-up study from early intrauterine life to age 1 year. Am J Obstet Gynecol 1999; 180: 849–58. 18. Rasanen J, Wood DC, Weiner S, Ludomirski A, Huhta JC. Role of the pulmonary circulation in the distribution of human fetal cardiac output during the second half of pregnancy. Circulation 1996; 94: 1068–73. 19. Tulzer G, Gudmundsson S, Rotondo KM et al. Acute fetal ductal occlusion in lambs. Am J Obstet Gynecol 1991; 165: 775–8. 20. Respondek ML, Kammermeier M, Ludomirsky A , Weil SR, Huhta JC. The prevalence and clinical significance of fetal tricuspid valve regurgitation with normal heart anatomy. Am J Obstet Gynecol 1994; 171: 1265–70. 21. Tulzer G, Gudmundsson S, Rotondo KM et al. Doppler in the evaluation and prognosis of fetuses with tricuspid regurgitation. J Matern Fetal Invest 1991; 1: 15–18. 22. Sutton MS, Gill T, Plappert T, Saltzman DH, Doubilet P. Assessment of right and left ventricular function in terms of force development with gestational age in the normal human fetus. Br Heart J 1991; 66: 285–9. 23. Rasanen J, Debbs RH, Wood DC et al. Human fetal right ventricular ejection force under abnormal loading conditions during the second half of pregnancy. Ultrasound Obstet Gynecol 1997; 10: 325–32. 24. Tulzer G, Gudmundsson S, Sharkey AM et al. Doppler echocardiography of fetal ductus arteriosus constriction versus increased right ventricular output. J Am Coll Cardiol 1991; 18: 532–6. 25. Rizzo G, Capponi A, Rinaldo D, Arduini D Romanini C. Ventricular ejection force in growth-retarded fetuses. Ultrasound Obstet Gynecol 1995; 5: 247–55.

576

Fetal Cardiology

26. Hecher K, Snijders R, Campbell S, Nicolaides K. Fetal venous, intracardiac, and arterial blood flow measurements in intrauterine growth retardation: relationship with fetal blood gases. Am J Obstet Gynecol 1995; 173: 10–15. 27. Rizzo G, Arduini D, Romanini C. Inferior vena cava flow velocity waveforms in appropriate- and small-for-gestational-age fetuses. Am J Obstet Gynecol 1992; 166: 1271–80. 28. Tulzer G, Gudmundsson S, Wood DC et al. Doppler in non-immune hydrops fetalis. Ultrasound Obstet Gynecol 1994; 4: 279–83. 29. Mahle WT, Rychik J, Tian ZY et al. Echocardiographic evaluation of the fetus with congenital cystic adenomatoid malformation. Ultrasound Obstet Gynecol 2000; 16: 620–4. 30. Quintero RA, Comas C, Bornick PW, Allen MH, Kruger M. Selective versus non-selective laser photocoagulation of placental vessels in twin-to-twin transfusion syndrome. Ultrasound Obstet Gynecol 2000; 16: 230–6. 31. Eidem BW, Edwards JM, Cetta F. Quantitative assessment of fetal ventricular function: establishing normal values of the myocardial performance index in the fetus. Echocardiography 2001; 18: 9–13. 32. Rizzo G, Arduini D. Fetal cardiac function in intrauterine growth retardation. Am J Obstet Gynecol 1991; 165: 876–82. 33. Stale H, Marsal K, Genner G et al. Aortic diameter pulse waves and blood flow velcocity in the small, for gestational age fetus. Ultrasound Med Biol 1991; 175: 471–8.

34. Falkensammer CB, Paul J, Huhta JC. Fetal congestive heart failure: correlation of Tei-index and Cardiovascular-score. J Perinat Med 2001; 29: 390–8. 35. Hofstaetter C, Huhta JC. Outcome assessment in hydrops fetalis using a cardiovascular score. Abstract presented at Society of Pediatric Research, Baltimore, MD, 2002. 36. Hofstaetter C, Hansmann M, Eik-Nes SH, Huhta JC, Luther SL. A cardiovascular profile score in the surveillance of fetal hydrops. J Matern Fetal Neonatal Med 2006; 19: 407–13. 37. Wieczorek A, Hernandez-Robles J, Ewing L et al. Prediction of outcome of fetal congenital heart disease using a cardiovascular profile score. Int J Ultrasound Obstet Gynecol 2008; 31: 284–8. 38. Mäkikallio K, Rasanen J, Mäkikallio T et al. Human fetal cardiovascular profile score and neonatal outcome in fetal growth restriction. J Ultrasound Obstet Gynecol 2008; 31: 48–54. 39. Sharif DS, Huhta JC, Moise KJ, Morrow RW, Yoon GY. Changes in fetal hemodynamics with terbutaline treatment and premature labor. J Clin Ultrasound 1990; 18: 85–9. 40. Maxwell D, Allan L, Tynan MJ. Balloon dilatation of the aortic valve in the fetus: a report of two cases. Br Heart J 1991; 65: 256–8. 41. Rychik J, Rome JJ, Collins MH, DeCampli WM, Spray TL. The hypoplastic left heart syndrome with intact atrial septum: atrial morphology, pulmonary vascular histopathology and outcome. J Am Coll Cardiol 1999; 34: 554–60.

Congestive heart failure in the fetus

577

Appendix Cardiovascular Profile Score Hydrops

Normal

–1 point

–2 points

None (2 points)

Ascites or pleural effusion or pericardial effusion

Skin edema

UV

UV

UV pulsations

Venous Doppler (umbilical vein) (ductus venosus)

DV (2 points)

DV

Heart size (Heart area/ chest area)

>0.20 and ≤ 0.35 (2 points)

0.35–0.50

>0.50 or <0.20

Cardiac function

Normal TV and MV RV/LV SF > 0.28 Biphasic diastolic filling (2 points)

Holosystolic TR or RV/LV SF < 0.28

Holosystolic MR or TR dP/dt < 400 or monophasic filling

Arterial Doppler (umbilical artery)

UA (2 points)

UA (AEDV)

UA (REDV)

The heart failure score is 10 if there are no abnormal signs and reflects 2 points for each of five categories: hydrops, venous Doppler, heart size, cardiac function, and arterial Doppler. AEDV, absent end-diastolic velocity; dp/dt, change in pressure over time of TR jet; DV, ductus venosus; LV, left ventricle, MR, mitral valve regurgitation; MV, mitral valve; SF, ventricular shortening fraction; TR, tricuspid valve regurgitation; TV, tricuspid valve; REDV, reversed end-diastolic velocity; RV, right ventricle; UV, umbilical vein.

578

Fetal Cardiology

Appendix Table 38.1 Summary of requirements of Cardiovascular Profile Score • • • •

Cardiac/chest area ratio M-mode of RV/LV Doppler of four valves Pulsed Doppler of the ductus venosus, cord umbilical vein and artery, and middle cerebral artery

Appendix Table 38.3 Studies showing use of the Cardiovascular Profile Score (CVP) in fetal congestive heart failure Group

Abnormal CVP

Mortality* or early delivery**

1. (CHD)

20% < 7

*87.5% vs 15.2%37

2. (Hydrops)

60% < 7

*73.5% vs 26.5%36

3. (IUGR)

11% < 6

100%38

4. (AVB)

*82% < 7

*100%

**26%

**100%43

32% < 5

100% (unpublished data)

5. (Digoxin)

Appendix Table 38.2 Summary of Cardiovascular Profile Score 1. Hydrops: effusion −1 point, skin edema −2 points 2. Venous Doppler: atrial reversal: ductus venosus −1 point, umbilical vein atrial pulsations − 2 points 3. Heart size: C/T area ratio > 0.35 − 1 point, > 0.5 or < 0.25 − 2 points 4. Cardiac function: RV/LV shortening fraction < 0.28 − 1 point or tricuspid valve regurgitation (holosystolic) −1 point or mitral regurgitation − 1 point or pulmonary or aortic valve regurgitation −1 point or valve regurgitation dP/dt < 400 mmHg/s −2 points or ventricular hypertrophy − 1 point or monophasic filling − 2 points (maximal deduction is two points for each category) 5. Umbilical artery: absent end-diastolic velocity −1 point, reversed diastolic velocity − 2 points

AVB, atrioventricular block; CHD, congenital heart disease; IUGR, intrauterine growth restriction.

Appendix Table 38.4 CVP categories most important in five groups of fetuses diagnosed in utero Group

Component markers

p value

1. CHD

Hydrops, cardiomegaly

< 0.05

2. Hydrops

Abnormal venous Doppler

3. IUGR

Abnormal venous Doppler

< 0.001

Abnormal function

< 0.001

Cardiomegaly 4. LAI–AVB 5. Digoxin

Hydrops

0.008 < 0.001

Cardiomegaly

0.003

Abnormal DV Doppler

0.007

Hydrops

0.01

LAI, left atrial isomerism; DV, ductus venosus.

39 Congenital cardiovascular malformations and the fetal and neonatal circulation Abraham M Rudolph Almost 50 years ago, the important role of the change in pulmonary circulation normally occurring after birth in influencing the manifestations of many congenital cardiovascular malformations was recognized.1 It is now appreciated that not only changes in the pulmonary circulation, but also in other sites in the circulation after birth, may have profound effects on the hemodynamics and clinical manifestations postnatally. In addition, the effects of various congenital cardiovascular lesions on the development of the pulmonary vasculature after birth were documented. The interrelationships between prenatal development of the circulation and the presence of congenital cardiovascular malformations had not previously been appreciated, but with the increasing application of fetal echocardiography, it is now becoming evident that congenital cardiovascular malformations may also have profound effects on normal development of the circulation in the fetus, and the effects may be progressive with intrauterine growth of the fetus. Furthermore, developmental changes in the circulation during fetal life may affect the manner in which cardiovascular malformations present, as well as the time during gestation that the changes become manifest. This chapter reviews current knowledge regarding interactions between congenital cardiovascular malformations and development of the circulation in the fetus and the interactions during postnatal cardiovascular adaptation. The individual lesions are not discussed in detail, because they are presented in other chapters.

Fetal cardiovascular system Alterations in blood flow patterns and volumes Ventricular development For several decades, the presence of a small or hypoplastic left or right ventricle at birth has been considered to be

related to inadequate volume of blood flow into or out of the chamber during fetal life. This concept was based on the presence of associated aortic or pulmonary valve obstruction, which was thought to limit the volume of blood entering the respective ventricle. In an autopsy study of infants with obstructive lesions of the aorta or aortic valve, associated with hypoplastic left ventricle, Lev et al noted that the foramen ovale was small and proposed that the restriction of flow into the left atrium was responsible for the ventricular and aortic anomalies.2 It is difficult, however, to exclude the possibility that the small foramen was the result of the ventricular and aortic anomalies, rather than their cause. Postnatally, blood flows serially through the pulmonary and systemic circulations; obstruction to inflow or outflow of either ventricle will thus affect the output of both ventricles. If obstruction is severe, blood supply to the body will be compromised. In the fetus, however, the presence of foramen ovale and ductus arteriosus shunts may permit a normal combined ventricular output to be maintained even if inflow or outflow of either ventricle is severely restricted (see Chapter 10). Thus, with tricuspid or pulmonary valve atresia, venous return is deflected through the foramen ovale to the left atrium and ventricle; blood flow to the lungs is provided from the aorta through the ductus arteriosus. In the presence of mitral or aortic valve obstruction, both pulmonary and systemic venous blood returns to the right atrium and is ejected by the right ventricle into the pulmonary artery; systemic and umbilical arterial blood is provided by flow through the ductus arteriosus. In those congenital malformations with severe obstructions, flow to one ventricle is reduced, while the output of the other ventricle is increased. One functioning ventricle in the fetus is capable of providing blood flow adequate for normal umbilical and fetal body requirements, as evidenced by normal body growth and absence of hypoxia in many of these fetuses. To attempt to assess the role of volume of blood flow into or out of the left ventricle on chamber growth, we simulated models of reduced inflow and of obstructed outflow, in fetal lambs at 90–120 days’ (0.6–0.8) gestation.3

580

Fetal Cardiology

Figure 39.1 Sections of fetal hearts midway between apex and atrioventicular sulcus. (a) Normal fetal heart showing left (LV) and right (RV) ventricles. (b) Fetus 10 days after banding ascending aorta to simulate aortic stenosis. Note marked decrease in left ventricular size.

Left ventricular inflow was reduced to about one-third of normal by prolonged inflation of a balloon inserted into the left atrial cavity. This resulted in a progressive reduction in the size of the left ventricular cavity, reaching about 50% of normal left ventricular size in about 7 days, and was also associated with a decrease in left ventricular muscle mass (Figure 39.1). Outflow obstruction of the left ventricle was accomplished by placing a polyvinyl snare around the ascending aorta, just above the origin of the coronary arteries. It was tightened to achieve a systolic pressure gradient from the left ventricle to the aorta, beyond the obstruction, of about 20 mmHg. Since the snare does not expand, it may be assumed that with growth of the aorta, the obstruction becomes progressively more severe. During the first week, the left ventricular wall became thickened, but within 2–3 weeks hypoplasia of the left ventricle began to develop, and subsequently the cavity became almost slit-like. These studies are important, because they indicate that in the normal fetus, development of ventricular chamber size is affected by reducing its input or output. The effects of outflow obstruction of the right ventricle were studied by placing a snare around the main pulmonary trunk in fetal lambs at about 60 days’ gestation. As the lamb developed in utero, the obstruction of the pulmonary trunk became progressively more severe, simulating pulmonary stenosis. Hemodynamic studies in the lambs, performed about 60 days after the snare was placed around the pulmonary artery, showed that right ventricular systolic pressure was elevated and right ventricular output was reduced. An increase in left ventricular output achieved almost normal combined ventricular outputs. The right ventricular myocardial mass increased, but the size of the right ventricular cavity varied greatly. In some fetuses, the cavity was almost obliterated and

appeared slit-like. In others, the right ventricle was markedly enlarged. The size of the right ventricle appeared to be related to the competence of the tricuspid valve. When the valve was small, the right ventricular cavity was small, but a large tricuspid orifice was associated with an enlarged right ventricle. The differences in response could be explained by the volume of blood being handled by the ventricle. If the tricuspid valve remained competent the right ventricular stroke volume was reduced by the outflow obstruction and the cavity size decreased. However, with tricuspid incompetence the ventricle ejected a large volume at high pressure, with a large proportion regurgitating into the right atrium; this was associated with enlargement of the right ventricle. This difference in the behavior of the right ventricle to outflow obstruction has been observed in human fetuses with pulmonary atresia, some having markedly hypoplastic right ventricles and others showing marked right ventricular enlargement, sometimes so great as to compromise lung growth in the fetus.4 The studies in the lamb fetus provide convincing evidence that interference with flow into or out of a ventricle during fetal development may result in hypoplasia, because the myocardium is normal prior to the introduction of the lesion and there are no other abnormalities. In human fetuses, several reports have suggested that aortic stenosis may result in a progressive decrease in left ventricular size and may progress to a hypoplastic left heart with advancing gestation.5,6 If an infant is born with a hypoplastic left ventricle, it may not be capable of providing an output adequate for the needs of the systemic circulation. Survival depends on providing flow to the systemic circulation from the right ventricle. Recently, it has been proposed that it may be possible to avoid severe hypoplasia of the ventricle by relieving the aortic stenosis

Congenital cardiovascular malformations and the fetal and neonatal circulation

as early as possible in the fetus by performing balloon angioplasty. Techniques have been developed to accomplish this by a percutaneous approach from the mother’s abdomen. At this time more than 100 procedures have been performed worldwide, with a high rate of success, very little risk to the fetus, and minimal risk to the mother. Despite the apparent success in relieving or reducing the aortic stenosis in the fetus and avoiding further reduction in ventricular size, until recently there have been relatively few infants in whom the left ventricle has been able to provide an adequate output after an angioplasty in utero.7,8 However, in the largest series recently reported from Boston Children’s Hospital, about one-third of fetuses with successful aortic balloon angioplasty went on to have a biventricular heart after birth.9 These disappointing results have raised the question whether the abnormal ventricular development is secondary to the abnormal aortic valve, or possibly the primary lesion. Ultrasound observations in fetuses have rarely documented aortic stenosis with normal left ventricular size and function, later progressing to hypoplastic ventricle. In most fetuses observed at about 16–20 weeks’ gestation, the left ventricle is large and dysfunctional and has an echodense subintimal layer. The diagnosis of aortic stenosis is considered, and with advancing gestation, the ventricular size is noted to become progressively smaller. It is quite possible that the ventricular muscle is abnormal as a result of exposure to infection or toxin, or as a result of a genetic anomaly. This could interfere with left ventricular output, and the aortic valve anomaly could be the result of the reduced flow. If output is severely reduced, possibly even aortic atresia could develop. It is likely that there are several causes of left ventricular hypoplasia. If the ventricular muscle damage is primary, due to toxic, infectious, or genetic factors, balloon angioplasty of the aortic valve in utero is probably contraindicated, because it is not likely to alter the course of the abnormal process. If, however, aortic stenosis is the primary problem, prenatal relief of the obstruction, as early as possible, is likely to improve the prospect of allowing the left ventricle to sustain adequate systemic blood flow after birth. It is thus important to develop techniques that define the cause of the ventricular abnormality.

Ascending aorta and aortic arch development Most of the blood ejected by the left ventricle into the ascending aorta is directed to the brain and upper extremities. In fetal lambs, about 33% of combined ventricular output (CVO) is ejected by the left ventricle; about 23–25% of CVO is distributed to the head and forelimbs, so that only about 10% of CVO passes across the aortic isthmus to the descending aorta (Chapter 10). Although precise measurements of flow across the aortic isthmus in

581

the human fetus are currently not available, it appears that it is similar to that in the lamb. This relatively low blood flow across the aortic isthmus is reflected in the morphology. The ratio of isthmus diameter to ascending aortic diameter in the human fetus is about 0.7, so that the cross-sectional area is about one-half. In several congenital cardiovascular malformations, the volume of blood flowing into the aorta is altered, and this is associated with an increase or decrease in aortic diameter. Thus, in the fetus with pulmonary atresia with a ventricular septal defect, all the blood from the right and the left ventricle is ejected into the ascending aorta, an amount that is at least twice the normal volume; this results in an ascending aorta with a cross-sectional area considerably greater than normal. In contrast, in the fetus with aortic atresia, no blood enters the aorta from the left ventricle; blood passes into the aortic arch and ascending aorta retrograde from the ductus arteriosus, and the ascending aorta conducts only the small amount of blood flowing into the coronary circulation. This results in a hypoplastic ascending aorta with a markedly reduced diameter. In the fetus with an intact ventricular septum or a small ventricular septal defect, all or most of the blood returning to the heart enters the left atrium and is ejected by the left ventricle into the aorta (Figure 39.2). Thus, as with pulmonary atresia, the ascending aorta is large. However, in the fetus with tricuspid atresia with transposition of the aorta and pulmonary artery, the left ventricle ejects directly into the main pulmonary artery. Blood flow into the ascending aorta is derived from the small right ventricle, which receives blood across a ventricular septal defect (Figure 39.3). As a result of the reduced blood flow into the ascending aorta, it is small. Also, because most of the blood ejected into the aorta is directed to the carotid and subclavian arteries, little crosses the aortic isthmus, which therefore is quite narrow. Flow through the ductus arteriosus compensates for the lack of contribution from the aortic isthmus to the descending aortic flow. In other malformations, flow into the ascending aorta may be compromised in the fetus by the presence of left ventricular outflow obstruction. This may occur as a result of valvar aortic stenosis, or subvalvar stenosis, as in Taussig–Bing anomaly, and is often associated with aortic arch narrowing and aortic coarctation. Aortic coarctation is frequently associated with congenital cardiovascular malformations in which left ventricular outflow is reduced; it has been suggested that the low blood flow across the isthmus is the cause of the localized aortic narrowing. Coarctation of the aorta is frequently not evident in utero, and presents as aortic obstruction postnatally in association with closure of the ductus arteriosus (see below). A shelf projects from the posterolateral wall of the aorta toward the entry of the ductus arteriosus. It has been proposed by Hutchins11 that the flow through the ductus from the pulmonary

582

Fetal Cardiology

Figure 39.2 Diagram of the heart of a fetus with tricuspid atresia and a small ventricular septal defect. Most of the blood is ejected by the left ventricle. The ascending aorta and aortic arch are large. The pulmonary artery derives blood supply through the ductus arteriosus and is quite small. Pressures and oxygen saturation (in circles) in chambers and great vessels are shown. m, mean pressure (reproduced with permission from reference 10).

artery to the aorta is large and, in association with the low isthmus flow, induces overgrowth at the junction of the aortic isthmus with the descending aorta; this region acts as a branch point. Experimental studies have demonstrated this phenomenon in flow models.

Ductus arteriosus size and orientation In the normal fetus, a large proportion of blood ejected by the right ventricle is deflected away from the lung through the ductus arteriosus to the descending aorta. In the fetal lamb, almost 85% of right ventricular output, or slightly more than half of the combined ventricular output, passes through the ductus; in the human fetus, pulmonary blood flow is relatively greater, so 35–40% of combined ventricular output traverses the ductus arteriosus (see Chapter 10). As a result of the magnitude and direction of flow, the ductus joins the descending aorta with an oblique inferior angle (Figure 39.4).

Figure 39.3 Diagram of the heart of a fetus with tricuspid atresia, a small ventricular septal defect, and aortopulmonary transposition. Blood supply to the ascending aorta is decreased, resulting in a narrow ascending and transverse aorta. The ductus arteriosus is large and supplies most of the flow to the descending aorta. Pressures and oxygen saturation (in circles) in chambers and great vessels are shown (reproduced with permission from reference 10).

Congenital cardiovascular malformations such as pulmonary atresia or marked right ventricular outflow stenosis, tricuspid atresia, and Ebstein’s malformation with severe tricuspid insufficiency are associated with reduced or absent flow from the right ventricle into the pulmonary artery. The normal pattern of a large flow from the pulmonary artery through the ductus is replaced by flow from the aorta to the pulmonary artery to provide pulmonary blood flow. Because the volume of flow is less than normal, the ductus arteriosus is usually somewhat narrower. The inferior angle between the aorta and ductus arteriosus varies in these infants. In infants with pulmonary atresia with a ventricular septal defect, the pulmonary arteries are small, and the inferior angle between the ductus and the descending aorta is acute (Figure 39.4). However, in infants with pulmonary atresia with an intact ventricular septum, the pulmonary arteries vary in size. When the right ventricle is fairly well developed with inflow, body, and outflow regions (tripartite), the pulmonary arteries are fairly well developed, and the inferior

Congenital cardiovascular malformations and the fetal and neonatal circulation

Aortic isthmus Ductus arteriosus

Pulmonary trunk Ascending aorta

Ascending aorta Pulmonary trunk

Left and right pulmonary arteries

Normal fetus

Aortic isthmus Ductus arteriosus

Left and right pulmonary arteries

Coronary artery Fetus with aortic atresia

Aortic isthmus Ascending aorta

Pulmonary trunk

583

there is no pulmonary arterial to aortic flow through the ductus arteriosus. With pulmonary atresia with intact ventricular septum, the pulmonary valve is open for some time during gestation, and then becomes progressively stenosed and eventually atretic. Flow from the pulmonary artery through the ductus is maintained until the stenosis becomes severe, thus accounting for the normal obtuse angle in those with a well-developed right ventricle.12,13 Cardiac lesions causing obstruction on the left side of the heart, such as interrupted aortic arch, aortic atresia, severe aortic or left ventricular outflow stenosis, and mitral atresia, are associated with either markedly reduced or absent blood flow across the aortic isthmus to the descending aorta. Furthermore, ascending as well as descending aortic flow may be derived from flow through the ductus arteriosus from the pulmonary artery. The ductus arteriosus is usually larger than normal in these fetuses, and the inferior angle at the junction of the ductus with the descending aorta is oblique. As mentioned above, the greater than normal flow across the ductus with a smaller than normal isthmus may create a branch point at the isthmus–descending aorta junction and induce the development of a shelf of aortic coarctation.

Ductus arteriosus

Effects of obstructive lesions Left and right pulmonary arteries

Fetus with pulmonary atresia

The role of obstructive lesions in reducing blood flow in chambers and great vessels has been discussed above. In this section, the possible effects of obstruction on other organ systems is presented.

Figure 39.4 The pulmonary trunk, ductus arteriosus, and aorta are depicted in the normal fetus, a fetus with aortic atresia, and a fetus with pulmonary atresia. In the normal fetus, a large volume of blood flows from the pulmonary artery through the ductus to the descending aorta. The ductus is large and joins the aorta with an oblique inferior angle. In the fetus with aortic atresia, the total output of the heart is ejected into the pulmonary artery and most passes through the ductus, which is very large and also joins the aorta with an oblique inferior angle. In the fetus with pulmonary atresia, blood flows from the aorta through the ductus to the pulmonary artery. This supplies only the pulmonary circulation.The ductus is small and the junction with the aorta has an acute inferior angle (reproduced with permission from reference 10).

angle between the ductus and the aorta is usually obtuse. If the right ventricle is poorly developed (unipartite), the pulmonary arteries are small and the angle between the ductus and descending aorta is usually acute. This difference can possibly be explained by the time during fetal development that there is interference of flow patterns. With pulmonary atresia and ventricular septal defect, the atresia is probably present early in development, so that

Ductus arteriosus obstruction and the pulmonary circulation In the fetus, pulmonary arterial and aortic pressures are almost identical through most of gestation, but in the last few weeks before birth of the lamb, the pulmonary arterial pressure is slightly higher, presumably because the ductus arteriosus becomes somewhat constricted. Acute compression of the ductus in fetal lambs results in an elevation of pulmonary arterial pressure and a slight fall of aortic pressure. We demonstrated experimentally that compression of the ductus arteriosus in fetal lambs induced an increase in the amount of smooth muscle in fifth generation pulmonary arterioles, the resistance vessels in the lungs.14 The possibility that this could interfere with the normal fall of pulmonary vascular resistance after birth and cause the clinical syndrome of persistent pulmonary hypertension in the newborn infant was considered. More recent studies in lambs have confirmed these observations.15,16 The possible causes of constriction of the ductus arteriosus in utero are not well defined.

584

Fetal Cardiology

As mentioned in Chapter 10, the ductus arteriosus is dilated by prostaglandins. Administration of salicylates or indomethacin to fetal lambs in utero resulted in constriction of the ductus with an increase in pulmonary arterial pressure.17 The association of persistent hypertension of the newborn with ingestion of non-steroidal antiinflammatory agents by the mother is now well recognized;18 this is presumably the result of prolonged pulmonary arterial hypertension in the fetus with resulting increased pulmonary arterial smooth muscle development. Constriction of the ductus reported in some fetuses with aortopulmonary transposition is discussed below.

Aortic arch obstruction and cerebral blood flow In fetuses with hypoplastic left heart and aortic atresia, blood supply to the cerebral circulation cannot be provided from left ventricular output and is derived from retrograde flow across the aortic isthmus and arch from the ductus arteriosus. Studies in fetuses and newborn infants with hypoplastic left heart have demonstrated that head circumference is less than normal.19 It has been proposed that this may be the result of interference with fetal brain development resulting from reduced cerebral blood flow. This concept is based on observations of Doppler flow velocity studies of major cerebral arteries in fetuses with hypoplastic left heart. These have shown that the PI (pulsatility index), which relates maximum systolic velocity and low diastolic velocity to mean velocity, is reduced in cerebral arteries in these fetuses as compared with normal fetuses.20 A low PI is interpreted to indicate that the cerebral vascular resistance is decreased and this is the result of reduced blood flow. However, although a reduced PI does suggest that vascular resistance in the vascular bed is low, this does not necessarily apply when there is a stenosis in the artery proximal to the site of Doppler recording, because the stenosis itself may alter the PI. The most likely cause of reduced flow into the aortic arch and cerebral vessels is the frequent presence of coarctation of the aorta at the site of the junction of the aortic isthmus and the ductus. This could well account for the altered Doppler flow pattern in cerebral arteries. An additional possible cause for decreased flow into the cerebral circulation, which has not to my knowledge been explored, is the possibility of a diastolic ‘steal’ effect in the aortic arch. During systole, the kinetic effect of blood ejected into the pulmonary artery and across the ductus arteriosus will carry it forward into the arch and cerebral vessels as well as to the descending aorta. However, during diastole, blood will flow to sites with lowest resistance. In the fetus, the large placental circulation has a low resistance, whereas the upper body and brain resistance is relatively high. Thus, blood will flow

preferentially to the placenta, and possibly blood in the aortic arch and its branches may flow retrograde to the descending aorta. Infants with hypoplastic left heart syndrome have been reported to show a high incidence of neurodevelopmental problems,21 as well as cerebral lesions with imaging procedures,22 and it has been suggested that reduced cerebral blood flow during fetal life may be responsible. However, the possibility that whatever genetic or toxic factors were responsible for the cardiac anomalies also caused the cerebral disturbance cannot as yet be excluded. Although the association of cerebral abnormality with aortic arch interruption has not been studied extensively, it does not appear to be a common concern. This may be related to the site of the interruption. When the interruption is between the innominate and left carotid arteries, an adequate brain blood flow could be provided to the right carotid artery from the innominate artery and to the left carotid artery via the ductus arteriosus and aortic isthmus. If the interruption is at the isthmus, blood to both carotid arteries will be supplied from the ascending aorta. If the interruption is proximal to the innominate artery, possible interference with cerebral flow could occur, but interruption at this site is quite rare.

Foramen ovale obstruction Lev et al2 reported the association of narrowing or absence of the foramen ovale with hypoplastic left heart complexes, and considered that the lack of flow into the left heart could be the cause of the left-sided hypoplasia. However, the abnormalities in the circulation that would result if left ventricular hypoplasia or aortic or mitral atresia were the primary lesion could well account for the abnormalities of the foramen ovale. Right ventricular output would be increased to maintain combined ventricular output, because left ventricular output is reduced. Pulmonary venous blood returning to the left atrium would not be ejected by the left ventricle and would therefore have to pass through the foramen ovale. Left atrial pressure will be elevated and the atrial septum displaced to the right, promoting closure of the foramen. The increase in left atrial and pulmonary venous pressure has been considered to be the cause of pulmonary lymphangiectasis that may be associated with hypoplastic left heart syndrome and absent foramen ovale.23 It is now recognized that the presence of obstructed or absent foramen ovale in infants born with hypoplastic left heart is associated with a high mortality both preoperatively, and following surgical palliation.24 This has stimulated attempts to open the foramen ovale urgently after birth before surgery is performed, if obstruction is identified. More recently, it has been suggested that, if foramen ovale obstruction is recognized in the fetus with hypoplastic left heart, attempts should be made to achieve

Congenital cardiovascular malformations and the fetal and neonatal circulation

an adequate opening in utero.25 It has recently been appreciated that echocardiographic evidence of a change in pulmonary venous flow, compatible with interference with forward flow, is useful in detecting the presence of foramen ovale obstruction.26

Effects of changes in blood oxygen content Increase of pulmonary arterial oxygen saturation Relatively little consideration has been given to possible effects that alterations in oxygen levels may have on the circulation in the fetus and neonate with congenital cardiovascular malformations. In 1974, I postulated that in fetuses with aortopulmonary transposition, pulmonary arterial oxygen saturation would be higher, and ascending aortic oxygen saturation lower, than normal.27 It was suggested that this could affect the pulmonary circulation and possibly oxygenation and metabolism of the brain. In studies of fetuses by echocardiography, several fetuses with aortopulmonary transposition were observed to have either a small foramen ovale or a narrowed ductus arteriosus or both.28 Jouannic et al reported that about 4% of infants with transposition had serious difficulty soon after

585

birth, and there was a frequent association of early severe hypoxemia with abnormalities of the foramen ovale or ductus arteriosus noted on fetal echocardiography.29 These observations could well be explained by higher than normal oxygen saturations in pulmonary arterial blood. In the fetus, well-oxygenated umbilical venous blood passing through the ductus venosus is preferentially directed through the foramen ovale to the left atrium and ventricle. Systemic venous blood from the superior and inferior caval veins preferentially flows through the tricuspid valve to the right ventricle (see Chapter 10). Normally, right ventricular blood with relatively low oxygen saturation is ejected into the pulmonary trunk, and measurements in fetal lambs indicate that the arteries of the lung and the ductus arteriosus are exposed to blood with an oxygen saturation of about 50% and a Po2 of about 18 mmHg (Figure 39.5). Blood ejected by the left ventricle into the ascending aorta has an oxygen saturation of about 65% with a Po2 of about 25 mmHg. Venous flow patterns in the umbilical vein, ductus venosus, and caval veins appear to be normal in the fetus with aortopulmonary transposition. However, well-oxygenated blood is ejected by the left ventricle into the pulmonary trunk, and thus the pulmonary circulation and the ductus arteriosus are exposed to blood with a higher than normal oxygen saturation and Po2. Based on assumptions that the proportions of combined ventricular output flowing through vessels and shunts are similar to those in normal

55

Descending aorta Lungs Ductus 55 Arteriosus Placenta

Pulmonary trunk 85

35

Aorta 65 50 Left ventricle

Right ventricle

Figure 39.5

Vena cava

Right atrium

Foramen ovale

85

Left atrium

48

Pulmonary veins

Diagram of circulation in the normal fetal lamb showing patterns of blood flow and oxygen saturations in cardiac chambers and great vessels. Note the higher oxygen saturation in the ascending compared with the descending aorta and low saturation in the pulmonary artery (reproduced with permission from reference 30).

586

Fetal Cardiology

60

Descending aorta

72

Lungs 78 Ductus arteriosus

Aorta 85 Placenta

Pulmonary trunk

45

35

Figure 39.6

78 Left ventricle

Right ventricle

Vena cava

Right atrium

85

Foramen ovale

Left atrium

68

Pulmonary veins

fetuses, it is estimated that the pulmonary vasculature and ductus arteriosus would be exposed to an oxygen saturation of about 72% and PO2 of ∼28 mmHg (Figure 39.6). The fetal pulmonary circulation is very sensitive to relatively small changes in PO2. In a study in fetal lambs, in which the ewe inhaled 100% oxygen, an increase in PO2 of 7 mmHg resulted in a three-fold increase of pulmonary blood flow.31 The lower pulmonary vascular resistance resulting from perfusion with relatively high PO2 in the fetus with transposition will result in an increase of flow through the pulmonary circulation and a reduction of flow through the ductus arteriosus. The increased pulmonary blood flow will result in greater venous return to the left atrium, elevating left atrial pressure; this could deviate the atrial septum to the right and thus tend to reduce foramen ovale size. The reorientation of flow from the pulmonary trunk could also have an effect on the size of the ductus arteriosus, because a smaller volume would flow through it to the descending aorta. The responsiveness of the pulmonary circulation to oxygen changes with gestational age. In fetal lambs, raising pulmonary arterial PO2 levels produces minimal pulmonary vasodilatation before about 100 days’ gestation (term ∼150 days); a progressive increase in vasodilator response is observed with increasing gestational age.32 A similar difference in response with gestational age has been reported in the human. Inhalation of oxygen by the mother did not affect pulmonary blood flow in fetuses of about 25 weeks’ gestation, but did increase flow in fetuses beyond 30 weeks’ gestation.33 Based on these

Diagram of circulation in the fetus with aortopulmonary transposition showing patterns of blood flow and oxygen saturations in cardiac chambers and great vessels. Oxygen saturation in the pulmonary artery, which arises from the left ventricle, is very high compared with normal. Saturation in the ascending aorta, which arises from the right ventricle, is relatively low. Data are based on estimates of volumes of blood flow in various vessels (reproduced with permission from reference 30).

observations, the effects of the higher oxygen saturation in pulmonary arterial blood in fetuses with transposition may not be observed until the third trimester. The higher PO2 in pulmonary arterial blood, in addition to reducing flow through the ductus arteriosus as a result of increased pulmonary blood flow, may have a direct vasoconstrictor effect on the ductus. The response of the ductus to oxygen also increases with advancing gestational age. In lamb fetuses below 90 days’ gestation, an increase of PO2 results in minimal constriction of the ductus, but the responsiveness increases with advancing gestation.34 Constriction of the ductus would tend to enhance diversion of blood into the pulmonary circulation. It is of interest that constriction of the ductus could provide an explanation for the occurrence of postnatal elevation of pulmonary vascular resistance observed in some infants with transposition with intact ventricular septum.28,35 Constriction of the ductus arteriosus in the fetus elevates pulmonary arterial pressure, and this induces an increase in pulmonary vascular smooth muscle development; this may result in persistent pulmonary hypertension in the newborn infant (see above). This same phenomenon may occur in fetuses with transposition. Whether the increase in pulmonary vascular resistance resulting from pulmonary arterial hypertension is opposed by the increase in pulmonary arterial PO2 in fetuses with aortopulmonary transposition is yet to be resolved.

Congenital cardiovascular malformations and the fetal and neonatal circulation

Decrease of ascending aortic oxygen saturation In the normal fetus, oxygen saturation of blood ejected from the left ventricle into the ascending aorta is relatively high – about 65%. This is due to preferential streaming of well-oxygenated umbilical venous blood passing through the ductus venosus, across the foramen ovale into the left atrium (Figure 39.5). In many congenital cardiovascular malformations, all venous blood returning to the heart is mixed almost completely, and blood ejected into the aorta and pulmonary arteries has similar oxygen saturations. Thus, in fetuses with tricuspid or pulmonary atresia with intact ventricular septum, all the venous blood enters the left atrium, whereas with mitral atresia, or aortic atresia with intact ventricular septum, all venous blood mixes in the right atrium. Based on magnitudes of venous return in the lamb fetus, it is estimated that the oxygen saturation of blood delivered to all parts of the body will be about 60%. This represents a small decrease in blood oxygen content delivered to the cerebral and coronary circulations, and, although it may induce some vasodilatation in these organs, the effect is probably minor. These lesions would also result in an increase above normal of oxygen saturation of pulmonary arterial blood, although the effect would not be as marked as with transposition (see above), but the possibility that pulmonary vascular resistance is lower than normal and the ductus arteriosus is constricted is yet to be determined. In fetuses with aortopulmonary transposition, the ascending aortic blood oxygen saturation would be reduced considerably, from a normal of ∼65%, to about 45% (Figure 39.6). A fall in oxygen saturation of blood perfusing the brain results in cerebral vasodilatation with increased pulmonary blood flow.36 The lower oxygen saturation is not likely to affect brain function and development, because even when oxygen delivery to the brain is decreased acutely, cerebral oxygen consumption is maintained at normal levels.37 There is evidence that cerebral blood flow is increased in fetuses with aortopulmonary transposition, because Doppler velocity studies of cerebral arteries have demonstrated a low pulsatility index, suggesting that cerebral vascular resistance is reduced.38 However, the effect of a prolonged increase in cerebral blood flow on brain development has not been examined. Furthermore, although the increased cerebral flow may permit adequate oxygen consumption in the unstressed fetus with transposition, there may be inadequate reserve to maintain cerebral oxygen consumption at normal levels if hypoxic stress occurs for any reason. The possibility that cerebral development may be affected is suggested by the observation that head circumference is somewhat reduced in infants born with transposition.19 The oxygen saturation of blood perfusing the coronary circulation is also reduced in the fetus with transposition.

587

During acute hypoxemia in fetal lambs, coronary blood flow increases markedly,36 and even with quite severe hypoxemia myocardial oxygen consumption is maintained at normal levels.39 However, as with the cerebral circulation, no information is available regarding possible deleterious effects of a prolonged increase in coronary blood flow. Also, the ability to maintain oxygen delivery in the event of hypoxic stress is questionable.

Relationship between congenital cardiovascular malformations and postnatal circulatory adjustments The changes in the circulation occurring normally after birth are discussed in Chapter 10. The main adjustments that occur are ventilation resulting in a rise in arterial oxygen tension, a fall in pulmonary vascular resistance, constriction of the ductus arteriosus, closure of the foramen ovale, elimination of the umbilical circulation, and closure of the ductus venosus. Blood flow patterns after birth are frequently altered by congenital cardiovascular malformations, and may account for the clinical manifestation of these anomalies,

Decrease in pulmonary vascular resistance Effect of abnormal communications Communications between the left and right ventricles, the aorta and pulmonary arteries, and the left and right atria are among the more common congenital cardiac anomalies. The magnitude and time course of changes in pulmonary vascular resistance after birth are important in determining the hemodynamic and clinical manifestations of these lesions after birth. After birth, the elimination of the relatively lowresistance umbilical–placental circulation, which arises from the aorta, results in an increase in systemic vascular resistance. Ventilation induces a marked fall in pulmonary vascular resistance (Chapter 10). Normally, separation of the left and right sides of the heart is achieved by closure of the ductus arteriosus and foramen ovale, and blood then flows serially through the pulmonary and systemic circulations, as in the adult. With the fall in pulmonary vascular resistance and closure of the ductus arteriosus after birth, pulmonary arterial and right ventricular pressures drop.

588

Fetal Cardiology

A large ventricular septal defect or a large patent ductus arteriosus tends to equalize the pressures in the left and right sides of the heart. With a large patent ductus, both systolic and diastolic pressures in the aorta and pulmonary artery are similar, whereas with a large ventricular septal defect, systolic pressures in the ventricles and great arteries are the same but diastolic pressure in the pulmonary artery may be lower than that in the aorta. With equal pressures, blood flows through the systemic and pulmonary circulations are determined by the relative resistances of the pulmonary and systemic circulations. As mentioned, systemic vascular resistance increases and pulmonary vascular resistance falls after birth. In the presence of a large communication at the ventricular or aortopulmonary level, blood flows preferentially into the pulmonary circulation and a left-to-right shunt develops after birth. The blood perfusing the pulmonary circulation includes systemic venous return and the blood shunted from the left to the right side of the heart. The pulmonary/systemic blood flow ratio is commonly used to indicate the magnitude of the blood that is shunted; the higher is the ratio, the greater is the amount of blood shunted. The blood shunted left to right reduces the volume of blood ejected by the left ventricle that enters the systemic circulation. Thus, if the pulmonary/systemic blood flow ratio is 2:1, only half the blood ejected by the left ventricle is distributed to the systemic circulation, and if the ratio is 3:1, only one-third of the left ventricular output reaches the systemic circulation. To maintain a systemic blood flow adequate to provide oxygen and nutritional needs, left ventricular output has to be increased. The blood shunted left to right enters the pulmonary circulation and returns through the pulmonary veins to the left atrium, with a resulting increase in left atrial and left ventricular end-diastolic pressures. As discussed in Chapter 10, the heart in the neonate is capable of increasing its output in response to elevated atrial filling pressures. During the early neonatal period, the decrease in pulmonary vascular resistance is achieved by relaxation of pulmonary vascular smooth muscle, but the fall is limited by the presence of the smooth muscle in the media of the arterioles. Therefore, the magnitude of the left-to-right shunt and increase of left atrial pressure is restricted. The hemodynamic changes associated with a large ventricular septal defect in the early neonatal period are depicted in Figure 39.7. In the normal infant, pulmonary vascular resistance continues to fall gradually for 6–8 weeks after birth, as the walls of the small pulmonary arteries become thinner as smooth muscle regresses. In the infant with a large left-toright shunt lesion, pulmonary vascular resistance also continues to fall gradually, resulting in a progressive increase in left to right shunt and left atrial pressure (Figure 39.8). In infants with a large ventricular septal defect or aortopulmonary communication, the fall in pulmonary vascular resistance is somewhat slower, and the levels achieved

Figure 39.7 Diagram of the circulation in the heart and great vessels of an infant with a large ventricular septal defect in the early neonatal period. Pulmonary vascular resistance has not yet dropped markedly. Oxygen saturations are shown in circles. Note moderate left-to-right shunt with a small increase in left atrial pressure (reproduced with permission from reference 10.)

are not as low as in the normal infant. This is probably the result of the persistence of the high pulmonary arterial pressure after birth. The progressive increase in the left-to-right shunt results in increasing left atrial pressure, eventually leading to transudation of fluid into the lungs, causing increased respiratory effort. The stress on the left ventricle to markedly increase its output is associated with sympathetic– adrenal and renin–angiotensin stimulation; this is responsible for many of the clinical features of congestive heart failure, such as sweating, fluid retention, increased metabolism, and failure to thrive. Hepatomegaly is a relatively late manifestation. The immediate objectives of treatment are to reduce sodium and fluid retention by diuresis, reduce afterload on the left ventricle by vasodilators, and possibly improve ventricular performance by use of inotropic agents. Following stabilization of the clinical condition, the communication is closed either surgically or, if possible, by an interventional catheterization procedure. Infants with smaller communications do not present with symptoms in the neonatal period. Small communications do not usually cause symptoms, and pulmonary

Congenital cardiovascular malformations and the fetal and neonatal circulation

589

stroke volume is very dependent on afterload, especially in the fetus and neonate (Chapter 10). The fall in pulmonary vascular resistance results in a decrease in afterload on the right ventricle, and thus right ventricular stroke volume is greater than that of the left. This results in greater emptying and a smaller right ventricular endsystolic volume; a larger volume flows into the right ventricle, and left-to-right shunting across the atrial septum occurs. The progressive fall in pulmonary vascular resistance over 6–8 weeks after birth permits an increasing left-to-right shunt. The decrease in pulmonary arterial and right ventricular pressure is associated with a decrease in right ventricular wall thickness and the left to right shunt is enhanced by the reduction in right ventricular compliance. Most infants with atrial septal defects do not have significant symptoms. The left atrial pressure is not elevated, and thus respiratory symptoms are not evident. The volume overload on the right ventricle increases gradually and, since right ventricular pressure is normal or only modestly increased, it is well tolerated. Occasionally infants with large atrial septal defects do manifest some evidence of cardiac failure, probably as a result of neurohormonal stimulation.

Figure 39.8 Diagram of the circulation in the heart and great vessels of an infant with a large ventricular septal defect after pulmonary vascular resistance has decreased markedly. Oxygen saturations are shown in circles. Note large left-to-right shunt with marked increase in left atrial pressure. Pulmonary arterial systolic pressure is high, but diastolic pressure is lower than in the aorta (reproduced with permission from reference 10).

arterial pressures fall normally after birth. Moderatesized communications may present with evidence of congestive failure 2–3 months after birth, in association with the fall in pulmonary vascular resistance and increasing shunt. The postnatal response to a large communication at the atrial level differs from that in ventricular or aortopulmonary communications. Associated with the decrease in pulmonary vascular resistance, pulmonary arterial pressure falls. If the atrial septal defect is large, pressures in the left and right atria will be similar. With equal pressure in the two chambers, flow into the left and right ventricles during ventricular diastole will be determined by their end-systolic volumes and by the compliance of the ventricular walls. During fetal life, the right and left ventricular pressures are equal and the thickness of the walls is similar; with the same pressure in the atria, flow into each ventricle would be similar, and no significant shunting should occur. However, left-to-right shunting has been observed soon after birth, and can be explained by the fall in pulmonary vascular resistance. Ventricular

Communications in preterm infants Premature infants with communications between the left and right sides of the heart frequently develop symptoms and signs soon after birth. Although it is not fully understood why this occurs, it could be related to a more rapid and greater decline in pulmonary vascular resistance after birth, inability of the left ventricle to increase its output adequately to maintain systemic blood flow when left-toright shunt is large, and greater permeability of the pulmonary capillaries to transudation of fluid into the lung tissues and alveoli. As mentioned in Chapter 10, the pulmonary circulation responds poorly to hypoxia in the sheep fetus prior to 90 days’ gestation and the response increases progressively to term. The preterm infant may therefore have a lower pulmonary vascular resistance at birth and also show a more rapid fall after birth, so that left-to-right shunt develops rapidly. Although left atrial pressure may not be markedly elevated, transudation of fluid into the lung is more likely to occur in the premature infant, because the pulmonary vessels are more permeable. This is further facilitated by the fact that plasma albumin concentrations are low, and this reduces intravascular osmotic pressure. Preterm infants frequently present with respiratory distress, requiring assisted ventilation due to surfactant deficiency. The presence of a left-to-right shunt lesion may prevent weaning the infant from assisted respiration, because pulmonary fluid accumulation interferes with establishment of normal respiration.

590

Fetal Cardiology

Administration of diuretic agents may improve ventilation, but it may be necessary to close the communication. The most common communication is a patent ductus arteriosus (see below), but ventricular or atrial septal defects, even of relatively small size, may be associated with difficulty in establishing normal respiration in preterm infants.

Closure of the ductus arteriosus The ductus arteriosus normally constricts after birth, and in the human infant is functionally closed within about 15 hours. As discussed above, during fetal life congenital cardiovascular malformations may affect development of the ductus, and the presence of the ductus may be important in providing blood flow to the pulmonary or systemic circulation if the right or left ventricular outflow is obstructed by a cardiac lesion. The ductus arteriosus may be very important to the survival of infants with various congenital malformations after birth, by providing blood flow to the pulmonary or systemic circulations. Preterm infants frequently have delayed closure of the ductus arteriosus; this may have serious implications for their survival (see below).

Decreased pulmonary blood flow In the fetus, the lung is not the organ of gas change; pulmonary blood flow is relatively low. In the presence of pulmonary atresia or tricuspid atresia with an intact ventricular septum, pulmonary blood flow is supplied through the ductus arteriosus in the fetus. Because the flow is much lower than occurs normally, the ductus is smaller. Postnatally, with the assumption of the function of gas exchange by the lung, pulmonary blood flow has to increase 8–10-fold. In the immediate postnatal period, while the ductus is open, pulmonary blood flow may be large enough to provide adequate oxygen uptake to meet metabolic needs. Constriction of the ductus results in a reduction in pulmonary blood flow with a decrease in oxygen uptake in the lungs (Figure 39.9). In these congenital cardiovascular malformations, complete admixture of systemic and pulmonary venous return occurs in the left atrium and ventricle. The oxygen saturation of this mixed blood is determined by the pulmonary/systemic blood flow ratio. The lower is the pulmonary blood flow, the lower is the arterial oxygen saturation. Normally, ductus arteriosus closure occurs within 10–15 hours after birth. However, in many infants with pulmonary or tricuspid atresia, the ductus remains patent for a considerably longer period. It is not known what factors are responsible for the delayed closure. It is possible that the fall in arterial oxygen saturation that occurs when the ductus constricts may then tend to relax it.

Figure 39.9 Pulmonary atresia in a newborn infant, showing that pulmonary blood flow is dependent on patency of the ductus arteriosus.Systemic and pulmonary venous blood mixes in the left atrium and total cardiac output is ejected by the left ventricle. Course of blood flow and oxygen saturations (in circles) are shown (reproduced with permission from reference 10).

Possibly, circulating prostaglandin concentrations could remain elevated for longer periods after birth. Prostaglandin is cleared from blood during passage through the pulmonary circulation, and since pulmonary blood flow is low, the postnatal decrease in prostaglandin concentrations may be delayed. When the ductus is constricted, it can usually be opened by infusing prostaglandin E; when it is markedly constricted, the relaxation may be delayed for 15–30 minutes or even more. It is therefore recommended that prostaglandin be infused early, to try to avoid severe constriction.40

Decreased systemic blood flow In fetuses with aortic atresia, the flow to the systemic circulation, as well as the umbilical–placental circulation, is provided through the ductus arteriosus; it is therefore widely patent in the fetus. After birth, systemic blood flow continues to be dependent on the ductus for flow from the pulmonary artery to the aorta (Figure 39.10). Postnatally, however, flow through the ductus is reduced, because umbilical–placental flow is eliminated. Constriction of the

Congenital cardiovascular malformations and the fetal and neonatal circulation

591

In infants with ductus-dependent systemic blood flow, prostaglandin infusion is usually effective in relaxing the ductus, and this may provide adequate flow for tissue requirements and permit planning for a procedure in an infant who has not suffered severe metabolic disturbances. A decision may be made to perform a heart transplant in some infants with aortic atresia, and, to promote survival of the infant until such time as an organ becomes available, a stent has been inserted into the ductus by an interventional catheterization procedure to maintain patency.41

Coarctation of the aorta

Figure 39.10 Course of the circulation and pressures and oxygen saturations (in circles) are shown in a newborn infant with aortic and mitral atresia. Systemic and pulmonary venous blood all returns to the right atrium and is ejected by the right ventricle. Blood flow into the aorta is dependent on patency of the ductus arteriosus. Constriction of the ductus interferes with systemic blood flow (reproduced with permission from reference 10).

ductus after birth reduces the systemic blood flow, resulting in a decrease in arterial pressure with poor pulse. If flow is greatly reduced, oxygen delivery to the tissues may be inadequate for metabolic needs, and an increase in anaerobic metabolism with lactic acidemia results, even though oxygen saturation is not greatly decreased. Decreased blood flow to the kidney may result in renal damage. The ductus arteriosus is necessary to provide systemic blood flow to the lower body in infants with aortic arch interruption; constriction of the ductus results in a reduction of blood flow with inadequate oxygen supply and metabolic acidemia. In these lesions, as in conditions with ductus-dependent pulmonary blood flow, the ductus arteriosus may remain open for much longer than normal after birth.

The presence of an aortic shelf in the region of the entrance to the ductus arteriosus, which narrows the lumen of the descending aorta in the fetus, has been noted in the fetus or neonate soon after birth. However, no evidence of obstruction to flow into the descending aorta may be induced while the ductus is open. Experimental studies in fetal lambs have demonstrated the role of the ductus arteriosus in determining the hemodynamic and clinical manifestations of aortic coarctation. An indentation was created by plicating the aortic wall opposite the entry of the ductus. After recovery from the surgical procedure, pressures were measured in the ascending and descending aorta before and after constriction of the ductus arteriosus. Prior to constriction of the ductus, no pressure difference was noted, but after closure, a considerable pressure gradient developed, indicating restriction of flow into the descending aorta.42 Similar observations have been made in infants. It is common for infants with aortic coarctation to have no manifestations in the immediate neonatal period and to have normal pulse in the upper and lower extremities, with equal blood pressures. Within days, or even up to 2–4 weeks, evidence of aortic coarctation, as manifested by weak pulse and decreased arterial pressures in the lower extremities, becomes manifest. This course can be explained by the behavior of the ductus arteriosus. While the ductus is widely patent, no significant obstruction of the descending aorta is evident (Figure 39.11), but closure of the ductus induces obstruction in the aorta. This results in an increase in ascending aortic pressure, and left ventricular systolic and end-diastolic and left atrial pressures increase (Figure 39.12). The clinical features are those of left ventricular failure such as increased respiratory effort, sweating, and poor perfusion of the peripheral circulation. In some infants, deterioration may occur rapidly as a result of acute onset of aortic obstruction. The ductus arteriosus normally is functionally closed within about 15 hours after birth. The delay in the onset of aortic obstruction in some infants could be explained by lack of constriction of the ductus at the aortic end. The ductus usually constricts nearest to the pulmonary

592

Fetal Cardiology

Figure 39.11 Circulation in a fetus with an aortic shelf opposite the ductus arteriosus orifice in the aorta. Note that patency of the ductus permits flow with no functional obstruction in the aorta. Course of flow and oxygen saturations (in circles) are shown (reproduced with permission from reference 10).

arterial attachment, and this progresses to the aortic end. It is not unusual to observe a persistent opening known as the ductus ampulla of the aorta for days or weeks. Closure of this region will result in aortic obstruction. It has been hypothesized that coarctation of the aorta may result from the extension of ductus tissue into the aortic wall, and that when the ductus constricts, this tissue also constricts, thus inducing aortic obstruction. Although this concept cannot be excluded, it does not explain the presence of the aortic shelf noted prenatally. Use of prostaglandin has been very effective in improving the clinical status of infants with aortic coarctation. Reducing ductus arteriosus constriction decreases the degree of aortic obstruction with improvement of symptoms. The risk of surgery to relieve aortic coarctation in young infants has been greatly improved by using prostaglandin to improve the clinical status prior to the procedure.

Ductus arteriosus in aortopulmonary transposition In infants with aortopulmonary transposition, systemic venous blood is ejected into the aorta, and pulmonary

Figure 39.12 Circulation in an infant with an aortic shelf showing that constriction of the ductus arteriosus induces obstruction in the descending aorta. Ascending aortic and left ventricular systolic and end-diastolic pressures have increased. Pressures and oxygen saturations (in circles) are shown (reproduced with permission from reference 10).

venous blood is ejected into the pulmonary artery. If there are no communications, systemic venous blood cannot enter the lungs to be oxygenated, and pulmonary venous blood cannot enter the systemic arterial circulation to provide oxygen to the tissues. In infants who do not have an associated ventricular septal defect to allow mixing of the two circulations, survival is dependent on patency of the foramen ovale or the ductus arteriosus, or both. During the early neonatal period, bidirectional shunting of blood may occur through the ductus; blood shunts from the pulmonary artery to the aorta during systole as a result of the kinetic force provided by left ventricular ejection. The short main pulmonary artery and orientation of the ductus arising from the pulmonary artery tend to favor this flow. During diastole, the lower resistance of the pulmonary circulation resulting from ventilation favors flow from the aorta to the pulmonary artery. Shunting from the pulmonary artery to the descending aorta can be recognized by the presence of a higher oxygen saturation or Po2 in blood in the lower extremities than in the right arm. This shunting decreases over a few days as pulmonary vascular resistance falls further, and if the ductus remains patent, shunting occurs only from the aorta to the pulmonary artery. During the early neonatal period,

Congenital cardiovascular malformations and the fetal and neonatal circulation

593

infusion of prostaglandin to maintain patency of the ductus may be helpful in providing bidirectional shunting and thus improve oxygenation in the infant.

Postnatal effects of congenital cardiovascular malformations on pulmonary circulation

Ductus arteriosus in preterm infants

The normal changes in the pulmonary circulation after birth are discussed in Chapter 10. In the presence of a large communication between the left and right ventricles or between the aorta and pulmonary artery, pressure in the pulmonary artery does not fall normally after birth, because the high pressure from the left side is transmitted to the right ventricle and pulmonary artery. The resistance arterioles in the lung retain some degree of vasoconstriction; this is essential to avoid transmission of the high pulmonary arterial pressure into the capillary circulation with resulting fluid transudation and pulmonary edema. The smooth muscle layer of the pulmonary arterioles does not undergo the regression seen normally, and the thicker medial layer in the small pulmonary arteries persists. The mechanisms responsible for the pulmonary arteriolar responses have not been resolved. It has been proposed that pulmonary hypertension affects the smooth muscle through a mechanical effect to prevent regression and to enhance further growth. Recently, however, the possibility that mitogens released from the endothelial or muscle layers may be involved has emerged. During the period that the pulmonary vascular resistance is elevated as a result of the increase in smooth muscle in the media, administration of vasodilators such as tolazoline, prostacyclin, or sildenafil will reduce resistance and increase the left-to-right shunt. If the communication between the ventricles or great arteries is not closed, progressive changes in the pulmonary circulation are likely to occur. Endothelial proliferation occurs, with thickening of the intima; this may not involve the whole circumference of the vessel. With progressive intimal thickening, the lumen is narrowed. In association with the intimal changes, fibroblast proliferation develops in both the media and the intima. Smooth muscle is gradually replaced and the lumen becomes quite narrow, With progression of these changes the pulmonary vascular resistance increases and the left-to-right shunt diminishes. When pulmonary vascular resistance approaches levels in the systemic circulation, right-to-left shunting across the communication with cyanosis may occur, first with exercise and later continuously. When these proliferative changes in the pulmonary circulation have occurred, pulmonary vasodilator agents produce almost no response. Closure of the communication at this stage is not likely to result in a significant fall of pulmonary vascular resistance and may be risky, because the high pulmonary vascular resistance may limit venous return to the left ventricle, causing syncope, or may induce right heart failure. The hypothesis was proposed that the shear on the intima created by high blood flow through constricted

The high incidence of persistent patency of the ductus arteriosus in premature babies and the possible mechanisms involved are discussed in Chapter 10. The hemodynamic and clinical manifestations of patent ductus arteriosus in the preterm infant differ in many ways from those in mature infants. Pulmonary vascular resistance falls very rapidly after birth, probably related to immaturity of the pulmonary circulation. This allows the rapid onset of a large left-to-right shunt with increased pulmonary blood flow, placing demands on the left ventricle to increase its output. The pulmonary vascular bed is more permeable to fluids as well as albumin in the preterm infant; this, combined with the relatively low plasma albumin concentration, favors transudation into the lung parenchyma and alveoli. Left ventricular output increases markedly after birth (see Chapter 10), and a left-to-right shunt places an additional burden on the ventricle to increase output. If the shunt is large, the ventricle may not be able to provide an output adequate to maintain flow to systemic tissues. Since left ventricular output is dependent on an adequate filling pressure, an additional factor that could compromise left ventricular output is the presence of incompetence of the foramen ovale. Enlargement of the left atrium due to the enhanced venous return may stretch the foramen and allow a leftto-right shunt. This will limit the ability for left atrial pressure to increase and maintain filling pressure of the ventricle. The presence of patency of the ductus is recognized clinically by the presence of a murmur in the upper left chest parasternally, and prominent pulse due to the high pulse pressure resulting from diastolic run-off through the shunt. The murmur does not always have the characteristic features of a patent ductus, namely a continuous murmur, and usually is present only in systole. When the ductus arteriosus is widely patent, no murmur may be audible. These infants often develop respiratory distress due to surfactant deficiency. This is aggravated by the pulmonary edema resulting from the left-to-right shunt. The presence of a patent ductus arteriosus frequently prolongs the need for assisted respiratory support. Inability of the left ventricle to maintain an output that is capable of providing adequate systemic blood flow may result in poor perfusion of many organs and is probably a contributor to the development of necrotizing enterocolitis in some of these infants. The use of prostaglandin synthesis inhibitors to constrict the ductus arteriosus in preterm infants is discussed in Chapter 10.

594

Fetal Cardiology

arterioles may induce endothelial proliferation. More recently the role of the release of mitogens or vascular elastase, which promote proliferation of various cell types, is being pursued.

References 1. Rudolph AM, Nadas AS. The pulmonary circulation and congenital heart disease. N Engl J Med 1962; 267: 968–74. 2. Lev M, Arcilla R, Rimoldi HJA, Licata RH, Gasul BM. Premature narrowing or closure of the foramen ovale. Am Heart J 1963; 65: 638–47. 3. Fishman NH, Hof RB, Rudolph AM. Models of congenital heart disease in fetal lambs. Circulation 1978; 58: 354–64. 4. Allan, LD, Crawford DC, Tynan MJ. Pulmonary atresia in prenatal life. J Am Coll Cardiol 1986; 8: 1131–6. 5. Axt-Fliedner R, Kreiselmaier P, Schwarze A, Krapp M, Gembruch U. Development of hypoplastic left heart syndrome after diagnosis of aortic stenosis in the first trimester by early echocardiography. Ultrasound Obstet Gynecol 2006; 28: 106–9. 6. Hornberger LK, Sanders SP, Rein AJ et al. Left heart obstructive lesions and left ventricular growth in the midtrimester fetus. A longitudinal study. Circulation 1995; 92: 1531–8. 7. Kohl T, Sharland G, Allan LD et al. World experience of percutaneous ultrasound-guided balloon valvuloplasty in human fetuses with severe aortic valve obstruction. Am J Cardiol 2000; 85: 1230–3. 8. Tworetzky W, Wilkins-Haug L, Jennings RW et al. Balloon dilation of severe aortic stenosis in the fetus: potential for prevention of hypoplastic left heart syndrome: candidate selection, technique, and results of successful intervention. Circulation 2004; 110: 2125–31. 9. Makikallio K, McElhinney DB, Levine JC et al. Fetal aortic valve stenosis and the evolution of hypoplastic left heart syndrome: patient selection for fetal intervention. Circulation 2006; 113: 1401–5. 10. Rudolph AM. Congenital Diseases of the Heart. Armonk, NY: Futura, 2001. 11. Hutchins GM. Coarctation of the aorta explained as a branch-point of the ductus arteriosus. Am J Pathol 1971; 63: 203–14. 12. Santos MA, Moll JN, Drumond C et al. Development of the ductus arteriosus in right ventricular outflow tract obstruction. Circulation 1980; 62: 818–22. 13. Santos MA, Azevedo VM. Angiographic morphologic characteristics in pulmonary atresia with intact ventricular septum. Arq Bras Cardiol 2004; 82: 420–5. 14. Levin DL, Hyman AI, Heymann MA, Rudolph AM. Fetal hypertension and the development of increased pulmonary vascular smooth muscle: a possible mechanism for persistent pulmonary hypertension of the newborn infant. J Pediatr 1978; 92: 265–9. 15. Wild LM, Nickerson PA, Morin FC 3rd. Ligating the ductus arteriosus before birth remodels the pulmonary vasculature of the lamb. Pediatr Res 1989; 25: 251–7.

16. Abman SH, Shanley PF, Accurso FJ. Failure of postnatal adaptation of the pulmonary circulation after chronic intrauterine pulmonary hypertension in fetal lambs. J Clin Invest 1989; 83: 1849–58. 17. Heymann MA, Rudolph AM. Effects of acetylsalicylic acid on the ductus arteriosus and circulation in fetal lambs in utero. Circ Res 1976; 38: 418–22. 18. Turner GR, Levin DL. Prostaglandin synthesis inhibition in persistent pulmonary hypertension of the newborn. Clin Perinatol 1984; 11: 581–9. 19. Manzar S, Nair AK, Pai MG, Al-Khusaiby SM. Head size at birth in neonates with transposition of great arteries and hypoplastic left heart syndrome. Saudi Med J 2005; 26: 453–6. 20. Kaltman JR, Di H, Tian Z, Rychik J. Impact of congenital heart disease on cerebrovascular blood flow dynamics in the fetus. Ultrasound Obstet Gynecol 2005; 25: 32–6. 21. Ikle L, Hale K, Fashaw L, Boucek M, Rosenberg AA. Developmental outcome of patients with hypoplastic left heart syndrome treated with heart transplantation. J Pediatr 2003; 142: 20–5. 22. Dent CL, Spaeth JP, Jones BV et al. Brain magnetic resonance imaging abnormalities after the Norwood procedure using regional cerebral perfusion. J Thorac Cardiovasc Surg 2006; 131: 190–7. 23. Rebelo M, Veiga E, Machado AJ, Pinto F, Kaku S. Hypoplastic left heart syndrome with in utero closed foramen ovale: case report. Rev Port Cardiol 2006; 3: 331–6. 24. Vlahos AP, Lock JE, McElhinney DB, van der Velde ME. Hypoplastic left heart syndrome with intact or highly restrictive atrial septum: outcome after neonatal transcatheter atrial septostomy. Circulation 2004; 109: 2326–33. 25. Marshall AC, van der Velde ME, Tworetzky W et al. Creation of an atrial septal defect in utero for fetuses with hypoplastic left heart syndrome and intact or highly restrictive atrial septum. Circulation 2004; 110: 253–8. 26. Taketazu M, Barrea C, Smallhorn JF, Wilson GJ, Hornberger LK. Intrauterine pulmonary venous flow and restrictive foramen ovale in fetal hypoplastic left heart syndrome. J Am Coll Cardiol 2004; 43: 1902–7. 27. Rudolph AM. Congenital Diseases of the Heart. Chicago, IL: Year book, 1974: 466–8. 28. Maeno YV, Kamenir SA, Sinclair B et al. Prenatal features of ductus arteriosus constriction and restrictive foramen ovale in d-transposition of the great arteries. Circulation 1999; 99: 1209–14. 29. Jouannic JM, Gavard L, Fermont L et al. Sensitivity and specificity of prenatal features of physiological shunts to predict neonatal clinical status in transposition of the great arteries. Circulation 2004; 110: 1743–6. 30. Rudolph AM. Aortopulmonary transposition in the fetus: speculation on pathophysiology and therapy. Pediatr Res 2007; 61: 375–80. 31. Konduri GG, Gervasio CT, Theodorou AA. Role of adenosine triphosphate and adenosine in oxygen-induced pulmonary vasodilation in fetal lambs. Pediatr Res 1993; 33: 533–9. 32. Lewis AB, Heymann MA, Rudolph AM. Gestational changes in pulmonary vascular responses in fetal lambs in utero. Circ Res 1976; 39: 536–41. 33. Rasanen J, Wood DC, Debbs RH et al. Reactivity of the human fetal pulmonary circulation to maternal

Congenital cardiovascular malformations and the fetal and neonatal circulation

34.

35.

36.

37.

hyperoxygenation increases during the second half of pregnancy: a randomized study. Circulation 1998; 97: 257–62. McMurphy DM, Heymann MA, Rudolph AM, Melmon KL. Developmental changes in constriction of the ductus arteriosus: Responses to oxygen and vasoactive substances in the isolated ductus arteriosus of the fetal lamb. Pediatr Res 1972; 6: 231–8. Dick M 2nd, Heidelberger K, Crowley D, Rosenthal A, Hees P. Quantitative morphometric analysis of the pulmonary arteries in two patients with D-transposition of the great arteries and persistence of the fetal circulation. Pediatr Res 1981; 15: 1397–401. Cohn HE, Sacks EJ, Heymann MA, Rudolph AM. Cardiovascular responses to hypoxemia and acidemia in fetal lambs. Am J Obstet Gynecol 1974; 120: 817–24. van Bel F, Sola A, Roman C, Rudolph AM. Role of nitric oxide in the regulation of the cerebral circulation in the lamb

38.

39.

40.

41.

42.

595

fetus during normoxemia and hypoxemia. Biol Neonate 1995; 68: 200–10. Jouannic JM, Benachi A, Bonnet D et al. Middle cerebral artery Doppler in fetuses with transposition of the great arteries. Ultrasound Obstet Gynecol 2002; 20: 122–4. Fisher DJ, Heymann MA, Rudolph AM. Fetal myocardial oxygen and carbohydrate consumption during acutely induced hypoxemia. Am J Physiol 1982; 242: H647–61. Gewillig M, Boshoff DE, Dens J, Mertens L, Benson LN. Stenting the neonatal arterial duct in duct-dependent pulmonary circulation: newer techniques, better results. J Am Coll Cardiol 2004; 43: 107–12. Chan KC, Mashburn C, Boucek MM. Initial transcatheter palliation of hypoplastic left heart syndrome. Catheter Cardiovasc Interv 2006; 68: 719–26. Rudolph AM, Heymann MA, Spitznas U. Hemodynamic considerations in the development of narrowing of the aorta. Am J Cardiol 1972; 30: 514–25.

40 Twin–twin transfusion syndrome: impact on the cardiovascular system Jack Rychik Introduction The twin–twin transtusion syndrome (TTTS) is a disorder seen in approximately 20% of monochorionic twin gestations.1,2 The phenomenon is increasingly recognized as the most important contributor to morbidity and mortality in twin gestations. If left undetected and untreated, the natural history is such that the risk of death in at least one of the twins approaches 90–100%. Despite the significance of this problem, our understanding of the pathophysiology of the disease and the development of effective treatment strategies is still incomplete and evolving. One can make the argument that TTTS is primarily a circulatory derangement – in essence a failed partnership between the vascular systems of two fetuses sharing the same womb and the same placenta. This failed circulatory arrangement results in serious negative consequences and sets a course for cardiovascular decompensation that is lethal if left unchecked. In this chapter, we will review our current understanding of TTTS from the cardiovascular perspective, propose a score for characterizing the magnitude of cardiovascular perturbation seen in this disease, and discuss some of the questions that continue to challenge investigators and clinical practitioners who deal with this intriguing condition.

Diagnosis and findings in twin–twin transfusion syndrome A multiple gestation pregnancy places each of the individual partners at risk. These risks have been known for quite some time. A medieval painting from the Muiderslot castle in Holland circa 1617 depicts twin boys, reported to be the children of the mayor of Amsterdam, Jacob Dierkszon De Graeff.3 In the painting, one twin has a ruddy face and the other an unusually pale white face (Figure 40.1).

Figure 40.1 A painting of ‘De Wikkeldinderen (The Swaddled Children)’ depicting a likely case of twin–twin transfusion syndrome, dated 1617 (reproduced with permission from reference 3).

These infants likely suffered from TTTS; both died soon after birth. TTTS is suspected in the presence of monochorionic, diamniotic twins in which there is size discrepancy between the twins of at least 10–20% and in which there is oligohydramnios in the smaller and polyhydramnios in the larger fetus. Often the degree of oligohydramnios is so severe that the smaller twin appears to be ‘shrink-wrapped’ in its own amniotic sac, with the membrane tightly adherent to it (Figure 40.2). With the larger twin exhibiting polyhydramnios, the smaller twin is pushed to a corner of the

598

Fetal Cardiology

Myocardium

Amniotic membrane

Amniotic fluid

Recipient twin

Donor twin

Umbilical cord

Portion of placenta (chorion) Maternal portion of placenta

In order to grade the severity of disease and to allow for rational analysis of treatment strategies and prognosis, Quintero et al developed a staging system for TTTS.11 The system is based upon a number of variables: presence of polyhydramnios (maximum vertical pocket > 8 cm) in the larger twin and oligohydramnios (maximum vertical pocket of < 2 cm) in the smaller twin; presence or absence of visualization of a bladder in the smaller twin; presence or absence of critically abnormal Doppler studies defined as absent or reversed diastolic umbilical arterial flow, reverse flow in the ductus venosus, or umbilical venous pulsations; presence or absence of hydrops. The staging system is listed in Table 40.1. Although a number of criticisms of the system have been put forth, by and large the Quintero score has endured as a standard tool used in gauging the magnitude of disease present, and has been extensively applied in clinical trials of various treatment modalities. One limitation of the Quintero staging classification lies in its inability to discriminate between various degrees of cardiovascular derangement, with no consideration given to the absence, presence, or degree of recipient twin cardiomyopathy, as will be discussed below.

Figure 40.2 Cartoon drawing depicting the larger recipient and smaller donor twin in the twin–twin transfusion syndrome. Note the severe oligohydramnios associated with the ‘stuck’ donor twin (reproduced with permission from reference 4).

uterus, limiting its mobility – hence it is commonly referred to as the ‘stuck twin’. Premature labor as a consequence of the polyhydramnios is common. TTTS is to be distinguished from other causes of intertwin size discrepancy such as intrauterine growth restriction, or the presence of congenital, genetic, or chromosomal anomalies, or infection in the smaller twin. The impact of TTTS on the outcome of affected twins is considerable. TTTS can result in the demise of the larger or smaller twin,5 serious neurological insult,6 and cardiovascular abnormalities.7–10 Death of one twin in a monochorionic system can subsequently lead to the rapid death of the partner twin. A dead twin can act as a vascular lowresistance sink, which through anastomotic intraplacental connections leads to hypotension in the survivor. This ‘bleed’ into the vascular system of the dead fetus can cause either death or neurological damage in the partner twin. The fetal neurological system can be deleteriously impacted in TTTS in a variety of manners including by the disease process itself, by the death of a cotwin, or as a consequence of premature birth. Studies have demonstrated a high prevalence for morphological abnormalities on brain imaging and an increased prevalence of neurocognitive impairment in survivors. Cardiovascular manifestations with the potential for long-term consequences are common in TTTS and will be discussed below.

How and why does twin–twin transfusion syndrome occur? The name ‘twin–twin transfusion syndrome’ originates from postnatal experience with the disease. Twins born with marked size discrepancy are often identified as having significant differences in hemoglobin, with one manifesting polycythemia and the other anemia. This led to the belief that a simple in utero exchange of blood between the twins was the cause of this ailment. Current findings suggest that the pathophysiology is much more complex. Fetal blood sampling commonly demonstrates no significant difference in hemoglobin concentrations between twins with manifestations of TTTS, and hence a simple intertwin blood ‘transfusion’ is unlikely to be the sole cause.

Table 40.1 Quintero staging criteria for twin–twin transfusion syndrome (TTTS) Stage

Findings

I

Polyhydramnios/oligohydramnios sequence but with visible bladder in smaller twin

II

Absent bladder in smaller twin

III

Abnormal Doppler studies

IV

Hydrops fetalis

V

Demise of one or both twins

Twin–twin transfusion syndrome

The current thinking about the primary mechanism of TTTS is that it originates as a placental vasculopathy. In monochorionic twins, placental vascular connections exist between the two circulatory systems of each of the twins (Figure 40.3). These connections consist of arterial-toarterial (A–A) anastomoses, venovenous (V–V) anastomoses, or arterial-to-venous (A–V) anastomoses.12 At both A–A and V–V anastomoses, there is an even bidirectional exchange of blood volume between the circulations; however, exchange at A–V anastomoses is unidirectional, based on the pressure gradient. In the balanced state, whatever uneven exchange takes place between twins at A–V anastomoses is counterbalanced by equilibration at the A–A and, less often, V–V anastomoses. TTTS is believed to occur when there is a paucity of adequate A–A connections to allow for equilibration; hence, A–V connections predominate, resulting in a disequilibrium of placental flow between the twins.13 One twin becomes a ‘donor’ while the other is a ‘recipient’ of transplacental blood flow. Ultimately the donor begins to manifest hypovolemia, and as a consequence oliguria resulting in oligohydramnios, while the recipient manifests hypervolemia and polyhydramnios. Recent data looking at the renovascular systems of twins in TTTS have shed important light on this complex process.14 Immunohistochemistry studies have demonstrated the renin–angiotensin system (RAS) to be upregulated in the donor twin. This makes sense, as it is the expected natural response to hypovolemia. The effects of angiotensin II released by the RAS in response to hypovolemia are to increase vasoconstriction and water retention in the donor and thereby promote maintenance of perfusion

pressure. Partner recipient twins have been identified as having downregulation of their renin–angiotensin system likely as a consequence of the hypervolemia, yet serum levels of RAS hormones such as renin are similar to those found in the donors. This suggests that RAS hormones found in the recipient are not intrinsically produced but in fact produced by the donor, and delivered to the recipient through vascular connections. Hence, not only does the recipient receive an increase in volume but also an increase in hormonal modulators that are typically released in response to low volume. These agents act as vasoconstrictors within the recipient, resulting in an increase in vascular resistance. In addition, other agents such as brain natriuretic peptide15 and endothelin,16 hormonal modulators released in the presence of heart failure, are elevated in the amniotic fluid of recipient twins with hydrops. Hence, there is a double insult that takes place – an increase in preload and a paradoxical increase in afterload – which, in combination, acts deleteriously upon the recipient cardiovascular system and is the cause of the cardiomyopathic changes seen in the recipient in TTTS.

Impact on the cardiovascular system and echocardiographic findings in twin–twin transfusion syndrome Important changes take place in the cardiovascular system of fetuses in TTTS, with the potential for a wide spectrum of findings7–10 (Figure 40.4–40.11). The donor heart rarely manifests cardiac abnormalities on echocardiography, as the myocardium deals with a decrease in preload and an increase in afterload quite well. The systolic performance of the donor heart is preserved, with no effect on valvular function.

Figure 40.3 Diagram depicting the possible vascular connections between twins in a monochorionic system. Unidirectional flow occurs between A–V connections, with equilibration of flow occurring at V–V and A–A connections. The hatched line represents the vascular equator between the two fetal circulations. A, arterial; V, venous.

599

Figure 40.4 Ascites in a recipient twin with severe cardiomyopathy.

600

Fetal Cardiology

Figure 40.5 Dilated atria in a recipient twin with severe cardiomyopathy. The atrial portion of the heart nearly fills the entire diameter of the chest. The atria are enlarged due to the diastolic dysfunction and poor compliance of the ventricles as well as the presence of significant atrioventricular valve regurgitation.

Ventricular cavity size and overall heart size are usually smaller than normal. Since the donor experiences an overall increase in systemic vascular resistance, analysis of placental vascular resistance by Doppler interrogation of the umbilical artery reveals an increase in the pulsatility index. This can be identified qualitatively by observing a low diastolic flow velocity or even absent or reversed flow at end-diastole in the spectral tracing of the umbilical artery. These findings suggest very high placental vascular resistance. The recipient heart bears the brunt of the disease in TTTS. At first, and at an early stage, one may identify ventricular cavity dilatation and mild ventricular hypertrophy. Mild atrioventricular valve regurgitation, likely as a consequence of cavity dilatation, can be seen. Doppler flow patterns in the umbilical artery and vein are normal, and both systolic and diastolic ventricular performance is preserved. As the disease progresses, ventricular thickening progresses with a consequential effect on ventricular compliance and diastolic function. Of note, it has been observed that it is the right ventricle more so than left which typically manifests the majority of change in TTTS. As the ventricle hypertrophies, it becomes less compliant, and Doppler parameters of ventricular filling begin to change. The normal twin peaks of inflow into the ventricle relating to passive filling (E-wave) and active atrial filling (A-wave) fuse into a single peak. Ductus venosus flow during atrial contraction diminishes, often becoming absent or reversed in conditions of a very stiff, non-compliant right ventricle. Finally, further upstream, the finding of pulsations in the umbilical vein can be seen, reflecting a severe degree of impediment to ventricular filling. As diastolic dysfunction progresses, ventricular systolic function is affected as well. Systolic ventricular dysfunction can be observed as a decrease in ventricular shortening fraction

and a worsening of atrioventricular valve regurgitation, initially on the right and then progressing to the left side of the heart. Ultimately severe ventricular dysfunction and severe atrioventricular insufficiency lead to low cardiac output, development of hydrops, and fetal death. Barrea et al reviewed 28 twin pairs affected by TTTS and found cardiomegaly due to right or left ventricular hypertrophy in 58% of recipient twins.10 Diastolic dysfunction of both right and left ventricles was present in two-thirds, and right ventricular systolic dysfunction with significant tricuspid regurgitation was seen in one-third. Progression of findings was associated with a higher perinatal mortality. Right ventricular pressure estimates from the peak velocity of the tricuspid regurgitant jet in the recipient with TTTS will typically demonstrate very high intracavitary values. This finding offers further support to the notion that the recipient twin heart is stressed by the presence of increased vascular resistance and increased afterload. The true blood pressure and ventricular cavity pressure in the human fetus is unknown. However, the typical peak systolic pressure for a newborn premature infant born at 24 weeks’ gestation is known to be approximately 30– 40 mmHg. The systolic pressure in the normal fetal heart in series with the low vascular resistance placenta should therefore be perhaps less, but certainly no higher, than this value. We have observed right ventricular pressures as high as 80–90 mmHg in the absence of any outflow tract obstruction, supporting the notion that an increase in vascular resistance is part of the pathophysiological process resulting in the cardiomyopathy of TTTS seen in the recipient twin. One fascinating phenomenon seen in TTTS is that of progressive development of right ventricular outflow tract obstruction in select recipient fetuses.9 This ‘acquired’ right ventricular outflow tract obstruction is phenotypically identical to the congenital heart defect of pulmonary atresia or pulmonary stenosis with a hypoplastic, hypertrophied right ventricle. This phenomenon of development of selective right ventricle outflow tract obstruction in otherwise structurally normal hearts begs the question of whether recipient TTTS cardiac changes may offer a clue as to the development of some forms of congenital heart disease. The process of TTTS demonstrates the plasticity of the fetal right ventricle, in that a structurally normal heart is dramatically affected by alterations in extrinsic loading conditions and hormonal modulators, these contributed by the cotwin donor. It is enticing to speculate that perhaps a similar mechanism of hormonal conditions and altered load very early in gestation may be the cause of right-sided obstructive anomalies in the singleton fetus born with these forms of congenital heart disease.17 This relates to the fundamental mechanisms of formation of congenital heart disease. Alterations in blood flow and other extrinsic variables may potentially lead to the development of ‘acquired–congenital’ heart disease at a period of time following completion of embryological formation of the human heart.

Twin–twin transfusion syndrome

(a)

601

(b)

(c)

Figure 40.6

(a) Four-chamber view of the heart in a recipient twin. Note the heart size which in cross-sectional area exceeds 50% of the chest area. (b) Color Doppler echocardiography demonstrates severe tricuspid and mitral valve regurgitation. (c) The right ventricular pressure estimate from the peak velocity of the spectral Doppler display is elevated at approximately 50 mmHg. PG, peak gradient; V, peak velocity.

Measures to assess the burden of cardiovascular impairment in twin–twin transfusion syndrome Gauging the magnitude of cardiovascular derangement in TTTS is important. One such measure of the magnitude of ventricular dysfunction is the myocardial performance index (MPI).18 The MPI is a ventricular geometry-independent

measure of combined systolic and diastolic ventricular performance. By performing Doppler sampling of inflow and outflow across the ventricle, one can measure the time intervals relating to isovolumic contraction, isovolumic relaxation, and ventricular ejection. The MPI is a measure of the ratio of combined isovolumic times to the ejection time. As global systolic and diastolic dysfunction worsens, MPI values increase. Raboisson and colleagues used the MPI to assess the ventricular performance status of recipient and donor twins with TTTS.19 Recipient twins exhibited higher MPI values relative to their donor partners.

602

Fetal Cardiology

Figure 40.7 Demonstration of a significant jet of pulmonary insufficiency in a recipient twin–twin transfusion syndrome fetus. Ao, aorta; MPA, main pulmonary artery; Pulm Insuff, pulmonary valve insufficiency.

Figure 40.9 Doppler sampling in the ductus venosus of a donor twin (a) and recipient twin (b). The donor has a normal Doppler signal with continuous but phasic forward flow in the ductus venosus. The recipient Doppler signal displays reversal of flow with atrial contraction suggesting a poorly compliant, and stiff right ventricle.

Figure 40.8 Doppler sampling in the umbilical cords of a donor twin (a) and a recipient twin (b). Umbilical arterial flow (pulsatile) is above the baseline, while umbilical venous flow is depicted below the baseline. Note the diminution in diastolic velocity in the donor umbilical artery flow as compared to the recipient. The arrows point out reversal of flow in the donor twin (below the baseline) suggesting markedly elevated placental vascular resistance.

The authors also found that ventricular dysfunction in the recipient was so characteristic of TTTS that one could reliably use the MPI to distinguish between TTTS and other causes of twin size discrepancy such as intrauterine growth restriction. We have been investigating the utility of various Dopplerderived measures of ventricular performance in order to help us improve our understanding of the pathophysiology of TTTS as well as help stratify patients by severity of disease. Szwast et al applied the MPI, as well as calculating the ventricular ejection force and cardiac output, in 22 twin pairs with TTTS.20 The ventricular ejection force describes the acceleration of blood across the pulmonary or aortic valve over a specific time interval, and is a reflection of systolic ventricular performance derived from Newton’s laws. A higher value corresponds to a greater force exerted in ejecting the ventricular volume of blood during systole. The combined right and left ventricle cardiac output (CCO) is a measure of total blood flow through the fetal heart, and is indexed to the estimated fetal weight in kilograms. The twin pairs were then compared to 36 age-matched singleton fetuses as normal controls. The findings are listed in Table 40.2. In the donor twins, right

Twin–twin transfusion syndrome

603

Figure 40.10

Figure 40.11

Doppler interrogation of flow across the mitral valve in a donor twin (a) and a recipient twin (b). The donor twin has a normal ‘double-peak’ inflow pattern (solid arrows) representing flow during passive diastolic filling (first peak) and during atrial contraction (second peak). The recipient twin has a ‘singlepeak’ inflow pattern (open arrow) suggesting fusion of the diastolic phases into a single phase. This represents a poorly compliant, and stiff ventricle. Note that often tachycardia can cause fusion of the normal double inflow peaks into a single peak; however, as can be seen in this case, the heart rate for the recipient is nearly identical to that of the donor twin.

Doppler interrogation of flow across the tricuspid valve in a donor twin (a) and a recipient twin (b). The donor has a normal ‘double-peak’ pattern (solid arrows) while the recipient has a ‘single-peak’ inflow pattern (open arrow).

ventricle (RV) and left ventricle (LV) MPI values were lower than in the recipient twins, and even lower than in normal control fetuses. Donor twins also had diminished RV and LV ejection forces compared to recipient twins, and diminished RV and LV ejection forces compared to normal control fetuses. Donor twins had lower CCO compared to recipient twins as well as compared to normal control fetuses. These findings are consistent with the belief that donor twins are volume depleted, but have preserved myocardial function. In contrast, the recipient twins had abnormally elevated RV and LV MPI compared to normal fetuses, although there was no significant difference between LV and RV ejection forces in recipient twins compared to normal fetuses. Finally, recipient twins had elevated CCO compared to their donor counterparts and compared to normal controls. This suggests that in our group of recipient twins, systolic function was still preserved, as ejection forces were increased while diastolic dysfunction was present. Application of these Doppler-derived parameters will help to identify fetuses with diastolic

dysfunction, prior to the onset of low cardiac output and hydrops, and may therefore be helpful in grading the magnitude of disease as the twin–twin transfusion process progresses. An important goal has been to provide a means for quantifying the cardiovascular burden present in TTTS by developing a cardiovascular score that may be used as an adjunct to the more general Quintero classification system. To that end, we have identified cardiovascular features which may be considered in such a score.21 Table 40.3 lists these features and the values given to the various findings of the Children’s Hospital of Philadelphia (CHOP) Cardiovascular Score for TTTS. The CHOP Cardiovascular Score is derived from echocardiographic data of 150 twin pairs referred for TTTS and incorporates features describing ventricular dilatation and hypertrophy, systolic function, valve regurgitation, and diastolic properties as described by Doppler indicators of ventricular compliance in the recipient twin. Umbilical arterial diastolic flow is considered in the donor twin. While right ventricular outflow tract obstruction is an end result in some cases of recipient cardiomyopathy in TTTS, we have observed situations in which there is reversed-size discrepancy in between the pulmonary artery and the aorta. Whereas the pulmonary artery diameter is normally larger than the aorta by approximately 25%, some

604

Fetal Cardiology

Table 40.2 Results for the twin-donor vs twin-recipient vs normal control are summarized. The subgroup of the normal control population with similar gestational age is used for statistical analysis. p values are sequentially recorded as p1, twin-donor vs twin-recipient; p2, twin-donor vs normal; and p3, twin-recipient vs normal (from reference 20) Variable

Twin-donor (n = 22)

Twin-recipient (n = 22)

Normal (n = 36)

p1

p2

p3

0.42

0.42

GA (weeks)

22.3 ± 2.3

22.3 ± 2.3

22.9 ± 2.0

Fetal weight (kg)

0.42 ± 0.22

0.56 ± 0.21

0.64 ± 0.27

< 0.001

< 0.05

0.23

RV MPI

0.38 ± 0.07

0.56 ± 0.09

0.42 ± 0.05

< 0.001

< 0.05

< 0.001

LV MPI

1

0.35 ± 0.07

0.54 ± 0.12

0.41 ± 0.05

< 0.001

< 0.05

< 0.001

RV EF (mN)

2.5 ± 1.5

6.1 ± 4.0

5.4 ± 3.3

< 0.001

< 0.001

0.53

LV EF (mN)

2.0 ± 1.7

5.7 ± 3.2

4.6 ± 2.0

< 0.001

< 0.001

0.19

CCO (ml/min/kg)

416 ± 74

568 ± 109

506 ± 86

< 0.001

< 0.001

< 0.05

GA, gestational age; RV, right ventricle; LV, left ventricle; MPI, myocardial performance index; EF, ejection force; CCO, combined cardiac output.

recipient fetuses exhibit an equal diameter between the pulmonary artery and aorta, or a pulmonary artery that is smaller than the aorta. We believe that this may reflect the spectrum of the disease, prior to the more severe finding of right ventricle outflow tract obstruction and pulmonary atresia. Altered compliance of the right ventricle can result in increased shunting away from the right side across the foramen ovale to the left ventricle. In such a circumstance, with diminished flow on the right – but increased flow on the left – pulmonary artery growth is inhibited while aortic growth is enhanced. We have therefore included abnormal size discrepancy between the pulmonary artery and aorta in our CHOP Cardiovascular TTTS Score. The maximum score suggesting the most severe degree of cardiovascular impairment is 20/20. Recently, we have applied this score at the Children’s Hospital of Philadelphia in prospectively monitoring all of our fetuses with TTTS. In addition to the score, which is derived from qualitative parameters, we have been routinely measuring MPI values in the recipient and donor twins as a supplemental means of assessing myocardial function. Table 40.4 describes the findings in 150 twin pairs with TTTS with respect to the Doppler-derived measures of umbilical artery pulsatility index, middle cerebral artery index, ductus venosus flow, MPI for right and left ventricles, and the parameters used to calculate the MPI values. It is only through just such a detailed and comprehensive characterization of the recipient and donor twins that one can appropriately gauge the effect of this disease, and the response of the cardiovascular system to various treatment strategies. Quantifying the degree of cardiovascular abnormality in the fetus with TTTS can also potentially be useful as a prognosticator of residual cardiovascular abnormalities after birth. These techniques will require further validation and analysis in a wide variety of cases. Such investigations are currently under way.

Treatment strategies for twin– twin transfusion syndrome A variety of treatment strategies have been proposed for TTTS. The serial removal of large volumes of amniotic fluid from the recipient – amnioreduction – has been demonstrated to reduce morbidity and mortality.22 Initially employed as a means of creating comfort for the mother, it soon became apparent that serial amnioreduction exerts a positive effect on fetal outcome as well. This may be related to improved placental circulation following relief of placental compression as well as prevention of preterm labor by reduction in abdominal/uterine size. Creation of a communication between the donor and recipient amniotic sacs, either intentionally or during amnioreduction procedures, has also been demonstrated to result in improved outcomes. Equilibration of amniotic volumes and pressure between the twins by creating a ‘microseptostomy’ can help; however, these communications commonly seal off, making this only a temporary measure. Ville and others pioneered the concept of direct interruption of the guilty vascular anastomoses within the placental plate by percutaneous laparoscopic laser techniques.23–26 Currently, this appears to be the most effective strategy for treating TTTS. In a large randomized trial of endoscopic laser photocoagulation therapy versus serial amnioreduction (The Eurofoetus Consortium Trial), laser therapy was identified as being superior as a first line treatment at all stages of severity of the disease.26 As compared to the amnioreduction group, the laser group had a higher likelihood of survival of at least one twin (76% vs 56%), had a lower incidence of periventricular leukomalacia (6% vs 14%), and were more likely

Twin–twin transfusion syndrome

605

Table 40.3 CHOP Cardiovascular Score for characterizing the severity of cardiovascular perturbation in the twin–twin transfusion syndrome (from reference 21) Parameter

Finding

Numerical score

Normal

0

Decreased diastolic flow

1

Recipient

Absent/reversed diastolic

2

Ventricular hypertrophy

None

0

Present

1

None

0

Mild

1

> Mild

2

None

0

Mild

1

> Mild

2

None

0

Mild

1

> Mild

2

None

0

Mild

1

> Mild

2

Double-peak

0

Single-peak

1

Double-peak

0

Single-peak

1

All antegrade

0

Absent diastolic flow

1

Reverse diastolic flow

2

No pulsations

0

Pulsations

1

PA > Ao

0

PA = Ao

1

PA < Ao

2

RV outflow obstruction

3

None

0

Donor Umbilical artery

Cardiac dilatation

Ventricular dysfunction

Tricuspid valve regurgitation

Mitral valve regurgitation

Tricuspid valve inflow Mitral valve inflow Ductus venosus

Umbilical vein Right-sided outflow tract

Pulmonary regurgitation

Present Maximum total cardiovascular score

1 20 points

PA, pulmonary artery; Ao, aorta; RV, right ventricle.

to be free of neurological complications at 6 months of age (52% vs 31%). Although laser therapy currently appears to be the more effective strategy, it is not a cure for the disease in all cases, as significant morbidity and

mortality in a large number clearly persist despite treatment. Further work needs to be done in determining the appropriate candidates and the optimal timing for laser therapy.

606

Fetal Cardiology

Table 40.4 Quantitative parameters of cardiovascular function evaluated in both the donor and recipient fetuses in 150 twin pairs with TTTS (from reference 21) Donor

Recipient

p value

Weight (g)

399 (235)

527 (285)

< 0.0001

UA S-wave peak velocity (cm/s)

29.2 (8.7)

41.7 (12.9)

< 0.0001

UA D-wave peak velocity (cm/s)

4.7 (4.3)

9.7 (4.5)

< 0.0001

UA PI

1.81 (0.8)

1.46 (0.4)

< 0.0001

MCA S-wave peak velocity (cm/s)

27.3 (8.1)

26.2 (8)

MCA D-wave peak velocity (cm/s)

5.1 (2.6)

5.1 (2.1)

NS

MCA PI

1.79 (0.4)

1.70 (0.4)

NS

DV A-wave peak velocity (cm/s)

18.5 (6.8)

13.3 (16.7)

< 0.01

DV S-wave peak velocity (cm/s)

51.1 (13)

53.3 (13.6)

NS

DV A/S ratio

0.38 (0.12)

0.24 (0.31)

< 0.001

TV closure-to-opening time (ms)

232 (18)

275 (32)

< 0.0001

PA ejection time (ms)

169 (13)

169 (21)

NS

RV MPI

0.38 (0.11)

0.69 (0.47)

< 0.0001

MV closure-to-opening time (ms)

222 (19)

263 (29)

< 0.0001

Ao ejection time (ms)

166 (15)

170 (17)

0.07

LV MPI

0.34 (0.12)

0.59 (0.28)

NS

< 0.0001

Ao, aorta; DV, ductus venosus; LV, left ventricle; MCA, middle cerebral artery; MPI, myocardial performance index (Tei index); MV, mitral valve; PA, pulmonary artery; RV, right ventricle; TV, tricuspid valve; UA, umbilical artery; NS, not significant.

Questions, speculations and long-term issues There are a number of unanswered questions concerning TTTS and the cardiovascular changes that take place. The outcome studies looking at therapeutic effectiveness have focused on perinatal survival and neurological outcome, with little focus on cardiovascular outcome. Which therapeutic interventions are most likely to result in cardiovascular improvement? Barrea and colleagues demonstrated that, despite amnioreduction, cardiovascular changes may persist or even progress.10 The impact of laser therapy on cardiovascular changes and the ability to prevent progression of right ventricular outflow tract obstruction, cardiac failure, or hydrops have not yet been investigated. It is through the application of descriptive cardiovascular tools such as the CHOP Cardiovascular Score and other quantitative indexes of function that such questions may be addressed. Intriguing questions remain concerning the pathophysiology of TTTS. Why do some severely affected recipient twins manifest ventricular dysfunction and heart failure while others develop ventricular hypertrophy and right ventricular outflow tract obstruction? We propose the possibility that perhaps it is not the severity of the disease alone that dictates this phenotype, but that inherent genetic

factors may play a role. It is well recognized in the mature postnatal heart that the myocardial response to stressors is, to a degree, under genetic control.27 It is therefore plausible to speculate that the fetal myocardial response to a severe increase in preload (volume exchange) and afterload (hormonal modulators) may similarly be under genetic control – for some twin recipients the response to these stressors is cardiac hypertrophy, decompensation, and failure, while for others it is hypertrophy and development of right ventricular outflow tract obstruction. In fact, the latter may be a positive adaptive response in those that have the ability to muster it, since few fetuses with right ventricle outflow tract obstruction experience fetal demise, unlike the recipients who develop cardiac decompensation and hydrops. Perhaps variables such as angiotensin converting enzyme genotype or other myocardial genotypes may influence the phenotypic direction that the recipient fetus will take. Angiotensin II can be a potent stimulant of myocardial hypertrophy in the mature heart; however, its direct myocardial affects are under complex genotypic control (i.e receptor type, density, and so on).27–30 Might the phenotypic variability seen in the recipient twin in TTTS be a result of a genetically controlled inherent response to elevated levels of angiotensin II? The genetic determinants of myocardial hypertrophy in the fetus are currently unknown, but will likely explain this phenomenon.

Twin–twin transfusion syndrome

The long-term implications of cardiovascular disease in the fetus with TTTS are potentially quite significant. The ‘Barker hypothesis’ proposes the concept of fetal origins of adult cardiovascular disease – ‘programming’ of the heart and vascular system takes place during fetal life, setting the risk for the development of late diseases such as hypertension and atherosclerosis.31 Cheung and colleagues demonstrated the impact of fetal TTTS on vascular dysfunction in infancy.32 TTTS donor twins at a mean of 9 months of age exhibited diminished arterial distensibility as measured by abnormal pulse-wave velocity examination. Gardiner and colleagues studied 27 twin pairs with TTTS after birth at a mean of 11 months of age, and found that laser therapy altered but did not abolish these vascular abnormalities of arterial distensibility.33 Are the survivors of fetal TTTS, at risk for significant cardiovascular difficulties as they grow into later childhood and into adults? What is the longterm cardiovascular burden of TTTS and to what degree do these patients carry added risk for cardiovascular disease as they grow into adulthood? These are intriguing questions yet to be answered.

Summary TTTS is a unique and complex disease process affecting monochorionic/diamniotic twins. The disease is primarily due to a placental vasculopathy with resultant transfer of both volume and hormonal modulators through intraplacental vascular connections from the donor to the recipient twin. Morbidity and mortality remain high despite the development of treatment strategies such as endoscopic laser photocoagulation. Complex changes take place within the cardiovascular system which contribute substantially to the outcome of this disease. The application of tools to better grade the degree of cardiovascular derangement will help in determining the efficacy of current and to-bedeveloped treatment strategies. The long-term impact of TTTS on the postnatal and mature adult cardiovascular system, although likely of significance, is yet to be fully understood.

References 1. Sebire NJ, Snijders RJ, Hughes K, Sepulveda W, Nicolaides KH. The hidden mortality of monochorionic twin pregnancies. Br J Obstet Gynaecol 1997; 104: 1203–7. 2. Harkness UF, Crombleholme TM. Twin-twin transfusion syndrome: where do we go from here? Semin Perinatol 2005; 29: 296–304. 3. Berger HM, de Waard F, Molenaar Y. A case of twin-to-twin transfusion in 1617. Lancet 2000; 356: 847–8. 4. Fisk NM, Galea P. Twin-twin transfusion—as good as it gets? N Engl J Med 2004; 351: 182–4.

607

5. Gonsoulin W, Moise KJ Jr, Kirshon B et al. Outcome of twin-twin transfusion diagnosed before 28 weeks of gestation. Obstet Gynecol 1990; 75: 214–16. 6. Haverkamp F, Lex C, Hanisch C, Fahnenstich H, Zerres K. Neurodevelopmental risks in twin-to-twin transfusion syndrome: preliminary findings. Eur J Paediatr Neurol 2001; 5: 21–7. 7. Zosmer N, Bajoria R, Weiner E et al. Clinical and echographic features of in utero cardiac dysfunction in the recipient twin in twin-twin transfusion syndrome. Br Heart J 1994; 72: 74–9. 8. Simpson LL, Marx GR, Elkadry EA, D’Alton ME. Cardiac dysfunction in twin-twin transfusion syndrome: a prospective, longitudinal study. Obstet Gynecol 1998; 92: 557–62. 9. Lougheed J, Sinclair BG, Fung Kee Fung K et al. Acquired right ventricular outflow tract obstruction in the recipient twin in twin-twin transfusion syndrome. J Am Coll Cardiol 2001; 38: 1533–8. 10. Barrea C, Alkazaleh F, Ryan G et al. Prenatal cardiovascular manifestations in the twin-to-twin transfusion syndrome recipients and the impact of therapeutic amnioreduction. Am J Obstet Gynecol 2005; 192: 892–902. 11. Quintero RA, Morales WJ, Allen MH et al. Staging of twintwin transfusion syndrome. J Perinatol 1999; 19: 550–5. 12. Bajoria R, Wigglesworth J, Fisk NM. Angioarchitecture of monochorionic placentas in relation to the twin-twin transfusion syndrome. Am J Obstet Gynecol 1995; 172: 856–63. 13. Galea P, Jain V, Fisk NM. Insights into the pathophysiology of twin-twin transfusion syndrome. Prenat Diagn 2005; 25: 777–85. 14. Mahieu-Caputo D, Meulemans A, Martinovic J et al. Paradoxic activation of the renin-angiotensin system in twin-twin transfusion syndrome: an explanation for cardiovascular disturbances in the recipient. Pediatr Res 2005; 58: 685–8. 15. Bajoria R, Ward S, Chatterjee R. Natriuretic peptides in the pathogenesis of cardiac dysfunction in the recipient fetus of twin-twin transfusion syndrome. Am J Obstet Gynecol 2002; 186: 121–7. 16. Bajoria R, Sullivan M, Fisk NM. Endothelin concentrations in monochorionic twins with severe twin-twin transfusion syndrome. Hum Reprod 1999; 14: 1614–18. 17. Clark, EB, Hu N, Frommelt P et al. Effect of increased pressure on growth in stage 21 chick embryos. Am J Physiol Heart Circ Physiol 1989; 257: H55–61. 18. Eidem BW, Edwards JM, Cetta F. Quantitative assessment of fetal ventricular function: establishing normal values of the myocardial performance index in the fetus. Echocardiography 2001; 18: 9–13. 19. Raboisson MJ, Fouron JC, Lamoureux J et al. Early intertwin differences in myocardial performance during the twin-totwin transfusion syndrome. Circulation 2004; 110: 3043–8. 20. Szwast A, Tian Z, McCann M et al. Impact of altered loading conditions on ventricular performance in fetuses with congenital cystic adenomatoid malformation and twin-twin transfusion syndrome. Ultrasound Obstet Gynecol 2007; 30: 40–6. 21. Rychik J, Tian Z, Bebbington M et al. The twin-twin transfusion syndrome: spectrum of cardiovascular abnormality and development of a cardiovascular score to assess severity of disease. Am J Obstet Gynecol 2007; 197: 392.e1–8.

608

Fetal Cardiology

22. Mari G, Roberts A, Detti L et al. Perinatal morbidity and mortality rates in severe twin-twin transfusion syndrome: results of the International Amnioreduction Registry. Am J Obstet Gynecol 2001; 185: 708–15. 23. Ville Y, Hyett J, Hecher K, Nicolaides K. Preliminary experience with endoscopic laser surgery for severe twintwin transfusion syndrome. N Engl J Med 1995; 332: 224–7. 24. De Lia JE, Kuhlmann RS, Lopez KP. Treating previable twin-twin transfusion syndrome with fetoscopic laser surgery: outcomes following the learning curve. J Perinat Med 1999; 27: 61–7. 25. Hecher K, Diehl W, Zikulnig L, Vetter M, Hackeloer BJ. Endoscopic laser coagulation of placental anastomoses in 200 pregnancies with severe mid-trimester twin-to-twin transfusion syndrome. Eur J Obstet Gynecol Reprod Biol 2000; 92: 135–9. 26. Senat MV, Deprest J, Boulvain M et al. Endoscopic laser surgery versus serial amnioreduction for severe twin-to-twin transfusion syndrome. N Engl J Med 2004; 351: 136–44. 27. Lips DJ, deWindt LJ, van Kraaij DJ, Doevendans PA. Molecular determinants of myocardial hypertrophy and

28.

29.

30.

31. 32.

33.

failure: alternative pathways for beneficial and maladaptive hypertrophy. Eur Heart J 2003; 24: 883–96. Yamazaki T, Komuro I, Yazaki Y. Role of the renin–angiotensin system in cardiac hypertrophy. Am J Cardiol 1999; 83: 53H– 7H. Sadoshima J, Izumo S. Molecular characterization of angiotensin II-induced hypertrophy of cardiac myocytes and hyperplasia of cardiac fibroblasts. Critical role of the AT1 receptor subtype. Circ Res 1993; 73: 413–23. Sadoshima J, Xu Y, Slayter HS et al. Autocrine release of angiotensin II mediates stretch-induced hypertrophy of cardiac myocytes in vitro. Cell 1993; 75: 977–84. Barker DJ. Fetal origins of cardiovascular disease. Ann Med 1999; 31 (Suppl 1): 3–6. Cheung YF, Taylor MJ, Fisk NM, Redington AN, Gardiner HM. Fetal origins of reduced arterial distensibility in the donor twin in twin-twin transfusion syndrome. Lancet 2000; 355: 1157–8. Gardiner HM, Taylor MJ, Karatza A et al. Twin-twin transfusion syndrome: the influence of intrauterine laser photocoagulation on arterial distensibility in childhood. Circulation 2003; 107: 1906–11.

41 Genetics and cardiac anomalies Eran Bornstein, David Seubert, and Mark I Evans Introduction Fetal cardiac anomalies are the most common type of congenital anomalies and occur in as many as 5–10/1000 live births. It is estimated that about 40 000 live births annually are complicated with a cardiac anomaly in the United States. Cardiac anomalies can present as an isolated finding or as part of a syndrome, manifesting a non-random occurrence of several malformations. The goal of the diagnostician, whether pediatric or perinatologist, is to distinguish a normal from an abnormal fetal heart, identify specific cardiac defects that may present, and search for associated extracardiac anomalies. In addition, the sonographic findings should be categorized as either a syndrome involving congenital heart disease, or as a primary, isolated cardiac anomaly. This distinction is important, as the implications of the two may vary significantly. Since chromosomal abnormalities commonly have phenotypic cardiac manifestations, genetic counseling as well as offering diagnostic genetic testing for specific disorders might be indicated. A thorough medical history plays an important role in identifying patients who are at increased risk for the development of congenital heart anomalies secondary to maternal disease, exposure to known cardiac teratogens, infectious etiology, and familial history suggestive of a genetic predisposition. For example, patients with diabetes mellitus have a markedly increased risk for congenital cardiac anomalies. Data from the Collaborative Perinatal Project in the 1970s showed that the infant of a diabetic mother has a 25.4/1000 risk of congenital cardiac anomalies, compared with an 8.1/1000 rate in non-diabetic mothers.1 Several studies have linked the increase in congenital malformations with poor preconception glycemic control.2 Hyperglycemia during organogenesis (5–8 weeks of gestation) has a major impact on abnormal development.3 A clinically useful marker of glycemic control over a period of 6–8 weeks is glycosylated hemoglobin (HbAlc), and its level has been directly correlated with the frequency of anomalies. Studies have shown that an HbA1c level of 5–6% is associated with a fetal malformation rate similar to that one of the general population, whereas a

level of 10% confers a 10-fold increase in the risk for congenital anomalies.4 Cardiac teratogenic effect has been documented following exposure to several drugs and medications. Among these teratogens are alcohol and isotretinoin (Accutane®). Fetal alcohol syndrome (FAS) is a condition with characteristic phenotypic manifestations that includes hypotonia, low birth weight, microcephaly, mental retardation, and typical facies. Fifty percent of these infants have associated congenital cardiac anomalies, which commonly include ventricular septal defects (VSDs), tetralogy of Fallot (TOF), and atrial septal defects (ASDs).5 Isotretinoin is a known potent teratogen that affects as many as 40% of exposed fetuses, and therefore should be used with appropriate contraception only.6 Unlike cases of maternal diabetes mellitus where a direct correlation exists between poor glycemic control and the presence of cardiac malformations, in FAS only a loose correlation between the degree of fetal exposure and the likelihood of anatomical abnormalities was detected.7,8 Infants of mothers with phenylketonuria (PKU) also have an increased risk of congenital heart anomalies. The National Maternal PKU Collaborative Study showed that 8.7% (6/69) of women with this condition and phenylalanine levels in excess of 10 mg/dl had fetuses with congenital heart anomalies.9 Rouse et al documented an increase in the frequencies of congenital abnormalities with increasing maternal phenylalanine levels. They therefore support the concept that women with PKU should begin a low-phenylalanine diet to achieve phenylalanine levels of < 360 μmol/l prior to conception and maintain these levels throughout pregnancy.10,11 A thorough personal and familial history regarding congenital heart defects should be obtained, focusing on previous children and other first-degree relatives. It is important to note that, while congenital heart defects in a first-degree relative increase the risk in the proband, the type of defect need not be the same condition as in the relative. Therefore, a mother with a congenital cardiac malformation may give birth to an infant with a different anomaly. As the mode of inheritance differs between different anomalies, the inherited risk depends on the

610

Fetal Cardiology

specific anomaly and the specific first-degree relative who is affected.12 As mentioned earlier, congenital cardiac malformation may be isolated or part of a congenital syndrome. A regional case–control study of the Baltimore–Washington area (Baltimore–Washington Infant Study) found that 73% of infants with a congenital cardiac malformation had an isolated lesion, whereas in 27% of these fetuses concurrent anomalies were detected in other systems.13 Copel et al found that in 7% of fetuses who had been initially diagnosed with extracardiac anomalies, further sonographic evaluation revealed additional cardiac malformations.14 Thus, the finding of an extracardiac anomaly should prompt the clinician to further investigate the fetal heart. An important issue that needs to be addressed is the likelihood of a chromosomal anomaly associated with fetal cardiac defects. The decision whether to proceed with invasive genetic testing is not easy in many cases. The specific cardiac defect, the presence of additional extracardiac malformations, and the association of the specific anomaly with known syndromes should all be taken into consideration before performing invasive testing. Based on epidemiological data in newborns, the association of congenital heart defects with chromosomal anomalies varies between 4 and 12%. Prenatally, this rate is probably higher due to early antenatal death that occurs in fetuses with chromosomal aberrations.15 The incidence of a chromosomal anomaly varies significantly in the literature. Copel et al reported an incidence as high as 32% of aneuploidy in fetuses with congenital heart defects, whereas Allan et al reported a lower rate of 16%.16,17 Additionally, Chaoui et al reported a 22% incidence of chromosomal anomaly in fetuses with diagnosed cardiac malformation. They concluded that in the presence of a fetal cardiac anomaly, offering invasive testing for fetal karyotype should be mandatory.15 Although the specific risk differs among different groups, the association with chromosomal anomalies is well established, and should be taken into consideration when dealing with cases of fetal cardiac malformation. Furthermore, the most common chromosomal anomalies are all associated with a marked increase in the incidence of congenital heart defects over the general population. The specific risk for cardiac anomalies in patients with trisomy 13, trisomy 18, trisomy 21, and monosomy X is 84%, 99%, 50%, and 35%, respectively.18,19 Various chromosomal abnormalities are associated with specific congenital cardiac malformations. For instance, fetuses with Turner syndrome (45,X) manifest with outflow tract obstructions, most commonly coarctation of the aorta. Fetuses with Down syndrome (trisomy 21) often present with a common atrioventricular canal in which there are both atrial and ventricular septal defects, along with mitral and tricuspid valves that are not ‘off-set’. Moyano et al reviewed the echocardiography results for

174 fetuses with trisomy 18. Images were non-diagnostic in only 12 cases (7%), all examined at < 15 weeks’ gestation. Abnormal cardiac findings were detected in 118 of the remaining 162 fetuses (73%), including 15 with functional anomalies. A wide spectrum of cardiac anomalies was detected, including VSD, TOF, left heart disease, and atrioventricular septal defects. Although previous pathological series of trisomy 18 found structural heart malformations in all cases, most but not all trisomy 18 fetuses are sonographically detectable.20 The detection of any characteristic findings should alert the clinician to the necessity to perform amniocentesis for karyotype evaluation. Cardiac anomalies are commonly detected as part of a large number of congenital syndromes. Some of these syndromes are associated only with congenital cardiac anomalies and may have variable cardiac features. These include several syndromes such as Apert, Ellis–van Creveld, Cornelia de Lange, VATER (vertebrae, anus, trachea, esophagus, renal), Smith–Lemli–Optiz, Fanconi anemia, and neurofibromatosis. Conversely, other syndromes are referred to as ‘cardiac’ as the cardiac malformations constitute a major and integral part of their clinical presentation. This group includes syndromes such as Holt–Oram, Noonan, Alagille, Williams, DiGeorge, and the velocardiofacial syndromes.

Sonographic examination The sonographic evaluation of the fetal heart is best conducted by a transabdominal approach at 18–22 weeks’ gestation, when the heart is large enough to be seen in the majority of patients.21–23 Some expert sonologists, especially those skilled in transvaginal cardiac sonography, recommend an early scan at 14–16 weeks of gestation.24 The early vaginal scan provides a better time frame for both karyotyping and pregnancy termination, if indicated and desired. Furthermore, the early transvaginal approach is superior to the transabdominal approach in patients who are morbidly obese.25 However, there is a consensus that the early anatomy scan cannot be used alone for fetal cardiac screening due to both the technical difficulty and the developmental nature of some cardiac anomalies. Therefore, a transabdominal scan at 20–22 weeks’ gestation should be subsequently obtained.21,22 The fetal heart must be assessed for its position and relative size in the chest as well as for its anatomical structure and contractility rate. The fetal heart normally occupies approximately one-third of the fetal chest and has a left axis deviation of approximately 45° (range 15–60°). Abnormal cardiac axis should alert the clinician to the possibility of cardiac and/or extracardiac malformations. Smith et al detected a strong association between left cardiac axis deviation and cardiac malformations,

Genetics and cardiac anomalies

specifically conotruncal anomalies.26 Comstock et al reported that a right cardiac axis is associated with an increased incidence of structural cardiac malformations as well as with polysplenia or asplenia.27 The normal cardiac rate ranges between 110 and 160 beats per minute. The structural evaluation of the fetal heart is accomplished by obtaining several sonographic views. The fourchamber view provides important structural information and may be obtained technically by applying one of two methods. The first is by obtaining an image of the fetal abdomen at the level of the circumference; the transducer should then slide cephalad until the four-chamber view is gained. Alternatively, the transducer may be placed adjacent to a fetal rib and then slide into an intercostal space for ultimate viewing. The exact method and location of the image gained will ultimately be determined by the fetal lie and maternal habitus. The four-chamber view enables the visualization of several cardiac structures. These structures include the relative chamber size, and the position of the tricuspid and mitral valves, noting that typically the tricuspid valve is located closer to the cardiac apex giving the characteristic ‘off-set’ valve phenomenon. Additionally, flapping of the foramen ovale into the left atrium and the presence of the moderator bands at the right ventricle should be seen. Typically, the left atrium lies closest to the fetal spine and, owing to the left axis deviation, the right ventricle lies closest to the fetal chest wall. The right and left atria should be approximately the same size, while the right ventricle often appears larger, although not significantly larger, than the left ventricle. This occurs secondary to the increased filling of the right ventricle relative to the left ventricle and results in the appearance of bowing of the right ventricle into the left ventricle. Care must be taken to visualize the intraventricular septum in order to rule out a VSD. This is best accomplished with the beam of the transducer perpendicular to the plane of the septum. Additionally, the use of the color Doppler modality may facilitate the diagnosis of a VSD by demonstrating flow across the septum, and may detect even small VSDs that are not visible by gray-scale alone.28,29 Therefore, the fourchamber cardiac view is the sine qua non of cardiac sonography. Copel et al demonstrated that in 96% of fetuses with abnormalities detected in the four-chamber view a subsequent fetal echocardiography confirmed the diagnosis of a structural cardiac malformation.14 The same study calculated that the four-chamber view examination yielded a sensitivity and specificity of 92% and 99.7%, respectively, in the detection of congenital heart anomalies. However, these numbers have been challenged by other studies, and may represent an overzealous estimation of the efficacy of the four-chamber view, for several reasons.30 These results are based on retrospective data; additionally, the sonographers were not blinded to the suspected cardiac malformation in these fetuses prior to the examination. A lower estimation of the routine yield

611

of the four-chamber view in non-tertiary centers was derived from the RADIUS (Routine Antenatal Diagnostic Imaging with Ultrasound) trial.31 It is likely that the true yield of the four-chamber view is somewhere in between these estimations. In recent years, the concept of the ‘five-chamber view’ has been used to extend the cardiac anatomical survey. This view incorporates the four-chamber view with views of the outflow tracts, including the aortic and pulmonary tracts. In a series presented by Kirk et al the four-chamber view detected 47% of 51 fetuses with cardiac abnormality, whereas the addition of the aortic root view increased the sensitivity to 78%.30 The initial cardiac sonographic evaluation has an important role as a screening method, identifying fetuses with suspected structural cardiac anomalies. Subsequently, such cases require an extensive examination by fetal echocardiogram. This examination should generally be conducted by a pediatric cardiologist or a perinatologist with special expertise in fetal echocardiography. The fetal echocardiogram consists of additional views including the aortic and ductal arches as well as the basic views that were mentioned earlier. Furthermore, modalities such as M-mode to image wall motions and Doppler studies to measure flow across valves are used. The echocardiogram can also enable us to identify the leaflets of the specific valves. Some of the indications for fetal echocardiography include a previous abnormal ultrasound screening, prior child or fetus with a known cardiac anomaly, significant first-degree familial history, fetal arrhythmias that are documented on more than one occasion, non-immune hydrops, known aneuploidy, maternal diabetes, maternal alcohol abuse, maternal collagen vascular diseases and maternal PKU.32 Recently, three- and four-dimensional reconstruction using a combination of spatiotemporal image correlation, ‘inversion mode’, and ‘B-flow’ imaging has shown promising results in evaluation of the fetal heart.33–36 This new technique consists of a consecutive summation of multiple two-dimensional images and permits the threedimensional reconstruction of the different structures and the definition of their spatial position. A recent study demonstrated that rendered images of the four-chamber view provide excellent visualization of the heart chambers, myocardium, septum, and mitral and tricuspid valves. Malformations such as VSD, abnormal differential insertion of the atrioventricular valves, and valve atresia were well visualized with the inversion mode. In addition, information that could not be obtained by a two-dimensional scan such as the spatial relationships between the outflow tracts as well as the connections between the great arteries and ventricular chambers were well visualized. Goncalves et al concluded that this new application generates information about the anatomic and pathologic characteristics of the fetal heart that cannot be obtained with two-dimensional fetal echocardiography. They further

612

Fetal Cardiology

proposed these modalities as a method to enhance our ability to evaluate the fetal heart and therefore provide improved prenatal diagnosis.33 Nevertheless, this modality has not yet been incorporated in routine clinical evaluation of the fetal heart and its utilization is confined to clinical trials and specialized centers only.

First trimester screening First trimester screening has a central role as an early screening method for both aneuploidy and congenital cardiac anomalies. An increased fetal nuchal translucency has been shown to be associated with the presence of such malformations in several studies.37–43 Hyett et al were one of the first groups to explore this association. They found that the majority of congenital abnormalities of the heart and great arteries were associated with subcutaneous edema in the nuchal region at 10–14 weeks’ gestation.37 These authors further detected that the prevalence of major cardiac defects was increasing from 0.8/1000 in fetuses with a nuchal translucency thickness of less than the 95th centile to as high as 63.5/1000 in fetuses with nuchal translucency thickness above the 99th percentile. In a subsequent study they further reported that 20 of 36 fetuses that were diagnosed as trisomy 21 by chorionic villus sampling at 10 to 13 weeks’ gestation following the detection of nuchal translucency measurements greater than 3 mm had detectable cardiac anomalies.44 The authors found that a septal defect was observed in one of the 11 fetuses with nuchal translucency thickness of 3 mm and in 19 of the 25 with translucency greater than 4 mm. Several other studies, including a large prospective multicenter study, published similar findings, further supporting the data that an increased first trimester nuchal translucency measurement is associated with a higher risk of major congenital heart defect even in chromosomally normal pregnancies.41 Nevertheless, other studies documented that increased nuchal translucency thickness for the detection of congenital cardiac malformations performed less well than expected.42,43 Recently, Simpson et al reported the analysis of the FASTER (First and Second Trimester Evaluation of Risk) database, stating that nuchal translucency was not an effective screening tool for cardiac disease in an unselected population. Their study, however, suffered from major limitations in its design, mainly as those cases with either cystic hygromas or diagnosed aneuploidy were removed from analysis, thereby excluding from the database the most important subjects for this analysis, rendering their conclusions incorrect, in our opinion.45 The clinical implication that patients with unexplained significant elevations of nuchal translucency should undergo a fetal echocardiogram is well accepted by all authors.41–43,45

Specific anomalies and syndromes The conotruncal group of cardiac malformations is a heterogeneous category that involves malformations of both the ventricles and the conotruncus, including the pulmonary artery and the aorta. The frequency of the different malformations involved in the syndrome varies. Interrupted aortic arch (IAA), truncus arteriosus communis, TOF, and VSD are the most common malformations that constitute the syndrome, and their prevalence is 52%, 36%, 16% and 10%, respectively.46 Rarely, other anomalies such as transposition of the great arteries (TGA) and double-outlet right ventricle (DORV) may also present as part of this syndrome. The conotruncal defects represent 20–30% of all congenital heart defects.47 Owing to the unique fetal circulation, conotruncal defects are asymptomatic in utero and antenatal sonographic diagnosis may not be certain. However, in the absence of an associated shunt defect such as a VSD, these anomalies may become symptomatic immediately after birth, with a clinical manifestation that varies from cyanosis to heart failure in the immediate neonatal period. Boudjemline et al reported a series of 337 conotruncal malformations diagnosed over 5.5 years.48 These malformations constituted 16.2% of the congenital heart anomalies, and within this group 56% were diagnosed with TOF or with pulmonary atresia and VSD. Major chromosomal abnormalities were detected in 28 of these cases (8.2%). Furthermore, among fetuses with a normal ‘standard’ karyotype, further evaluation by fluorescent in situ hybridization (FISH) revealed a deletion of chromosome 22q11.2 in 54 of 237 cases (22.8%). Twenty-nine cases (8.3%) had an associated polymalformation syndrome. McElhinney et al prospectively followed 50 consecutive patients with truncus arteriosus. They assessed correlations between the anatomic features and the presence of chromosome 22q11 deletion. They concluded that a chromosome 22q11 deletion is common in patients with truncus arteriosus (40%), and that those with abnormal sidedness and/or branching of the aortic arch are significantly more likely to carry the deletion.49 Transposition of the great arteries can appear as the complete or the corrected form, and its incidence is 2/10 000 live births.47 The more common form, a complete TGA, consists of the aorta originating from the right ventricle and the pulmonary artery originating from the left ventricle. This can be sonographically visualized as absence of the typical crossing of the great vessels and/or by the ability to visualize both great vessels exiting the base of the heart simultaneously. Without an associated VSD, the four-chamber cardiac view will appear normal. The complete form of TGA may occur with or without an associated VSD and/or pulmonary stenosis. The corrected form consists of position changes of both the arterial

Genetics and cardiac anomalies

vessels and the atria. In corrected TGA, the pulmonary artery exits from the left ventricle that is connected to the right atrium, while the aorta exits from the right ventricle which is connected to the left atrium. In both forms, there may be additional associated defects, the most common of which is a VSD. It is difficult to quote the exact incidence of chromosomal anomalies associated with TGA, since this is a diverse condition. Several investigators have reported cases of isolated subsets with this condition, whereas association with cardiac syndromes has been reported as well. Several ‘cardiac syndromes’ have been reported to arise from specific chromosomal deletions, hence termed ‘deletion syndromes’. The most common of these deletions, located within the chromosome region of 22q11, has been described in several syndromes to be associated with cardiac malformations and dysmorphic features. These syndromes include the DiGeorge syndrome, the velocardiofacial syndrome, the conotruncal anomaly facial syndrome, the ‘Opitz’ GBBB syndrome, and the Cayler cardiofacial syndrome.50–52 As previously stated, 22q11.2 is the most frequent microdeletion syndrome and is estimated to occur in 1/6000 live births.53 It accounts for over 700 cases per year in the United States, and for at least 1/68 of all cardiac malformations at birth. The clinical manifestations of the 22q11.2 deletion syndrome may include cardiac anomalies, immune deficits, hypocalcemia, feeding difficulties secondary to palatal abnormalities, imperforated anus, renal anomalies, hypospadias, club feet, syndactyly, polydactyly, neural tube defects, short stature, speech delay, and neurobehavioral difficulties. The DiGeorge and velocardiofacial syndromes are the most commonly associated syndromes with 22q11 rearrangements. These conditions result from a failure to correctly form the third and fourth branchial arches during embryologic development. In more than 90% of these patients, deletions of chromosome 22q11.2 encompass about 3 MB of genomic DNA.54 Carlson et al studied the extent of the deletion on chromosome 22q11 in velocardiofacial syndrome by genotyping 151 such patients and performing haplotypic analysis in 105, using 15 consecutive polymorphic markers in 22q11.55 The authors found that 83% had a deletion, and more than 90% of these had a similar deletion of approximately 3 Mb, suggesting that the sequences flanking the common breakpoints were susceptible to rearrangement. Edelman et al investigated the hypothesis that most patients with DiGeorge or velocardiofacial syndromes had similar deletions in the 3-Mb region of chromosome 22q11.56 The authors developed a model of hamster–human somatic hybrid cell lines from such patients, and showed that breakpoints occurred within similar low copy repeats, termed LCR 22s. Sergi et al reported a series of seven infants with a deletion of the locus 22q11 with dysmorphic features and complex congenital heart defects including DORV, atresia or stenosis of the pulmonary valve, VSD, TOF, and major

613

aortopulmonary collateral arteries.57 Subsequently, Saitta et al reported that a significant aberrant interchromosomal exchange event during meiosis I in the proximal region of the affected chromosome 22 is the likely etiology causing the deletion.58 Hofbeck et al examined 44 patients with a conotruncal anomaly of VSD and pulmonary atresia to describe the prevalence and clinical spectrum of monosomy 22q11.2.59 The authors recorded the type of collateral lung perfusion along with the associated facial and immunological abnormalities. They found that monosomy 22q11.2 was present in 10 children (23%) with major aortopulmonary collateral arteries, an abnormality that was associated with decreased success of corrective surgery. The authors concluded that, in children with pulmonary atresia and VSD, monosomy 22q11.2 is preferentially associated with major aortopulmonary collateral arteries. In most patients with this microdeletion and such findings, surgical repair will require a different therapeutic approach. Further investigation was done by Giglio et al who reported that deletion of a 5-cM region on the short arm of chromosome 8 (8p23) was associated with a spectrum of congenital heart defects including conotruncal defects, ASD, atrioventricular canal defects, and pulmonary valve stenosis.60 The authors studied 12 patients, of whom there were seven with and five without a congenital heart defect. Their analysis detected that subjects with 8p deletions distal to D8S1706, at approximately 10 cM from the 8p telomere, did not have congenital heart defects, whereas subjects with a deletion that included the more proximal region suffered from the spectrum of heart defects reported in patients with 8p distal deletions. The most sensitive and widely used diagnostic test for the detection of 22ql1.2 deletion is FISH using probes from the commonly deleted region. Alternatively, polymerase chain reaction (PCR) can be performed to confirm failure to inherit a parental allele in the region or to determine copy number. Prenatal diagnosis is also available, particularly when a conotruncal cardiac defect is identified or when a parent carries a deletion. It is important to stress the importance of a thorough genetic counseling session prior to testing. This is done in order to provide the patient with current information regarding the natural history of the disorder, the testing options available, and the sensitivity and positive predictive value, as well as the limitations of the available tests. Recently, Soares et al investigated the prevalence of deletion 22q11 in fetuses with congenital cardiac anomalies. They concluded that in the presence of heart defects associated with other congenital anomalies, karyotyping is mandatory, and if clinical features are compatible, 22q11 microdeletion should be specifically sought with FISH techniques.61 The detection of chromosomal anomalies has a significant impact on both the prognosis and the follow-up of such patients. Subsequently, such cases may require specialized genetic counseling to these families.

614

Fetal Cardiology

Additional ‘deletion syndromes’ that are associated with congenital cardiac anomalies are Williams and Alagille syndromes. Williams is a rare syndrome, presenting in 5/100 000 live births. It may result from a sporadic mutation or be due to an autosomal dominant inheritance with high penetrance. Cardiac findings are a key in this syndrome, manifesting as supravalvular arterial stenosis. Additional anomalies seen in Williams syndrome are characteristic facies, cognitive impairment, and infantile hypercalcemia. Brewer et al investigated the genetics of 16 children and adolescents with a firm clinical diagnosis of Williams syndrome. They demonstrated positive results in all 16 subjects by using the FISH technique, employing the elastin gene probe. They therefore concluded that the detection of such hemizygosity by FISH constitutes a useful confirmatory diagnostic test.62 Alagille syndrome, a rare syndrome occurring in 1/100 000 live births, has several cardiac manifestations, including pulmonary stenosis, peripheral arterial stenosis, and TOF. Extracardiac manifestations include skeletal and ocular anomalies, characteristic facies, and neonatal jaundice secondary to bile duct paucity. It is inherited via both autosomal dominant inheritance with reduced penetrance and sporadic inheritance. A specific mutation in the JAG1 gene on chromosome 20p12 has been identified as the disease gene. Krantz et al found that JAG1 mutations are detected in about 70% of patients with Alagille syndrome, and include total gene deletions as well as protein truncating, splicing, and missense mutations. They did not find a phenotypic difference between patients with deletion of the entire JAG1 gene and those with intragenic mutations, suggesting that haploinsufficiency for JAG1 is the mechanism causing this syndrome.63 Recently, Warthen et al aggressively screened a cohort of 247 well-defined patients. They found that an increase in the JAG1 mutation detection rate to a level of 94% was accomplished by combining rigorous clinical phenotyping with a combination of mutation detection techniques, including FISH, genomic and cDNA sequencing, and quantitative PCR.64 Since the recurrence rate may be as high as 50%, families with a child with Alagille syndrome should undergo comprehensive genetic counseling. Approaches to consider in such cases include testing of the parents for possible mosaicism as well as performing pregestational diagnosis or FISH if medically indicated and desired by the patient. Holt–Oram is another rare ‘cardiac syndrome’ in which 85% of affected subjects demonstrate an ASD, VSD, and/ or atriovenricular (AV) block. Bilateral or unilateral upper limb anomalies present in all of these patients, and may include thenar abnormalities, hypoplastic radius, and carpal bone anomalies. The gene for this syndrome has been previously mapped to chromosome 12q24.1 by linkage. Subsequently, Li et al have identified the gene for the disorder as TBX5.65 Tissue in situ hybridization studies on human embryos from day 26 to day 52 of gestation revealed expression of TBX5 in the heart and limbs, a

finding which is consistent with the role of this gene in human embryonic development. An additional cardiac syndrome in which a specific gene abnormality has been identified is Noonan syndrome. It is a far more prevalent syndrome compared to the previous ones discussed, with an incidence of 1/1000–2000 live births. The clinical manifestations are variable, and may consist of cardiac anomalies in about 80% of cases, dysmorphism, short stature, webbed neck, and intelligence quotient (IQ) abnormalities (range from 50 to 130). Typical cardiac anomalies associated with this syndrome include pulmonary stenosis, cardiomyopathy, ASD, VSD, and a patent ductus arteriosus. Prenatal sonography plays an important role in the detection of this syndrome by detecting findings such as increased nuchal translucency, cystic hygroma, polyhydramnios, cardiac malformations, and cardiomyopathy. The mode of inheritance is either autosomal dominant or sporadic. Tartaglia et al demonstrated that mutations in PTPN11, the gene that encodes the nonreceptor type protein tyrosine phosphatase SHP-2, is the cause for Noonan syndrome in approximately 50% of cases of this genetically heterogeneous disorder. They further detected that 70% of subjects carrying the PTPN11 gene mutations had pulmonic stenosis, whereas only 6% of subjects carrying this mutation had hypertrophic cardiomyopathy.66 Double-outlet right ventricle (DORV) is a malformation in which most of the aorta and pulmonary artery arise from the right ventricle. The aorta and main pulmonary artery are classically located side-by-side. DORV usually has an associated VSD. DORV may occur in association with a subaortic or subpulmonary VSD, a doubly committed, juxta-arterial VSD, or a non-committed VSD. The hemodynamics depends on the position of the septal defect and the presence or absence of associated cardiac malformations. Likewise, the development of postnatal cyanosis varies depending on the degree and position of septal defects and associated cardiac anomalies. As reported by Smith et al, most fetuses with DORV can be identified antenatally as having a cardiac malformation. However, it may be difficult to distinguish this defect from other conotruncal abnormalities.67 The authors described 20 fetuses with antenatally detected conotruncal defects in which DORV was included in the differential diagnosis. After the exclusion of three fetuses from the study, postnatal confirmation showed that 10 actually had DORV. The authors subsequently detected two additional cases in the medical records with a postnatal diagnosis of DORV. Of these 12 cases, nine had a normal karyotype. The classic findings in tetralogy of Fallot (TOF) consist of a combination of VSD with an associated overriding aorta, pulmonary stenosis of varying degree, and associated right ventricular hypertrophy. The condition may merge with the diagnosis of VSD with associated pulmonary atresia in cases where there is no continuity between the right ventricle and the pulmonary trunk. It is important to

Genetics and cardiac anomalies

realize that the fourth portion of the tetrad, namely right ventricular hypertrophy, does not exist in utero, and is detectable postnatally. Sonographic evaluation of this condition demonstrates an inverse relationship between the size of the aorta and that of the pulmonary artery. A large aortic root is often the sonographic finding detected, leading to the possibility of this condition. Postnatal manifestations include cyanosis and heart failure, and are dependent on the degree of pulmonary stenosis and the size of the VSD. Lee et al correlated prenatal echocardiographic findings with neonatal outcomes in a large population affected by TOF.68 The authors identified 17 fetuses with confirmed VSD and an overriding aorta, with varying degrees of right ventricular outflow obstruction after delivery. Five fetuses demonstrated enlarged aortic roots during the initial scan, and only two of the 10 fetuses with a measurable pulmonary artery demonstrated pulmonary valve stenosis in the initial sonographic evaluation. Prior to delivery, two fetuses were thought to have only a VSD. The authors concluded that the sonographic findings in fetuses with TOF vary significantly, and may consist of a VSD and aortic septal override only, since the pulmonary artery stenosis is not always present at the initial sonographic examination. Interestingly enough, all patients in this study had normal karyotypes. Endocardial cushion defects are cardiac malformations that evolve from abnormalities in structures that originate from the endocardial cushions and the atrial and ventricular septa. These defects are also referred to as common atrioventricular canal defects and atrioventricular septal defects. There are three major types of endocardial cushion defects, as described by Vick and Titus.69 A complete endocardial cushion defect consists of an ostium primum ASD with a non-restrictive VSD of the atrioventricular canal type and a common atrioventricular valve. A partial endocardial cushion defect consists of an ostium primum ASD with or without a left mitral valve defect and additional restrictive defects in the ventricular septum. The third type of an endocardial cushion defect is an isolated VSD of the atrioventricular canal type. These malformations can be diagnosed antenatally by sonographic visualization of the atrial and septal defects. In addition, in the complete form of this condition, the characteristic offset appearance of the tricuspid and mitral valves is lost, and the valves insert at the same level. Usually this condition has no clinical in utero manifestations unless there is a major valvular incompetence associated, in which case fetal heart failure may occur. The finding of an endocardial cushion defect in utero should prompt the clinician to obtain a fetal karyotype. Down syndrome is by far the most common chromosomal abnormality associated with this cardiac anomaly. Torfs and Christianson detected a risk ratio of 1/1009 in comparing infants without and with Down syndrome for endocardial cushion defects.70 Bhatia et al performed a

615

postnatal echocardiography evaluation of children with Down syndrome.71 They detected cardiac anomalies in 44% of these children, with endocardial cushion defects being the most common anomaly. Their result may underestimate the prevalence of these malformations, as a large number of pregnancies with Down syndrome fetuses and cardiac anomalies miscarry or terminate electively and, thus, are not represented in this analysis. The Baltimore– Washington Infant Study on congenital cardiovascular malformations associated with chromosome anomalies found that endocardial cushion defects were the most predominant cardiac abnormality associated with Down syndrome (60.1%) and rarely occurred as an associated lesion (2.8%).72 Carmi et al further analyzed data from the Baltimore–Washington Infant Study, comparing the isolated occurrence of endocardial cushion defect with its occurrence as part of a syndrome.73 The authors concluded that ‘syndromic’ endocardial cushion defects tend to be of the complete atrioventricular canal type, and are less frequently associated with left cardiac anomalies than the isolated form. Korenberg et al found that the presence of a third copy of the band 21q22 was required for the phenotypic expression of many features of Down syndrome, including endocardial cushion defects, mental retardation, and the characteristic facies.74 Using autoradiograms of quantitative southern blots of DNA from two affected sisters, their carrier father, and a normal control, the authors defined the region responsible for these characteristics. This region may include parts of bands 21q22.2 and 21q22.3, but must exclude the genes SOD1 and APP and most of band 21q22.1. Hiltgen et al developed an animal model producing atrioventricular septal defects with a high prevalence.75 This murine model of trisomy 16 included failure of the endocardial cushions to fuse. The authors detected a significant reduction in both the raw cardiac mesenchyme cells and the cellular density in this model. There appeared to be a delay in the expression of cytotaxin and fibronectin during cushion development in the murine model. The authors concluded that failure of endocardial cushion fusion in the murine model may be related to either an abnormal shape of the cushions or an inhibition or delay in the induction, transformation, or seeding of cardiac mesenchymal cells. Kurnit et al developed a stochastic model for atrioventricular canal malformations in Down syndrome.76 The authors used a computer simulation to model the normal anatomic sequences of cushion-tocushion and cushion-to-septum fusion in atrioventricular canal development. Low values of intercellular adhesiveness engendered simulations resembling normal atrioventricular canal development, whereas higher values of adhesiveness yielded deficiencies of atrioventricular canal development as seen in Down syndrome. VSD and ASD may occur in combination with endocardial cushion defects or present as an isolated anomaly.

616

Fetal Cardiology

Van Karnebeek and Hennekam used the Human Cytogenetics Database to search for all autosomal anomalies associated with one or more congenital heart defects.77 The database search showed that both ASD and VSD were associated with multiple chromosome loci, suggesting that heterogeneity existed with these types of congenital heart defect. However, the authors noted duplications of both the p and q arms of chromosomes 3 and 4 as well as a duplication of chromosome 8qter, PS and duplication of the terminal long arm of chromosome 8. Hypoplastic left heart consists of a normal-sized right ventricle with a very small left ventricle. The left ventricle may have associated mitral and/or aortic atresia. Real-time sonography demonstrates a left ventricle that is significantly smaller than the right one with additional restricted movement of the mitral valve into the left ventricle. This condition is usually asymptomatic in utero, secondary to perfusion from the right ventricle through the ductus arteriosus. Postnatally, this condition is universally fatal without an immediate surgical correction, and is responsible for about 25% of cardiac related neonatal deaths in the first week of life. There is no single chromosomal abnormality associated with hypoplastic left heart. In fact, most fetuses with this condition carry a normal karyotype. Reis et al described a series of 219 deliveries of infants with hypoplastic left heart syndrome.78 Antenatal diagnosis was made in 82 (37%), whereas 137 (63%) were diagnosed in the neonatal period. Only 32 cases underwent karyotype analysis, with eight (25%) cases having an abnormal result. Seven of the abnormal karyotypes were 45,X, and one was trisomy 21. Other studies in the literature have reported associations of hypoplastic left heart with several genetic anomalies such as trisomy 16q,79 trisomy 9,80 46,X, i(Xq),81 dup 12p,82 and the Ullrich–Turner syndrome with mosaicism 45,X/46,XX/47,XXX.83 An increased incidence of hypoplastic left heart with telomeric deletions of chromosome 11q was recently detected using the Human Cytogenetics Database.77 Coarctation of the aorta usually occurs between the ductus arteriosus and the opening of the left subclavian artery. Three theories have been suggested to explain this condition: a true malformation due to embryologic abnormalities, an aberrant ductal tissue in the aortic wall causing narrowing of the isthmus at the time of ductal closing (Skodiac theory), and the result of an intrauterine hemodynamic perturbation from the aorta to the pulmonary artery through the ductus arteriosus.84 Associated cardiac anomalies are extremely common and consist mostly of VSD, ASD, and conotruncal anomalies. As many as 50% of these cases have additional mitral and/or aortic valve abnormality.85 Conduction abnormalities may be present as well, leading to complete heart block in severe cases.86 The association between coarctation of the aorta and Turner syndrome is well established. Mazzanti and

Cacciari described a series of 594 patients with Turner syndrome, 6.9% of whom had aortic coarctation.87 Additionally, 12.5% had a bicuspid aortic valve, and 3.2% had other aortic valve disease. The authors found that patients with a 45,X karyotype had the greatest prevalence of partial anomalous pulmonary venous drainage and aortic coarctation, whereas a bicuspid aortic valve and aortic valve disease were more common in patients with X chromosome structural abnormalities. Hou et al found that congenital cardiac anomalies were present in 22.4% of 66 patients with Turner syndrome. In their study, 27% of these anomalies were diagnosed as aortic coarctation.88 Gotzsche et al found a significantly increased prevalence of cardiac malformations in subjects who were 45,X as compared to those who were mosaic monosomy X (38% versus 11%). This was primarily due to the difference in prevalence of aortic valve abnormalities and aortic coarctation between the two.89 In an antenatal series of 36 fetuses with sonographically diagnosed cardiac anomalies, five were Turner syndrome.90 The most common anomaly observed in these patients was hypoplastic aortic arch in combination with hypoplasia of the left ventricle and left VSD.

Conclusions The finding of a structural cardiac anomaly should prompt further evaluation including fetal echocardiography, detailed anatomical survey searching for additional anomalies in other systems, genetic counseling, and specific tests looking for an underlying chromosomal aneuploidy or deletion. Conversely, the elucidation of an abnormal karyotype should lead to a thorough, detailed cardiac ultrasound examination looking for associated anomalies. It is well appreciated that, for example, 50% of children with Down syndrome have significant cardiac pathology. We need to increase our understanding of the interplay between aneuploid or otherwise genomically derived abnormalities and cardiac pathology. Hopefully, over the next decade, we will learn how to improve our diagnosis and treatment of such problems.

References 1. Mitchell SC, Sellman AH, Westphal MC, Park J. Etiologic correlates in a study of congenital heart disease in 56,109 births. Am J Cardiol 1971; 28: 653–7. 2. Rosenn B, Miodovnik M, Combs CA. Glycemic thresholds for spontaneous abortion and congenital malformations in insulin-dependent diabetes mellitus. Obstet Gynecol 1994; 84: 515–20. 3. Freinkel N. Diabetic embriopathy and fuel-mediated organ teratogenesis: lessons from animal models. Horm Metab Res 1988; 20: 463–75.

Genetics and cardiac anomalies

4. Greene MF, Hare JW, Cloherty JP, Benaceraff BR, Soeldner JS. First-trimester haemoglobin A1 and risk for major malformation and spontaneous abortion in diabetic pregnancy. Teratology 1989; 39: 225–31. 5. Lipson T. Fetal alcohol syndrome. Aust Fam Physician 1988; 17: 385–6. 6. Lammer EJ, Chen DT, Hoar RM et al. Retinoic acid embryopathy. N Engl J Med 1985; 313: 837–41. 7. Brent R, Fawcett L, Beckman DA. Principles of teratology. In: Evans MI, Johnson MP, Yaron Y, Drugan A, eds. Prenatal Diagnosis. New York: McGraw-Hill, 2006: 79–112. 8. Abel E, Sokol RJ, Hannigan JH, Nordstrom BA. The effects of maternal drinking during pregnancy. In: Evans MI, Johnson MP, Yaron Y, Drugan A, eds. Prenatal Diagnosis. New York: McGraw-Hill, 2006: 123–30. 9. Platt LD, Koch R, Azen C et al. Maternal phenylketonuria collaborative study, obstetrical aspects, and outcomes: the first 6 years. Am J Obstet Gynecol 1992; 166: 1150–60. 10. Rouse B, Azen C, Koch R et al. Maternal Phenylketonuria Collaborative Study (MPKUCS) offspring: facial anomalies, malformations, and early neurological sequelae. Am J Med Genet 1997; 69: 89–95. 11. Rouse B, Matalon R, Koch R et al. Maternal phenylketonuria syndrome: congenital heart defects, microcephaly, and developmental outcomes. J Pediatr 2000; 136: 57–61. 12. Callan NA, Maggio M, Steger S, Kan JS. Fetal echocardiography: indications for referral, prenatal diagnoses, and outcomes. Am J Perinatol 1991; 8: 390–4. 13. Ferencz D, Rubin JD, McCarter RJ et al. Cardiac and noncardiac malformations: observations in a population based study. Teratology 1987; 35: 367–78. 14. Copel JA, Pilu G, Green J, Hobbins JC, Kleinman CS. Fetal echocardiographic screening for congenital heart disease: the importance of the four chamber view. Am J Obstet Gynecol 1986; 157: 648–55. 15. Chaoui R, Korner H, Bommer C et al. Prenatal diagnosis of heart defects and associated chromosomal aberrations. Ultraschall Med 1999; 20: 177–84. 16. Copel JA, Pilu G, Kleinman CS. Congenital heart disease and extracardiac anomalies: associations and indications for echocardiography. Am J Obstet Gynecol 1986; 154: 1121–32. 17. Allan LD, Sharland GK, Chita SK, Lockart S, Maxwell DJ. Chromosomal anomalies in fetal congenital heart disease. Ultrasound Obstet Gynecol 1991; 1: 8–11. 18. Nora JJ, Nora AH. Update on counselling the family with a first degree relative with a congenital heart defect. Am J Med Genet 1988; 29: 137–42. 19. Wallgren EI, Landtman B, Rapola J. Extra-cardiac malformations associated with congenital heart defects. Eur J Cardiol 1978; 7: 15–24. 20. Moyano D, Huggon IC, Allan LD. Fetal echocardiography in trisomy 18. Arch Dis Child Fetal Neonatal Ed 2005; 90: F520–2. 21. Allan L, Benacerraf B, Copel JA et al. Isolated major congenital heart disease. Ultrasound Obstet Gynecol 2001; 17: 370–9. 22. Allan LD. Cardiac anatomy screening: what is the best time for screening in pregnancy? Curr Opin Obstet Gynecol 2003; 15: 143–6. 23. Copel JA, Kleinman CS. Heart and vascular malformations. In: Evans MI, Johnson MP, Yaron Y, Drugan A, eds. Prenatal Diagnosis. New York: McGraw-Hill, 2006: 227–38.

617

24. Timor-Tritsch IE, Bashiri A, Monteagudo A, Arslan A. Qualified and trained sonographers in the US can perform early anatomy scans between 11 and 14 weeks. Am J Obstet Gynecol 2004; 191: 1247–52. 25. Timor-Tritsch IE. As technology evolves, so should its application: shortcomings of the ‘18-week anatomy scan’. J Ultrasound Med 2006; 25: 423–8. 26. Smith RS, Comstock CH, Kirk JS, Lee W. Ultrasonographic left cardiac axis deviation: a marker for fetal anomalies. Obstet Gynecol 1995; 85: 187–91. 27. Comstock CH, Smith R, Lee W, Kirk JS. Right fetal cardiac axis: clinical significance and associated findings. Obstet Gynecol 1998; 91: 495–9. 28. Axt-Fliender R, Schwartz A, Smrcek J et al. Isolated ventricular septal defects detected by colour Doppler imaging: evolution during fetal and first year of postnatal life. Ultrasound Obstet Gynecol 2006; 27: 266–73. 29. Kurjak A, Kupesis S. Color Doppler in congenital defects. In: Evans MI, Johnson MP, Yaron Y, Drugan A, eds. Prenatal Diagnosis. New York: McGraw-Hill, 2006: 331–42. 30. Kirk JS, Riggs TW, Comstock CH et al. Prenatal screening for cardiac anomalies: the value of routine addition of the aortic root to the four-chamber view. Obstet Gynecol 1994; 84: 427–31. 31. Ewigman BG, Crane JP, Frigoletto FD et al. Effect of prenatal ultrasound screening on perinatal outcome. RADIUS Study Group. N Engl J Med 1993; 329: 821–27. 32. Winter RM, Baraitser M. Multiple Congenital Anomalies: a Diagnostic Compendium. London: Chapman and Hall Medical, 1991. 33. Goncalves LF, Espinoza J, Lee W et al. A new approach to fetal echocardiography: digital casts of the fetal cardiac chambers and great vessels for detection of congenital heart disease. J Ultrasound Med 2005; 24: 415–24. 34. Chaoui R, Hoffmann J, Heling KS. Three-dimensional (3D) and 4D color Doppler fetal echocardiography using spatiotemporal image correlation (STIC). Ultrasound Obstet Gynecol 2004; 23: 535–45. 35. DeVore GR, Polanco B, Sklansky MS, Platt LD. The ‘spin’ technique: a new method for examination of the fetal outflow tracts using three-dimensional ultrasound. Ultrasound Obstet Gynecol 2004; 24: 72–82. 36. Goncalves LF, Espinoza J, Romero R et al. Four-dimensional ultrasonography of the fetal heart using a novel Tomographic Ultrasound Imaging display. J Perinat Med 2006; 34: 39–55. 37. Hyett JA, Perdu M, Sharland GK, Snijders RS, Nicolaides KH. Increased nuchal translucency at 10–14 weeks of gestation as a marker for major cardiac defects. Ultrasound Obstet Gynecol 1997; 4: 242–6. 38. Weiner Z, Goldstein I, Bombard A, Applewhite L, Itzkovits-Eldor J. Screening for structural fetal anomalies during the nuchal translucency ultrasound examination. Am J Obstet Gynecol 2007; 197: e1–5. 39. Sonek J. First trimester ultrasonography in screening and detection of fetal anomalies. Am J Med Genet C Semin Med Genet 2007; 145: 45–61. 40. Dane B, Dane C, Sivri D et al. Ultrasound screening for fetal major abnormalities at 11–14 weeks. Acta Obstet Gynecol Scand 2007; 86: 666–70. 41. Bahado-Singh RO, Wapner R, Thom E et al. First Trimester Maternal Serum Biochemistry and Fetal Nuchal Translu-

618

42.

43.

44.

45.

46.

47.

48.

49.

50.

51. 52.

53.

54.

55.

56.

57.

58.

59.

Fetal Cardiology

cency Screening Study Group. Elevated first-trimester nuchal translucency increases the risk of congenital heart defects. Am J Obstet Gynecol 2005; 192: 1357–61. Mavrides E, Cobian-Sanchez F, Tekay A et al. Limitations of using first-trimester nuchal translucency measurement in routine screening for major congenital heart defects. Ultrasound Obstet Gynecol 2001; 17: 102–5. Hafner E, Schuller T, Metzenbauer M, Schuchter K, Philipp K. Increased nuchal translucency and congenital heart defects in a low-risk population. Prenat Diagn 2003; 23: 985–9. Hyett JA, Moscoso G, Nicolaides KH. First-trimester nuchal translucency and cardiac septal defects in fetuses with trisomy 21. Am J Obstet Gynecol 1995; 172: 1411–13. Simpson LL, Malone FD, Bianchi DW et al. Nuchal translucency and the risk of congenital heart disease. Obstet Gynecol 2007; 109: 376–83. Goldmuntz E, Clark BJ, Mitchell LE et al. Frequency of 22q11 deletions in patients with conotruncal defects. J Am Coll Cardiol 1998; 32: 492–8. Fyler DC, Buckley LP, Hellenbrand WE. Report of the New England Regional Cardiac Program. Pediatrics 1980; 65: 375–461. Boudjemline Y, Fermont L, LeBidois J et al. Prenatal diagnosis of conotruncal heart diseases. Arch Mal Coeur Vaiss 2000; 93: 583–6. McElhinney DB, Driscoll DA, Emanuel BS, Goldmuntz E. Chromosome 22q11 deletion in patients with truncus arteriosus. Pediatr Cardiol 2003; 24: 569–73. Driscoll DA, Salvin J, Sellinger B et al. Prevalence of 22q11 microdeletions in DiGeorge and velocardiofacial syndromes: implications for genetic counselling and prenatal diagnosis. J Med Genet 1993; 30: 813–17. Driscoll DA. Genetic basis of DiGeorge and velocardiofacial syndromes. Curr Opin Pediatr 1994; 6: 702–6. McDonald-McGinn DM, Driscoll DA, Bason L et al. Autosomal dominant ‘Opitz’ GBBB syndrome due to a 22q11.2 deletion. Am J Med Genet 1995; 59: 103–13. Botto LD, May K, Fernhoff PM et al. A population-based study of the 22q11.2 deletion: phenotype, incidence, and contribution to major birth defects in the population. Pediatrics 2003; 112: 101–7. Amati F, Conti E, Novelli A et al. Atypical deletions suggest five 22q11.2 critical regions related to DiGeorge/ velo-cardio-facial syndrome. Eur J Hum Genet 1999; 7: 903–9. Carlson C, Sirotkin H, Pandita R et al. Molecular definition of 22q11 deletions in 151 velo-cardio-facial syndrome patients. Am J Hum Genet 1997; 61: 620–9. Edelmann L, Pandita RK, Palanisamy N et al. A common molecular basis for rearrangement disorders on chromosome 22q11. Hum Mol Genet 1999; 8: 1157–67. Sergi C, Serpi M, Muller-Navia J et al. CATCH 22 syndrome: report of 7 infants with follow up data and review of the recent advancements in the genetic knowledge of the locus 22q11. Pathologica 1999; 91: 166–72. Saitta SC, Harris SE, Gaeth AP et al. Aberrant interchromosomal exchanges are the predominant cause of the 22q11.2 deletion. Hum Mol Genet 2004; 13: 417–28. Hofbeck M, Leipold G, Rauch A, Buheitel G, Singer H. Clinical relevance of monosomy 22q11.2 in children with

60.

61.

62.

63.

64.

65.

66.

67.

68.

69.

70.

71.

72.

73.

74.

75.

76.

77.

pulmonary atresia and ventricular septal defect. Eur J Pediatr 1999; 158: 302–7. Giglio S, Graw SL, Gimelli G et al. Deletion of a 5-cM region at chromosome 8p23 is associated with a spectrum of congenital heart defects. Circulation 2000; 102: 432–7. Soares G, Alvares S, Rocha C et al. Congenital heart defects and chromosomal anomalies including 22q11 microdeletion (CATCH 22). Rev Port Cardiol 2005; 24: 349–71. Brewer CM, Morrison N, Tolmie JL. Clinical and molecular cytogenetic (FISH) diagnosis of Williams syndrome. Arch Dis Child 1996; 74: 59–61. Krantz ID, Piccoli DA, Spinner NB. Clinical and molecular genetics of Alagille syndrome. Curr Opin Pediatr 1999; 11: 558–64. Warthen DM, Moore EC, Kamath BM et al. Jagged1 (JAG1) mutations in Alagille syndrome: increasing the mutation detection rate. Hum Mutat 2006; 27: 436–43. Li QY, Newbury-Ecob RA, Terrett JA et al. Holt-Oram syndrome is caused by mutations in TBX5, a member of the Brachyury (T) gene family. Nat Genet 1997; 15: 21–9. Tartaglia M, Kalidas K, Shaw A et al. PTPN11 mutations in Noonan syndrome: molecular spectrum, genotypephenotype correlation, and phenotypic heterogeneity. Am J Hum Genet 2002; 70: 1555–63. Smith RS, Comstock CH, Kirk JS et al. Double-outlet right ventricle: an antenatal diagnostic dilemma. Ultrasound Obstet Gynecol 1999; 14: 315–19. Lee W, Smith RS, Comstock CH et al. Tetralogy of Fallot: prenatal diagnosis and postnatal survival. Obstet Gynecol 1995; 86: 583–8. Vick GW, Titus JL. Defects of the atrial septum including the atrioventricular canal. In: Garson A, Bricker J, McNamara DG, eds. The Science and Practice of Pediatric Cardiology. Philadelphia: Lea & Febiger, 1990: 1037–42. Torfs CP, Christianson RE. Anomalies in Down syndrome individuals in a large population-based registry. Am J Med Genet 1998; 77: 431–8. Bhatia S, Verma IC, Shrivastava S. Congenital heart disease in Down syndrome: an echocardiographic study. Indian Pediatr 1992; 29: 1113–16. Ferencz C, Neill CA, Boughman JA et al. Congenital cardiovascular malformations associated with chromosome abnormalities: an epidemiologic study. J Pediatr 1989; 114: 79–86. Carmi R, Boughman JA, Ferencz C. Endocardial cushion defect: further studies of ‘isolated’ versus ‘syndromic’ occurrence. Am J Med Genet 1992; 43: 569–75. Korenberg JR, Kawashima H, Pulst SM et al. Molecular definition of a region of chromosome 21 that causes features of the Down syndrome phenotype. Am J Hum Genet 1990; 47: 236–46. Hiltgen GG, Markwald RR, Litke LL. Morphogenetic alterations during endocardial cushion development in the trisomy 16 (Down syndrome) mouse. Pediatr Cardiol 1996; 17: 21–30. Kurnit DM, Aldridge JF, Matsuoka R, Matthysse S. Increased adhesiveness of trisomy 21 cells and atrioventricular canal malformations in Down syndrome: a stochastic model. Am J Med Genet 1985; 20: 385–99. Van Karnebeek CD, Hennekam RC. Associations between chromosomal anomalies and congenital heart defects database search. Am J Med Genet 1999; 84: 158–66.

Genetics and cardiac anomalies

78. Reis PM, Runch MR, Bove EL, van de Ven CJ. Obstetric management of 219 infants with hypoplastic left heart syndrome. Am J Obstet Gynecol 1998; 179: 1150–4. 79. Bacino CA, Lee B, Spikes AS, Shaffer LG. Trisomy 16q in a female newborn with a de novo X;16 translocation and hypoplastic left heart. Am J Med Genet 1999; 82: 128–31. 80. Saneto RP, Applegate KE, Frankel DG. Atypical manifestations of two cases of trisomy 9 syndrome: rethinking development delay. Am J Med Genet 1998; 80: 42–5. 81. Shah A, Fay JE, Ford S, Holden JJ. 46,X,i(Xq) karyotype in a patient with hypoplastic left heart. Clin Genet 1985; 28: 178–9. 82. Qazi QH, Kanchanapoomi R, Cooper R, Madahar C, Beller E. Dup(12p) and hypoplastic left heart. Am J Med Genet 1981; 9: 195–9. 83. Bidot-Lopez P, Matisoff D, Talner NS, Hsia YE. Hypoplastic left heart in a patient with 45,X/46,XX/47,XXX mosaicism. Am J Med Genet 1978; 2: 341–3. 84. Pilu G, Jeanty P. Prenatal diagnosis of congenital heart disease. In: Reece EA, Hobbins JC, eds. Medicine of the

85. 86. 87.

88.

89.

90.

619

Fetus and Mother. Philadelphia: Lippincott-Raven, 1999: 561–2. Becker AE. Anomalies associated with coarctation of the aorta. Circulation 1970; 41: 1067–75. Machado MV, Tynan MJ, Curry PV, Allan LD. Fetal complete heart block. Br Heart J 1998; 60: 512–15. Mazzanti L, Cacciari E. Congenital heart disease in patients with Turner’ s syndrome. Italian Study Group for Turner Syndrome (ISGTS). J Pediatr 1998; 133: 688–92. Hou JW, Hwu WL, Tsai WY et al. Cardiovascular disorders in Turner’s syndrome and its correlation to karyotype. J Formos Med Assoc 1993; 92: 188–9. Gotzsche CO, Krag-Olsen B, Nielsen J, Sorensen KE, Kristensen BO. Prevalence of cardiovascular malformations and association with karyotypes in Turner’ s syndrome. Arch Dis Child 1994; 71: 433–6. Gembruch U, Baschat AA, Knopfle G, Hansmann M. Results of chromosomal analysis in fetuses with cardiac anomalies as diagnosed by first and early second trimester echocardiography. Ultrasound Obstet Gynecol 1997; 10: 391–6.

42 Cardiac defects in chromosomally abnormal fetuses Jon Hyett and Alex Gooi Introduction Chromosomal abnormalities are associated with high rates of perinatal death and infant morbidity. Similarly, cardiac abnormalities are the commonest form of severe congenital abnormality, resulting in stillbirth, and neonatal and childhood death, and are a major cause of childhood morbidity. It is therefore not surprising that the detection of chromosomal abnormalities and congenital heart defects forms two key areas of screening in prenatal diagnosis and that there is a significant interaction between the two. Approximately 50% of infants with trisomy 21 are affected by congenital heart disease, and the prevalence is even higher in other, more lethal chromosomal abnormalities. Similarly, a high proportion of fetuses with structural cardiac defects have an underlying chromosomal abnormality. This chapter reviews the associations that have been described between structural cardiac defects and various chromosomal abnormalities. Prenatal detection rates for cardiac defects in low-risk populations are recognized as being relatively poor, and this chapter also reviews some first trimester screening tools for aneuploidy that demonstrate cardiac function and may also be useful markers for congenital heart disease in euploid fetuses. Some progress has recently been made in identifying the underlying genetic anomalies associated with cardiac defects specific to chromosomal abnormality, and this is also discussed.

Associations between structural cardiac defects and chromosomal abnormality One of the main problems in assessment of the association between cardiac defects and chromosomal abnormality is the fact that all chromosomal abnormalities have

significant rates of intrauterine lethality, and consequently the prevalence and type of cardiac defects seen within a population will vary markedly depending on the stage of ascertainment. The Baltimore–Washington Infant Study, which reported structural cardiac defects in 3.7/1000 children, found that 12.7% of liveborn infants with congenital cardiovascular malformations had a chromosomal abnormality.1 Although the original study did not include data for stillbirths or terminations for fetal abnormality, reanalysis of the data to allow for intrauterine lethality suggested that the prevalence of chromosomal abnormalities in fetuses with cardiac defects identified at the time of the 18–20-week scan would be as high as 40%.2 This is certainly supported by studies describing populations of fetuses with cardiac defects, which have suggested a prevalence of chromosomal abnormality of 28–42%.3–5 Further follow-up of these fetuses shows a high rate (73%) of intrauterine and neonatal lethality, a fact supported by postmortem data suggesting that the prevalence of cardiac defects in a population of stillborn fetuses (21/1000 fetuses) was 10 times higher than seen in infancy and childhood.5,6 The postmortem data also suggest that the types of cardiac defects seen depend on the characteristics of the population being studied –with an increased prevalence of ventricular septal defects, pulmonary atresia, and double-outlet right ventricle in stillborn fetuses compared to liveborn infants.6 This situation is further complicated by the impact of prenatal diagnosis. Many parents who find they have a fetus affected by a severe cardiac abnormality will choose to terminate the pregnancy, particularly if a chromosomal abnormality is also found.5,7 Consequently, future series describing liveborn infants will be very different from those being examined prenatally.8 The commonest chromosomal abnormalities seen after the first trimester of pregnancy are the trisomies: 21 (Down syndrome), 18 (Edwards syndrome), and 13 (Patau syndrome), and 45,X (Turner syndrome). The pattern of structural malformation, including the type of cardiac defects, seen in these conditions are well described, allowing the sonologist to form an opinion of the likely

622

Fetal Cardiology

association based on the cardiac anomaly and the presence of extracardiac abnormalities. The recognized associations between these chromosomal abnormalities and cardiac defects are listed in Table 42.1. Whilst trisomies 13 and 18 are essentially lethal, a significant proportion of fetuses with trisomy 21 or 45,X survive, and the cardiac defects seen in these infants were therefore the first to be described. From a prenatal perspective, the first ultrasound view commonly used to screen for cardiac defects was the four-chamber view, and consequently the postnatal association between trisomy 21 and endocardial cushion defects was rapidly recognized in prenatal life.

syndrome are seen during the anomaly scan it is important to return to the heart and repeat the examination looking for the specific anomalies associated with this condition. Langford et al have recently reviewed a series of 125 fetuses diagnosed with an AVSD prenatally and calculated the relative risk for chromosomal abnormality when other variables, such as maternal and gestational age, are taken into account. They found that the presence of an AVSD increased the risk for trisomy 21 by 107 (95% confidence interval (CI) 87–127) times – making this the strongest marker for trisomy 21 that has been described.16 The prognosis for fetuses with an atrioventricular septal defect is very variable – depending on the wide range of associated cardiac and extracardiac anomalies – but it is now clear that fetuses with Down syndrome that have a simple AVSD have lower mortality and morbidity from surgical repair than their more complex euploid counterparts.17,18 Evans reported a series of cardiac postmortem findings in infants who had had Down syndrome in 1950, drawing attention to the association with septal defects.19 This has been followed by several studies reporting echocardiographic findings in live born infants as well as findings, by both ultrasound scan and postmortem, in prenatal cohorts (Table 42.2).9,20–25 All series suggest that atrioventricular septal defects and ventricular septal defects are the predominant anomalies seen in this chromosomal abnormality, being the anomalies found in up to 75% of Down cases with heart defects. Whilst the prevalence of septal defects is approximately 30% in the liveborn series, it has been reported to be as high as 45–55% in the prenatal/postmortem series.20,21,23,25 There are several possible explanations for this discrepancy. First, there may be spontaneous

Trisomy 21 An atrioventricular septal defect (AVSD) is described in 7.5% of infants with congenital heart disease, and has a very distinctive association with trisomy 21.1,9 Recent series of fetuses found to have an AVSD describe chromosomal abnormalities in 40–49% of cases, most commonly trisomy 21, but also reporting cases of trisomy 18 and 13.10–12 Roughly half of the AVSDs described are seen in association with other cardiac anomalies, including both left and right atrial isomerism, a small left ventricle and/or aorta, and congenital heart block, and there is also a strong association with extracardiac anomalies. The prevalence of chromosomal abnormality in fetuses with isomerism has been shown to be very low.13,14 Interestingly, the fetus with Down syndrome is most likely to have a simple AVSD with no other identifiable anomalies – which is also the type most likely to be missed at the 20-week anomaly scan.15 If soft markers for Down

Table 42.1

Cardiac defects associated with chromosomal abnormality

Atrioventricular septal defect

Trisomy 21

Trisomy 18

+++

+

Ventricular septal defect

+++

+++

Atrial septal defect

++

++

Patent ductus arteriosus

Trisomy 13

45,X

Del22q11

++

+

++

++

++ +

Tetralogy of Fallot

++

+

Double-outlet right ventricle

+

+

Common arterial trunk

+

Transposition of the great arteries

+ +

No common chromosomal associations

Interrupted aortic arch

++

Coarctation of the aorta

+

+

+++

Hypoplastic left heart

+

+

+

Univentricular heart

+

Pulmonary stenosis with intact ventricular septum

++

No common chromosomal associations

Cardiac defects in chromosomally abnormal fetuses

closure of some septal defects in utero – which appears to be common in infancy and has been reported prenatally (Figure 42.1).26–28 Second, intrauterine mortality (which is 30% in trisomy 21 fetuses between 12 and

Figure 42.1 Perimembranous ventricular septal defect (arrow) partially guarded by the septal leaflet of the tricuspid valve in a fetal heart from a 13-week trisomy 21 fetus. T, septal leaflet of the tricuspid valve; C, crista supraventricularis; P, pulmonary valve; RV, right ventricle. Scale bar: 1 mm.

Table 42.2

623

40 weeks) may be higher in fetuses with a cardiac defect compared to those without.29 Third, there may be some selection bias in the prenatal postmortem series, as many of these fetuses had increased nuchal translucency thickness (NT), and there is a recognized association between increasing NT and both cardiac defects and intrauterine lethality.24,30 As the rates of aneuploidy associated with the prenatal finding of an atrioventricular septal defect or ventricular septal defect appear to be very high, it is normal practice to offer amniocentesis to determine the fetal karyotype in such cases.3,4,31–33 However, it is important to recognize that in unselected populations, the prenatal detection rate of an isolated ventricular septal defect is in fact very low, and that using this feature as an isolated method of screening for trisomy 21 may not be very useful.34–36 This is reinforced by the data of Palidini et al, who reviewed the cardiac findings in a series of fetuses found to have trisomy 21 prenatally.9 Fetuses were referred for echocardiography either with a diagnosis of trisomy 21 or with a suspected cardiac defect in circumstances where the karyotype was not known. In all, 56% of the 41 fetuses known to have trisomy 21 had a cardiac defect, 44% of which were atrioventricular and 48% of which were ventricular septal defects. In the group of 274 fetuses suspected of having a cardiac defect but with no karyotypic description, 9/21 (43%) cases with an atrioventricular septal defect were found to have trisomy 21 but 0/39 (0%) of those with a ventricular septal defect had this chromosomal abnormality. This study suggests that whilst karyotyping fetuses with an atrioventricular septal defect is definitely useful,

Studies describing the prevalence of cardiac defects+ in trisomy 21

Study

Population

AVSD

VSD

ASD

ToF

DORV

CoA

Other

Hyett et al. 1994 / 199724,25

prenatal / postmortem

24%

19%

–

–

–

50%*

4% Bicuspid AV 4% Bicuspid PV

Palidini et al. 20009 prenatal / ultrasound

46%

27%

5%

2%

2%

Evans 195019

postnatal / postmortem

14%

36%

29%

7%

–

–

Tubman et al. 199120

postnatal / echo

17%

7%

10%

1%

Hoe et al. 199021

postnatal / echo

6%

26%

–

–

4% PS

1% –

–

6% HOCM 3% ‘Complex’

Wells et al. 1994

22

Vida et al. 200523

postnatal / echo

19%

15%

14%

4%

–

2%

PS 3%

postnatal / echo

6%

22%

11%

–

–

–

<1% Ebsteins

+Data on liveborn infants with a patent ductus arteriosus is not included. *These were described as narrowing of the aortic isthmus rather than coarctation of the aorta. In all cases the nuchal translucency thickness at 11–14 weeks was ≥3.5 mm. AVSD – atrioventricular septal defect; VSD – ventricular septal defect; ASD – atrial septal defect; ToF – tetralogy of Fallot; DORV – double outlet right ventricle; CoA – coarctation of aorta; PS – pulmonary stenosis; HOCM – hypertrophic obstructive cardiomyopathy; AV – aortic valve; PV – pulmonary valve.

624

Fetal Cardiology

Figure 42.2 Scanning electron micrograph of the parietal aspect of the right ventricle showing marked dysplasia of both the pulmonary (P) and tricuspid (T) valves in a trisomy 18 fetus at 12 weeks of gestation. Scale bar: 500 mm.

there may be less value in karyotyping all those with a ventricular septal defect than previously thought.

Trisomy 18 Ventricular septal defects are in fact more prevalent in fetuses with trisomy 18 than in fetuses with trisomy 21, although this is not immediately apparent in most postnatal data sets as the live-birth prevalence of trisomy 21 is 10 times higher than that of trisomy 18.3,5 In a series of 23 fetuses found to be affected by trisomy 18 through first trimester screening, the fetal heart was examined after termination of pregnancy and a cardiac defect found in all cases.37 This was most commonly a ventricular septal defect (in 19/23 (83%) cases), although there was a very high prevalence of valvular abnormalities (in 19/23 (83%) cases) as well (Figure 42.2). These findings are broadly similar to those described postnatally, either through echocardiography or at postmortem examination.38–41 The earlier, postnatal studies had suggested that valvular dysplasia was a secondary pathological feature that developed later in gestation, but the findings of this first trimester series suggest that this is not necessarily the case. There was an imperforate valve in four of the nine cases with nuchal translucency thickness of > 7 mm compared to one of the 14 cases with translucency of < 7 mm, and although the postnatal series suggest that the valvular

abnormalities are of no hemodynamic significance, the high rate of valvular atresia/agenesis in this first trimester series suggests that this anomaly may be associated with a high rate of intrauterine lethality. It is also interesting that the normally relatively rare finding of a persistent left superior vena cava was more prevalent with increasing nuchal translucency, and it has been suggested that this may be the consequence of venous congestion of the head and neck resulting in increased flow through the left anterior cardinal vein and persistence, rather than obliteration of this vessel. Four recent series have described prenatal detection rates of between 94 and 100% for trisomy 18.42–45 DeVore demonstrated that the detection rate for this chromosomal abnormality increased from 77 to 97% by including a detailed examination of the fetal heart,42 and in the other three series cardiac anomalies were the commonest anomalies found (ranging from 47 to 84% of cases).43–45 In each case the commonest cardiac anomaly seen was a ventricular septal defect. Other authors have also commented on the value of defining the cardiac defect as a means to making the diagnosis of trisomy 18. In one series of 3151 pregnancies undergoing first trimester (nuchal translucency) screening followed by detailed echocardiography for those with NT > 3.0 mm at 16 weeks’ gestation, all five cases of trisomy 18 had increased nuchal translucency and a cardiac defect – in comparison to only two of the nine cases of trisomy 21 having both features.46 The largest series of prenatal echocardiographic findings of trisomy 18 fetuses also reports a large cohort that had nuchal translucency screening, and therefore frequently had a relatively early echo. In this series 118/162 (73%) of the trisomy 18 fetuses had cardiac defects detected during the prenatal echo – with the same diversity of findings as described by other authors.47 Another case report has suggested that evaluation of the ductus venosus, which has been shown to have an abnormal (reversed) A-wave in many fetuses with cardiac abnormality, may lead to the diagnosis of trisomy 18 in cases where the nuchal translucency was in fact within normal limits.48 From a clinical perspective, it would appear to be sensible to recognize the strong association between trisomy 18 and ventricular septal defects. Further cardiac evaluation to look for evidence of valvular abnormalities is worthwhile, as is further assessment of the fetus to look for other structural abnormalities or markers of aneuploidy. These features should be visible in the majority of cases when a targeted examination is performed, but the diagnosis of chromosomal abnormality can only be confirmed by an invasive test such as amniocentesis.

Trisomy 13 Trisomy 13 is relatively rare, being the associated chromosomal anomaly in only 6/129 (4.7%) cases of congenital

Cardiac defects in chromosomally abnormal fetuses

heart disease reported by Tennstedt et al and in 5/355 (1.4%) cases reported by Palidini et al, compared to 15% of cases associated with trisomy 21 in each of these series.4,49 Given the small numbers of cases reported in most series it is difficult to make comments about the pattern of cardiac abnormalities commonly seen with this chromosomal abnormality. There are, however, some pathological data, related to a series of 15 fetuses diagnosed through a nuchal translucency screening program – although it is, once again, important to remember the potential impact of intrauterine lethality on the pattern of malformations that would be expected at later gestations. All 15 of these fetuses affected by trisomy 13 had a cardiac defect, and these were often suggestive of cell migration abnormalities.25 The commonest defects seen were atrioventricular and ventricular septal defects (Figure 42.3), and there were also a variety of valvular defects, including agenesis of the pulmonary valve that was not observed in the other chromosomal abnormalities reported in this series. The great arteries were also abnormal in all cases. In three cases there was truncus arteriosus – also a finding unique to trisomy 13. Other specimens showed evidence of narrowing of the aortic isthmus. Although the prevalence of cardiac defects in this pathological series is higher than that reported through prenatal obstetric ultrasound screening or postnatal echocardiography, the types of defect reported are similar.38,50,51 In a series of 28 cases of trisomy 13 fetuses the commonest structural anomalies were central nervous system (CNS) or facial abnormalities – in comparison to trisomy 18 where cardiac defects were the commonest findings.51 Some 54% of cases had a cardiac defect identified during prenatal ultrasound assessment, the predominant features being a ventricular septal defect, significant ventricular disproportion, and abnormalities of the outflow tracts and great arteries. In contrast to trisomy 21, the intracardiac defect seen in trisomy 13 is often a component of a more complex cardiac anomaly, and abnormalities of the outflow tracts and great arteries are significant features.

Del 22q11 Although a common arterial trunk was seen in 3/15 (20%) of the first trimester trisomy 13 specimens, this does not appear to be such a common association postnatally. In a series of 23 cases of a common arterial trunk, only two (8.7%) had a major chromosomal abnormality (one trisomy 13, one trisomy 22), but six of 19 (31.6%) had a microdeletion for 22q11.52 This microdeletion, which can now be recognized at a molecular level using fluorescent in situ hybridization (FISH), appears to have a very significant association with conotruncal anomalies. The prevalence of the 22q11 microdeletion has been estimated at 1 in 4000 pregnancies, and consequently, in liveborn

625

Figure 42.3 The septal aspects of the right ventricle showing a type 1 atrioventricular septal defect (O). The right ventricular outflow tract has collapsed partially during processing of the specimen (arrow). A, right atrium; V, common atrioventricular valve; C, crista supraventricularis. Scale bar: 500 mm.

infants this may in fact be the second commonest chromosomal abnormality seen in association with cardiac defects – after trisomy 21.53 Microdeletion of 22q11 presents with a variety of clinical phenotypes, and is recognized as the underlying chromosomal aberration in DiGeorge, velocardiofacial, and CATCH-22 (cardiac defects, abnormal facies, thymic hypoplasia, cleft palate, and hypocalcemia, associated with 22q11 deletion) syndromes.54 Postnatally, affected infants are frequently hypocalcemic, immunodeficient, have some dysmorphic features including palatial anomalies, and have learning difficulties, features that cannot be described prenatally. Cardiac anomalies are seen in approximately 75% of affected infants and may be the most recognizable feature from an antenatal perspective. Several groups have also described the prevalence of a 22q11 microdeletion in a variety of conotruncal anomalies, including tetralogy of Fallot (16% prevalence), truncus arteriosus (35% prevalence), and an interrupted aortic arch (50% prevalence).55–57 Other conotruncal anomalies seem to have a weaker association – such as a double-outlet right ventricle, where only 5% of cases showed this association – but even with this level of risk inclusion of FISH assessment of the 22q11 region should be requested during fetal karyotyping.58 Recently, Chaoui et al have suggested that once an anomaly of the outflow septum or aortic arch is recognized, the fetal thymus should be assessed, as hypoplasia has a strong association with positive diagnosis of a 22q11 microdeletion and this may allow clinicians to be more selective about the cases they choose to karyotype.59

626

Fetal Cardiology

Turner syndrome Two large series of liveborn female infants affected by Turner syndrome (45,X) have examined the prevalence of associated cardiac defects. These gave an overall prevalence of structural cardiac defects in 26% and 23% of cases, including abnormalities of the aortic valve (predominant bicuspid) in 18% and 15.7% and coarctation of the aorta in 10% and 6.9% of cases, respectively.60,61 This is significantly different from an admittedly much smaller series (n = 6) of hearts of Turner syndrome fetuses examined at 12–14 weeks of pregnancy, where the prevalence of cardiac defects, and in particular of coarctation of the aorta, was 100%.25 In all six cases there was tubular hypoplasia of the aortic arch (Figure 42.4), consistent with the postnatal finding of coarctation of the aorta, and two of the five hearts available for examination had an anomaly: in one case a ventricular septal defect and in another a bicuspid aortic valve. Infants with Turner syndrome may have several phenotypic features, including webbing of the neck, which is thought to be the sequela of jugular lymphatic obstruction resulting in the development of cystic hygroma in fetal life.62 In liveborn children, a triad of findings: the chromosomal abnormality 45,X, the phenotypic appearance of webbing in the posterior aspect of the neck, and coarctation of the aorta, are recognized as being strongly associated with one another.63,64 In the fetal pathology data presented here, all six cases had increased nuchal translucency, and this feature had led to invasive testing by chorionic villus sampling and determination of the chromosomal abnormality25 – so it may be the bias of patient selection that leads to a markedly increased prevalence of coarctation of the aorta in this data set. The high incidence of cardiac defects in Turner syndrome fetuses was confirmed in a study of 53 45,X fetuses that had prenatal echocardiography.65 Forty-seven of these cases were identified on the basis of increased nuchal translucency in the first trimester of pregnancy and intrauterine death or termination of pregnancy occurred in 51 of the cases, leaving two liveborn fetuses, both with normal hearts. Reviewing the fetal data, 62% of these fetuses had a cardiac defect, including 45% with coarctation of the aorta and 13% with a hypoplastic left heart.

Other chromosomal abnormalities An analysis of data combined from 11 European congenital anomaly registries found that 1738 of the 7758 anomalies were associated with chromosomal abnormality.66 In all, 114 of these chromosomal abnormalities were described as rare autosomal anomalies, including deletions, duplications, atypical trisomies, and unbalanced rearrangements. In 13.2% of these cases, a cardiac defect was seen prenatally. These data suggest that although other

Figure 42.4 The ascending aorta (Ao) is hypoplastic and the aortic arch (arrow) is extremely hypoplastic in this 12-week fetus with Turner syndrome. The ductus arteriosus (D) is dilated. PT, pulmonary trunk. Scale bar: 1 mm.

chromosomal abnormalities are far less common than those described above, there is a potential association with structural cardiac defects, and karyotyping should always be considered if these chromosomal anomalies are to be recognized prenatally. From the other perspective, many fetuses with macrodeletions or duplications may have a cardiac defect. This includes the deletions 3q, 4q, 5p, 8p, 9p, 11q, 13q, 18p, and 18q and the duplications 1p, 2p, 2q, 3q, 5p, 8p, 13q, and 16q.67 The breakpoints for these anomalies often vary between cases, making the prediction of associated anomalies difficult, and there is in fact often little correlation between cases even in circumstances where the breakpoints are the same. The development of molecular genetic investigations such as fluorescent in situ hybridization has led to the recognition that some conditions, such as Williams syndrome, are actually caused by a microdeletion – in this case del 7q11.23. Williams syndrome had previously been described as an autosomal dominant genetic condition, and is recognized as being associated with supravalvular aortic stenosis and/or peripheral pulmonary stenosis.68 Less subtle chromosomal aberrations that have clear associations with specific cardiac defects include tetrasomy 22p (cat-eye syndrome) with a 30% prevalence of cardiac defects, frequently total anomalous pulmonary venous drainage, and tetrasomy 12p (Pallister–Killian syndrome), which has a 25% prevalence of cardiac defects, predominantly ventricular septal defects.69,70 Septal defects are also

Cardiac defects in chromosomally abnormal fetuses

627

the major cardiac feature in fetuses with Wolf–Hirschhorn syndrome (del 4p), and are seen in approximately 30% of cases.71 The study of Baena et al, which found that 6.6% of chromosomal abnormalities were atypical and that 13.2% of these cases had an underlying cardiac defect that was recognized prenatally, also reported that 20.6% of these fetuses had a cystic hygroma.66 The significance of nuchal edema and cystic hygroma in relation to both cardiac defects and chromosomal abnormalities is discussed in more detail in the section below, but it is interesting to see that this association seems to be as important in atypical chromosomal aberrations as it is in those that are more commonly described.

Ultrasound markers of chromosomal abnormality and/or cardiac defects Although tertiary centers have been shown to offer highly sensitive and specific prenatal ultrasound screening for cardiac defects, routine assessment of the fourchamber view and outflow tracts at the 20-week morphology scan have been shown to be less successful as a screening tool.34–36,72,73 In order to improve the antenatal detection of cardiac defects, it may be helpful to consider the role of other morphological abnormalities seen during screening for chromosomal abnormalities that are easier to identify but are recognized as having an association with cardiac defects. Hydrops fetalis, defined as excessive accumulation of fluid in at least two serous cavities or body tissues, is readily identifiable during an antenatal scan. It is associated with a high rate of perinatal mortality and neonatal morbidity and represents the end-stage of many disease processes. Since the introduction of rhesus-D prophylaxis against immune causes, the two most common underlying etiologies are structural cardiac defects and chromosomal abnormalities, which occur in approximately 20% and 15% of cases, respectively.74–80 The commonest chromosomal abnormality associated with fetal hydrops is Turner syndrome – reported in up to 15% of cases – which has the associations with coarctation of the aorta described above.74 Whilst hydrops, which describes fluid accumulation at multiple sites, is rare, fluid accumulating in the posterior aspect of the neck is more common. This is defined as either nuchal edema or cystic hygroma (on the basis of morphological characteristics) in the second trimester and as increased nuchal translucency in the first trimester. As prenatal scans were initially performed at 18–20 weeks, the associations of nuchal edema and cystic hygroma were first described at this stage. Both are associated with

Figure 42.5 An ultrasound image of a fetus of 12 weeks’ gestation showing the measurement of fetal nuchal translucency. The measurement of 3.0 mm increases the risk of a chromosomal abnormality above the background risk based upon maternal age.

chromosomal abnormality, although the cystic hygroma is more commonly seen with 45,X and nuchal edema with trisomies.81,82 Pathologically, a cystic hygroma (45,X) is associated with obstruction of the jugular lymphatics with a paucity of peripheral lymphatic channels, whilst nuchal edema (trisomy 21) is associated with an increase in the number of small peripheral lymphatics.83–85 In contrast, in the first trimester, not only are the morphological characteristics of nuchal swellings harder to define, but those that would be described as being a hygroma are as commonly associated with trisomies as with 45,X, and the feature that is most predictive of an association with chromosomal abnormality is the thickness of this swelling.86–88 It therefore makes sense to describe these lesions as one group – using the term increased nuchal translucency (Figure 42.5). Although pathological observations show that lymphatic anomalies are still present in these first trimester fetuses, the precise etiology of nuchal translucency, and why this fluid is seen in very high proportions of chromosomally abnormal fetuses at 12 weeks in comparison to the proportion having nuchal edema or cystic hygroma at 20 weeks, are not fully understood.89–92 It has been suggested that chromosomally abnormal fetuses with a large nuchal translucency measurement are more likely to have both intracardiac and great artery anomalies.24,25,92 The assessment of nuchal translucency thickness has been shown to be a very effective method of screening for Down syndrome, and forms the basis of universal screening programs in many countries.93,94 The fact that increased nuchal translucency also has a strong association with

628

Fetal Cardiology

cardiac defects means that this can potentially also be used to screen for cardiac defects. This was first reviewed in a retrospective series of 29 154 pregnancies primarily screened for chromosomal abnormality at 11–14 weeks of pregnancy.95 In this series, 56% (95% CI 42–70%) of the fetuses later found to have a cardiac defect (by postmortem, fetal echocardiography, or postnatal examination) had increased nuchal translucency, suggesting that this tool was a more powerful marker in screening a low-risk population than the traditional technique of assessing the four-chamber view and outflow tracts at 20 weeks. Many other groups have examined the role of nuchal translucency as a screening tool for cardiac defects in euploid fetuses, and these studies suggest that the potential is more modest than was originally described.96–98 In some series, the apparent low sensitivity of the test is due to population bias in the manner that increased nuchal translucency is defined.99,100 Despite this, virtually all these studies have shown a 7% prevalence of major cardiac defects in a population with NT > 99th centile, twice the rate seen in fetuses of diabetic women, who are considered to merit detailed fetal echocardiography.101 Virtually all clinicians agree that although nuchal translucency screening has not replaced formal ultrasound evaluation of the heart at a later stage of pregnancy, the finding of increased nuchal translucency in a euploid fetus warrants further careful investigation. In some centers, this has led to the development of skills in first trimester echocardiography which, in specialist hands, appears to have high sensitivity and specificity for the detection of cardiac anomalies.102–105 Molecular studies in first trimester fetuses suggest that the myocardium of fetuses with increased nuchal translucency and chromosomal abnormality have some cardiac dysfunction.106 This leads to the investigation of other potential markers of hemodynamic imbalance, and although no significant change in the myocardial performance index of fetuses with increased nuchal translucency could be demonstrated, there does appear to be an association with abnormal flow through the ductus venosus.107,108 The ductus venosus shunts oxygenated blood from the umbilical vein directly to the coronary and cerebral circulations, with high pulsatile flow and forward velocity throughout the cardiac cycle. This vessel has been used to assess cardiac dysfunction in fetuses with severe intrauterine growth restriction that have evidence of myocardial decompensation.109 In a series of 515 pregnancies having cytogenetic testing at 11–14 weeks’ gestation, 80% of cases with chromosomal abnormality were also found to have an abnormal ductus venosus – defined by absent or reversed flow in the ‘A’ wave. The specificity of this test appears to be very high (approximately 99%), and in 7/17 (41%) euploid fetuses in this series that had increased nuchal translucency and an abnormal ductus venosus waveform, a major cardiac defect was found at a later date.109 This has been supported by several other data sets, which have suggested that measurement of ductal flow in terms

of the pulsatility for veins (PVIV) is more reliable, and that inclusion of this tool in first trimester testing improves the sensitivity and specificity of screening for chromosomal abnormality and is a powerful marker for euploid fetuses that have major cardiac defects.110–114 In comparison to its value in the first trimester, which appears to extend to the early part (15–17 weeks) of the second trimester, studies have shown that the ductus venosus is not markedly abnormal in fetuses with cardiac defects in later pregnancy, and its use in screening for cardiac defects in euploid fetuses is therefore probably restricted to the 11–14-week scan.115,116 Another cardiac marker of blood flow, tricuspid regurgitation, has also recently been reported as being associated with both chromosomal abnormalities and cardiac defects, although this too seems to be an association limited to early gestations.117 In a population of fetuses referred for detailed echocardiography at 11–14 weeks, after the identification of increased nuchal translucency (> 4 mm), tricuspid regurgitation was seen in 27% of cases. Some 83% of these were subsequently found to have a chromosomal abnormality, most commonly trisomy 21, although all types of karyotypic anomaly were seen.118 Tricuspid regurgitation was assessed by transabdominal scan, using pulsed wave Doppler. A sample gate of 3 mm was positioned across the tricuspid valve in an apical four-chamber view such that the angle of insonation was < 20°. Valvular flow was assessed three times, and tricuspid regurgitation defined as flow toward the atrium for at least half of systole with a velocity of > 60 cm/s.119 This method has been shown to be highly reproducible by sonographers with experience in fetal echocardiography, and a second larger prospective study found tricuspid regurgitation in 4.4% of chromosomally normal fetuses, 67.5% of fetuses affected by trisomy 21, and 33% of fetuses affected by trisomy 18.119 A third study performed by the same group, again involving a selected population with a high prevalence of increased nuchal translucency, chromosomal abnormalities and cardiac defects, defined tricuspid regurgitation more stringently, with flow > 80 cm/s.120 In this study, tricuspid flow was successfully examined in 96.8% of cases and tricuspid regurgitation was found in 8.5% of chromosomally normal fetuses and 65.1% of those with trisomy 21. Further analysis of the euploid population found that tricuspid regurgitation was present in 46.9% of those cases that had a structural cardiac defect and in only 5.6% of euploid fetuses with no cardiac defect. Likelihood ratios describing the associations between tricuspid regurgitation and trisomy 21 or isolated cardiac defects were calculated – being 7.7 and 8.4, respectively. The authors have noted an association with crown– rump length and nuchal translucency thickness, and more precise likelihood ratios can be used based on these parameters. Only one other group has published any data related to the finding of tricuspid regurgitation in early pregnancy – with data collected at 14–16 weeks of pregnancy.

Cardiac defects in chromosomally abnormal fetuses

These authors reported that tricuspid regurgitation was a much commoner finding, being present in 83.4% of cases, and consequently was of no clinical value in identifying fetuses with either chromosomal or cardiac defects.121 The accompanying editorial to this paper questioned the validity of the technique used (spatiotemporal image correlation, STIC) and the stringency of the criteria used for diagnosis. It is important to note, however, that whilst the initial data suggest that the assessment of tricuspid regurgitation will be a valuable tool in screening for chromosomal abnormality, with the potential adjunctive benefit of identifying fetuses at high risk for cardiac defects, further data are needed to demonstrate that this can be applied outside of a large tertiary institution. Similarly, there are currently no large prospective studies detailing the effectiveness of this tool in screening for cardiac defects, and, due to the relatively low prevalence of disease in a low-risk population, this is likely to take some time to generate. We appear to be in the process of a shift of emphasis from second trimester (18–20 weeks) to first trimester screening. Whilst this has primarily occurred due to the success of combined first trimester screening for chromosomal abnormality, it has had the added benefit of enabling universal assessment of cardiac defects with the cardiac ‘marker’ increased nuchal translucency. It is highly likely that as more sonographers develop their confidence with a rudimentary assessment of the heart in the first trimester, potentially including the markers of an abnormal ductus venosus or tricuspid regurgitation in the process of first trimester screening, that the detection rate of cardiac anomalies at this early stage of pregnancy will increase.

The molecular basis of cardiac defects seen in chromosomally abnormal fetuses The recognition of specific associations between chromosomal abnormalities and cardiac defects may be of some value in identifying genes, and genetic mutations involved in cardiogenesis. In conditions such as trisomy 21, which is commonly associated with atrioventricular septal defects, cardiac anomalies appear to be associated with the distal region of chromosome 21q, and abnormal expression of genes such as DS-CAM have been implicated.122 Gene dosage is frequently not 1.5 times that expected in euploid fetuses, and appears to vary between different tissues for the same gene, suggesting that the relationship between trisomy and phenotypic outcome is not merely related to gene dosage but that there are more complex interactions in play.123 This makes assessment of candidate genes difficult, and consequently there has been minimal process in

629

determining the underlying molecular genetic cause of cardiac defects in trisomies 21, 18, and 13. Molecular techniques such as FISH that can readily identify microdeletions, linkage analysis, and DNA mutation analysis have developed rapidly over the last 10–15 years, making investigation of the molecular basis of disorders such as DiGeorge syndrome easier. The genome project has also described the gene loci for some chromosomal regions – so that a short list of genes can be identified based on the location of the microdeletion. One study, comparing groups of patients with clinical characteristics of DiGeorge syndrome who were identified as having, or not having, the 22q11 microdeletion has demonstrated involvement of the T-box transcription factor TBX1, and this relationship is now being investigated further in knockout mouse models.124 In a similar manner, the underlying molecular anomaly associated with Williams syndrome (del 7q11) has been recognized as haploinsufficiency of the Elastin gene.125 At another level, mutation analysis has now been used to identify the underlying molecular mutation associated with many genetic syndromes that include cardiac anomalies, and in some cases molecular testing can be used to reach a firm diagnosis prenatally.126,127

Conclusions The strong association between cardiac defects and chromosomal abnormality initially recognized in populations of infants and neonates is even stronger in fetal life. The heart should be carefully examined in fetuses that have other markers of chromosomal abnormality, as the identification of a particular cardiac defect may help the clinician to define the level of risk for aneuploidy and help the patient make a decision with respecting to karyotyping. The identification of structural anomalies, or the unexpected finding of chromosomal abnormality, should lead to review of the heart to ensure that no defect is present. First trimester screening for chromosomal abnormality is becoming more prevalent, and there is a well recognized association between increased nuchal translucency and cardiac defects. Whilst this association is not strong enough to advocate using nuchal translucency as an isolated screening tool, it is important to recognize the risk of cardiac defects in fetuses with increased nuchal translucency and arrange appropriate expert echocardiography at a later stage. Other first trimester ultrasound screening tools for chromosomal abnormality may also have relevance to the detection of cardiac defects – although further data are needed to support this. The strong association of certain defects with specific chromosomal abnormalities, in particular microdeletions, has allowed characterization of genes involved in cardiogenesis and is likely to add to our understanding of this complex process in the longer term.

630

Fetal Cardiology

References 1. Ferencz C, Rubin JD, McCarter RJ et al. Congenital heart disease: prevalence at livebirth. Am J Epidemiol 1985; 121: 31–6. 2. Berg KA, Clark EB, Astemborski JA et al. Prenatal detection of cardiovascular malformations by echocardiography: an indication for cytogenetic evaluation. Am J Obstet Gynecol 1988; 159: 477–81. 3. Allan LD, Sharland GK, Milburn A et al. Prospective diagnosis of 1,006 consecutive cases of congenital heart disease in the fetus. J Am Coll Cardiol 1994; 23: 1452–8. 4. Tennstedt C, Chaoui R, Korner H, Dietel M. Spectrum of congenital heart defects and extracardiac malformations associated with chromosomal abnormalities: results of a seven year necropsy study. Heart 1999; 82: 34–9. 5. Boldt T, Andersson S, Eronen M. Outcome of structural heart disease diagnosed in utero. Scand Cardiovasc J 2002; 36: 73–9. 6. Samanek M, Goetzova J, Benesova D. Distribution of congenital heart malformations in an autopsied child population. Int J Cardiol 1985; 8: 235–50. 7. Fesslova V, Nava S, Villa L. Evolution and long term outcome in cases with fetal diagnosis of congenital heart disease: Italian multicentre study. Heart 1999; 82: 594–9. 8. Allan LD. Atrioventricular septal defects in the fetus. Am J Obstet Gynecol 1999; 181: 1250–3. 9. Paladini D, Tartaglione A, Agangi A et al. The association between congenital heart disease and Down syndrome in prenatal life. Ultrasound Obstet Gynecol 2000; 15: 104–8. 10. Fesslova V, Villa L, Nava S et al. Spectrum and outcome of atrioventricular septal defect in fetal life. Cardiol Young 2002; 12: 18–26. 11. Huggon IC, Cook AC, Smeeton NC et al. Atrioventricular septal defects diagnosed in fetal life: associated cardiac and extra-cardiac abnormalities and outcome. J Am Coll Cardiol 2000; 36: 593–601. 12. Ashok M, Thangavel G, Indrani S et al. Atrioventricular septal defect--associated anomalies and aneuploidy in prenatal life. Indian Pediatr 2003; 40: 659–64. 13. Berg C, Geipel A, Smrcek J et al. Prenatal diagnosis of cardiosplenic syndromes: a 10 year experience. Ultrasound Obstet Gynecol 2003; 22: 451–9. 14. Splitt M, Burn J, Goodship J. Defects in the determination of left-right asymmetry. J Med Genet 1996; 33: 498–503. 15. Machlitt A, Heling KS, Chaoui R. Increased cardiac atrialto-ventricular length ratio in the fetal four-chamber view: a new marker for atrioventricular septal defects. Ultrasound Obstet Gynecol 2004; 24: 618–22. 16. Langford K, Sharland G, Simpson J. Relative risk of abnormal karyotype in fetuses found to have an atrioventricular septal defect (AVSD) on fetal echocardiography. Prenat Diagn 2005; 25: 137–9. 17. Delisle MF, Sandor GG, Tessier F et al. Outcome of fetuses diagnosed with atrioventricular septal defect. Obstet Gynecol 1999; 94: 763–7. 18. Lange R, Guenther T, Busch R, Hess J, Schreiber C. The presence of Down syndrome is not a risk factor in complete atrioventricular septal defect repair. J Thorac Cardiovasc Surg 2007; 134: 304–10.

19. Evans PR. Cardiac anomalies in mongolism. Br Heart J 1950; 12: 258–62. 20. Tubman TRJ, Shields MD, Craig BG et al. Congenital heart disease in Down’s syndrome: two year prospective early screening study. Br Med J 1991; 302: 1425–7. 21. Hoe TS, Chan KC, Boo NY. Cardiovascular malformations in Malaysian neonates with Down’s syndrome. Singapore Med J 1990; 31: 474–6. 22. Wells GL, Barker SE, Finley SC et al. Congenital heart disease in infants with Down’s syndrome. South Med J 1994; 87: 724–7. 23. Wessel MW, Los FJ, Wladimiroff JW. Prenatal ultrasound findings and final outcome in fetal trisomy 21. Presented at Seventh World Congress on Ultrasound in Obstetrics and Gynecology, Washington, DC, 1997. 24. Hyett JA, Moscoso G, Nicolaides KH. First-trimester nuchal translucency and cardiac septal defects in fetuses with trisomy 21. Am J Obstet Gynecol 1995; 172: 1411– 13. 25. Hyett JA, Moscoso G, Nicolaides KH. Abnormalities of the heart and great arteries in first trimester chromosomally abnormal fetuses. Am J Med Genet 1997; 69: 207– 16. 26. Moe DG, Guntheroth WG. Spontaneous closure of uncomplicated ventricular septal defect. Am J Cardiol 1987; 60: 674–8. 27. Rao PS. Further observations on the spontaneous closure of physiologically advantageous ventricular septal defects in tricuspid atresia: surgical implications. Ann Thorac Surg 1983; 35: 121–31. 28. Nir A, Weintraub Z, Oliven A, Kelener J, Lurie M. Anatomic evidence of spontaneous intrauterine closure of a ventricular septal defect. Pediatr Cardiol 1990; 11: 208–10. 29. Snijders RJM, Sundberg K, Holzgreve W, Henry G, Nicolaides KH. Maternal age- and gestation-specific risk for trisomy 21. Ultrasound Obstet Gynecol 1999; 13: 167–70. 30. Hyett JA, Sebire NJ, Snijders RJM, Nicolaides KH. Intrauterine lethality of trisomy 21 fetuses with increased nuchal translucency thickness. Ultrasound Obstet Gynecol 1996; 7: 101–3. 31. Paladini D, Calabro R, Palmiere S et al. Prenatal diagnosis of congenital heart disease and fetal karyotyping. Obstet Gynecol 1993; 81: 679–82. 32. Schwanitz G, Zerres K, Gembruch U et al. Prenatal detection of heart defects as an indication for chromosome analysis. Ann Genet 1990; 33: 79–83. 33. Raymond FL, Simpson JM, Sharland GK et al. Fetal echocardiography as a predictor of chromosomal abnormality. Lancet 1997; 350: 930. 34. Grandjean H, Larroque D, Levi S et al. The performance of routine ultrasonographic screening of pregnancies in the Eurofetus study. Am J Obstet Gynecol 1999; 181: 446–54. 35. Wong SF, Chan FY, Cincotta RB et al. Factors influencing the prenatal detection of structural heart diseases. Ultrasound Obstet Gynecol 2003; 21: 19–25. 36. Chew C, Halliday JL, Riley MM et al. Population-based study of antenatal detection of congenital heart disease by ultrasound examination. Ultrasound Obstet Gynecol 2007; 29: 619–24.

Cardiac defects in chromosomally abnormal fetuses

37. Hyett JA, Moscoso G, Nicolaides KH. Cardiac defects in 1st-trimester fetuses with trisomy 18. Fetal Diagn Ther 1995; 10: 381–6. 38. Musewe NN, Alexander DJ, Teshima I et al. Echocardiographic evaluation of the spectrum of cardiac anomalies associated with trisomy 13 and trisomy 18. J Am Coll Cardiol 1990; 15: 673–7. 39. Balderston SM, Shaffer EM, Washington RL et al. Congenital polyvalvular disease in trisomy 18: echocardiographic diagnosis. Pediatr Cardiol 1990; 11: 138–42. 40. Van Praagh S, Truman T, Firpo A et al. Cardiac malformations in trisomy 18: a study of 41 postmortem cases. J Am Coll Cardiol 1989; 13: 1586–97. 41. Matsuoka R, Misugi K, Goto A et al. Congenital heart anomalies in the trisomy 18 syndrome, with reference to congenital polyvalvular disease. Am J Med Genet 1983; 14: 657–68. 42. DeVore GR. Second trimester ultrasonography may identify 77–97% of fetuses with trisomy 18. J Ultrasound Med 2000; 19: 565–76. 43. Papp C, Ban Z, Szigeti Z et al. Role of second trimester sonography in detecting trisomy 18: a review of 70 cases. J Clin Ultrasound 2007; 35: 68–72. 44. Yang JH, Chung JH, Shin JS et al. Prenatal diagnosis of trisomy 18: report of 30 cases. Prenat Diagn 2005; 25: 119–22. 45. Yeo L, Guzman ER, Day-Salvatore D et al. Prenatal detection of fetal trisomy 18 through abnormal sonographic features. J Ultrasound Med 2003; 22: 581–90. 46. Cheng PJ, Liu CM, Chueh HY et al. First-trimester nuchal translucency measurement and echocardiography at 16 to 18 weeks of gestation in prenatal detection for trisomy 18. Prenat Diagn 2003; 23: 248–51. 47. Moyano D, Huggon IC, Allan LD. Fetal echocardiography in trisomy 18. Arch Dis Child Fetal Neonatal Ed 2005; 90: F520–2. 48. Martínez Crespo JM, Del Río M, Gómez O et al. Prenatal diagnosis of hypoplastic left heart syndrome and trisomy 18 in a fetus with normal nuchal translucency and abnormal ductus venosus blood flow at 13 weeks of gestation. Ultrasound Obstet Gynecol 2003; 21: 490–3. 49. Paladini D, Russo MG, Teodoro A et al. Prenatal diagnosis of congenital heart disease in the Naples area during the years 1994–1999 – the experience of a joint fetal-pediatric cardiology unit. Prenat Diagn 2002; 22: 545–52. 50. Lehman CD, Nyberg DA, Winter TC et al. Trisomy 13 syndrome: prenatal US findings in a review of 33 cases. Radiology 1995; 194: 217–22. 51. Papp C, Szigeti Z, Toth-Pal E et al. Ultrasonographic findings of fetal aneuploidies in the second trimester – our experiences. Fetal Diagn Ther 2008; 23: 97–105. 52. Volpe P, Paladini D, Marasini M et al. Common arterial trunk in the fetus: characteristics, associations, and outcome in a multicentre series of 23 cases. Heart 2003; 89: 1437–41. 53. Burn J, Wilson DI, Cross I et al. The clinical significance of 22q11 deletion. In: Clark EB, Markwald RR, Takao A, eds. Developmental Mechanisms of Heart Disease. New York: Futura, 1995: 559–67. 54. Ryan AK, Goodship JA, Wilson DI et al. Spectrum of clinical features associated with interstitial chromosome

55.

56.

57.

58.

59.

60.

61.

62.

63.

64.

65. 66.

67.

68.

69. 70.

71.

72.

631

22q11 deletions: a European collaborative study. J Med Genet 1997; 34: 798–804. Goldmuntz E, Clark BJ, Mitchell LE et al. Frequency of 22q11 deletions in patients with conotruncal defects. J Am Coll Cardiol 1998; 32: 492–8. Takahashi K, Kido S, Hoshino K et al. Frequency of a 22q11 deletion in patients with conotruncal cardiac malformations: a prospective study. Eur J Pediatr 1995; 154: 878–81. Webber SA, Hatchwell EL, Barber JCK et al. Importance of microdeletions of chromosomal region 22q11 as a cause of selected malformations of the ventricular outflow tracts and aortic arch: a three-year prospective study. J Pediatr 1996; 129: 26–32. Raymond FL, Simpson JM, Mackie CM et al. Prenatal diagnosis of 22q11 deletions: a series of five cases with congenital heart defects. J Med Genet 1997; 34: 679–82. Chaoui R, Kalache KD, Heling KS et al. Absent or hypoplastic thymus on ultrasound: a marker for deletion 22q11.2 in fetal cardiac defects. Ultrasound Obstet Gynecol 2002; 20: 546–52. Gotzsche CO, Krag-Olsen B, Nielsen J et al. Prevalence of cardiovascular malformations and association with karyotype in Turner’s syndrome. Arch Dis Child 1994; 71: 433–6. Mazzanti L, Cacciari E. Congenital heart disease in patients with Turner’s syndrome. Italian Study Group for Turner Syndrome. J Pediatr 1998; 133: 688–92. Chervenak FA, Isaacson G, Blakemore KJ et al. Fetal cystic hygroma. Cause and natural history. N Engl J Med 1983; 309: 822–5. Clark EB. Neck web and congenital heart defects: a pathogenic association in 45 X-O Turner syndrome? Teratology 1984; 29: 355–61. Berdahl LD, Wenstrom KD, Hanson JW. Web neck anomaly and its association with congenital heart disease. Am J Med Genet 1995; 56: 304–7. Surerus E, Huggon IC, Allan LD. Turner’s syndrome in fetal life. Ultrasound Obstet Gynecol 2003; 22: 264–7. Baena N, De Vigan C, Cariati E et al. Prenatal detection of rare chromosomal autosomal abnormalities in Europe. Am J Med Genet 2003; 118A: 319–27. Prasad C, Chudley AE. Genetics and cardiac anomalies: the heart of the matter. Indian J Pediatr 2002; 69: 321–32. Zalzstein E, Moes CA, Musewe NN et al. Spectrum of cardiovascular anomalies in Williams-Beuren syndrome. Pediatr Cardiol 1991; 12: 219–23. Freedom RM, Gerald PS. Congenital heart disease and the ‘Cat Eye’ syndrome. Am J Dis Child 1973; 126: 16–18. Reynolds JF, Daniel A, Kelly TE et al. Isochromosome 12p mosaicism (Pallister mosaic aneuploidy or Pallister-Killiak syndrome): report of 11 cases. Am J Med Genet 1987; 27: 257–74. Battaglia A, Carey JC, Cederholm P et al. Natural history of Wolf-Hirschhorn syndrome: experience with 15 cases. Pediatrics 1999; 103: 830–6. Copel JA, Pilu G, Green J et al. Fetal echocardiographic screening for congenital heart disease: the importance of the four-chamber view. Am J Obstet Gynecol 1987; 157: 648–55.

632

Fetal Cardiology

73. Stümpflen I, Stümpflen A, Wimmer M et al. Effect of detailed fetal echocardiography as part of routine prenatal ultrasonographic screening on detection of congenital heart disease. Lancet 1996; 348: 854–7. 74. Machin GA. Hydrops revisited: literature review of 1414 cases published in the 1980s. Am J Med Genet 1989; 34: 366–90. 75. Hansmann M, Gembruch U, Bald R. New therapeutic aspects in nonimmune hydrops fetalis based on 402 prenatally diagnosed cases. Fetal Ther 1989; 4: 29–36. 76. McCoy MC, Katz VL, Gould N et al. Nonimmune hydrops after 20 weeks’ gestation: review of 10 years’ experience with suggestions for management. Obstet Gynecol 1995; 85: 578–82. 77. Lallemand AV, Doco-Fenzy M, Gaillard DA. Investigation of nonimmune hydrops fetalis: multidisciplinary studies are necessary for diagnosis – review of 94 cases. Pediatr Dev Pathol 1999; 2: 432–9. 78. Ismail KMK, Martin WL, Ghosh S et al. Etiology and outcome of hydrops fetalis. J Matern Fetal Med 2001; 10: 175–81. 79. Trainor B, Tubman R. The emerging pattern of hydrops fetalis – incidence, aetiology and management. Ulster Med J 2006; 75: 185–6. 80. Abrams ME, Meredith KS, Kinnard P et al. Hydrops fetalis: a retrospective review of cases reported to a large national database and identification of risk factors associated with death. Pediatrics 2007; 120: 84–9. 81. Azar GB, Snijders RJ, Gosden C et al. Fetal nuchal cystic hygromata: associated malformations and chromosomal defects. Fetal Diagn Ther 1991; 6: 46–57. 82. Nicolaides KH, Azar G, Snijders RJ et al. Fetal nuchal oedema: associated malformations and chromosomal defects. Fetal Diagn Ther 1992; 7: 123–31. 83. Van der Putte SCJ. Lymphatic malformations in human fetuses. Virchows Arch 1977; 376: 233–46. 84. Chitayat D, Kalousek DK, Bamforth JS. Lymphatic abnormalities in fetuses with posterior cervical cystic hygroma. Am J Med Genet 1898; 33: 352–6. 85. Miyabara S, Sugihara H, Maehara N et al. Significance of cardiovascular malformations in cystic hygroma: a new interpretation of the pathogenesis. Am J Med Genet 1989; 34: 489–501. 86. Cullen MT, Gabrielli S, Green JJ et al. Diagnosis and significance of cystic hygroma in the first trimester. Prenat Diagn 1990; 10: 643–51. 87. Malone FD, Ball RH, Nyberg DA et al. First-trimester septated cystic hygroma: prevalence, natural history, and pediatric outcome. Obstet Gynecol 2005; 106: 288–94. 88. Molina FS, Avgidou K, Kagan KO et al. Cystic hygromas, nuchal edema, and nuchal translucency at 11–14 weeks of gestation. Obstet Gynecol 2006; 107: 678–83. 89. Haak MC, Bartelings MM, Jackson DG et al. Increased nuchal translucency is associated with jugular lymphatic distension. Hum Reprod 2002; 17: 1086–92. 90. Bekker MN, van den, Akker NM, Bartelings MM et al. Nuchal edema and venous-lymphatic phenotype disturbance in human fetuses and mouse embryos with aneuploidy. J Soc Gynecol Investig 2006; 13: 209–16.

91. von Kaisenberg CS, Nicolaides KH, Brand-Saberi B. Lymphatic vessel hypoplasia in fetuses with Turner syndrome. Hum Reprod 1999; 14: 823–6. 92. Hyett JA, Moscoso G, Nicolaides KH. Increased nuchal translucency in trisomy 21 fetuses: relation to narrowing of the aortic isthmus. Hum Reprod 1995; 10: 3049–51. 93. Nicolaides KH, Azar G, Byrne D et al. Fetal nuchal translucency: ultrasound screening for chromosomal defects in the first trimester of pregnancy. Br Med J 1992; 304: 867–9. 94. Snijders RJM, Noble P, Sebire N et al. UK multicentre project on assessment of risk of trisomy 21 by maternal age and fetal nuchal translucency thickness at 10–14 weeks of gestation. Lancet 1998; 352: 343–6. 95. Hyett J, Perdu M, Sharland G et al. Using fetal nuchal translucency to screen for major congenital cardiac defects at 10–14 weeks of gestation: population based cohort study. BMJ 1999; 318: 81–5. 96. Makrydimas G, Sotiriadis A, Huggon IC et al. Nuchal translucency and fetal cardiac defects: a pooled analysis of major fetal echocardiography centers. Am J Obstet Gynecol 2005; 192: 89–95. 97. Westin M, Saltvedt S, Bergman G et al. Is measurement of nuchal translucency thickness a useful screening tool for heart defects? A study of 16,383 fetuses. Ultrasound Obstet Gynecol 2006; 27: 632–9. 98. Müller MA, Clur SA, Timmerman E et al. Nuchal translucency measurement and congenital heart defects: modest association in low-risk pregnancies. Prenat Diagn 2007; 27: 164–9. 99. Simpson LL, Malone FD, Bianchi DW et al. Nuchal translucency and the risk of congenital heart disease. Obstet Gynecol 2007; 109: 376–83. 100. Hyett J, Sonek J, Nicolaides K. Nuchal translucency and the risk of congenital heart disease. Obstet Gynecol 2007; 109: 1455–6. 101. Hyett J. Does nuchal translucency have a role in fetal cardiac screening? Prenat Diagn 2004; 24: 1130–5. 102. Carvalho JS, Moscoso G, Tekay A et al. Clinical impact of first and early second trimester fetal echocardiography on high risk pregnancies. Heart 2004; 90: 921–6. 103. Lombardi CM, Bellotti M, Fesslova V et al. Fetal echocardiography at the time of the nuchal translucency scan. Ultrasound Obstet Gynecol 2007; 29: 249–57. 104. Becker R, Wegner RD. Detailed screening for fetal anomalies and cardiac defects at the 11–13-week scan. Ultrasound Obstet Gynecol 2006; 27: 613–18. 105. Yagel S, Cohen SM, Messing B. First and early second trimester fetal heart screening. Curr Opin Obstet Gynecol 2007; 19: 183–90. 106. Hyett JA, Brizot ML, Von Kaisenberg CS et al. Cardiac gene expression of atrial natriuretic peptide and brain natriuretic peptide in trisomic fetuses. Obstet Gynecol 1996; 87: 506–10. 107. Huggon IC, Turan O, Allan LD. Doppler assessment of cardiac function at 11–14 weeks’ gestation in fetuses with normal and increased nuchal translucency. Ultrasound Obstet Gynecol 2004; 24: 390–8. 108. Montenegro N, Matias A, Areias JC et al. Increased fetal nuchal translucency: possible involvement of early cardiac failure. Ultrasound Obstet Gynecol 1997; 10: 265–8.

Cardiac defects in chromosomally abnormal fetuses

109. Matias A, Montenegro N. Ductus venosus blood flow in chromosomally abnormal fetuses at 11 to 14 weeks of gestation. Semin Perinatol 2001; 25: 32–7. 110. Borrell A, Perez M, Figueras F et al. Reliability analysis on ductus venosus assessment at 11–14 weeks’ gestation in a high risk population. Prenat Diagn 2007; 27: 442–6. 111. Borrell A, Gonce A, Martinez JM et al. First-trimester screening for Down syndrome with ductus venosus Doppler studies in addition to nuchal translucency and serum markers. Prenat Diagn 2005; 25: 901–5. 112. Oh C, Harman C, Baschat AA. Abnormal first trimester ductus venosus blood flow: a risk factor for adverse outcome in fetuses with normal nuchal translucency. Ultrasound Obstet Gynecol 2007; 30: 192–6. 113. Bilardo CM, Muller MA, Zikulnig L et al. Ductus venosus studies in fetuses at high risk for chromosomal or heart abnormalities: relationship with nuchal translucency measurement and fetal outcome. Ultrasound Obstet Gynecol 2001; 17: 288–94. 114. Favre R, Cherif Y, Kohler M et al. The role of fetal nuchal translucency and ductus venosus Doppler at 11–14 weeks of gestation in the detection of major congenital heart defects. Ultrasound Obstet Gynecol 2003; 21: 239–43. 115. Hung JH, Fu CY, Lu JH, Hung CY. Ductus venosus blood flow resistance and congenital heart defects in the second trimester. J Clin Ultrasound 2008; 36: 72–8. 116. Gembruch U, Meise C, Germer U et al. Venous Doppler ultrasound in 146 fetuses with congenital heart disease. Ultrasound Obstet Gynecol 2003; 22: 345–50. 117. Gembruch U, Smrcek JM. The prevalence and clinical significance of tricuspid valve regurgitation in normally grown fetuses and those with intrauterine growth retardation. Ultrasound Obstet Gynecol 1997; 9: 374–82.

633

118. Huggon IC, DeFigueiredo DB, Allan LD. Tricuspid regurgitation in the diagnosis of chromosomal anomalies in the fetus at 11–14 weeks of gestation. Heart 2003; 89: 1071–3. 119. Falcon O, Faiola S, Huggon I et al. Fetal tricuspid regurgitation at the 11+0 to 13+6 week scan: association with chromosomal defects and reproducibility of the method. Ultrasound Obstet Gynecol 2006; 27: 609–12. 120. Faiola S, Tsoi E, Huggon IC et al. Likelihood ratio for trisomy 21 in fetuses with tricuspid regurgitation at the 11 to 13+6 week scan. Ultrasound Obstet Gynecol 2005; 26: 22–7. 121. Messing B, Porat S, Imbar T et al. Mild tricuspid regurgitation: a benign fetal finding at various stages of pregnancy. Ultrasound Obstet Gynecol 2005; 26: 606–10. 122. Korenberg JR, Chen XN, Schipper R et al. Down syndrome phenotypes: the consequences of chromosomal imbalance. Proc Natl Acad Sci USA 1994; 91: 4997–5001. 123. Li CM, Guo M, Salas M et al. Cell type specific over expression of chromosome 21 genes in fibroblasts and fetal hearts with trisomy 21. BMC Med Genet 2006; 7: 24–38. 124. Yagi H, Furutani Y, Hamada H et al. Role of TBX1 in human del22q11.2 syndrome. Lancet 2003; 362: 1366–73. 125. Urbán Z, Riazi S, Seidl TL et al. Connection between elastin haploinsufficiency and increased cell proliferation in patients with supravalvular aortic stenosis and WilliamsBeuren syndrome. Am J Hum Genet 2002; 71: 30–44. 126. Brennan P, Young ID. Congenital heart malformations: aetiology and associations. Semin Neonatol 2001; 6: 17–25. 127. Pierpont ME, Basson CT, Benson DW et al. Genetic basis for congenital heart defects: current knowledge. Circulation 2007; 115: 3015–38.

43 Associated anomalies in congenital heart disease Christoph Berg, Ulrich Gembruch, and Annegret Geipel Extracardiac malformations are present in 20% of infants with congenital heart defects.1 In prenatal series the percentage is considerably higher, owing to the fraction of fetuses with major cardiac anomalies or chromosomal anomalies that either die in utero or undergo termination of pregnancy and therefore are not represented in postnatal series. A further bias leading to an increased rate of associated extracardiac malformations in prenatal cardiac series is due to the fact that the detection of a major extracardiac anomaly often prompts fetal echocardiography and therefore enhances the detection rate of cardiac defects in the subset of fetuses with extracardiac anomalies. In a prospective trial in a non-selected population, Tegnander et al reported extracardiac anomalies in 65% of fetuses diagnosed with major cardiac malformations (47% in combination with chromosomal anomalies).2 However, if also less severe cardiac defects are taken into account the incidence is considerable lower. In a preceding study incorporating critical as well as non-critical cardiac malformations, the same group reported rates of extracardiac anomalies and chromosomal defects of 26% and 9%, respectively.3 Series from referral centers for fetal echocardiography are reporting rates of extracardiac malformations in their cohorts ranging from 29 to 37%, with percentages of chromosomal anomalies ranging from 18 to 26%, respectively.4,5 Complex and critical cardiac defects have a clear association with extracardiac anomalies and/or abnormal karyotype (Table 43.1), and these have a significantly worse outcome.2,3,6 Furthermore, a cardiac defect is more likely to be recognized if an extracardiac anomaly and/or an abnormal karyotype is present.2,3 In this context, increased nuchal translucency has evolved to be the most important extracardiac anomaly leading to the diagnosis of a cardiac defect.7 Therefore, the prenatal detection of a cardiac defect has to prompt a meticulous examination of the extracardiac anatomy and, vice versa, a detailed cardiac scan should follow the detection of any extracardiac malformation (Table 43.2). As some cardiac defects are clearly associated with extracardiac anomalies and/or aneuploidies

while others are not, a profound knowledge of the pattern of associated conditions will enable the examiner, on the one hand, to perform a targeted sonography of the fetal anatomy and, on the other hand, to avoid invasive procedures unlikely to reveal an abnormal karyotype. This chapter aims to assist the examiner in both respects: what anomalies are most likely to be associated in the presence of a specific cardiac defect, and what cardiac defects have to be taken into consideration when isolated or combined extracardiac anomalies are detected?

Specific cardiac defects and their association with extracardiac malformations Atrioventricular septal defect Atrioventricular septal defect (AVSD) is one of the most frequently diagnosed cardiac defects in the fetal period.5,9,10 It is found as an isolated cardiac defect or as part of complex heart lesions, and is frequently associated with chromosomal anomalies, ambiguities of the situs, extracardiac malformations, and non-karyotypic syndromes.11–14 The prognosis of AVSD strongly depends on the associated conditions as well as the additional cardiac and extracardiac malformations. These parameters that ultimately form the basis for parental counseling differ largely between prenatal and postnatal series, with more severe cases in prenatal cohorts.14 Furthermore, the associated conditions and therefore also the extracardiac anomalies reported in prenatal AVSD series depend on the referral base of the reporting centers. Tegnander et al, who analyzed 30 149 fetuses in an unselected population, found 17/21 (80.9%) cases with AVSD to be associated with chromosomal anomalies (14 with trisomy 21, one with trisomy 18, and two others), and only 2/21 (9.5%) had complex cardiac malformations.2

636

Fetal Cardiology

Table 43.1 Frequency of extracardiac and chromosomal anomalies in single types of congenital heart disease in an Italian multicenter study covering 847 fetuses with prenatally diagnosed congenital heart disease (modified after reference 5) Type of congenital heart disease

Extracardiac anomalies (%)

Chromosomal anomalies (%)

Atrioventricular defect

13.8

47.1

Univentricular defect

17.8

6.7

Hypoplastic left heart syndrome

10.9

4.2

Tricuspid atresia

34.3

8.6

Tetralogy of Fallot

25.0

26.7

Double-outlet right ventricle

19.3

45.2

Truncus arteriosus communis

21.4

28.6

Transposition of great arteries

25.6

2.6

Corrected transposition of great arteries

5.6

0

Ebstein/dysplasia of tricuspid valve

6.2

6.2

Ventricular septal defect

37.1

37.2

Atrial septal defect ostium secundum

16.1

3.2

Aortic coarctation

12.5

20.8

Aortic stenosis

13

17.4

Pulmonary stenosis

25.9

3.7

Dilated cardiomyopathy

28.6

0

0

0

57.6

0

6.2

12.5

Myocarditis Hypertrophic cardiomyopathy Tumors

Conversely, in a representative study of 301 fetuses with AVSD from a tertiary referral center for fetal echocardiography, Huggon et al14 reported 107 (35.5%) fetuses with abnormal karyotype (80.4% trisomy 21, 12.1% trisomy 18, 3.7% trisomy 13, 3.8% others), 101 (33.6%) with heterotaxy syndromes, and 30 (10%) with extracardiac anomalies not related to the forenamed associations. In this study, complex cardiac malformations were present in 146 (48.5%) cases. This spectrum is comparable to that reported in previous prenatal AVSD series from tertiary referral centers11–15 with higher incidences of complex cardiac malformations and lower incidences of aneuploidies than seen in unselected prenatal populations.2 AVSD with a normal karyotype is strongly associated with more complex cardiac malformations and unbalanced ventricular morphology, particularly in heterotaxy syndromes. In contrast, fetuses with chromosomal anomalies are strongly associated with the balanced and isolated type of AVSD, namely those with trisomy 21.11–15 Therefore, a balanced AVSD without additional intracardiac malformations is most likely to be associated with the extracardiac anomalies occurring in trisomy 21, 18, and 13, while the unbalanced complex type of AVSD with additional intracardiac malformations is most likely to be

associated with anomalies of the situs and the concomitant extracardiac malformations typical for heterotaxy syndromes. In prenatal AVSD series, isolated extracardiac anomalies and non-karyotypic syndromes are reported to be present in 13–26%.14,16 Isolated extracardiac malformations may affect all organ systems and include among others: hydrocephaly, cleft lip/palate, meningocele, diaphragmatic hernia, tracheoesophageal fistula, omphalocele, duodenal atresia, polycystic kidneys, and anomalies of the extremities.14,16 Multiple extracardiac malformations are most frequently associated with VACTERL (vertebral anomalies, anal atresia, cardiac defect, tracheoesophageal fistula, renal and limb abnormalities) and CHARGE (coloboma of the eye, heart defects, atresia of the choanae, retardation of growth and/or development, genital and/or urinary abnormalities, ear abnormalities) associations, less frequently with Ellis–van Creveld syndrome, Cornelia de Lange syndrome and Smith–Lemli–Opitz syndrome, among others.14–17 Details on the extracardiac anomalies associated with the above named conditions are provided later in this chapter.

Associated anomalies in congenital heart disease

Table 43.2 Extracardiac malformations frequently associated with cardiac anomalies8 Central nervous system (2–15%) hydrocephalus microcephalus agenesis of the corpus callosum encephalocele Dandy–Walker malformation neural tube defect Mediastinum (10–40%) diaphragmatic hernia tracheoesophageal fistula (VACTERL association) Gastrointestinal (12–22%) esophageal atresia duodenal atresia abnormal situs visceralis anorectal anomalies Abdominal wall (14–30%) omphalocele ectopia cordis Genitourinary (5–40%) hydronephrosis renal agenesis renal dysplasia horseshoe kidney Vascular (5–10%) single umbilical artery persistent right umbilical vein agenesis of ductus venosus

Ventricular septal defect Ventricular septal defects (VSDs) are the most frequently diagnosed cardiac defects in the first year of life.1 Basically, they comprise muscular and perimembranous (inlet and outlet) defects, although a great variety of classification systems exist.1,18 The rates of associated intra- and extracardiac anomalies depend on the localization as well as the size of the defect. In the prenatal period ventricular septal defects are less frequently diagnosed than in the postnatal period, and in prenatal series the subset of cases with associated anomalies is overrepresented.18,19 The reason for this discrepancy is that a considerable part of the VSD cannot be detected in the four-chamber view, and therefore escapes the basic cardiac scan if the outflow tract is not examined and if the attention of the examiner is not attracted by the presence of extracardiac anomalies.

637

Furthermore, color Doppler sonography is often required for the diagnosis of small VSDs. In their non-selected prospective cohort, Tegnander et al, reported 279 VSDs that were diagnosed in the prenatal or postnatal period.2 Isolated muscular VSDs were present in 62%, small perimembranous VSDs (isolated or in association with minor cardiac defects) in 24%, large VSDs in 3%, and 11% occurred in association with conotruncal anomalies. Isolated muscular defects were associated with aneuploidies and extracardiac malformations in 1.2% and 4.4%, respectively. In small perimembranous defects, aneuploidies and extracardiac malformations were present in 23.5% and 8.8%, respectively. None of the muscular VSDs and only 13% of the isolated perimembranous defects were detected prenatally. All of the detected perimembranous VSDs were associated with aneuploidies. Associated aneuploidies and extracardiac malformations were present in 44% and 22% of the fetuses with large VSDs, and in 30% and 23% of the fetuses with conotruncal anomalies, respectively. In these two groups the prenatal detection rate reached 52%; again, mainly those fetuses with associated aneuploidies or extracardiac malformations were diagnosed prenatally. Paladini et al, reported aneuploidies and extracardiac malformations in 47% and 33%, respectively, in their series of 68 isolated VSDs.18 The aneuploidies included trisomy 18 (21%), trisomy 21 (19%), and trisomy 13 (3%). Although the high rate of associated conditions in this study is certainly biased by the referral base, a distinct association between the site of the defect and the karyotype could be demonstrated: 69% of the fetuses with trisomy 18 had a malalignment VSD, while 82% of the fetuses with trisomy 21 had a perimembranous posterior defect of the inlet.18 The extracardiac malformations in isolated VSD with normal karyotype may affect all organ systems. In their study of 146 fetuses with VSD detectable only by color Doppler, Axt-Fliedner et al, reported 18 (12%) euploid fetuses with extracardiac anomalies. Most frequent were anomalies of the central nervous system, neural tube defects, abdominal wall defects, and skeletal anomalies. Multiple malformations were present in four fetuses.19 Respondek et al, who studied 100 consecutive fetuses with cardiac defects reported five euploid fetuses with extracardiac anomalies among 12 fetuses with isolated VSD. The diagnoses included hydrocephaly (n = 2), gastroschisis, polycystic kidneys, and diaphragmatic hernia.16

Conotruncal anomalies Conotruncal anomalies comprise common arterial trunk, transposition of the great arteries (atrioventricular concordance with ventriculoarterial discordance), and tetralogy of Fallot. When the group of conotruncal defects is expanded to outflow tract defects, double-outlet right

638

Fetal Cardiology

ventricle is added to the group.20 The concept of grouping these cardiac defects originates from a presumed common etiology: an early failure or delay of the circular outflow tract in attaining an elliptical configuration, and a resulting disturbance of the cushions that form semilunar valves and the conotruncal septum.21 However, from an epidemiological point of view, it is doubtful that they represent equivalent malformations.20 This is emphasized by the different spectrum of associated malformations within this group of cardiac defects. In their review of 12 932 live births and stillbirths with congenital heart defects from three registries, Harris et al found tetralogy of Fallot, double-outlet right ventricle (DORV) and common truncus to be significantly associated with extracardiac malformations, while transposition of the great arteries was inversely correlated with extracardiac anomalies. Furthermore, DORV did not cluster with any of the three other conotruncal anomalies. The only common feature of the conotruncal anomalies was their occurrence in Patau syndrome.20 Apparently, the conotruncal defects and the outflow tract defects are rather heterogeneous from an epidemiologic point of view, and it is doubtful that they should be grouped together.20 In particular, dextrotransposition of the great arteries (atrioventricular concordance with ventriculoarterial discordance) as well as levotransposition (atrioventricular discordance with ventriculoarterial discordance) is most likely to result from looping anomalies, is rarely associated with aneuploidies and extracardiac malformations, and should therefore be considered separately.

Tetralogy of Fallot Extracardiac anomalies are present in 28–30% of infants born with tetralogy of Fallot (TOF).22 In prenatal collectives the subset of fetuses with associated conditions is considerably larger. A recent study from a tertiary referral center in 129 prenatally diagnosed TOF reported 55 (43%) cases with aneuploidies (33% trisomy 21, 16% trisomy 18, 16% trisomy 13, 27% 22q11 microdeletions, and 7% miscellaneous).23 Extracardiac anomalies occurred in 65 (50%) cases, and in 22/37 (59%) cases in which chromosomal abnormalities had been excluded. The associated defects varied in type and severity, and included talipes, tracheoesophageal fistula, cleft lip, abdominal wall defects, ventriculomegaly, single umbilical artery, and renal anomalies.23 In their study of 61 fetuses with conotruncal anomalies, Tometzki et al, found 6/18 (33%) cases with TOF to be associated with extracardiac malformations and two (11%) with aneuploidies.24 The extracardiac malformations in euploid fetuses included omphalocele, diaphragmatic hernia (n = 2), renal agenesis, and pentalogy of Cantrell. In a non-selected prenatal collective, 5/7 (71%) fetuses with TOF had extracardiac anomalies, three of them in combination with aneuploidies.2

While the two variants of TOF – pulmonary atresia with ventricular septal defect and TOF with absent pulmonary valve syndrome – have a spectrum and incidences of associated aneuploidies and extracardiac malformations similar to the classic type of TOF, they differ largely concerning the association with microdeletion 22q11. Vesel et al reported 40% chromosomal anomalies and 19% extracardiac malformations in their series of 27 fetuses with pulmonary atresia and ventricular septal defect.25 Amongst 18 fetuses with TOF with absent pulmonary valve syndrome, Volpe et al reported 57 % chromosomal anomalies and 28% extracardiac malformations.26 Furthermore, monosomy 22q11 is found in 14% of classic TOF, 21% of pulmonary atresia with ventricular septal defect (40% in the presence of major aortopulmonary collateral arteries), and 37% of tetralogy of Fallot with absent pulmonary valve syndrome.27,28 As there is no specific extracardiac malformation associated with tetralogy of Fallot and its variants, a thorough search of the fetal anatomy is mandatory upon detection. Karyotyping has to be considered, especially in the presence of multiple anomalies, as well as a search for 22q11 microdeletions.

Double-outlet right ventricle DORV is a rare anomaly with a high degree of complexity and variation. It is frequently associated with additional intracardiac malformations, extracardiac malformations, anomalies of the situs, and aneuploidies. The reported rates of aneuploidies as well as extracardiac malformations vary considerably. In a series of 19 fetuses diagnosed with DORV, Kim et al reported 21% chromosomal anomalies and 36% heterotaxy syndromes.29 In another series of 22 cases with DORV, 14% were associated with aneuploidies and another 14% with major extracardiac malformations in euploid fetuses.24 In their multicenter series of 847 fetuses with cardiac malformations, Fesslowa et al reported 45% chromosomal anomalies and 19% extracardiac malformations among the 31 cases of DORV.5 The most frequently diagnosed aneuploidies were trisomy 18 (26%), trisomy 13 (13%), and trisomy 21 (10%). The associated extracardiac anomalies frequently reported in euploid fetuses with DORV include Dandy–Walker anomaly, hydrocephalus, absence of corpus callosum, diaphragmatic hernia, and the anomalies commonly associated with VACTERL association and heterotaxy syndromes.6,16,24

Common arterial trunk Common arterial trunk is frequently associated with aneuploidies and extracardiac anomalies. Fesslova et al reported 29% aneuploidies and 21% extracardiac malformations among 14 cases with common arterial trunk in their series.5

Associated anomalies in congenital heart disease

Likewise, Tometzki et al found 33% aneuploidies and another 33% with extracardiac anomalies in their cohort.24 Microdeletions of chromosome 22q11 are present in onethird of the cases,27 namely those with associated anomalies of the aortic arch (e.g. right aortic arch, interrupted aortic arch) and its branches (e.g. aberrant right subclavian artery).30 The leading aneuploidies in common arterial trunk are trisomy 13 and trisomy 18.5,24,30 The extracardiac anomalies in euploid fetuses include anophthalmos, hydrocephalus, duodenal atresia, imperforate anus, and those occurring in CHARGE association.4,30,31

Hypoplastic left heart syndrome Hypoplastic left heart syndrome (HLHS) is not a single entity but is rather a spectrum of congenital heart malformations characterized by severe hypoplasia of the left ventricle and left ventricular outflow tract. In ‘classical’ HLHS the aortic valve is atretic, with either atresia or severe hypoplasia of the mitral valve. In some instances, the term has been applied to other lesions including critical aortic stenosis with severe hypoplasia of the left ventricle, unbalanced atrioventricular septal defect, and severe coarctation of the aorta.32 Hence, it is not surprising that the reported rates of associated conditions vary widely in published series. In ‘classical’ HLHS the incidence of karyotypic anomalies is 4–10%.10,33,34 Allan et al reported two cases of trisomy 18 and one case of monosomy X in their series of 30 fetuses with HLHS, resulting in an incidence of aneuploidies of 10%. Two fetuses had extracardiac anomalies (7%) – a tracheoesophageal fistula and multiple small bowel atresias, respectively.34 Brackley et al found similar incidences among their 87 fetuses.35 Seven had chromosomal anomalies (8%), including two cases of trisomy 13, one case of monosomy X, and four structural chromosomal anomalies. The 6/80 (7%) extracardiac anomalies in euploid fetuses included renal hypoplasia, agenesis of corpus callosum, Dandy–Walker variant, omphalocele, hemivertebrae, and tracheoesophageal fistula.35

Tricuspid atresia Tricuspid atresia is highly associated with additional cardiac anomalies and less frequently with other conditions. Aneuploidies and extracardiac anomalies have been reported in 2–9% and 19–34%, respectively.5,10,20,36 In their series of 58 fetuses in which information was available, Wald et al reported one case each of trisomy 21 and trisomy 18 and one structural chromosomal anomaly.36 Some 19% of the 58 fetuses had extracardiac anomalies. These included cleft lip, and gastrointestinal, urogenital, and musculoskeletal defects. Multiple malformations

639

were present in two fetuses, one of them with VACTERL association.

Anomalies of the aortic arch Right and double aortic arch Anomalies of the aortic arch that concern the vessel course and/or its branching pattern include right aortic arch (RAA) with mirror-image branching; RAA with aberrant left subclavian or innominate artery; double aortic arch (DAA); circumflex retroesophageal aortic arch; or left aortic arch with an aberrant right subclavian artery.37 Depending on the type of anomaly, important associated conditions are intra- and extracardiac malformations, postnatal tracheal or esophageal compression, left subclavian steal syndrome due to constriction of the aberrant artery, and chromosomal anomalies, namely microdeletion of the DiGeorge critical region of chromosome 22q11.38–44 The risk of concomitant congenital heart disease is over 90% with the mirror-image branching type and only 10% with the RAA and aberrant left subclavian artery. Double aortic arch usually occurs as an isolated finding.44,45 The most common association is tetralogy of Fallot where the incidence of RAA (usually the mirror-image branching pattern) ranges from 13 to 35%.42,46,47 Other frequent associations are pulmonary atresia with ventricular septal defect, and truncus arteriosus with incidences of RAA of 31–36% and 15–36%, respectively.42,46 In our own series of 71 cases with RAA, 28 (39%) had extracardiac anomalies.44 Of these, 68% were abnormalities of the situs in heterotaxy syndromes and 18% were associated with microdeletion 22q11. The latter included hydramnios, esophageal atresia, cleft lip/palate, spina bifida, and clubbed feet.

Coarctation of the aorta Coarctation accounts for 7.1–8.3% of CHD in fetal series, and is significantly associated with chromosomal and extracardiac anomalies.4,5,9 Paladini et al reported 29% chrosomosomal anomalies and 18% major extracardiac anomalies in their series of 68 fetuses.48 The most frequent diagnoses among the 20 fetuses with chromosomal aberrations were monosomy X (35%), trisomy 21 (15%), trisomy 18 (15%), trisomy 13 (15%), and microdeletion 22q11 (5%). The extracardiac anomalies in euploid fetuses included agenesis of the corpus collosum, encephalocele, esophageal atresia, anorectal atresia, bilateral renal agenesis, pyelectasis, polycystic kidney, hypospadia, clubfoot, eye and ear anomalies, osteogenesis imperfecta, thanatophoric dwarfism, Ellis–van Creveld syndrome, Cornelia de Lange syndrome, Roberts syndrome, and Jeune syndrome.48

640

Fetal Cardiology

Interrupted aortic arch Interrupted aortic arch is a rare, severe form of congenital heart disease; it can be subdivided into three groups according to the site of interruption: type A, interruption distal to the left subclavian artery; type B, interruption between the left carotid and left subclavian arteries; and type C, interruption between the innominate artery and the left carotid artery.30 Type A is rarely associated with 22q11 deletion,49,50 whereas type B is associated with 22q11 deletion and other extracardiac features in 50–80% of cases (only part of them detectable at prenatal ultrasound),51,52 often in the context of specific syndromes, namely DiGeorge syndrome or velocardiofacial syndrome; type C is by far the least common form of interrupted aortic arch, accounting for fewer than 5% of cases. Among nine fetuses with interrupted aortic arch detected prenatally, Volpe et al reported six with type B and three with type A. Five had a microdeletion 22q11 (four type B cases and an unusual association with one type A case). All cases were de novo. None of the nine cases was associated with extracardiac anomalies.53 However, in a subsequent study in 141 fetuses with conotruncal anomalies or anomalies of the aortic arch (28 of them with 22q11 microdeletion), the same group reported that the association with 22q11 microdeletions is significantly predicted by the presence of associated ultrasound findings: thymic hypo/aplasia, intrautetine growth restriction (IUGR), and additional aortic arch anomalies.54

Specific cardiac defects less likely to be associated with extracardiac anomalies While most of the cardiac anomalies are to a certain extent associated with extracardiac anomalies and aneuploidies, this is not true for transposition of the great arteries, left and right outflow tract obstruction with intact ventricular septum, and tricuspid valve dysplasia including Ebstein’s anomaly.

Transposition of the great arteries Transposition of the great arteries exists in two distinct variants: atrioventricular concordance with ventriculoarterial discordance (d-TGA) and atrioventricular discordance with ventriculoarterial discordance (congenitally corrected TGA or l-TGA). Both variants are highly associated with additional cardiac malformations but rarely associated with aneuploidies and extracardiac malformations.6 Most prenatal series report rates of aneuploidies and extracardiac malformations ranging 0–7% and 13–26%,

respectively.4,5,16,24 Fesslova et al reported one case each of trisomy 21 and trisomy 18 in their series of 39 fetuses with d-TGA.5 Tometzki et al reported one case of trisomy 13 among 15 cases of d-TGA.24 The negative correlation between transposition and an abnormal karyotype is especially true for fetuses with atrioventricular and ventriculoarterial discordancy.55 The reported extracardiac anomalies in transposition of the great arteries in euploid fetuses include hydrocephalus, diaphragmatic hernia, renal anomalies, and anomalies of the situs associated with heterotaxy syndromes.24,55

Left ventricular outflow tract obstruction In critical aortic stenosis, the incidence of karyotypic abnormalities and extracardiac anomalies is exceedingly low. Two large prenatal series reported no such abnormalities in their cohorts.10,33 Fesslova et al, reported one case of monosomy X and three cases with extracardiac anomalies among 23 fetuses with aortic stenosis.5

Right ventricular outflow tract obstruction with intact ventricular septum In the presence of an intact ventricular septum, pulmonary atresia and pulmonary stenosis are isolated findings in most cases. Allan et al reported 5% chromosomal anomalies among their 55 cases.10 Todros et al found one case of trisomy 22, one Noonan syndrome, and one case with dysplastic kidneys among their 33 cases, resulting in an incidence of 3% and 6% of aneuploidies and extracardiac anomalies, respectively.56

Tricuspid dysplasia and Ebstein’s anomaly Although sporadic cases with aneuploidies, namely Down syndrome, have been reported,57,58 most cases with tricuspid dysplasia occur as isolated lesions.59 Sharland et al reported two cases (5%) with aneuploidy in their series of 38 fetuses with tricuspid dysplasia or Ebstein’s anomaly.60 Similar incidences are reported in other large prenatal and neonatal series.5,10,20 Extracardiac anomalies are reported equally rarely, and include neural tube defects, and craniofacial defects, and anomalies of the central nervous system and the limbs.6,61

Associated anomalies in congenital heart disease

Specific extracardiac malformations and their association with cardiac defects Head and central nervous system Anomalies of the central nervous system are seen in 4% of fetuses indentified to have congenital heart disease.5 The anomalies most frequently associated with cardiac defects are ventriculomegaly, hydrocephalus, Dandy– Walker malformation, holoprosencephaly, and agenesis of the corpus callosum, while in neural tube defects (except for encephalocele in Meckel–Gruber syndrome), porencephaly, and hydranencephaly no specific association has been described.22 Because the etiology of microcephaly is heterogeneous, including chromosomal aberrations, viral infections, and environmental agents (e.g. phenytoin, ethanol), congenital heart disease is often present.8

Ventriculomegaly and hydrocephalus Identification of isolated mild ventriculomegaly (defined as atrium of the lateral ventricles between 10 and 15 mm) (Figure 43.1) presents a counseling dilemma because it can represent a normal physiologic variant or can be the epiphenomenon of a heterogeneous group of pathologic processes that include increased intraventricular pressure, primary neuronal loss, and abnormalities of brain development such as those seen with chromosomal anomalies. Mild dilatation of the cerebral ventricles should therefore prompt a careful search for associated anomalies including cardiac defects, whose presence carries a poor prognosis.62 Among 82 fetuses with mild ventriculomegaly analyzed by Vergani et al,62 34/82 (42%) had associated anomalies,

641

7/82 (9%) had aneuploidies and 5/82 (6%) had cardiac defects (one in combination with trisomy 21). Severe ventriculomegaly (atrium of the lateral ventricles > 15 mm) is less frequently associated with aneuploidies and cardiac defects. In the series of Gaglioti et al, 60 fetuses had a ventricular width > 15 mm and 116 < 15 mm. All nine cases with aneuploidy and all six cases with cardiac defects (four of them in aneuploidies) occurred in the group < 15 mm.63 Likewise, Breeze et al found only one case of trisomy 21 and none with associated cardiac defects in their series of 20 fetuses with severe ventriculomegaly.64

Agenesis of the corpus callosum In fetuses with prenatally diagnosed agenesis of the corpus callosum, about one-third of cases are isolated and two-thirds are complicated by associated structural defects and/or an underlying chromosomal abnormality. In a recent study of 117 cases, 42% had other structural anomalies and 28% had chromosomal aberrations.65 Therefore, the incidence of cardiac defects in complete agenesis of the corpus callosum is high. Among 19 cases of prenatally diagnosed partial agenesis of the corpus callosum, Volpe et al found five (26%) cases with associated cardiac malformations, three of them in combination with chromosomal anomalies.66

Dandy–Walker malformation Dandy–Walker malformation (Figure 43.2) is frequently associated with intracranial anomalies, extracranial malformations, and aneuploidies. There seems to be little difference in the spectrum of associated findings between Dandy–Walker malformation and Dandy– Walker variant.67 Among 99 fetuses with Dandy–Walker

Figure 43.1

Figure 43.2

Mild ventriculomegaly in a fetus with Down syndrome and atrioventricular septal defect.

Dandy–Walker malformation in a fetus with multiple malformations including a large ventricular septal defect.

642

Fetal Cardiology

Figure 43.3 Alobar holoprosencephaly in a fetus with Patau syndrome and common arterial trunk.

malformation (50/99) and Dandy–Walker variant (49/99) in the series of Ecker et al, 85% had associated sonographic anomalies, 49% had cardiac anomalies, and 21% had aneuploidies. Ulm et al found 46% additional malformations, 18% cardiac anomalies, and 29% aneuploidies in their series of 28 fetuses. Sixty percent of the cardiac defects occurred in aneuploidies.68 A recent prenatal series of 10 euploid fetuses reported only one case with associated ventricular septal defect.

Holoprosencephaly Holoprosencephaly (Figure 43.3) is causally heterogeneous and is associated with various chromosomal abnormalities, single gene disorders, teratogens, and maternal diabetes.69 Cardiac anomalies are frequently detected, particularly in association with trisomy 13 where the incidence of cardiac defects is 90–94%.70 In the series of Ong et al, 30/113 (27%) had trisomy 13, 12/113 (11%) had other aneuploidies, 81/113 (72%) had additional malformations, and 24/113 (21%) had cardiac defects.69

Figure 43.4 Bilateral hydrothorax in a fetus with Noonan syndrome and interrupted aortic arch.

the most well known (see VACTERL association later in this chapter). Polyhydramnios and a small or invisible fetal stomach are the most common ultrasound findings when esophageal atresia is suspected prenatally.73 However, these ultrasound findings have a low positive predictive value in diagnosing esophageal obstruction, and the false-positive rate has been high.74 Conversely, only 20–44% of affected infants are detected prenatally.73,75 In a recent series of 48 cases, Brandtberg et al reported associated anomalies in 79% and an abnormal karyotype in 23%. Cardiac anomalies were present in 21/48 (44%) of the cases and in 10/27 (37%) fetuses with a normal karyotype.73 Most frequent other anomalies in fetuses with a normal karyotype were urogenital anomalies (48%), imperforate anus (41%), musculoskeletal anomalies (26%), and vertebral anomalies (26%), consistent with the spectrum of malformations seen in VACTERL association.

Hydrothorax

Chest Anomalies of the chest likely to be associated with cardiac defects are esophageal atresia, aplasia of the thymus (see 22q11 microdeletion later in this chapter), hydrothorax, and diaphragmatic hernia. Pulmonary sequestration and congenital cystic adenomatoid malformation of the lung are isolated lesions in the vast majority of cases.71,72

Esophageal atresia Esophageal atresia is found in connection with a number of associations/syndromes, VATER or VACTERL being

Pleural effusion may be associated with a number of underlying conditions including non-immune hydrops, intrathoracic mass, diaphragmatic hernia, trisomy 21, monosomy X, Noonan syndrome, and infection.76 Isolated hydrothorax (Figure 43.4) is most often caused by congenital chylothorax, a primary lymphatic abnormality. Associated malformations and aneuploidies are found in about 21–25% and 4–7% respectively, and obviously worsen the outcome.77,78 Rustico et al found five (9%) cardiac anomalies among their 53 patients with prenatally diagnosed pleural effusion.78 Likewise, Waller et al reported 13 (5%) cardiac defects in their series of 246 fetuses with pleural effusion.79

Associated anomalies in congenital heart disease

643

Figure 43.5

Figure 43.6

Dextroposition of the heart and herniated stomach in leftsided diaphragmatic hernia. A perimembranous ventricular septal defect was also present.

‘Double bubble’ sign in a fetus with Down syndrome, duodenal atresia, and atrioventricular septal defect.

Diaphragmatic hernia

of pregnancy, with dilated small bowel segments. Heydanus et al reported one case of cystic fibrosis as the only associated extraintestinal condition among their 11 cases with prenatally diagnosed dilated bowel loops.83 Likewise, Surana and Puri found only two cardiac defects and two cases with cystic fibrosis among their 59 fetuses with jejunal or ileal atresia.85 Anorectal atresia has been rarely recognized as dilated colon in the lower abdomen or as calcified intraluminal meconium at prenatal sonography. However, in these cases the incidence of additional malformations exceeds 90%, consisting mainly of genitourinary malformations but also of the cardiovascular malformations occurring in VACTERL association.86

Congenital diaphragmatic hernia (Figure 43.5) is frequently associated with other major malformations, karyotype anomalies, and syndromes. In a study from 20 registries of congenital malformations including 187 cases of diaphragmatic hernia, Garne et al reported 38% associated major malformations, karyotype anomalies, and syndromes. Eleven percent had associated cardiac anomalies.80 Other authors found 33–47% fetuses with associated malformations and abnormal karyotype in their series. Cardiac defects were present in 8–9%.81,82

Abdomen and abdominal wall Bowel obstruction

Ventral wall defects

The major forms of gastrointestinal abnormality diagnosed prenatally are bowel atresias. Duodenal atresia (Figure 43.6) is associated with trisomy 21 in 30–60% of cases, and, therefore, the incidence of associated cardiac defects in prenatally diagnosed ‘double bubble’ sign is quite high, ranging from 20 to 30%.8,83 Among 275 cases of duodenal obstruction, Murshed et al found Down syndrome in 30%, Down syndrome plus cardiac malformation in 14%, cardiac anomalies without Down syndrome in 23%, and other gastrointenstinal disorders in 42%.84 Among their 10 cases with prenatally diagnosed duodenal atresia, Heydanus et al reported one case of tetralogy of Fallot and one case of atrioventricular septal defect, both occurring among six cases of trisomy 21.83 On the other hand, cardiac defects are only found in 3–5% of fetuses diagnosed with jejunal or ileal atresia.8,85 The affected fetuses mostly become apparent after 24 weeks

Fetal abdominal wall defects can be broadly categorized into gastroschisis, omphalocele, limb-body wall complex, cloacal and bladder exstrophy, ectopia cordis, and urachal cyst.87 Gastroschisis is isolated in most cases, and is typically not associated with chromosomal abnormalities. In their series of 109 fetuses with abdominal wall defects, Fratelli et al found associated anomalies in only 2/40 (5%) cases with gastroschisis, while 51/67 (76%) cases with omphalocele were associated with aneuploidies or additional malformations.88 Omphalocele has two distinct etiologies: (1) failure of the normal infolding process of the lateral borders of the embryonic disc (folding failure of the caudal borders results in bladder exstrophy and folding failure of the cephalic portion in ectopia cordis); (2) failure of the normal herniation of small intestine into the umbilical

644

Fetal Cardiology

Figure 43.7

Figure 43.8

Large omphalocele containing liver and ascites.

Small omphalocele containing small bowel only in a fetus with Edwards syndrome and a large perimembranous ventricular septal defect.

cord to reverse by the 12th week.8 The former mechanism leads to large omphaloceles containing liver, colon, and other intra-abdominal organs (Figure 43.7), and may be associated with bladder exstrophy or ectopia cordis, and sternal and anterior diaphragmatic defects, while the latter results in omphaloceles containing small bowel only (Figure 43.8). Unsurprisingly, the associated anomalies are different between the two variants. Fetuses with omphalocele with liver protruding into the sac have lower rates of abnormal karyotypes than fetuses with only the bowel in the sac. The most frequent chromosomal anomalies associated with omphalocele are trisomy 18 and 13, where cardiac defects are regularly associated.87,89 Brantberg et al reported 15 cases with cardiac defects among their cohort of 36 fetuses with omphalocele and a normal karyotype, including two fetuses with Beckwith–Wiedemann syndrome (see non-karyotypic syndromes later in this chapter) and two with Cantrell’s pentalogy.89 Pentalogy of Cantrell90 results from a folding failure of the caudal borders of the embryonic disc and features omphalocele, anterior diaphragmatic hernia, sternal cleft, ectopia cordis, and structural cardiac defects. Hornberger et al reported 7/13 (54%) cases with tetralogy of Fallot and 6/13 (46%) with double-outlet right ventricle in their cohort of children with ectopia cordis, most of whom also displayed features of pentalogy of Cantrell.91

renal agenesis, horseshoe kidney, bilateral dysplastic kidneys, obstruction of the pelviureteric junction, and megacystis,92,93 while unilateral lesions and obstructions of the vesicoureteric junction are less frequently associated with extrarenal anomalies. Associated anomalies are seen in more than 50% of fetuses with renal agenesis. The major associated anomalies are cardiac defects (25%), the VATER association (27%), digital anomalies (15%), and Mullerian anomalies (20%).94 Horseshoe kidney is associated with many other anomalies including urogenital, central nervous, gastrointestinal, musculoskeletal, and cardiovascular defects.95,96 Greenwood et al reported cardiac defects in 44% of neonates with horseshoe kidneys.92 Large ‘bright’, or hyperechogenic, kidneys (Figure 43.9) represent a difficult diagnostic dilemma, particularly in the presence of a normal amout of amniotic fluid, since their underlying etiologies are relatively diverse. The differential diagnosis includes obstruction, autosomal recessive (infantile) polycystic kidney disease, autosomal dominant (adult) polycystic kidney disease, Beckwith– Wiedemann syndrome, Meckel–Gruber syndrome, and trisomy 13.97 In the latter three, cardiac defects will be frequently present. Twelve percent of fetuses with obstruction of the pelviureteric junction have other extrarenal abnormalities such as anorectal anomalies, congenital heart disease, VATER association, and esophageal atresia, but no particular pattern exists unless associated with chromosomal abnormality.98 Associated anomalies are seen in up to 43% of posterior urethral valves and the resulting megacystis. These include malrotation, anal atresia, the VATER association, and cardiac anomalies.

Urogenital Cardiac anomalies are present in 8% of children born with genitourinary malformations.92 The strongest association with cardiac defects has been reported for bilateral

Associated anomalies in congenital heart disease

645

anomaly, but there is a widely differing detection rate in the second trimester. In a targeted echocardiography study, Paladini et al reported a 56% incidence of congenital heart disease in fetuses with known Down syndrome, which included ventricular septal defect (48%), atrioventricular septal defect (44%), tetralogy of Fallot (4%), and coarctation of the aorta (4%). Conversely, 53% of cases with AVSD and normal visceral situs were associated with Down syndrome.114 DeVore reported cardiac abnormalities in 76% of fetuses with Down syndrome, including structural and functional (tricuspid regurgitation, pericardial effusion) findings.101

Edwards syndrome Figure 43.9 Large and hyperechogenic kidneys in Meckel–Gruber syndrome.

Multiple malformations in aneuploidies and non-karyotypic syndromes Down syndrome A number of referral centers with expertise in targeted ultrasound examination report sonographic abnormalities in 60–90% of second trimester fetuses with Down syndrome in selected high-risk patients.99–103 Sonographic findings are commonly described as structural defects (major abnormalities) or minor abnormalities (markers). A wide variety of ultrasound markers have been associated with fetal Down syndrome during the second trimester.99,102,104–106 The most frequently reported findings in fetuses with trisomy 21 include thickened nuchal fold, shortened long bones (humerus, femur), hyperechogenic bowel, renal pyelectasis, and intracardiac hyperechogenic foci. More recently, nasal bone hypoplasia, larger iliac wing angle, and shortened fingers have been added to the list of sonographic markers.107–112 Systematic evaluation of those multiple markers is commonly referred to as a second trimester genetic sonogram. As the list of sonographic markers is growing, there is a high chance to identify at least one marker by routine ultrasound. Congenital malformations associated with Down syndrome are less frequent than with trisomy 18 or trisomy 13 and will be identified in about 20–30% of fetuses.99,100,102,103,109,113 Those anomalies include heart defects, fetal hydrops, brachycephaly, ventriculomegaly, Dandy–Walker anomaly, agenesis of the corpus callosum, duodenal atresia, and abnormalities of the fingers, among others. Heart defects represent the most common

Most fetuses with trisomy 18 have multiple abnormalities, and 80–90% are detected sonographically in centers with expertise.115–118 The most characteristic syndromal patterns involve the central nervous system, the limbs, and the cardiac system. Typical sonographic findings include choroid plexus cysts, a strawberry-shaped head and cisterna magna abnormalities, clenched fingers, radial defects, and clubbed or rocker-bottom feet. They may also have micrognathia, omphalocele, and umbilical cord abnormalities.115–118 Pathological series suggest that congenital heart disease is universal in trisomy 18.70,119,120 However, in prenatal series the incidence of detected cardiac anomalies varies depending on the gestational age at scanning. Moyano et al reported 27% sonographically normal hearts in the first trimester and 15% in the second trimester, respectively.121 The types of cardiac defects seen in trisomy 18 are more varied than those associated with trisomy 21, and some of them might not be detectable at prenatal scan, such as small ventricular septal defect and atrial septal defects. In Moyano’s series, 17% had atrioventricular septal defect, 14% ventricular septal defect, 13% coarctation of the aorta, 5% tetralogy of Fallot, and 26% had no detected anomaly.121 Fetal growth restriction is a common finding. The prevalence increases with gestational age from about 30% in the early second trimester to 90% in the third trimester.115,122 Severe fetal growth restriction in combination with polyhydramnios in a third trimester fetus should prompt a detailed search for associated malformations, since infants with trisomy 18 do not benefit from altered perinatal management.

Patau syndrome Fetuses with trisomy 13 present with a cluster of characteristic sonographic abnormalities. The most frequently observed abnormalities include holoprosencephaly with the associated facial defects (hypotelorism, absent nose,

646

Fetal Cardiology

cyclopia, median clefts), neural tube defects, polydactyly (primarily of the hands), cardiac defects, omphalocele, enlarged polycystic kidneys, and growth restriction.123,124 Heart defects are present in up to 94% of fetuses with trisomy 13 in the first trimester. Most common are ventricular septal defects and a variety of valvular defects. Tetralogy of Fallot with absent pulmonary valve syndrome and truncus arteriosus are specifically associated with trisomy 13.70 The three characteristic features in Meckel–Gruber syndrome – postaxial polydactyly, occipital encephalocele, and polycystic kidneys – overlap with the syndromal pattern in trisomy 13; however, while Meckel–Gruber syndrome tends to present with severe oligohydramnios as early as 16 weeks,125 the amniotic fluid volume in trisomy 13 may remain unchanged. To reliably distinguish between the two diagnoses, fetal karyotyping is required. This is important despite the lethal condition in both, as Meckel–Gruber syndrome is an autosomal recessive disease with a 25% recurrence risk in affected families.

Triploidy The most striking sonographic sign in triploidy (69,XXX; 69,XXY; 69,XYY) in the second trimester is an early onset severe asymmetrical growth restriction. Whereas the head appears appropriate for gestational age, the fetal abdomen and the extremities are growth-restricted. Oligohydramnios and placentomegaly are common features too. Structural fetal defects are observed antenatally in 70–90%. The central nervous system (ventriculomegaly, Dandy–Walker malformation, agenesis of the corpus callosum) is commonly affected. Other sonographic abnormalities seen in triploidy include facial dysmorphism (micrognathia), limb anomalies (syndactyly, clinodactyly, talipes), congenital heart defects, neural tube defects, and renal malformations.126–129 In the series of Mittal et al, postmortem examination revealed cardiac defects in 4/20 (20%) fetuses with prenatally diagnosed triploidy (atrial septal defect, ventricular septal defect, pulmonary stenosis, and common arterial trunk); however, none of these had been detected at sonography.128

Turner syndrome Clinical features of monosomy X include short stature, webbing of the neck, ovarian dysfunction, and cardiovascular abnormalities. Typical sonographic signs in this syndrome include huge septated cystic hygroma of the neck, increased nuchal translucency, cardiac defects, and renal malformations.130–133 A congenital heart defect is found in about 30% of postnatal cases of Turner syndrome. Left-sided obstructive

defects predominate, especially bicuspid aortic valve and coarctation of the aorta.134 There is a higher incidence of structural heart disease diagnosed in those cases that present during fetal life.130–132 Surerus et al reported cardiac malformations in 62% (33/53) of fetuses with Turner syndrome investigated in the first and mid-second trimester. Coarctation of the aorta (45%) and hypoplastic left heart syndrome (13%) were the most common diagnoses. A markedly increased nuchal translucency was frequently associated (47/53).131

Syndromes associated with 22q11 microdeletion The clinical presentation of monosomy 22q11 includes patients with DiGeorge (DGS), Shprintzen, velocardiofacial (VCFS), and conotruncal anomaly face syndromes. A microdeletion of 22q11 has been shown to be associated with these anomalies in more than 80% of cases.135,136 Subsequently, the syndromes have been grouped together under the acronym ‘CATCH-22’: cardiac defects, abnormal facies, thymus aplasia or hypoplasia, cleft palate, hypocalcemia, and 22 denoting the deletion on chromosome 22.137 Recently it was suggested to abandon the acronym CATCH-22 and instead to use mainly deletion 22q11. Conotruncal malformations are a major feature of 22q11 microdeletion and are present in 70–80% of the affected patients.27,138 The distribution of the various types of conotruncal malformations in neonates with 22qdel is random;51 nevertheless, various cardiac phenotypes have been proved to be almost as specific for 22qdel as interrupted aortic arch type B or pulmonary atresia with ventricular septal defect and multiple aortopulmonary collateral arteries.28,51 In a series of 261 consecutive fetuses with conotruncal defects and a normal karyotype, 54 (21%) had a 22q11 deletion.27 Among these, 26% had tetralogy of Fallot, 20% pulmonary atresia with ventricular septal defect, 19% interrupted aortic arch, 17% common arterial trunk, 11% absent pulmonary valve syndrome, and 7% DORV. It has been proposed to test all fetuses with conotruncal defects for monosomy 22q11. However, the prevalence of 22q11 deletion has been reported to be fewer than 20% of patients with tetralogy of Fallot, leading to 80% unnecessary tests.139 In addition, recent postnatal studies have shown that 22q11 deletion is very rare in patients with isolated tetralogy of Fallot.27,140 Conversely, the vast majority of patients with tetralogy of Fallot and 22q11 deletion exhibit the other phenotypic features of the syndrome, namely abnormal facial appearance, thymic aplasia or hypoplasia, and neonatal hypocalcemia.141 In a fetal population, these extracardiac anomalies are difficult to identify. Extracardiac anomalies are also frequent in fetuses with conotruncal defects, but previous studies

Associated anomalies in congenital heart disease

found no statistical difference between fetuses with and without 22qdel in their series, even for renal anomalies that were present in 30% of the fetuses with 22qdel.27,142 In their series of 151 fetuses with tetralogy of Fallot, Boudjemline et al found increased nuchal translucency (in the first trimester), polyhydramnios, and growth restriction (in late pregnancy) to be more frequent in fetuses with 22q11 deletion as well as pulmonary arterial abnormalities. When these different features were present in the same fetus with tetralogy of Fallot, 22q11 deletion could be predicted with a sensitivity of 88%.143 Absence or hypoplasia of the thymus might be a further sonographic target. Chaoui et al analyzed 149 fetuses with congenital heart disease and normal karyotype. Seventysix fetuses had conotruncal anomalies. 22q11.2 deletion was present in 10 cases (6.7%), all of which had conotruncal anomalies (13.1%). Thymic hypoplasia or absence was suspected in 11 cases with conotruncal anomaly. Nine of these 11 had the deletion; two cases were false-positive. One fetus with a normal-sized thymus had deletion of 22q11.2 (sensitivity 90%, specificity 98.5%, positive predictive value 81.8%, and negative predictive value 99.2%).144 In a study of Barrea et al in 16 fetuses with cardiac anomalies at risk for 22q11 deletion, the thymus was absent at prenatal sonography in all six cases with 22q11 deletion and present in all cases without.145 There are 763 syndromes listed on the London Dysmorphology Database where cardiac abnormalities can form part of the spectrum, and many of these are very rare.146 In a recent review, Pajkrt et al summarized the more common associations with cardiac abnormalities that may be detected prenatally (Table 43.3).

Thrombocytopenia–absent radius syndrome Thrombocytopenia–absent radius (TAR) syndrome is a congenital malformation syndrome of unknown inheritance that features bilateral absence of the radii with conservation of the thumbs and thrombocytopenia.17 Other structural anomalies in TAR syndrome may include an absent ulna and/or humerus in 50%, lower leg involvement in 40–47%, and renal anomalies in 23%.147 Greenhalgh et al reported 34 patients with TAR syndrome of whom five (15%) had cardiac anomalies, including atrial septal defect, ventricular septal defect, and atrioventricular septal defect.147

Cornelia de Lange syndrome Cornelia de Lange syndrome is a well-described autosomal dominant multiple malformation syndrome typically involving proportionate small stature, developmental delay, specific facial features, major malformations

647

(particularly the cardiac, gastrointestinal, and musculoskeletal systems), and behavioral abnormalities.17,148 Main sonographic features are asymmetrical upper limb defects (27–58%) and fetal growth restriction (in late pregnancy), which is present in 80–100% of cases.17,149,150 Non-skeletal anomalies include diaphragmatic hernia, nuchal webbing, duodenal atresia, renal dysplasia, cleft palate, and genital anomalies.150,151 Cardiac anomalies have been reported in 14–70% of cases, most commonly VSD but also atrial septal defect (ASD), pulmonary stenosis, tetralogy of Fallot, mitral atresia, stenosis or coarctation of the aorta, AVSD, and single ventricle.17,152,153

Holt–Oram syndrome Holt–Oram syndrome is characterized by bilateral upper limb deformities, predominantly involving the radial ray, and congenital heart defects.17 The characteristic findings are a thumb anomaly and/or radial aplasia. Cardiac anomalies are present in about 90% of patients and most commonly include atrial septal defect, common arterial trunk, mitral valve anomalies, and tetralogy of Fallot.154

VACTERL association VATER is an acronym for the association of vertebral defects (V), anal atresia (A), tracheoesophageal fistula with esophageal atresia (TE), and radial and renal dysplasia (R). The further observation of cardiac anomalies (C) and limb anomalies (L) expanded the name to VACTERL.17 Vertebral anomalies and tracheoesophageal defects are present in 60–65% of patients, anal atresia in 55–60%, renal and cardiac anomalies in 70–75%, and limb defects and thoracic anomalies in 35–45%. Other common findings include ear abnormalities, facial cleft, and genitourinary anomalies.155–157 Among 15 cases of VATER association with cardiac defects reported by the Euroscan study group, five had complex cardiac defects, three had tetralogy of Fallot, three had atrial septal defects, two had ventricular septal defects, and two had tricuspid atresia.158

Short rib–polydactyly syndromes Short rib–polydactyly syndromes (Saldino–Noonan syndrome (type I), Majewski syndrome (type II), Verma– Naumoff syndrome (type III), and Beemer–Langer syndrome (type IV)) are lethal skeletal dysplasias inherited in an autosomal recessive fashion. They are characterized by generalized shortening of the long bones, thoracic hypoplasia, short ribs, polydactyly, and multiple visceral anomalies. They should be differentiated from the potentially surviving conditions of Jeune syndrome and Ellis– van Creveld syndrome that also feature shortened long

Nuchal edema/hydrops

AVSD, ASD

VSD, ASD, tetralogy of Fallot

Campomelic dysplasia

Smith–Lemli–Opitz

AVSD, common atrium

Ellis–van Creveld

Tetralogy of Fallot, ASD

VATER association

Wide variety

ASD

Holt–Oram

Short rib–polydactyly

VSD

Cornelia de Lange

Skeletal dysplasia

ASD, VSD, AVSD

TAR

Radial anomalies

Most common cardiac anomaly

Syndrome

? – associated with microdeletion on 1q21.1

Inheritance

Polydactyly, cataracts, two- or three-toe syndactyly, renal anomalies, ambiguous genitalia, FGR

Bowing of femur and tibia, talipes, hypoplastic scapulae, 11 pairs of ribs, small chest, micrognathia

Short limbs, short ribs, polydactyly, renal anomalies

Hypoplasia of thorax, short ribs, polydactyly

Vertebral defects, tracheoesophageal fistula, cleft lip, ambiguous genitalia

Absent or hypoplastic thumb, or more severe reduction defect of upper limb

AR – mutations in DHCR7

AD – de novo mutations in SOX9

AR – mutations in EVC or EVC2

AR

?

AD – mutations in TBX5

FGR, microcephaly, abnormal AD – most are de novo mutations in NIPBL profile, micrognathia, oligodactyly, thumb hypoplasia

Thumbs present, absent ulna or humerus, lower leg involvement, renal anomalies

Associated ultrasound features

Non-karyotypic syndromes frequently associated with congenital heart defects (modified after reference 17)

Cardiac plus

Table 43.3

Amniotic fluid urinary cholesterol metabolites

Positive family history and parental examination

Low maternal serum PAPP-A

Fetal blood sampling for platelet count

Other aids to prenatal diagnosis in ‘new’ cases

648 Fetal Cardiology

Tetralogy of Fallot, VSD Conotruncal malformations, tetralogy of Fallot Various VSD Cardiac rhabdomyomas

Goldenhar CHARGE Beckwith–Wiedemann Meckel–Gruber Tuberous sclerosis

Left ventricular hypertrophy, pulmonary stenosis

Occasional renal, intracerebral lesions

Large echogenic kidneys, polydactyly, encephalocele

Omphalocele, visceromegaly, macroglossia, macrosomia

Microphthalmia, genital anomalies

Vertebral and thoracic anomalies, renal anomalies

Short femur, renal anomalies, polyhydramnios

AD – mutations in TSC1 or TSC2

AR

AD – most are de novo mutations

AD – mostly de novo mutations in CHD7

Sporadic

AD – 40% have mutations in PTPN11

Positive family history, parental examination, fetal MRI scan

Positive family history and parental examination

TAR, thrombocytopenia–absent radius syndrome; FGR, fetal growth restriction; AR, autosomal recessive; AD, autosomal dominant; ASD, atrial septal defect; VSD, ventricular septal defect; AVSD, atrioventricular septal defect; PAPP-A, pregnancy-associated plasma protein A; MRI, magnetic resonance imaging.

Miscellaneous

Noonan

Associated anomalies in congenital heart disease 649

650

Fetal Cardiology

toes, and overlying skin dimples. Ambiguous genitalia occur in the majority of patients with an XY karyotype. Most cases are sporadic.161 A third of the patients have heart malformations, mainly VSD, ASD, and tetralogy of Fallot.17

Smith–Lemli–Opitz syndrome

Figure 43.10 Bilateral postaxial hexadactyly in a fetus with Ellis–van Creveld syndrome and atrioventricular septal defect.

bones/ribs and polydactyly.17 Although all of these syndromes are associated with cardiac defects, the incidence in each type varies. Diglio et al reviewed 28 patients with short rib–polydactyly syndromes and reported a wide variety of cardiac anomalies, including transposition of the great arteries (18%), coarctation of the aorta (14%), hypoplastic right or left heart (14%), AVSD (11%), and VSD (11%).159

Ellis–van Creveld syndrome (chondroectodermal dysplasia) Ellis–van Creveld (ECV) syndrome160 is a non-lethal, autosomal recessive skeletal dysplasia characterized by short limbs, short ribs, postaxial polydactyly (Figure 43.10), cardiac and renal anomalies, and dysplastic nails and teeth. Congenital cardiac defects occur in 60% of affected individuals.17 In a review of 76 patients with cardiac anomalies and ECV syndrome, 29% had an AVSD, 22% a common atrium, and 18% a combination of both. The remaining cases presented with ASD (11%), VSD (8%), or other defects (11%), including situs inversus.159

Campomelic dysplasia Campomelic dysplasia is a frequently lethal skeletal disorder caused by a severe defect in cartilage development. It is characterized by anterior bowing of the femur and tibia, a large head, a flat nasal bridge, talipes equinovarus, fan-like

Smith–Lemli–Opitz syndrome (SLOS)162 is caused by a disorder of metabolism and features a wide variety of anomalies including postaxial polydactyly, second- and third-toe syndactyly, cataracts, renal anomalies (including cystic kidneys), ambiguous genitalia in males, fetal growth restriction (in late pregnancy), nuchal edema, or hydrops.17,163 A defect in cholesterol synthesis is now recognized as the cause of SLOS, and leads to hypocholesterolemia with elevated levels of 7-dehydrocholesterol (7-DHC). The high levels of 7-DHC are thought to have teratogenic effects.164 It is inherited in an autosomal recessive manner. Almost half of the patients have heart defects, with a strong predominance of atrioventricular septal defect and ventricular septal defect, and, less often, hypoplastic left heart, teralogy of Fallot, tricuspid atresia, and coarctation of the aorta.146,163,165

Noonan syndrome Noonan syndrome166 is a multiple congenital anomaly syndrome comprising typical facial changes and various somatic abnormalities, including short stature, lymphedema, genital anomalies, and cardiac defects (mainly pulmonary stenosis and hypertrophic cardiomyopathy). It is inherited in an autosomal dominant manner.17 The most common sonographic findings in previous prenatal series were polyhydramnios (58%), cystic hygroma (42%), increased nuchal translucency or fetal hydrops (33%), and cardiac anomalies (29%).167–169 However, due to the diversity of prenatal presentation the diagnosis of Noonan syndrome is challenging.

Goldenhar syndrome (hemifacial microsomia) Goldenhar syndrome170 is a sporadic malformation syndrome featuring facial and vertebral anomalies.17 Facial abnormalities are bilateral, but asymmetrical, and consist of ear deformities, ear dystopia (85%), and eye anomalies such as epibulbar dermoids (65–70%) and upper eye lid colombomas (25–30%), facial clefting (30–35%), and hemifacial macrosomia (75–80%). Vertebral anomalies are present in 70–75% of cases and affect mainly the cervical and upper thoracic regions.155,170 Thoracic defects,

Associated anomalies in congenital heart disease

renal anomalies, and cardiac defects are further features of Goldenhar syndrome.17 Cardiac defects are present in up to 35% of cases, with tretralogy of Fallot and ventricular septal defect being most frequent amongst a wide variety of other cardiac defects.171

CHARGE syndrome CHARGE172 is an acronym for coloboma of the eye (C), heart anomaly (H), choanal atresia (A), retardation of mental and somatic development (R), genital anomalies (G), ear abnormalities (E) and/or deafness. Facial palsy, cleft palate, and dysphagia are commonly associated. Cardiac defects of various types occur in CHARGE syndrome with incidences ranging from 50 to 85%,173,174 and include atrial or ventricular septal defect, atrioventricular septal defect, tetralogy of Fallot, double-outlet right ventricle, and right aortic arch.174

Beckwith–Wiedemann syndrome Beckwith–Wiedemann syndrome175 (exomphalos–macroglossia–gigantism syndrome) is sporadic in most cases and features macrosomia, macroglossia, visceromegaly, omphalocele, and a propensity to develop childhood tumors.176 However, a number of additional other malformations are frequently present, namely cardiovascular defects. Greenwood et al reported structural cardiac anomalies in 7/13 (58%) cases with Beckwith–Wiedemann syndrome. No specific type of cardiac abnormality predominated.177

651

and interruption of the inferior vena cava with azygos continuation.14,180–189 Typical findings in right isomerism are bilateral morphologic right atrial appendages (right atrial isomerism), viscerocardiac heterotaxy (situs ambiguus; with incoherent laterality of heart axis, stomach, portal sinus, or gallbladder), multiple severe cardiac anomalies (with a predominance of atrioventricular septal defect, right outflow tract obstruction, anomalies of ventriculoarterial connections, and anomalous pulmonary venous return), bilateral morphologic right (trilobed) lungs with eparterial bronchi, an absent spleen (asplenia), and a malpositioned inferior vena cava, which may be anterior or juxtaposed to the aorta.14,181,182,184–187,190,191 Among the 78 fetuses diagnosed with heterotaxy syndromes in our centers over a period of 15 years, 91% had major congenital heart defects (89% in left isomerism and 95% in right isomerism). In left isomerism the most frequently diagnosed cardiac malformations were atrioventricular septal defect (70%), right outflow tract obstruction (38%), and double-outlet right ventricle (20%). In right isomerism 60% had an atrioventricular septal defect, 45% right outflow tract obstruction, 36% anomalous pulmonary venous drainage, and 27% each had doubleoutlet right ventricle and transposition of the great arteries.184,188,191 As the cardiac malformations associated with heterotaxy syndromes show a considerable overlap during the prenatal period, and since the postnatal diagnostic criteria include features that are not reliably assessed in utero (e.g. lung lobulation, bronchial branching pattern, and spleen status), prenatal diagnosis of left and right isomerism has traditionally relied on the presence of heart block, cardiac defects, interruption of the inferior vena cava, and juxtaposition of the inferior vena cava and aorta in previous prenatal publications.14,184–187

Heterotaxy syndromes Heterotaxy is defined as the abnormal arrangement of viscera across the left–right axis, differing from complete situs solitus and complete situs inversus.178,179 There are two recognized variants of heterotaxy: left isomerism and right isomerism. Left isomerism is associated with paired left-sided viscera while right-sided viscera may be absent. In contrast, right isomerism features paired right-sided viscera while left-sided viscera may be absent. Both variants are associated with complex cardiac malformations. Typical findings in left isomerism are bilateral morphologic left atrial appendages (left atrial isomerism), viscerocardiac heterotaxy (situs ambiguus; with incoherent laterality of heart axis, stomach, portal sinus, or gallbladder), multiple cardiac anomalies (with a predominance of atrioventricular septal defect and right outflow tract obstruction), congenital heart block, bilateral morphologic left (bilobed) lungs with hyparterial bronchi, multiple splenules (polysplenia), intestinal malrotation,

References 1. Ferencz C, Rubin JD, McCarter RJ et al. Congenital heart disease: prevalence at livebirth. The Baltimore–Washington Infant Study. Am J Epidemiol 1985; 121: 31–6. 2. Tegnander E, Williams W, Johansen OJ, Blaas HG, Eik-Nes SH. Prenatal detection of heart defects in a nonselected population of 30,149 fetuses—detection rates and outcome. Ultrasound Obstet Gynecol 2006; 27: 252–65. 3. Tegnander E, Eik-Nes SH, Johansen OJ, Linker DT. Prenatal detection of heart defects at the routine fetal examination at 18 weeks in a non-selected population. Ultrasound Obstet Gynecol 1995; 5: 372–80. 4. Paladini D, Russo M, Teodoro A et al. Prenatal diagnosis of congenital heart disease in the Naples area during the years 1994–1999 — the experience of a joint fetal-pediatric cardiology unit. Prenat Diagn 2002; 22: 545–52. 5. Fesslova V, Nava S, Villa L. Evolution and long term outcome in cases with fetal diagnosis of congenital heart

652

6.

7.

8.

9.

10.

11.

12. 13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

Fetal Cardiology

disease: Italian multicentre study. Fetal Cardiology Study Group of the Italian Society of Pediatric Cardiology. Heart 1999; 82: 594–9. Pradat P, Francannet C, Harris JA, Robert E. The epidemiology of cardiovascular defects, part I: a study based on data from three large registries of congenital malformations. Pediatr Cardiol 2003; 24: 195–221. Carvalho JS, Mavrides E, Shinebourne EA, Campbell S, Thilaganathan B. Improving the effectiveness of routine prenatal screening for major congenital heart defects. Heart 2002; 88: 387–91. Copel JA, Pilu G, Kleinman CS. Congenital heart disease and extracardiac anomalies: associations and indications for fetal echocardiography. Am J Obstet Gynecol 1986; 154: 1121–32. Brick DH, Allan LD. Outcome of prenatally diagnosed congenital heart disease: an update. Pediatr Cardiol 2002; 23: 449–53. Allan LD, Sharland GK, Milburn A et al. Prospective diagnosis of 1,006 consecutive cases of congenital heart disease in the fetus. J Am Coll Cardiol 1994; 23: 1452–8. Machado MV, Crawford DC, Anderson RH, Allan LD. Atrioventricular septal defect in prenatal life. Br Heart J 1988; 59: 352–5. Allan LD. Atrioventricular septal defect in the fetus. Am J Obstet Gynecol 1999; 181: 1250–3. Fesslova V, Villa L, Nava S et al. Spectrum and outcome of atrioventricular septal defect in fetal life. Cardiol Young 2002; 12: 18–26. Huggon IC, Cook AC, Smeeton NC, Magee AG, Sharland GK. Atrioventricular septal defects diagnosed in fetal life: associated cardiac and extra-cardiac abnormalities and outcome. J Am Coll Cardiol 2000; 36: 593–601. Delisle MF, Sandor GG, Tessier F, Farquharson DF. Outcome of fetuses diagnosed with atrioventricular septal defect. Obstet Gynecol 1999; 94: 763–7. Respondek ML, Binotto CN, Smith S et al. Extracardiac anomalies, aneuploidy and growth retardation in 100 consecutive fetal congenital heart defects. Ultrasound Obstet Gynecol 1994; 4: 272–8. Pajkrt E, Weisz B, Firth HV, Chitty LS. Fetal cardiac anomalies and genetic syndromes. Prenat Diagn 2004; 24: 1104–15. Paladini D, Palmieri S, Lamberti A et al. Characterization and natural history of ventricular septal defects in the fetus. Ultrasound Obstet Gynecol 2000; 16: 118–22. Axt-Fliedner R, Schwarze A, Smrcek J et al. Isolated ventricular septal defects detected by color Doppler imaging: evolution during fetal and first year of postnatal life. Ultrasound Obstet Gynecol 2006; 27: 266–73. Harris JA, Francannet C, Pradat P, Robert E. The epidemiology of cardiovascular defects, part 2: a study based on data from three large registries of congenital malformations. Pediatr Cardiol 2003; 24: 222–35. Yu IT, Hutchins GM. Truncus arteriosus malformation: a developmental arrest at Carnegie stage 14. Teratology 1996; 53: 31–7. Copel JA, Pilu G, Kleinman CS. Extracardiac anomalies and congenital heart disease. Semin Perinatol 1993; 17: 89–105.

23. Poon LC, Huggon IC, Zidere V, Allan LD. Tetralogy of Fallot in the fetus in the current era. Ultrasound Obstet Gynecol 2007; 29: 625–7. 24. Tometzki AJ, Suda K, Kohl T, Kovalchin JP, Silverman NH. Accuracy of prenatal echocardiographic diagnosis and prognosis of fetuses with conotruncal anomalies. J Am Coll Cardiol 1999; 33: 1696–701. 25. Vesel S, Rollings S, Jones A et al. Prenatally diagnosed pulmonary atresia with ventricular septal defect: echocardiography, genetics, associated anomalies and outcome. Heart 2006; 92: 1501–5. 26. Volpe P, Paladini D, Marasini M et al. Characteristics, associations and outcome of absent pulmonary valve syndrome in the fetus. Ultrasound Obstet Gynecol 2004; 24: 623–8. 27. Boudjemline Y, Fermont L, Le B, idois J et al. Prevalence of 22q11 deletion in fetuses with conotruncal cardiac defects: a 6-year prospective study. J Pediatr 2001; 138: 520–4. 28. Chessa M, Butera G, Bonhoeffer P et al. Relation of genotype 22q11 deletion to phenotype of pulmonary vessels in tetralogy of Fallot and pulmonary atresiaventricular septal defect. Heart 1998; 79: 186–90. 29. Kim N, Friedberg MK, Silverman NH. Diagnosis and prognosis of fetuses with double outlet right ventricle. Prenat Diagn 2006; 26: 740–5. 30. Volpe P, Paladini D, Marasini M et al. Common arterial trunk in the fetus: characteristics, associations, and outcome in a multicentre series of 23 cases. Heart 2003; 89: 1437–41. 31. Duke C, Sharland GK, Jones AM, Simpson JM. Echocardiographic features and outcome of truncus arteriosus diagnosed during fetal life. Am J Cardiol 2001; 88: 1379–84. 32. Simpson JM. Hypoplastic left heart syndrome. Ultrasound Obstet Gynecol 2000; 15: 271–8. 33. Raymond FL, Simpson JM, Sharland GK, Ogilvie Mackie CM. Fetal echocardiography as a predictor of chromosomal abnormality. Lancet 1997; 350: 930. 34. Allan LD, Apfel HD, Printz BF. Outcome after prenatal diagnosis of the hypoplastic left heart syndrome. Heart 1998; 79: 371–3. 35. Brackley KJ, Kilby MD, Wright JG et al. Outcome after prenatal diagnosis of hypoplastic left-heart syndrome: a case series. Lancet 2000; 356: 1143–7. 36. Wald RM, Tham EB, McCrindle BW et al. Outcome after prenatal diagnosis of tricuspid atresia: a multicenter experience. Am Heart J 2007; 153: 772–8. 37. Yoo SJ, Min JY, Lee YH et al. Fetal sonographic diagnosis of aortic arch anomalies. Ultrasound Obstet Gynecol 2003; 22: 535–46. 38. Chaoui R, Schneider MB, Kalache KD. Right aortic arch with vascular ring and aberrant left subclavian artery: prenatal diagnosis assisted by three-dimensional power Doppler ultrasound. Ultrasound Obstet Gynecol 2003; 22: 661–3. 39. Bronshtein M, Lorber A, Berant M et al. Sonographic diagnosis of fetal vascular rings in early pregnancy. Am J Cardiol 1998; 81: 101–3.

Associated anomalies in congenital heart disease

40. Patel CR, Lane JR, Spector ML, Smith PC. Fetal echocardiographic diagnosis of vascular rings. J Ultrasound Med 2006; 25: 251–7. 41. Rauch R, Rauch A, Koch A et al. Laterality of the aortic arch and anomalies of the subclavian artery – reliable indicators for 22q11.2 deletion syndromes? Eur J Pediatr 2004; 163: 642–5. 42. Hastreiter AR, D’Cruz IA, Cantez T, Namin EP, Licata R. Right-sided aorta. I. Occurrence of right aortic arch in various types of congenital heart disease. II. Right aortic arch, right descending aorta, and associated anomalies. Br Heart J 1966; 28: 722–39. 43. Webber SA, Hatchwell E, Barber JC et al. Importance of microdeletions of chromosomal region 22q11 as a cause of selected malformations of the ventricular outflow tracts and aortic arch: a three-year prospective study. J Pediatr 1996; 129: 26–32. 44. Berg C, Bender F, Soukup M et al. Right aortic arch detected in fetal life. Ultrasound Obstet Gynecol 2006; 28: 882–9. 45. Valletta EA, Pregarz M, Bergamo-Andreis IA, Boner AL. Tracheoesophageal compression due to congenital vascular anomalies (vascular rings). Pediatr Pulmonol 1997; 24: 93–105. 46. Glew D, Hartnell GG. The right aortic arch revisited. Clin Radiol 1991; 43: 305–7. 47. Yoo SJ, Lee YH, Kim ES et al. Tetralogy of Fallot in the fetus: findings at targeted sonography. Ultrasound Obstet Gynecol 1999; 14: 29–37. 48. Paladini D, Volpe P, Russo MG et al. Aortic coarctation: prognostic indicators of survival in the fetus. Heart 2004; 90: 1348–9. 49. Lindsay EA, Goldberg R, Jurecic V et al. Velo-cardio-facial syndrome: frequency and extent of 22q11 deletions. Am J Med Genet 1995; 57: 514–22. 50. Takahashi K, Kuwahara T, Nagatsu M. Interruption of the aortic arch at the isthmus with DiGeorge syndrome and 22q11.2 deletion. Cardiol Young 1999; 9: 516–18. 51. Lewin MB, Lindsay EA, Jurecic V et al. A genetic etiology for interruption of the aortic arch type B. Am J Cardiol 1997; 80: 493–7. 52. Rauch A, Hofbeck M, Leipold G et al. Incidence and significance of 22q11.2 hemizygosity in patients with interrupted aortic arch. Am J Med Genet 1998; 78: 322–31. 53. Volpe P, Marasini M, Caruso G, Gentile M. Prenatal diagnosis of interruption of the aortic arch and its association with deletion of chromosome 22q11. Ultrasound Obstet Gynecol 2002; 20: 327–31. 54. Volpe P, Marasini M, Caruso G et al. 22q11 deletions in fetuses with malformations of the outflow tracts or interruption of the aortic arch: impact of additional ultrasound signs. Prenat Diagn 2003; 23: 752–7. 55. Paladini D, Volpe P, Marasini M et al. Diagnosis, characterization and outcome of congenitally corrected transposition of the great arteries in the fetus: a multicenter series of 30 cases. Ultrasound Obstet Gynecol 2006; 27: 281–5. 56. Todros T, Paladini D, Chiappa E et al. Pulmonary stenosis and atresia with intact ventricular septum during prenatal life. Ultrasound Obstet Gynecol 2003; 21: 228–33.

653

57. Silva SR, Bruner JP, Moore CA. Prenatal diagnosis of Down’s syndrome in the presence of isolated Ebstein’s anomaly. Fetal Diagn Ther 1999; 14: 149–51. 58. Bauk L, Espinola-Zavaleta N, Munoz-Castellanos L. Ebstein’s malformation in the setting of Down’s syndrome. Cardiol Young 2003; 13: 370–2. 59. Hornberger LK, Sahn DJ, Kleinman CS, Copel JA, Reed KL. Tricuspid valve disease with significant tricuspid insufficiency in the fetus: diagnosis and outcome. J Am Coll Cardiol 1991; 17: 167–73. 60. Sharland GK, Chita SK, Allan LD. Tricuspid valve dysplasia or displacement in intrauterine life. J Am Coll Cardiol 1991; 17: 944–9. 61. Siebert JR, Barr M Jr, Jackson JC, Benjamin DR. Ebstein’s anomaly and extracardiac defects. Am J Dis Child 1989; 143: 570–2. 62. Vergani P, Locatelli A, Strobelt N et al. Clinical outcome of mild fetal ventriculomegaly. Am J Obstet Gynecol 1998; 178: 218–22. 63. Gaglioti P, Danelon D, Bontempo S et al. Fetal cerebral ventriculomegaly: outcome in 176 cases. Ultrasound Obstet Gynecol 2005; 25: 372–7. 64. Breeze AC, Alexander PM, Murdoch EM et al. Obstetric and neonatal outcomes in severe fetal ventriculomegaly. Prenat Diagn 2007; 27: 124–9. 65. Fratelli N, Papageorghiou AT, Prefumo F et al. Outcome of prenatally diagnosed agenesis of the corpus callosum. Prenat Diagn 2007; 27: 512–17. 66. Volpe P, Paladini D, Resta M et al. Characteristics, associations and outcome of partial agenesis of the corpus callosum in the fetus. Ultrasound Obstet Gynecol 2006; 27: 509–16. 67. Ecker JL, Shipp TD, Bromley B, Benacerraf B. The sonographic diagnosis of Dandy-Walker and Dandy-Walker variant: associated findings and outcomes. Prenat Diagn 2000; 20: 328–32. 68. Ulm B, Ulm MR, Deutinger J, Bernaschek G. Dandy-Walker malformation diagnosed before 21 weeks of gestation: associated malformations and chromosomal abnormalities. Ultrasound Obstet Gynecol 1997; 10: 167–70. 69. Ong S, Tonks A, Woodward ER, Wyldes MP, Kilby MD. An epidemiological study of holoprosencephaly from a regional congenital anomaly register: 1995–2004. Prenat Diagn 2007; 27: 340–7. 70. Hyett J, Moscoso G, Nicolaides K. Abnormalities of the heart and great arteries in first trimester chromosomally abnormal fetuses. Am J Med Genet 1997; 69: 207–16. 71. Ierullo AM, Ganapathy R, Crowley S et al. Neonatal outcome of antenatally diagnosed congenital cystic adenomatoid malformations. Ultrasound Obstet Gynecol 2005; 26: 150–3. 72. Adzick NS, Harrison MR, Crombleholme TM, Flake AW, Howell LJ. Fetal lung lesions: management and outcome. Am J Obstet Gynecol 1998; 179: 884–9. 73. Brantberg A, Blaas HG, Haugen SE, Eik-Nes SH. Esophageal obstruction – prenatal detection rate and outcome. Ultrasound Obstet Gynecol 2007; 30: 180–7. 74. Stringer MD, McKenna KM, Goldstein RB et al. Prenatal diagnosis of esophageal atresia. J Pediatr Surg 1995; 30: 1258–63.

654

Fetal Cardiology

75. Sparey C, Jawaheer G, Barrett AM, Robson SC. Esophageal atresia in the Northern Region Congenital Anomaly Survey, 1985–1997: prenatal diagnosis and outcome. Am J Obstet Gynecol 2000; 182: 427–31. 76. Goldstein RB. The thorax. In: Nyberg D, McGahan JP, Pretorius D, Pilu G, eds. Diagnostic Imaging of Fetal Anomalies. Philadelphia: Lippincott Williams & Wilkins, 2003: 381–420. 77. Deurloo KL, Devlieger R, Lopriore E et al. Isolated fetal hydrothorax with hydrops: a systematic review of prenatal treatment options. Prenat Diagn 2007; 27: 893–9. 78. Rustico MA, Lanna M, Coviello D, Smoleniec J, Nicolini U. Fetal pleural effusion. Prenat Diagn 2007; 27: 793–9. 79. Waller K, Chaithongwongwatthana S, Yamasmit W, Donnenfeld AE. Chromosomal abnormalities among 246 fetuses with pleural effusions detected on prenatal ultrasound examination: factors associated with an increased risk of aneuploidy. Genet Med 2005; 7: 417–21. 80. Garne E, Haeusler M, Barisic I et al. Congenital diaphragmatic hernia: evaluation of prenatal diagnosis in 20 European regions. Ultrasound Obstet Gynecol 2002; 19: 329–33. 81. Dommergues M, Louis-Sylvestre C, Mandelbrot L et al. Congenital diaphragmatic hernia: can prenatal ultrasonography predict outcome? Am J Obstet Gynecol 1996; 174: 1377–81. 82. Geary MP, Chitty LS, Morrison JJ et al. Perinatal outcome and prognostic factors in prenatally diagnosed congenital diaphragmatic hernia. Ultrasound Obstet Gynecol 1998; 12: 107–11. 83. Heydanus R, Spaargaren MC, Wladimiroff JW. Prenatal ultrasonic diagnosis of obstructive bowel disease: a retrospective analysis. Prenat Diagn 1994; 14: 1035–41. 84. Murshed R, Nicholls G, Spitz L. Intrinsic duodenal obstruction: trends in management and outcome over 45 years (1951–1995) with relevance to prenatal counselling. Br J Obstet Gynaecol 1999; 106: 1197–9. 85. Surana R, Puri P. Small intestinal atresia: effect on fetal nutrition. J Pediatr Surg 1994; 29: 1250–2. 86. Nyberg D, Neilsen I. Abdomen and gastrointestinal tract. In: Nyberg D, McGahan JP, Pretorius D, Pilu G, eds. Diagnostic Imaging of Fetal Anomalies. Philadelphia: Lippincott Williams & Wilkins, 2003: 547–603. 87. Gibbin C, Touch S, Broth RE, Berghella V. Abdominal wall defects and congenital heart disease. Ultrasound Obstet Gynecol 2003; 21: 334–7. 88. Fratelli N, Papageorghiou AT, Bhide A et al. Outcome of antenatally diagnosed abdominal wall defects. Ultrasound Obstet Gynecol 2007; 30: 266–70. 89. Brantberg A, Blaas HG, Haugen SE, Eik-Nes SH. Characteristics and outcome of 90 cases of fetal omphalocele. Ultrasound Obstet Gynecol 2005; 26: 527–37. 90. Cantrell JR, Haller JA, Ravitch MM. A syndrome of congenital defects involving the abdominal wall, sternum, diaphragm, pericardium, and heart. Surg Gynecol Obstet 1958; 107: 602–14. 91. Hornberger LK, Colan SD, Lock JE, Wessel DL, Mayer JE Jr. Outcome of patients with ectopia cordis and significant intracardiac defects. Circulation 1996; 94: II32–7.

92. Greenwood RD, Rosenthal A, Nadas AS. Cardiovascular malformations associated with congenital anomalies of the urinary system. Observations in a series of 453 infants and children with urinary system malformations. Clin Pediatr (Phila) 1976; 15: 1101–4. 93. Gucer S, Ince T, Kale G et al. Noncardiac malformations in congenital heart disease: a retrospective analysis of 305 pediatric autopsies. Turk J Pediatr 2005; 47: 159–66. 94. Newbould MJ, Lendon M, Barson AJ. Oligohydramnios sequence: the spectrum of renal malformations. Br J Obstet Gynaecol 1994; 101: 598–604. 95. Boatman DL, Kolln CP, Flocks RH. Congenital anomalies associated with horseshoe kidney. J Urol 1972; 107: 205–7. 96. Kolln CP, Boatman DL, Schmidt JD, Flocks RH. Horseshoe kidney: a review of 105 patients. J Urol 1972; 107: 203–4. 97. Winyard P, Chitty L. Dysplastic and polycystic kidneys: diagnosis, associations and management. Prenat Diagn 2001; 21: 924–35. 98. Bosman G, Reuss A, Nijman JM, Wladimiroff JW. Prenatal diagnosis, management and outcome of fetal ureteropelvic junction obstruction. Ultrasound Med Biol 1991; 17: 117–20. 99. Vintzileos AM, Campbell WA, Guzman ER et al. Secondtrimester ultrasound markers for detection of trisomy 21: which markers are best? Obstet Gynecol 1997; 89: 941–4. 100. Nyberg DA, Luthy DA, Resta RG, Nyberg BC, Williams MA. Age-adjusted ultrasound risk assessment for fetal Down’s syndrome during the second trimester: description of the method and analysis of 142 cases. Ultrasound Obstet Gynecol 1998; 12: 8–14. 101. DeVore GR. Trisomy 21: 91% detection rate using second-trimester ultrasound markers. Ultrasound Obstet Gynecol 2000; 16: 133–41. 102. Bromley B, Lieberman E, Shipp TD, Benacerraf BR. The genetic sonogram: a method of risk assessment for Down syndrome in the second trimester. J Ultrasound Med 2002; 21: 1087–96. 103. Hobbins JC, Lezotte DC, Persutte WH et al. An 8-center study to evaluate the utility of mid-term genetic sonograms among high-risk pregnancies. J Ultrasound Med 2003; 22: 33–8. 104. Vergani P, Locatelli A, Piccoli MG et al. Critical reappraisal of the utility of sonographic fetal femur length in the prediction of trisomy 21. Prenat Diagn 2000; 20: 210–14. 105. Nyberg DA, Souter VL. Sonographic markers of fetal trisomies: second trimester. J Ultrasound Med 2001; 20: 655–74. 106. Hansmann M. Trisomy 21 in the mid trimester: sonographic phenotyping of the fetus is the key. Ultrasound Obstet Gynecol 2004; 23: 531–4. 107. Bromley B, Lieberman E, Shipp TD, Benacerraf BR. Fetal nose bone length: a marker for Down syndrome in the second trimester. J Ultrasound Med 2002; 21: 1387–94. 108. Bellini C, Boccardo F, Campisi C, Bonioli E. Congenital pulmonary lymphangiectasia. Orphanet J Rare Dis 2006; 1: 43.

Associated anomalies in congenital heart disease

109. Cicero S, Sonek JD, McKenna DS et al. Nasal bone hypoplasia in trisomy 21 at 15–22 weeks’ gestation. Ultrasound Obstet Gynecol 2003; 21: 15–18. 110. Massez A, Rypens F, Metens T, Donner C, Avni FE. The iliac angle: a sonographic marker of trisomy 21 during the midtrimester: dependency of fetal lying? Eur Radiol 2003; 13: 2075–81. 111. Vintzileos A, Walters C, Yeo L. Absent nasal bone in the prenatal detection of fetuses with trisomy 21 in a high-risk population. Obstet Gynecol 2003; 101: 905–8. 112. Maymon R, Tovbin Y, Dreazen E, Weinraub Z, Herman A. All five digits of the hands of fetuses with Down syndrome are short. Ultrasound Obstet Gynecol 2004; 23: 557–60. 113. Rotmensch S, Liberati M, Bronshtein M et al. Prenatal sonographic findings in 187 fetuses with Down syndrome. Prenat Diagn 1997; 17: 1001–9. 114. Paladini D, Tartaglione A, Agangi A et al. The association between congenital heart disease and Down syndrome in prenatal life. Ultrasound Obstet Gynecol 2000; 15: 104–8. 115. Shields LE, Carpenter LA, Smith KM, Nghiem HV. Ultrasonographic diagnosis of trisomy 18: is it practical in the early second trimester? J Ultrasound Med 1998; 17: 327–31. 116. DeVore GR. Second trimester ultrasonography may identify 77 to 97% of fetuses with trisomy 18. J Ultrasound Med 2000; 19: 565–76. 117. Tongsong T, Sirichotiyakul S, Wanapirak C, Chanprapaph P. Sonographic features of trisomy 18 at midpregnancy. J Obstet Gynaecol Res 2002; 28: 245–50. 118. Bronsteen R, Lee W, Vettraino IM et al. Second-trimester sonography and trisomy 18. J Ultrasound Med 2004; 23: 233–40. 119. Kinoshita M, Nakamura Y, Nakano R et al. Thirty-one autopsy cases of trisomy 18: clinical features and pathological findings. Pediatr Pathol 1989; 9: 445–57. 120. Moerman P, Fryns JP, Goddeeris P, Lauweryns JM. Spectrum of clinical and autopsy findings in trisomy 18 syndrome. J Genet Hum 1982; 30: 17–38. 121. Moyano D, Huggon IC, Allan LD. Fetal echocardiography in trisomy 18. Arch Dis Child Fetal Neonatal Ed 2005; 90: F520–2. 122. Nyberg DA, Kramer D, Resta RG et al. Prenatal sonographic findings of trisomy 18: review of 47 cases. J Ultrasound Med 1993; 12: 103–13. 123. Tongsong T, Sirichotiyakul S, Wanapirak C, Chanprapaph P. Sonographic features of trisomy 13 at midpregnancy. Int J Gynaecol Obstet 2002; 76: 143–8. 124. Lehman CD, Nyberg DA, Winter TC 3rd et al. Trisomy 13 syndrome: prenatal US findings in a review of 33 cases. Radiology 1995; 194: 217–22. 125. Chaumoitre K, Brun M, Cassart M et al. Differential diagnosis of fetal hyperechogenic cystic kidneys unrelated to renal tract anomalies: a multicenter study. Ultrasound Obstet Gynecol 2006; 28: 911–17. 126. Jauniaux E, Brown R, Rodeck C, Nicolaides KH. Prenatal diagnosis of triploidy during the second trimester of pregnancy. Obstet Gynecol 1996; 88: 983–9. 127. Rijhsinghani A, Yankowitz J, Strauss RA et al. Risk of preeclampsia in second-trimester triploid pregnancies. Obstet Gynecol 1997; 90: 884–8.

655

128. Mittal TK, Vujanic GM, Morrissey BM, Jones A. Triploidy: antenatal sonographic features with post-mortem correlation. Prenat Diagn 1998; 18: 1253–62. 129. Blaicher W, Ulm B, Ulm MR et al. Dandy-Walker malformation as sonographic marker for fetal triploidy. Ultraschall Med 2002; 23: 129–33. 130. Gembruch U, Baschat AA, Knöpfle G, Hansmann M. Results of chromosomal analysis in fetuses with cardiac anomalies as diagnosed by first- and early second-trimester echocardiography. Ultrasound Obstet Gynecol 1997; 10: 391–6. 131. Surerus E, Huggon IC, Allan LD. Turner’s syndrome in fetal life. Ultrasound Obstet Gynecol 2003; 22: 264–7. 132. Bronshtein M, Zimmer EZ, Blazer S. A characteristic cluster of fetal sonographic markers that are predictive of fetal Turner syndrome in early pregnancy. Am J Obstet Gynecol 2003; 188: 1016–20. 133. Isaksen CV, Eik-Nes SH, Blaas HG et al. A correlative study of prenatal ultrasound and post-mortem findings in fetuses and infants with an abnormal karyotype. Ultrasound Obstet Gynecol 2000; 16: 37–45. 134. Saenger P, Wikland KA, Conway GS et al. Recommendations for the diagnosis and management of Turner syndrome. J Clin Endocrinol Metab 2001; 86: 3061–9. 135. Driscoll DA. Genetic basis of DiGeorge and velocardiofacial syndromes. Curr Opin Pediatr 1994; 6: 702–6. 136. Driscoll DA. Molecular and genetic aspects of DiGeorge/ velocardiofacial syndrome. Methods Mol Med 2006; 126: 43–55. 137. Wilson DI, Burn J, Scambler P, Goodship J. DiGeorge syndrome: part of CATCH 22. J Med Genet 1993; 30: 852–6. 138. Ryan AK, Goodship JA, Wilson DI et al. Spectrum of clinical features associated with interstitial chromosome 22q11 deletions: a European collaborative study. J Med Genet 1997; 34: 798–804. 139. Maeda J, Yamagishi H, Matsuoka R et al. Frequent association of 22q11.2 deletion with tetralogy of Fallot. Am J Med Genet 2000; 92: 269–72. 140. Takahashi K, Kido S, Hoshino K et al. Frequency of a 22q11 deletion in patients with conotruncal cardiac malformations: a prospective study. Eur J Pediatr 1995; 154: 878–81. 141. Amati F, Mari A, Digilio MC et al. 22q11 deletions in isolated and syndromic patients with tetralogy of Fallot. Hum Genet 1995; 95: 479–82. 142. Stewart TL, Irons MB, Cowan JM, Bianchi DW. Increased incidence of renal anomalies in patients with chromosome 22q11 microdeletion. Teratology 1999; 59: 20–2. 143. Boudjemline Y, Fermont L, Le Bidois J et al. Can we predict 22q11 status of fetuses with tetralogy of Fallot? Prenat Diagn 2002; 22: 231–4. 144. Chaoui R, Kalache KD, Heling KS et al. Absent or hypoplastic thymus on ultrasound: a marker for deletion 22q11.2 in fetal cardiac defects. Ultrasound Obstet Gynecol 2002; 20: 546–52. 145. Barrea C, Yoo SJ, Chitayat D et al. Assessment of the thymus at echocardiography in fetuses at risk for 22q11.2 deletion. Prenat Diagn 2003; 23: 9–15. 146. Winter RM, Baraitser M. The London Dysmorphology Database. Oxford: Oxford University Press, 2001.

656

Fetal Cardiology

147. Greenhalgh KL, Howell RT, Bottani A et al. Thrombocytopenia-absent radius syndrome: a clinical genetic study. J Med Genet 2002; 39: 876–81. 148. Kline AD, Krantz ID, Sommer A et al. Cornelia de Lange syndrome: clinical review, diagnostic and scoring systems, and anticipatory guidance. Am J Med Genet A 2007; 143: 1287–96. 149. Kliewer MA, Kahler SG, Hertzberg BS, Bowie JD. Fetal biometry in the Brachmann-de Lange syndrome. Am J Med Genet 1993; 47: 1035–41. 150. Sekimoto H, Osada H, Kimura H et al. Prenatal findings in Brachmann-de Lange syndrome. Arch Gynecol Obstet 2000; 263: 182–4. 151. Aitken DA, Ireland M, Berry E et al. Second-trimester pregnancy associated plasma protein-A levels are reduced in Cornelia de Lange syndrome pregnancies. Prenat Diagn 1999; 19: 706–10. 152. Jackson L, Kline AD, Barr MA, Koch S. de Lange syndrome: a clinical review of 310 individuals. Am J Med Genet 1993; 47: 940–6. 153. Mehta AV, Ambalavanan SK. Occurrence of congenital heart disease in children with Brachmann-de Lange syndrome. Am J Med Genet 1997; 71: 434–5. 154. Bruneau BG, Logan M, Davis N et al. Chamber-specific cardiac expression of Tbx5 and heart defects in HoltOram syndrome. Dev Biol 1999; 211: 100–8. 155. Bergmann C, Zerres K, Peschgens T et al. Overlap between VACTERL and hemifacial microsomia illustrating a spectrum of malformations seen in axial mesodermal dysplasia complex (AMDC). Am J Med Genet A 2003; 121: 151–5. 156. Martinez-Frias ML, Frias JL. VACTERL as primary, polytopic developmental field defects. Am J Med Genet 1999; 83: 13–16. 157. Weaver DD, Mapstone CL, Yu PL. The VATER association. Analysis of 46 patients. Am J Dis Child 1986; 140: 225–9. 158. Stoll C, Clementi M. Prenatal diagnosis of dysmorphic syndromes by routine fetal ultrasound examination across Europe. Ultrasound Obstet Gynecol 2003; 21: 543–51. 159. Digilio MC, Marino B, Ammirati A et al. Cardiac malformations in patients with oral-facial-skeletal syndromes: clinical similarities with heterotaxia. Am J Med Genet 1999; 84: 350–6. 160. Ellis RWB, van Creveld S. A syndrome characterised by ectodermal dysplasia, polydactyly, chondrodysplasia and congenital morbus cordis. Arch Dis Child 1940; 15: 65–84. 161. Promsonthi P, Wattanasirichaigoon D. Prenatal diagnosis of campomelic dysplasia with three-dimensional ultrasound. Ultrasound Obstet Gynecol 2006; 27: 583–5. 162. Smith DW, Lemli L, Opitz JM. A newly recognized syndrome of multiple congenital anomalies. J Pediatr 1964; 64: 210–17. 163. Hyett JA, Clayton PT, Moscoso G, Nicolaides KH. Increased first trimester nuchal translucency as a prenatal manifestation of Smith-Lemli-Opitz syndrome. Am J Med Genet 1995; 58: 374–6. 164. Clayton PT. Disorders of cholesterol biosynthesis. Arch Dis Child 1998; 78: 185–9.

165. Lin AE, Ardinger HH, Ardinger RH Jr, Cunniff C, Kelley RI. Cardiovascular malformations in Smith-Lemli-Opitz syndrome. Am J Med Genet 1997; 68: 270–8. 166. Noonan JA, Ehmke DA. Associated noncardiac malformations in children with congenital heart disease. J Pediatr 1963; 63: 468–70. 167. Nisbet DL, Griffin DR, Chitty LS. Prenatal features of Noonan syndrome. Prenat Diagn 1999; 19: 642–7. 168. Menashe M, Arbel R, Raveh D, Achiron R, Yagel S. Poor prenatal detection rate of cardiac anomalies in Noonan syndrome. Ultrasound Obstet Gynecol 2002; 19: 51–5. 169. Achiron R, Heggesh J, Grisaru D et al. Noonan syndrome: a cryptic condition in early gestation. Am J Med Genet 2000; 92: 159–65. 170. Gorlin RJ, Jue KL, Jacobsen U, Goldschmidt E. Oculoauriculovertebral dysplasia. J Pediatr 1963; 63: 991–9. 171. Pierpont ME, Moller JH, Gorlin RJ, Edwards JE. Congenital cardiac, pulmonary, and vascular malformations in oculoauriculovertebral dysplasia. Pediatr Cardiol 1982; 2: 297–302. 172. Hall BD. Choanal atresia and associated multiple anomalies. J Pediatr 1979; 95: 395–8. 173. Tellier AL, Cormier-Daire V et al. CHARGE syndrome: report of 47 cases and review. Am J Med Genet 1998; 76: 402–9. 174. Gilbert-Barness E, Debich-Spicer D. Cardiovascular system defects. In: Embryo and Fetal Pathology. Cambridge: Cambridge University Press, 2004: 428–69. 175. Wiedemann HR. Complexe malformatif familial avec hernie ombilicale et macroglossie - un ‘syndrome nouveau’? J Genet Hum 1964; 13: 223–32. 176. Nyberg D, Jeanty P, Glass I. Syndromes and multiple anomaly conditions. In: Nyberg D, McGahan JP, Pretorius D, Pilu G, eds. Diagnostic Imaging of Fetal Anomalies. Philadelphia: Lippincott Williams & Wilkins, 2003: 133–220. 177. Greenwood RD, Somer A, Rosenthal A, Craenen J, Nadas AS. Cardiovascular abnormalities in the BeckwithWiedemann syndrome. Am J Dis Child 1977; 131: 293–4. 178. Bowers PN, Brueckner M, Yost HJ. The genetics of leftright development and heterotaxia. Semin Perinatol 1996; 20: 577–88. 179. Lin AE, Ticho BS, Houde K, Westgate MN, Holmes LB. Heterotaxy: associated conditions and hospital-based prevalence in newborns. Genet Med 2000; 2: 157–72. 180. Peoples WM, Moller JH, Edwards JE. Polysplenia: a review of 146 cases. Pediatr Cardiol 1983; 4: 129–37. 181. Van Praagh S, Kakou-Guikahue M, Hae-Seong K et al. Atrial situs in patients with visceral heterotaxy and congenital heart disease: conclusions based on findings in 104 postmortem cases. Coeur 1988; 19: 484–502. 182. Winer-Muram HT, Tonkin IL. The spectrum of heterotaxic syndromes. Radiol Clin North Am 1989; 27: 1147–70. 183. Ho SY, Cook A, Anderson RH, Allan LD, Fagg N. Isomerism of the atrial appendages in the fetus. Pediatr Pathol 1991; 11: 589–608. 184. Berg C, Geipel A, Smrcek J et al. Prenatal diagnosis of cardiosplenic syndromes: a 10-year experience. Ultrasound Obstet Gynecol 2003; 22: 451–9.

Associated anomalies in congenital heart disease

185. Phoon CK, Villegas MD, Ursell PC, Silverman NH. Left atrial isomerism detected in fetal life. Am J Cardiol 1996; 77: 1083–8. 186. Atkinson DE, Drant S. Diagnosis of heterotaxy syndrome by fetal echocardiography. Am J Cardiol 1998; 82: 1147–9, A10. 187. Lin JH, Chang CI, Wang JK et al. Intrauterine diagnosis of heterotaxy syndrome. Am Heart J 2002; 143: 1002–8. 188. Berg C, Geipel A, Kamil D et al. The syndrome of left isomerism: sonographic findings and outcome in prenatally diagnosed cases. J Ultrasound Med 2005; 24: 921–31.

657

189. Berg C, Geipel A, Kohl T et al. Atrioventricular block detected in fetal life: associated anomalies and potential prognostic markers. Ultrasound Obstet Gynecol 2005; 26: 4–15. 190. Ivemark BL. Implications of agenesis of the spleen on the pathogenesis of conotruncus anomalies in childhood. Acta Paediatr Scand 1955; 44(Suppl 104): 1–116. 191. Berg C, Geipel A, Kamil D et al. The syndrome of right isomerism— prenatal diagnosis and outcome. Ultraschall Med 2006; 27: 225–33.

44 The neonate with congenital heart disease – medical and interventional management Ulrike Herberg

Introduction The sensitivity and specificity of prenatal diagnosis of congenital heart disease are constantly improving, owing to technical and scientific progress in the area of fetal echocardiography. Therefore, physicians working in the area of prenatal diagnostics are increasingly confronted with the pre- and postnatal physiology and alternatives for therapy of complex cardiac defects. Counseling of the parents concerned and the planning of pre-, peri-, and postnatal disease management are done in close cooperation between the gynecologist, pediatric cardiologist, cardiac surgeon, and neonatologist. This chapter gives an insight into the importance of peri- and postnatal adaptation of the circulation and the available choices of drug and interventional therapy of congenital heart diseases. Chapter 45 deals in more detail with the options of neonatal cardiac surgery.

Transition from fetal to postnatal circulation The fetal circulation differs uniquely from the postnatal circulation in several ways (Figure 44.1): • the parallel connection and pressure adaptation of the two circulations • the high resistance of the undeveloped lung, preventing significant lung perfusion • the low resistance of the placenta, which supplies oxygen and nutrition • the bypass of relatively highly oxygenated blood through the ductus venosus and foramen ovale via the left atrium and left ventricle to the heart and the brain • the bypass of blood from the right ventricle through the ductus arteriosus to the lower body.

Transition from fetal to postnatal circulation for the healthy fetus Conversion of the fetal to the postnatal circulation involves elimination of the umbilical–placental circulation, establishment of the pulmonary circulation, and separation of the pulmonary and systemic circulation by closure of fetal channels: 1. With the first breath after birth, the lungs of the newborn expand and become filled with air. In consequence, the pulmonary vascular resistance decreases 5–10-fold, leading to an increased pulmonary blood flow. 2. The increased pulmonary blood flow increases the filling and pressure of the left atrium. The pressure in the left atrium then exceeds the pressure in the right atrium. The valve of the foramen ovale is pressed against the atrial septum and either closes functionally or causes a reversal of the interatrial shunt (left-to-right shunt) (Figure 44.2a). 3. The decrease in the pulmonary vascular resistance causes first a cross shunt and then, because of further decrease, a left-to-right shunt across the ductus arteriosus. Owing to an increase of the oxygen concentration and changes in levels of circulating mediators, the ductus arteriosus constricts. 4. The peripheral vascular resistance increases following closure of the umbilical vessels and loss of the vascular system of the placenta. The two previously parallel circulations are now connected in series. After further reduction of the pulmonary vascular resistance and functional closure of the foramen ovale and ductus arteriosus, the majority of the right ventricular stroke volume (cardiac output) reaches the left ventricle after oxygenation (Figure 44.2b). The left ventricle pumps the blood into the upper and lower parts of the body,

660

Fetal Cardiology

The pressure and blood flow of a circulatory compartment is regulated by the resistance of the following vascular system. The hemodynamic changes during and after birth can be divided into two phases: • the immediate perinatal changes during the first few minutes after delivery and establishment of the neonatal circulation over the first 10–15 hours after birth • gradual adaptation to adult pressure and resistance conditions during the first 6 weeks of life. These chronological processes are of great relevance for the hemodynamic function of congenital heart defects and consequently for any timing of therapeutic interventions.

Establishment of the neonatal circulation The neonatal circulation is characterized by sensitive adaptation of the above mentioned flow and resistance conditions. Fetal connections are closed only functionally and not anatomically.

Figure 44.1 Percentages of the combined ventricular output for the late gestation human fetus. Right and left ventricle are working parallel, blood is pumped from the right ventricle through the pulmonary trunk to the descendent aorta via the ductus arteriosus. The fetal right ventricle ejects about 56% of the combined ventricular output, perfusing the lower body and the placenta. The remaining output is handled by the left ventricle, with most of this blood being directed to the coronary vessels and the head. Antenatally only 10% of the combined ventricular output crosses the isthmus region to the descending aorta. The direction of flow is indicated by arrows (adapted with permission from reference 1). Ao, aorta; DA, ductus arteriosus; IVC, inferior vena cava; LA, left atrium; LV, left ventricle; PA, pulmonary artery; PV, pulmonary venous return; RA, right atrium; RV, right ventricle; SVC, superior vena cava.

working against a slightly higher resistance in comparison to the resistance of the pulmonary arteries. The previously dominant right ventricle takes over the low pressure system of the pulmonary circulation. In comparison to the prenatal system, the left ventricle now operates with an increased pressure and volume load. Consequently, equalization of pressure or volume load between the two separate circulatory systems is impossible once the short-circuit is completely closed.

Closure of the ductus arteriosus The ductus arteriosus narrows increasingly during the first 2–7 hours of life. Following clamping of the umbilical cord, a cross shunt via the ductus arteriosus persists for a few hours. With further fall of the pulmonary vascular resistance the cross shunt will be replaced by a left-to-right shunt. Contraction and shortening of the ductus arteriosus, migration of smooth muscle cells from the media, and the development of intimal cushions close the mature ductus arteriosus within 3–4 days.3–5 Classic theories of normal spontaneous ductal closure include a combination of direct smooth muscle contractility in response to the postnatal increase in Po2, reduction in the levels of prostaglandins, and also a decrease in the sensitivity of ductal tissue to the dilating influences of prostaglandins. Rapid histological changes lead to obliteration of the lumen and prevent reopening. These processes result in an anatomical occlusion after 2–3 weeks. The contraction of the ductal tissue causes an anatomical narrowing of the descending part of the aorta in the area of the aortic isthmus. Therefore, an aortic isthmus stenosis becomes apparent after complete closure of the duct. Heart defects with duct-dependent pulmonary or systemic perfusion become symptomatic with progressive ductus constriction (see Chapter 39). The higher incidence of open ductus arteriosus among premature infants compared to the incidence in mature newborns is caused by the lack of differentiation of the ductal tissue (related to gestational age), disturbance of the oxygenation, and acid–base balance. In the premature newborn, in contrast to the mature infant, oxygen has

Management of the neonate with congenital heart disease

661

Figure 44.2 Transitional circulation. (a) Transitional circulation after birth in the human and (b) in the animal model. Ventilation produces a fall in pulmonary vascular resistance, and with the greater rise in pulmonary blood flow, reduces the flow across the ductus arteriosus (cross shunt in the human). Left atrial pressure becomes higher than right atrial pressure, resulting in a partial closure of the foramen ovale with a left-to-right shunt. The increased peripheral vascular resistance after umbilical cord occlusion increases the systemic arterial pressure. (b) The percentages of the combined cardiac output after ventilation with oxygen and umbilical cord occlusion in fetal lambs. The left ventricular output increases from 35% in the fetal lamb to 59% and is slightly higher than that of the right ventricle (41%). The occlusion of the umbilical cord increases the systemic arterial pressure and the left-to-right shunt across the ductus arteriosus (adapted from reference 2).

a less vasoconstrictive effect on the ductus arteriosus, whilst nitric oxide (NO) and prostaglandin E2 have a higher vasodilatory effect.5 After temporary closure the premature ductus arteriosus may reopen spontaneously.

Functional closure of the foramen ovale and ductus venosus The closure of the foramen ovale is functional; even long after the neonatal period a slight residual shunt can be found on echocardiographic examination. For some congenital heart defects, therefore, the foramen ovale can serve as a relief mechanism in cases of right or left ventricular overload (hypoplastic left heart syndrome, tricuspid atresia, Ebstein’s anomaly). The ductus venosus will be closed in 75% of all cases by the seventh day of life.6

Function of the ventricle The right ventricle supplies blood only to the pulmonary circulation (Figure 44.3). The myocardium of the right ventricle, which has been the dominant ventricle in the fetus, adapts to the falling pulmonary resistance and pressure. However, a moderately increased pressure and volume load due to congenital heart disease can be compensated by the in utero trained right ventricle. The left ventricle, however, has to deal relatively abruptly with higher pressure and volume performance, because the pulmonary venous return and – after loss of the placental vascular bed – the systemic vascular resistance are increased. A further increase in afterload (e.g. due to aortic valve stenosis or coarctation) or preload (e.g. due to an increased volume burden by a large ductus arteriosus or ventricular septal defect) can be tolerated only to a limited

662

Fetal Cardiology

Figure 44.3 Serial circulation after closure of fetal communications. (a) In serial circulation, percentages of the total cardiac output are similar in the pulmonary and systemic circulation. (b) Circled figures represent percentages of oxygen saturation; the other numbers represent intracavity pressures for the infant after completed transition. There is no mixing of pulmonary and systemic circulations after closure of fetal shunts.

extent and leads to early decompensation. The diastolic function – myocardial relaxation and compliance – also undergoes a maturation process during the first 6 months of life and is especially affected during the first 2 weeks after birth.7 Postnatal adaptation to the increased volume and pressure load is mainly achieved by hypertrophy of the muscle cells, and to a much lesser extent by hyperplasia. Hypertrophy of the muscle cells is reinforced by hypoxemia and reduces the compliance of the myocardium.8 It is important to keep this in mind when faced with cyanotic heart disease in single ventricle physiology.

Pulmonary vascular resistance The pulmonary vascular resistance is mainly determined by the number of peripheral pulmonary vessels and their state of vasoconstriction. This is the most sensitive section of the transitional circulation. The clearance of the alveolar fluid occurs during the first 4–6 hours of life. The pulmonary vascular resistance

is reduced by the opening of the alveoli, ventilation with oxygen, loss of vasoconstrictors produced in utero, and local vasodilators.9 Pulmonary resistance vessels constrict under the influence of hypoxemia, hypercapnia, or acidosis. Neonatal drug treatment and ventilation makes particular use of the pronounced reactivity of the pulmonary vessel bed.

Adaptation of the pulmonary vascular resistance to the ‘adult’ situation The outstanding studies of Rudolph and Heymann made a major contribution to the understanding of the postnatal adaptation of healthy newborns and newborns with congenital heart defects. Within 24 hours after birth the pulmonary arterial pressure drops to half of the systemic pressure and after 6 weeks of life reaches adult values10,11 (Figure 44.4). The postnatal lung development is anatomically characterized by vascular remodeling, muscular involution, and development of the alveoli.9,14,15

Management of the neonate with congenital heart disease

663

grow and develop normally in utero. If there is obstruction of an efferent vessel, the blood is able to pass via the ductus arteriosus and/or foramen ovale to the neighboring circulatory system. A connection between the two circulatory systems – for example, a ventricular septal defect – does not lead to any net shunting, because there is pressure equalization between both ventricles. The time point of closure of the fetal communications, for example the ductus arteriosus, is identical in both the healthy mature newborn and the mature newborn with a congenital heart defect. Therefore, intrauterine compensated heart defects may decompensate during the transition phase from the fetal to the postnatal circulation. The sequence of transition determines the moment of decompensation and also the timing of any therapeutic intervention for the various congenital heart defects.

Clinical presentation of duct-dependent heart defects Hypoplasia or obstruction of the right or left ventricle, obstruction or atresia of the aortic arch, and the transposition of the great arteries are the major duct-dependent heart defects. The persistence of fetal communications is essential for an adequate postnatal circulation and oxygenation. Since the transition of the circulation and closure of the vital fetal shunts occur at the same time as in healthy newborns, newborns with duct-dependent heart defects initially go through a symptomless postnatal adaptation, but then develop symptoms after a few hours of life (Table 44.1). Chapter 39 deals in more detail with the pathophysiology of perinatal transition in these cases.

Figure 44.4 The changes in pulmonary arterial pressure, blood flow, and vascular resistance that occur around birth. After the initial rapid decrease in pulmonary vascular resistance and pulmonary arterial blood pressure, there is a slow, progressive decrease, with adult levels being reached after 6 weeks (reproduced with permission from reference 11, data from references 12 and 13).

Transition from fetal to postnatal circulation for the fetus with congenital heart disease— implications for the perinatal management The parallel connection of the circulatory systems permits most fetuses with congenital cardiac malformations to

Presentation of heart defects with duct-dependent systemic circulation The hypoplastic left heart syndrome can be used as an example of the duct-dependent systemic circulation. In hypoplastic left heart syndrome, the oxygenated pulmonary venous blood reaches the systemic circulation via the foramen ovale and the right ventricle. The right ventricle pumps both the systemic and the venous return into the pulmonary artery and via the ductus arteriosus to the aorta. The aortic arch, the head, and the coronary arteries are perfused retrogradely with relatively lowoxygenated blood (Figure 44.5). Progressive narrowing of the ductus arteriosus, which acts as a connection to the systemic blood flow, leads to a fall in the blood pressure and underperfusion of the central and peripheral organs – including the coronary circulation. Tachypnea, hypotension, pronounced acidosis, and moderate cyanosis lead to the erroneous presumptive diagnosis of neonatal sepsis or metabolic disorder. Progressive ductus arteriosus

664

Fetal Cardiology

Table 44.1 Clinical presentation of duct-dependent heart disease. Duct-dependent pulmonary flow and inadequate mixing presents as early cyanosis that is unresponsive to oxygen therapy. In duct-dependent systemic blood flow, compromise of systemic perfusion occurs with ductal constriction Early cyanosis

Hypotension, acidosis congestive heart failure

Hypotension, acidosis > 24 h

Duct-dependent pulmonary blood flow

Duct-dependent systemic blood flow

Duct-dependent blood-flow of the lower body

Pulmonary atresia

Hypoplastic left heart syndrome

Critical pulmonary stenosis

Critical aortic stenosis

Severe tetralogy of Fallot

Interrupted aortic arch

Critical coarctation

Duct-dependent mixing Transposition of great arteries

Figure 44.5 Circulation in hypoplastic left heart syndrome (HLHS). (a) A schematic diagram of blood flow, O2 saturations (circled values), and pressures (values in boxes). (b) A depiction of the blood flow and saturations in the normal neonate. In HLHS, systemic blood flow is served by the right ventricle via the pulmonary artery and ductus arteriosus, with retrograde perfusion of the upper trunk and coronary arteries. Pulmonary venous return enters the right heart through the mildly restrictive foramen ovale. The resultant arterial O2 saturation of 75–80% represents balanced pulmonary and systemic blood flow (Qp:Qs = 1) (adapted from reference 17). Qp, pulmonary blood flow (l/min); Qs, systemic blood flow (l/min).

Management of the neonate with congenital heart disease

665

Congenital heart defect with duct-dependent pulmonary circulation Obstructions of the right ventricle such as pulmonary atresia and heart disease with severe right ventricular dysfunction such as severe Ebstein’s disease have a ductdependent pulmonary circulation (Figure 44.6). Clinical signs are severe cyanosis within a few hours of birth, which shows no improvement after treatment with oxygen (hyperoxia test).

Figure 44.6 Circulation in pulmonary atresia with hypoplastic main pulmonary artery. Pulmonary blood flow is served by the left ventricle via the aorta and the ductus arteriosus which has a typical vertical shape. The systemic venous return drains across the open foramen ovale into the left atrium (right-to-left shunt). There is no antegrade flow across the pulmonary valve, and the right ventricle is decompressed by tricuspid regurgitation (adapted from reference 17).

occlusion leads to acute coronary hypoperfusion, myocardial infarction, shock, and multiorgan failure. Early use of prostaglandin derivatives can prevent ductal occlusion for a limited period of time. The functionally single right ventricle serves the pulmonary and systemic circulation. The proportion of the ventricular output that goes to the pulmonary or systemic vascular bed is determined by the relative resistance to flow within the two circuits. As the pulmonary resistance continues to drop, more blood preferentially reaches the pulmonary circulation (Qp) than the systemic circulation (Qs). The resulting increased lung perfusion (Qp > Qs) causes a severe volume overload of the right ventricle, which may already suffer from ischemia due to coronary hypoperfusion. Therefore, intensive medical interventions aim to recognize duct-dependent heart defects as early as possible, as well as regulating pulmonary and systemic perfusion.

Congenital heart disease with duct-dependent mixing The simple transposition of the great arteries is characterized by parallel arrangement of both circulations: the highly saturated pulmonary venous return reaches the left ventricle and recirculates via the pulmonary artery into the lung; in parallel the systemic circulation is provided with less oxygen-rich blood via the right ventricle and the aorta. In transposition of the great arteries with an intact ventricular septum, only a small portion of blood is exchanged by mixing between the two circulations (Figure 44.7a). The monitored oxygen saturation reflects the magnitude of mixing between the systemic and pulmonary circulation. To a certain extent, an open duct facilitates an exchange between the two circulations; however, the limiting factor is not the size of the ductus, but that of the atrial shunt through the foramen ovale. The atrial cross shunt enables a flow of oxygenated blood from the left atrium to the right atrium and consequently to the systemic circulation (Figure 44.7b). Interventional balloon atrioseptostomy allows to create an unrestrictive interatrial shunt and to increase the oxygen saturation.

Pulmonary hypertension in persistent fetal circulation and total anomalous pulmonary venous return In persistent pulmonary hypertension a disturbance of the circulatory transition process is observed. The pulmonary vascular resistance does not drop but instead remains at a high level or even increases. In this situation the right-to-left shunts at the atrial and ductal levels persist. The reasons for pulmonary hypertension are manifold (meconium aspiration, hypoplastic lungs, asphyxia) and beyond the scope of this chapter. The therapeutic options to decrease the pulmonary vascular resistance are ventilation with oxygen, hyperventilation, surfactant, NO, use of vasodilators, and special ventilation regimes such as high-frequency ventilation or the employment of extracorporeal membrane oxygenation.11 Total anomalous pulmonary venous return with obstruction is difficult to distinguish from other causes of persistent pulmonary hypertension. All the pulmonary veins drain anomalously into the systemic venous circulation

666

Fetal Cardiology

Figure 44.7 Circulation in transposition of the great arteries. Parallel connection of the systemic and pulmonary circulation. Arterial O2 saturation reflects mixing between pulmonary and systemic circulations. (a) The ductus arteriosus is already closed, the atrial right-to-left shunt is restrictive and allows minimal mixing. The oxygen saturation of the aorta is similar to the right ventricle, and hence the neonate appears deeply cyanotic. (b) Opening of the arterial duct by prostaglandin E1 (PGE1) and most important of the foramen ovale by balloon-atrial septostomy allows increased mixing and improvement of cyanosis (adapted from reference 17).

rather than the left atrium. The systemic perfusion is dependent on the right-to-left shunt via the ductus arteriosus and the foramen ovale. The assessment of the clinical presentation of a neonate with total anomalous pulmonary venous return is typically based on whether the anomalous draining vein is obstructed or not. Infants with obstruction present with serious, sometimes suprasystemic pulmonary hypertension, profound hypoxemia, and low cardiac output. Therapy in this condition is decompression with a palliative balloon atrial septostomy, combined, if necessary, with reopening of the ductus and emergency surgical intervention. A prenatal increase in the pulmonary vascular resistance, as seen in cases of intrauterine constriction of the ductus arteriosus or pulmonary venous obstruction in hypoplastic left heart syndrome, causes hypertrophy of the media of the pulmonary vessels prenatally and severe pulmonary hypertension postnatally.

Pulmonary hypertension in cardiac shunt defects All congenital heart defects with large-volume shunts at the ventricular and especially at the great artery level with unrestricted pulmonary blood flow show an extremely high risk of developing pulmonary hypertension: these include common truncus arteriosus, transposition of the great arteries with ventricular septal defect or with persistent ductus arteriosus, single ventricle without pulmonary stenosis, large ventricular septal defect and large persistent ductus arteriosus. For most other congenital heart defects the adaptation from the fetal to the neonatal circulation causes a gradual increase in the pressure or volume load without life-threatening effects. This includes, in particular, shunt lesions such as ventricular septal defects, atrioventricular septal defects, and mild valvular stenosis. Chapter 39

Management of the neonate with congenital heart disease

667

Figure 44.8 Postnatal changes in infants with and without large ventricular septal defects. In children with large ventricular septal defect (VSD, dotted line) high pulmonary arterial pressure persists. The pulmonary vascular resistance falls more slowly after birth and does not reach normal levels. Associated with the fall in pulmonary vascular resistance, pulmonary blood flow increases due to left-to-right shunting. After a variable period, the pulmonary vascular resistance begins to increase due to secondary changes of the pulmonary vessels; this is associated with a fall of pulmonary blood flow (adapted with permission from reference 1).

deals in more detail with the pathophysiology of these defects. The presentation of heart failure due to shunt connections depends on the size of the defect and the rapidity of the fall in pulmonary vascular resistance. For many patients, especially those newborns suffering from Down syndrome, the reduction in pulmonary resistance is not as fast or as pronounced as in healthy newborns; in consequence these children develop less severe heart failure. The normal development of the pulmonary arteries proceeds more slowly. The pulmonary artery resistance remains three to four times that of normal, but still allows for a significantly increased pulmonary flow (Figure 44.8). For these children the excessive pulmonary flow also results in reactive changes to the pulmonary vessels. Dysfunction of the endothelium causes a diminished production of vasoconstrictors such as NO. Increased production of vasoconstrictors leads to vasoconstriction, stimulating the proliferation of smooth muscle cells and fibroblasts and increasing the production of prothrombotic substances. Extended muscularization of the peripheral arteries and decreased diameter of the pulmonary vascular bed is histologically detectable.9,14 In consequence, the development of pulmonary hypertension is reversible only for a short time. Without punctual palliative or corrective intervention, serious pulmonary vessel changes with irreversible pulmonary hypertension (Eisenmenger syndrome) occur. In this situation the patient is rendered inoperable. Early corrective measures within the first 3–6 months of life or palliative interventions during the neonatal period such as pulmonary artery banding prevent this condition. Further congenital heart defects that may cause deterioration during the neonatal or infant period are listed in Table 44.2.

Cardiac disease without suspected problems In isolated small ventricular septal defects, atrial septal defects and small ductus arteriosus there is no need for therapy in the neonatal period. Despite large left-to-right shunts at the atrial level, the neonate is not compromised (see Chapter 39).

Drug or interventional treatment of congenital heart disease in the neonate (Table 44.4) Principles of intensive therapy The objective is to provide a sufficient oxygen supply and to maintain systemic perfusion. In neonates with heart defects dependent on the persistence of fetal circulatory communications, the ductus arteriosus can be maintained patent by treatment with prostaglandins E1 and E2 (PGE1 and PGE2). The atrial shunt essential in children with transposition of the great arteries can be enlarged by interventional balloon atrial septostomy on the intensive care unit or in the catheter laboratory. The pulmonary vascular resistance that decreases gradually postpartum can also be influenced by medical or ventilatory interventions. Above all, preoperative stabilization and optimal organ perfusion has to be established following previous circulatory insufficiency or multiorgan

Postnatal physiology

Compromise of systemic perfusion with ductal constriction; low output syndrome, shock

Increased left ventricular pressure load

Compromise of systemic perfusion with ductal constriction; lowoutput syndrome, acidosis; shock

Reduced systolic and diastolic ventricular function

Congenital heart delect

Critical aortic stenosis

Moderate aortic stenosis

Hypoplastic left heart syndrome

Hypertrophic (obstructive) cardiomyopathy

Beta-blockers, diuretics Exclude metabolic or mitochondrial disease Exclude associated syndromes Examine family members

Parental counseling PGE Avoidance of right ventricular volume and pressure overload Therapeutic goal: O2-saturations between 70–80%, PCO; 45–50 mmHg, PO2 35 mmHg, balancing of pulmonary and systemic vascular resistance to achieve adequate flow into both vascular beds

+

–

Careful follow-up

PGE Univentricular (staged Norwood-procedures) or biventricular approach?

+

–

Perinatal management

Ductdependent

Palliative in patients with high surgical risk; stenting of the ductus arteriosus and surgical bilateral banding of the pulmonary arteries

First choice: modified Norwood - Sano procedure or cardiac transplantation

Myotomy-myectomy Transluminal myocardial ablation Dual chamber pacemaker Cardiac transplantation

After Norwood-procedure: bidirectional cavopulmonary shunt (Glenn shunt), modified [fenestrated] Fontan repair [followed by interventional occlusion of fenestrations]; interventional occlusion of collaterals; balloon dilatation of recoarctation; balloon dilatation of surgical anastomoses, late cardiac transplantation

Interventional balloon angioplasty if transaortic gradient >50mmHg or symptoms

Valve regurgitation: Ross-procedure, valve replacement

Restenosis: redilatation, surgical revalvuloplasty

Biventricular approach: surgical valvotomy; tunnel subaortic stenosis with hypoplastic aortic valve: Ross-Konnoprocedure; Univentricular: Norwood-Sanoprocedure; Alternative: cardiac transplantation

First choice: interventional balloon valvuloplasty in borderline cases: ductal stenting and bilateral pulmonary banding

Interventions and problems in medium- and long-term follow-up

Surgical intervention in the neonate

Transcatheter intervention in the neonate

Table 44.2 Postnatal physiology and hemodynamics, medical, interventional and surgical therapy in the neonatal period and mid- to long-term interventions in various congenital heart defects

668 Fetal Cardiology

Double outlet right ventricle without pulmonary stenosis

Double inlet left (single) ventricle

L-transposition of the great arteries

D-transposition of the great arteries

Interrupted aortic arch Mitral stenosis

Aortic coarctation

Pulmonary overflow Congestive heart failure

Dependent on associated lesions inflow in the left ventricle is disturbed, decompression of the left atrium across the foramen ovale Parallel outputs of systemic right and pulmonary left ventricle O2-saturation dependent on magnitude of mixing between both circulations (foramen ovale, ductus, VSD) Dependent on associated lesions (VSD, PS, WPWsyndrome) Hemodynamics dependent on pulmonary stenosis

Critical coarctation: ductal dependent perfusion of the lower limb, hypertension of the upper limb, increased left ventricular pressure load See critical coarctation

–

–

(+)

Treatment of heart failure

Dependent on associated pulmonary stenosis or subaortic stenosis

PGE Dependent on the O2-saturation (foramen ovale, VSD): balloon atrial septostomy

In single ventricle hemodynamic: PGE and atrial septostomy

(+)

+

PGE

PGE in critical coarctation Check for other cardiac malformations

+

(+)

Severe pulmonary stenosis: balloon dilatation

Balloon atrial septostomy

Palliative: mitral valve balloon valvotomy (rare)

Balloon dilatation in circumscribed stenosis Balloon dilatation and/or stenting in critically ill neonates

Dysfunction of the right systemic ventricle

If arterial switch not possible: staged repair after pulmonary banding, Rastelli-repair

Balloon dilatation of restenosis or stent placement

Restenoses: balloon dilatation or stent placement Hypertension

(Continued )

Modified Fontan repair Severe pulmonary stenosis: BlalockTaussig-Shunt No pulmonary stenosis: pulmonary banding Severe subaortic stenosis: Damus-Kaye-Stansel Pulmonary banding or Surgical: modification dependent early corrective surgery on the position of the VSD and associated malformations

Classic repairs of associated lesions

Arterial switch operation day 7–14

Supravalvular ring: surgical resection; severe mitral valve stenosis associated with left ventricular hypoplasia: Norwood palliation

Surgical repair

First choice: surgical repair: resection and end-to-end repair or subclavian flap

Management of the neonate with congenital heart disease 669

Careful follow up

Increased pressure load of the right ventricle, reduced pulmonary blood flow

Pulmonary stenosis

–

See critical pulmonary stenosis

Pulmonary atresia with intact ventricular septum

PGE Biventricular or univentricular repair? (dependent on right ventricular morphology, the development of pulmonary arteries and the presence of right ventricular dependent coronary artery circulation)

PGE

+

Ductal dependent pulmonary flow increase pressure load of the right ventricle right-to-left shunt across foramen ovale

Critical pulmonary stenosis/ membranous pulmonary atresia

+

Associated with severe pulmonary stenosis: PGE

(+)

See Tetralogy of Fallot

Perinatal management

Double outlet right ventricle with pulmonary stenosis

Ductdependent

Postnatal physiology

Congenital heart delect

Table 44.2 Continued

Balloon dilatation if right ventricular pressure 2/3 to left systemic pressure or if symptoms

First choice: balloon dilatation Membranous PA: perforation of membrane and balloon dilatation in cases with well developed right ventricle and pulmonary arteries. otherwise as palliative procedure in poor surgical patients Ductal stenting

In severe pulmonary stenosis: balloon dilatation

Transcatheter intervention in the neonate

Intraventricular tunnel repair/Rastelli procedure Redilatation in restenosis Homograft in severe pulmonary insufficiency or stenosis 1.5 ventricle repair when RV is hypoplastic

Redilatation in restenosis Homograft in severe pulmonary insufficiency or stenosis 1.5 ventricle repair or univentricular repair when RV is hypoplastic

Redilatation in restenosis Homograft in severe pulmonary insufficiency or stenosis

Surgical valvotomy and/ or palliation by BlalockTaussig shunt

Surgical valvotomy and/ or reconstruction of the right ventricular outflow tract and/or palliation by Blalock-Taussig-shunt to maintain sufficient pulmonary blood supply

Rare: surgical valvuloplasty

Interventions and problems in medium- and long-term follow-up

In severe pulmonary stenosis: Blalock-Taussig shunt

Surgical intervention in the neonate

670 Fetal Cardiology

PGE in TA + PS or TA + PA or TA + transposition + restrictive VSD or coarctation

PGE in (functionally) pulmonary atresia; therapy of right heart failure

(+)

(+)

Right-to-left shunt across foramen ovale, hemodynamics dependent on associated lesions

Hemodynamics dependent on right ventricular hypoplasia and tricuspid insufficiency, right-to-left shunt across foramen ovale, arrhythmia (WPWsyndrome, first degree AV-block)

Right sided volume loading

Tricuspid atresia

Ebstein’s anomaly

Atrial septal defect

–

PGE

+

Pulmonary perfusion dependent on ductus arteriosus and main aortopulmonary collaterals

Pulmonary atresia with ventricular septal defect or tetralogy of Fallot with servere pulmonary stenosis and main aortopulmonary collaterals (MAPCAS)

–

In severe pulmonary stenosis PGE Beta-blockers to prevent hypoxic spells

(+)

Hemodynamics dependent on severity of right ventricular outflow obstruction and hypoplasia of the pulmonary valve and arteries

Tetralogy of Fallot

In case of MAPCAS without central pulmonary arteries: Unifocalization. Right ventricularpulmonary artery-conduit Interventional coil occlusion of main aortopulmonary collaterals Right ventricular dysfunction Ventricular arrhythmia

Palliative: systemico-topulmonary artery shunt with or without right ventricular outflow tract reconstruction Early corrective surgery

–

Balloon dilatation in pulmonary stenosis

–

(Continued )

Interventional or surgical occlusion; atrial septal defect with partial anomalous pulmonary venous connection: surgical repair

Tricuspid valce reconstruction, valve replacement; 1.5 ventriclerepair; modified Fontan repair

Modified Fontan repair

Repair Interventional coil occlusion of main aortopulmonary collaterals Homograft in pulmonary insufficiency Right ventricular dysfunction Ventricular arrhythmia

Blalock-Taussig shunt in severe pulmonary stenosis Early repair

Atrial septostomy Blalock-Taussig shunt in when foramen ovale is ductal dependent restrictive pulmonary perfusion; coarctation-repair, pulmonary banding in pulmonary overflow

Palliative: stenting of the ductus arteriosus or main aortopulmonary collaterals

Balloon valvotomy in severe pulmonary stenosis (and high surgical risk)

Management of the neonate with congenital heart disease 671

Postnatal physiology

Right-to-left shunt across foramen ovale permits maintenance of systemic perfusion; pulmonary venous congestion; ductus arteriosus patency provides relief in pulmonary venous obstruction

Left ventricular volume overloading with falling pulmonary vascular resistance, pulmonary edema, congestive heart failure

pulmonary vascular resistance and size of atrial and ventricular communication determine clinical presentation

Left ventricular volume overloading with falling pulmonary vascular resistance, pulmonary edema, congestive heart failure

Dependent on pulmonary perfusion, mostly pulmonary congestion and ventricular volume overloading

Congenital heart delect

Total anomalous pulmonary venous return

Ventricular septal defect

Atrioventricular septal defect

Persistent ductus arteriosus Botalli

Truncus arteriosus communis

Table 44.2 Continued

rare

Surgical repair

Surgical ligation or In the neonatal period: interventional division in the occlusion only in large symptomatic neonate communications after indomethacin-failure

Avoidance of hypoxemia, acidosis, fluid restriction; treatment by indomethacin or ibuprofen in premature babies

Anticongestive therapy

–

–

(+)

In rare cases with unbalanced atrioventricular septal defect: single ventricle physiology Anticongestive therapy

Anticongestive therapy

–

immediate repair

Atrial septostomy as palliative procedure before repair

Pulmonary banding in cases with large communication and left ventricular outflow stenosis or coarctation if primary repair not possible

PGE in pulmonary venous obstruction (partial bypass of obstructed pulmonary venous return)

(+)

Surgical intervention in the neonate

Transcatheter intervention in the neonate

–

Perinatal management

Ductdependent

Reconstruction or replacement of homograft/conduit in homograft insufficiency or stenosis

Age >6–12 months: interventional occlusion

Early repair at 3-4 months if VSD is large; risk of pulmonary vascular obstructive disease greater in Down syndrome

Repair at the end of first year or earlier depending on communication size; otherwise risk of developing pulmonary vascular obstructive disease; in small VSD no surgical repair

Residual pulmonary venous obstruction at the anastomotic site: reoperation Proximal pulmonary vein obstruction: dilatation and stent-implantation Arrhythmias

Interventions and problems in medium- and long-term follow-up

672 Fetal Cardiology

–

Congestive heart failure

Arrhythmia, obstruction

Dilated cardiomyopathy

Cardiac tumors

Symptomatic

Anticongestive therapy Exclude coronary abnormalies as BlandWhite-Garland syndrome Exclude metabolic and inherited disease Exclude syndromes

Anticongestive therapy

–

Coronary abnormalities: reconstruction of abnormal origin of coronary artery

Myocardial biopsy

–

Surgical removal of shunt lesion

Interventional occlusion of shunts

Surgical removal in obstruction check for tuberous sclerosis in rhabdomyoma

Cardiac transplantation

PS, pulmonary stenosis; PGE, prostaglandin E1; +, ductal dependent; (+), sometimes ductal dependent; –, not ductal dependent; VSD, ventricular septal defect; WPW, Wolff Parkinson White; TA, tricuspid atresia.

–

–

Prenatal or neonatal a. Shunt closed alter birth: recovery of cardiac failure due to the (fetal extracardiac shunts dominant) right a. twin-twinventricle transfusion b. aneurysm of vein b. Shunt patent after birth: biventricular of Galen, volume load, hemangioma, pulmonary edema, teratoma; pulmonary pulmonary hypertension sequestration

Management of the neonate with congenital heart disease 673

674

Fetal Cardiology

failure. Preoperatively, restored organ function is essential for survival in the peri- and postoperative periods for surgical interventions with or without extracorporeal circulation.

high-dose prostaglandin therapy, PGE1 treatment can be used only as a short-term palliative measure. Further palliative or corrective interventions should be performed within the first 2 weeks following birth.

Opening of the ductus arteriosus with prostaglandin E

Regulation of pulmonary perfusion

The effect of PGE1 and PGE2 on ductal tissue is dependent on the applied dose, the age of the newborn, the oxygen concentration, and the arterial pH. Whilst a small dose is sufficient in neonates with a still patent ductus arteriosus, a significantly higher dose with a correspondingly greater incidence of side-effects is necessary in newborns with an obstructed or already closed ductus arteriosus (Table 44.3). Side-effects are dose-dependent. An additional oxygen supply should be avoided. Because of the high tendency of spontaneous ductus obstruction even with

Table 44.3 Relevant side-effects of prostaglandin treatment. Adapted from reference 18

Pulmonary perfusion can be influenced sensitively by regulation of pulmonary vascular resistance. High resistance reduces pulmonary perfusion, and low resistance increases pulmonary perfusion. Oxygen supply and hyperventilation (decreased CO2), NO, prostacyclin, and other vasodilators reduce pulmonary arterial resistance and are used in situations with decreased pulmonary perfusion, for example in persistent fetal circulation or pulmonary–hypertensive crisis.19 Hypercapnia, acidosis, and hypoxemia intensify pulmonary vasoconstriction (Figure 44.9).

Balance of pulmonary and systemic perfusion in single ventricle physiology In neonates with single ventricle physiology and ductdependent pulmonary or systemic flow (e.g. hypoplastic left heart syndrome or pulmonary atresia), the extent of pulmonary (Qp) or systemic (Qs) perfusion depends on the relation between pulmonary resistance and systemic resistance. In the case of increased pulmonary resistance, the pulmonary blood flow decreases (Qp lower) and the systemic blood flow increases. In single ventricle physiology, a balanced ratio of pulmonary to systemic blood flow has to be achieved in order to avoid ventricular overload and dysfunction. The ideal situation is an equalization of pulmonary and systemic blood flow with Qp = Qs. The blood oxygen concentration can therefore

Hypotension Bradycardia Apnea Hypoventilation Sepsis Enterocolitis Tissue fragility Temperature elevation Hyperexcitability Thrombocytopenia

Pulmonary vascular resistance (PVR)

Raised PVR

Acidosis, high pCO2, low pO2, high hematocrit, low lung volume, mediators

Reduced lung volume (atelectasis; diaphragmatic hernia) perfusion/ventilation-mismatch

Lowered PVR

Alkalosis, low pCO2, high pO2,NO, prostacyclin, tolazoline, vasodilators

Figure 44.9 Factors and therapeutic options increasing (↑) or decreasing (↓) pulmonary vascular resistance.

Management of the neonate with congenital heart disease

be seen as a sensitive marker for the relation of pulmonary to systemic perfusion: Qp = Qs: Sao2 75%, Qp >> Qs: Sao2 >85%, Qp << Qs: Sao2 < 70% (Figure 44.10). In a balanced situation, the Sao2 is approximately 75%, Pco2 ∼ 45 mmHg, Po2 ∼ 35 mmHg. If the pulmonary blood flow increases (Qp high), the oxygen saturation increases to the detriment of the systemic circulation (Qs reduced), and is then followed by hypotension and reduced organ perfusion. In a balanced situation, the pulmonary blood flow remains high enough to allow for sufficient oxygenation (for more details see ‘Examples for the management of hypoplastic left heart syndrome’ below and Figure 44.11).

Therapy for cardiac insufficiency Even in healthy neonates both systolic and diastolic myocardial reserves are limited. Pressure and volume loads

675

are only tolerated restrictedly, so that an additional volume load has to be avoided. The therapeutic measures are volume restriction and beta-blockade, diuretics, angiotensin-converting enzyme inhibitors, and, in critically ill children, phosphodiesterase inhibitors or catecholamines together with vasodilators.

Examples for the management of critical congenital heart defects Hypoplastic left heart syndrome – single ventricle physiology with duct-dependent systemic perfusion The systemic circulation is supplied by the right ventricle and ductus arteriosus Botalli. Spontaneous occlusion of the ductus arteriosus and pulmonary overflow result in hypoperfusion of the systemic circulation, hypotension, and shock.

Figure 44.10 Preoperative physiology in hypoplastic left heart syndrome. Schematic diagram of O2 saturations (circles) and intracavity pressures. (a) Ideal preoperative physiology (Qp = Qs). The resultant O2 saturation of 75–80% represents balanced pulmonary and systemic blood flow (Qp: Qs = 1). (b) Falling pulmonary vascular resistance (PVR) resulting in increased pulmonary perfusion (Qp), decreased systemic perfusion (Qs), and systemic hypotension (Qp >> Qs). The relation of Qp and Qs is calculated by: Qp:Qs = [So2 (aorta) − So2 (SVC)]: [So2 (pulmonary vein) − So2 (pulmonary artery)].

676

Fetal Cardiology

Pathophysiology:

Clinical signs:

Therapy:

Pulmonary vascular resistance (PVR)⇓ Pulmonary blood flow (Qp)⇑

Pulmonary vascular resistance (PVR)⇑ Pulmonary blood flow (Qp)⇓

SaO2 >85% Blood pressure (mean) <40 mmHg Oliguria/anuria

SaO2 <70% Blood pressure normal

pH⇓ PCO2⇑

pH⇑ PCO2⇓ (PO2)⇑ Inhaled nitric oxide Prostacyclin

Intensive therapy includes maintenance of patency or reopening of the ductus arteriosus, optimization of right ventricular function, and regulation of pulmonary perfusion. In children with prenatally diagnosed hypoplastic left heart syndrome, low-dose prostaglandin E1 can be used immediately postpartum, so that ventilation is not necessary. If possible, mechanical ventilation should be avoided. With the physiological reduction of pulmonary vascular resistance, the pulmonary flow (Qp) increases to the disadvantage of the systemic flow (Qs). Systemic perfusion is reduced (Qp > Qs) (Figure 44.10b). To increase the systemic blood flow, the pulmonary blood flow has to be reduced by raising the pulmonary vascular resistance, and the systemic resistance can be lowered by vasodilators. The pulmonary vascular resistance can be increased by pulmonary vasoconstriction (Figure 44.9). Pulmonary vasoconstriction can be achieved by avoidance of oxygen therapy, or, in the case of ventilation, by additional hypoventilation (Figure 44.11). The therapeutic goal is an equilibrium between pulmonary and systemic perfusion (Qp = Qs), which corresponds to an oxygen saturation of about 70–80% (Figure 44.10a). In a balanced situation, the Sao2 is approximately 75%, Pco2 ∼ 45 mmHg, Po2 ∼ 35 mmHg. Higher oxygen saturations indicate an excessive pulmonary overflow with a risk of additional volume overload of the right ventricle, and are therefore undesirable. Oxygen saturations under 70% indicate pulmonary hypoperfusion (Qp < Qs) caused by increased pulmonary vascular resistance or atelectasis. A further decisive parameter for the relation of pulmonary to systemic perfusion is the size of the atrial shunt, which should be somewhat restrictive.16 Only in rare cases with a significantly restrictive foramen ovale and severe hypoxemia is an atrial septostomy required. If hypoplastic left heart syndrome is diagnosed after ductal occlusion, the intensive care physician is often confronted with an ischemic right ventricle as the systemic ventricle and with shock-induced multiorgan failure. In comparison to children with prenatally diagnosed

Figure 44.11 Therapeutic algorithms for neonates with ductal-dependent systemic flow.

hypoplastic left heart syndrome, children diagnosed postnatally show a worse preoperative state, with a tendency toward higher postoperative mortality and worse neurological outcome.20–22 Risk factors for an unfavourable outcome are prematurity, low birth weight (< 2.5 kg), and complex lesions.23,24 In these cases, in patients awaiting transplantation or, for whatever reason, not suitable for Norwood-I procedures, ductal stenosis can be managed by interventional catheterization. More and more centers perform interventional stent implantation with or without atrial septostomy. In cases of pulmonary hyperperfusion this may be combined with surgical bilateral pulmonary banding.25

Pulmonary atresia – duct-dependent pulmonary perfusion (Figure 44.5) Pulmonary perfusion is provided by the left ventricle and aorta via the ductus arteriosus. Spontaneous ductal occlusion results in pulmonary hypoperfusion and hypoxemia. Intensive therapy includes maintenance of patency or reopening of the ductus arteriosus, optimization of left ventricular function, and regulation of pulmonary perfusion. In neonates with prenatally diagnosed pulmonary atresia, only low-dose PGE1 is necessary to keep the ductus arteriosus open. An increased pulmonary vascular resistance found in instances with hypoplastic pulmonary vascular beds can be reduced by administration of oxygen or hyperventilation. If necessary, systemic pressure can be enhanced by treatment with catecholamines. Operation (systemic-to-pulmonary shunt or right ventricular outflow repair) or interventional perforation and balloon dilatation of the pulmonary valve can be performed after stabilization. Ebstein’s disease with functional pulmonary atresia and fetuses with right ventricular myocardial disease and duct dependent pulmonary circulation (right-sided cardiomyopathy) present with cyanosis and hypotension. Enlargement

Management of the neonate with congenital heart disease

of the right ventricle and dilatation of the right atrium result in lung hypoplasia and also in left ventricular dysfunction. The therapeutic goal is to reduce pulmonary vascular resistance by adequate oxygenation and – if necessary – addition of inhaled NO, opening of the ductus arteriosus, optimizing ventricular function and to take advantage of the physiologic changes during perinatal adaptation. Following the physiologic fall of pulmonary vascular resistance during perinatal transition, right ventricular pressure and volume load are reduced and the right ventricle is capable to serve the pulmonary circulation after a few days. If antegrade flow across the pulmonary artery is established, the neonate can be gradually weaned off prostaglandins and ventilation without the need of further interventions. Transposition of the great arteries without ventricular septal defect: shunt-dependent mixing (Figure 44.7) The pulmonary and systemic circulations are connected in parallel. The O2 saturation is dependent on the magnitude of mixing between both circulations. Ductal occlusion or restrictive foramen ovale leads to severe hypoxemia. Intensive therapy includes maintenance of patency or reopening of the ductus arteriosus and creating an unrestricted interatrial shunt by balloon atrioseptostomy. In the neonate without a ventricular septal defect and with a restrictive foramen ovale, the ductus arteriosus is kept open by PGE1, and balloon atrioseptostomy is performed immediately. If good mixing can be achieved at the atrial level, further treatment with PGE1 is not necessary. In transposition of the great vessels, the left ventricle serves the pulmonary circulation via the pulmonary artery. As a result of the physiological reduction in the pulmonary vascular resistance, the left ventricular myocardial mass is subsequently reduced. Therefore, the anatomic correction has to be performed by an arterial switch in the first 2 weeks after birth, to ensure a sufficiently developed and trained left ventricle. The importance of early intervention is pointed out in the study of Bonnet et al,26 which compares the outcomes of neonates with transposition diagnosed prenatally versus postnatally. The clinical condition at arrival on the neonatal intensive care unit, including metabolic acidosis and multiorgan failure, was significantly worse in the postnatal group. Prenatal diagnosis of transposition of the great arteries reduced postoperative mortality (0/68 neonates diagnosed prenatally vs 15/250 (6%) neonates diagnosed postnatally) and morbidity.

Established interventional procedures and therapy Interventional transcatheter procedures in neonates can be used as life-saving palliative procedures as well as to supplement or even replace open surgical interventions.

677

Balloon atrioseptostomy Principle Transcutaneous balloon atrioseptostomy introduced by Rashkind and Miller in 1966 is still the standard procedure for dilatation of the interatrial shunt. The therapeutic goal is to achieve thorough mixing of blood from both circulations or atrial decompression.27,28 Indications These are transposition of the great arteries, restrictive foramen ovale in tricuspid valve atresia or double-outlet right ventricle, single ventricle with mitral stenosis, pulmonary atresia with intact ventricular septum (if a systemicto-pulmonary shunt has to be placed), and total anomalous pulmonary venous return. Methods On the intensive care unit or in the catheter laboratory, an uninflated balloon catheter is advanced via the right atrium and foramen ovale into the left atrium (Figure 44.12). For newborns younger than 3 or 4 days old, a transumbilical access is chosen; in older neonates a transvenous access is appropriate. After inflating the balloon with diluted contrast material the septum is torn by jerking back the catheter. A successful balloon atrioseptostomy should result in an increased saturation (about 10% increase) or decreased interatrial gradient to approximately 3 mmHg. Owing to increasing thickness of the atrial septum in neonates older than 1 month, a Park–Blade atrioseptostomy is required.29 Results In more than 85% of patients, balloon atrioseptostomy is successful; only in isolated cases is a reseptostomy necessary. Complications This intervention is of low risk with a mortality of 0.7%. The relevant risks are blood loss, perforation, arrhythmia, and thromboembolism with subsequent cerebral infarction.

Balloon dilatation of valves and blood vessels Principle The physical principle is the transmission of a controlled radial force on the tissue by a rigid balloon. The mechanism of vessel enlargement by balloon dilatation is by tearing the intima and media. Dissection of the intima

678

Fetal Cardiology

Figure 44.12 Balloon atrial septostomy. The balloontipped catheter is advanced through the patent foramen ovale to the left atrium, where it is inflated with diluted contrast (a). The balloon is jerked back to the right atrium to tear the septum (b). Posterior-anterior projection. RA, right atrium; LA, left atrium.

Figure 44.13 Balloon dilatation of coarctation in a neonate. Angiography of the descendent aorta before (a) and after (c) successful angioplasty (b). (a) The diameter of the stenotic segment (D2) and the descendent aorta (D1) is defined using a calibrated marker catheter. (b) The balloon catheter is advanced to the stenotic segment and expanded. After full inflation of the balloon, only a subtle waist (arrow) is seen. (c) Note the increased diameter of the circumscribed coarctation. The aortic arch remains small. Lateral projection. AAO, ascending aorta; DAO, descending aorta; arrows indicate the coarctation.

Management of the neonate with congenital heart disease

and/or media and the subsequent development of an aneurysm has to be avoided by appropriate selection of balloon diameter, inflation pressure, length, and balloon compliance.

Method First, the measurement of the annulus or vessel diameter is precisely defined during angiography by means of a calibrated marker-catheter (Figure 44.13). A balloon catheter with a specific final diameter is chosen. Via a guidewire, the uninflated dilatation balloon is advanced to the stenotic vessel or valve. Dilatation is performed by filling the balloon with a contrast–water mixture. At the narrowest position of the valve or vessel the balloon is inflated under low pressure. By the stenotic vascular segment a waist is formed, which should disappear when the balloon is maximally inflated. This procedure may need to be repeated several times to optimize dilatation. After dilatation, angiography and measurement of the pressure gradients across the dilated segment are repeated.

Risks The risks include valvular insufficiency, aneurysm, vascular dissection, or perforation. Redilatations of valves and vessels are possible. Femoral vessel access is the site of first choice for neonates, but in the first 3–4 days after birth – if not yet occluded – the umbilical vein or artery can also be used as an access point.

679

Balloon dilatation of the pulmonary valve in neonates (Figure 44.14) Indications Interventional balloon dilatation of the pulmonary valve is the treatment of first choice in neonates with critical and extreme pulmonary valve stenosis. It is used as a palliative procedure in cases of membranous pulmonary atresia with an intact ventricular septum and tetralogy of Fallot.

Method In neonates with critical pulmonary stenosis a balloon dilatation is performed under PGE1-maintained patency of the ductus arteriosus. A sequential balloon dilatation with a balloon diameter measuring 120–130% of the size of the annulus is recommended to achieve maximal and prolonged relief of pulmonary stenosis. In patients with membranous pulmonary atresia, perforation of the obstructive membrane with a wire or, in isolated cases, with a radiofrequency- or laser-assisted catheter is also possible. After successful balloon dilatation, prostaglandin infusion can be withdrawn gradually if a sufficiently contractile right ventricle allows enough antegrade flow.

Results Even in cases of critical pulmonary valve stenosis, the success rate for dilatation is 88–95%, and is defined by a

Figure 44.14 Interventional balloon dilatation of the pulmonary valve. Right ventricular angiograms before (a) and after (c) balloon pulmonary valvuloplasty in a neonate with pulmonary stenosis. Arrows indicate the site of pulmonary stenosis. (b) The waist (arrow) of the balloon catheter produced by the stenotic valve. Lateral projection. PS, pulmonary stenosis.

680

Fetal Cardiology

reduction of the gradient of ≤ 30 mmHg or morphometric increase of the tricuspidal and pulmonary valve diameter or right ventricular size.30,31 In the neonatal period the mortality rate is 3.7–8%,23,24 with total mortality in the whole pediatric population being 0.4%. In long-term follow-up, redilatations are necessary in 6%. The freedom of reintervention after 1, 2, and 8 years is 90%, 84%, and 84%, respectively. A relevant pulmonary insufficiency occurs in 9%.30 On the other hand, the mortality rate following surgery is 25%, with a reintervention rate of 25%.32 In patients with an immobile dysplastic pulmonary valve, only limited success can be achieved. In pulmonary atresia, valve perforation and dilatation are used to increase antegrade pulmonary blood flow, stimulate right ventricular growth, and allow biventricular surgical repair.33 The results of pulmonary valvuloplasty depend on right ventricular morphology and diameter of the pulmonary annulus.31 In cases of insufficient oxygenation, a Blalock–Taussig shunt and/or a right ventricular outflow tract patch has to be placed (6% in reference 30). In patients with tetralogy of Fallot, dilatation of the stenotic pulmonary artery is performed in patients in whom very early correction is not possible.34

Conclusion In preterm newborns and neonates, interventional balloon dilatation is the treatment of first choice for pulmonary valve stenosis.

Balloon aortic valvuloplasty (Figure 44.15) Indication Critical aortic valve stenosis in neonates is the indication. In contrast to pulmonary valve stenosis, dilatation of aortic valve stenosis is much more difficult in newborns, and is therefore associated with higher rates of morbidity and mortality. Neonates with critical aortic valve stenosis are severely ill, deteriorate easily, and present as emergency cases. Without intervention, most will die within weeks. As aortic stenosis is already relevant in utero, these newborns present with fibroelastosis of the left ventricle, reduced systolic and diastolic left ventricular function, and, in some cases, severe mitral insufficiency. In general, before performing cardiac catheterization, it has to be carefully assessed whether the patient would benefit

Figure 44.15 Interventional balloon dilatation of a critical aortic stenosis in a neonate. The left ventriculogram (a) shows a dilated and low contractile left ventricle and severe mitral regurgitation. The left atrium is enlarged. There is minimal antegrade flow across the aortic valve. The retrograde aortogram (b) shows the doming of the aortic valve (arrow) before dilatation (c). After angioplasty (d) the leaflets are mobilized, there is a mild aortic regurgitation. Posterior– anterior projection. Arrow indicates the site of aortic stenosis.

Management of the neonate with congenital heart disease

from a biventricular circulation.35 In patients with a very small left ventricle and a small aortic and mitral annulus, a univentricular correction by staged Norwood procedures or cardiac transplantation may be preferred.35,36

Method Critically ill children have a duct-dependent circulation (prostaglandin administration) and need inotropic support. In a transumbilical or transfemoral approach the aortic valve is crossed. Transvalvular pressure gradients in critical duct-dependent aortic stenosis may not be directly proportional to the severity of the disease, because the left ventricular function is, in most cases, extremely compromised and the cardiac output is provided by the right ventricle via the ductus. To avoid a severe and poorly tolerated aortic insufficiency, a low-profile balloon measuring up to 90% of the annulus diameter is used. The postinterventional mortality rate depends on the left ventricular function, the degree of mitral insufficiency, and the development of pulmonary hypertension. Therefore, positive inotropic support has to be used even after successful dilatation.

Results A success rate of 97–100% is faced with an early mortality rate of 10–20%, compared to a surgical mortality rate of 10–25%.37–39 The mortality rate is much higher for critical aortic valve stenosis associated with malformations such as mitral stenosis, aortic arch hypoplasia, hypoplastic aortic annulus, hypoplastic left ventricle, and duct-dependent circulation (the mortality rate of duct-dependent circulations is 38% in contrast to 5% for duct-independent circulations).39,40 Vascular complications occur, especially following access via the femoral artery. Therefore, the umbilical artery, carotid artery or the femoral vein with an antegrade approach can be used as alternative vascular access points. Reinterventions such as redilatation, surgical commissurotomy, or valve replacement are necessary in patients treated by interventional or surgical procedures.37,39,40 The rate of reintervention is 50–60% after 5 years,37,40,41 the freedom of intervention is 64% after 8 years,34 and eventfree survival is 34.2% after 10 years, and 27.4% after 15 years.34 Following balloon dilatation, 6–15% of patients develop significant aortic insufficiency.38,41

Conclusion Interventional valvuloplasty of critical aortic valve stenosis is a procedure with high morbidity and mortality. Alternative surgical procedures are associated with comparable risks. In cases with unfavorable morphology (borderline mitral valve area, small aorta, small left ventricle), interventional palliative therapy (stenting of

681

the duct with or without pulmonary banding) may be performed to achieve time before deciding for either uni- or biventricular repair. Selection of the preferred treatment depends on the severity of the condition of the patient, additional cardiovascular abnormalities, and the practical experience of the center with the individual procedures.

Balloon dilatation of aortic coarctation (Figure 44.13) Interventional balloon dilatation of native coarctation is still controversial. Surgical procedures have a slightly higher mortality, but require reintervention less frequently.42

Indications Indications for surgical or transcatheter intervention are hypertension and/or congestive heart failure. Balloon angioplasty of native aortic coarctation is performed in circumscribed short-segment stenosis. Narrowing of a long segment of the aorta or stenosis associated with a hypoplastic aortic arch usually needs surgical augmentation aortoplasty. Results Both interventional and surgical procedures have successfully eliminated stenosis in over 80% of patients (defined as a reduction in the gradient to less than 20 mmHg). Balloon dilatation is significantly less invasive than surgical procedures, but is associated with a higher recurrence rate and the risk of femoral artery injury.42 In newborns, restenosis due to residual ductal tissue occurs in 40–50%; in older children the recurrence rate is significantly lower.43 Recoarctation is associated with a younger age of the neonate, a smaller diameter of the aortic isthmus, and a small diameter of the tightest part of the aortic coarctation before and after dilatation.44 In contrast, the reintervention rate following surgical procedures is only 28%.42 There is little doubt about balloon coarctation angioplasty as the procedure of choice in the treatment of recurrent coarctation after surgical resection. Conclusion Interventional balloon dilatation of native aortic coarctation shows a high reintervention rate compared to the more invasive surgical approach. In critically ill patients, interventional balloon dilatation of coarctation with or without stenting has proved to be a very useful procedure by providing sufficient relief of the coarctation until a later operative correction under optimal conditions can be performed.

682

Fetal Cardiology

Figure 44.16 Stenting of the ductus arteriosus Botalli. (a) The pigtail catheter enters the ductus arteriosus via the pulmonary artery. The angiography of the ducts shows the duct-dependent perfusion of the aorta. (b) The unexpanded stent, mounted on a balloon catheter, is positioned in the ductus arteriosus. (c) The balloon is inflated and the stent expands. (d) After removal of the balloon catheter a repeat angiography is performed showing optimal position of the stent. Lateral projection (courtesy of D Schranz, MD PhD, University of Giessen, Germany).

Balloon dilatation of other blood vessels Native or postoperative pulmonary artery stenosis, stenotic branch pulmonary arteries, or surgical shunts such as the modified Blalock–Taussig anastomosis can also be successfully dilated in neonates. If restenosis occurs and redilatation is necessary, intravascular stents can be inserted successfully. In older children, balloon dilatation is a suitable procedure for vascular stenosis of femoral arteries, renal arteries, systemic veins, collaterals, and surgically placed shunts.

Intravascular stents Vascular stenosis can be dilated by balloon angioplasty, but, owing to elastic ‘recoil’ of the vascular wall, stenosis may recur shortly after dilatation. Intravascular stents are inserted to avoid the intrinsic ‘recoil’ of the vascular wall, so that long-term success can be achieved. Indication The placement of intravascular stents is performed in stenotic ductus arteriosus with duct-dependent defects,

stenosis of a surgically placed Blalock–Taussig shunt, and severe coarctation in ciritically ill children. Indications in older children are pulmonary artery stenosis, peripheral pulmonary artery stenosis, aortic coarctation, stenosis of surgically placed shunts and conduits (baffle after atrial switch operation), systemic veins, major aortopulmonary collateral arteries, and coronary stenosis after Kawasaki syndrome.45 Valvular stents are already in use as percutaneous valve replacements.46 Stents are also used to maintain the patency of atrial defects and fenestrations, which serve as communications between the venous and atrial systemic circulations. Procedure Stents are placed on a dilatation balloon catheter and are advanced to the stenosis in a long sheet via a guidewire (Figure 44.16). The balloon is inflated and expands the stent, which remains in position after the balloon catheter is removed. Redilatation of a stent is possible.

Ductal stenting (Figures 44.16 and 44.17) Implantation of a ductal stent is used in duct-dependent defects with prostaglandin-resistant ductal stenosis.

Management of the neonate with congenital heart disease

683

and feasible means of palliation in neonates and young infants.

Interventional occlusion of cardiac defects and connecting vessels Interventional occlusion of persistent ductus arteriosus (PDA) is discussed as follows. Indications Significant hemodynamic PDA at any time, or persistence of smaller PDA for longer than 6–12 months because of the risk of endocarditis, is an indication for interventional occlusion.

Figure 44.17 Interventional stenting of the ductus arteriosus and bilateral surgical pulmonary banding in hypoplastic left heart syndrome. Arrows indicate the narrowing of the pulmonary left (LPA) and right (RPA) arteries after bilateral surgical banding. A stent was placed in the stenotic ductus arteriosus to offer unrestricted flow from the pulmonary artery (PA) to the aorta. The bilateral pulmonary banding prevents excessive pulmonary blood flow. This approach allows bridging to a combined Norwood I and II operation at once or a cardiac transplantation (‘The Giessen Procedure’, courtesy of D Schranz, MD PhD, University of Giessen, Germany). Posterior–anterior projection.

Neonates with hypoplastic left heart syndrome awaiting cardiac transplantation, with prematurity, with low birth weight, or not suitable for Norwood-I palliation can be treated by interventional ductal stenting in association with surgical bilateral banding of the pulmonary artery.23–25 In patients with ductal restenosis the stent is redilated. In duct-dependent pulmonary circulation, stents can be placed as an alternative for a surgical aortopulmonary shunt, but the implantation is technically demanding owing to the more tortuous ductal anatomy.48 Risks and complications Critical limitations are in-stent stenosis and relative stenosis during childhood growth. Neointimal proliferation and thrombosis lead to stent stenosis or occlusion. Stents with intimal proliferation and relative stenosis due to childhood growth can be expanded by redilatation in the short to medium term; long-term results are not yet available. With improved features of stents, maintaining ductal patency by percutaneous stent implantation offers a safe

Procedure Angiography of the ductus is performed to evaluate the configuration and size of the PDA (Figure 44.18). The hemodynamic and angiographic data are reviewed, and an optimal device is selected. Ductal access is made via the aorta or pulmonary artery. If angiography does not show an optimal position, the coil can be retrieved and repositioned. If the coil position is acceptable and the ductus is closed, the coil can be released. As different coil sizes are available, an optimal coil can be selected for each ductal diameter and form. From the multitude of available devices, retrievable products (that is controlled-release and repositionable) have prevailed. If repeat angiography after implantation reveals a significant residual shunt, an additional device is implanted in the same session. After occlusion and complete endothelialization of the implant 6 months after implantation, endocarditis prophylaxis is no longer required.

Results The occlusion rate of all devices is 90–95%. The risks are embolization and the presence of a residual shunt.49–52 In cases where the residual shunt persists for more than 6 months, a coil-in-device implantation is performed. Implantations can even be performed in neonates, but as the catheters are relatively large, the body weight should be at least 3.5 kg. Interventional occlusions of other blood vessels (arteriovenous malformations, such as arteriovenous vascular connections of the brain, hepatic hemangioma, pulmonary sequester or aortopulmonary collaterals) become symptomatic in the neonatal period and lead to cardiac insufficiency. They can also be occluded successfully with coils, balloons, or vascular plugs.53 Further indications for interventional occlusion are surgically placed aortopulmonary shunts, fenestrations, venovenous collaterals, and coronary fistulas.54

684

Fetal Cardiology

Figure 44.18 Interventional occlusion of a persistent ductus arteriosus (PDA) using a coil. (a) Angiogram using a marker-catheter demonstrating the PDA. (b) PDA measurements. (c) Implantation of the coil: most coil loops are placed into the PDA, only the proximal loop is positioned in the pulmonary artery end of the PDA. (d) After control of position and occlusiveness of the coil, the coil is released and a repeat aortogram performed. Note the lack of residual shunt. Lateral projection (courtesy of DA Redel, MD PhD, Bonn, Germany).

Atrial septal defects (Figure 44.19) Indications Atrial septal defect (ASD II) without anomalous pulmonary venous return is an indication, as is surgically performed fenestration, for example fenestration of a Fontan-tunnel. For occlusion of atrial septal defects several devices are available. Until now, only occlusion procedures that can be used in children older than 6 months have been developed, as isolated atrial septal defects without anomalous origin of pulmonary veins rarely become symptomatic in early infancy.

Procedure Using transesophageal echo, the size of the defect, the margins of the residual septum (at least 3–4 mm), and the atrial size are measured, because the diameter of the device should measure 150–200% of the size of the unstretched defect. Anomalous pulmonary venous return is also excluded. After balloon sizing and measurement of the stretched diameter of the ASD II, the device is implanted. Usually, double umbrella devices or stent-like occluders are utilized.

Results Dependent on proper patient selection, implantations are possible in 95%. The occlusion rate is 90–98% with a stent such as the Amplatzer device; in children under 2 years of age the occlusion rate is 83%.55–58 Complications are embolization, thrombosis, and arrhythmia.

Ventricular septal defects The percutaneous interventional occlusion of a ventricular septal defect is still limited to selected cases.59 The procedure enables the occlusion of a ventricular septal defect without the need for a cardiopulmonary bypass. Indications These include multiple ventricular septal defects, and perimembranous or muscular ventricular septal defects.55,59

Myocardial biopsy To diagnose dilative cardiomyopathy, myocardial biopsy is also performed in neonates. Usually a right ventricular

Management of the neonate with congenital heart disease

685

Figure 44.19 Interventional occlusion of an atrial septal defect in a 1-year-old infant. (a) Transesophageal echocardiography: large atrial septal defect of secundum type with right-to-left shunt. (b) Cineangiographic frame of the Amplatzer septal-occluder after implantation. (c) Transthoracic echocardiography reveals good device position (courtesy of J Breuer, MD PhD, University of Bonn, Germany). ASD, atrial septal defect.

Drug or interventional treatment of arrhythmia

Figure 44.20 Standard catheter positions for electrophysiologic study. The mapping catheters can be used to pace or record from various cardiac sites. The catheter in the right atrium is in close proximity to the sinoatrial node (HRA, high right atrium). The His bundle electrogram (HBE) is obtained by positioning the catheter just across the tricuspid valve. Further standard pacing and recording sites are the right ventricular apex and coronary sinus. Posterior–anterior projection.

approach is used. For diagnosis of metabolic disease, a metabolic workup has to be completed, occasionally including fibroblast cultures and skeletal muscle biopsies.

Prenatal drug therapy of fetal tachy- and bradycardias can sometimes be extremely difficult, and may therefore lead to severe cardiac insufficiency and hydrops fetalis. In contrast, tachycardias in newborns usually have a good prognosis. Although there is approximately 50% incidence of recurrence of supraventricular tachycardia and atrial flutter in the neonatal period,60 in the first year of life spontaneous remission can be expected in 30–90%, in particular for paroxysmal tachycardias.61,62 Postnatally, conventional and transesophageal electrocardiogram (ECG) makes discovery of the precise diagnosis easier. By transesophageal overdrive pacing, and, as a second choice, with lowenergy cardioversion, even refractory atrial reentry tachycardias such as atrial flutter can be interrupted. Accessory pathways such as are seen in Wolff–Parkinson– White syndrome – which constitute over 70% of all supraventricular neonatal tachycardias – or atrioventricular node-reentry tachycardia (10–12% of all supraventricular neonatal tachycardias) are the most frequent tachycardias in the first year of life.63 These reentry tachycardias use the atrioventricular node as a pathway and can be interrupted by rapid intravenous administration of adenosine, which blocks the atrioventricular node for a few seconds.64 Adenosine allows the differentiation between atrial reentry tachycardia, atrial flutter, and ectopic tachycardia as well as ventricular tachycardia. In ventricular tachycardia no effect is seen, and in atrial flutter or atrial ectopic tachycardia the adenosine-induced atrioventricular block allows detection of the atrial focus. Owing to the limited systolic and diastolic myocardial reserve in the newborn, high-frequency (usually > 230 beats/minute) and long-lasting tachycardias are poorly

686

Fetal Cardiology

Figure 44.21 Successful radiofrequency ablation of atrial flutter. Termination of atrial flutter recorded on surface (I-aVF) and intracardiac electrograms.

tolerated. Repeat occurrences of supraventricular tachycardias make drug treatment in neonates difficult, and sometimes require multiple drug combinations. Bearing in mind the spontaneous remission of paroxysmal supraventricular tachycardia, drug treatment can be paused at the end of the first year of life. The data regarding freedom from recurrence of untreated supraventricular tachycardia are limited, and this may be in the range of 25–60%. Above all, ectopic atrial and multifocal atrial tachycardias are difficult to manage and may lead to secondary congestive cardiomyopathy. In the presence of persistent long-lasting tachycardias, an interventional catheter-ablation procedure using a radiofrequency catheter provides conclusive therapy, particularly in cases with accessory pathways (for example Wolff–Parkinson–White syndrome). Radiofrequency ablation can even be applied in newborns with refractory tachycardia.65,66 First of all, the focus or ectopic pathway is localized by means of a mapping catheter (Figure 44.20) and induction of the tachyarrhythmia. The lesion is then thermocoagulated (Figure 44.21). The probability of spontaneous pathway degeneration as well as the limitations of vascular access and possible interventional complications in neonates have to be carefully weighed in comparison to multiple drug treatment. However, after the first year of life, radiofrequency ablation is a sensible alternative to several years of drug treatment. The mortality rate of pediatric radiofrequency ablation is 0.22%, with an efficacy rate greater than 85%.66 Absence of recurrence is dependent on the underlying mechanism. Serious complications are complete heart block, cardiac perforation, and cerebrovascular accident. Ventricular tachycardia, if not complicated by a structural cardiac defect or by the existence of long-QT syndrome, is a rarity in neonates and has a high tendency of spontaneous remission of 89%.67 Long-QT syndrome (LQTS) is a familial disease characterized by prolonged and abnormal repolarization,

associated with a high risk of ventricular arrhythmias and sudden death.68 Hallmarks are the prolongation of corrected QT (QTc) intervals and the occurrence of torsades de points. A widely used scoring system allows calculation of the probability of LQTc if suspicion is given.68 Serial ECGs should be obtained, because the QTc value could vary. Furthermore, screening ECGs from other family members will be needed. Diagnosis, risk assessment, and management are increasingly being guided by genespecific diagnoses.69 Beta-blockers, cardiac pacing as adjuvant therapy in bradycardia, and implantable cardioverter defibrillator improves survival; in LQTS III patients, Na+ channel blockers are suggested. Bradycardic rhythm disorders, especially congenital complete atrioventricular block, may require a temporary pacemaker, where cardiac insufficiency or symptoms are present. An intravenous pacemaker with a small lumen can be placed into the right ventricle in the intensive care unit and used as a temporary measure until a permanent pacemaker can be implanted. Surgical implantation of a pacemaker is possible even in the smallest of children. Owing to the potential risk of late venous thrombosis, epimyocardial electrodes with VVI(R) units are implanted (VVI, ventricle paced, ventricle sensed, ventricle inhibited; R, rate responsive). The use of transvenous dual-chamber pacemakers in neonates is controversial because of the incidence of vascular complications in the long term.

Conclusion As a result of the continuous improvement of prenatal diagnosis, cardiac defects and rhythm disorders can be detected earlier and with a higher sensitivity. The cooperation of obstetricians working in the field of prenatal diagnosis and pediatric cardiologists is extremely important

Management of the neonate with congenital heart disease

for the care of the fetus suffering from cardiac disease. Counseling and support of the parents after diagnosis, planning and monitoring of intrauterine treatment as well as planning of delivery is ensured by gynecologists in cooperation with pediatric cardiologists, neonatologists and cardiac surgeons. Postpartum, further treatment in specialized centers with the possibilities of interventional and surgical therapy enables interdisciplinary care of the newborn with congenital heart disease.

References 1. Rudolph AM. Congenital Diseases of the Heart. Armonk, NY: Futura, 2000; 3–43: 121–52. 2. Rudolph AM. The fetal circulation and its adjustments after birth. In: Moller JH, Hoffmann JIE, eds. Pediatric Cardiovascular Medicine, 1st edn. Philadelphia: Churchill Livingstone, 2000; 4: 60–4. 3. Lim MK, Hanretty K, Houston AB et al. Intermittent ductal patency in healthy newborn infants: demonstration by colour flow mapping. Arch Dis Child 1992; 67: 1217–18. 4. Shiraishi H, Yanagisawa M. Bidirectional flow through the ductus arteriosus in normal newborns: evaluation by Doppler color flow imaging. Pediatr Cardiol 1991; 12: 201–5. 5. Clyman RI. Patent ductus arteriosus in the premature infant. In: Taeusch HW, Ballard RA, eds. Schaffer and Avery’s Diseases of the Newborn, 8th edn. Philadelphia: WB Saunders, 2004; 816–26. 6. Fugelseth D, Lindemann R, Liestol K et al. Ultrasonographic study of ductus venosus in healthy neonates. Arch Dis Child Fetal Neonatal Ed. 1997; 77: 131–4. 7. Kozak-Barany A, Jokinen E, Rantonen T et al. Efficiency of left ventricular diastolic function increases in healthy fullterm infants during the first months of life. A prospective follow-up study. Early Hum Dev 2000; 57: 49–59. 8. Rudolph AM. Myocardial growth before and after birth. Clinical implications. Acta Paediatr 2000; 89: 129–33. 9. Levy M, Maurey C, Dinh-Xuan AT et al. Developmental expression of vasoactive and growth factors in human lung. Role in pulmonary vascular resistance adaptation at birth. Pediatr Res 2005; 57: 21R–5R. 10. Rudolph AM. The changes in the circulation after birth. Their importance in congenital heart disease. Circulation 1970; 41: 343–59. 11. Fineman JR, Heymann MA, Morin FC. Fetal and postnatal circulations: pulmonary and persistent pulmonary hypertension of the newborn. In: Allen HD, Adams FH, Moss AJ, eds. Moss and Adams’ Heart Disease in Infants, Children, and Adolescents, 6th edn. Philadelphia: Lippincott Williams & Wilkins, 2001: 41–52. 12. Morin FC III, Egan E. Pulmonary hemodynamics in fetal lambs during the development at normal and increased oxygen tension. J Appl Physiol 1992; 73: 213–18. 13. Soifer SJ, Morin FC III, Kaslow DC, Heymann MA. The developmental effects of prostaglandin D2 on the pulmonary and systemic circulations in the newborn lamb. J Dev Physiol 1983; 5: 237–50.

687

14. Martin LD, Wetzel RC. Regulation of pulmonary vascular tone and blood flow. In: Nichols DG, Cameron DE, Greeley WJ et al, eds. Critical Heart Disease in Infants and Children, 1st edn. St Louis: Mosby, 1995: 75–122. 15. Steinhorn RH, Fineman JR. The pathophysiology of pulmonary hypertension in congenital heart disease. Artif Organs 1999; 22: 970–4. 16. Nicolson SC, Steven JM, Jobes DR. Hypoplastic left heart syndrome. In: Nichols DG, Cameron DE, Greeley WJ et al, eds. Critical Heart Disease in Infants and Children, 1st edn. St Louis: Mosby, 1995: 863–84. 17. Mullins CE, Mayer DC, eds. Congenital Heart Disease. A Diagrammatic Atlas. New York: Alan R Liss, 1988. 18. Lucron H, Chipaux M, Bosser G et al. Complications of prostaglandin E1 treatment of congenital heart disease in paediatric medical intensive care. Arch Mal Coeur Vaiss 2005; 98: 524–30. 19. Wernovsky G, Chang AC, Wessel DL. Intensive care. In: Allen HD, Adams FH, Moss AJ, eds. Moss and Adams’ Heart Disease in Infants, Children, and Adolescents, 6th edn. Philadelphia: Lippincott Williams & Wilkins, 2001: 350–81. 20. Mahle WT, Clancy RR, McGaurn SP et al. Impact of prenatal diagnosis on survival with early neurologic morbidity in neonates with the hypoplastic left heart syndrome. Pediatrics 2001; 107: 1277–82. 21. Tworetzky W, McElhinney DB, Reddy VM et al. Improved surgical outcome after fetal diagnosis of hypoplastic left heart syndrome. Circulation 2001; 103: 1269–73. 22. Brown KL, Ridout DA, Hoskote A et al. Delayed diagnosis of congenital heart disease worsens preoperative condition and outcome of surgery in neonates. Heart 2006; 92: 1298–302. 23. Artrip JH, Campbell DN, Ivy DD et al. Birth weight and complexity are significant factors for the management of hypoplastic left heart syndrome. Ann Thorac Surg 2006; 82: 1252–7. 24. Lim DS, Peeler BB, Matherne GP et al. Risk-stratified approach to hybrid transcatheter-surgical palliation of hypoplastic left heart syndrome. Pediatr Cardiol 2006; 27: 91–5. 25. Akintuerk H, Michel-Behnke I, Valeske K et al. Stenting of the arterial duct and banding of the pulmonary arteries: basis for combined Norwood stage I and II repair in hypoplastic left heart. Circulation 2002; 105: 1099–103. 26. Bonnet D, Coltri A, Butera G et al. Detection of transposition of the great arteries in fetuses reduces neonatal morbidity and mortality. Circulation 1999; 99: 916–18. 27. Rashkind WJ, Miller WW. Creation of an atrial septal defect without thoracotomy: a palliative approach to complete transposition of the great arteries. J Am Med Assoc 1966; 1996: 991–2. 28. O’Laughlin MP, Mullins CE. Therapeutic cardiac catheterization. In: Garson AJr, Bricker JT, Fisher DJ, Neish SR, eds. The Science and Practice of Pediatric Cardiology, 2nd edn. Baltimore: Williams and Wilkins, 1997: 2415–45. 29. Park SC, Zuberbuhler JR, Neches WH et al. A new atrial septostomy technique. Cathet Cardiovasc Diagn 1975; 1: 195–201. 30. Tabatabaei H, Boutin C, Nykanen DG et al. Morphologic and hemodynamic consequences after percutaneous balloon

688

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43. 44.

45.

46.

Fetal Cardiology

valvotomy for neonatal pulmonary stenosis: medium-term follow up. J Am Coll Cardiol 1996; 27: 473–8. Gournay V, Piechaud JF, Delogu A et al. Balloon valvotomy for critical stenosis or atresia of pulmonary valve in newborns. J Am Coll Cardiol 1995; 26: 1725–31. Hanley FL, Sade RM, Freedom RM et al. Outcomes in critically ill neonates with pulmonary stenosis and intact ventricular septum: a multiinstitutional study. Congenital Heart Surgeons Society. J Am Coll Cardiol 1993; 22: 183–92. Humpl T, Soderberg B, McCrindle BW et al. Percutaneous balloon valvotomy in pulmonary atresia with intact ventricular septum: impact on patient care. Circulation 2003; 108: 826–32. Arnold, R. Is balloon dilation a reasonable option for the palliation of Fallot’s tetralogy. Cardiol Young 1999; 9: 4–5. Colan SD, McElhinney DB, Crawford EC et al. Validation and re-evaluation of a discriminant model predicting anatomic suitability for biventricular repair in neonates with aortic stenosis. J Am Coll Cardiol 2006; 47: 1858–65. Lofland GK, McCrindle BW, Williams WG et al. Critical aortic stenosis in the neonate: a multi-institutional study of management, outcomes, and risk factors. Congenital Heart Surgeons Society. J Thorac Cardiovasc Surg 2001; 121: 10–27. McCrindle BW, Blackstone EH, Williams WG et al. Are outcomes of surgical versus transcatheter balloon valvotomy equivalent in neonatal critical aortic stenosis? Circulation 2001; 104 (Suppl 1): I152–8. Egito ES, Moore P, O’Sullivan J et al. Transvascular balloon dilation for neonatal critical aortic stenosis: early and midterm results. J Am Coll Cardiol 1997; 29: 442–7. Hawkins JA, Minich LL, Tani LY et al. Late results and reintervention after aortic valvotomy for critical aortic stenosis in neonates and infants. Ann Thorac Surg 1998; 65: 1758–62. Gaynor JW, Bull C, Sullivan ID et al. Late outcome of survivors of intervention for neonatal aortic valve stenosis. Ann Thorac Surg 1995; 60: 122–5. McElhinney DB, Lock JE, Keane JF et al. Left heart growth, function, and reintervention after balloon aortic valvuloplasty for neonatal aortic stenosis. Circulation 2005; 111: 451–8. Ovaert C, Benson LN, Nykanen D, Freedom RM. Transcatheter treatment of coarctation of the aorta. Pediatr Cardiol 1998; 19: 27–44. Rao PS. Interventional pediatric cardiology. State of the art and future directions. Pediatr Cardiol 1998; 19: 107–24. Rao PS, Chopra PS, Kocscik R et al. Surgical versus balloon therapy for aortic coarctation in infants < 3 months old. J Am Coll Cardiol 1994; 23: 1470–83. Okubo M, Benson LN. Intravascular and intracardiac stents used in congenital heart disease. Curr Opin Cardiol 2001; 26: 84–91. Coats L, Tsang V, Khambadkone S et al. The potential impact of percutaneous pulmonary valve stent implantation on right ventricular outflow tract re-intervention. Eur J Cardiothorac Surg 2005; 27: 536–43.

47. Shaffer KM, Mullins CE, Grifka RG et al. Intravascular stents in congenital heart disease: short- and long-term results from a large single-center experience. J Am Coll Cardiol 1998; 31: 661–7. 48. Alwi M, Choo KK, Latiff HA et al. Initial results and medium-term follow-up of stent implantation of patent ductus arteriosus in duct-dependent pulmonary circulation. J Am Coll Cardiol 2004; 44: 438–45. 49. Moore JW, Schneider DJ, Dimeglio D. The Duct-Occlud device: design, clinical results, and future directions. J Interv Cardiol 2001; 14: 231–8. 50. Ebeid MR, Masura J, Hijazi ZM. Early experience with the Amplatzer ductal occluder for closure of the persistently patent ductus arteriosus. J Intervent Cardiol 2001; 14: 33–6. 51. Hofbeck M, Bartolomaeus G, Buheitel G et al. Safety and efficacy of interventional occlusion of patent ductus arteriosus with detachable coils: a multicentre experience. Eur J Pediatr 2000; 159: 331–7. 52. Breuer J. Interventioneller Verschluss des persistierenden Ductus botalli und verschiedener Gefässmalformationen. Monatsschr Kinderheilkd 2001; 149: 1000–10. 53. Rothman A. Pediatric cardiovascular embolization therapy. Pediatr Cardiol 1998; 19: 74–84. 54. Sieverding L, Breuer J. Interventional occlusion of congenital vascular malformations with the detachable Cook coil system. J Interv Cardiol 2001; 14: 313–18. 55. Hein R, Buscheck F, Fischer E et al. Atrial and ventricular septal defects can safely be closed by percutaneous intervention. J Interv Cardiol 2005; 18: 515–22. 56. Fischer G, Stieh J, Uebing A et al. Experience with transcatheter closure of secundum atrial septal defects using the Amplatzer septal occluder: a single centre study in 236 consecutive patients. Heart 2003; 89: 199–204. 57. Vogel M, Berger F, Dahnert I et al. Treatment of atrial septal defects in symptomatic children aged less than 2 years of age using the Amplatzer septal occluder. Cardiol Young 2000; 10: 534–7. 58. Formigari R, Di Donato RM, Mazzera E et al. Minimally invasive or interventional repair of atrial septal defects in children: experience in 171 cases and comparison with conventional strategies. J Am Coll Cardiol 2001; 37: 1707–12. 59. Michel-Behnke I, Le TP, Waldecker B et al. Percutaneous closure of congenital and acquired ventricular septal defects – considerations on selection of the occlusion device. J Interv Cardiol 2005; 18: 89–99. 60. van Engelen AD, Weijtens O, Brenner JI et al. Management outcome and follow-up of fetal tachycardia. J Am Coll Cardiol 1994; 24: 1371–5. 61. Deal BJ, Keane JF, Gillette PC, Garson AJr. Wolff-ParkinsonWhite syndrome and supraventricular tachycardia during infancy: management and follow-up. J Am Coll Cardiol 1985; 5: 130–5. 62. Perry JC, Garson A Jr. Supraventricular tachycardia due to Wolff–Parkinson–White syndrome in children: early disappearance and late recurrence. J Am Coll Cardiol 1990; 16: 1215–20. 63. Weindling SN, Saul JP, Walsh EP. Efficacy and risks of medical therapy for supraventricular tachycardia in neonates and infants. Am Heart J 1996; 131: 66–72.

Management of the neonate with congenital heart disease

64. Paul T, Pfammatter JP. Adenosine: an effective and safe antiarrhythmic drug in pediatrics. Pediatr Cardiol 1997; 18: 118–26. 65. Blaufox AD, Felix GL, Saul JP; Pediatric Catheter Ablation Registry. Radiofrequency catheter ablation in infants
689

67. Pfammater JP, Paul T. Idiopathic ventricular tachycardia in infancy and childhood: a multicenter study on clinical profile and outcome. Working group on disrhythmias and electrophysiology of the Association for European Pediatric Cardiology. J Am Coll Cardiol 1999; 33: 2067–72. 68. Chiang CE. Congenital and acquired long QT syndrome. Current concepts and management. Cardiol Rev 2004; 12: 222–34. 69. Collins KK, Van Hare GF. Advances in congenital long QT syndrome. Curr Opin Pediatr 2006; 18: 497–502.

45 Infants with congenital heart disease in the first year of life Andrew J Parry and Frank L Hanley

Introduction For every 10 000 live births, 23 children have congenital heart disease which requires surgery before the age of 1 year. These children fall into two categories: those who require surgery within the first year of life, and those who, philosophically, we consider would do better if complete correction were performed during this time (Table 45.1). For those who require surgery during the first year of life, a second decision must be made as to what treatment is most appropriate: palliative or corrective. For patients with ‘single ventricle’ anatomy there is little choice; in cyanotic neonates and those who are duct-dependent, pulmonary blood flow must be secured by means of a systemic-to-pulmonary arterial shunt, usually a modified Blalock–Taussig shunt, whereas in those neonates with an unprotected pulmonary vascular bed a pulmonary artery band is required. The group of patients with two-ventricle anatomy, however, may be approached in two different ways: initial palliation or primary complete repair. Studies have shown that establishing normal intracardiac hemodynamics as early as can be safely achieved maximizes the opportunity for optimal cardiac and pulmonary development and results in the best outcome for all organ systems. We must therefore define what the phrases ‘as can be safely achieved’ and ‘appropriate lesions’ mean. Miniaturization of extracorporeal perfusion equipment and improvements in optics over the past decade allow us now to operate safely on children of 1500 g or more, although smaller children have been treated successfully in specialized units. With these very small infants the greatest risk is often during the postoperative period when the insult to the child of a period of extracorporeal circulation, and problems primarily related to physical size, may be overwhelming. Physical size is therefore a major determinant of what can be safely achieved. In some units great exper-

tise has been developed in caring for these very small infants, which has led to a significant reduction in perioperative mortality. However, not all units have adopted this course, and therefore perform initial palliation. In these units the definition of ‘as can be safely achieved’ means correcting children significantly later. Taking this to an extreme, there is evidence that it is possible to reduce serious cardiac abnormalities postnatally by intervening in the fetus. Secondary cardiac development is dependent on hemodynamic forces rather than on genetic programming, and what may initially be a relatively simple lesion, such as valvar stenosis, may cause a major secondary problem, ventricular hypoplasia, which requires a single-ventricle circulation. Intervening early in fetal life to normalize the hemodynamics may stimulate ventricular growth to the extent that a single-ventricle circulation is avoided. This possibility is considered later. The other key consideration is what constitutes an ‘appropriate lesion’? There is a small group of patients which we believe are better managed by delaying complete correction until later, either because no benefit can be demonstrated for operating during the first year of life (such as most children with an atrial septal defect) or because a complete repair in a small individual is more likely to cause significant long-term problems than if the child were initially palliated (such as a child with transposition of the great arteries and pulmonary atresia). While opinions vary as to which operations should be included within this group, there is general acceptance that, despite the advances in pediatric cardiac surgical techniques, there remains a role for palliation even in the modern age. Currently, for many lesions, treatment pathways are well worked out and approaches are similar between various units. This discussion will therefore focus primarily on the current areas of contention, although reference will be made to the treatment of the more well established lesions.

692

Fetal Cardiology

Table 45.1 Congenital cardiac lesions appropriate for intervention during the first year of life Essential to intervene Inadequate systemic blood flow mitral stenosis critical aortic stenosis interrupted aortic arch coarctation of the aorta hypoplastic left heart syndrome Inadequate pulmonary blood flow pulmonary atresia with ventricular septal defect pulmonary atresia with intact ventricular septum tricuspid atresia severe tetralogy of Fallot Excessive pulmonary blood flow arterial level shunts: truncus arteriosus aortopulmonary window persistent ductus arteriosus ventricular level shunts atrial level shunts Inadequate intracardiac mixing transposition of the great arteries Pulmonary venous obstruction Vascular abnormalities vascular rings anomalous origin of the coronary arteries Preferable to intervene Tetralogy of Fallot

Essential lesions Inadequate systemic blood flow

have variously been shown to be low ejection fraction, the presence of endocardial fibroelastosis, high left ventricular end-diastolic pressure, elevated mean pulmonary artery pressure, and low left ventricular volume.1 However, the most widely adopted approach (the ‘Rhodes’ score, DS) weights four factors to determine the adequacy of left ventricular structures (body surface area (BSA), indexed aortic root dimension (iROOT), ratio of long axis of the left ventricle to long axis dimension of the heart (LAR), and the indexed mitral valve area (iMVA)) by the equation: DS = 14.0(BSA) + 0.943(iROOT) + 4.78(LAR) + 0.157(iMVA) − 12.3 This has been found to be predictive of death after a twoventricle repair with confidence limits of 88–91% if the score is less than −0.35.2 In patients with aortic arch obstruction, however, this assessment may not be valid.3,4 In this group, relieving the outflow tract obstruction and providing adequate preload by closing intracardiac shunts has been found to allow adequate growth of the left ventricle within a short period. Failure in these series was always due to the presence of mitral stenosis. In some children with critical aortic stenosis who, by other criteria, should ideally undergo a single-ventricle repair, aortic arch reconstruction as for the Norwood stage I repair, a Ross–Konno left ventricular outflow tract (LVOT) reconstruction, and an aggressive, total left ventricular resection of endomyocardial fibroelastosis may permit a two-ventricle repair. Experience with this approach is limited, but it may prove appropriate in the management of children with borderline left ventricles but severe endocardial fibroelastosis.5

Aortic stenosis Aortic stenosis is part of a spectrum which, in its most severe form, results in complete atresia of the outflow tract. Following resuscitation there are two key issues: first, is the child best served by a single- or two-ventricle repair? And second, which form of two-ventricle intervention is most appropriate? Resuscitation is similar regardless of the ultimate adequacy of the left ventricle. The child is treated like a child with single-ventricle anatomy by maintaining the ductus arteriosus; coronary perfusion is achieved by retrograde flow down the ascending aorta if stenosis is severe. So long as the atrial septal defect is unrestrictive, this will allow enough time for more definitive treatment to be planned, though this should not be delayed, as left ventricular function may continue to deteriorate during this time. Single- or two-ventricle palliation? Predicting those who will tolerate a two-ventricle repair is of paramount importance. Risk factors for poor outcome

What form of two-ventricle palliation? When the left ventricle is considered to be of adequate size to support a two-ventricle circulation, the type of palliation needs to be decided upon. Nowadays this is either by balloon valvuloplasty or open surgical valvotomy. Which approach is adopted primarily depends on unit philosophy, but comparison of outcome is difficult, as aortic stenosis is frequently associated with other cardiac lesions (such as Shone complex or mitral stenosis). The primary success rate of balloon valvuloplasty is 97–100%6,7 with a 6–21% early mortality6,8 usually in children with other severe abnormalities or small, nonapex-forming left ventricles, and 5–6% develop early, significant aortic regurgitation. Late mortality (up to 6 years) is 0–4%, and 9–20% require reintervention (repeat balloon valvuloplasty in two-thirds and surgery in one-third), giving an overall freedom from reintervention rate at 3 years of 67%; children with a duct-dependent circulation have a higher mortality (38% as opposed to 5% for non-duct-dependent children).

Congenital heart disease in the first year of life

For surgical aortic commissurotomy, early mortality for infants is 0–18%9,10 with 10-year survival of 73–100%.9,11 Freedom from reintervention is 55% at 10 years, with 21% requiring aortic valve replacement. As these results are very similar it depends on local bias as to which palliation is adopted. However, many of these patients return for further intervention, and the question arises as to whether there is not a more definitive treatment that can be offered. While repeat balloon valvuloplasty or surgical valvotomy is possible, the Ross–Konno aortoventriculoplasty may be a more definitive treatment. Operative risk is low,1,12 and it allows excellent relief of left ventricular outflow tract obstruction. While there will be the inevitable requirement for pulmonary homograft replacement with growth, encouraging early experience with this procedure obliges us to consider its adoption when primary palliation fails.

Hypoplastic left heart syndrome The management of hypoplastic left heart syndrome (HLHS) starts prenatally, as awareness of the diagnosis allows the child to be safely stabilized at birth; it has been shown that prenatal diagnosis is associated with an improved stage I survival. Except when there is pulmonary venous obstruction (including restriction of the interatrial communication) when surgery should be immediate, time should be allowed for the pulmonary vascular resistance to fall somewhat before surgery is undertaken. Surgical palliation is either by the ‘Norwood’ route or primary cardiac transplantation. The classic Norwood procedure13 has recently been modified so that pulmonary blood flow is provided by a right ventricular to pulmonary artery conduit (Sano shunt14) rather than by the traditional modified Blalock–Taussig shunt, though the overall benefits of this are disputed. Comparing conventional surgery to primary cardiac transplantation, data show that similar results can be achieved with either strategy in units with adequate activity.15 However, procurement of organs for transplantation remains the limiting factor. Palliation of children with HLHS using interventional catheter techniques has been attempted, either to stabilize patients awaiting primary transplantation, or to delay the need for the stage I Norwood procedure until the child is more robust.16 This involves stenting the ductus arteriosus, banding the branch pulmonary arteries, and achieving an unrestrictive interatrial communication. Interest in this approach is being revived, especially for low birth weight children.

Mitral stenosis At the severe end of the spectrum, mitral stenosis is managed as for HLHS. In milder forms, when assessment concludes that mitral valve size is adequate for a two-ventricle

693

repair, the options are balloon valvuloplasty or surgical repair. Which treatment is better is uncertain, as balloon valvuloplasty is infrequently used; in the only series comparing these two approaches, 2-year survival was similar for the two groups (80 vs 85%, respectively) although the surgical patients were a worse group.17 Surgery carries an operative risk of between 0 and 30%,18,19 although mitral stenosis is rarely an isolated lesion and the other abnormalities are optimally addressed at the same time. Longterm survival and freedom from reintervention at 2–7 years are 40–90%18,19 and 60–71% for balloon valvuoplasty and surgery, respectively, with 40–54% requiring mitral valve replacement.20,21

Interrupted aortic arch Although interrupted aortic arch is rarely a surgical emergency, there should not be undue delay in the definitive treatment of this lesion as lower body perfusion is precariously dependent on ductal patency. Therefore, after resuscitation using prostaglandin to maintain the arterial duct, surgical repair should be undertaken. Historically a two-stage approach (initial aortic arch reconstruction with pulmonary artery banding and later ventricular septal defect closure) was used, though now the consensus tends towards a single-stage approach. For a single-stage approach, early mortality is 0–12% and late mortality 20–25%22,23 even in small, premature infants, compared to 15–37% and 4–26%,22,24 respectively, for the two-stage approach, though it should be emphasized that these results pertain to different surgical eras. Despite good early results, however, freedom from reintervention at 5 years is only 55–62% predominantly for arch reobstruction (treated by surgery or balloon aortoplasty) and one-third for LVOT obstruction. When left ventricular outflow tract obstruction is also noted at presentation, the surgical risk is significantly higher, with 42% early and 50% late mortality.22 There is, however, little agreement on what constitutes an adequate left ventricular outflow tract, with indexed cross-sectional area and absolute aortic annulus size being proposed. Current surgical practice is to repair the arch using an end-to-side anastomosis having excised all ductal tissue, though an augmented subclavian flap may be useful particularly in complex interrupted aortic arch (such as with truncus arteriosus).25 The placement of foreign materials should be avoided. Associated subaortic narrowing should be addressed at the first operation and the ventricular septal defect closed. When LVOT obstruction is severe but left ventricular size is adequate, repair can be performed with a Damus–Kaye–Stansel pulmonary artery/aortic anastomosis with closure of the ventricular septal defect to the pulmonary artery. The repair is completed using a conduit between the right ventricle and pulmonary artery.

694

Fetal Cardiology

Coarctation of the aorta (infantile type) Infantile aortic coarctation is associated with persistence of the ductus arteriosus, aortic arch hypoplasia, and the presence of other intracardiac defects. If unsuspected, the child may present in extremis with low cardiac output and severe metabolic acidosis. Stabilization can usually be achieved in neonates by starting an infusion of prostaglandin E1 (PGE1), which may both reopen the ductus arteriosus and relax the ‘ductal sling’ of contractile cells that strangle the aortic isthmus. While this temporizing measure usually allows recovery from the effects of hypoperfusion, this is an unstable circulation and any benefit derived will be short-lived; definitive treatment should therefore not be imprudently delayed. The optimal definitive management of aortic coarctation remains controversial. Balloon aortoplasty of native aortic coarctation has shown a 91–100% primary success rate.26,27 However, for children aged less than 1 year the recurrence rate is 50–83%,27,28 though many of these can undergo successful redilatation. This is also true of recoarctation after surgery, and balloon aortoplasty is now the treatment of choice in this situation, giving an 88–91% success rate. Historically, surgery has been the mainstay of treatment; patch aortoplasty, subclavian flap aortoplasty, and coarctation resection with end-to-end anastomosis have been used. The first of these has a high recurrence rate (50%) and is therefore not used nowadays.29 Of the other two, rigorous analysis of early and late mortality has not demonstrated any advantage of either technique (3–8.6% and 4% vs 0–5% and 4–5%, respectively18,19,30,31); mortality in most series is due to associated cardiac defects. Further, medium-term recurrence rate is identical with the two techniques (6% and 3.6–4.3%, respectively). Younger age at the time of initial repair is a risk factor for recurrence, and a rate of 25–44% has been reported for low birth weight infants.25,32 However, the risk of maintaining these neonates on prostaglandin infusions until they have grown is high, and delaying surgical intervention cannot be reasonably advocated.

Inadequate pulmonary blood flow Initial management of patients with inadequate pulmonary blood flow involves PGE1 infusion to maintain ductal patency. Thereafter, when the child has been stabilized, more definitive intervention can be undertaken using a surgical shunt or stenting the ductus arteriosus.

Pulmonary atresia with ventricular septal defect This lesion is considered by some to be an extreme form of tetralogy of Fallot, and management of children with

confluent, good sized pulmonary arteries is directed by similar considerations as for other forms of tetralogy (see below). Consequently, in many units, complete correction is performed in the neonatal period even in premature neonates as small as 1400 g, with excellent results.33 In other units, where the prevailing philosophy advocates initial palliation, a Blalock–Taussig shunt is performed. In some patients the pulmonary arteries are severely hypoplastic or absent, and the patient may have major aortopulmonary collateral arteries. The management of these children is more controversial. When all lung segments are supplied by the hypoplastic pulmonary arteries it is necessary to promote growth of these vessels as early as possible by performing a central aortopulmonary artery shunt (the diminutive branch pulmonary arteries should be avoided, as the risk of vessel stenosis and occlusion is high), reconstructing the right ventricular outflow tract directly (when there is short segment pulmonary atresia), or creating an aortopulmonary window by anastomosing the main pulmonary artery directly to the aorta. Using such techniques, full correction may later be achieved in up to 100% of patients;34 small collateral vessels may later be coiled by interventional catheter techniques. For children in whom pulmonary blood flow is primarily dependent on major aortopulmonary collateral arteries rather than native pulmonary arteries, afferent conduits to the lungs must be constructed from the collaterals, a process known as ‘unifocalization’. While correction may be performed in stages, with each lung being unifocalized to a shunt before the intracardiac repair and central reconstruction of the pulmonary arteries is undertaken, this approach produces very varied results, with 21–86% of patients subsequently achieving complete repair,35,36 with a mortality rate of 12–14%.35,37 The alternative approach is to establish a normal circulation as early as feasible using only native tissue, by performing complete unifocalization at the same procedure as undertaking intracardiac repair. This would maximize the opportunity for adequate vessel growth, and can be achieved in up to 90% of patients38 with an early mortality of 9–10.6%.39,40 Some of the early deaths are due to low cardiac output state owing to the distal pulmonary vascular bed being too underdeveloped to allow biventricular repair, and as experience has increased it has been shown that on occasion staged repairs may still be most appropriate. In our experience 66% of patients have been able to undergo complete primary repair, while 27% require the ventricular septal defect to be left open initially. However, owing to poorly developed collateral vessels, 7% of patients still require staged unifocalizations through thoracotomies. Overall, the probability of complete repair at 2 years is 88% with a 3-year survival of 80%.40 In survivors, reinterventions are frequently required to dilate or stent anastomoses and coil collaterals.38,40

Congenital heart disease in the first year of life

Pulmonary atresia with intact interventricular septum Atresia of the pulmonary valve causes secondary problems which define what treatment can be offered. These problems are right heart hypoplasia and coronary sinusoids. When the atresia is long-standing, as suggested from the ‘flow related theory’ of cardiac development, the absence of output causes a lack of stimulus to right heart growth, and the right ventricle and tricuspid valve are secondarily hypoplastic. In the absence of contraindications, and when the tricuspid valve is not too hypoplastic, relieving the right ventricular outflow tract obstruction and promoting right ventricular output may allow later right ventricular growth. The presence of sinusoids is an independent risk factor for death, with a 1-year survival of 50–83%41,42 for those with sinusoids compared to 92–97.6% for those without them.41,43 When the coronary circulation is truly dependent on the right ventricle (i.e. no connection of the coronary artery/arteries to the aorta), palliation along the singleventricle route is all that is possible, as decompressing the ventricle will lead to myocardial ischemia. However, when the sinusoids simply interconnect with an essentially normal coronary system, ligation of the sinusoidal connections and decompression of the ventricle may be possible. Further, if the fistulae are confined to a single coronary artery territory, decompression may still be feasible if the amount of left ventricular myocardium ‘at risk’ is small.44 Sinusoids are predominantly confined to those with very small tricuspid annulus to body surface area index (z-value), however, and as studies suggest that the main predictor of a twoventricle repair is the z-value of the tricuspid valve,45 some centers do not attempt right ventricular decompression in patients with very small tricuspid valves. Right ventricular decompression may be achieved by surgery or transvascular balloon valvuloplasty with radiofrequency or laser perforation of the valve plate when complete atresia is present. Primary success with balloon valvuloplasty is 75–90%, with 5–14% mortality.46,47 Of survivors, 29–55% require further balloon dilatation or surgery.46,48 In comparison, surgery is associated with an 8–19% mortality and a 65–100% requirement for further intervention,47,49 though these patients often represent the worse end of the spectrum. The key to long-term success is stimulating adequate growth of the right ventricle both by relieving outflow obstruction and providing adequate preload. It has therefore been argued that a non-restrictive atrial septal defect is detrimental to the growth of the right ventricle, and some surgeons deliberately control flow across the interatrial septum to promote ventricular growth.50 However, despite these efforts the tricuspid valve and right ventricle may still not grow. Depending on how aggressively a two-ventricle circulation is pursued, survival is 50–98% at 1 year43,46 and 67.5–98% at 6 years.41,43 The lower survival rate reflects when a philosophy of right ventricular decompression is pursued in

695

all patients, while the higher survival is for a group in whom the final outcome was assumed from the indexed size of the tricuspid valve at the time of presentation. Inevitably, the number of patients achieving a two-ventricle repair is much smaller with the latter approach, 30%43 as opposed to 61–90%.41,49

Tricuspid atresia In the absence of a ventricular septal defect, primary palliation is as for other forms of inadequate pulmonary blood flow (see above). However, in a subgroup of neonates with an associated ventricular septal defect and no right ventricular outflow tract obstruction, the pulmonary circulation needs to be protected from excessive blood flow. Pulmonary artery banding is therefore required in the neonatal period. Later palliation requires a staged single-ventricle approach by which returning systemic blood bypasses the heart and directly enters the pulmonary arteries. While this is completed in childhood, the first stage, the bidirectional cavopulmonary shunt, should be undertaken during the first year of life to prevent the complications of chronic volume loading of the heart, which causes deterioration in cardiac function. Early reduction of volume overload is associated with improved long-term functional status and an increased aerobic exercise capacity.51 However, due to the physiologically high pulmonary vascular resistance in the neonate, a cavopulmonary shunt cannot be performed too early, and it has been found that performing a cavopulmonary shunt earlier than 1 month of age has a high failure rate and requirement for shunt takedown.52 In general, therefore, a bidirectional cavopulmonary shunt should be performed when patients are between 3 and 6 months of age, even if systemic arterial saturations are adequate at this time.

Severe tetralogy of Fallot When infants are severely cyanosed some intervention is necessary, either palliation or complete correction. There is evidence that long-term function of the heart and other organs is better if complete repair is undertaken early, but the decision on whether to palliate initially or completely repair depends on the philosophy of the unit; the arguments are therefore discussed in more depth below, under ‘Desirable lesions’.

Excessive pulmonary blood flow Children with abnormal communications between the chambers of the heart or the great vessels develop shunts, with blood flowing from the high-pressure area to the

696

Fetal Cardiology

Atrial level shunts

low-pressure area. At birth, as the pulmonary vascular resistance is physiologically high, the shunt is small, but increases as the pulmonary vascular resistance falls, which leads to volume loading of the heart and causes pulmonary vascular disease. The rate of development of pulmonary vascular disease is dependent on the shear stress that the endothelial cells are exposed to, and therefore children with arterial level shunts are at greatest risk.

Atrial septal defects only require closure within the first year of life if the child has other problems and it is anticipated that maximizing the efficiency of the cardiovascular system will benefit the child. As 14–66% of these defects close spontaneously, intervention before the age of 1 year should be considered extraordinary.

Arterial level shunts

Inadequate intracardiac mixing

These include aortopulmonary window, truncus arteriosus and persistent ductus arteriosus. Pulmonary endothelial damage occurs within days of birth, and may become irreversible within weeks of life. Further, run-off into the lungs may be so great that systemic perfusion is compromised, with systemic arterial diastolic pressures so low that cardiac and intestinal perfusion is at risk. Full correction is nowadays undertaken in the neonatal period except when other medical issues (such as necrotizing enterocolitis) preclude the use of cardiopulmonary bypass. In such cases, banding of the branch pulmonary arteries may be considered. Persistent ductus arteriosus (PDA), however, is a variable lesion; the shunt may be so large that urgent closure in the neonatal period is essential to permit adequate systemic perfusion, or the duct may partially close so that the lungs are minimally affected. For severely premature infants, it is suggested that any shunt exacerbates chronic lung disease and therefore all PDAs should be closed, regardless of how small. Initial medical therapy using non-steroidal anti-inflammatory agents is used if not contraindicated, while surgery is reserved for those who fail medical treatment. Other children with small shunts may undergo closure later using percutaneous techniques.

Transposition of the great arteries

Ventricular level shunts Closure of ventricular septal defects is undertaken during the first year of life if the defect is large and there is pulmonary overcirculation. Pulmonary vascular disease develops slower than for arterial level shunts as the shear stress is less, though with large defects irreversible changes may occur within months. Initial management is medical, but as the operation is usually relatively straightforward, delay in surgery is difficult to justify. More problematic are children with multiple ventricular septal defects, as the small size of the heart may preclude successful closure of all the defects, and the proportion of septum splinted by the repair may be prohibitive. In these children, pulmonary artery banding may be more advisable, with debanding of the pulmonary artery and direct closure of the remaining defects performed once the child has grown. This has the added advantage that some of the muscular defects may close in the interim.

Treatment of this abnormality is surgical with, nowadays, anatomic correction in the form of the ‘arterial switch procedure’ being the standard approach. However, initial management is medical/interventional, and surgery is often postponed for a few days until the neonate has overcome the insults of birth and the pulmonary vascular resistance has started to fall. Evidently, there has to be mixing of the saturated and destaturated blood. If transposition is associated with a ventricular septal defect, this alone may allow adequate mixing owing to the higher pulmonary blood flow. If the interventricular septum is intact, an interatrial communication is required, which is now performed by a Rashkind balloon atrial septostomy. Prostaglandin E1 may be necessary also to maintain patency of the ductus arteriosus. Definitive treatment by the arterial switch operation involves transecting the ascending aorta and main pulmonary artery and reanastomosing them in the correct orientation. In addition, the first branches of the aorta, the coronary arteries, must also be translocated without compromising coronary artery flow.

Pulmonary venous obstruction This lesion remains a surgical emergency; interventional palliation by balloon venoplasty and stenting of the communicating vein achieves little but to delay definitive treatment. Infusing prostaglandin rarely improves the situation, and prolonged obstruction quickly induces structural changes in the pulmonary veins. Surgery involves anastomosing the venous confluence to the back of the left atrium.

Vascular abnormalities Vascular rings Vascular rings always require surgery to relieve the obstruction. At surgery the ring is divided and all associated fibrous tissue dissected off to allow complete relief of obstruction. Classically this has been performed by open thoracotomy, though nowadays an endoscopic approach may be used.53

Congenital heart disease in the first year of life

Although operative mortality is low, patients are frequently left with tracheobronchomalacia, and this is the commonest cause of death.54 Long-term ventilation with or without tracheostomy is often required. Some surgeons excise the airway segment at the time of repair with end-to-end reconstruction, or suspend the malacic segment within a tube of polytetrafluoroethylene, though these are rarely performed. Intraluminal stenting may also be used. Despite the severe early problems, 70–97% are asymptomatic at long-term follow-up.55,56

Anomalous origin of the coronary arteries from the pulmonary artery Regardless of the apparent severity of left ventricular dysfunction at the time of presentation, after emergency resuscitation surgery should be undertaken. Presentation is usually in infancy, as coronary steal is not significant until the pulmonary vascular resistance has fallen. Historically, the origin of the left coronary artery was tied off at the pulmonary artery, but outcome was poor and nowadays revascularization is performed. Ideally this is achieved by reimplanting the coronary ostium into the aorta, but when this cannot be achieved a Takeuchi procedure (intrapulmonary artery baffling of the coronary ostium to a surgically created aortopulmonary window) may be done. Alternatively, the coronary artery may be grafted using the saphenous vein or prosthetic tubing, or the subclavian artery may be anastomosed to the coronary ostium. Surgical mortality is currently 0–20% with revascularization;57,58 mitral regurgitation, which is frequently present, is an independent risk factor for death.59 It is, however, not recommended that mitral valve repair be undertaken at the primary operation, as in up to 62% of patients mitral valve function improves with revascularization alone. Postoperative mechanical support is required in up to 25% of patients due to the preoperative insult, yet long-term outcome is excellent, with cardiac function and regional wall motion returning to normal at rest in infants within 7 months.60 However, myocardial flow reserve is significantly impaired late-postoperatively, with associated impairment of exercise capacity.61

Desirable lesions Having considered lesions which require intervention during the first year of life, there are some lesions that are better treated by intervention during the first year of life, although this is more a philosophical issue. The lesion par

697

excellence that falls into this category is tetralogy of Fallot. Correcting the circulation in patients with this lesion as early as possible has been shown to minimize the morbidity and mortality associated with the abnormal circulation, and allows other organ systems to develop more normally. For example, patients with tetralogy of Fallot develop fibrosis of the ventricular muscle due to chronic myocardial ischemia from the hypertrophy and cyanosis,62,63 and have malformed lungs due to impaired pulmonary angiogenesis and alveogenesis.64 The argument hinges on the risk/benefit ratio of early complete correction; if it can be achieved with no increase in morbidity or mortality compared to initial palliation there is no advantage to the latter approach. As experience has grown, mortality has fallen to 0–4%, and the morbidity of early complete repair is no higher than for a two-stage approach. Based on this evidence, many now pursue a policy of complete surgical correction in all patients with tetralogy of Fallot, regardless of age or size, except in a few with severe coexistent medical problems in whom a bypass run is considered high risk. Other centers, based on the same evidence, consider that initial palliation is preferable. Each center, dependent on the local experience and philosophy, must determine for itself what is their optimal approach for the management of these children.

Fetal cardiac surgery If it is considered that life begins at conception, cardiac intervention in the first year of life includes the fetal period. Though experimental at present, there is a good theoretical basis for considering this strategy: to reduce mortality, to improve outcome in survivors, and to reduce damage to surrounding organs.

Preventing death Recent prenatal echocardiographic studies have shown that the in utero prognosis is much poorer for fetuses with congenital heart disease than had been appreciated previously, with mortality rates of 12–18%.65,66 In addition, postnatally, the transition from a fetal to a neonatal circulation is a period of high risk as the two circulations become separated soon after birth (see Chapter 10), and lesions that are not necessarily lethal before birth may become lethal after birth. The risk is highest when cardiac disease is unexpected and the infants are born in centers inexperienced in the delivery and resuscitation of such individuals; overall survival for neonates with cardiac malformations amenable to biventricular repair may be as high as 96% for those diagnosed prenatally, whereas for a similar cohort not diagnosed prenatally survival is 76%.67

698

Fetal Cardiology

Improving outcome Early intrauterine intervention may also improve the outcome for the affected fetus due to an improvement in either gross or ultrastructural cardiac development.

Improving gross cardiac development Following the rotation and folding of the cardiac tube and the formation of the cardiac structures, the flow-directed theory of cardiac development hypothesizes that further growth and development of the cardiac chambers and great vessels is primarily directed by the volume and pressure of blood flowing in the heart.68 Therefore, if inflow to a chamber is restricted, for example by a stenotic valve, there will be inadequate stimulus for that chamber to grow and it will become hypoplastic. A relatively simple primary lesion (a stenotic valve) may therefore cause such disturbances in intracardiac flow patterns that a far more complex secondary malformation (ventricular hypoplasia) results. Theoretically, if the primary lesion can be identified and adequately relieved early enough during intrauterine development, normal cardiac growth may be reestablished. Clinical studies support this theory of secondary cardiac development. Patients with tricuspid atresia and normally related great vessels may be divided into two groups depending on the presence of a ventricular septal defect (VSD). In patients without a VSD the right ventricle and pulmonary valve are severely hypoplastic, and after birth they require a systemic–pulmonary artery shunt to maintain adequate pulmonary blood flow after closure of the ductus arteriosus. However, in the subgroup of patients with a VSD, right ventricular growth may be entirely normal, as may the right ventricular outflow tract, and in these patients a pulmonary artery band is frequently required to limit pulmonary blood flow. Blood is able to bypass the blockage caused by the tricuspid atresia and access the right ventricle via the VSD. Developmental hemodynamic forces are therefore normalized and distal right heart structures develop normally. Evidently, although both these hearts have the substrate to develop tricuspid valve stenosis/atresia, normal growth of other right heart structures may be achieved if normal flow patterns can be established. Similar results have been obtained using animal models. If obstruction to ventricular inflow is produced by inflating a balloon within the left atrial cavity, left ventricular output falls acutely by 70%, and within 7 days the chamber size may decrease by up to 50%.69 There is an associated fall in the left ventricular/right ventricular weight ratio which is directly proportional to the length of time that the balloon has been in place, suggesting a direct causal effect. This leads to an early form of hypoplastic left heart syndrome. Similarly, obstructing ventricular outflow by banding the ascending aorta and reducing left ventricular output by 36% causes left ventricular hyperplasia over the

first 10 days. However, over a longer period (30–60 days) the left ventricular cavity becomes obliterated, with a decrease in the left ventricular/right ventricular weight ratio. These changes are more pronounced in fetuses with tighter aortic bands.

Improving ultrastructural cardiac development Soon after birth the ability for cardiac growth by hyperplasia and angiogenesis (as occurs in the fetus) is lost, and the neonatal heart responds to stress by hypertrophy, with little or no angiogenesis, resulting in a very different ultrastructure of the organ. In utero intervention may allow normal cardiac ultrastructural development; in structurally normal hearts the response of the fetal myocardium to experimentally induced left ventricular outflow tract obstruction is one of hyperplasia.69 In humans, the age at which this switch occurs is unknown, but in rats the hyperplastic response is known to be lost within 3 weeks of birth and angiogenesis by 7 weeks.

Reducing damage to surrounding organs The development of other intrathoracic structures is dependent on adequate physical space within the chest, and their growth may be compromised if the heart is enlarged. In patients with pulmonary atresia and intact interventricular septum with a dilated ventricle that survived to term in one study, the neonatal mortality was 100% due to an inability to ventilate the infant adequately after birth, owing to severe lung hypoplasia caused by the grossly enlarged heart.70 This is similar for fetuses with Ebstein’s anomaly. If the cardiac distension could be prevented early enough, adequate lung growth may be permitted and postnatal survival ensured.

Requirements for fetal cardiac intervention If fetal cardiac surgery is to be attempted, the lesion for which the intervention is required must have such a poor outlook with conventional surgical approaches that the increased risk of in utero intervention is justified. These will mainly be lesions that result in single ventricle physiology, chief of which are hypoplastic left heart syndrome and severe forms of Ebstein’s anomaly and pulmonary atresia with intact interventricular septum. Second, there must be a simple primary abnormality that can be readily dealt with. Third, the intervention must be simple, quick, and reliable. Finally, the diagnosis of the defect must be

Congenital heart disease in the first year of life

practical early enough during fetal development to allow time for adequate catch-up growth.

Specific issues There are three practical issues which confront the fetal cardiac interventionalist: size, tissue structure, and response of the fetoplacental unit to cardiac bypass.

Size The development of miniaturized bypass circuits and oxygenators now allows us to operate on infants weighing as little as 1000 g clinically. Experimentally, bypass has been performed on fetuses weighing as little as 500 g.

Tissue structure Fetal tissues are extremely friable due to the high water content of 88% at 18 weeks’ gestation, compared to 69% at term and 60% in adulthood.71,72 There is also a gradual accumulation of collagen: 2.4 g/kg, 16.8 g/kg, and 45.7 g/kg, respectively,73 with a progressive maturation from type I to type III with increased crosslinking.74 These friable tissues may preclude the use of conventional techniques and the use of laser-assisted tissue welding and laser scalpels is being investigated.

Fetoplacental unit response to bypass The greatest obstacle to fetal cardiac surgery is the reaction of the fetoplacental unit to bypass. In the fetus, the systemic circulation lies in parallel with the placental circulation. Relative changes in the vascular resistances of the two beds will therefore affect the perfusion of the other bed (analogous to a Blalock–Taussig shunt). Initial experiments showed that fetal bypass caused an increase in blood flow to all fetal organs but a decrease to

699

the placenta by 25–65%.75,76 This change persisted even after bypass, and was caused by an increase in vascular resistance at the level of the cotyledons and umbilical vein.77 Inevitably this led to respiratory acidosis and hypoxemia, but while systemic perfusion was initially well maintained, the progressive hypoxemia secondarily caused tissue hypoxia and metabolic acidosis (Figure 45.1). These changes in placental vascular resistance resemble the effects of postnatal cardiopulmonary bypass, which is due to a ‘whole body inflammatory response’, and it has been demonstrated that there are similarities between the two. Minimizing the inflammatory response by inhibiting eicosanoid metabolism (using indomethacin and corticosteroids) maintains adequate placental blood flow for a period,78,79 but later a more insidious metabolic acidosis resistant to this pharmacological manipulation develops, again caused by elevation in placental vascular resistance. This is part of the fetal stress response,80 and blocking it using a high spinal anesthetic, in combination with the other techniques outlined above, has permitted long-term survival after cardiac bypass in 80% of fetuses experimentally.81 The trigger of the inflammatory and stress responses is unknown; it may be the contact of fetal blood with foreign surfaces, or the hemodilution which accompanies bypass. Both these issues have been addressed using a novel bypass pump, the Hemopump® (Johnson & Johnson Interventional Systems, Rancho Cordova, CA, USA).82 This miniaturized axial flow pump with a priming volume of only 15 ml (150 ml for a conventional circuit) also minimizes contact surface area. Using this circuit, white cell activation is significantly reduced, and even without pharmacological manipulations, long-term fetal survival has been achieved in 89%.83 Other attempts to improve fetal survival following in utero bypass have included using pulsatile perfusion84 and increasing flow rates to above 300 ml/kg per min.85 However, neither of these studies attempted long-term fetal survival, and their efficacy therefore remains in doubt.

Intervention or open surgery? Placenta (lungs)

Heart

Body

Figure 45.1 The balance of fetal cardiac output. As the pulmonary and systemic circulations are in parallel with a Blalock–Taussig shunt, so the placental and systemic circulations are in parallel in the fetus. Relative changes in the impedances of these two circuits will cause dramatic changes in the distribution of the cardiac output in the fetus.

The discussion above has concentrated mainly on issues associated with open cardiac operations. From the experimental results it is evident that the fetal stress response is a major factor impacting on the success of the operation. While efforts to understand the trigger for the stress response and to control its effects are under investigation, there remains the opportunity for less stressful procedures, such as interventional cardiological techniques, to be immediately applicable. Many of the issues are the same, such as size and delicacy of the fetal structures. However, percutaneous placement of wires and balloons without the necessity for full fetal exposure may reduce the morbidity of the intervention (see Chapter 44).

700

Fetal Cardiology

Fetal cardiac pacing The area in which fetal cardiac intervention is closest to clinical implementation is cardiac pacing. One in 20 000 fetuses develops congenital complete heart block86 and of these, 57–80% have structural heart disease associated with an in utero mortality of up to 86%.87,88 In contrast, in those without severe structural heart disease, the in utero mortality is 11–25%.88,89 Hemodynamic studies suggest that as the fetal heart can accommodate down to a rate of 40 beats per minute there is no inherent myocardial dysfunction in these fetuses, and that accelerating the cardiac rate by pacing may reverse the cardiac failure. The aim of intervention is to treat the fetal cardiac failure adequately to allow safe progression of pregnancy to a safe gestational age, generally 32 weeks. In cases where transplacental or direct intrafetal therapy has failed there are two options. The first is early delivery and immediate neonatal pacing, accepting the higher risks associated with premature delivery,90 initially using temporary epicardial leads if the child is too small for a permanent system. If this is not feasible due to fetal maturity, fetal cardiac pacing may be undertaken. Experimentally both epicardial91 and endocardial92 approaches have been successfully used acutely and chronically in fetuses with induced heart block.91 Further, using conductance catheter techniques, studies have demonstrated normal contractility, a maintained inotropic response, and an intact force–frequency relationship for the paced fetuses,93 suggesting that the fetal myocardium tolerates pacing as well as the neonatal myocardium. To date there have been six reported cases of attempted in utero fetal pacing, clinically accessing the fetal endocardium using transthoracic/transmyocardial, transhepatic, or intracaval routes. The pacing generator was placed external to the fetus. Although effective pacing was achieved in all cases, capture was lost hours after the procedure, presumably due to lead dislodgement by fetal movement. A number of groups are currently involved in improving pacemaker and lead technology to minimize the risk of lead displacement in this scenario. Fetal cardiac intervention is in its infancy. Although there are theoretical advantages to this approach, the final proof of its efficacy is awaited. Certainly there are still hurdles which must be overcome, but significant progress has been made over the past 10 years. It is likely that the final proof will be available in the foreseeable future.

References 1. Hammon JW Jr, Lupinetti FM, Maples MD et al. Predictors of operative mortality in critical valvular aortic stenosis presenting in infancy. Ann Thorac Surg 1988; 45: 537–40. 2. Rhodes LA, Colan SD, Perry SB et al. Predictors of survival in neonates with critical aortic stenosis. Circulation 1991; 84: 2325–35.

3. Serraf A, Piot JD, Lacour-Gayet F et al. Biventricular repair approach in ducto-dependent neonates with hypoplastic but morphologically normal left ventricle. J Am Coll Cardiol 1999; 33: 827–34. 4. Tani LY, Minich LL, Pagotto LT et al. Left heart hypoplasia and neonatal aortic arch obstruction: is the Rhodes left ventricular adequacy score applicable? J Thorac Cardiovasc Surg 1999; 118: 81–6. 5. Reddy VM, Rajasinghe HA, McElhinney DB et al. Extending the limits of the Ross procedure. Ann Thorac Surg 1995; 60: S600–3. 6. Carminati M, Giusti S, Spadoni I et al. Balloon aortic valvuloplasty in the first year of life. J Interv Cardiol 1995; 8: 759–66. 7. Robinson BV, Brzezinska-Rajszys G, Weber HS et al. Balloon aortic valvotomy through a carotid cutdown in infants with severe aortic stenosis: results of the multi-centre registry. Cardiol Young 2000; 10: 225–32. 8. Gatzoulis MA, Rigby ML, Shinebourne EA, Redington AN. Contemporary results of balloon valvuloplasty and surgical valvotomy for congenital aortic stenosis. Arch Dis Child 1995; 73: 66–9. 9. Alexiou C, Langley SM, Dalrymple-Hay MJ et al. Open commissurotomy for critical isolated aortic stenosis in neonates. Ann Thorac Surg 2001; 71: 489–93. 10. Gildein HP, Kleinert S, Weintraub RG et al. Surgical commissurotomy of the aortic valve: outcome of open valvotomy in neonates with critical aortic stenosis. Am Heart J 1996; 131: 754–9. 11. Hawkins JA, Minich LL, Tani LY et al. Late results and reintervention after aortic valvotomy for critical aortic stenosis in neonates and infants. Ann Thorac Surg 1998; 65: 1758–62. 12. Calhoon JH, Bolton JW. Ross/Konno procedure for critical aortic stenosis in infancy. Ann Thorac Surg 1995; 60(Suppl): S597–9. 13. Norwood WI, Kirklin JK, Sanders SP. Hypoplastic left heart syndrome: experience with palliative surgery. Am J Cardiol 1980; 45: 87–91. 14. Sano S, Ishino K, Kawada M et al. Right ventricle-pulmonary artery shunt in first-stage palliation of hypoplastic left heart syndrome. J Thorac Cardiovasc Surg 2003; 126: 504–9. 15. Jacobs ML, Blackstone EH, Bailey LL. Intermediate survival in neonates with aortic atresia: a multi-institutional study. J Thorac Cardiovasc Surg 1998; 116: 417–31. 16. Gibbs JL, Wen C, Watterson KG et al. Stenting of the arterial duct combined with banding of the pulmonary arteries and atrial septectomy or septostomy: a new approach to palliation for the hypoplastic left heart syndrome. Br Heart J 1993; 69: 551–5. 17. McElhinney DB, Sherwood MC, Keane JF et al. Current management of severe mitral stenosis; outcomes of transcatheter and surgical therapy in 108 infants and children. Ciruclation 2005; 112: 707–14. 18. Rajasinghe HA, Reddy VM, van Son JA et al. Coarctation repair using end-to-side anastomosis of descending aorta to proximal aortic arch. Ann Thorac Surg 1996; 61: 840–4. 19. Cobanoglu A, Thyagarajan GK, Dobbs JL. Surgery for coarctation of the aorta in infants younger than 3 months: endto-end repair versus subclavian flap angioplasty: is either operation better? Eur J Cardiothorac Surg 1998; 14: 19–25.

Congenital heart disease in the first year of life

20. Moore P, Adatia I, Spevak PJ et al. Severe congenital mitral stenosis in infants. Circulation 1994; 89: 2099–106. 21. Uva MS, Galletti L, Gayet FL et al. Surgery for mitral valve disease in the first year of life. J Thorac Cardiovasc Surg 1995; 109: 164–74. 22. Schreiber C, Eicken A, Vogt M et al. Repair of interrupted aortic arch: results after more than 20 years. Ann Thorac Surg 2000; 70: 1896–9. 23. Hirooka K, Fraser CD Jr. One-stage neonatal repair of complex aortic arch obstruction or interruption. Recent experience at Texas Children’s Hospital. Texas Heart Inst J 1997; 24: 317–21. 24. Mainwaring RD, Lamberti JJ. Mid- to long-term results of the two-stage approach for type B interrupted aortic arch and ventricular septal defect. Ann Thorac Surg 1997; 64: 1782–5. 25. Hakimi M, Clapp SK, Walters HL 3rd et al. Arch growth after staged repair of interrupted aortic arch using carotid artery interposition. Ann Thorac Surg 1997; 64: 503–7. 26. Fletcher SE, Nihill MR, Grifka RG et al. Balloon angioplasty of native coarctation of the aorta: midterm follow-up and prognostic factors. J Am Coll Cardiol 1995; 25: 730–4. 27. Park Y, Lucas VW, Sklansky MS et al. Balloon angioplasty of native aortic coarctation in infants 3 months of age and younger. Am Heart J 1997; 134: 917–23. 28. Rao PS, Galal O, Smith PA, Wilson AD. Five- to nine-year follow-up results of balloon angioplasty of native aortic coarctation in infants and children. J Am Coll Cardiol 1996; 27: 462–70. 29. Backer CL, Paape K, Zales VR et al. Coarctation of the aorta. Repair with polytetrafluoroethylene patch aortoplasty. Circulation 1995; 92(Suppl 2): 132–6. 30. Backer CL, Mavroudis C, Zias EA et al. Repair of coarctation with resection and extended end-to-end anastomosis. Ann Thorac Surg 1998; 66: 1365–70. 31. Jahangiri M, Shinebourne EA, Zurakowski D et al. Subclavian flap angioplasty: does the arch look after itself? J Thorac Cardiovasc Surg 2000; 120: 224–9. 32. Bacha EA, Almodovar M, Wessel DL et al. Surgery for coarctation of the aorta in infants weighing less than 2 kg. Ann Thorac Surg 2000; 71: 1260–4. 33. Reddy VM, Liddicoat JR, McElhinney DB et al. Routine primary repair of tetralogy of Fallot in neonates and infants less than three months of age. Ann Thorac Surg 1995; 60(Suppl): S592–6. 34. Metras D, Chetaille P, Kreitmann B et al. Pulmonary atresia with ventricular septal defect, extremely hypoplastic pulmonary arteries, major aorto-pulmonary collaterals. Eur J Cardiothorac Surg 2001; 20: 590–6. 35. Murthy KS, Rao SG, Naik SK et al. Evolving surgical management for ventricular septal defect, pulmonary atresia, and major aortopulmonary collateral arteries. Ann Thorac Surg 1999; 67: 760–4. 36. Tchervenkov CI, Salasidis G, Cecere R et al. One-stage midline unifocalization and complete repair in infancy versus multiple-stage unifocalization followed by repair for complex heart disease with major aortopulmonary collaterals. J Thorac Cardiovasc Surg 1997; 114: 727–35. 37. Iyer KS, Mee RB. Staged repair of pulmonary atresia with ventricular septal defect and major systemic to pulmonary artery collaterals. Ann Thorac Surg 1991; 51: 65–72.

701

38. Reddy VM, Liddicoat JR, Hanley FL. Midline one-stage complete unifocalization and repair of pulmonary atresia with ventricular septal defect and major aortopulmonary collaterals. J Thorac Cardiovasc Surg 1995; 109: 832–44. 39. Lofland GK. The management of pulmonary atresia, ventricular septal defect, and multiple aorta pulmonary collateral arteries by definitive single stage repair in early infancy. Eur J Cardiothorac Surg 2000; 18: 480–6. 40. Reddy VM, McElhinney DB, Amin Z et al. Early and intermediate outcomes after repair of pulmonary atresia with ventricular septal defect and major aortopulmonary collateral arteries: experience with 85 patients. Circulation 2000; 101: 1826–32. 41. Ekman Joelsson BM, Sunnegardh J, Hanseus K et al. The outcome of children born with pulmonary atresia and intact ventricular septum in Sweden from 1980 to 1999. Scand Cardiovasc J 2001; 35: 192–8. 42. Powell AJ, Mayer JE, Lang P, Lock JE. Outcome in infants with pulmonary atresia, intact ventricular septum, and right ventricle-dependent coronary circulation. Am J Cardiol 2000; 86: 1272–4. 43. Jahangiri M, Zurakowski D, Bichell D et al. Improved results with selective management in pulmonary atresia with intact ventricular septum. J Thorac Cardiovasc Surg 1999; 118: 1046–55. 44. Giglia TM, Mandell VS, Connor AR et al. Diagnosis and management of right ventricle-dependent coronary circulation in pulmonary atresia with intact ventricular septum. Circulation 1992; 86: 1516–28. 45. Hanley FL, Sade RM, Blackstone EH et al. Outcomes in neonatal pulmonary atresia with intact ventricular septum. A multiinstitutional study. J Thorac Cardiovasc Surg 1993; 105: 406–23. 46. Ovaert C, Qureshi SA, Rosenthal E et al. Growth of the right ventricle after successful transcatheter pulmonary valvotomy in neonates and infants with pulmonary atresia and intact ventricular septum. J Thorac Cardiovasc Surg 1998; 115: 1055–62. 47. Alwi M, Geetha K, Bilkis AA et al. Pulmonary atresia with intact ventricular septum percutaneous radiofrequency-assisted valvotomy and balloon dilation versus surgical valvotomy and Blalock Taussig shunt. J Am Coll Cardiol 2000; 35: 468–76. 48. Wang JK, Wu MH, Chang CI et al. Outcomes of transcatheter valvotomy in patients with pulmonary atresia and intact ventricular septum. Am J Cardiol 1999; 84: 1055–60. 49. Sano S, Ishino K, Kawada M et al. Staged biventricular repair of pulmonary atresia or stenosis with intact ventricular septum. Ann Thorac Surg 2000; 70: 1501–6. 50. Laks H; in discussion of Jahangiri M, Zurakowski D, Bichell D et al. Improved results with selective management in pulmonary atresia with intact ventricular septum. J Thorac Cardiovasc Surg 1999; 118: 1046–5. 51. Mahle WT, Wernovsky G, Bridges ND et al. Impact of early ventricular unloading on exercise performance in preadolescents with single ventricle Fontan physiology. J Am Coll Cardiol 1999; 34: 1637–43. 52. Reddy VM, McElhinney DB, Moore P et al. Outcomes after bidirectional cavopulmonary shunt in infants less than 6 months old. J Am Coll Cardiol 1997; 29: 1365–70.

702

Fetal Cardiology

53. Burke RP, Wernovsky G, van der, Velde M et al. Videoassisted thoracoscopic surgery for congenital heart disease. J Thorac Cardiovasc Surg 1995; 109: 499–507. 54. Chun K, Colombani PM, Dudgeon DL, Haller JA Jr. Diagnosis and management of congenital vascular rings: a 22-year experience. Ann Thorac Surg 1992; 53: 597–602. 55. Anand R, Dooley KJ, Williams WH, Vincent RN. Follow-up of surgical correction of vascular anomalies causing tracheobronchial compression. Pediatr Cardiol 1994; 15: 58–61. 56. Rivilla F, Utrilla JG, Alvarez F. Surgical management and follow-up of vascular rings. Z Kinderchirurg 1989; 44: 199–202. 57. Wollenek G, Domanig E, Salzer-Muhar U et al. Anomalous origin of the left coronary artery: a review of surgical management in 13 patients. J Cardiovasc Surg (Torino) 1993; 34: 399–405. 58. Backer CL, Hillman N, Dodge-Khatami A, Mavroudis C. Anomalous origin of the left coronary artery from the pulmonary artery: successful surgical strategy without assist devices. Semin Thorac Cardiovasc Surg Pediatr Cardiac Surg Annu 2000; 3: 165–72. 59. Schwartz ML, Jonas RA, Colan SD. Anomalous origin of left coronary artery from pulmonary artery: recovery of left ventricular function after dual coronary repair. J Am Coll Cardiol 1997; 30: 547–53. 60. Jin Z, Berger F, Uhlemann F et al. Improvement in left ventricular dysfunction after aortic reimplantation in 11 consecutive paediatric patients with anomalous origin of the left coronary artery from the pulmonary artery. Early results of a serial echocardiographic follow-up. Eur Heart J 1994; 15: 1044–9. 61. Singh TP, Di Carli MF, Sullivan NM et al. Myocardial flow reserve in long-term survivors of repair of anomalous left coronary artery from pulmonary artery. J Am Coll Cardiol 1998; 31: 437–43. 62. Kirklin JW, Blackstone EH, Pacifico AD et al. Routine primary repair vs two stage repair of tetralogy of Fallot. Circulation 1979; 60: 373–86. 63. Castañeda AR, Freed MD, Williams RG, Norwood WI. Repair of tetralogy of Fallot in infancy: early and late results. J Thorac Cardiovasc Surg 1977; 74: 372–81. 64. Kirklin JW, Blackstone EH, Kirklin JK et al. Surgical results and protocols in the spectrum of tetralogy of Fallot. Ann Surg 1983; 198: 251–65. 65. Fesslova’ V, Nava S, Villa L. Evolution and long term outcome in cases with fetal diagnosis of congenital heart disease: Italian multicentre study. Heart 1999; 82: 594–9. 66. Allan LD, Crawford DC, Handerson R, Tynan M. Spectrum of congenital heart disease detected echocardiographically in prenatal life. Br Heart J 1985; 54: 523–6. 67. Copel JA, Tan AS, Kleinman CS. Does a prenatal diagnosis of congenital heart disease alter short-term outcome? Ultrasound Obstet Gynecol 1997; 10: 237–41. 68. Rose V, Clark LE. Etiology of congenital heart disease. In: Freedom RM, Benson LN, Smallhorn JF, eds. Neonatal Heart Disease. New York: Springer-Verlag, 1992: 3–13. 69. Fishman NH, Hof RB, Ruolph AM, Heymann MA. Models of congenital heart disease in fetal lambs. Circulation 1978; 58: 354–64.

70. Allan LD, Crawford DC, Tynan MJ. Pulmonary atresia in prenatal life. J Am Coll Cardiol 1986; 8: 1131–6. 71. Widdowson EM, McCance RA, Spray CM. Chemical composition of human body. Clin Sci 1951; 10: 113–25. 72. Widdowson EM, Spray CM. Chemical development in utero. Arch Dis Child 1951; 26: 205–14. 73. Hallock GG, Rice DC, Merkel JR, DiPaolo BR. Analysis of collagen content in the fetal wound. Ann Plast Surg 1988; 21: 310–15. 74. Lovvorn HN III, Cheung DT, Nimni ME et al. Relative distribution and crosslinking of collagen distinguish fetal from adult sheep wound repair. J Pediatr Surg 1999; 34: 218–23. 75. Richter RC, Slate RK, Rudolph AM, Turley K. Fetal blood flow during hypothermic cardiopulmonary by-pass in utero. J Cardiovasc Surg 1985; 26: 86. 76. Bradley SM, Hanley FL, Duncan BW et al. Fetal cardiac bypass alters regional blood flows, arterial blood gases, and hemodynamics in sheep. Am J Physiol 1992; 263: H919–28. 77. Bradley SM, Hanley FL, Jennings RW et al. The site of increased placental vascular resistance caused by fetal cardiopulmonary bypass in sheep. Surg Forum 1991; 72: 196–7. 78. Sabik JF, Assad RS, Hanley FL. Prostaglandin synthesis inhibition prevents placental dysfunction after fetal cardiac bypass. J Thorac Cardiovasc Surg 1992; 103: 733–42. 79. Sabik JF, Heinemann MK, Assad RS, Hanley FL. High-dose steroids prevent placental dysfunction after fetal cardiac bypass. J Thorac Cardiovasc Surg 1996; 111: 536–44. 80. Fenton KN, Heinemann MK, Hickey PR et al. Inhibition of the fetal stress response improves cardiac output and gas exchange after fetal cardiac bypass. J Thorac Cardiovasc Surg 1994; 107: 1416–22. 81. Fenton KN, Zinn HE, Heinemann MK et al. Long-term survivors of fetal cardiac bypass in lambs. J Thorac Cardiovasc Surg 1994; 107: 1423–7. 82. Reddy VM, Liddicoat JR, Klein JR et al. Fetal cardiac bypass using an in-line axial flow pump to minimize extracorporeal surface and avoid priming volume. Ann Thorac Surg 1996; 62: 393–400. 83. Reddy VM, Liddicoat JR, Klein JR et al. Long-term fetal outcome after fetal cardiac bypass: fetal survival to full term and organ abnormalities. J Thorac Cardiovasc Surg 1996; 111: 536–44. 84. Champsaur G, Parisot P, Martinot S et al. Pulsatility improves hemodynamics during fetal bypass. Experimental comparative study of pulsatile versus steady flow. Circulation 1994; 90: II47–50. 85. Hawkins JA, Clark SM, Shaddy RE, Gay WA Jr. Fetal cardiac bypass: improved placental function with moderately high flow rates. Ann Thorac Surg 1994; 57: 293–7. 86. Michaëlsson M, Engle MA. Congenital complete heart block: an international study of the natural history. In: Breast YAN, Engle MA, eds. Cardiovascular Clinics. Philadelphia, PA: FA Davis, 1972: 89–101. 87. Machado MVL, Tynan MJ, Curry PVL, Allan LD. Fetal complete heart block. Br Heart J 1988; 60: 512–15.

Congenital heart disease in the first year of life

88. Schmidt KG, Ulmer HE, Silverman NH et al. Perinatal outcome of fetal comlete atrioventricular block: a multicenter experience. J Am Coll Cardiol 1991; 17: 1360–6. 89. Groves AMM, Allan LD, Rosenthal E. Outcome of isolated congenital complete heart block diagnosed in utero. Heart 1996; 75: 190–4. 90. Deloof E, Devlieger H, Van Hoestenberghe R et al. Management with a staged approach of the premature hydropic fetus due to complete congenital heart block. Eur J Pediatr 1997; 156: 521–3.

703

91. Scagliotti D, Shimokochi DD, Pringle KC. Permanent cardiac pacemaker implant in the fetal lamb. Pacing Clin Electrophysiol 1987; 10: 1253–61. 92. Kikuchi Y, Shiraishi H, Igarashi H et al. Cardiac pacing in fetal lambs: intrauterine transvenous cardiac pacing for fetal complete heart block. Pacing Clin Electrophysiol 1995; 18: 417–23. 93. Liddicoat JR, Klein JR, Reddy VM et al. Hemodynamic effects of chronic prenatal ventricular pacing for the treatment of complete atrioventricular block. Circulation 1997; 96: 1025–30.

46 Genetic counseling in families with congenital heart defects Klaus Zerres and Sabine Rudnik-Schöneborn Congenital heart disease occurs in around 1% of births and is therefore the largest group for which advice is sought, usually regarding the risk for further affected children, but increasingly concerning the offspring of a successfully treated patient. Although around 90% of congenital heart defects are not familial (Table 46.1), it seems increasingly likely that genetic factors are involved in most cases. The increase of genetic knowledge as well as advances in prenatal echocardiography offer more tools for counseling, but require precise information. Among the large amount of knowledge, the textbook by Nora et al1 on cardiovascular disease is still an excellent comprehensive monograph about the subject. Owing to rapidly increasing genetic knowledge, current online information providing details about all aspects of certain conditions is essential. The most important genetic database including relevant references is Online Mendelian Inheritance in Man (http:// www.ncbi.nlm.nih.gov/Omim/searchomim.html),2 based on Victor McKusick’s classical catalogs.

Genetic basis of congenital heart defects The majority of congenital heart defects do not cluster within families, indicating that genetic factors are less important in their etiology. A small but important group of congenital heart defects and most syndromes with congenital heart defects as a major sign follow monogenic modes of inheritance. The major characteristics are shown in Table 46.2. The vast majority are believed to be ‘multifactorially’ inherited, indicating that the combination of genetic factors as well as often unknown exogenous factors may cause the defect. The genetic predisposition is assumed to be polygenic, with a greater number of relevant genes involved. The essential distinguishing factor from Mendelian disorders is that a single genetic locus cannot be held responsible for the condition, and that it is the result of the additive

effect or interaction of a number of genetic loci and of a number of external factors. Recent findings in other diseases underline that epigenetic mechanisms might be responsible for gene expression and could also explain transmission to the next generation. Similar observations in congenital heart defects can be expected. The sum of these factors determines a person’s liability to be affected with a particular disorder, and the liability should show a more or less ‘normal’ distribution in the population, with most people having an intermediate degree of liability and a smaller number of each end of the distribution curve having unusually low or unusually high liabilities (Figure 46.1). An important group comprises those who are actually affected, whose liability is above a postulated ‘threshold’ for the disorder. The liability of relatives of a patient with the disorder will be distributed in a similar way to that of the general population, but the curve will be shifted toward a higher liability because of the increased genetic component. The following practical aspects, which are in contrast to the rules of classical Mendelian modes of inheritance (see Table 46.2) are relevant and are of importance for genetic counseling: (1) increased risk is greatest among closest relatives and decreases rapidly with distance of relationship; (2) the risk of recurrence depends on the incidence

Table 46.1

Causes of congenital heart defects

Cause Multifactorial

Percentage 70

Genetic chromosome abnormalities monogenic defects

20–25 3–5

Exogenous intrauterine infections

1

other teratogenic effects

1

706

Fetal Cardiology

Table 46.2

Modes of inheritance in monogenic disorders

Mode of inheritance

Characteristics

Autosomal dominant

Usually further first-degree family members are affected. In severe diseases spontaneous mutations might occur. The clinical picture is often variable (variable expressivity) and sometimes not evident at all (incomplete penetrance). Risk to children depending on the penetrance up to 50%, usually regardless of gender. Often late-onset diseases. Predictive testing in case of known gene and identified mutation possible. In severe disorders, often new mutations occur in a parental germ cell. In this case, parents are unaffected and recurrence risk to sibs is low (but 50% to own children)

Autosomal recessive

Mostly single patients, siblings affected, generally no further affected family members in other branches. Both parents are gene carriers (heterozygous). Risk to siblings 25%, risk to children usually low (<1%) except in a case of parental consanguinity. Clinical picture often severe and similar in siblings. Heterozygosity testing in case of known gene defects generally possible

X-linked recessive

Usually only boys/males affected, females only rarely and often milder. If mother is a carrier, recurrence risk to affected brothers is 50%, 50% of sisters are carriers. Risk to children of affected males: none, all daughters are gene carriers. About a third of cases of severe disease represent new mutations in the maternal germline. Genetic testing in a case of known genes is possible

Principles of genetic counseling First-degree relatives of affected individuals

No. of individuals

Normal population

Threshold

Affected individuals

Shift of curve of predisposition

Figure 46.1 Model for multifactorial inheritance of congenital heart defects. The liability to the malformation in the general population follows an approximately normal distribution, with individuals exceeding a certain threshold value being affected. First-degree relatives have a similar normal distribution of liability, but the curve is shifted to the right by the increased genetic component. Thus, a greater proportion will exceed the threshold and will be affected.

of the disorder, with a higher risk the more common is the disease; (3) the risk to offspring is approximately equivalent to the risk for sibs, unless there is evidence of a significant proportion of isolated cases being the result of new dominant mutations; (4) there is often an unequal sex incidence, and the risk is higher for relatives of a patient of the sex in which the condition is less common; (5) the risk may be greater when the disorder is more severe; (6) the risk is increased when multiple family members are affected.

Owing to the different etiologies of congenital heart defects and the individual reasons for asking for genetic counseling, the possible consequences for families might be completely different. Genetic counseling is in many aspects different from classical medical care in families with a certain disease. As a consequence, specific principles have been formulated. In 1975 a committee of the American Society of Human Genetics (ASHG) proposed a definition of genetic counseling that was subsequently adopted by the society: ‘Genetic counseling is a recommendation process which deals with the human problems associated with the occurrence or risk of recurrence of a genetic disorder in a family.’ Although the face of genetic counseling has changed continuously since 1975, the basic goals remain as they were then. The following basic principles have been generally accepted: 1. The decision to utilize genetics services should be entirely voluntary. Information should be made available and tests offered when appropriate, but patients and families should have the right to make decisions – particularly about genetic testing and reproduction – unencumbered by pressure or any intimation that a particular course is fiscally or socially irresponsible. 2. Ideally, genetic services, including counseling, diagnosis, and treatment, should be equally available to all who need and choose to use them. 3. A central feature of genetic counseling is a firm belief in the importance of patient education. Typical patient education regarding a particular disorder includes information about: (1) the features, natural history, and range of variability of the condition in question; (2) its genetic (or non-genetic) basis; (3) how it can be diagnosed and managed; (4) the chances that it can occur or

Genetic counseling in families with congenital heart defects

recur in various family members; (5) the economic, social, and psychological impacts – positive as well as negative – it may have; (6) resources that are available to help families deal with the challenges the disorder presents; and (7) strategies for amelioration or prevention that the family may wish to consider. 4. One of the principal issues regarded by geneticists is that all ‘relevant information’ should be disclosed. Now, however, with the complexity of genetic knowledge and technology, achieving this level of client education is often impracticable. Moreover, full disclosure of all ‘relevant information’ could overwhelm even the most sophisticated patient. It is critical for the counselor to disclose any information relevant to decision making in ways that the client can interpret and act on. 5. Although the counselor can use clinical judgment in choosing what information is most likely to be important and helpful in a client’s adjustment to a diagnosis or in decision making, it should be presented fairly, not with the purpose of encouraging a particular course of action.

Molecular genetics and genetic counseling Recent decades have seen a remarkable expansion and development of molecular genetics, affecting not just the field of inherited disorders, but the whole of medicine. Not only has molecular genetics given insights into many monogenic disorders whose basis was previously largely or incompletely unknown, it has transformed the way we view many disorders, especially those with a significant genetic component. Genetic counseling has been affected profoundly by these molecular advances. Not only have they resulted in valuable specific tests that allow carrier detection and prenatal diagnosis where these were previously unreliable or impossible, but also we are gaining insight into the basis of the variability in expression that is characteristic of so many genetic disorders, and which provides some of the most difficult problems of genetic counseling.

Diseases with localized genes which have not yet been identified After localizing a gene to a specific chromosomal region, the identification of the responsible gene is the next step in research. Since the identification of a gene nowadays often follows within a short period after its localization, the use of indirect genotype analysis has decreasing importance in genetic counseling. Because linkage studies are based on a knowledge of certain haplotypes without knowledge about the responsible mutation itself, this method has major

707

limitations. First, exact clinical diagnosis in an affected family member is necessary. It is a widespread misunderstanding that knowledge of the localization of a gene allows establishment of a diagnosis in a single patient. Linkage studies result in a conclusion about the risk status in persons belonging to a family where specific markers are known to segregate with a disease mutation. Second, genetic heterogeneity of a condition (different genetic entities which usually cannot phenotypically be distinguished in a single person) can lead to misinterpretation. An impressive example is hypertrophic cardiomyopathy, with very many different gene loci. Indirect genotype analysis leads to false results under the assumption of the wrong gene locus. In small families, a certain localization can neither be proven nor be excluded. Third, the family has to be informative. Depending on the individual marker constellation, there can be situations where it might not be possible to identify the at-risk haplotype in a family. With the use of multiple flanking and highly polymorphic genetic markers, the majority of families in most nephropathies with known gene loci are informative, allowing disclosure of the haplotype at risk. In case of a recombination between the analyzed flanking markers, however, a risk estimation can be impossible as well.

Conditions with known gene defects The situation in conditions where the gene defect is known differs in many aspects from those where only the chromosomal localization of the gene is known. The identification of the responsible mutation usually allows a definite diagnosis of the condition without further investigation or clinical examination. However, even in diseases where a gene is identified, specific problems can limit the practical application of gene testing: 1. The structure of many genes is extremely complex, and hence it is often difficult to identify the responsible mutation. In these diseases, DNA mutation analysis is far from being a routine diagnostic method. 2. Similar to the situation in indirect genotype analysis, the detection of a specific mutation can be impossible in a case of the existence of more than one gene. If a patient turns out to be negative upon mutation analysis of one gene, no diagnostic conclusions can be drawn if there are several gene loci. However, in patients where the mutation has been identified in a specific gene, the diagnosis can be clearly established, and other genetic loci or disease causes can be ruled out. 3. Another important question that is often forwarded to the medical geneticist is whether the identification of a mutation can predict the individual clinical course. There are diseases where a rough correlation between

708

Fetal Cardiology

Table 46.3 Selected syndromes with congenital heart disease in at least 50% of patients. Modified from reference 3 Disorder

Main extracardiac features

Heart defect

Inheritance

Del22q11 syndrome, DiGeorge syndrome, velocardiofacial syndrome

Short stature, cleft palate, nasal speech, thymus aplasia/hypoplasia, transient hypocalcemia, urogenital anomalies, mild mental retardation, slender fingers

Conotruncal defects, aortic arch anomalies, septal defects

Mostly new deletion of the region 22q11.2, 15–20% familial

Williams syndrome

Short stature, elfin face, mental retardation, renal anomalies, abnormal dentition

SVAS, pulmonary stenosis, stenosis of large vessels

Mostly new deletion of the region 7q11.23

Monosomy 1p36

Microbrachycephaly, large anterior fontanel, brachydactyly, urogenital anomalies, mental retardation

Septal defects, valvular anomalies, non-compaction cardiomyopathy

Mostly new deletion of the region 1p36

9q subtelomere deletion syndrome

Microbrachycephaly, arched eyebrows, mental retardation, urogenital anomalies, obesity

Septal defects, artery and valve stenosis, TOF

Mostly new deletion of the region 9q34.3

Microdeletion syndromes

Syndromes of known genetic basis, genetic testing possible Noonan syndrome

Short stature, pterygium colli, sternum deformities, cryptorchidism, facial dysmorphism

Pulmonary stenosis, hypertrophic cardiomyopathy

Heterogeneous, mostly AD new mutations (more than five known genes)

CFC syndrome/ Costello syndrome

C = cardio, F = facio, C = cutaneous manifestations, short stature, sparse hair, skin abnormalities

Pulmonary stenosis, septal defects

See Noonan syndrome

Alagille syndrome (arteriohepatic dysplasia)

Liver dysfunction (cholestasis), vertebral arch defects, renal dysfunction, typical face

Right-sided defects, peripheral pulmonary artery stenosis

AD, mostly new mutations (JAG1 gene)

CHARGE syndrome

C = coloboma, H = heart defects, A = choanal atresia, R = mental retardation, G = genital hypoplasia (males), E = ear anomalies, deafness

TOF, PDA, DORV, ASD, VSD

AD, mostly new mutations (CHD7 gene)

Smith–Lemli–Opitz syndrome

Short stature, failure to thrive, microcephaly, syndactyly of 2–3 toes, postaxial polydactyly, genital abnormalities (males), severe mental retardation

Endocardial cushion defects, hypoplastic left heart, septal defects

AR (biallelic mutations in the DHCR7 gene)

Kartagener syndrome

Bronchiectasis, sinusitis, infertility

Dextrocardia

AR, heterogeneous

Holt–Oram syndrome (heart–hand syndrome)

Upper limb defect

Septal defects, conduction defects

AD, heterogeneous (TBX5 gene)

Marfan syndrome

Increased height, arachnodactyly, ectopia lentis, skeletal manifestations, joint laxity, dural ectasia

Aortic dilatation, cardiac valve insufficiency

AD, heterogeneous (FBN1 gene), neonatal onset cases represent new mutations

Syndromes of unknown etiology Kabuki syndrome

Short stature, long palpebral fissures, large protruding ears, fingertip pads, mental retardation

Variable malformations with altered hemodynamics

Unknown, mostly sporadic

VATER/VACTERL association

V = vertebral anomalies, A = anal atresia, TE = tracheoesophageal fistula, R = radial limb defect/renal anomaly

Septal defects

Unknown, mostly sporadic, axial mesodermal dysplasia

SVAS, supravalvular aortic stenosis; TOF, tetralogy of Fallot; PDA, patent ductus arteriosus; DORV, double-outlet right ventricle; ASD, atrial septal defect; VSD, ventricular septal defect; AD, autosomal dominant; AR, autosomal recessive; XL, X-linked.

Genetic counseling in families with congenital heart defects

the type or localization of a mutation and the clinical phenotype exists; in many cases, however, an individual prediction is impossible since all family members share the same mutation. It is still unknown which other genetic or non-genetic factors may modify the severity of the disease. While the diagnosis of a disease-causing mutation does not depend on a clinical phenotype, genetic diagnoses may differ from conventional methods and may have different implications.

Prenatal diagnosis Prenatal prediction is in most cases limited to severe autosomal recessive or X-linked conditions. There is only exceptionally a request for prenatal diagnosis in autosomal dominant conditions. This observation can be made in nearly all autosomal dominant diseases (e.g. Huntington disease, tuberous sclerosis, neurofibromatosis, breast cancer). One reason might be the fact that always one parent is affected, and that the decision to terminate a pregnancy statistically in 50% of cases is obviously unacceptably high. The question of terminating a pregnancy which is connected with a prenatal diagnosis, when early treatment is not available, is always an important issue that has to be discussed in detail with the parents before they plan the pregnancy.

Predictive testing The principle is to predict whether a person in a family with a known inherited disease is most likely to be a gene carrier and will develop the disease in the future, although there are no clinical symptoms when tested. Although in many late-onset diseases no prevention is available, the knowledge of a person at risk of being a gene carrier can be of importance. Predictive testing differs from any conventional diagnosis and evaluation of clinical symptoms. Predictive testing has enormous consequences for persons at risk. Different aspects have to be discussed with the individuals who want to undergo predictive testing. Some of the arguments in favor of an early diagnosis are aspects of family planning, therapeutic possibilities, easing the burden of uncertainty, and planning occupation/finance. Insurance and occupational problems are only some of the aspects that have to be discussed intensively. Predictive testing in children is to be considered only when the diagnosis is of any benefit for the child, for example in diseases where a specific preclinical therapy is available. In the majority of diseases this is not the case, and a routine clinical examination is sufficient. As soon as there are clinical signs of the disease, the child can be regarded as a patient and should be taken under medical care.

709

In late-onset diseases, the decision for genetic testing should be made by the individual at risk. There is usually no reason for prenatal prediction of late-onset disease.

Heterozygosity testing In autosomal recessive conditions, heterozygosity testing (identification of healthy gene carrier) is applicable only in families with a known gene mutation or where specific genetic markers are known to segregate with the responsible gene defect. The situation is different in X-linked diseases where a genotype analysis can identify female carriers for an X-linked condition, which usually has important implications for family planning and possible prenatal diagnosis. The recurrence risk is 50% for males to be affected. The relevance of being heterozygous for a recessive disorder, which usually has no consequences for the gene carrier, is often misunderstood, and needs detailed information before testing. Knowledge of the heterozygous state can be important information for future family planning. Detection of the heterozygosity status in a relative of an affected person is therefore often of limited value, as long as no testing can be applied in a spouse. Prenatal testing in order to evaluate whether the fetus is heterozygous should not usually be performed.

Genetic counseling before applying genetic testing is essential According to several guidelines of the various national Societies for Human Genetics and Medical Boards, possible limitations of DNA analysis and its consequences should be discussed with the family.

Counseling in families with congenital heart defects The disclosure of the underlying basis of congenital heart defects is one of the major aims in cardiology. Although around 90% are not obviously familiar, it seems increasingly likely that important genetic factors are involved in most cases, and specific genes and chromosomal regions are starting to be identified. A current update on monogenic defects in human disease and congenital defects is given with the database of Victor McKusick’s Inheritance in Man (http://www.ncbi.nlm.nih.gov/Omim/searchomim. html).2 Thus, the first task in genetic counseling of families

710

Fetal Cardiology

Table 46.4 Frequency of heart disease in newborns with selected chromosome abnormalities. From reference 1 Chromosomal syndrome

Percentage

Frequent heart defects

Trisomy 18

99

VSD, pulmonary stenosis

Trisomy 13

90

VSD, dextrocardia

Trisomy 21

50

AV canal, VSD + ASD

4p- (Wolf–Hirschhorn) syndrome

40

ASD, VSD

20

VSD, ASD

5p- (cri du chat) syndrome X0 (Turner syndrome)

20–30

XXY (Klinefelter syndrome)

55

SVAS, ASD Mitral prolapse

AV, atrioventricular.

Table 46.5 Frequency of chromosomal aberrations in the presence of a congenital heart defect. From reference 4 Heart defect

Chromosome aberration (%)

AV canal

69

VSD + ASD

32

ASD

27

VSD

18

TOF

10

SVAS Valve stenosis (aortic or pulmonary)

6 4–5%

Hypoplastic left heart

4

Transposition of great vessels

0.9

Table 46.6 Overall risks in congenital heart disease. From reference 5 Heart defect

Risk (%)

Population incidence

0.5

Sibs of isolated case

2–3

Half-sibs or second-degree relative

1–2

Offspring of isolated case father affected

2–3

mother affected

5–6

Two affected sibs or sib and parent affected

10

More than two affected first-degree relatives

∼50

is to ensure that a Mendelian lesion has been deemed unlikely, particularly if abnormalities additional to the cardiac lesion are present. Table 46.3 lists selected syndromes of different etiologies with congenital heart disease in at least 50% of patients. Congenital heart disease is also prominent in chromosomal disorders, particularly the autosomal trisomies (Tables 46.4 and 46.5). Among the identified enviromental causes, rubella is still the most important, but congenital heart defects are produced by almost all the less specific teratogens and enviromental factors, which should be carefully enquired for, even though it may not be possible to prove cause and effect in an individual case. Lithium is specifically associated with Ebstein’s anomaly. The offspring of diabetic women also appear to be a high-risk group. Monozygous twinning is itself a risk factor for congenital heart disease, in that twins have an increased risk (around 1.5%). Genetic advice is most frequently sought for future sibs of an affected child, or sometimes for more distant relatives. However, risks other than for first-degree relatives are low in the absence of multiple cases or an identified Mendelian basis. Information is now becoming available for the offspring of affected individuals, and there is increasing evidence that risks are higher for offspring of affected females than of males. Overall risks are summarized in Table 46.6, but whenever possible, a specific anatomical diagnosis should be used as the basis for risk estimates. When recurrence happens, the defect is the same as previously in only about half the cases (i.e. clinical concordance). That is, a discordance of the site and severity of heart defects in families with more than one affected individual can be seen in a similar proportion of cases. This is relevant to counseling because it may mean that a sib of a proband with a correctable defect may have a fatal or untreatable lesion, or vice versa.

Genetic counseling in families with congenital heart defects

The most common discordant lesion in humans is ventricular septal defect (VSD), followed by pulmonary stenosis, aortic stenosis, transposition of the great arteries, and tricuspid atresia.

Risks to more distant relatives Data are inadequate, but the excess risk for second-degree relatives of an isolated case of congenital heart disease is certainly under 1%, and it is doubtful whether third-degree relatives have a significantly increased risk. Families where there are several affected members, none of whom is a first-degree relative, are not infrequently encountered. The possibility of a variable Mendelian form should be seriously considered here.

711

includes multiple system involvement, and is only rarely limited to cardiovascular defects. It is clearly understood that the counseling must relate to the karyotype. Since most chromosome anomalies are non-hereditary, the recurrence risk to sibs is small, if familial translocations or inherited structural defects can be excluded. A considerable number of conditions are caused by small chromosomal deletions that often escape conventional cytogenetic analysis. By means of fluorescent in situ hybridization (FISH) or other molecular cytogenetic methods, many of these small deletions can be made visible and have important implications for genetic counseling. If the same microdeletion can be detected in a parent or a relative, the recurrence risk to a sib is 50%, with a broad spectrum of phenotypic manifestations in some of these syndromes. Congenital heart disease is a leading feature in velocardiofacial syndrome/DiGeorge syndrome and Williams–Beuren syndrome.

Multiple cases Family clusters of congenital heart defects are not uncommon; their occurrence should prompt a careful search for a Mendelian or chromosomal syndrome. After two affected children the risk of congenital heart disease in future sibs is approximately tripled, regardless of whether the affected individuals have the same heart defect or not. This gives risks ranging from 5% for the rare defects to 10% for a common abnormality such as ventricular septal defect. A similar risk would be likely for future children where an affected parent has an affected child, although data to confirm this are not yet available. Numbers are insufficient to give individual estimates for specific defects. The occurrence of more distant affected relatives does not increase the risks given in Table 46.6. In exceptional families with more than two affected first-degree relatives, risks are likely to approach 50%.

Velocardiofacial syndrome/ DiGeorge syndrome Patients with velocardiofacial syndrome exhibit a broad spectrum of phenotypic abnormalities. Most important features are short stature, intellectual impairment, velopharyngeal incompetence, and cardiac defects (conotruncal defects, aortic arch anomalies, septal defects). The condition is autosomal dominantly inherited, with most patients representing new mutations on the basis of a microdeletion of chromosome 22q11. Inherited deletions also occur in about 15–20% of patients. The corresponding deletions are only rarely visible by conventional cytogenetic analysis.

Williams–Beuren syndrome Chromosomal disorders and microdeletion syndromes Heart defects are important manifestations of chromosomal aberrations. The most common lesions are ventricular and atrial septal defects, patent ductus, and pulmonary stenosis. Among fetuses with a prenatally detected congenital heart defect and/or intrauterine growth restriction, a chromosomal disorder can be detected in up to 40% of cases.6 The frequency of congenital heart defects in chromosomal aberrations ranges from near 100% (trisomy 18) to minimal (Table 46.4). While the heart lesion seldom points specifically toward a chromosome abnormality, endocardial cushion defects are frequently seen in Down syndrome, and Turner syndrome is predominantly associated with coarctation of the aorta. A chromosomal disorder normally

This syndrome is characterized by facial dysmorphism (elfin face), hypercalcemia, congenital heart defect (supravalvular aortic stenosis, pulmonary valvular stenosis, peripheral pulmonary artery stenosis, and ventricular and atrial septal defects), and mental retardation. Most affected individuals represent sporadic cases. The syndrome is caused by a small deletion including the elastin gene on chromosome 7q11. A rare mutation in the elastin gene causes isolated supravalvular aortic stenosis and cutis laxa.

Specific anomalies In the following, the genetic basis of the most common heart malformations with relevance to prenatal diagnosis will be discussed in more detail.

712

Fetal Cardiology

Atrioventricular septal defects While the majority are multifactorially inherited (recurrence risk about 2–3% to first-degree relatives), families following autosomal dominant inheritance are known. Atrioventricular septal defect (AVSD) exhibits locus heterogeneity. Linkage analysis has established a locus on 1p31–p21 (AVSD1), and deletion mapping has identified a locus on 3p25 (AVSD2) which has been found to be due to mutations in the CRELD1 gene. AVSD may also be caused by a mutation in the GJA1 gene on chromosome 6q21–q23.2. AVSD is classically associated with Down syndrome, and is seen in other syndromes such as Holt–Oram syndrome and atrial septal defect (ASD) with atrioventricular conduction defects and in rare cases of familial Noonan syndrome. These syndromes are too rare to affect the general recurrence risks, but should be considered in patients with extracardiac manifestations. Concordance in sibs and first-degree relatives of patients with ASD is about 50% in combined series.

Ventricular septal defects The recurrence risks of 3% to sibs and 2–9% to offspring apply to severe ventricular septal defects, mostly those in patients requiring surgery. It is doubtful whether the risks are as high for relatives of patients with asymptomatic or transient defects. The risk for offspring seems to be increased in those cases where the mother is affected. The concordance rate with VSD recurring in first-degree relatives varies from 30 to 60%. Tetralogy of Fallot (TOF) is the most common discordant lesion, indicating that VSD can be regarded as a forme fruste of TOF in these families. TOF can be caused by mutations in the human homolog of rat Jagged-1 (JAG1), or in the gene encoding the cardiac-specific homeobox Nkx2.5 (CSX). There is also a wellrecognized association with 22q11 microdeletion and trisomy 21. Mutations in the ZFPM2 gene have been identified in sporadic cases of TOF, as have mutations in the GDF1 gene.

ASD. There is evidence for a small fraction following autosomal recessive inheritance. In practice, a recessive mode of inheritance should be considered in families with two affected children.

Endocardial cushion defect The interpretation of figures for recurrence risks in families with endocardial cushion defects is difficult. The sib recurrence risk in a larger series of studies is 2.5%; those of children from affected sibs differ greatly. The offspring risk is given as 14% for children of affected mothers and about 1% for those of affected fathers. Concordance is about 90% in affected families, which is very high.

Pulmonary stenosis It is estimated that about 10% of patients with pulmonary stenosis have Noonan syndrome. Pulmonary stenosis can also be found in related diseases such as LEOPARD syndrome or neurofibromatosis. If Noonan syndrome is present in patients with pulmonary stenosis, the risk of Noonan syndrome in offspring is 50%, and the risk for some form of cardiovascular disease is about 25%. Pulmonary or aortic stenosis generally have a recurrence risk of 2–5% to first-degree relatives.

References 1.

2. 3.

4.

Hypoplastic left heart syndrome The overall recurrence risk of about of 3.2% is higher than predicted for multifactorial inheritance alone. Hypoplastic left heart syndrome (HLHS) can be caused by a mutation in the GJA1 gene. Only about 25% of recurrences are HLHS, the remaining including a variety of anomalies, VSD, and

5. 6.

Nora JJ, Berg K, Nora AH. Cardiovascular Diseases. Genetics, Epidemiology and Prevention. New York: Oxford University Press, 1991. McKusick’s Online Mendelian Inheritance in Man: http:// www.ncbi.nlm.nih.gov/Omim/ searchomim.html. Jones KL, Smith DW. Smith’s Recognizable Patterns of Human Malformation, 5th edn. Philadelphia: WB Saunders Company, 2004. Pradat P, Francannet C, Harris JA, Robert E. The epidemiology of cardiovascular defects, part I: a study based on data from three large registries of congenital malformations. Pediatr Cardiol 2003; 24: 195–221. Harper PS. Practical Genetic Counselling 5th edn. Oxford: Butterworth-Heinemann, 1998. Schwanitz G, Zerres K, Gembruch U et al. Prenatal detection of heart defects as an indication for chromosome analysis. Ann Genet 1990; 33: 79–93.

47 Intrapartum evaluation of fetal well-being Yoram Sorokin and Sean C Blackwell

Introduction It has long been known that labor is a risk factor for fetal mortality and for neonatal morbidity and mortality. Through research carried out in the 1950s, 1960s, and early 1970s, obstetricians obtained a better understanding of fetal respiratory physiology and human fetal physiology in response to the labor process. It provided a basis for diagnostic techniques to detect possible fetal well-being and compromise. It was appreciated that clinical management could change fetal conditions. In 1961, Saling introduced intermittent scalp pH measurement as the first technique for direct assessment of fetal well-being during labor.1 Fetal heart rate (FHR) monitoring technologies were developed in the 1950s and 1960s by Hammacher et al,2 Hon and Quilligan,3 Caldeyro-Barcia et al,4 and others. By the late 1960s and early 1970s, equipment for intrapartum fetal evaluation was commercially available. In the 1970s, obstetricians had very optimistic expectations that, with intrapartum surveillance (utilizing continuous FHR monitoring and intermittent fetal scalp pH determinations), intrapartum stillbirths and neonatal neurological injuries caused by intrapartum hypoxia could be significantly reduced or eliminated. The hope was that with continuous electronic FHR monitoring, ‘early asphyxia’ would be recognized; through timely obstetrical intervention, asphyxia-induced brain damage, or neonatal death, would be avoided. Continuous electronic fetal monitoring (EFM) was introduced into widespread clinical practice before evidence from randomized clinical trials demonstrated either efficacy or safety. In the 1970s and 1980s, continuous electronic FHR monitoring became routine in most hospitals in the United States and the Western world. During the last 30 years, thousands of articles have been written on this topic. Initial retrospective studies evaluated 135 000 patients and showed a more than three-fold improvement in the intrapartum fetal death rate for the electronically monitored group versus the control group with intermittent auscultation (IA).5,6 Many randomized trials were performed comparing the efficacy and safety of

routine continuous electronic fetal monitoring (EFM) with intermittent auscultation (IA) for intrapartum surveillance. Recently, there has been increasing utilization of central monitoring systems and computerized recognition and interpretation of fetal heart rate patterns.7–11

Fetal heart rate patterns Over the past 40 years, many scientific articles have been written concerning the definitions of FHR patterns and recommendations for their interpretation. For many years there was confusion about terminology and definitions used to describe and interpret FHR patterns. In 1997, recommendations for research guidelines for the interpretation of fetal heart rate were published.12 The recommendations resulted from a National Institute of Child Health and Human Development (NICHD) Research Planning Workshop that had the purpose to assess the research status of this area and publish research recommendations.12 This consensus report suggested standardized and unambiguous definitions of FHR patterns for the purposes of improving research studies on the reliability and validity of FHR interpretation as well as studies of the relationship between FHR patterns and outcome. Four major parameters of FHR were defined: baseline rate, baseline FHR variability, accelerations, and decelerations (variable, early, late, and prolonged). The most recent American College of Obstetricians and Gynecologists (ACOG) Practice Bulletin published in 2005, entitled ‘Intrapartum Fetal Heart Rate Monitoring’,13 included the NICHD Research Planning Workshop FHR pattern definitions and descriptions.12

Baseline FHR baseline is the mean FHR over a given 10-minute period (rounded to the nearest 5 beats/min). The normal range is between 110 and 160 beats/minute. Bradycardia is

714

Fetal Cardiology

defined as a decrease in FHR baseline below 110 beats/ minute for 10 minutes or longer. Mild degrees of bradycardia may occur in the second stage of labor, and often immediately before birth. The fetus is generally able to tolerate bradycardia by compensating with an increased stroke volume. However, this ability to compensate with increased stroke volume in response to bradycardia breaks down at severe decreases in FHR, below 60 beats/minute. Catastrophic events associated with bradycardia include umbilical cord prolapse, umbilical cord occlusion, or uterine rupture. Congenital heart block due to the presence of anti-Ro and/or anti-La antibodies from maternal collagen vascular disease may be another uncommon cause.14 Tachycardia, a FHR baseline > 160 beats/min for 10 minutes or longer, may occur due to maternal pyrexia, medications (β-sympathomimetics, cocaine), chorioamnionitis, fetal hypoxia, fetal anemia, or tachyarrythmias. In tachycardia there is increased sympathetic and/or decreased parasympathetic autonomic tone, which is associated with decreased FHR variability. Chronically instrumented fetal sheep exposed to umbilical cord occlusions for 1 minute every 2.5 minutes develop a fall in nadir of FHR decelerations and a rise in interocclusion fetal heart rate, or tachycardia, due to increased catecholamine activity.15,16 Tachycardia, in cases of intrapartum acidemia, does not appear in isolation. If tachycardia is seen with normal FHR variability and no periodic changes, it should be assumed to be due to other causes.

hypoxemia; FHR variability increased when the fetus became mildly or moderately hypoxemic.17 The increases in FHR variability during mild and moderate hypoxemia were abolished by atropine. Propranolol (beta-blocker) had no effect on FHR variability in cases of either mild or moderate hypoxemia. A recent study demonstrated that the initial response to experimental acute progressive asphyxia typically included an immediate, transient, increase in FHR variability.18 However, the terminal fetal compromise with profound acidemia and hypotension were accompanied by an increase in FHR variability in some fetuses and a decrease in others.18 The importance of FHR variability in evaluating fetal well-being cannot be underestimated, for it is a sensitive predictor of fetal acid–base status.6,19–21 Several experts believe that with respect to pattern interpretation, there is an overemphasis on the presence of decelerations, and an underappreciation of FHR variability.6,19–21 Absent variability may be associated with severe fetal acidemia and/or previous fetal central nervous system (CNS) injury. Decreased variability can be caused by fetal acidemia, but also by various medications (e.g. CNS depressants). Fetal quiet sleep epochs are associated with the absence of accelerations and decreased variability, but these episodes at term often last approximately 20 minutes and only rarely longer than 60 minutes.12

Accelerations Variability Baseline FHR variability is a fluctuation in baseline FHR of at least two cycles per minute. It represents the balance between sympathetic and parasympathetic systems and is influenced by gestational age. One of the modifications found in the NICHD workshop recommendation is that FHR variability should not be divided into ‘short-term’ and ‘long-term’ components, but should be evaluated in toto.12 The fluctuations in baseline FHR are visually quantitated as the amplitude of the peak-to-trough in beats per minute. Grades of fluctuation are subdivided into: undetectable, minimal (≤ 5 beats/min), moderate (6–25 beats/ min), and marked (> 25 beats/min). This definition is adequate for clinical visual interpretations, although in fact two characteristics of FHR variability are recognized. Short-term variability, beat-to-beat changes of FHR, may exist independently of long-term variability (continuous unidirectional change in baseline FHR over a period of time), but long-term variability cannot exist independently of short-term variability. Under normoxemic conditions, cardiac vagal blockade (atropine) causes an increase in FHR baseline and a decrease in FHR variability. In studies that induced acute mild and moderate fetal hypoxemia, but no acidosis, baseline heart rate did not change significantly with mild hypoxemia, but decreased with moderate

A FHR acceleration is defined as an abrupt (peak within 30 seconds) increase in FHR above the previously calculated baseline by at least 15 beats/minute that lasts at least 15 seconds, and less than 2 minutes from the onset to return to FHR baseline. The NICHD Research Planning Workshop recommends that before 32 weeks, the criteria for FHR acceleration should be different: the acme must be at least 10 beats/minute over the baseline and lasting at least 10 seconds, but less than 2 minutes.12 Accelerations are often associated with fetal movements and/or uterine contractions. The presence of FHR accelerations (whether spontaneous or induced) is considered a reassuring sign of fetal well-being and indicates that the fetus is not acidotic (pH < 7.20).22–24 Intrapartum stimulation tests induce FHR accelerations. They appear to be useful to rule out fetal acidemia in the setting of non-reassuring FHR pattern.25

Decelerations There are three basic types of FHR decelerations: late, variable, and early. A late deceleration is a gradual but shallow decrease from the baseline FHR that is associated with a uterine contraction. The deceleration is delayed in timing with onset, nadir, and recovery of the FHR deceleration

Intrapartum evaluation of fetal well-being

occurring after the beginning, peak, and ending of the contraction. Late decelerations are associated with uteroplacental insufficiency and can be divided into two categories.26–28 A reflex late deceleration is seen with normal FHR variability, thus signifying normal CNS integrity; it occurs due to transient hypoxia associated with uterine contractions; it is caused by a vagal reflex. The second type, a non-reflex late deceleration, is caused by direct myocardial hypoxic depression (or failure), as well as vagal activity; it signifies a risk of ‘central asphyxia.’ These are seen with decreased or absent FHR variability and are much more ominous. If late decelerations occur in coordination with > 50% uterine contractions, they are termed recurrent. Most cases of late decelerations reflect reduced fetal reserve rather than myocardial hypoxia or acidosis.15 Late decelerations are of value in the identification of fetuses who are at risk of hypoxia, when the late decelerations appear with additional findings, such as reduced FHR variability. The incidence of late deceleration is low. A variable deceleration is a visually apparent abrupt decrease in FHR below the baseline, of at least > 15 beats/ minute, lasting > 15 seconds, but less than 2 minutes. These decelerations are termed variable because they are variable in shape, depth, duration, and onset relative to uterine contractions. Prior-to-labor conditions associated with variable decelerations include decreased amniotic fluid volume, nuchal cord, knotted cord, or a body cord. Variable decelerations are not associated with fetal compromise per se; however, during the deceleration there may be transient hypoxemia. Although not caused by uteroplacental insufficiency, if severe, variable decelerations (e.g. FHR decrease to < 60 beats/minute, lasting > 60 seconds) that become repetitive may lead to fetal hypoxia and acidemia. However, this does not usually occur without noticeable change in other aspects of the FHR pattern, including decreased variability and increased FHR baseline. Most FHR decelerations that are related to labor are variable. Variable decelerations are vagally mediated; umbilical cord occlusion, dural stimulation during fetal head compression, increased intracranial pressure, and reduced cerebral perfusion are likely causes of variable FHR decelerations.15,29 The vagal response may be due to either chemoreceptor or baroreceptor input.15,29 The cord occlusion mechanism occurs mostly in the first stage of labor;15,29 other mechanisms occur mostly in the second stage.15,29 A recent review summarized our understanding of the pathophysiological mechanisms of fetal responses to hypoxia, based on recent experimental studies in chronically instrumented near-term fetal sheep in utero, and described the pathophysiology of intrapartum FHR decelerations.15 Asphyxia was produced in fetal sheep by repeated complete occlusion of the umbilical cord for 1 minute.15,30,31 Comparison was performed between two groups of fetuses, concerning the effect of umbilical cord occlusions every 5 minutes, consistent with early labor,

715

and occlusions every 2.5 minutes, consistent with late first stage and second stage labor.15 Once deep decelerations are present no deceleration pattern is necessarily benign. Fetuses with normal placental reserve had a remarkable ability to adapt to repeated hypoxia; when subjected to deep and brief decelerations they could compensate fully for surprisingly prolonged intervals before development of profound acidosis and hypotension.15 The adaptive ability of the term human fetus is illustrated by the consistent finding that neonatal complications and the need for resuscitation are extremely uncommon below an acute base deficit of 10–12 mmol/l.32 However, even in healthy sheep fetuses, prolonged series of brief variable decelerations, if repeated sufficiently frequently, led ultimately to severe repeated hypotension and profound metabolic acidosis.15 Fetuses with preexisting hypoxia were vulnerable even to relatively infrequent periods of additional hypoxia in early labor.15 Changes in FHR associated with such deterioration, during repeat deep decelerations, develop progressively and surprisingly slowly, even during frequent occlusions. When the fetuses deteriorated, the sequence of events included increasing amplitude of deceleration, more rapid rate of initial deceleration, a rising baseline, initial increase, and then loss of baseline variability, and finally brief overshoot immediately after the deceleration.15 The authors recommended simplifying terminology, and educating clinicians in the physiological mechanisms of FHR decelerations and the patterns of FHR change that indicate progressive loss of fetal compensation.15 Early decelerations are defined as visually apparent gradual decreases in baseline associated with uterine contractions. They are coincident in timing with the nadir of the deceleration occurring at the same time as the peak of the contraction. They are thought to be due to pressure on the fetal head as it moves down the birth canal. The mechanism is one of reflex, slowing of FHR mediated via the vagus nerve. They are innocuous and are not associated with fetal compromise. They may be confused with late decelerations given their similar shallow appearance, but their timing in relation to contraction is critical. The presence of FHR accelerations, normal FHR variability, and cervical dilatation of 4–5 cm are other features consistent with early decelerations.

Sinusoidal pattern Sinusoidal FHR pattern consists of smooth, sine wave-like regular oscillations of the baseline FHR. The pattern lasts at least 10 minutes and has a relatively fixed period of 3–5/ minute and an amplitude of 5–15 beats/minute fluctuating above and below the baseline. There is an absence of shortterm variability or FHR accelerations.6,12,13,20,33 This pattern is always non-reactive, and is strongly associated with poor outcome; the association is with fetal hypoxia, often resulting from severe fetal anemia, as in Rh isoimmunization

716

Fetal Cardiology

or fetal–maternal hemorrhage. A derangement of nervous control of the heart, secondary to central or peripheral ischemia, has been hypothesized to result in sinusoidal heart rate. The pattern was also described in some cases of normal infants born without depression or acid–base abnormalities.34 Such patterns, also called pseudosinusoidal, have also been reported following intrapartum administration of the analgesics alphaprodine (Nisentil®), butorphanol (Stadol®), or meperidine (Demerol®), and in association with amnionitis.6 The pseudosinusoidal pattern has increased cyclicity and presence of short-term variability; FHR is reactive before and after the episode, and the pattern is benign. True sinusoidal patterns are quite rare, and are non-reassuring.

Fetal heart rate interpretation and management The interpretation of tests of intrapartum fetal well-being is performed with the context of the entire obstetrical situation; it should include maternal and fetal factors (i.e. the specific underlying pathology), as well as the course, anticipated duration, and outcome of labor.6,12,20,35 For instance, the effect of gestational age of the fetus (the difference in fetal CNS maturation at 24 compared to 38 weeks), maternal

medical disorder (e.g. hypertension or diabetes), intrinsic fetal pathology (e.g. growth restriction, fetal anemia, fetal infection), and prior results of fetal assessment are several important factors that impact on the clinical significance of different FHR patterns.6,12,20,35 It is obvious that with the number of variables one has to consider, and the often imprecise nature of the information, the medical decision can be difficult. The obstetrical community has not reached a broad consensus on a standardized approach to assessment and management of most FHR monitoring patterns. Sometimes the interpretation of the pattern is clear. When the FHR pattern is normal and reassuring, no intervention is necessary. In some non-reassuring patterns the need for intervention is clear. More often, however, the observation is a mixture of reassuring and non-reassuring patterns and the management is not so simple. Several studies suggest that FHR interpretation is plagued by poor inter- and intraobserver reliability.36–40 There are several patterns and combination of patterns for which there exists widespread agreement.12 There is greater agreement in the interpretation of FHR tracings if the tracing is reassuring.41 A pattern with normal FHR baseline, presence of FHR accelerations, normal (moderate) FHR variability, and absence of FHR decelerations confirms an extremely high predictability of a normally oxygenated fetus with normal acid–base status (Figure 47.1). This is considered a reassuring

Figure 47.1 Normal fetal heart rate (FHR) tracing: normal baseline, spontaneous accelerations, no decelerations, and normal variability.

Intrapartum evaluation of fetal well-being

pattern.12 When this pattern is obtained it is nearly always associated with a newborn who is vigorous at birth. Abnormal patterns clearly associated with fetal acidemia are those with absent variability, recurrent late or severe variable decelerations, or sustained bradycardia (Figure 47.2). These patterns are consistent with hypoxia that is predictive of current or impending fetal asphyxia, so severe that the fetus is at risk for neurological and other fetal damage or death.12 A fetus with absent FHR variability and severe variable or late decelerations is at significantly high risk to have profound acidosis, and expedited delivery is indicated. A fetus with a prolonged FHR deceleration or bradycardia, for example less than 60 beats/minute, with absent variability is also at significantly increased risk to have acidemia, and expedited delivery is also indicated. There are many FHR patterns that lie between the very reassuring patterns and those considered ominous by the NICHD Workshop.12 In intermediate FHR patterns there is no consensus concerning clinical management. Fetuses have, at various times during labor, patterns, and combinations of patterns, that have some non-reassuring characteristics, including repetitive late decelerations with accelerations and normal variability, variable decelerations with slow return to baseline or late component, absent variability with no decelerations associated with contractions, fetal tachycardia with decreased variability and without decelerations, blunted patterns, and checkmark patterns.41 Non-reassuring patterns are non-specific, and

717

cannot reliably predict whether a fetus is well oxygenated, depressed, or acidotic. One of the problems with non-reassuring FHR patterns is that factors other than hypoxia, such as intrauterine infection, fetal sleep states, congenital abnormality (developmental or acquired), or drugs, may lead to a non-reassuring FHR pattern. Non-reassuring FHR patterns that are associated with hypoxia do not depict the severity of the hypoxia, and it is difficult to predict the progress of the hypoxia if labor continues (Figure 47.3). An internal fetal scalp electrode (and/or intrauterine pressure catheter) could be placed when there is a need for a more accurate look at FHR variability, or a need for better correlation of the timing of deceleration with uterine contractions. It is clear that many fetuses with various types of decelerations will not be born with mixed or metabolic acidosis. Clinical observations and retrospective surveys of outcome following monitoring suggest that the presence of FHR variability can help in specificity for interpretation of the vigorous, well-oxygenated fetus.2,19,20,42,43 The presence of normal FHR variability is almost invariably associated with a vigorous well-oxygenated neonate at birth (Figure 47.4). Moderate levels of FHR variability are a strong indication that the fetus is coping well with labor and is unlikely to have significant acidosis. A recent study demonstrated that the most significant intrapartum FHR parameter to predict the development of significant acidemia is the presence of minimal/absent variability for at least 1 hour,

Figure 47.2 Abnormal FHR tracing: normal baseline, no accelerations, and decreased variability. The patient presented to hospital complaining of decreased fetal movements, and underwent emergent cesarean delivery after 15 minutes of arrival. The neonate was depressed at birth, but had a normal umbilical artery pH. Within 6 hours of life, the neonate began having generalized tonic–clonic seizures. This pattern and case is illustrative of the possibility that the fetus may be neurologically injured prior to labor and recover to a normal acid–base status, but still manifest a significantly abnormal FHR pattern.

718

Fetal Cardiology

Figure 47.3 Abnormal FHR tracing: tachycardia, repetitive late decelerations, and decreased variability. The uterine contraction pattern is consistent with excessive contractions due to recent maternal cocaine use and placental abruption. After approximately 1 hour after this section of the tracing, which did not improve despite resuscitative efforts, a cesarean section was performed for ‘nonreassuring fetal status remote from delivery’. The umbilical artery pH was 7.26 and the neonate had no problems after birth.

Figure 47.4 Abnormal FHR tracing: normal baseline, repetitive severe variable decelerations, and normal variability. The patient had spontaneous vaginal delivery approximately 20 minutes after this section of tracing. The neonate was vigorous at birth with normal Apgar scores and normal umbilical artery pH values. Despite the recurrent severe variable decelerations until delivery, maintenance of normal FHR variability during this time period indicated normal fetal acid–base status and was predictive of the favorable outcome.

as a solitary abnormal finding or in conjunction with late decelerations in the absence of accel erations.44 A recent review suggests that moderate FHR variability was strongly associated (98%) with an umbilical pH > 7.15 or newborn

vigor (5-minute Apgar score ≥ 7).45 Undetectable or minimal FHR variability, in the presence of late or variable decelerations, was the most consistent predictor of newborn acidemia, but the association was only 23%.45

Intrapartum evaluation of fetal well-being

Hypoxia usually results in decelerations first. While the type of deceleration usually helps in the understanding of the cause of hypoxia, the persistence of some decelerations and the duration and depth of variable and prolonged FHR decelerations helps in determination of the severity of hypoxia. However, the evolution of patterns is very important. A pattern of persistence of late decelerations that cannot be reversed, followed by loss of reactivity and loss of variability, usually reflects developing acidosis from hypoxia. The management of the FHR deceleration is dependent on the type of deceleration as well as the presence or absence of FHR variability and/or FHR accelerations. However, a fetus with normal FHR variability that has pattern evolution to reduced variability is extremely unlikely to have intrapartum hypoxia unless decelerations precede or are simultaneous with this change. Thus, in this setting, the FHR variability change is most likely due to fetal sleep cycle or medication, and aggressive management may not be warranted. The fetus generally tolerates decelerations or bradycardias that are not < 80 beats/minute, because of compensatory interactions that maintain cardiac output. However, at severely reduced heart rates, particularly < 60 beats/ minute, it is likely that cardiac output, and hence umbilical blood flow, cannot be maintained.29 Correction of the deceleration, or bradycardia, is dependnt on the etiology. For repetitive severe variable decelerations due to cord occlusion, maternal positional changes, or intrauterine amnioinfusion, may be required. For repetitive late decelerations or bradycardia due to uteroplacental insufficiency, interventions such as maternal oxygen therapy, intravenous hydration, positional changes, or medications to improve maternal blood pressure may be used. As mentioned above, other insults may occur in labor, including infection, fetal hemorrhage, thrombosis, and anemia, which may alter fetal physiology and thus impact on fetal response to the causes of FHR abnormalities. In addition to hypoxia, any insult, or physiological variant, that causes neurologic depression usually results in decreased FHR variability and elimination of reactivity. Several professional societies have issued documents on use of EFM that include guidelines for interpretation, assessment, and management.13,46–48 Current ACOG recommendations include some form of fetal monitoring for all women in labor.13 Since most clinical trials excluded subjects at high risk for adverse outcomes, the relative safety of IA in such cases is uncertain. The labor of parturients with high-risk conditions (e.g. suspected fetal growth restriction, preeclampsia, and type 1 diabetes) should be monitored continuously.13 For other patients the choice of technique (EFM or IA) is based on a variety of factors and is left to the judgment of the woman and her clinician. When EFM is used during labor, in patients without complications, the ACOG recommendation is to review the FHR tracing approximately every 30 minutes in the first

719

stage of labor and every 15 minutes in the second stage. In patients with complications (e.g. fetal growth restriction, preeclampsia) the corresponding frequency is approximately every 15 minutes in the first stage of labor and every 5 minutes during the second stage. There are no comparative data for optimal frequency for IA in the absence of risk factors. One method is to evaluate the FHR at least every 15 minutes in the active phase of the first stage of labor and at least every 5 minutes in the second stage.13 The use of EFM is nearly universal, estimated at 80% in the USA. Intermittent auscultation requires a 1:1 nurse/ patient ratio. Logistically, it may be difficult to adhere to ACOG guidelines for IA, for this reason. A prospective study found that the protocol of IA was successfully completed in only 3% of cases.49 EFM is easier, is cheaper, and provides more data. Economic considerations have been suggested as a motivating factor for hospitals to use EFM and not IA. Many providers still believe, despite data from randomized trials, that EFM is better than IA, in part because much more data are provided with EFM. A persistently non-reassuring FHR tracing requires evaluation for possible causes,13 including hypotension, response to medicine or drug, maternal position, cord prolapse, maternal fever, placental abruption, or umbilical cord occlusion. Initial assessment and treatment should include attempts to correct the problem and/or improve fetal oxygenation, for example discontinuation or reduction of oxytocin, change in maternal position, treatment with intravenous ephedrine for correction of hypotension secondary to regional anesthesia, or amnioinfusion.13,50 A recent Cochrane Database Systematic Review found that there is not enough evidence to support the use of prophylactic oxygen therapy for women in labor, or to evaluate its effectiveness for non-reassuring fetal status.13,51 β-Mimetic therapy appears to be able to reduce the number of fetal heart rate abnormalities and perhaps reduce uterine activity. However, there is not enough evidence based on clinically important outcomes to evaluate the use of β-mimetics for suspected fetal distress.13,52 If the non-reassuring pattern is not improved, ancillary tests may be performed, and determination should be made concerning the need for surgical intervention and the urgency of the surgical intervention.

Ancillary tests to electronic fetal monitoring Electronic FHR monitoring has a very high false-positive rate. Several ancillary tests may help to ensure fetal wellbeing when faced with a non-reassuring FHR tracing. These ancillary tests include: fetal scalp stimulation, fetal vibroacoustic stimulation, and fetal scalp pH assessment. The use of scalp puncture, Allis clamp, and vibroacoustic and digital stimulation to provoke FHR accelerations, and

720

Fetal Cardiology

the use of intermittent fetal scalp blood sampling for pH, can help in fetal assessment during labor.24,25 In at least 50% of patients with non-reassuring patterns FHR accelerations will be elicited in response to stimulation. The presence of accelerations (spontaneous or induced) correlates with normal fetal pH, and reduces the need for further assessment, such as fetal scalp sampling for fetal pH assessment. The presence of a response (FHR acceleration) correlates with a normal scalp pH (> 7.20). A meta-analysis of 11 studies of four techniques of intrapartum fetal stimulation showed that each test was a reliable method to exclude acidosis if accelerations were noted after stimulation.25 Digital scalp and vibroacoustic stimulation tests are less invasive, and preferred.13 Digital scalp stimulation is the easiest to use routinely because no device is required and rupture of membranes is not necessary. When there is a non-reassuring pattern with lack of FHR acceleration, fetal stimulation may be performed. When there is a FHR acceleration after stimulation, acidosis is unlikely. If the fetus responds with FHR acceleration but the non-reassuring FHR pattern continues, repeat stimulation testing is necessary. If the fetus does not respond with a FHR acceleration, or the non-reassuring pattern persists, a fetal scalp sample may be obtained for acid–base determination, for further reassurance. Fetal scalp blood sampling, for determination of pH, is useful and should be considered when there is a non-reassuring FHR pattern and failure to elicit a FHR acceleration. The use of fetal scalp sampling for fetal scalp pH may result in fewer cesarean deliveries for the indication of non-reassuring fetal status. The use of scalp sampling has decreased, especially in the USA, and is not available in many hospitals. Fetal scalp sampling has poor sensitivity (35%) and poor positive predictive value (PPV) (9%) for predicting umbilical arterial pH < 7, and poor sensitivity (50%) and poor PPV (3%) for identifying newborns with hypoxic–ischemic encephalopathy.53

Fetal pulse oximetry Fetal pulse oximetry was approved recently by the Food and Drug Administration (FDA). After transcervical insertion, the intrauterine fetal oxygen sensor is placed against the fetal cheek. It is then connected to an oximetry monitor and a continuous reading is displayed. Fetal membranes must be ruptured, the cervix must be adequately dilated (usually > 2 cm), and the fetal station must be low enough (−2 or lower) for the sensor to touch the fetus. Often the sensor must be adjusted throughout labor due to fetal positional changes. Fetal oxygen concentrations are examined between contractions; values > 30% are considered normal. In 2000, Garite et al, published the results of a multicenter clinical trial that offered some promise for routine use. Eligibility for the trial included term pregnancies in active labor which developed an abnormal FHR

pattern. A total of 1010 women were randomized to EFM alone or EFM + continuous fetal pulse oximetry. There was a reduction of > 50% in the number of cesarean deliveries performed because of non-reassuring fetal status in the EFM + oximetry group. However, there was no net difference in overall cesarean delivery rate (EFM + fetal oximetry 29% vs EFM alone 26%; p = 0.49) because of an increase in cesarean deliveries performed because of dystocia in the oximetry group.54 In 2006, Bloom et al, published the results of the NICHD Maternal Fetal Medicine Unit (MFMU) Network study of fetal pulse oximetry.55 Eligibility criteria for the study were nulliparity, gestational age > 36 weeks, and active labor. An abnormal FHR was not required for study entry, with a subanalysis performed for those with non-reassuring pattern. There was no significant difference in the overall rate of cesarean delivery between groups (EFM + fetal oximetry 26.3% vs EFM alone 27.5%; p = 0.31). The rates of cesarean delivery associated with the separate indications of a non-reassuring fetal heart rate and dystocia were similar between the two groups. This very large study showed that knowledge of the fetal oxygen saturation was not associated with a reduction in the rate of cesarean delivery or with an improvement in the condition of the newborn.55 The clinical use of fetal pulse oximetry as an adjunct to electronic fetal monitoring has not been demonstrated. There is insufficient evidence of any benefit, and use of fetal pulse oximetry in clinical practice was not supported by the ACOG.13

Fetal ST wave analysis The STAN® S31 fetal heart monitor analyzes changes in the T-wave and ST segment of the fetal electrocardiogram (ECG) obtained via a spiral electrode attached to the scalp. A Cochrane Review of three randomized trials found that ST waveform analysis, when used as an adjunct to continuous EFM, was associated with a significant reduction in the number of neonates with severe metabolic acidosis at birth (umbilical artery pH < 7.05 and base deficit > 12 mmol/l).56 The STAN S31 fetal heart rate monitor was recently approved by the FDA for use as an adjunct to the assessment of non-reassuring FHR tracings, in pregnancies over 36 weeks, in labor, with vertex presentation and ruptured membranes. Although the technique is promising, there are inadequate clinical data at the present time for recommendation of its use.

Fetal heart rate interpretation and management future directions One known problem is observer inconsistency in interpretation of the FHR record. A multicenter comparative study

Intrapartum evaluation of fetal well-being

of 17 experts and an intelligent computer system in management of labor using EFM, patient information, and fetal blood sampling found the intelligent computer system to be indistinguishable from the experts in 50 cases examined, but the intelligent computer system was more consistent.10 The study demonstrated the potential to improve the interpretation of EFM and decrease intervention. Similar trials of the intelligent computer system may prove it to be useful. Parer and Ikeda57 recently proposed a framework for the standardized management of intrapartum FHR patterns. It includes the classification of 134 FHR monitor patterns according to available data concerning risk of fetal acidemia and the probability of evolution to a more serious pattern as an indicator of urgency of preparation for delivery. Proposed management of the five color-coded categories includes the rapidity with which preparation for delivery should be made, based on the likelihood of evolution of the FHR pattern to a pattern with a higher risk of acidemia. This preliminary approach was used in two institutions to demonstrate feasibility but was not subjected to appropriate prospective testing. Education could be of value. A retrospective cohort observational study found that a compulsory fetal monitoring education program in one institution was associated with a significant reduction in the incidence of babies born with low Apgar scores and with neonatal hypoxic– ischemic encephalopathy.58

Fetal distress, asphyxia The term ‘fetal distress’ is often used by professionals to describe the clinical situation that includes a concern that the function of the maternal–fetal physiologic unit is altered so that fetal death or serious fetal injury may occur. There is no agreed definition of ‘fetal distress’. The term is imprecise and non-specific. In fact an ACOG Committee Opinion recommended that in communication between clinicians caring for the woman and those caring for the neonate, the term ‘fetal distress’ should be replaced by the term ‘non-reassuring fetal status’. This term should be followed by a further description of the specific FHR findings (e.g. repetitive variable decelerations, fetal tachycardia, or bradycardia).59 In the past, the terms ‘fetal asphyxia’ and ‘birth asphyxia’ have also been used. In the Greek origin of the word asphyxia it has the meaning of pulseless, or stopping of the pulse. Asphyxia refers to acidosis resulting from progressive hypoxia in utero. The ACOG Committee Opinion also declared that the term birth asphyxia is too non-specific and should not be used.59 Instead, ACOG has advocated more specific language and definitions for criteria required to define an acute intrapartum hypoxic event sufficient to cause cerebral palsy (as modified by the ACOG Task Force on Neonatal Encephalopathy and Cerebral

721

Palsy from the template provided by the International Cerebral Palsy Task Force59,60). Criteria to define an acute intrapartum hypoxic event as sufficient to cause cerebral palsy59 include essential criteria (must meet all four): 1. Evidence of a metabolic acidosis in fetal umbilical cord arterial blood obtained at delivery (pH < 7 and base deficit ≥ 12 mmol/l); 2. Early onset of severe or moderate neonatal encephalopathy in infants born at 34 or more weeks of gestation; 3. Cerebral palsy of the spastic quadriplegic or dyskinetic type. 4. Exclusion of other identifiable etiologies, such as trauma, coagulation disorders, infectious conditions, or genetic disorders. Criteria that collectively suggest an intrapartum timing (within close proximity to labor and delivery, e.g. 0–48 hours) but are non-specific to asphyxia insults: 1. A sentinel (signal) hypoxic event occurring immediately before or during labor; 2. A sudden and sustained fetal bradycardia or the absence of fetal heart rate variability in the presence of persistent, late, or variable decelerations, usually after a hypoxic sentinel event when the pattern was previously normal; 3. Apgar scores of 0–3 beyond 5 minutes; 4. Onset of multisystem involvement within 72 hours of birth; 5. Early imaging study showing evidence of acute non-fatal cerebral abnormality.

Efficacy of fetal heart rate monitoring Recent moves toward evidence-based healthcare have led to increasing utilization of systemic reviews, and metaanalysis. Sutton and Abrams61 recently reviewed the use of Bayesian methods in meta-analysis and evidence synthesis, and illustrated the main concepts with a meta-analysis examining the evidence relating the effect of electronic fetal heart rate monitoring on perinatal mortality. Virtually all professional organizations believe that some form of monitoring is necessary during labor, although there are no trials comparing either IA or EFM with no monitoring at all in the control group.41 The recommendations for FHR assessment during labor are based upon protocols used in randomized clinical trials that compared IA and EFM. A recent systematic review of all randomized controlled trials comparing the efficacy and safety of routine continuous electronic FHR monitoring during labor with intermittent auscultation (IA) found 13 published randomized controlled trials.62 Four trials did

722

Fetal Cardiology

not fulfill the selection criteria; the remaining nine trials included 18 561 pregnant women and their 18 695 infants in both high- and low-risk pregnancies from seven clinical centers in the United States, Europe, and Australia. The benefits once claimed for EFM are clearly more modest than once believed, and appear to be primarily in the prevention of early-onset neonatal seizures.62 Long-term implications of neonatal seizures appear to be less serious than once believed. The incidences of abnormal neurological consequences (e.g. cerebral palsy) and perinatal death are not consistently lower among children monitored with electronic methods relative to those monitored with intermittent FHR auscultation. In several trials, EFM was associated with an increased incidence of cesarean delivery and operative vaginal delivery. A 2006 Cochrane Database Systematic Review63 found that when continuous EFM was compared to IA there was no significant difference in overall perinatal death rate. Although EFM was associated with halving of neonatal seizures, no significant difference was detected in the rate of cerebral palsy. There was a significant increase in cesarean deliveries and instrumental deliveries associated with continuous EFM. Data for subgroups of low-risk, high-risk, preterm pregnancies and high-quality trials were consistent with the overall results. Access to fetal blood sampling did not influence the difference in neonatal seizures or any other prespecified outcome. Many institutions use the ‘labor admission test’. It refers to EFM for 20–30 minutes upon admission to labor and delivery for the assessment of fetal status and selection of fetuses at increased risk of non-reassuring FHR patterns during labor that might benefit from continuous EFM during labor (as compared to IA). A systematic review of three randomized trials and 11 observational studies found no evidence that the labor admission test has benefits in low-risk women.64 Routine use of the labor admission test led to higher rates of continuous EFM and intervention, and no reduction in neonatal morbidity.

Unfulfilled expectations from fetal heart rate There were several reasons behind the disappointing story of FHR monitoring and the results of randomized trials.6,13,20,21,46,47,62,63 Expectations were too high. It was assumed that most cases of cerebral palsy are caused by intrapartum asphyxia. In fact, only 10% of cerebral palsy cases are caused by intrapartum asphyxia.65,66 With a cerebral palsy rate of 2/1000 in infants born at term, FHR monitoring is expected to prevent an event that occurs 1/5000 (10% of 2/1000) in term pregnancies. Cerebral palsy has been stable over time.67 Data are limited in quantity, but EFM did not result in a reduction in cerebral

palsy.62,63 The positive predictive value of non-reassuring FHR patterns to predict cerebral palsy in singleton newborn at term is 0.14%.68 A recent review found that, although intrapartum EFM abnormalities correlated with umbilical cord base excess, and its use is associated with decreased neonatal seizures, it has no effect on perinatal mortality or pediatric neurologic morbidity.69 It is now clear that nomenclature and definitions of FHR patterns used in the prior original randomized trials were not standardized. Interobserver and intraobserver consistency was poor. Due to the low frequency of neonatal death, sample sizes were too low to demonstrate differences in mortality rate. Falsely equating absence of evidence with evidence of absence should be avoided. Experts have argued that the randomized trials were inadequate because three conditions were not met: establishment of the (1) reliability of FHR interpretation and the (2) validity of FHR interpretation, and (3) a causal relationship between FHR patterns and adverse outcome.21,70 Given the widespread use of EFM in clinical practice, it is extremely unlikely that a randomized trial of intrapartum EFM versus IA (or no monitoring at all) will ever be undertaken. The continued limitations of EFM and the lack of quality data supporting its use serve as a reminder for obstetrics of the hazards of adopting and implementing a new technology without adequately testing it with high-quality research.

References 1. Saling E. Neues Vorgehen zur Untersuchung des Kindes unter der Gebrut. Arch Gynakol 1961; 197: 108. 2. Hammacher K, Huter KA, Bokelmann J, Werners PH. Foetal heart rate frequency and perinatal condition of foetus and newborn. Gynaecologia (Basel) 1968; 166: 348–60. 3. Hon EH, Quilligan EJ. The classification of fetal heart rate II. A revised working classification. Conn Med 1967; 31: 779–84. 4. Caldeyro-Barcia R, Casacuberta C, Busros R et al. Correlation of intrapartum changes in fetal heart rate with fetal blood oxygen and acid-base balance. In: Adamsons K, ed. Diagnosis and Treatment of Fetal Disorders. New York: Springer-Verlag, 1968: 205. 5. National Institutes of Health. Antenatal diagnosis. Report of consensus development conference. NIH publication no 79-1973. Bethesda, MD: NIH, 1979. 6. Freeman RK, Garite TJ, Nageotte MP. Fetal Heart Rate Monitoring, 3rd edn. Baltimore, MD: Williams & Wilkins, 2003. 7. Strachan BK, Sahota DS, van Wijngaarden WJ et al. Computerised analysis of the fetal heart rate and relation to acidaemia at delivery. Br J Obstet Gynaecol 2001; 108: 848–52. 8. Ayres-de-Campos D, Bernardes J, Garrido A et al. SisPorto 2.0: a program for automated analysis of cardiotocograms. J Matern Fetal Med 2000; 9: 311–18. 9. Hamilton EF, Glaude MC, Macieszczak M. Accuracy of computerized detection of fetal heart rate patterns. Obstet Gynecol 2001; 97: S4.

Intrapartum evaluation of fetal well-being

10. Keith RD, Beckley S, Garibaldi JM et al. A multicentre comparative study of 17 experts and an intelligent computer system for managing labour using the cardiotocogram. Br J Obstet Gynaecol 1995; 102: 688–700. 11. Agrawal SK, Doucette F, Gratton R et al. Intrapartum computerized fetal heart rate parameters and metabolic acidosis at birth. Obstet Gynecol 2003; 102: 731–81. 12. National Institute of Child Health and Human Development Research Planning Workshop. Electronic fetal heart rate monitoring. Research guidelines for interpretation. Am J Obstet Gynecol 1997; 177: 1385–90. 13. American College of Obstetricians and Gynecologists. Intrapartum Fetal Heart Rate Monitoring. Practice Bulletin no 70. Washington, DC: ACOG, 2005: 1–9. 14. Scott JS, Maddison PJ, Taylor PV et al. Connective-tissue disease, antibodies to ribonucleoprotein, and congenital heart block. N Engl J Med 1983; 309: 209–12. 15. Westgate JA, Wibbens B, Bennet L et al. The intrapartum deceleration in center stage: a physiologic approach to the interpretation of fetal heart rate changes in labor. Am J Obstet Gynecol 2007; 197: 236.e1–11. 16. Bennet L, Westgate JA, Lui YC et al. Fetal acidosis and hypotension during repeated umbilical cord occlusions are associated with enhanced chemoreflex responses in nearterm fetal sheep. J Appl Physiol 2005; 99: 1477–82. 17. Yu ZY, Lumbers ER, Gibson KJ, Stevens AD. Effects on hypoxemia on fetal heart rate, variability and cardiac rhythm. Clin Exp Pharmacol Physiol 1998; 25: 577–87. 18. Westgate JA, Gunn AJ, Gunn TR. Antecedents of nenonatal encephalopathy with fetal acidaemia at term. Br J Obstet Gynaecol 1999; 106: 774–82. 19. Boehm FH. FHR variability: key to fetal well being. Contemp Ob Gyn 1977; 9: 57. 20. Parer JT. Handbook of Fetal Heart Rate Monitoring, 2nd edn. Philadelphia: Saunders, 1997. 21. Parer JT, King T. Fetal heart rate monitoring: is it salvageable? Am J Obstet Gynecol 2000; 182: 982–7. 22. Tejani N, Mann LI, Bhakthavathsalan A. Correlation of fetal heart rate patterns and fetal pH with neonatal outcome. Obstet Gynecol 1976; 48: 460–3. 23. Clark SL, Gimovsky ML, Miller FC. Fetal heart rate response to scalp blood sampling. Am J Obstet Gynecol 1982; 144: 706–8. 24. Clark SL, Gimovsky ML, Miller FC. The scalp stimulation test: a clinical alternative to fetal scalp blood sampling. Am J Obstet Gynecol 1984; 148: 274–7. 25. Skupski DW, Rosenberg CR, Eglinton GS. Intrapartum fetal stimulation tests: a meta-analysis. Obstet Gynecol 2002; 99: 129–34. 26. Martin CB Jr, de Haan J, van der Wildt B et al. Mechanisms of late decelerations in the fetal heart rate. A study with autonomic blocking agents in fetal lambs. Eur J Obstet Gynecol Reprod Biol 1979; 9: 361–73. 27. Parer JT, Harris JL, Kruger TR. Fetal oxygen consumption and mechanisms of heart rate response during artificially produced late decelerations of fetal heart rate in sheep. Am J Obstet Gynecol 1980; 136: 478–2. 28. Harris JL, Kruger TR, Parer JT. Mechanisms of late decelerations of the fetal heart rate during hypoxia. Am J Obstet Gynecol 1982; 144: 491–6. 29. Ball RH, Parer JT. The physiologic mechanisms of variable deceleration. Am J Obstet Gynecol 1992; 166: 1683–9.

723

30. de Haan HH, Gunn AJ, Gluckman PD. Fetal heart rate changes do not reflect cardiovascular deterioration during brief repeated umbilical cord occlusions in near-term fetal lambs. Am J Obstet Gynecol 1997; 176: 8–17. 31. de Haan HH, Gunn AJ, Williams CE, Gluckman PD. Brief repeated umbilical cord occlusions cause sustained cytotoxic cerebral edema and focal infarcts in near-term fetal lambs. Pediatr Res 1997; 41: 96–104. 32. Low JA, Lindsay BG, Derrick EJ. Threshold of metabolic acidosis associated with newborn complications. Am J Obstet Gynecol 1997; 177: 1391–4. 33. Garite TJ, Modanlou HD, Freeman RK. Sinusoidal fetal heart rate pattern. Am J Obstet Gynecol 1981; 139: 226–7. 34. Young BK, Katz M, Wilson SJ. Sinusoidal fetal heart rate. I. Clinical significance. Am J Obstet Gynecol 1980; 136: 587–93. 35. Cibils LA. On intrapartum fetal monitoring. Am J Obstet Gynecol 1996; 174: 1382–9. 36. Hefland M, Marton K, Ueland K. Factors involved in the interpretation of fetal monitor tracings. Am J Obstet Gynecol 1981; 151: 737–42. 37. Peck T. Physicians subjectivity in evaluating oxytocin challenge tests. Obstet Gynecol 1980; 56: 13–16. 38. Beaulieu MD, Fabia J, Leduc B et al. The reproducibility of intrapartum cardiotocogram assessments. Can Med J 1982; 127: 214–16. 39. Nielson PV, Stigsby B, Nickelsoen C, Nim J. Intra- and inter-observer variability in the assessments of intrapartum cardiotocograms. Acta Obstet Gynecol Scand 1987; 66: 421–4. 40. Blix E, Sviggum O, Koss KS, Oian P. Inter-observer variation in assessment of 845 labour admission tests: comparison between midwives and obstetricians in clinical settings and two experts. BJOG 2003; 110: 1–5. 41. Freeman RK. Problems with intrapartum fetal heart rate monitoring interpretation and patient management. Obstet Gynecol 2002; 100: 812–26. 42. Krebs HB, Petres RE, Dunn LJ et al. Intrapartum fetal heart rate monitoring. I. Classification and prognosis of fetal heart rate patterns. Am J Obstet Gynecol 1979; 133: 261–3. 43. Paul RH, Suidan AK, Yeh S et al. Clinical fetal monitoring. VII. The evaluation significance of intrapartum baseline FHR variability. Am J Obstet Gynecol 1975; 123: 206–10. 44. William KP, Galerneau F. Interpartum fetal heart rate patterns in the prediction of neonatal acidemia. Am J Obstet Gynecol 2003; 188: 820–3. 45. Parer JT, King T, Flanders S et al. Fetal acidemia and electronic fetal heart rate patterns: Is there evidence of an association? J Matern Fetal Neonatal Med 2006; 19: 289–94. 46. Society of Obstetricians and Gynecologists of Canada Policy Statement. Fetal health surveillance in labor. J Obstet Gynaecol Can 2002; 24: 342–55. 47. Royal College of Obstetricians and Gynecologists Clinical Effectiveness Support Unit. The use of electronic fetal monitoring: evidence based clinical guideline no 8. London: RCOG Press, 2001. 48. International Federation of Gynecology and Obstetrics Workshop on the Guidelines for the Use of Fetal Heart Rate Monitoring. Int J Gynaecol Obstet 1987; 25: 159–67. 49. Morrison JC, Chez BF, Davis ID et al. Intrapartum fetal heart rate assessment: monitoring by auscultation or electronic means. Am J Obstet Gynecol 1993; 168: 63–6.

724

Fetal Cardiology

50. Hofmeyr GJ. Amnioinfusion for umbilical cord compression in labour. Cochrane Database Syst Rev 2000; (2): CD000013. 51. Fawole B, Hofmeyr GJ. Maternal oxygen administration for fetal distress. Cochrane Database Syst Rev 2003; (4): CD000136. 52. Kulier R, Hofmeyr GJ. Tocolytics for suspected intrapartum fetal distress. Cochrane Database Syst Rev 2000; (2): CD000035. 53. Kruger K, Hallberg B, Blennow M et al. Predictive value of fetal scalp blood lactate concentration and pH markers of neurologic disability. Am J Obstet Gynecol 1999; 181: 1072–8. 54. Garite TJ, Dildy GA, McNamara H et al. A multicenter controlled trial of fetal pulse oximetry in the intrapartum management of nonreassuring fetal heart rate patterns. Am J Obstet Gynecol 2000; 183: 1049–58. 55. Bloom SL, Spong CY, Thom E et al. Fetal pulse oximetry and cesarean delivery. N Engl J Med 2006; 355: 2195–202. 56. Neilson JP. Fetal electrocardiogram (ECG) for fetal monitoring during labour. Cochrane Database Syst Rev 2003; (2): CD000116. 57. Parer JT, Ikeda T. A framework for standardized management of intrapartum fetal heart rate patterns. Am J Obstet Gynecol 2007; 197: 26.e1–6. 58. Draycott T, Sibanda T, Owen L et al. Does training in obstetric emergencies improve neonatal outcomes? BJOG 2006; 113: 177–82. 59. American College of Obstetricians and Gynecologists. Inappropriate use of the terms fetal distress and birth asphyxia. Committee Opinion 326. Washington, DC: ACOG, 2005. 60. American College of Obstetricians and Gynecologists, American Academy of Pediatricians. Neonatal encephalopathy and

61.

62.

63.

64.

65. 66. 67.

68.

69.

70.

cerebral palsy: defining the pathogenesis and pathophysiology. Washington, DC: ACOG, 2003. Sutton AJ, Abrams KR. Bayesian methods in meta-analysis and evidence synthesis. Stat Methods Med Res 2001; 4: 277– 303. Thacker SB, Stroup DF, Chang M. Continuous electronic heart rate monitoring for fetal assessment during labor. Cochrane Database Syst Rev 2001; (3): CD000063. Alfirevic Z, Devane D, Gyte G. Continuous cardiotocography (CTG) as a form of electronic fetal monitoring (EFM) for fetal assessment during labour. Cochrane Database Syst Rev 2006; (3): CD006066. Blix E, Reiner LM, Klovning A, Oian P. Prognostic value of the labour admission test and its effectiveness compared with auscultation only: a systematic review. BJOG 2005; 112: 1595–604. Nelson KB. What proportion of cerebral palsy is related to birth asphyxia? J Pediatr 1988; 112: 572–4. Blair E, Stanley FJ. Intrapartum asphyxia: a rare cause of cerebral palsy. J Pediatr 1988; 112: 515–19. Clark SL, Hankins GD. Temporal and demographic trends in cerebral palsy – fact and fiction. Am J Obstet Gynecol 2003; 188: 628–33. Nelson KB, Dambrosia JM, Ting TY, Grether JK. Uncertain value of electronic fetal monitoring in predicting cerebral palsy. N Engl J Med 1996; 324: 613–18. Graham EM, Petersen SM, Christo DK, Fox HE. Intrapartum electronic fetal heart rate monitoring and the prevention of perinatal brain injury. Obstet Gynecol 2006; 108: 656–66. Paneth N, Bommarito M, Stricker J. Electronic fetal monitoring and later outcome. Clin Invest Med 1993; 16: 159–65.

48 Cardiac disease in pregnancy David Planer, Haim D Danenberg, and Chaim Lotan Introduction Pregnant women with heart disease require specific care in light of the hemodynamic changes that occur during pregnancy. Management must be based on an understanding of the specific anatomic and hemodynamic defects, the nature of previous surgical correction, and the risk of pregnancy for the mother as well as expected fetal outcome. Very few heart conditions pose high enough risks to advise a patient to avoid pregnancy. Careful and continuous multidisciplinary monitoring during pregnancy and the postpartum period will lead to a normal delivery and ensure the health of both mother and child.1

Hemodynamic changes during pregnancy The hemodynamic changes during pregnancy are usually well tolerated by the mother, but can exacerbate the clinical signs of existing heart disease or induce cardiac failure. These changes include an increase in total blood volume with relative anemia, increase in both heart rate and stroke volume, and a decrease in systemic and pulmonary vascular resistance with a slight decrease in blood pressure. Hemodynamic changes also result in an augmented cardiac output starting at the 5th week of gestation and reaching its peak (30–50% above baseline level) at the 32nd week. The uterus requires up to 18% of the cardiac output, and oxygen consumption is increased by about 30%, owing to the increased metabolic need of both the mother and the fetus. Cardiac output may be compromised by decreased venous return due to compression of the enlarged uterus, which may be alleviated by lying in the left lateral position. Cardiac output further increases during labor, which may also lead to a marked increase in systolic and diastolic blood pressure during uterine contractions. Immediately after delivery, the preload is raised by uterine blood autotransfusion and caval decompression; these changes subside over the following 2 weeks.

Besides hemodynamic changes, pregnancy is also associated with a hypercoagulable state due to venous stasis, increase in vitamin K-dependent clotting factors, and lowered protein S levels.

Special considerations and risk stratification of specific cardiac disorders Successful neonatal surgery for complex congenital heart anomalies has significantly increased survival, and this has resulted in a larger number of women with congenital heart disease living to reproductive age. Although most women with heart disease have successful pregnancies, the treatment of a woman with known or suspected heart disease should be multidisciplinary and include obstetricians, neonatologists, anesthetists, and cardiologists. This team should be involved in patient management and follow-up from pregnancy planning, to delivery and the postpartum period.2 A detailed list of the unique aspects of specific cardiac disorders is presented in Table 48.1.

High-risk patients The most important predicting factor for the risk of pregnancy in a woman with heart disease is the functional capacity. Any woman that reaches New York Heart Association (NYHA) functional class III or IV (shortness of breath at rest or during minimal physical activity), whatever the underlying condition, is considered to be at high risk. These patients have no cardiovascular reserve, and further increase in demand could result in hemodynamic compromise. A few other clinical conditions are considered as high risk: 1. Pulmonary hypertension: Severe pulmonary hypertension and especially Eisenmenger syndrome (shunt reversal due to increased right-sided pressures) carries the higher risk of up to 50% maternal mortality during

726

Fetal Cardiology

Table 48.1

Special considerations and risk stratification of specific cardiac disorders

Cardiac lesion

Pathophysiological considerations

Accompanying phenomena

Risk stratification

Special considerations

Left-to-right shunts: ASD VSD PDA

Increased cardiac output is counterbalanced by decreased systemic resistance

Arrhythmias, pulmonary hypertension, CHF paradoxical emboli (ASD)

Small shunt – low risk Large shunt – intermediate risk

LV outflow tract obstruction: AS3

Limited elevation of cardiac output; abnormal reaction to low preload

Hypotension, CHF

Severe AS – high risk

Coarctation of the aorta

Imbalance of upper–lower body blood pressure

IUGR and premature labor due to uterine hypoperfusion

Corrected – low risk Uncorrected – intermediate risk

Pulmonic stenosis

Increased hemodynamic load on a stenosed valve

Right heart failure, atrial arrhythmias

Corrected or mild – low risk Severe – intermediate risk

Should be corrected prior to pregnancy Balloon valvulotomy should be considered in the pregnant symptomatic patient

Cyanotic heart disease: tetralogy TGA single ventricle4

Increased right-toleft shunt due to reduced systemic resistance and increased cardiac output

Prematurity and fetal death (∼40% each) Maternal cardiac events

Repaired lesions without residual cardiac dysfunction – low risk Unrepaired or palliated lesions – intermediate risk

High significance to the type of correction and residual function

Marfan syndrome

Hemodynamic stress and hormonal effect compromise diseased aortic wall

Aortic dilatation and dissection, aortic regurgitation Maternal and fetal mortality

Aortic root or valvular involvement – high risk

Women with normal aortic root should undergo serial echocardiograms Women with evident aortic root involvement should be offered abortion

Eisenmenger syndrome and pulmonary vascular obstructive disease

Increased right-toleft shunting due to reduced systemic resistance

Increased cyanosis, spontaneous abortions, preterm deliveries and high perinatal mortality High maternal mortality

High risk

Early termination of pregnancy should be recommended

Mitral stenosis

Hypervolemia and tachycardia increase LA pressure

Atrial fibrillation, CHF

Intermediate risk

Valvular insufficiency: mitral and aortic (rheumatic)

Usually well tolerated due to decreased systemic resistance

Peripartum cardiomyopathy5–7

LV systolic dysfunction

Symptomatic and severe AS should be corrected prior to pregnancy Balloon valvulotomy should be considered in the pregnant symptomatic patient

Normal LV function – low risk

CHF, arrhythmias

Past peripartum cardiomyopathy with no residual dysfunction – intermediate risk Ventricular dysfunction – high risk

AS, aortic stenosis; ASD, atrial septal defect; CHF, congestive heart failure; IUGR, intrauterine growth restriction; LA, left atrium; LV, left ventricle; PDA, patent ductus arteriosus; TGA, transposition of the great arteries; VSD, ventricular septal defect. Patients with severe CHF (New York Heart Association (NYHA) class III and IV) are considered high-risk as well as patients with severe pulmonary hypertension of any cause.

Cardiac disease in pregnancy

pregnancy.8 Women with severe pulmonary vascular disease adapt poorly to hemodynamic changes that occur during pregnancy, and especially during labor and the peripartum period. In addition, any further increase in pulmonary vascular resistance, such as in a case of pulmonary thrombosis, can be fatal due to severe hypoxemia – even in a previously mildly symptomatic woman. Fetal prognosis is also poor: only 15–20% of pregnancies progress to term, almost half suffer from severe intrauterine growth restriction (IUGR), and the mortality rate is considerably high. 2. Cyanosis: Low oxygen saturation in maternal arterial blood is a bad outcome predictor. In patients with cyanotic heart disease, morbidity from congestive heart failure, arrhythmias, and infective endocarditis is about 30% and mortality is around 2%. Fetal outcome is also poor due to impaired fetal growth. In women with uncorrected cyanotic heart disease, attention should be paid to the increased risk of thrombosis due to secondary erythrocytosis. Prophylaxis with heparin should be considered. 3. Left ventricular outflow tract obstruction: Severe outflow tract obstruction (such as in aortic stenosis) results in a fixed cardiac output which, in the presence of increased demand, causes increases in left ventricular, left atrial, and pulmonary vein pressures. This, in turn, may lead to pulmonary congestion and, in severe cases, cardiogenic shock.

Low-risk patients Patients with small shunts, no pulmonary hypertension, and mild to moderate left ventricular outflow tract obstruction usually tolerate physiologic hemodynamic changes during pregnancy well. These women should be followed up both clinically and with echocardiography before and during pregnancy to anticipate any deterioration and to provide updated evaluation before delivery or any other intervention indicated. The cardiac disease in pregnancy (CARPREG) investigators developed a simple risk score (Table 48.2) to predict the risk of cardiac events during pregnancy. The predictive value of this index was evaluated prospectively in more than 700 pregnancies, in patients with both congenital and acquired heart diseases.2,9,10

Principles of management General In the evaluation of a pregnant woman with cardiovascular disease, the safety of the mother takes precedence over that of the fetus. Assessment of the patient should include

727

Table 48.2 Cardiac disease in pregnancy (CARPREG) risk index Predictor

Definition

Points

Functional class

NYHA III or IV or cyanosis

1

Previous cardiovascular event

Heart failure, TIA, stroke or arrhythmia

1

Left ventricular obstruction

MV area < 2 cm2, AV area < 1.5 cm2, peak outflow gradient > 30 mmHg

1

LV systolic dysfunction

Ejection fraction < 40%

1

Total points The risk score was found to accurately predict the risk of adverse cardiac event during pregnancy: 0 points 4–10%, 1 point 25–30%, two points and above > 60%. TIA, transient ischemic attack; MV, mitral valve; AV, aortic valve.5

both anatomical and hemodynamic evaluation based on her medical history and physical examination, 12-lead electrocardiogram (ECG), Doppler echocardiography (the ideal examination for the ‘cardiac’ pregnant woman, providing a high diagnostic yield with no reported damage to the mother or fetus), and pulse oximetry. Diagnostic procedures such as X-radiography, computed tomography, radionuclide scan, and cardiac catheterization expose the fetus to harmful radiation and should be avoided throughout pregnancy, especially in the early months after conception.11 Evaluation should cover the underlying cardiac lesion, additional associated risk factors, maternal functional status, the possibility of further palliative or corrective surgery, maternal prognosis and child-caring ability, and the fetal risk for congenital heart disease. For a woman in whom pregnancy is considered an intermediate or high risk (Tables 48.1 and 48.2), a multidisciplinary team approach and close follow-up are recommended, starting as early as possible. Specific cardiovascular disorders carry an unacceptable risk of mortality to both the mother and the fetus and should generally be avoided (Table 48.3). Recently, the European Society of Cardiology published guidelines on the management of cardiovascular diseases during pregnancy.12

Congenital heart disease Right ventricular outflow tract obstruction Right ventricular outflow tract (RVOT) obstruction, such as pulmonary valve stenosis, is usually well tolerated

728

Fetal Cardiology

Table 48.3 pregnancy

Cardiac contraindications for

Severe pulmonary hypertension (≥ 0.75 systemic pressure) Eisenmenger syndrome Severe obstructive cardiac lesions (intervention recommended prior to conception)

ventricle. Although hemodynamic changes and increased burden on the ventricle can cause a compensatory increase in systemic venous pressure and finally a low cardiac output state, a recently published series concluded that most women can tolerate pregnancy well without long-term sequelae, but with a considerable rate of non-cardiac events (mainly miscarriages).14

Marfan syndrome, aortic root ≥ 4 cm Severe cyanosis Congestive heart failure class III and IV Peripartum cardiomyopathy in the past with severe heart failure or residual left ventricular dysfunction.

during pregnancy. In cases of severe pulmonary valve stenosis, right heart failure and tricuspid regurgitation might worsen due to volume overload during pregnancy. Patients with severe RVOT obstruction should therefore consider treatment prior to conception. Pulmonic balloon valvotomy is the treatment of choice when right ventricular failure symptoms develop during pregnancy. The risk in patients with uncorrected tetralogy of Fallot (TOF) is usually high, and depends mainly on the maternal oxygen saturation.4 During pregnancy, volume overload on the right side and decreased systemic vascular resistance lead to an increase in right-to-left shunt that exacerbates cyanosis. Patients with baseline saturation below 85% are considered to be at high risk, and should be monitored carefully during pregnancy. Patients with well-corrected TOF without residual right ventricle (RV) obstruction or RV failure are considered to be at low risk. All patients with TOF should have genetic counseling before pregnancy.

Transposition of the great arteries In patients with congenitally corrected transposition of the great arteries (TGA), the right ventricle works under systemic pressures. These women usually do well during pregnancy, although tricuspid regurgitation might increase and supraventricular arrhythmias might ensue. Patients with surgically corrected TGA in good functional class usually tolerate pregnancy well, but close monitoring is advised.13

Post-Fontan operation The fontan procedure is the connection of the systemic venous return to the pulmonary artery, and is the treatment of choice in cases of single ventricle and tricuspid atresia. In these patients cardiac output depends on the pressure gradient between the vena cava and the systemic

Coarctation of the aorta Aortic coarctation should be repaired before pregnancy. Hypertension might be difficult to control in women with uncorrected coarctation, but toxemia is very rare. Although fetal growth is usually normal, overtreatment of blood pressure might compromise distal segment blood pressure and placental perfusion, which may result in fetal death. The most common cause of death in these patients is aortic rupture. A sudden increase in blood pressure is a risk factor for aortic rupture, and therefore beta-blockers and restriction of physical activity are recommended.

Left-to-right shunts Left-to-right shunts consist of an abnormal communication between the left and the right systems, at the atrial level (atrial septal defect, ASD), ventricular level (ventricular septal defect, VSD), or the great vessels (patent ductus arteriosus, PDA).

ASD ASD is the most common heart defect in pregnant women. This lesion is asymptomatic in many cases and many women are unaware of the lesion. Uncorrected ASD has an increased risk for atrial arrhythmias as well as paradoxical embolus, where a thrombus in the venous system migrates through the ASD to the systemic system. (A mandatory condition of this phenomenon is a temporary shunt reversal during coughing or valsalva.)

VSD Small VSDs are usually well tolerated during pregnancy. Symptoms are usually derived from left ventricle volume overload and depend on shunt magnitude. Large noncorrected VSDs induce increased pulmonary vascular resistance, which leads to shunt reversal and cyanosis in most patients who survive through adulthood. This is called the Eisenmenger syndrome and carries an extremely high risk of up to 50% mortality during pregnancy, and is thus considered a contraindication for pregnancy. A complex malformation such as complete atrioventricular (AV) canal, or a triad of ASD, VSD, and single atrioventricular

Cardiac disease in pregnancy

RA

729

LA Ao ASD

10

AV

RV

VSD

PDA

LV 5 PA

Figure 48.1 Complete atrioventricular (AV) canal in a patient with Down syndrome. This complex malformation is a triad of atrial septal defect (ASD) primum, common AV valve, and ventral septal defect (VSD). RA, right atrium; RV, right ventricle; LA, left atrium; LV, left ventricle.

valve (Figure 48.1), seen mostly in patients with Down syndrome, is associated with poor prognosis.

PDA Most cases of patent ductus arteriosus with restrictive shunt tolerate pregnancy well (Figure 48.2). An increased risk for endocarditis exists in non-corrected cases. Non-restrictive shunts are usually diagnosed and easily corrected in childhood, but in rare undiagnosed cases, increases in pulmonary vascular resistance can lead to the Eisenmenger syndrome.

Valvular heart disease Valvular heart disease in pregnancy comprises acquired, mainly rheumatic valvular disease, and congenital heart or inherited connective tissue disease (Marfan, Ehlers– Danlos, and so on). In developing countries the most common symptomatic valvular disease in pregnancy is still rheumatic mitral stenosis.15,16

Mitral stenosis Increased cardiac output and heart rate during pregnancy augment the transvalvular pressure gradient, which leads to an elevated pulmonary venous pressure. This might unmask a previously asymptomatic mitral stenosis. The transmitral gradient increases mainly during the second

Figure 48.2 Patent ductus arteriosus (PDA) with a communicating flow between the distal aortic arch (Ao) and the pulmonary artery (PA) as demonstrated by transesophageal color Doppler. A continuous loud murmur found on routine examination led to the diagnosis of PDA in a 42-year-old pregnant woman.

and third trimesters. Close clinical and echocardiographic monitoring is warranted even in the asymptomatic woman (Figure 48.3). The medical treatment principle is to lower the heart rate and prevent atrial fibrillation, which are major risks for congestive heart failure exacerbation. Treatment with beta-blockers is indicated in any symptomatic woman or when the estimated pulmonary artery pressure is above 50 mmHg. Loop diuretics are indicated when pulmonary congestion persists despite normal heart rate and rhythm. If the patient remains symptomatic despite ideal medical therapy, there is a major risk for pulmonary edema and fetal death during delivery. In these women, mitral valvotomy should be considered. Surgical repair or replacement of the mitral valve during pregnancy carries a high risk (up to 30%) of fetal death. Hence, in most developed countries the procedure of choice nowadays is percutaneous mitral valvotomy. This procedure, when done in selected patients in highly experienced centers, is usually safe but not without risks, and should therefore be reserved only for symptomatic patients.17

Aortic stenosis Symptomatic aortic stenosis is rare in pregnant women. The most common etiology is congenital, and in cases

730

Fetal Cardiology

V

V

Figure 48.3

LA

5

5

LAA

LV (a)

LV (b)

of rheumatic valvular disease, mitral stenosis is usually present as well. Fixed cardiac output is a major hemodynamic problem and might compromise safety of delivery. Therefore, severe, symptomatic cases should be treated before delivery. The procedure of choice in most cases is balloon valvotomy, which should be carried out only in highly experienced centers.18

Mitral and aortic regurgitation Chronic regurgitant valve disease is usually well tolerated during pregnancy. Although increased cardiac output and blood volume may lead to a volume overload, this is usually compensated by the physiological afterload reduction resulting from systemic vasodilatation. Acute valvular regurgitation is usually poorly tolerated since the atria (in cases of mitral regurgitation) and ventricle (in the cases of aortic regurgitation) are not dilated, and have a poor compliance to the sudden increase in blood volume. These patients may develop progressive heart failure symptoms. Symptomatic patients should be treated with diuretics and vasodilators (nitrates and calcium channel blockers). Surgery is usually not recommended during pregnancy, and should be reserved only for refractory heart failure.

Prosthetic heart valves Choosing the correct kind of heart valve prosthesis in women of childbearing years who need valve replacement is controversial, and should be carried out in accordance with the patient’s will and needs. Mechanical valves expose the patient and the fetus to the risks of anticoagulant therapy (see below), whereas patients with biological prostheses have a higher live-birth rate and usually do not require anticoagulant therapy. However, bioprostheses have a relatively higher rate of early structural valve deterioration that will require recurrent operation within a few years.19

LA

Rheumatic mitral stenosis. (a) The mitral valve leaflets are thickened and calcified with limited opening (arrows), the left atrium (LA) is dilated, the left atrial appendage (LAA) is demonstrated without a thrombus. (b) Characteristic color Doppler flow through a narrowed mitral orifice is demonstrated.

Endocarditis prophylaxis The American College of Cardiology/American Heart Association guidelines for the management of patients with valvular heart disease do not recommend endocarditis prophylaxis in patients with valvular heart disease undergoing uncomplicated vaginal delivery or cesarean section, unless infection is suspected. Antibiotics are optional, but seem to be standard practice for high-risk patients with prosthetic heart valves, a previous history of endocarditis, complex congenital heart disease, or surgically constructed systemic–pulmonary conduits.3

Marfan syndrome Marfan syndrome is a dominantly inherited fibrillin-1 deficiency, affecting connective tissues in all organs, with dominant involvement of the heart, eyes, and the skeletal structure. About 80% of women with Marfan syndrome have cardiac involvement. The most common lesion is mitral valve prolapse with mitral regurgitation, but the most fatal problem is aortic root aneurysm with rupture and dissection. All women should undergo full cardiac evaluation before pregnancy and close echocardiographic follow-up during pregnancy. The risk of aortic dissection is highest in the third trimester and the peripartum due to the high shear force associated with hemodynamic changes during delivery. The risk of acute aortic dissection is low if the aortic root diameter is lower than 4 cm.20 The risk of acute dissection during pregnancy is about 10% if the aortic diameter is above 4.5 cm; thus, for these women, it is advised to have elective aortic root replacement before surgery. If possible, preservation of the aortic valve is recommended to avoid the need for chronic anticoagulation. Most women with surgically corrected aortic root will have an uncomplicated pregnancy. All women should be treated with beta-blockers throughout pregnancy. Normal delivery is recommended for women with an aortic root diameter below 4.5 cm, and cesarean section is advised if the aortic root is larger. Postpartum hemorrhage should be expected.21

Cardiac disease in pregnancy

731

Figure 48.4 FL TL FL

(a)

(b)

Acute aortic dissection of the ascending aorta is a medical emergency, and surgical intervention is indicated (Figure 48.4). Fetal outcome post-emergency aortic root repair is usually poor. Genetic counseling is advised in all pregnancies of parents diagnosed with Marfan syndrome.

Acquired heart disease Cardiomyopathies Peripartum cardiomyopathy (PPCM) is an acquired form of dilated cardiomyopathy that, by definition, develops during the last month of pregnancy or within 5 months of delivery.5 The clinical signs and symptoms are as in any other form of congestive heart failure. Sometimes, the presenting symptom might be arrhythmia (ventricular or atrial), or a thromboembolic event. PPCM might deteriorate rapidly, and require the administration of inotropic drugs or a left ventricular assist device. Few cases will end in an urgent transplantation. In most cases the clinical course is unpredictable; paradoxically, the most fulminant cases, in the event of survival, tend to resolve rapidly. This phenomenon gives rationale to treatment with an assist device as a bridge to spontaneous recovery, or, in the worst cases, a bridge to transplantation. The diagnosis of PPCM is established by clinical and echocardiographic evaluation within the temporal definitions of PPCM. Idiopathic or post-myocarditis cardiomyopathy tends to exacerbate during pregnancy, but in cases when there is no previously documented myocardial dysfunction, there is no way to distinguish between the etiologies. Myocardial biopsy usually demonstrates non-specific inflammation and usually does not change the management, and hence is not indicated in most cases. The etiology for PPCM is unknown, but assumed to be an immune reaction against fetal antigen. The immune hypothesis is the rationale for immunosuppressive therapy and treatment with immunoglobulins; however, evidence to support these strategies is weak.6

Acute aortic dissection in a patient with Marfan syndrome. Transesophageal echo (a) demonstrating a dilated ascending aorta with a flap (arrows) extending from the aortic root. Color Doppler imaging (b) demonstrates blood flow in the true lumen (TL) with flow through the intimal tear into the false lumen (FL).

The medical treatment of PPCM is the standard treatment for heart failure. Conservative measures, such as salt restriction and limitation of activity, are extremely important. Digoxin can be administered safely, whereas diuretics may impair uterine blood flow, and their initiation during pregnancy is not recommended unless absolutely necessary. Continuation of diuretic therapy initiated prior to pregnancy does not seem to be problematic.22 Angiotensin converting enzyme (ACE) inhibitors are contraindicated, because of maternal–fetal transfer and increased risk for the fetus. A major question in a woman with PPCM is the risk of subsequent pregnancies. Elkayam and colleagues7 found that the relapse rate of PPCM with clinically significant deterioration and death in subsequent pregnancies is high. In subgroup analysis it was found that when left ventricle (LV) systolic function was completely recovered, the risk of symptomatic heart failure in subsequent pregnancies was 21% (compared to 44% in women with abnormal LV function), but death occurred only in the group with abnormal LV function. These points should be thoroughly discussed before a subsequent pregnancy in any woman with PPCM. Other causes of dilated cardiomyopathy (DCM) are rare among pregnant women. Pregnancy is not advised for women with idiopathic DCM, and pregnancy termination is recommended for every case of ejection fraction below 50% or when the LV is significantly dilated.

Hypertrophic cardiomyopathy Hypertrophic cardiomyopathy is usually well tolerated during pregnancy. Treatment is indicated in cases of symptomatic women, and is based on controlling heart rate and rhythm with beta-blockers that should be continued during delivery.23 Patients with atrial fibrillation should undergo cardioversion, and in cases of recurrent atrial fibrillation, treatment with anticoagulation is indicated. For women with ventricular arrhythmia, or those at high risk for ventricular arrhythmias (syncope, severe hypertrophy, and family history of sudden death), an implantable

732

Fetal Cardiology

cardioverter defibrillator (ICD) is recommended. Treatment with amiodarone is indicated in resistant cases. Genetic counseling is strongly recommended for all cases.

electrical direct current cardioversion, which may be used safely.27 The indications for an ICD for primary and secondary prevention of life-threatening ventricular arrhythmias are as in non-pregnant patients. Pregnancy itself does not increase the frequency of ICD discharges.28

Coronary artery disease Although atheromatous coronary artery disease is usually rare in the age group of women of childbearing years, older pregnant women and women with major cardiac risk factors (long-standing diabetes, familial hypercholesterolemia, and family history of premature coronary artery disease) can develop symptomatic atherosclerotic disease. Evaluation of the patient with suspected coronary artery disease (CAD) is not different from that of non-pregnant women. The exercise stress test is the preferred initial evaluation method. Thallium or other radioactive isotope perfusion scanning is associated with high-dose radiation, so it is generally not in use among pregnant women. Cardiac catheterization and coronary computed tomography (CT) angiography are not contraindicated, but the benefits should be weighed against the risk of radiation exposure. Rare causes of angina pectoris such as coronary anomalies and coronary arteritis (Takayasu arteritis, and so on) are relatively more common in this age group. Known CAD is not a contraindication for pregnancy, but full assessment and treatment is advised prior to conception.12

Arrhythmia Evaluation of arrhythmia is the same as in the nonpregnant patient. Underlying illness and precipitating factors should be treated if possible. The use of betablockers during pregnancy was anecdotally reported to cause intrauterine growth restriction, apnea at birth, and fetal bradycardia. These concerns were not confirmed in large studies, and beta-blockers (including sotalol) are considered relatively safe and are used when indicated in the treatment of supraventricular tachycardia (SVT), hypertrophic cardiomyopathy, and hyperthyroidism.24 Owing to maternal–fetal transfer, fetal and neonatal monitoring of heart rate (and neonatal blood glucose and respiratory status) is warranted. Verapamil is also considered safe for the treatment of SVT; however, it should be discontinued at the onset of labor to prevent dysfunctional labor and postpartum hemorrhage. Digoxin has not shown untoward effects in the treatment of arrhythmia, and neither has quinidine in therapeutic doses. Little information is available on the use of procainamide, propafenone, and disopyramide during pregnancy, but so far no adverse fetal effects have been described.25 Amiodarone use was reported in several cases to result in fetal hypothyroidism.26 Arrhythmia causing hemodynamic compromise may require

Cardiovascular drug treatment during pregnancy Most cardiovascular drugs cross the placenta and are secreted in breast-milk. The risk/benefit ratio should be well considered prior to the administration of any medication during pregnancy and lactation. The American Heart Association/American College of Cardiology guidelines classify cardiovascular drugs according to safety3 (Table 48.4).

Anticoagulation Pregnant patients with cardiovascular disorders may require systemic anticoagulation for prophylaxis and treatment of venous or arterial thromboembolic disease or because of an artificial heart valve. This exposes the mother and fetus to significant risks. Irrespective of whether heparin or warfarin is used, only two-thirds of pregnancies will result in a healthy delivery. Maternal morbidity and mortality are more common with heparin, owing to treatment failure (e.g. mechanical valve thrombosis) and hemorrhage. Both heparin and warfarin may cause uteroplacental bleeding and increase the risk for aborting. Long-term systemic heparinization may cause osteoporosis. Heparin does not cross the placental barrier and has no effect on the fetus. Low-molecular-weight heparin (LMWH) offers several advantages over unfractionated heparin, including the greater ease of administration and more predictable dose-response, and less osteoporosis and thrombocytopenia. However, the data on LMWH use for mechanical valves in pregnancy are still limited, and valve thrombosis has been reported.29 Warfarin derivatives cross the placental membrane and cause fetal anticoagulation. Central nervous system abnormalities including blindness may occur, most probably owing to bleeding and scarring. Furthermore, warfarin given in the first trimester may have a teratogenic potential, causing embryopathy manifested by nasal hypoplasia and skeletal malformations. The incidence of coumadin embryopathy has not yet been established; however, an estimate of 4–10% seems reasonable. This effect is probably dose-related, with low occurrence in patients treated by less than 5 mg coumadin. In a study of 43 women carrying 58 fetuses and treated by warfarin, a significantly lower rate of fetal complications was reported in those receiving a daily dose of ≤ 5 mg.30

Cardiac disease in pregnancy

Table 48.4

733

Cardiovascular drug safety during pregnancy

Safety

Drug

Potential adverse effect

Safe

Digoxin

Low birth weight

Quinidine

Toxic levels may cause premature labor and 8th cranial nerve damage

Lidocaine

Toxic levels and fetal acidosis may cause CNS depression

Adenosine

None reported. Use during first trimester limited to few patients

β-Adrenergic blockers

IUGR, birth apnea, hypoglycemia, hyperbilirubinemia, premature uterine contractions

Hydralazine

None reported

Potentially safe

Heparin

Potentially unsafe

Sodium nitroprusside

Thiocyanate toxicity in high doses, mortality in animal studies

Diuretics

Impaired uterine blood flow and placental hypoperfusion, thrombocytopenia, jaundice, hyponatremia, bradycardia

Warfarin

Uterine fetal hemorrhage, embryopathy, CNS abnormalities

Amiodarone

IUGR, prematurity, hypothyroidism

ACE inhibitors

IUGR, premature delivery and low birth weight, skull ossification defect, oligohydramnios, neonatal renal failure, anemia and death, limb contractures, patent ductus arteriosus

Disopyramide Mexiletene

Premature uterine contractions Fetal and neonatal bradycardia, IUGR, low Apgar scores, hypoglycemia, neonatal hyperthyroidism

Unsafe

Yet undefined, limited information

Propafenone

None reported

Ca2+ channel blockers

Fetal distress due to maternal hypotension

Organic nitrates

Fetal bradycardia

ACE, angiotensin converting enzyme; CNS, central nervous system; IUGR, intrauterine growth restriction Apgar, activity, pulse, grimace, appearance, respiration.

Most authors recommend systemic heparinization during the critical 6–12 weeks of gestation and over the last 2 weeks of pregnancy to allow rapid reversal for delivery. In the first trimester, high-risk patients with a history of thromboembolism or an older-generation mechanical prosthesis in the mitral position should receive continuous heparin intravenously to prolong the partial thromboplastin time (PTT) to 2–3 times that of control. Subcutaneous heparin adjusted to obtain similar PTT levels may be used in low-risk patients. The use of LMWH is recommended for deep vein thrombosis (DVT) and pulmonary emboli prophylaxis and treatment; however, its use in patients with mechanical valves is controversial; if used, monitoring of anti-Xa levels is mandatory. After the 36th week of pregnancy, intravenous heparin should be administered in anticipation of labor. In the absence of significant bleeding, heparin may be resumed 4–6 hours after delivery. Warfarin can be administered for the remainder of gestation with little risk of embryopathy. Alternatively, high-risk patients may be given coumadin throughout pregnancy until the 36th week. However, informed consent for this regimen should be obtained,

and the risks explained to the patient and her partner. The lowest possible effective dose (preferably ≤ 5 mg) should be administered.3

Cardiovascular complications of tocolytic drugs Tocolytics are used to decrease uterine contractility usually at the third trimester of pregnancy. The tocolytic agents have β-sympathomimetic activity and may cause tachycardia and antidiuresis. Up to 4.4% of female patients exposed to tocolytic drugs developed pulmonary edema, the etiology of which is as yet undefined. In a review of 58 reported cases the most commonly used agents were terbutaline, isoxsuprine, ritodrine, and salbutamol. The average duration of treatment was 54 hours and included discontinuation of tocolytic therapy, supplemental oxygen, and intravenous diuretics, with prompt and effective response. Maternal and fetal mortality was rarely observed.31

734

Fetal Cardiology

Delivery management in the cardiac patient The time, route, and interventions for delivery of the patient with heart disease should be planned in advance.1 The mode of delivery should be discussed between the obstetricians, anesthesiologists, cardiologists, and the patient. With a number of exceptions, vaginal delivery with a facilitated second stage is recommended for the pregnant woman with cardiac disease. Cesarean section should be reserved for obstetric reasons only, except for women anticoagulated with coumadin derivatives, patients with Marfan syndrome with dilated aortic root and/or aortic dissection, women with fixed obstructive lesions (as in aortic stenosis), or when early delivery is essential for maternal survival. Pain relief is important, since pain aggravates the circulatory burden of labor. Optimal anesthesia and analgesia should be considered prior to delivery. Epidural anesthesia is the mode of choice for most cases; however, hypotension should be avoided. Epidural fentanyl is advantageous in cyanotic patients with shunt lesions, as it does not lower peripheral vascular resistance. Intensive echocardiographic and hemodynamic monitoring is recommended for severe maternal heart disease. Right heart catheterization and monitoring has been advocated during labor in preload-dependent cardiac patients (Eisenmenger syndrome, severe aortic stenosis). Patients at increased risk should be monitored for at least 72 hours after delivery, owing to prolonged hemodynamic instability.

Summary In most cases of women with congenital or acquired heart disease, pregnancy is not contraindicated. Understanding the physiological and hemodynamic changes during pregnancy and labor, and their influence on the preexisting cardiac problem, are essential for the management of women with cardiovascular disease. Each woman should undergo a comprehensive multidisciplinary evaluation before pregnancy, when possible. A woman with heart disease planning pregnancy should be aware of the anticipated risks, and all decisions concerning time and mode of delivery should be planned well in advance.

References 1. Uebing A, Steer PJ, Gatzoulis MA et al. Pregnancy and congenital heart disease. BMJ 2006; 332: 401–6. 2. Khairy P, Ouyang DW, Landzberg MJ et al. Pregnancy outcomes in women with congenital heart disease. Circulation 2006; 113: 517–24.

3. Bonow RO, Carabello BA, Kanu C et al. ACC/AHA 2006 guidelines for the management of patients with valvular heart disease: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (writing committee to revise the 1998 Guidelines for the Management of Patients With Valvular Heart Disease): developed in collaboration with the Society of Cardiovascular Anesthesiologists: endorsed by the Society for Cardiovascular Angiography and Interventions and the Society of Thoracic Surgeons. Circulation 2006; 114: e84–231. 4. Presbitero P, Somerville J, Rabajoli F et al. Pregnancy in cyanotic congenital heart disease. Outcome of mother and fetus. Circulation 1994; 89: 2673–6. 5. Pearson GD, Veille JC, Rahimtoola S et al. Peripartum cardiomyopathy: National Heart, Lung, and Blood Institute and Office of Rare Diseases (National Institutes of Health) workshop recommendations and review. JAMA 2000; 283: 1183–8. 6. Sliwa K, Fett J, Elkayam U. Peripartum cardiomyopathy. Lancet 2006; 368: 687–93. 7. Elkayam U, Tummala PP, Rao K et al. Maternal and fetal outcomes of subsequent pregnancies in women with peripartum cardiomyopathy. N Engl J Med 2001; 344: 1567–71. 8. Weiss BM, Hess OM. Pulmonary vascular disease and pregnancy: current controversies, management strategies, and perspectives. Eur Heart J 2000; 21: 104–15. 9. Siu SC, Sermer M, Harrison DA et al. Risk and predictors for pregnancy-related complications in women with heart disease. Circulation 1997; 96: 2789–94. 10. Siu SC, Sermer M, Colman JM et al. Prospective multicenter study of pregnancy outcomes in women with heart disease. Circulation 2001; 104: 515–21. 11. Mossman KL, Hill LT. Radiation risks in pregnancy. Obstet Gynecol 1982; 60: 237–42. 12. Expert consensus document on management of cardiovascular diseases during pregnancy. Eur Heart J 2003; 24: 761–81. 13. Therrien J, Barnes I, Somerville J. Outcome of pregnancy in patients with congenitally corrected transposition of the great arteries. Am J Cardiol 1999; 84: 820–4. 14. Drenthen W, Pieper PG, Roos-Hesselink JW et al. Pregnancy and delivery in women after Fontan palliation. Heart 2006; 92: 1290–4. 15. Stout KK, Otto CM. Pregnancy in women with valvular heart disease. Heart 2007; 93: 552–8. 16. Reimold SC, Rutherford JD. Clinical practice. Valvular heart disease in pregnancy. N Engl J Med 2003; 349: 52–9. 17. Esteves CA, Munoz JS, Braga S et al. Immediate and longterm follow-up of percutaneous balloon mitral valvuloplasty in pregnant patients with rheumatic mitral stenosis. Am J Cardiol 2006; 98: 812–16. 18. Elkayam U, Bitar F. Valvular heart disease and pregnancy part I: native valves. J Am Coll Cardiol 2005; 46: 223–30. 19. Elkayam U, Bitar F. Valvular heart disease and pregnancy: part II: prosthetic valves. J Am Coll Cardiol 2005; 46: 403–10. 20. Meijboom LJ, Vos FE, Mulder BJ et al. Pregnancy and aortic root growth in the Marfan syndrome: a prospective study. Eur Heart J 2005; 26: 914–20.

Cardiac disease in pregnancy

21. Milewicz DM, Dietz HC, Miller DC. Treatment of aortic disease in patients with Marfan syndrome. Circulation 2005; 111: e150–7. 22. Lindheimer MD, Katz AI. Sodium and diuretics in pregnancy. N Engl J Med 1973; 288: 891–4. 23. Thaman R, Varnava A, Hamid MS et al. Pregnancy related complications in women with hypertrophic cardiomyopathy. Heart 2003; 89: 752–6. 24. deSwiet M. Antihypertensive drugs in pregnancy. Br Med J (Clin Res Ed) 1985; 291: 365–6. 25. Chow T, Galvin J, McGovern B. Antiarrhythmic drug therapy in pregnancy and lactation. Am J Cardiol 1998; 82: 581–621. 26. Bartalena L, Bogazzi F, Martino E et al. Effects of amiodarone administration during pregnancy on neonatal

27. 28.

29.

30.

31.

735

thyroid function and subsequent neurodevelopment. J Endocrinol Invest 2001; 24: 116–30. Rosemond RL. Cardioversion during pregnancy. JAMA 1993; 269: 3167. Natale A, Davidson T, Newby K et al. Implantable cardioverter-defibrillators and pregnancy: a safe combination? Circulation 1997; 96: 2808–12. Chan WS, Anand S, Ginsberg JS. Anticoagulation of pregnant women with mechanical heart valves: a systematic review of the literature. Arch Intern Med 2000; 160: 191–6. Vitale N, De Feo M, Cotrufo M. Dose-dependent fetal complications of warfarin in pregnant women with mechanical heart valves. J Am Coll Cardiol 1999; 33: 1637–41. Pisani RJ, Rosenow EC 3rd. Pulmonary edema associated with tocolytic therapy. Ann Intern Med 1989; 110: 714–18.

49 Maternal diseases and therapies affecting the fetal cardiovascular system Salim Kees and Eyal Schiff

The biological interactions generally referred to as ‘fetomaternal interactions’ come into play during intrauterine life and are remarkably harmonious for an association between two genetically disparate individuals. This unique relationship exposes all fetal organs including the heart to the to the influence of maternal health. Fetal cardiac development occurs in the first trimester of pregnancy; however, maternal health the fetal heart in later pregnancy. This chapter deals with the various mechanisms whereby maternal health influences the fetal cardiovascular system.

Congenital and genetic heart diseases Cardiac malformations are the most common congenital malformations and occur in approximately 1% of liveborn children, and a much higher proportion of spontaneous abortions and stillbirths.1,2 The commonest form of congenital heart disease (CHD) is a ventricular septal defect, which occurs in 30–40% of all children with CHD. Approximately 5–8% of CHD cases are due to chromosomal abnormalities; 3% are due to classical Mendelian gene defects, with a correspondingly high recurrence risk in first-degree relatives.2 Genetic mutations can affect a variety of functions in early heart development, and have been associated with human congenital heart disease.3 It has been reported that the risk of any congenital heart defect is substantially higher if the affected parent is the mother rather than the father.4 The recurrence risk is 2–6% for future siblings, and 1–10% for offspring. However, there are families with a much higher transmission risk.2 Women of 35 years of age or older have twice the risk of fetal death compared with their younger counterparts, even when controlled for recognized coexisting conditions

that contribute to fetal death.5 Advanced maternal age is associated with a significantly increased risk of congenital malformations not caused by aneuploidy. Hollier et al6 examined the effect of maternal age on the incidence of non-chromosomal malformations. The odds ratio for cardiac defects was 3.95 times higher in infants of women above 40 years of age (95% confidence interval (CI) 1.70–9.17) than in those of women aged 20–24 years.6 The risk of congenital heart disease in a pregnancy complicated by maternal congenital heart disaese is 2–4%, which is 3–5 times the incidence of heart disease in the general population. In addition, such defects are mostly concordant, i.e. the defect in the child is usually the same as that of the mother.7 However, if the mother’s defect was due to an external factor such as exposure to a teratogenic drug, or rubella, the fetus would have a similar risk of birth defects to that in the general population. Fetal echocardiography is presently considered the optimal modality for diagnosing and evaluating fetal congenital heart disease.8 Heart defects which can be recognized prenatally by ultrasound are associated with a chromosomal aberration in 40% of cases.9 Therefore, fetal karyotyping should be offered whenever a cardiac defect is diagnosed.9,10 The diagnosis of a chromosomal aberration in the presence of a cardiac malformation will affect the prognosis and obstetric and neonatal management.10 Congenital heart defects are prevalent in patients with autosomal aneuploidies such as trisomies 21, 18, and 13 and other chromosomal anomalies such as duplications, deletions, and mosaic trisomies.11 Turner syndrome is associated with a high prevalence of congenital heart defects. Coarctation of the aorta is the most frequent, accounting for 80–95% of anomalies. Other cardiac anomalies include bicuspid aortic valve, aortic stenosis, and hypoplastic left heart syndrome.12–15 Fragile X syndrome is associated with similar cardiac defects to those seen in connective tissue abnormalities such as Marfan and Ehler–Danlos syndromes. Aortic root dilatation is found in 52% of patients with fragile

738

Fetal Cardiology

Infectious causes of congenital cardiac anomalies

Figure 49.1 Two rhabdomyomas (arrows) are seen in this four-chamber view of the heart (reproduced with permission from reference 31).

X syndrome, and 22% have mitral valve prolapse.16,17 Marfan syndrome is characterized by abnormalities in the skeletal, cardiovascular, and ocular Systems and occurs with an incidence of 1–2/20 000. There is an autosomal dominant inheritance pattern with high penetrance. Premature death usually occurs due to cardiovascular abnormalities.18 Almost all affected persons have mitral valve prolapse, and 40–80% show aortic root enlargement.19–21 Neurofibromatosis, another autosomal dominant disorder, may be associated with cardiomyopathy,22 and involvement of the entire arterial tree from the proximal aorta down to the small arteries.23,24 Noonan syndrome is a relatively common genetic disease affecting between 1/1000 and 1/1500 newborns.25 Approximately 50% of patients with Noonan syndrome have congenital heart defects.26 Although almost all heart defects have been reported in this syndrome, the characteristic cardiac abnormalities are pulmonary stenosis, atrial septal defect, and hypertrophic cardiomyopathy.27 Cardiac tumors have been reported to occur in approximately two-thirds of patients with tuberous sclerosis28–30 (Figure 49.1). This disease is an autosomal dominant condition with the potential to involve the skin, brain, lungs, kidneys, and heart.32 Other cardiac lesions include endocardial fibroelastosis, aortic stenosis, pulmonary stenosis, arrhythmias, and sarcomas.33 Ultrasonography or echocardiography can diagnose cardiac rhabdomyosarcoma in utero. An association between tuberous sclerosis and Wolff–Parkinson– White syndrome has been reported.34

Infections contracted during pregnancy are often passed unnoticed; however, the causative agent may still cross the placenta and infect the developing embryo and fetus.35 Few of these agents will affect the fetal heart specifically.36 Rubella virus, Toxoplasma gondii, and coxsackievirus B are the principal infectious agents that can affect the fetal heart.37 The damage caused by these organisms is dependent on the organism, gestational age at the time of infection, and maternal immunity status, i.e. whether it is a primary infection or whether an effective immune response has previously been mounted.38,39 The embryo is most susceptible to cardiac malformations between 19 and 45 days of gestation. However, subsequent myocarditis can disrupt cardiac structures that have previously developed normally.38 Common sonographic abnormalities, although nonspecific, may be indicative of fetal viral infections. These include growth restriction, ascites, hydrops, ventriculomegaly, intracranial calcifications, hydrocephaly, microcephaly, cardiac anomalies, hepatosplenomegaly, echogenic bowel, placentomegaly, and abnormal amniotic fluid volume. Some sonographic findings may be pathognomonic, enabling the diagnosis of a specific congenital syndrome (e.g. eye and cardiac anomalies in congenital rubella syndrome).40 Febrile illnesses are associated with a two-fold higher risk of any heart defect in the offspring if contacted in the first trimester of pregnancy.41 Specific groups of defects that have been shown to be associated with maternal febrile illness include pulmonic stenosis, all right-sided obstructive defects, tricuspid atresia, coarctation of the aorta, all left-sided obstructive defects, conotruncal defects, and ventricular septal defects (VSDs).42 Disordered apoptosis has been postulated to be a possible mechanism. Apoptosis is known to be involved in cardiac morphogenesis. In the development of the cardiac outflow tract, apoptosis is responsible for the common arterial trunk dividing into the ascending aorta and pulmonary arteries. Fever and infection can disrupt this normal pathway.43 Infectious agents can affect the heart by one or more of three pathways: inhibition of cell growth, cytolysis, or interference with blood supply.44 Although recognition of cardiac malformations both antenatally and neonatally has been improved by non-invasive ultrasound imaging, our ability to verify the causality of these anomalies as a result of maternal infection remains limited with regard to most suspected infectious organisms.39 When a cardiac structural anomaly is identified in utero, congenital infections should be suspected and confirmed or excluded by amniocentesis or cordocentesis. Rubella is the only infectious agent for which specific congenital cardiac anomalies have been identified.38

Maternal diseases and therapies affecting the fetal cardiovascular system

Fetal damage and the development of congenital anomalies generally occur after infection in the first 16 weeks of gestation.45 Miller et al46 have demonstrated the inverse correlation between risk, severity, and frequency of fetal infection to gestational age at the time of maternal infection. Generally, the earlier is the rubella infection, the more severe are the fetal damage and malformation. Cardiovascular anomalies occur in at least two-thirds of severely affected fetuses.47 The cardiovascular anomalies associated with rubella infection are patent ductus arteriosus, pulmonary artery branch stenosis, pulmonary vein stenosis, ventricular septal defect, atrial septal defect, coarctation of the aortic isthmus, and tetralogy of Fallot.46,47 Myocarditis can also accompany congenital rubella infection leading to death in utero and to neonatal morbidity and mortality.48 Congenital toxoplasmosis usually occurs as a result of primary maternal infection. The risk of fetal infection increases during pregnancy, although the risk of severe disease decreases.49 Desmonts and Couvreur50 found that 17% of first trimester pregnancies with acute maternal toxoplasmosis and 24% of second trimester pregnancies resulted in infected infants. However, severe infection or stillbirths occurred in 75% of women after first trimester infection, and 0% after third trimester infection. The most useful tests for confirmation of fetal infection are ultrasound examination, and amniocentesis for the detection of toxoplasma DNA in amniotic fluid.51,52 Cardiac malformations have been reported to occur sporadically in severely affected infants with multiorgan involvement.38 Myocarditis with lymphocytic infiltrates and focal necrosis has been reported in the neonate;53 in utero heart failure and hydrops may also occur.54,55 Coxsackieviruses A and B are known to be capable of inducing myocarditis leading to either fetal or neonatal cardiac failure.56 Case reports suggest that these viruses may infect the fetus at any time in gestation, since nonimmune hydrops fetalis from cardiac failure has been seen early in pregnancy.57 The most common malformations associated with this group of viruses are patent ductus arteriosus and ventricular septal defects.58 The presence of maternal antibody to cytomegalovirus (CMV) before conception provides substantial protection against the deleterious effects of CMV infection on the newborn. Primary maternal CMV infection during pregnancy may be associated with severe sequelae.59 However, there are only a few case reports of fetal cardiovascular abnormalities associated with cytomegalovirus.60 These include: fetal heart block, supraventricular tachycardia, myocarditis, and cardiomyopathy (Figure 49.2)56–60,62,63. Maternal first trimester mumps infection has been reported to be associated with fetal cardiac fibroelastosis,64,65 but this association remains controversial. An association has been reported66,67 between fetal parvovirus B19 infection and fetal anomalies and hydrops fetalis. The incidence of parvovirus B19 infection in cases of ‘idiopathic’ nonimmune hydrops fetalis (NIHF) has been reported to be

739

Figure 49.2 An enlarged heart with dilated chambers, right greater than left, is seen in the fetus with cytomegalovirus. SP, spine (reproduced with permission from reference 61).

approximately 16%,68 when polymerase chain reaction (PCR)-based methods are used. Parvovirus infection can cause severe fetal anemia as a result of fetal erythroid progenitor cell infection combined with a shortened halflife of erythrocytes, causing high output cardiac failure and therefore non-immune hydrops fetalis. Myocardial infarction and myocarditis have been attributed to parvovirus infection.69,70 The P antigen expressed on fetal cardiac myocytes enables parvovirus B19 to infect myocardial cells and produce myocarditis that aggravates the cardiac failure.71 Fetal hydrops and myocarditis have been reported secondary to congenital infection with Trypanosoma cruzi causing Chagas disease.72 Vertical infection with human immunodeficiency virus (HIV) leads to an increased risk of dilated cardiomyopathy and inappropriate left ventricular hypertrophy (20%). Over a 2-year period, approximately 10% of HIV-infected children had congestive heart failure (CHF) or required treatment with cardiac medications.73–75

Metabolic and endocrine causes of cardiac abnormalities The fetal heart is vulnerable to metabolic abnormalities resulting from maternal diseases or teratogens. These may influence the fetal heart during organogenesis or in later gestation. The cardiac effects of metabolic abnormalities depend on the type of teratogen, concentration in the fetus, and the time of exposure. Some examples are given below. The incidence of malformations and life-threatening neonatal disorders in gestational diabetes has fallen

740

Fetal Cardiology

Figure 49.3 Hypertrophic cardiomyopathy. There is hypertrophic thickening of the interventricular septum (IVS) in this fetus of a diabetic mother. Typically, the obstructive hypertrophy is transient and usually not clinically significant. LA, left atrium; RA, right atrium; LV, left ventricle; RV, right ventricle (reproduced with permission from reference 61).

dramatically.76 Most patients with pregestational diabetes can can be assured of a safe pregnancy and a healthy infant. However, there is still a small risk of congenital malformations.77 Diabetes appears to induce malformations before the seventh week of gestation.78 Although the exact mechanisms responsible for diabetic embryopathy remain obscure, one hypothesis is that abnormal glucose disrupts expression of a regulatory gene in the embryo, leading to apoptotic cellular changes thought to be embryotoxic.79 Oxidative stress and generation of free radicals is another possible mechanism, as antioxidants have been reported to prevent diabetic embryopathy in animal studies.80–82 The absolute risk of a cardiovascular malformation has been reported as 8.5 per 100 live births in infants of diabetic mothers. Infants of mothers with gestational diabetes requiring insulin in the third trimester are 20 times more likely to have major cardiovascular defects than infants of non-diabetic mothers.83 In the Baltimore– Washington Infant Study,84 double-outlet right ventricle and truncus arteriosus were found to be the most common sequelae of overt maternal diabetes. Shields et al85 have evaluated the role of fetal echocardiography for predicting congenital heart disease in pregnancies complicated by overt diabetes. Detailed fetal echocardiographic imaging in all patients with initial glycosylated hemoglobin (HbA1c) levels above the upper limit of normal of 6.1% was recommended.85 The importance of fetal echocardiographic imaging in diabetic pregnancies has also been confirmed by other authors.86 The most common cardiovascular anomalies seen in infants of diabetic mothers are transposition of the great vessels,

ventricular septal defect, single ventricle, and hypoplastic left heart.87 Myocardial hypertrophy (Figure 49.3) may still occur in infants of diabetic mothers despite good metabolic control, as reflected by the decreased fraction of glycosylated fetal hemoglobin.88–90 Maternal hyperglycemia has been reported91 to result in fetal hyperinsulinemia and neonatal asymmetric septal hypertrophy with impairment in diastolic function. Fetal hyperinsulinemia coupled with a normally increased level of insulin receptors in the fetal heart causes myocardial cell hyperplasia and hypertrophy from increased fat and protein synthesis, in addition to increased glycogen deposition.91,92 Using computerized analysis of the fetal heart rate (FHR), Weiner et al93 found that the FHR pattern appears to be different in fetuses of well-controlled diabetic mothers in comparison to fetuses of non-diabetic mothers. Phenylketonuria is an autosomal recessive inborn error of phenylalanine metabolism due to a defect in the enzyme phenylalanine hydroxylase, that converts phenylalanine to tyrosine.94 Untreated maternal phenylketonuria is associated with a > 6-fold increased risk of heart defects95,96 and increased risk of spontaneous abortion and maternal phenylketonuria syndrome, characterized by low birth weight, congenital heart disease, microcephaly, childhood growth failure, and cognitive impairment.97,98 Platt et al98 found that among 414 infants born to women with phenylketonuria, 31 (7.5%) were born with congenital heart disease. The most common defects were: ventricular septal defect, tetralogy of Fallot, atrial septal defect, patent ductus arteriosus, and coarctation of the aorta. Control of blood phenylalanine levels through a phenylalaninerestricted diet significantly diminished the occurrence of congenital abnormalities in the offspring of women with hyperphenylalaninemia. No cardiac defects were detected in patients with first trimester maternal phenylalanine levels less than 10 mg/dl.98 Fetal and neonatal hyperthyroidism are usually caused by transplacental passage of thyroid-stimulating immunoglobulins.99 Thyroid stimulating immunoglobulins are generally believed to be the cause of hyperthyroidism in Graves disease. Thyroid stimulating immunoglobulins may directly affect the fetal cardiovascular system. Tachycardia is the most common symptom,100 and heart failure is the most common cause of death.101–103 Tachycardia and heart failure may occur despite normal maternal thyroid function. Moreover, such antibodies may be produced after thyroidectomy, radioiodine, or immune destruction as in Hashimoto thyroiditis.99,104 Occasionally, fetal thyrotoxicosis is accompanied by fetal cardiomegaly.105 Cordocentesis might be helpful for assessing fetal thyroid function, reaching a definite diagnosis, and guiding therapy.104 Thyrotoxicosis may be a cause of non-immune hydrops, due to high-output cardiac failure.106,107 This form of non-immune hydrops can be treated with propylthiouracil.108

Maternal diseases and therapies affecting the fetal cardiovascular system

Drug exposure Drugs and environmental toxins cause approximately 2% of all congenital defects (1). A possible relationship could exist between a genetic predisposition to certain congenital anomalies and prenatal exposure to certain agents, which could cause expression of the anomaly. Angiotensin converting enzyme (ACE) inhibitors are contraindicated during the second and third trimesters of pregnancy, as in utero exposure is associated with ACE inhibitor fetopathy.109,110 In contrast, the use of ACE inhibitors has not been linked to adverse birth outcomes in the first trimester of pregnancy. However, recently Cooper et al111 studied the association between exposure to ACE inhibitors during the first trimester of pregnancy and the risk of congenital malformations. Infants exposed to ACE inhibitors were found to be at increased risk for malformations of the cardiovascular system (risk ratio 3.72, 95% CI 1.89–7.30). Cardiac malformations are common in fetal hydantoin syndrome. Ventricular and atrial septal defects, coarctation of the aorta, and patent ductus arteriosus are the most frequent.112–114 Adverse fetal effects after exposure

741

to phenytoin (Food and Drug Administration (FDA) category D) in utero are initiated via a common pharmacological mechanism: blockage of ion channels in the developing heart resulting in bradyarrhythmias, hemodynamic alterations, and hypoxia/reoxygenation damage.115,116 It should be noted that currently available data are incapable of resolving the controversy as to whether the malformations are due to the epilepsy or the anticonvulsant therapy.42 Lithium (FDA category D) has been reported to be teratogenic, increasing the incidence of defective tricuspid valves, and arterialization of the right ventricle (Ebsteins’ anomaly) (Figure 49.4).117–119 Isotretinoin is a potent human teratogen; maternal ingestion early in pregnancy leads to a distinct clinical pattern of anomalies.120 It has been associated with a significant incidence of congenital heart disease, especially transposition of the great vessels, truncus arteriosus, and other anomalies of the aorta.121,122 It has been postulated that these anomalies are mediated by a deficiency of branchial arch mesenchyme.123 Two studies have shown an association between maternal exposure to trimethoprim sulfonamide (FDA category C)

Figure 49.4 Magnified sagittal views in a fetus of 20 weeks’ gestation with Ebstein’s anomaly. (a) The inferior vena cava (IVC) drains into the dilated right atrium (RA); the eustachian valve (EV) can be seen marking the approximate position of the atrioventricular groove. The right ventricle (RV) and aorta (Ao) are also seen. (b) The displacement of the valve leaflets, especially of the posteroinferior leaflet (curved arrow), is demonstrated during systole when the leaflets are apposed. The tricuspid valve ring, which is part of the artioventricular groove, is marked by small arrowheads. (c) In a diastolic frame, the displaced posteroinferior leaflet is shown tethered to the diaphragmatic surface of the right ventricular wall. The diagrams on the right represent the important anatomic features of the echocardiogram (reproduced with permission from reference 31).

742

Fetal Cardiology

treatment during the second or third month of gestation and cardiovascular malformations. The risks were reduced if the mother received folic acid supplementation.124,125 Alterations in fetal heart rate variability and baseline may be seen with maternal administration of magnesium sulfate, commonly used as a tocolytic agent.126–131 However, this effect is transient and reversible and should not be considered as a sign of fetal distress.132 Peaceman et al133 investigated the effects of magnesium sulfate tocolysis on the biophysical profile. Fifty percent of fetuses had a non-reactive cardiotocogram while only 18% demonstrated sustained fetal breathing movements. Magnesium sulfate therapy can result in maternal hypothermia and a decrease in the fetal heart rate and beat-to-beat variability.134 β-Sympathomimetic drugs diffuse across the placenta. They may have the same serious cardiovascular effects in the fetus as in the mother.135 The most common fetal cardiovascular effects of sympathomimetics are tachycardia and tachyarrhythmias. β-Sympathomimetic agents have also been associated with fetal hydrops, stillbirth, neonatal cardiac failure, myocardial ischemia and infarction, and neonatal death.136 Prolonged therapy with β-sympathomimetics can affect the fetal myocardium.137 Interventricular septal thickening has been reported,138 correlating with duration of exposure. This finding was transient, however, and absent at follow-up.138 The mechanism of β-mimetic toxicity appears to be due to increased myocardial intracellular calcium leading to overexcitation and cell necrosis.136,139 Indomethacin is used to inhibit premature labor, but its use has been limited as it may constrict the ductus arteriosus. The ductus arteriosus connects the pulmonary artery to the descending aorta. It is essential in fetal life as it diverts most of the deoxygenated right ventricular output away from the lungs and into the descending aorta and thereby to the placenta for reoxygenation.140 Patency is maintained during gestation by locally produced and circulating prostaglandins (PGEs). As gestation proceeds, the ductus becomes less sensitive to the dilating effect of prostaglandins and more sensitive to constricting factors such as PGE synthetase inhibitors.141 Indomethacin appears to cause transient constriction of the ductus arteriosus in some fetuses, even after shortterm use.142 Intrauterine closure of the human ductus arteriosus has also been reported following maternal ingestion of salicylates, and glucocorticoids.143 A similiar effect has been reported after diclofenac administration with the development of reversible right ventricular hypertension.141 Corticosteroids and indomethacin have a synergistic effect on the incidence and severity of fetal ductus arteriosus constriction. Short-term treatment may lead to a transient effect, and has no deleterious effects on fetal or neonatal cardiac function.144 Ductal constriction can occur at any gestational age. If indomethacin tocolysis is used, weekly fetal echocardiography is indicated for the duration of therapy.145,146 The use of indomethacin should be restricted to gestational ages of < 32 weeks. In multiple

gestations each fetus should be evaluated by echocardiography, as the ductal response may vary between individual fetuses.147 Cocaine crosses the placenta and can be found in the urine of newborns. Cocaine exposure during prenatal life appears to predispose infants to structural cardiovascular malformations, electrocardiographic abnormalities, and possibly cardiopulmonary autonomic dysfunction.148 A single ventricle may result from maternal cocaine ingestion by inducing coronary occlusion in the developing fetal heart.149 An increased frequency of cardiovascular malformations was observed among 214 infants with neonatal toxicology screens showing the prevalence of cocaine in one study;150 peripheral pulmonic stenosis was the leading diagnosis, and present in greater numbers than in the general population. Cocaine is also associated with transient ST segment abnormalities in the infant. These abnormalities are consistent with transient myocardial ischemia.151 Cocaine-exposed infants show significantly greater sleep state effects on the heart rate compared to control infants.152 Marijuana use is associated with a two-fold increase in risk of isolated simple VSD.153 Maternal use of marijuana has been found to be associated with a slight increase in risk for Ebstein’s anomaly.154 Several studies have documented a wide range of teratogenic effects of alcohol consumption during pregnancy, including cardiac defects. It has been suggested that ethanol may produce fetal tissue edema and affect the turgor of the primitive cardiac loop.155 Cardiac anomalies are frequently present in fetal alcohol syndrome; the most common anomaly is atrial septal defect, but other cardiac anomalies have been reported.156 Occupational exposure to organic solvents is associated with an increased risk of VSDs;157 dyes, lacquers, and paints are associated with conotruncal malformations,158 and mineral oil products with coarctation of the aorta.159 Thalidomide (FDA category X) is a cardiac teratogen causing several malformations ranging from ventricular and atrial septal defects (ASDs) to complex conotruncal defects.160

Autoimmune disorders Collagen vascular diseases that cause vasculopathies (such as lupus) may affect the fetal cardiovascular system by causing uteroplacental insufficiency, leading to intrauterine growth restriction, stillbirth, or fetal distress during labor.161 Placental pathology in systemic lupus erythematosus (SLE) is characterized by decidual vasculopathy and infarction, and when combined with antiphospholipid syndrome (APS), infarction can be extensive.162 Isolated congenital heart block (CHB) in utero is strongly associated with autoantibodies to intracellular

Maternal diseases and therapies affecting the fetal cardiovascular system

ribonucleoproteins SSB/La, SSA/Ro, and SSA/Ro. Fetal and neonatal diseases are presumed to be due to the transplacental passage of these immunoglobulin G (IgG) autoantibodies from the mother with SLE, or Sjogren syndrome.163–165 Maternal anti-SSA(B) antibody can cause fetal myocarditis, CHB, hydrops fetalis, and intrauterine fetal death.166 The exact mechanism whereby these autoantibodies bind the fetal conduction system and elicit a local inflammatory response is still unclear.167 When anti-SSA/Ro antibodies are present in sera of mothers with connective tissue diseases (CTD), the incidence of CHB has been reported to be 1–2% in live births. Recently, other cardiac manifestations have been reported in children born to mothers bearing anti-SSA/Ro antibodies, including transient fetal first-degree heart block, prolongation of corrected QT interval, sinus bradycardia, lateonset cardiomyopathy, endocardial fibroelastosis, and other cardiac malformations.168 This chapter summarizes the maternal conditions or exposures associated with an increased risk for cardiac defects. These conditions can be divided into two groups: (1) those that have a definite relation with fetal cardiac abnormalities such as maternal rubella infection, phenylketonuria, diabetes, thalidomide exposure, vitamin A retinoids, and indomethacin tocolysis, and (2) possible risk factos that may change the risk for cardiac defects in the offspring such as maternal febrile illness. Fetal echocardiography is recommended in those cases of maternal conditions thought to affect the fetal cardiac structure and function. Although not conclusive, periconceptional intake of multivitamin supplements containing folic acid may reduce the risk of congenital cardiovascular defects in offspring, similar to the known risk reduction for neural tube defects seen with folic acid.42

References 1. Hoffman JI. Congenital heart disease: incidence and inheritance. Pediatr Clin North Am 1990; 37: 25–43. 2. Nora JJ, Nora AH. Update on counseling the family with a first-degree relative with a congenital heart defect. Am J Med Genet 1988; 29: 137–42. 3. Payne RM, Johnson MC, Grant JW, Strauss AW. Toward a molecular understanding of congenital heart disease. Circulation 1995; 91: 494–504. 4. Nora JJ, Nora AH. Maternal transmission of congenital heart diseases: new recurrence risk figures and the questions of cytoplasmic inheritance and vulnerability to teratogens. Am J Cardiol 1987; 59: 459–63. 5. Fretts RC, Schmittdiel J, McLean FH, Usher RH, Goldman MB. Increased maternal age and the risk of fetal death. N Engl J Med 1995; 333: 953–7. 6. Hollier LM, Leveno KJ, Kelly MA, MCIntire DD, Cunningham FG. Maternal age and malformations in singleton births. Obstet Gynecol 2000; 96: 701–6.

743

7. Schmaltz AA, Neudorf U, Winkler UH. Outcome of pregnancy in women with congenital heart disease. Cardiol Young 1999; 9: 88–96. 8. Stumpflen I, Stumpflen A, Wimmer M, Bernaschek G. Effect of detailed fetal echocardiography as part of routine prenatal ultrasonographic screening on detection of congenital heart disease. Lancet 1996; 348: 854–7. 9. Schwanitz G, Zerres K, Gembruch U et al. Prenatal detection of heart defects as an indication for chromosome analysis. Ann Genet 1990; 33: 79–83. 10. Paladini D, Calabro R, Palmieri S, D’Andrea T. Prenatal diagnosis of congenital heart disease and fetal karyotyping. Obstet Gynecol 1993; 81: 679–82. 11. Pierpont MEM, Gorlin RJ, Moller JH. Chromosomal abnormalities. In: Pierpont MEM, Moller JH, eds. The Genetics of Cardiovascular Disease. Boston: Martinus Nijhoff Publishing, 1987: 69–74. 12. Subramaniam PN. Turner’s syndrome and cardiovascular anomalies: a case report and review of the literature. Am J Med Sci 1989; 297: 260–2. 13. Reis PM, Punch MR, Bove EL, van de Ven CJ. Outcome of infants with hypoplastic left heart and Turner syndromes. Obstet Gynecol 1999; 93: 532–5. 14. Mazzanti L, Cacciari E. Congenital heart disease in patients with Turner’s syndrome. Italian Study Group for Turner Syndrome (ISGTS). J Pediatr 1998; 133: 688–92. 15. Mazzanti L, Prandstraller D, Tassinari D et al. Heart disease in Turner’s syndrome. Helv Paediatr Acta 1988; 43: 25–31. 16. Sreeram N, Wren C, Bhate M, Robertson P, Hunter S. Cardiac abnormalities in the fragile X syndrome. Br Heart J 1989; 61: 289–91. 17. Waldstein G, Hagerman R. Aortic hypoplasia and cardiac valvular abnormalities in a boy with fragile X syndrome. Am J Med Genet 1988; 30: 83–98. 18. Roberts WC, Honig HS. The spectrum of cardiovascular disease in the Marfan syndrome: a clinico-morphologic study of 18 necropsy patients and comparison to 151 previously reported necropsy patients. Am Heart J 1982; 104: 115–35. 19. Brown OR, DeMots H, Kloster FE et al. Aortic root dilatation and mitral valve prolapse in Marfan’s syndrome: an echocardiographic study. Circulation 1975; 52: 651–7. 20. Pyeritz RE, McKusick VA. The Marfan syndrome: diagnosis and management. N Engl J Med 1979; 300: 772–7. 21. Geva T, Hegesh J, Frand M. The clinical course and echocardiographic features of Marfan’s syndrome in childhood. Am J Dis Child 1987; 141: 1179–82. 22. Riccardi VM. Type 1 neurofibromatosis and the pediatric patient. Curr Probl Pediatr 1992; 22: 66–106. 23. Salyer WR, Salyer DC. The vascular lesions of neurofibromatosis. Angiology 1974; 25: 510–19. 24. Donaldson MC, Ellison LH, Ramsby GR. Hypertension from isolated thoracic aortic coarctation associated with neurofibromatosis. J Pediatr Surg 1985; 20: 169–71. 25. Mendez HM, Opitz JM. Noonan syndrome: a review. Am J Med Genet 1985; 21: 493–506. 26. Nora JJ, Nora AH, Sinha AK, Spangler RD, Lubs HA. The Ullrich-Noonan syndrome (Turner phenotype). Am J Dis Child 1974; 127: 48–55.

744

Fetal Cardiology

27. Achiron R, Heggesh J, Grisaru D et al. Noonan syndrome: a cryptic condition in early gestation. Am J Med Genet 2000; 92: 159–65. 28. Gibbs JL. The heart and tuberous sclerosis. An echocardiographic and electrocardiographic study. Br Heart J 1985; 54: 596–9. 29. Jozwiak S, Kawalec W, Dluzewska J et al. Cardiac tumours in tuberous sclerosis: their incidence and course. Eur J Pediatr 1994; 153: 155–7. 30. Smith HC, Watson GH, Patel RG, Super M. Cardiac rhabdomyomata in tuberous sclerosis: their course and diagnostic value. Arch Dis Child 1989; 64: 196–200. 31. Callen PW, ed. Ultrasonography in Obstetrics and Gynecology, 4th edn. Philadelphia: Saunders, 2000: 102. 32. Cowley CG, Tani LY, Judd VE, Shaddy RE. Sinus node dysfunction in tuberous sclerosis. Pediatr Cardiol 1996; 17: 51–2. 33. Muhler EG, Turniski-Harder V, Engelhardt W, von Bernuth G. Cardiac involvement in tuberous sclerosis. Br Heart J 1994; 72: 584–90. 34. O’Callaghan FJ, Clarke AC, Joffe H et al. Tuberous sclerosis complex and Wolff-Parkinson-White syndrome. Arch Dis Child 1998; 78: 159–62. 35. Ornoy A, Tenenbaum A. Pregnancy outcome following infections by coxsackie, echo, measles, mumps, hepatitis, polio and encephalitis viruses. Reprod Toxicol 2006; 21: 446–57. 36. Ghidini A, Lynch L. Prenatal diagnosis and significance of fetal infections. West J 1993; 159: 366–73. 37. Rosenberg HS. Cardiovascular effects of congenital infections. Am J Cardiovasc Pathol 1987; 1: 147–56. 38. Blattner RJ, Williamson AP, Heys FM. Role of viruses in the etiology of congenital malformations. Prog Med Virol 1973; 15: 1–41. 39. Overall JC Jr. Intrauterine virus infections and congenital heart disease. Am Heart J 1972; 84: 823–33. 40. Degani S. Sonographic findings in fetal viral infections: a systematic review. Obstet Gynecol Surv 2006; 61: 329–36. 41. Botto LD, Lynberg MC, Erickson JD. Congenital heart defects, maternal febrile illness, and multivitamin use: a population-based study. Epidemiology 2001; 12: 485–90. 42. Jenkins KJ, Correa A, Feinstein JA et al. American Heart Association Council on Cardiovascular Disease in the Young. Noninherited risk factors and congenital cardiovascular defects: current knowledge: a scientific statement from the American Heart Association Council on Cardiovascular Disease in the Young: endorsed by the American Academy of Pediatrics. Circulation 2007; 115: 2995–3014. 43. Watanabe M, Choudhry A, Berlan M et al. Developmental remodeling and shortening of the cardiac outflow tract involves myocyte programmed cell death. Development 1998; 125: 3809–20. 44. Zeichner SL, Plotkin SA. Mechanisms and pathways of congenital infections. Clin Perinatol 1988; 15: 163–88. 45. Miller E, Cradock-Watson JE, Pollock TM. Consequences of confirmed maternal rubella at successive stages of pregnancy. Lancet 1982; 2: 781–4. 46. Miller E, Cradock-Watson JE, Pollock TM. Consequences of confirmed maternal rubella at successive stages of pregnancy. Lancet 1982; 2: 781–4.

47. Catalano LW Jr, Sever JL. The role of viruses as causes of congenital diseases. Annu Rev Microbiol 1971; 25: 255–82. 48. Ainger LE, Lawyer NG, Fitch CW. Neonatal rubella myocarditis. Br Heart J 1966; 28: 691–7. 49. Stray-Pedersen B. Toxoplasmosis in pregnancy. Baillieres Clin Obstet Gynaecol 1993; 7: 107–37. 50. Desmonts G, Couvreur J. Congenital toxoplasmosis. A prospective study of 378 pregnancies. N Engl J Med 1974; 290: 1110–16. 51. Beazley DM, Egerman RS. Toxoplasmosis. Semin Perinatol 1998; 22: 332–8. 52. Holliman RE. Congenital toxoplasmosis: prevention, screening and treatment. J Hosp Infect 1995; 30(Suppl): 179–90. 53. Garcia AG, Torres AC, Pegado CS. Congenital toxoplasmic myocarditis: case report of an unusual presentation. Ann Trop Paediatr 1985; 5: 227–30. 54. Barron SD, Pass RF. Infectious causes of hydrops fetalis. Semin Perinatol 1995; 19: 493–501. 55. Zornes SL, Anderson PG, Lott RL. Congenital toxoplasmosis in an infant with hydrops fetalis. South Med J 1988; 81: 391–3. 56. Benirschke K, Swartz WH, Leopold G, Sahn D. Hydrops due to myocarditis in a fetus. Am J Cardiovasc Pathol 1987; 1: 131–3. 57. Kawana R. Cardiovascular diseases due to viruses. Heart Vessels Suppl 1985; 1: 91–6. 58. Bates HR Jr. Coxsackie virus B3 calcific pancarditis and hydrops fetalis. Am J Obstet Gynecol 1970; 106: 629–30. 59. Fowler KB, Stagno S, Pass RF et al. The outcome of congenital cytomegalovirus infection in relation to maternal antibody status. N Engl J Med 1992; 326: 663–7. 60. Karn K, Julian TM, Ogburn PL Jr. Fetal heart block associated with congenital cytomegalovirus infection. A case report. J Reprod Med 1984; 29: 278–80. 61. Rumack CM, Wilson SR, Charboneau JW, eds. Diagnostic Ultrasound, 2nd edn. St Louis: Mosby-Year Book, 1998; 2: 1151. 62. Lewis PE 2d, Cefalo RC, Zaritsky AL. Fetal heart block caused by cytomegalovirus. Am J Obstet Gynecol 1980; 136: 967–8. 63. Filloux F, Kelsey DK, Bose CL, Veasy LG, Gooch WM. Hydrops fetalis with supraventricular tachycardia and cytomegalovirus infection. Clin Pediatr (Phila) 1985; 24: 534–6. 64. St Geme JW Jr, Noren GR, Adams P Jr. Proposed embryopathic relation between mumps virus and primary endocardial fibroelastosis. N Engl J Med 1966; 275: 339–47. 65. St Geme JW Jr, Peralta H, Farias E, Davis CW, Noren GR. Experimental gestational mumps virus infection and endocardial fibroelastosis. Pediatrics 1971; 48: 821–6. 66. Ni J, Bowles NE, Kim YH et al. Viral infection of the myocardium in endocardial fibroelastosis. Molecular evidence for the role of mumps virus as an etiologic agent. Circulation 1997; 95: 133–9. 67. Wright C, Hinchliffe SA, Taylor C. Fetal pathology in intrauterine death due to parvovirus B19 infection. Br J Obstet Gynaecol 1996; 103: 133–6. 68. Essary LR, Vnencak-Jones CL, Manning SS, Olson SJ, Johnson JE. Frequency of parvovirus B19 infection in

Maternal diseases and therapies affecting the fetal cardiovascular system

69.

70.

71. 72.

73.

74.

75.

76.

77.

78. 79.

80.

81.

82.

83.

84.

nonimmune hydrops fetalis and utility of three diagnostic methods. Hum Pathol 1998; 29: 696–701. Respondek M, Bratosiewicz J, Pertynski T, Liberski PP. Parvovirus particles in a fetal-heart with myocarditis: ultrastructural and immunohistochemical study. Arch Immunol Ther Exp (Warsz) 1997; 45: 465–70. Katz VL, McCoy MC, Kuller JA, Hansen WF. An association between fetal parvovirus B19 infection and fetal anomalies: a report of two cases. Am J Perinatol 1996; 13: 43–5. Ergaz Z, Ornoy A. Parvovirus B19 in pregnancy. Reprod Toxicol 2006; 21: 421–35. Okumura M, Aparecida dos, Santos V, Camargo ME, Schultz R, Zugaib M. Prenatal diagnosis of congenital Chagas’ disease (American trypanosomiasis). Prenat Diagn 2004; 24: 179–81. Starc TJ, Lipshultz SE, Kaplan S et al. Cardiac complications in children with human immunodeficiency virus infection: Pediatric Pulmonary and Cardiac Complications of Vertically Transmitted HIV Infection (P2C2 HIV) Study Group, National Heart, Lung, and Blood Institute. Pediatrics 1999; 104: e14. Hornberger LK, Lipshultz SE, Easley KA et al. Cardiac structure and function in fetuses of mothers infected with HIV: the prospective PCHIV multicenter study. Am Heart J 2000; 140: 575–84. Lipshultz SE, Easley KA, Orav EJ et al; Pediatric Pulmonary and Cardiovascular Complications of Vertically Transmitted HIV Infection (P(2)C(2) HIV) Study Group. Cardiovascular status of infants and children of women infected with HIV-1 (P(2)C(2) HIV): a cohort study. Lancet 2002; 360: 368–73. Cousins L. Etiology and prevention of congenital anomalies among infants of overt diabetic women. Clin Obstet Gynecol 1991; 34: 481–93. Rose BI, Graff S, Spencer R, Hensleigh P, Fainstat T. Major congenital anomalies in infants and glycosylated hemoglobin levels in insulin-requiring diabetic mothers. J Perinatol 1988; 8: 309–11. Kousseff BG. Diabetic embryopathy. Curr Opin Pediatr 1999; 11: 348–52. Phelan SA, Ito M, Loeken MR. Neural tube defects in embryos of diabetic mice: role of the Pax-3 gene and apoptosis. Diabetes 1997; 46: 1189–97. Viana M, Herrera E, Bonet B. Teratogenic effects of diabetes mellitus in the rat: prevention by vitamin E. Diabetologia 1996; 39: 1041–6. Reece EA, Wu YK. Prevention of diabetic embryopathy in offspring of diabetic rats with use of a cocktail of deficient substrates and an antioxidant. Am J Obstet Gynecol 1997; 176: 790–7. Siman CM, Eriksson UJ. Vitamin E decreases the occurrence of malformations in the offspring of diabetic rats. Diabetes 1997; 46: 1054–61. Becerra JE, Khoury MJ, Cordero JF, Erickson JD. Diabetes mellitus during pregnancy and the risks for specific birth defects: a population-based case-control study. Pediatrics 1990; 85: 1–9. Ferencz C, Rubin JD, McCarter RJ, Clark EB. Maternal diabetes and cardiovascular malformations: predominance of double outlet right ventricle and truncus arteriosus. Teratology 1990; 41: 319–26.

745

85. Shields LE, Gan EA, Murphy HF, Sahn DJ, Moore TR. The prognostic value of hemoglobin A1c in predicting fetal heart disease in diabetic pregnancies. Obstet Gynecol 1993; 81: 954–7. 86. Meyer-Wittkopf M, Simpson JM, Sharland GK. Incidence of congenital heart defects in fetuses of diabetic mothers: a retrospective study of 326 cases. Ultrasound Obstet Gynecol 1996; 8: 8–10. 87. Kitzmiller JL, Gavin LA, Gin GD et al. Preconception care of diabetes. Glycemic control prevents congenital anomalies. JAMA 1991; 265: 731–6. 88. Weber HS, Copel JA, Reece EA, Green J, Kleinman CS. Cardiac growth in fetuses of diabetic mothers with good metabolic control. J Pediatr 1991; 118: 103–7. 89. Gandhi JA, Zhang XY, Maidman JE. Fetal cardiac hypertrophy and cardiac function in diabetic pregnancies. Am J Obstet Gynecol 1995; 173: 1132–6. 90. Veille JC, Hanson R, Sivakoff M, Hoen H, Ben-Ami M. Fetal cardiac size in normal, intrauterine growth retarded, and diabetic pregnancies. Am J Perinatol 1993; 10: 275–9. 91. Breitweser JA, Meyer RA, Sperling MA, Tsang RC, Kaplan S. Cardiac septal hypertrophy in hyperinsulinemic infants. J Pediatr 1980; 96: 535–9. 92. Buchanan TA, Kitzmiller JL. Metabolic interactions of diabetes and pregnancy. Annu Rev Med 1994; 45: 245–60. 93. Weiner Z, Thaler I, Farmakides G et al. Fetal heart rate patterns in pregnancies complicated by maternal diabetes. Eur J Obstet Gynecol Reprod Biol 1996; 70: 111–15. 94. Scriver CR, Clow CL. Phenylketonuria: epitome of human biochemical genetics (first of two parts). N Engl J Med 1980; 303: 1336–42. 95. Lenke RR, Levy HL. Maternal phenylketonuria and hyperphenylalaninemia: an international survey of the outcome of untreated and treated pregnancies. N Engl J Med 1980; 303: 1202–8. 96. Levy HL, Guldberg P, Guttler F et al. Congenital heart disease in maternal phenylketonuria: report from the Maternal PKU Collaborative Study. Pediatr Res 2001; 49: 636–42. 97. Kesby G. Repeated adverse fetal outcome in pregnancy complicated by uncontrolled maternal phenylketonuria. J Paediatr Child Health 1999; 35: 499–502. 98. Platt LD, Koch R, Hanley WB et al. The international study of pregnancy outcome in women with maternal phenylketonuria: report of a 12-year study. Am J Obstet Gynecol 2000; 182: 326–33. 99. Zimmerman D. Fetal and neonatal hyperthyroidism. Thyroid 1999; 9: 727–33. 100. Polak M. Hyperthyroidism in early infancy: pathogenesis, clinical features and diagnosis with a focus on neonatal hyperthyroidism. Thyroid 1998; 8: 1171–7. 101. Page DV, Brady K, Mitchell J, Pehrson J, Wade G. The pathology of intrauterine thyrotoxicosis: two case reports. Obstet Gynecol 1988; 72: 479–81. 102. Maxwell KD, Kearney KK, Johnson JW, Eagan JW Jr, Tyson JE. Fetal tachycardia associated with intrauterine fetal thyrotoxicosis. Obstet Gynecol 1980; 55(Suppl): 18S–22S. 103. Houck JA, Davis RE, Sharma HM. Thyroid-stimulating immunoglobulin as a cause of recurrent intrauterine fetal death. Obstet Gynecol 1988; 71: 1018–19.

746

Fetal Cardiology

104. Polak M, Leger J, Luton D et al. Fetal cord blood sampling in the diagnosis and the treatment of fetal hyperthyroidism in the offsprings of a euthyroid mother, producing thyroid stimulating immunoglobulins. Ann Endocrinol (Paris) 1997; 58: 338–42. 105. Ting MK, Hsu BR, Huang YY, Lin JD, Chen TC. Recurrent fetal thyrotoxicosis in a woman with Graves’ disease: case report. Chang Keng I Hsueh Tsa Chih 1999; 22: 492–7. 106. Watson WJ, Fiegen MM. Fetal thyrotoxicosis associated with nonimmune hydrops. Am J Obstet Gynecol 1995; 172: 1039–40. 107. Stulberg RA, Davies GA. Maternal thyrotoxicosis and fetal nonimmune hydrops. Obstet Gynecol 2000; 95: 1036. 108. Treadwell MC, Sherer DM, Sacks AJ, Ghezzi F, Romero R. Successful treatment of recurrent non-immune hydrops secondary to fetal hyperthyroidism. Obstet Gynecol 1996; 87: 838–40. 109. Briggs GG. Drug effects on the fetus and breast-fed infant. Clin Obstet Gynecol 2002; 45: 6–21. 110. Tabacova S, Little R, Tsong Y, Vega A, Kimmel CA. Adverse pregnancy outcomes associated with maternal enalapril antihypertensive treatment. Pharmacoepidemiol Drug Saf 2003; 12: 633–46. 111. Cooper WO, Hernandez-Diaz S, Arbogast PG et al. Major congenital malformations after first-trimester exposure to ACE inhibitors. N Engl J Med 2006; 354: 2443–51. 112. Grech V, Vella C. Atrioventricular septal defect with separate right and left atrioventricular valvar orifices in a patient with foetal hydantoin syndrome. Cardiol Young 1999; 9: 73–4. 113. Anderson RC. Cardiac defects in children of mothers receiving anticonvulsant therapy during pregnancy. J Pediatr 1976; 89: 318–19. 114. Arpino C, Brescianini S, Robert E et al. Teratogenic effects of antiepileptic drugs: use of an international database on malformations and drug exposure (MADRE). Epilepsia 2000; 41: 1436–43. 115. Danielsson B, Skold AC, Azarbayjani F, Ohman I, Webster W. Pharmacokinetic data support pharmacologically induced embryonic dysrhythmia as explanation to Fetal Hydantoin Syndrome in rats. Toxicol Appl Pharmacol 2000; 163: 164–75. 116. Azarbayjani F, Danielsson BR. Pharmacologically induced embryonic dysrhythmia and episodes of hypoxia followed by reoxygenation: a common teratogenic mechanism for antiepileptic drugs? Teratology 1998; 57: 117–26. 117. Zalzstein E, Koren G, Einarson T, Freedom RM. A casecontrol study on the association between first trimester exposure to lithium and Ebstein’s anomaly. Am J Cardiol 1990; 65: 817–18. 118. Sipek A. Lithium and Ebstein’s anomaly. Cor Vasa 1989; 31: 149–56. 119. Long WA, Willis PW 4th. Maternal lithium and neonatal Ebstein’s anomaly: evaluation with cross-sectional echocardiography. Am J Perinatol 1984; 1: 182–4. 120. Benke PJ. The isotretinoin teratogen syndrome. JAMA 1984; 251: 3267–9. 121. Lammer EJ, Chen DT, Hoar RM et al. Retinoic acid embryopathy. N Engl J Med 1985; 313: 837–41. 122. Stern RS, Rosa F, Baum C. Isotretinoin and pregnancy. J Am Acad Dermatol 1984; 10: 851–4.

123. Hummler H, Korte R, Hendrickx AG. Induction of malformations in the cynomolgus monkey with 13-cis retinoic acid. Teratology 1990; 42: 263–72. 124. Hernandez-Diaz S, Werler MM, Walker AM, Mitchell AA. Folic acid antagonists during pregnancy and the risk of birth defects. N Engl J Med 2000; 343: 1608–14. 125. Czeizel AE, Rockenbauer M, Sorensen HT, Olsen J. The teratogenic risk of trimethoprim-sulfonamides: a population based case-control study. Reprod Toxicol 2001; 15: 637–46. 126. Hamersley SL, Landy HJ, O’Sullivan MJ. Fetal bradycardia secondary to magnesium sulfate therapy for preterm labor. A case report. J Reprod Med 1998; 43: 206–10. 127. Hallak M, Martinez-Poyer J, Kruger ML et al. The effect of magnesium sulfate on fetal heart rate parameters: a randomized, placebo-controlled trial. Am J Obstet Gynecol 1999; 181: 1122–7. 128. Babaknia A, Niebyl JR. The effect of magnesium sulfate on fetal heart rate baseline variability. Obstet Gynecol 1978; 51(Suppl): 2s–4s. 129. Wright JW, Ridgway LE, Wright BD, Covington DL, Bobitt JR. Effect of MgSO4 on heart rate monitoring in the preterm fetus. J Reprod Med 1996; 41: 605–8. 130. Atkinson MW, Belfort MA, Saade GR, Moise KJ Jr. The relation between magnesium sulfate therapy and fetal heart rate variability. Obstet Gynecol 1994; 83: 967–70. 131. Hiett AK, Devoe LD, Brown HL, Watson J. Effect of magnesium on fetal heart rate variability using computer analysis. Am J Perinatol 1995; 12: 259–64. 132. Lin CC, Pielet BW, Poon E, Sun G. Effect of magnesium sulfate on fetal heart rate variability in preeclamptic patients during labor. Am J Perinatol 1988; 5: 208–13. 133. Peaceman AM, Meyer BA, Thorp JA, Parisi VM, Creasy RK. The effect of magnesium sulfate tocolysis on the fetal biophysical profile. Am J Obstet Gynecol 1989; 161: 771–4. 134. Cardosi RJ, Chez RA. Magnesium sulfate, maternal hypothermia, and fetal bradycardia with loss of heart rate variability. Obstet Gynecol 1998; 92: 691–3. 135. Thorkelsson T, Loughead JL. Long-term subcutaneous terbutaline tocolysis: report of possible neonatal toxicity. J Perinatol 1991; 11: 235–8. 136. Katz VL, Seeds JW. Fetal and neonatal cardiovascular complications from beta-sympathomimetic therapy for tocolysis. Am J Obstet Gynecol 1989; 161: 1–4. 137. Gemelli M, De Luca F, Manganaro R et al. Transient electrocardiographic changes suggesting myocardial ischaemia in newborn infants following tocolysis with betasympathomimetics. Eur J Pediatr 1990; 149: 730–3. 138. Nuchpuckdee P, Brodskyr N, Porat R. Ventricular septal thickness and cardiac function in neonates after in utero ritodrine exposure. J Pediatr 1986; 109: 687–91. 139. Petersen R, Carter LS, Chescheir NC, Katz VL, Cefalo RC. Effects of terbutaline sulfate on fetal cardiac function. Am J Obstet Gynecol 1989; 161: 509–12. 140. Clyman RI. Ibuprofen and patent ductus arteriosus. N Engl J Med 2000; 343: 728–30. 141. Rein AJ, Nadjari M, Elchalal U, Nir A. Contraction of the fetal ductus arteriosus induced by diclofenac. Case report. Fetal Diagn Ther 1999; 14: 24–5.

Maternal diseases and therapies affecting the fetal cardiovascular system

142. Moise KJ Jr, Huhta JC, Sharif DS et al. Indomethacin in the treatment of premature labor. Effects on the fetal ductus arteriosus. N Engl J Med 1988; 319: 327–31. 143. Mielke G, Steil E, Breuer J, Goelz R. Circulatory changes following intrauterine closure of the ductus arteriosus in the human fetus and newborn. Prenat Diagn 1998; 18: 139–45. 144. Levy R, Matitiau A, Ben Arie A et al. Indomethacin and corticosteroids: an additive constrictive effect on the fetal ductus arteriosus. Am J Perinatol 1999; 16: 379–83. 145. Eronen M, Pesonen E, Kurki T, Ylikorkala O, Hallman M. The effects of indomethacin and a beta-sympathomimetic agent on the fetal ductus arteriosus during treatment of premature labor: a randomized double-blind study. Am J Obstet Gynecol 1991; 164: 141–6. 146. Vermillion ST, Scardo JA, Lashus AG, Wiles HB. The effect of indomethacin tocolysis on fetal ductus arteriosus constriction with advancing gestational age. Am J Obstet Gynecol 1997; 177: 256–9. 147. Moise KJ Jr. Effect of advancing gestational age on the frequency of fetal ductal constriction in association with maternal indomethacin use. Am J Obstet Gynecol 1993; 168: 1350–3. 148. Lipshultz SE, Frassica JJ, Orav EJ. Cardiovascular abnormalities in infants prenatally exposed to cocaine. J Pediatr 1991; 118: 44–51. 149. Shepard TH, Fantel AG, Kapur RP. Fetal coronary thrombosis as a cause of single ventricular heart. Teratology 1991; 43: 113–17. 150. Lipshultz SE, Frassica JJ, Orav EJ. Cardiovascular abnormalities in infants prenatally exposed to cocaine. J Pediatr 1991; 118: 44–51. 151. Mehta SK, Finkelhor RS, Anderson RL et al. Transient myocardial ischemia in infants prenatally exposed to cocaine. J Pediatr 1993; 122: 945–9. 152. Regalado MG, Schechtman VL, Del Angel AP, Bean XD. Cardiac and respiratory patterns during sleep in cocaineexposed neonates. Early Hum Dev 1996; 44: 187–200. 153. Williams LJ, Correa A, Rasmussen S. Maternal lifestyle factors and risk for ventricular septal defects. Birth Defects Res A Clin Mol Teratol 2004; 70: 59–64.

747

154. Ferencz C, Correa-Villasenor A, Loffredo CA, eds. Genetic and Environmental Risk Factors of Major Cardiovascular Malformations: The Baltimore–Washington Infant Study: 1981–1989. Armonk, NY: Futura Publishing, 1997. 155. Clarren SK, Smith DW. The fetal alcohol syndrome. N Engl J Med 1978; 298: 1063–7. 156. Sardor GG, Smith DF, MacLeod PM. Cardiac malformations in the fetal alcohol syndrome. J Pediatr 1981; 98: 771–3. 157. Tikkanen J, Heinonen OP. Risk factors for ventricular septal defect in Finland. Public Health 1991; 105: 99–112. 158. Tikkanen J, Heinonen OP. Risk factors for conal malformations of the heart. Eur J Epidemiol 1992; 8: 48–57. 159. Tikkanen J, Heinonen OP. Risk factors for coarctation of the aorta. Teratology 1993; 47: 565–72. 160. Smithells RW, Newman CG. Recognition of thalidomide defects. J Med Genet 1992; 29: 716–23. 161. Long WA, ed. Fetal and Neonatal Cardiology. Philadelphia: WB Saunders, 1990: 139–14. 162. Meng C, Lockshin M. Pregnancy in lupus. Curr Opin Rheumatol 1999; 11: 348–51. 163. Buyon JP. Neonatal lupus: bedside to bench and back. Scand J Rheumatol 1996; 25: 271–6. 164. Hamilton RM, Lee-Poy M, Kruger K, Silverman ED. Investigative methods of congenital complete heart block. J Electrocardiol 1998; 30(Suppl): 69–74. 165. Paredes RA, Morgan H, Lachelin GC. Congenital heart block associated with maternal primary Sjogren’s syndrome. Case report. Br J Obstet Gynaecol 1983; 90: 870–1. 166. Klassen LR. Complete congenital heart block: a review and case study. Neonatal Netw 1999; 18: 33–42. 167. Smeenk RJ. Immunological aspects of congenital atrioventricular block. Pacing Clin Electrophysiol 1997; 20: 2093–7. 168. Clancy RM, Neufing PJ, Zheng P et al. Impaired clearance of apoptotic cardiocytes is linked to anti-SSA/Ro and -SSB/La antibodies in the pathogenesis of congenital heart block. J Clin Invest 2006; 116: 2413–22.

50 Congenital heart disease and the central nervous system: a perinatal perspective Amanda Shillingford and Jack Rychik

Introduction An estimated 35 000 children are born each year in the United States with congenital heart disease, and onethird of these children are expected to undergo palliative or corrective surgical repair in the first year of life.1 Advances in surgical techniques and perioperative management, as well as improvements in prenatal diagnosis and management, have contributed to the decreased morbidity and improved survival of children born with congenital heart disease.2 Accordingly, increased attention has been directed toward understanding the longerterm developmental and functional outcomes of this growing population of survivors. During the past decade, investigators have identified an increased incidence of neurodevelopmental and neurobehavioral impairment in school-age children with congenital heart disease. Studies have shown that cognitive function for children with congenital heart disease as a whole is generally within the normal range.3–5 However, these children are at risk for deficits in visual–spatial and visual–motor skills and gross and fine motor skills, as well as impairment of speech, language, and executive functioning.3,4,6–10 More recent reports have also cited inattention and hyperactivity as emerging issues for the school-age population.11–14 The long-term impact that these morbidities will have on these children as they approach adulthood may contribute to school failure, low self-esteem, poor social skills, and inability to maintain employment.15,16 In order to improve the neurological and functional outcomes in this population, it is important to appreciate the influential contributing factors. In this chapter we discuss the phenomenon of central nervous system deficits found in congenital heart disease and review the perinatal elements which may contribute.

Possible etiologies to the central nervous system abnormalities seen in congenital heart disease A variety of factors may potentially contribute to the central nervous system (CNS) abnormalities seen in congenital heart disease. These can be broadly categorized as related to the surgical correction (perioperative), or inherently native and unrelated to the surgery (preoperative).

Perioperative causes The etiology of neurological impairment after cardiac surgery is multifactorial. The durations of cardiopulmonary bypass and deep hypothermic circulatory arrest (as well as related reperfusion injury and embolic phenomena) are frequently implicated as significant causes of neurologic impairment in the population of children undergoing neonatal cardiac surgery. The precise impact that these necessary surgical support techniques have on the neonatal brain remains poorly understood, but durations of hypothermic circulatory arrest longer than 40 minutes have been associated with a greater incidence of neurologic impairment.5,8,17–20 Modifications of intraoperative support techniques, such as temperature management, hemodilution, and the specific acid–base strategy have also been implicated as risk factors for neurologic complications.21,22 Other uncontrollable factors include lower parental intelligence quotient (IQ) and socioeconomic status, which have been associated with worse performance on subsequent childhood neurodevelopmental testing.8,23

750

Fetal Cardiology

An association between acute postoperative events, such as seizures and tone abnormalities, and later neurodevelopmental impairment exists.3,24–26 Hospital length of stay has also been shown to be a predictor for poor neurodevelopmental outcome.25,27 In addition, periventricular leukomalacia, a finding noted on brain imaging, has been reported in up to 54% of neonates after cardiac surgery.28,29 Periventricular leukomalacia is characterized by lesions in the cerebral white matter, and is typically associated with cerebral ischemia and hypoxia (Figure 50.1). The presence of this central nervous system abnormality in infants has been linked to developmental delay, motor delay, and attention deficit hyperactivity disorder in children without congenital heart disease.30,31

heart disease as well as a wide spectrum of neurological deficits. Genetic predispositions may also render an individual more susceptible to adverse neurological effects after cardiac surgery. Gaynor and associates have found an association between the apolipoprotein ε2 allele and worse performance on developmental testing in infants 1 year after cardiac surgery.33 Longer follow-up of this cohort is ongoing. There may be other genes that control the host response to potential injuries that the infant with congenital heart disease could incur. The concept of a genetically controlled host response influencing neurodevelopmental outcome as much as the severity of any insult or ‘stressor’ applied is an intriguing one, and is worthy of further investigation.

Preoperative causes

CNS structural abnormalities

While there are a number of perioperative events which can potentially affect the clinical status of children with congenital heart disease, the prenatal condition of the fetus is likely a significant contributor to eventual outcome. Preoperative neurological findings are abnormal in up to 50% of some neonates with congenital heart disease, including seizures, tone abnormalities, motor function deficits, and poor orienting responses.6,32 There are numerous prenatal factors to consider which can contribute to the neurological impairment appreciated in older children with congenital heart disease.

Congenital central nervous system anatomical abnormalities are known to coexist with the presence of some forms of congenital heart disease. These anatomical abnormalities include microcephaly, holoprosencephaly, agenesis of the corpus callosum, and delayed closure of the operculum.29,34–36 Periventricular leukomalacia in the newborn is a marker of fetal and early postnatal cerebral ischemia, and was detected by magnetic resonance imaging in 17% of a cohort of preoperative newborns with congenital heart disease.29 Head circumference at birth can be considered a predictor for fetal brain growth and development, and microcephaly is indicative of suboptimal brain growth. Limperopoulos and associates published a series of studies evaluating the neurological and functional outcomes of newborns with congenital heart disease (excluding hypoplastic left heart syndrome).6,7,10,25 The authors found microcephaly (defined as head circumference less than or equal to the 2nd centile) present in 31% of the

Genetics Genetic syndromes, such as trisomy 21, Williams syndrome, Noonan syndrome, and microdeletion of chromosome 22, are associated with a high incidence of congenital

Figure 50.1 Axial (primary) and coronal and sagittal reconstructions demonstrating white matter hyperintensities (arrows) in a periventricular watershed distribution in a newborn with hypoplastic left heart syndrome.

Congenital heart disease and the central nervous system

newborns, and the presence of preoperative microcephaly was significantly associated with abnormal neurologic examinations, fine motor impairment, and behavioral difficulties.25 A high incidence of preoperative microcephaly has similarly been reported in newborns with hypoplastic left heart syndrome, although a direct correlation between microcephaly and specific neurologic impairment has not been delineated.34,37

Anthropometric growth and the phenomenon of ‘brain-sparing’ The etiology of congenital central nervous system anomalies in newborns with congenital heart disease may be related to abnormalities of cerebral blood flow and oxygen delivery resulting from intrauterine physiological alterations directly related to the abnormal cardiac anatomy that is present. Animal studies have shown that redistribution of cardiac output occurs when a fetus without congenital heart disease is exposed to nutritional or oxygen deficiency.38–41 The socalled ‘brain-sparing effect’ refers to the observation that in a growth-restricted infant, the degree of body growth impairment is more significant than head growth impairment. The impetus for this phenomenon is thought to be related to fetal autoregulatory mechanisms which influence preferential shunting of cardiac output to the most vital organs (such as the brain) when the oxygen supply is limited. In the fetus with congenital heart disease, the associated physiologic and anatomic alterations may impede the efficacy of such autoregulatory mechanisms to sufficiently augment cerebral blood flow. Several studies in the literature have explored this relationship. Rosenthal extensively described the newborn growth parameters in 251 non-syndromic infants with transposition of the great arteries, tetralogy of Fallot, hypoplastic left heart syndrome, and coarctation of the aorta, compared to infants without congenital heart disease.42 Infants with transposition of the great arteries had similar birth weights to controls, but the head circumferences in this group were disproportionately smaller than the birth weights. In the fetus with transposition of the great arteries, relatively highly oxygenated blood returning via the ductus venosus is directed across the foramen ovale to the left ventricle, while relatively deoxygenated blood is directed toward the right ventricle. Hence, the oxygen content of the blood supply to the head and neck vessels is lower than normal since the aorta arises from the right ventricle, while the systemic blood flow to the lower body originates predominantly from the ductus arteriosus, which carries more highly oxygenated blood from the left ventricle and ductus venosus. In infants with tetralogy of Fallot, birth weight, birth length, and head circumference were all smaller compared to controls. The intrauterine physiologic alteration in this lesion consists of significant

751

intracardiac mixing resulting in blood containing a lower than normal oxygen content that is ejected into the ascending aorta. Similar to infants with tetralogy of Fallot, infants with hypoplastic left heart syndrome were smaller than controls for all growth patterns. However, the head circumference was disproportionately smaller than the weight. In hypoplastic left heart syndrome, there is significant intracardiac mixing so that any blood ejected antegrade into the aorta contains a lower oxygen content. In addition, the aortic arch is hypoplastic, resulting in a relative anatomic obstruction to cerebral blood flow. Thus, oxygen and nutrient delivery to the fetal brain may be impaired because of a decreased volume of blood delivery and decreased oxygen content of the cerebral blood supply. Finally, the infants with coarctation of the aorta were found to have smaller birth weights and birth lengths, but slightly larger head circumferences compared to controls. These findings are consistent with the observations in the previous lesions since the anatomic obstruction to blood flow is typically near the aortic isthmus and is distal to the origin of the head and neck vessels. Thus, the torso and lower extremities receive predominately deoxygenated blood from the ductus arteriosus, while the head and neck and vessels display normal fetal physiologic oxygen delivery (Table 50.1). The findings of Rosenthal provide an intriguing platform from which to further investigate the relationship between physiologic alterations related to different forms of congenital heart disease and growth characteristics in utero. However, no causal relationship can be concluded from these data. Moreover, the strong possibility of general growth impairment as part of a developmental syndrome of congenital heart disease cannot be overlooked in this discussion. In order to further explore the association between fetal anatomy and growth abnormalities, a study from the Children’s Hospital of Philadelphia investigated the relationship between aortic arch size and head circumference in 129 newborns with hypoplastic left heart syndrome.37 In this cohort, birth weight and head circumference were smaller than expected for the normal newborn population, and microcephaly (head circumference less than or equal to the 3rd centile) was present in 12% of patients. Consistent with previous findings for hypoplastic left heart syndrome, head circumference was disproportionately smaller than birth weight. In addition, an important morphometric relationship between the head and the heart was noted. The presence of microcephaly was associated with a smaller ascending aortic diameter (Figure 50.2), suggesting that the degree of anatomic obstruction to cerebral blood flow attenuates the normal autoregulatory responses to increase cerebral blood flow. The consequent decrease in the delivery of oxygen and nutrients to the developing brain may result in growth impairment. Still, the precise correlation of this relationship between aortic size and head circumference is speculative, since the study

752

Fetal Cardiology

Table 50.1 Summary of Rosenthal’s description of anthropometric differences among newborns with various types of congenital heart disease (CHD) compared to matched controls42 CHD lesion

Head circumference

Weight

Length

TGA (n = 69)

↓

—

—

TOF (n = 66)

↓

↓

↓

Coarctation (n = 65)

↑

↓

↓

↓↓

↓

↓

HLHS (n = 51)

TGA, transposition of the great arteries; TOF, tetralogy of Fallot; Coarctation, coarctation of the aorta; HLHS, hypoplastic left heart syndrome.

8 7 0

6

Aorta size (mm)

5

0

4 3 2 1 p = 0.03

0 Non-microcephalic

Microcephalic

Ascending aorta

p = 0.62 Non-microcephalic

Microcephalic

Transverse aorta

Figure 50.2 These box and whisker plots demonstrate the distribution of data for the aorta diameters in microcephalic and non-microcephalic infants with hypoplastic left heart syndrome (HLHS). The ascending aorta diameters are shown on the left, and the transverse aorta diameters are shown on the right. The open boxes represent the distributions around the median aortic diameters for nonmicrocephalic newborns. The yellow boxes represent the distributions around the median aortic diameters of the microcephalic newborns. The solid line in the middle of the box corresponds to the median value and the entire box incorporates the data points falling into the lower and upper quartiles. Extreme values are represented by the vertical lines. Open circles represent outliers. Microcephalic infants with HLHS have a smaller ascending aorta, but not transverse aorta, in comparison to normocephalic infants with HLHS (p < 0.05).

does not provide direct information about cerebral blood flow and oxygen delivery in the fetus.

Neonatal cerebral blood flow Recently, sophisticated imaging techniques have been implemented for the study of cerebral blood flow. Licht and associates utilized a novel magnetic resonance imaging (MRI) technique to measure preoperative cerebral blood flow in 25 newborns with various forms of

congenital heart disease.36 The mean quantity of cerebral blood flow for the cohort was lower than the reported normal values. Seven patients had periventricular leukomalacia, and the presence of this finding was significantly associated with lower baseline cerebral blood flow values. In this small cohort, there were no significant associations between the specific congenital heart defect and cerebral blood flow. Microcephaly was present in 24% of the group, but the relationship between cerebral blood flow and head circumference was not investigated.

Congenital heart disease and the central nervous system

Ricci and associates created a single-ventricle physiology model in a piglet in order to investigate the acute effects of hypoxemia on cerebral oxygen delivery.43 Cerebral blood flow was measured immediately after surgical creation of single-ventricle anatomy using stable isotope microspheres. While cerebral blood flow was acutely increased in a control group that underwent a sham operation, there was no significant change in the single-ventricle group. Therefore, despite an acute drop in cerebral arterial oxygen saturation, no comparable increase in cerebral blood flow was observed to augment oxygen delivery. In addition, oxygen extraction by the cerebral tissue did not increase in order to compensate for oxygen delivery. The authors concluded that the findings suggest that normal autoregulatory mechanisms in the fetal brain may be impaired when exposed to single-ventricle physiology and hypoxemia.

Fetal cerebral blood flow Fetal cardiac physiology is distinct from newborn physiology since oxygenation occurs in the placenta, and fetal anatomic structures are designed to shunt highly oxygenated blood to the head and neck vessels. Thus, specific assessment of fetal cerebral blood flow is essential in order to appreciate derangements caused by abnormal cardiac anatomy. There are few studies in the literature that have investigated this relationship. Donofrio and colleagues utilized Doppler ultrasound of the middle cerebral artery and umbilical artery in order to calculate the resistance index of each region, defined as the difference between peak systolic velocity and end-diastolic velocity divided by peak systolic velocity.44 Thirty-six fetuses with congenital heart disease were studied and compared to normal. The mean ratio of cerebral resistance to placental resistance was lower in fetuses with congenital heart disease compared to controls. The measurements of mean placental resistance index alone were insignificant between the congenital heart disease group and the normal group, indicating that the difference in mean cerebral resistance index values were important factors. Moreover, only one of the control patients had a ratio less than one, while 44% of the congenital heart disease patients had a ratio less than one. Thus, the congenital heart disease group demonstrated lower cerebral resistance compared to the control group. Five different cardiac lesions were represented (hypoplastic left heart syndrome, left ventricular outflow tract obstruction, hypoplastic right heart syndrome, tetralogy of Fallot, and transposition of the great arteries), and the cerebral vascular resistance varied between the lesions. Fetuses with hypoplastic left and right heart were the most significantly affected, while the fetuses with tetralogy of Fallot and transposition were less affected compared to controls. No fetus with left ventricular outflow obstruction had a ratio of less than one. The in utero physiologic

753

alteration of hypoplastic left and right heart lesions involves intracardiac mixing resulting in the delivery of relatively deoxygenated blood to the head and neck vessels. Thus, the assumption is that in these two variations of singleventricle lesions, hypoxemia induces a decrease in the cerebral vascular resistance in order to augment oxygen delivery to the developing brain. In the fetus with hypoplastic left heart syndrome, cerebral blood flow is also restricted because of anatomic obstruction due to a diminutive ascending aorta and transverse aortic arch. In transposition of the great arteries and tetralogy of Fallot, hypoxemic blood is delivered to the ascending aorta. However, these lesions contain two ventricles which may allow for changes in the cardiac output that allow augmentation of cerebral blood flow and ultimately increased oxygen delivery. The latter observation is only speculative, and requires more aggressive investigation of fetal myocardial mechanics in association with measures of cerebral blood flow (Table 50.2). Kaltman and associates similarly utilized Doppler ultrasound of the middle cerebral artery and umbilical artery in order to calculate the pulsatility index in two regions of fetal blood flow.45 The pulsatility index is defined as the difference between the peak systolic and peak diastolic velocity divided by the mean velocity, and is a predictor for the vascular resistance. Fetuses with congenital heart disease (hypoplastic left heart syndrome, left-sided obstructive lesion, and right-sided obstructive lesion) were compared to normal fetal controls. The pulsatility index in the hypoplastic left heart syndrome group was significantly lower compared to the normal group, indicating that cerebral vascular resistance is lower in this group (Figures 50.3 and 50.4). Alternatively, the right-sided obstructive lesion

Table 50.2 Simplified summary of Donofrio’s study of 36 fetuses with various forms of CHD is shown.44 The cerebral-to-placental resistance (CPR) index was assessed. In this cohort, the umbilical artery resistance index ratio was similar in all CHD groups as well as the control group (see text for details) CHD lesion

Vascular resistance in the MCA

HLH

↓↓

HRH

↓↓

LVOTO

—

TOF

↓

TGA

↓

HLH, hypoplastic left heart; HRH, hypoplastic right heart; LVOTO, left ventricular outflow tract obstruction; TOF, tetralogy of Fallot; TGA, transposition of the great arteries; MCA, middle cerebral artery.

754

Fetal Cardiology

Figure 50.3

group had a significantly higher pulsatility index compared to the normal group, suggesting that the cerebral vascular resistance was higher in this group. The pulsatility index for the left-sided obstructive lesion group was lower than in normals, but higher than in the hypoplastic left heart syndrome group (Figure 50.5). The findings from this study suggest that fetal autoregulatory mechanisms affect the cerebral blood flow by altering the cerebral vascular resistance, and the impetus may be in part dependent on the volume of blood (and consequent overall oxygen delivery) ejected into the head and neck vessels. As previously noted, the fetus with hypoplastic left heart syndrome has two obstacles hindering adequate oxygen delivery to the developing brain. Because of intracardiac mixing, the blood that is delivered to the head and neck vessels is

Visualization of the middle cerebral artery (MCA) arising from the circle of Willis on color Doppler flow imaging of the fetal head.

3.0 2.0 * 0.7

Pl Z score

1.0 0.0

–0.13

–0.13 ** –1.1

–1.0 –2.0 –3.0 Normal (125) LSOL (21)

Figure 50.5

Figure 50.4 (a) Doppler spectral tracing of middle cerebral artery flow in a normal fetus. Arrows point to diastolic flow. Note the relatively low diastolic flow velocity in proportion to the systolic flow velocity. This low ratio suggests high vascular resistance which is seen in the normal fetus. (b) Doppler spectral tracing of middle cerebral artery flow in a fetus with hypoplastic left heart syndrome. Arrows point to the relatively increased diastolic flow velocity in proportion to systolic flow velocity, in comparison to the ratio seen in the normal fetus. This increase in diastolic flow suggests a lower vascular resistance, possibly a protective mechanism to increase cerebral blood flow in a fetus with structural impediment to such flow due to the presence of aortic hypoplasia.

RSOL (23)

HLHS (34)

Data from Kaltman et al.45 Graph of the mean and standard deviations for the pulsatility index (PI) Z scores for fetuses with normal heart (n = 125), left-sided obstructive lesions (LSOL, n = 21), right-sided obstructive lesions (RSOL, n = 23), and hypoplastic left heart syndrome (HLHS, n = 34). Normal PI values were similar to published data, as the mean Z score was −0.13. Similarly, fetuses with left-sided obstructive disease but with two good-sized ventricles and no evidence for aortic hypoplasia had normal PI Z scores. Fetuses with right-sided obstructive lesions had a significantly increased PI Z score (*p < 0.05) in comparison to normal, suggesting an increased cerebrovascular resistance. Fetuses with hypoplastic left heart syndrome had a significantly decreased PI Z score (**p < 0.01), suggesting a decreased cerebrovascular resistance relative to normal. These data support the concept that cerebrovascular resistance, as reflected by middle cerebral artery PI, is influenced by cardiac structure and blood flow patterns. In right-sided disease blood is shunted away toward the left side of the heart and the aorta, and therefore aortic flow and hence cerebral flow are potentially increased. In an attempt to control and limit cerebral blood flow, cerebrovascular resistance is therefore increased. In left-sided disease aortic flow is diminished and cerebral blood flow is often achieved through retrograde flow from the ductus arteriosus, and hence cerebrovascular resistance is diminished in an attempt to promote increased cerebral blood flow.

Congenital heart disease and the central nervous system

relatively deoxygenated. In addition, antegrade blood flow into the head and neck vessels is often severely obstructed, and the cerebral vasculature is supplied retrogradely by the ductus arteriosus into a hypoplastic aortic arch. The concept that cerebral vascular resistance is lower in hypoplastic left heart syndrome supports the theory that fetal autoregulatory mechanisms attempt to compensate for decreased oxygen and nutrient delivery by increasing the total volume of cerebral blood flow. Alternatively, fetuses with right-sided obstructive lesions may have a compensatory increase in the systemic (or left-sided) cardiac output as blood is anatomically directed away from the right toward the left side of the heart. The finding that the cerebral vascular resistance is higher in the group of rightsided anatomical lesions compared to controls also supports the theory that fetal autoregulatory mechanisms exist to control the oxygen and nutrient delivery to the brain by limiting the volume of antegrade flow via increased resistance. Finally, the left-sided obstructive lesions have antegrade flow in the ascending aorta with varying degrees of anatomic obstruction. This group of lesions demonstrates mildly elevated cerebral vascular resistance, which suggests that modification of the cerebral vascular resistance is implemented but not to the same extent as in hypoplastic left heart syndrome. The findings of cerebral resistance variability seen in the Kaltman study strongly support the notion that cerebral blood flow in the fetus is directly affected by the type of congenital heart disease present and the subsequent alterations in intracardiac flow patterns that take place. Modena and associates utilized Doppler ultrasound in 71 fetuses with congenital heart disease and compared the results to normal controls.46 These authors found no significant difference between the congenital heart disease group and the control group with respect to the mean values of middle cerebral artery pulsatility index or for mean values of the cerebral to placental resistance ratio. However, a significantly higher proportion of patients in the congenital heart disease group had middle cerebral artery pulsatility index values that were less than the 5th centile, based on published normal values (5/71 fetuses with congenital heart disease versus 0/71 controls).47 A similar trend was noted for the cerebral to placental resistance ratio (8/71 fetuses with congenital heart disease versus 2/71 controls). No significant differences were identified between the congenital heart disease groups with intracardiac mixing compared to those without intracardiac mixing. However, only six of the fetuses were in the non-mixing group for the comparison. This study may not have identified differences in middle cerebral artery pulsatility indexes as the group of congenital heart disease fetuses was anatomically heterogeneous, and structural specificity dictating an increased or decreased delivery of blood flow to the brain may play a role. Still, the above study provides additional supporting evidence that regulation of cerebral blood flow

755

is clearly abnormal in the fetus with congenital heart disease compared to the normal fetus. It is important to appreciate that while the described investigations provide compelling evidence supporting the relationship between changes in cerebral vascular resistance and the hemodynamic effects of specific forms of congenital heart disease, no study directly quantifies oxygen and nutrient delivery to the developing brain. Moreover, no studies have prospectively followed these fetuses postnatally in order to correlate changes in cerebral blood flow with abnormal neurological outcome. Still, reports in the literature have clearly demonstrated that a significant proportion of children who have undergone cardiac surgery for congenital heart disease are at increased risk for neurological deficits. Abnormal fetal and neonatal cerebral blood flow patterns as a consequence of structural heart disease may be influencing brain development. In the future, a great deal of information will be learned as more sophisticated fetal imaging techniques become available, and larger prospective studies are designed to follow the clinical outcomes of these fetuses into childhood and young adulthood.

References 1. Ferencz C, Rubin JD, McCarter RJ et al. Congenital heart disease: prevalence at livebirth. The Baltimore–Washington Infant Study. Am J Epidemiol 1985; 121: 31–6. 2. Welke KF, Shen I, Ungerleider RM. Current assessment of mortality rates in congenital cardiac surgery. Ann Thorac Surg 2006; 82: 164–70. 3. Bellinger DC, Wypij D, Kuban KC et al. Developmental and neurological status of children at 4 years of age after heart surgery with hypothermic circulatory arrest or low-flow cardiopulmonary bypass. Circulation 1999; 100: 526–32. 4. Bellinger DC, Wypij D, duDuplessis AJ et al. Neurodevelopmental status at eight years in children with dextrotransposition of the great arteries: the Boston Circulatory Arrest Trial. J Thorac Cardiovasc Surg 2003; 126: 1385–96. 5. Forbess JM, Visconti KJ, Hancock-Friesen C et al. Neurodevelopmental outcome after congenital heart surgery: results from an institutional registry. Circulation 2002; 106: I95–102. 6. Limperopoulos C, Majnemer A, Shevell MI et al. Neurodevelopmental status of newborns and infants with congenital heart defects before and after open heart surgery. J Pediatr 2000; 137: 638–45. 7. Limperopoulos C, Majnemer A, Shevell MI et al. Functional limitations in young children with congenital heart defects after cardiac surgery. Pediatrics 2001; 108: 1325–31. 8. Wernovsky G, Stiles KM, Gauvreau K et al. Cognitive development after the Fontan operation. Circulation 2000; 102: 883–9. 9. Bellinger DC, Bernstein JH, Kirkwood MW, Rappaport LA Newburger JN. Visual-spatial skills in children after open-heart surgery. J Dev Behav Pediatr 2003; 24: 169–79.

756

Fetal Cardiology

10. Majnemer A, Limperopoulos C, Shevell M et al. Long-term neuromotor outcome at school entry of infants with congenital heart defects requiring open-heart surgery. J Pediatr 2006; 148: 72–7. 11. Mahle WT, Clancy RR, Moss EM et al. Neurodevelopmental outcome and lifestyle assessment in school-aged and adolescent children with hypoplastic left heart syndrome. Pediatrics 2000; 105: 1082–9. 12. Kirshbom PM, Flynn TB, Clancy RR et al. Late neurodevelopmental outcome after repair of total anomalous pulmonary venous connection. J Thorac Cardiovasc Surg 2005; 129: 1091–7. 13. Shillingford AJ, Glanzman MM, Ittenbach RF et al. Inattention, hyperactivity, and school performance in a population of school-age children with complex congenital heart disease. Pediatrics 2008; 121: e759–67. 14. Hovels-Gurich HH, Konrad K, Skorzenski D et al. Attentional dysfunction in children after corrective cardiac surgery in infancy. Ann Thorac Surg 2007; 83: 1425–30. 15. Huntington DD, Bender WN. Adolescents with learning disabilities at risk? Emotional well-being, depression, suicide. J Learn Disabil 1993; 26: 159–66. 16. Naylor MW, Staskowski M, Kenney MC, King CA. Language disorders and learning disabilities in school-refusing adolescents. J Am Acad Child Adolesc Psychiatry 1994; 33: 1331–7. 17. Clancy RR, McGaurn SA, Wernovsky G et al. Risk of seizures in survivors of newborn heart surgery using deep hypothermic circulatory arrest. Pediatrics 2003; 111: 592–601. 18. Oates RK, Simpson JM, Turnbull JA, Cartmill TB. The relationship between intelligence and duration of circulatory arrest with deep hypothermia. J Thorac Cardiovasc Surg 1995; 110: 786–92. 19. Wypij D, Newburger JW, Rappaport LA et al. The effect of duration of deep hypothermic circulatory arrest in infant heart surgery on late neurodevelopment: the Boston Circulatory Arrest Trial. J Thorac Cardiovasc Surg 2003; 126: 1397–403. 20. Bellinger DC, Jonas RA, Rappaport LA et al. Developmental and neurologic status of children after heart surgery with hypothermic circulatory arrest or low-flow cardiopulmonary bypass. N Engl J Med 1995; 332: 549–55. 21. Jonas RA, Wypij D, Roth SJ et al. The influence of hemodilution on outcome after hypothermic cardiopulmonary bypass: results of a randomized trial in infants. J Thorac Cardiovasc Surg 2003; 126: 1765–74. 22. du Plessis AJ, Jonas RA, Wypij D et al. Perioperative effects of alpha-stat versus pH-stat strategies for deep hypothermic cardiopulmonary bypass in infants. J Thorac Cardiovasc Surg 1997; 114: 991–1000. 23. Goldberg CS, Schwartz EM, Brunberg JA et al. Neurodevelopmental outcome of patients after the Fontan operation: a comparison between children with hypoplastic left heart syndrome and other functional single ventricle lesions. J Pediatr 2000; 137: 646–52. 24. Rappaport LA, Wypij D, Bellinger DC et al. Relation of seizures after cardiac surgery in early infancy to neurodevelopmental outcome. Boston Circulatory Arrest Study Group. Circulation 1998; 97: 773–9.

25. Limperopoulos C, Majnemer A, Shevell MI et al. Predictors of developmental disabilities after open heart surgery in young children with congenital heart defects. J Pediatr 2002; 141: 51–8. 26. Limperopoulos C, Majnemer A, Rosenblatt B et al. Association between electroencephalographic findings and neurologic status in infants with congenital heart defects. J Child Neurol 2001; 16: 471–6. 27. Newburger JW, Wypij D, Bellinger DC et al. Length of stay after infant heart surgery is related to cognitive outcome at age 8 years. J Pediatr 2003; 143: 67–73. 28. Galli KK, Zimmerman RA, Jarvik GP et al. Periventricular leukomalacia is common after neonatal cardiac surgery. J Thorac Cardiovasc Surg 2004; 127: 692–704. 29. Mahle WT, Tavani F, Zimmerman RA et al. An MRI study of neurological injury before and after congenital heart surgery. Circulation 2002; 106(Suppl 1): 1109–14. 30. Back SA. Perinatal white matter injury: the changing spectrum of pathology and emerging insights into pathogenetic mechanisms. Ment Retard Dev Disabil Res Rev 2006; 12: 129–40. 31. Miller SP, Shevell MI, Patenaude Y, O’Gorman AM. Neuromotor spectrum of periventricular leukomalacia in children born at term. Pediatr Neurol 2000; 23: 155–9. 32. Limperopoulos C, Majnemer A, Shevell MI et al. Neurologic status of newborns with congenital heart defects before open heart surgery. Pediatrics 1999; 103: 402–8. 33. Gaynor JW, Gerdes M, Zackai EH et al. Apolipoprotein E genotype and neurodevelopmental sequelae of infant cardiac surgery. J Thorac Cardiovasc Surg 2003; 126: 1736–45. 34. Glauser TA, Rorke LB, Weinberg PM, Clancy RR. Congenital brain anomalies associated with the hypoplastic left heart syndrome. Pediatrics 1990; 85: 984–90. 35. van Houten JP, Rothman A, Bejar R. High incidence of cranial ultrasound abnormalities in full-term infants with congenital heart disease. Am J Perinatol 1996; 13: 47–53. 36. Licht DJ, Wang J, Silvestre DW et al. Preoperative cerebral blood flow is diminished in neonates with severe congenital heart defects. J Thorac Cardiovasc Surg 2004; 128: 841–9. 37. Shillingford A, Ittenbach RF, Marino B et al. Aortic morphometry and microcephaly in hypoplastic left heart syndrome. Cardiology Young 2007; 17: 189–95. 38. Cohn HE, Sacks EJ, Heymann MA, Rudolph AM. Cardiovascular responses to hypoxemia and acidemia in fetal lambs. Am J Obstet Gynecol 1974; 120: 817–24. 39. Rudolph AM. The fetal circulation and its response to stress. J Dev Physiol 1984; 6: 11–19. 40. Lumbers ER, Yu ZY, Gibson KJ. The selfish brain and the barker hypothesis. Clin Exp Pharmacol Physiol 2001; 28: 942–7. 41. Bishai JM, Blood AB, Hunter CJ Longo LD, Power GG. Fetal lamb cerebral blood flow (CBF) and oxygen tensions during hypoxia: a comparison of laser Doppler and microsphere measurements of CBF. J Physiol 2003; 546: 869–78. 42. Rosenthal GL. Patterns of prenatal growth among infants with cardiovascular malformations: possible fetal hemodynamic effects. Am J Epidemiol 1996; 143: 505–13. 43. Ricci M, Lombardi P, Galindo A et al. Single-ventricle physiology reduces cerebral oxygen delivery in a piglet model. Ann Thorac Surg 2005; 80: 686–93.

Congenital heart disease and the central nervous system

44. Donofrio MT, Bremer YA, Schieken RM et al. Autoregulation of cerebral blood flow in fetuses with congenital heart disease: the brain sparing effect. Pediatr Cardiol 2003; 24: 436–43. 45. Kaltman JR, Di H, Tian Z, Rychik J. Impact of congenital heart disease on cerebrovascular blood flow dynamics in the fetus. Ultrasound Obstet Gynecol 2005; 25: 32–6.

757

46. Modena A, Horan C, Visintine J et al. Fetuses with congenital heart disease demonstrate signs of decreased cerebral impedance. Am J Obstet Gynecol 2006; 195: 706–10. 47. Baschat AA, Gembruch U. The cerebroplacental Doppler ratio revisited. Ultrasound Obstet Gynecol 2003; 21: 124–7.

Index

Note: Page numbers in italic denote figures or tables. abdominal wall defects 643–4 abortion 106, 621 absent pulmonary valve syndrome and hydrops fetalis 496, 496 in tetralogy of Fallot 314, 315, 316, 496 absolute cardiac flow 532 absolute flow velocities 531, 532 diastolic 161 systolic 161 action potentials 435–6 activated T cells 35 acute aortic dissection 731, 731 adenosine 476, 686, 733 direct fetal application 476 adhesion molecules 29 adrenal artery 166 ‘adrenal gland sparing’ 393 advanced dynamic flow (ADF) 87–8 afterload 142–3, 143, 157 in fetal growth restriction 539 in hydrops fetalis 493 myocardial response 159, 386, 390, 394 through gestation 167, 168–9 in twin–twin transfusion syndrome 599 ventricular compensation 562 Alagille syndrome 103, 614 inheritance 708 alcohol teratogenicity 609, 742 alveolar units 148 American College of Radiology (ACR) antepartum obstetric guidelines 186 American Institute of Ultrasound in Medicine (AIUM) antepartum obstetric guidelines 186 American Society of Human Genetics (ASHG) 706 amiodarone 473, 474, 475–6, 475, 733 direct fetal application 476–7, 478 during pregnancy 732 amniotic cavity pressures 135 anatomy, normal 19–26, 20 early 186 anemia, fetal 363, 378, 379 fetal blood sampling 380

in hydrops fetalis 489, 503, 504 peak flow velocities in 393, 393 treatment 574 ultrasound investigation 392–3 aneuploidies 102 and heart rate changes 161 and septal defects 623, 637 aneurysm of left ventricle 364 angiogenesis natural killer (NK) cells 32 placental 29–32 villous anomalies 51–3 angiotensin-converting enzyme (ACE) inhibitors 574, 733, 741 angiotensin II 606 angipoietins (ANG) 31, 31 anomalous pulmonary artery branching 697 anomalous pulmonary venous connection 266–7, 266, 417 partial (PAPVC) 420–1, 421 prognosis 422 total (TAPVC) 422 diagnosis, 3D/4D US 229 incidence 107 and isomerisms 248, 357, 357 STIC acquisition 229 anorectal atresia 643 anthropometric growth 751–2 antiarrhythmics 472–3 direct fetal application 476–7 effectiveness 477 monitoring 470 placental effects 474 protocols 471, 474–7 risks 470 aorta anatomy 21–2 blood velocity 139, 140, 140 and cardiovascular anomalies 583 double-vessel sign 242, 243, 245 flow velocity waveforms in FGR 541 4D US SPIC imaging 210 postnatal dilatation 345, 345 see also ascending aorta; descending aorta

aortic arch circulation fetal 332, 337, 338 postnatal 332, 337, 338 development and blood flow/volume changes 581–2 ‘digital casts’ 213 echocardiography 332 4D US B-flow image 191 4D US SPIC imaging 203, 206, 210, 213, 214 hypothetical double aortic arch model 329–30, 330, 332 interrupted 300 longitudinal views 180, 180 normal 329, 331–4 obstruction 584 reconstruction 299 segments 329, 329 in tetralogy of Fallot 313, 313 aortic arch anomalies 297–300, 329–42 aberrant right subclavian artery 334 airway compression 340 associated abnormalities 336–7, 339 categories 329 cervical aortic arch 331 circumflex retroesophageal aortic arch 331, 334, 339 color/power Doppler 332 double aortic arch 331, 338 morphology 334 echocardiography 331–6 scanning maneuver 333 incidence 336–7, 339 left aortic arch defect 330 aberrant right subclavian artery 334 and morphogenesis 330–1 postnatal manifestations/ management 339–41 right aortic arch defect 232, 330–1 aberrant left subclavian artery 336, 337 mirror-image branching 335, 336 tubular hypoplasia 626 in Turner syndrome 626 vascular rings 331, 334–5, 339–40

760

Index

aortic arch anomalies (Continued) surgical division 341 vascular sling 331, 335–6, 340 without vascular ring/sling 331, 336 aortic atresia 25, 292, 294, 583 four-chamber view 294 long-axis view 295 in neonate 591 prognosis 301 restrictive atrial septum 294, 295 aortic coarctation see coarctation of aorta aortic–left ventricular tunnel 292, 296–7 and hydrops fetalis 497 prognosis 301 aortic regurgitation 297 maternal, during pregnancy 730 aortic–right ventricular tunnel 496–7 aortic stenosis 292, 295–6, 296 critical 296, 297, 396 and hydrops fetalis 498 perinatal/neonatal management 668 postnatal physiology/ hemodynamics 668 pressure gradient calculation 498 prognosis 301 extracardiac anomalies 636 frequency in CHD 115 incidence 107 in infants 692–3 maternal, during pregnancy 729–30 mild/moderate 296 perinatal/neonatal management 668 postnatal physiology/ hemodynamics 668 supravalvular 104, 710 surgery 515, 516 surgical repair 301 treatment 580–1 and ventricle development 579–80, 580 aortic valve anatomy 21, 306, 306 balloon valvuloplasty 680–1 flow velocity waveforms 536 in tetralogy of Fallot 313, 313 aortic valve anomalies 292, 294–7 aortopulmonary transposition 582, 592–3 oxygen saturation effects 587 see also transposition of great arteries (TGA) Apert syndrome 104, 114 arrhythmias differential diagnosis 461 electrophysiology 436, 438–9 in hydrops fetalis 494 mechanisms 445 arrhythmia-induced cardiomyopathy 363–4 arterial annuli 344 arterial duct see ductus arteriosus arterial flow 164

arterial flow velocity waveform 162 time to peak velocity 164 arterial level shunts 696 arterial–placental–venous interaction 415, 416 arterial pole 10 myocardial recruitment 10, 12 septation/valve formation 6–7 arterial trunks 3D/4D ultrasound 232 variants 25 arterial valves, perforate/imperforate 25 arteriovenous fistula, coronary 394–5 arteriovenous malformations 505 ascending aorta anatomy 21–2 development and blood flow/volume changes 581–2 oxygen saturation 587 in tetralogy of Fallot 313, 313 ASDs see atrial septal defects (ASDs) asphyxia 713, 720–1 asplenia 347, 349, 350, 351, 360 asymmetric development 348 asymmetric organs development 239 features 348 normal 348 atria arrangement 22–3, 23, 24 development 2, 3, 3, 101 echocardiography 175–6 3D/4D 229 premature atrial contraction (PAC) 437, 439, 442 bigeminal 452, 453, 454 primitive 10 systolic flow velocities 439 wall motion 438 see also left atrium; right atrium atrial appendages 354, 355 atrial bigeminy 443 atrial ectopic tachycardia (AET) 444–5, 463 persistence 478 atrial fibrillation 464 atrial flutter 437, 444, 444, 461 mechanisms 464 radiofrequency ablation 686, 686 treatment 475, 477 atrial isomerisms 455–6 atrial level shunts 696 atrial natriuretic peptide (ANP) 50, 487, 547, 563, 571 atrial pacing 466–7 atrial septal defects (ASDs) 281–5 anatomy 281–2 with aortic arch anomalies 337 chromosomal anomalies 615–16 frequency 710 complications 284

coronary sinus ASD 281, 281, 284 fetal diagnosis 282–4 frequency in CHD 115 incidence 107 maternal, during pregnancy 726, 728 natural history/outcome 284–5 ostium primum (partial AVSD) 281, 281, 282, 284 associated disorders 284 four-chamber view 282, 283, 284 ostium secundum 281, 281, 282 extracardiac anomalies 636 perinatal/neonatal management 671 postnatal physiology/ hemodynamics 671 repair 284–5 sinus venosus ASD 281, 281, 284 spontaneous closure 284 atrial septostomy 276 atrial situs determination 354–5 isomerisms 241, 243–5, 349 left 245–7, 245, 246 prognosis 248–50 right 246, 247–8, 247, 248 atrioventricular block see heart block atrioventricular canal 728, 729 development 2, 3–4 atrioventricular canal defect 189 atrioventricular conduction system 437–8 atrioventricular conduction time 436 reference values, age-matched 438, 440 atrioventricular (endocardial) cushions abnormalities 101, 106, 615 inheritance 712 development 9 atrioventricular delay 438 atrioventricular dissociation 456 diagnosis 359 isorhythmic 453 in left isomerism 359, 359 atrioventricular fibrous annulus 15 atrioventricular interval 439 atrioventricular junction anatomy 20 biventricular 23–4 concordant/discordant 23–4, 24 specimen examination 23–4 3D/4D ultrasound 229–30 univentricular 24 atrioventricular node 435, 436 formation 4 atrioventricular reentry tachycardia 437, 444, 461, 463 and hydrops fetalis 500 orthodromic 464, 466 spontaneous disappearance 464 atrioventricular septal defects (AVSDs) 229–30 complete 285–7 anatomy 285

Index

associated lesions 286 fetal diagnosis 285–6 four-chamber view 285–6, 285 natural history/outcome 286–7 repair 286 surgery survival 287 in double-outlet ventricles 321 and extracardiac anomalies 635–6, 636 frequency in CHD 115, 115 and hydrops fetalis 495 inheritance 712 and isomerisms 356, 356 in left atrial isomerism 246, 356 partial 281, 281, 282, 284 perinatal/neonatal management 672 postnatal physiology/ hemodynamics 672 in right atrial isomerism 248, 248 simple 343 in transposition of great arteries 343 in trisomy 13 625, 625 in trisomy 21 622, 622, 623 unbalanced 285, 286, 357 atrioventricular valves flow velocity waveforms 535 formation 3–4, 4 4D US SPIC imaging 208 ‘thick slice’ rendering 209 in hydrops fetalis 501 insufficiency 286 3D surface rendering 209 augmented atrial contraction 548 automatic arrhythmias of sinus node 437 autonomic innervation 160–1 maturation 161 autosomal dominant inheritance 706 prenatal diagnosis 709 autosomal recessive inheritance 706 prenatal diagnosis 709 AVSDs see atrioventricular septal defects (AVSDs) azygos vein malformations 243, 244, 245 balloon aortoplasty 694 balloon atrioseptostomy 677, 678 balloon dilatation aortic coarctation 678, 681 other vessels 682 pulmonary valve 679–80, 679 valves/blood vessels 677, 679 balloon valvuloplasty 276, 515–17, 516 aortic 680–1, 680, 692 in infants 692, 695 Baltimore–Washington Infant Study 610, 621, 740 baroreflexes 144–5 basic fibroblastic growth factor (bFGF) 30–1 Beckwith–Wiedemann syndrome 649, 651 beta blockers 732, 733

beta sympathomimetics 742 bigeminal premature atrial contractions 452, 453, 454 biliary atresia 359–60 biventricular hypertrophy 572 blastocyst implantation 30 blood flow coronary artery 387 coronary sinus 388 human fetus 138, 586 lamb fetus 137, 159, 585 liver area human fetus 139, 159 lamb fetus 137, 159 in neonate 591, 663 to organs, by gestation time 138 pulmonary 138–9 in SVT 501 tumor obstruction 405 velocity determination 161 venous, in SVT 468–9 volumes per heart chamber/blood vessel human fetus 139 lamb fetus 137 blood flow impedance 139 placental disorders 51 in pregnancy 45 reduction mechanism 50 blood gases 136 perinatal 145 postnatal 147–8 blood velocity abnormal placentation 47–8 arterial flow 139–41 placental 43, 43, 46 and pressure waves 549–50 venous flow 141–2 blood viscosity, fetal 43, 44 bone morphogenic protein (BMP) 9 Bourneville–Pringle syndrome 114 bowel obstruction 643 bradyarrhythmias 440, 444 bradycardia 141, 439, 440 catastrophic events 714 definition 713–14 and dilated cardiomyopathy 364 and isomerisms 356, 359 M-mode echocardiography 454, 455 peak flow velocities in coronary arteries 395 prenatal presentation 441 prenatal treatment 685 pulsed Doppler evaluation 455 sinoatrial 443 sustained 451, 452 in TGA 321 see also sinus bradycardia bradydysrhythmia 449–60 causes 453 clinical presentation 451–3

761

complete heart block 453–4 isolated 454–5 prevention 456 and structural heart disease 455–6 treatment, prenatal 456–8 diagnostic approach 449–50 pathophysiology 450–1 brain natriuretic peptide (BNP) 564 ‘brain sparing’ effect 393, 538–9, 540, 751–2 branch pulmonary arteries 165–6 bronchial morphology 23, 23 bronchopulmonary anatomy, normal 353 bronchopulmonary heterotaxy 347, 349, 352, 353, 354 bulbus cordosis 414 bypass, fetal 699 calcium channel blockers 733 calcium ions in contractility 160 in impulse formation/conduction 435 transport 143 campomelic dysplasia 648, 650 Cantrell’s pentalogy 644 capacitive micromachined ultrasound transducers (CMUTs) 96, 97 capillaries, fetal 48–9, 50 cardiac anatomy see anatomy, normal cardiac anomalies 121 and chromosomal anomalies 610, 613–16 structural cardiac defects 621–33 Doppler ultrasound 611 duct-dependent heart defects 663, 665 duct-dependent mixing 665 duct-dependent pulmonary circulation 664, 665 duct-dependent systemic circulation 663, 664, 665 therapeutic algorithms 676 and extracardiac anomalies 635–57, 637 prevalence 635 specific cardiac defects extracardiac malformations less likely 640 extracardiac malformations likely 635–40 specific extracardiac malformations 641–7, 650–1 abdomen/abdominal wall 637, 643–4 chest/mediastinum 637, 642–3 gastrointestinal 637 head/CNS 637, 641–2 multiple chromosomal malformations 645–7, 650–1 non-karyotypic syndromes 648–9 urogenital 637, 644 vascular 637 genetics 609–19

762

Index

cardiac anomalies (Continued) M-mode echocardiography 611 secondary to maternal disease 609 sonographic examination 610–12 first trimester screening 612 structural defects 616 and chromosomal anomalies 621–33 associations 621–7, 622 candidate genes 629 molecular basis of defect 629 ultrasound markers 627–9 in hydrops fetalis 484, 494–9 in syndromes 610 cardiac apex orientation 25–6 cardiac cell recruitment 13 cardiac conduction system 436 correlation with blood flow 436 development 160–1 cardiac cycle 156 diastole 156 systole 157 ventricles, individual 165 cardiac decompensation 555, 555 ductus venosus flow 554 venous flow 547 cardiac defects see cardiac anomalies cardiac dextropositions 241 cardiac disease in pregnancy (CARPREG) risk index 727 cardiac disease, maternal 725–35 arrhythmia 732 cardiovascular drug treatment 732–3 anticoagulation 732–3 tocolytic drugs 733 congenital heart disease acquired 731–2 coronary artery disease 732 right ventricular outflow tract obstruction 727–9 valvular 729–31 contraindications for pregnancy 728 delivery management 734 low-risk patients 727 management principles 727 risk stratification 725, 726, 727 cardiac failure 563 prognosis 564 treatment 573–4 see also congestive heart failure cardiac flow, absolute 532 cardiac function evaluation 553 factors 564 impaired 555 cardiac function indices 157–8 cardiac functionality through gestation 168 cardiac insufficiency treatment 675 cardiac levoposition 241 cardiac malpositions/isomerisms 239–50

classification 243 diagnosis 243–5 cardiac morphogenesis see morphogenesis cardiac output 150–1, 584 balance 700 calculation 490 combined ventricular output distribution late gestation 660 perinatal 145, 661 human fetus 138, 562 lamb fetus 137–8, 150 postnatal 150 determinants 142–4 and distribution 136–8, 562 estimation 164 and heart rate 142 lamb fetus 159 maternal, during pregnancy 725 and myocardial performance 143–4 percentage ejected by each ventricle human fetus 139 lamb fetus 137 perinatal 146, 147 perinatal, factors affecting 562–3 and preload/afterload 142–3 redistribution 573 right/left ratio (RCO/:CO) 532 normal ranges 537–8 through gestation 169 volumes per heart chamber/blood vessel 137 cardiac pacing, fetal 700 cardiac rhythms analysis 452, 453 irregular 439, 441 normal 450 cardiac shunt defects 666–7 ‘cardiac’ syndromes 610, 613 cardiac tumors 401–12 in children/adults 401 extracardiac anomalies 636 fetal 401–10 diagnosis 401–2, 404 echocardiography 401–2, 402, 404 3D ultrasound 402, 404 and echogenic foci 410, 410 family counseling 404 fibromas 401, 408–9 hemangiomas 401, 409 and hydrops fetalis 484, 499–500, 503–5, 504 magnetic resonance imaging 404, 405 outcome 405 prenatal management 404–5 prevalence 401 regression 405, 407 rhabdomyomas 401, 402, 404, 405–8 teratomas 401, 409

perinatal/neonatal management 673 postnatal physiology/ hemodynamics 673 cardiac volume evaluation 532, 533 cardinal veins 413 anomalies 417, 417 cardiogenesis 101 primary 9–10 secondary 10, 11 cardiomegaly 273, 377 and congestive heart failure 570–1 in Ebstein’s anomaly 491 in hydrops fetalis 490 and tachyarrhythmias 500 cardiomyopathies 363–7, 375–84 induced-induced 363–4 classification 375 in hydrops fetalis 502–5 immune-mediated 378 maternal, during pregnancy 731 tachycardia-induced 501 treatment 505 see also dilated cardiomyopathy; hypertrophic cardiomyopathy; non-compacted cardiomyopathy; peripartum cardiomyopathy; restrictive cardiomyopathy; toxic cardiomyopathy cardiosplenic syndromes 241, 243 cardiothoracic index/ratios 364, 503, 504 congestive heart failure 565 supraventricular tachycardia 468 cardiovascular anomalies congenital 579–95 in hydrops fetalis 494 and postnatal changes/effects 587–93 ductus arteriosus aortopulmonary transposition 592–3 preterm infants 592–3 ductus arteriosus closure 590–3 coarctation of aorta 591–2 pulmonary flow decrease 590 systemic flow decrease 590–1 pulmonary circulation 593–4 pulmonary vascular resistance decrease 587–90 abnormal communications 587–9 communications in preterm infants 589–90 cardiovascular control mechanisms 155–7 cardiac cycle 156 diastole 156 systole 157 cardiac function indices 157–8 cardiovascular development 1–8, 101–2, 186

Index

secondary cardiac development 698 segments 1–2, 2, 3, 4, 5 transitional zones 2–5, 2 Cardiovascular Profile Score 561, 573, 577 categories 578 development 569–70 requirements 578 summary 578 use in congestive heart failure 578 cardiovascular system blood flow/volume changes ascending aorta/aortic arch development 581–2 ductus arteriosus size/ orientation 582–3 ventricular development 579–81 blood oxygen content changes ascending aortic oxygen saturation 587 pulmonary arterial oxygen saturation 585–6 function in twin–twin transfusion syndrome 606 obstructive lesions 583–5 aortic arch obstruction 584 ductus arteriosus obstruction 583–4 foramen ovale obstruction 584–5 quantitative analysis 161 cat-eye syndrome 103, 117 CATCH-22 syndrome 103, 106, 324, 646 three-vessel echocardiography view 181 catecholamine 150 central nervous system anomalies in CHD 749–57 genetics 750 perioperative causes 749–50 preoperative causes 750 effects 749 structural defects 750–1 cerebral circulation aortic arch obstruction 584 blood flow in CHD 427 fetal 752–5 neonatal 752–3 Doppler indexes 165 middle cerebral artery 428–30, 429 cerebral palsy 721, 722 cerebral–placental resistance (CPR) index 753, 753 cervical aortic arch 331 Cesarean section 734 CFC syndrome 708 Chagas disease 739 CHAOS (congenital high airway obstruction syndrome) 520 chaotic tachycardia 463 CHARGE association 103, 114 and complete AVSD 286 extracardiac anomalies 636, 649, 651 inheritance 708

CHD see congenital heart disease (CHD) chemoreflexes 144 Children’s Hospital of Philadelphia (CHOP) Cardiovascular Score 603, 604, 605 growth abnormality study 751 Children’s Hospital of Philadelphia (CHOP) Cardiovascular Score 606 chloride ions 435 chromosomal anomalies 711 and cardiac anomalies 610, 613–16, 621–33 associations 621–7, 622 congenital heart disease 102, 103, 117 and conotruncal anomalies 323, 324 in hydrops fetalis 506 intrauterine mortality 621, 623 and isomerisms 249–50 in left heart malformations 300 survival 104 chromosome deletions 103 chromosome duplications 103 chromosome microdeletions 103 circulations fetal 133–53, 158–9, 561–2 aortic shelf 592 blood gases 136 cardiac output determinants 142–4 and distribution 136–8, 158, 159 and heart rate 142 and myocardial performance 143–4 postnatal 150–1 and preload/afterload 142–3 course 133–5 heart/great vessels 135 porta hepatis region 134 diagrammatic representation 159 flow velocity contours 139–42 arterial flow 139–41 venous flow 141–2 human fetal circulation 138–9, 586 lamb fetus 585 oxygen saturation 136 maternal oxygen administration 136–9 oxygenated/systemic blood admixture 135 persistent 665–6 regulation 144–5 baroreflexes 144–5 chemoreflexes 144 transitional 563 vascular pressures 135–6 perinatal 146, 147 with ventricular septal defect high pulmonary vascular resistance 588

763

low pulmonary vascular resistance 589 vs postnatal circulation 561, 659 intervillous 42–4 perinatal birth-associated changes 145–51 persistent pulmonary hypertension of newborn 148 pulmonary circulation 147–8 postnatal adaptation of pulmonary vascular resistance 662 cardiac output 150–1 ductus arteriosus closure 148–9 in preterm infant 149–50 ductus venosus flow 151 establishment 660–2 hepatic flow 151 myocardial changes, morphological 151 vs fetal circulation 561, 659 transitional, fetal to postnatal 646, 661 duct-dependent heart defects 663, 665 fetus with CHD 663–7 without suspected problems 667 healthy fetus 659–63 pulmonary vascular resistance in 563, 659, 661, 662–3 umbilicoplacental 48–53 uterine, maternal 42 uteroplacental 41–8 villous 48–9 circumflex retroesophageal aortic arch 331, 334, 339 coarctation of aorta balloon dilatation 681 chromosomal anomalies 300, 616, 622, 623 development 300 diagnosis 297–300, 298, 299, 300 and ductus arteriosus closure 581–2, 591–2 extracardiac anomalies 636, 639 frequency in CHD 115 identification 297 incidence 107 infantile 694 maternal, during pregnancy 726, 728 mimicking cardiomyopathy 379 with mitral valve disease 297 outcome 302 perinatal/neonatal management 669 postnatal physiology/ hemodynamics 669 prognosis 301–2 screening outcomes 120 in TGA 345 in Turner syndrome 626 cocaine exposure 742 Collaborative Perinatal Project 609

764

Index

color Doppler imaging B-mode/B-flow 86–7 beam alignment 532 color flow imaging 73–5 color velocity imaging 73–4 power Doppler imaging 74 4D US 202, 212, 220–1, 220–1 spatiotemporal image correlation 202, 221–2 tissue Doppler 93, 95 color flow imaging 73–5 color velocity imaging 73–4 committees for human research 524 common arterial trunk 12, 25, 323 chromosomal anomalies 622, 625 extracardiac anomalies 636, 638–9 conduction abnormal 436, 437 atrioventricular conduction system 437–8 normal 435–6 propagation 15 sinoatrial conduction system 14, 15 conduction system see cardiac conduction system congenital heart disease (CHD) 691 abnormal four-chamber view 121 detection rates 122 associated extracardiac anomalies 635–57 associated non-karyotypic syndromes 648–9 associated syndromes 114 causes 705 and central nervous system 749–57 cerebral–placental resistance (CPR) index 753, 753 and congestive heart failure 568–9 definition 101 detection, prenatal 83, 106–7, 111 diagnosis, 3D/4D US 229–33 environmental factors 710 epidemiology 101–10 chromosomal anomalies 102, 103 environmental factors 105–6, 105 genetic abnormalities 102, 105 incidence 106–8, 107 familial, as screening indication 112, 113 familial risk 609–10 genetic basis 705–6 infants 691–703 early intervention 691 intervention desirable 691, 692, 697 intervention essential 691, 692–7, 692 multiple cases 711 natural course 191–3 neonatal intervention 668–73 outcome 119 perinatal management 668–73

postnatal physiology/ hemodynamics 668–73 risks 737 overall 710 to wider family 711 in utero development 191–3 congenital high airway obstruction syndrome (CHAOS) 520 congenital malformations EPDCs in 15 inlet tract 14–15 outlet tract 10, 12 congestive heart failure 561–79 cardiomegaly 570–1 cardiovascular profile score 573 causes 564 and congenital heart disease 568–9 definition 561 Doppler ultrasound arterial 573 ductus venosus 565, 569–70 umbilical 569–70 venous 570 ductal constriction/occlusion 568 ductus arteriosus 567 fetal cardiac interventions 574–5 fetal (intrauterine) growth restriction (FGR) (IUGR) 567 in hydrops fetalis 490, 503 hydrops fetalis 563–4, 569 myocardial function, abnormal 571–3 pulmonary valve 567, 568–9 right ventricle 567–8 treatment 573–4 tricuspid valve 566–7, 569 twin–twin transfusion syndrome 568 conotruncal anomalies see ventricular outflow tracts, congenital malformations continuous wave (CW) Doppler 72, 532 contractility 142, 157, 158 analysis with echocardiography 174 echocardiography 93 and maturation 159, 160 Cornelia de Lange syndrome 103, 114 and complete AVSD 286 extracardiac anomalies 647, 648 coronary arteries course 388 functional anatomy 385 intramural 344, 345 ultrasound examination 387–8 pulsed wave images 389 coronary arteriovenous fistula 394–5 coronary circulation blood flow in animal model 386–7 embryology 385–6 flow velocity waveforms 535 functional anatomy 385 ultrasound examination 385–99 technique 387–90

coronary artery examination 387–8 coronary sinus examination 388–90 setup 387 coronary sinus atrial septal defect 281, 281, 284 coronary sinuses 207 blood flow 392 course 390 measurement 391, 392 ultrasound examination 388–90 visualization 390 coronary sinusoids see sinusoids corpus callosum agenesis 641 corticosteroids 742 cortisol 150 Costello syndrome 708 coumadin 732, 734 counseling, prenatal fetal cardiac tumors 404 genetic counseling 705–12 conditions with known gene defects 707, 709 conditions with unidentified localized genes 707 definition 706 disclosure 707 families with congenital heart defects 709–11 and molecular genetics 707, 709 principles 706–7 timing 709 left heart malformations 300, 302 tachyarrhythmias 469 coxsackieviruses 739 cri-du-chat syndrome 103, 117, 710 crista superventricularis 306 cyanotic heart disease, maternal 726, 727 cystic hygroma 627 cytokines 29, 32, 33, 34–5 cytomegalovirus 739 cytotoxic T lymphocytes (CTLs) 34 Dandy–Walker malformation 641–2, 641 del 22q11 see microdeletion 22q11 ‘deletion’ syndromes 613, 614 delivery management 734 descending aorta anatomy 22 Doppler indexes 166 echocardiography 175 4D US SPIC imaging 207 development see cardiovascular development dexamethasone 456, 501, 502 congestive heart failure 574 dextrocardia 25, 239 mirror-image 239 visceral/atrial situs 355 diabetes, maternal 44

Index

and atrial isomerism 350 and hypertrophic cardiomyopathy 365–6, 381, 383 risk of cardiac anomalies 609, 740 as screening indication 113, 114 diaphragmatic hernia 642, 643 diastasis 156 diastole 436 ventricular 155 DiGeorge syndrome 103, 106, 114, 613, 711 and conotruncal anomalies 324 inheritance 708 and ostium primum ASD 284 ‘digital casts’ 224 digoxin 456, 472, 474, 475, 477 congestive heart failure 574 direct fetal application 476, 478 during pregnancy 732, 733 transplacental 470, 501, 505 dilated cardiomyopathy 104, 363–4, 364 ascites 378 atrioventricular regurgitation 377 with cardiomegaly 377 causes 378–9, 383 conditions mimicking 379 echocardiographic features 383 echographic features 363–7 extracardiac anomalies 636 and hydrops fetalis 503 intrauterine therapy 381 investigation 379–80 left ventricle involvement 375, 376, 377, 378 maternal, during pregnancy 731 outcome 378, 383 perinatal/neonatal management 673 postnatal physiology/ hemodynamics 673 reported series 375, 378 right ventricle involvement 375, 376, 377, 378 secondary 503–5 discordant atrioventricular connection 319 discordant ventriculoarterial connection 317, 319 disopyramide 733 diuretics 733 Doppler flows in early gestation 155–72 Doppler indexes 77–8 extracardiac arterial 165–7 cerebral circulation 165 descending aorta/visceral blood flow 166 pulmonary circulation 165–6 uteroplacental blood flow 167 extracardiac venous 167–8 Doppler motion analysis 71 Doppler spectrum 71

Doppler ultrasound 71–80 abnormal placentation 47–8 advanced dynamic flow (ADF) 87–8 baseline 80 beam alignment 531–2 cardiac function evaluation 531–46 cardiovascular system 161–2 color in 4D US 202, 212, 220–1 color ‘quality’ 80 directional power 87–8 ductus venosus 430–3 echocardiographic indexes, normal 537–8 echocardiography technique 531–7 cardiac circulation 532–6 atrioventricular valves 535 coronary blood flow 535 outflow tracts 535 pulmonary vessels 536 venous circulation 533–5 parameters 532 principles 531–2 reproducibility 536–7 extracardiac, in CHD 427–34 fetal growth restriction 538–40, 542–4 fetal heart rate 450 in hydrops fetalis 490, 491 venous velocimetry 493, 495 mechanism of action 71–2, 72 middle cerebral artery 428–30 modes B-mode/B-flow 87, 91, 202, 210–11, 212, 214 color in 86–7 color flow imaging 73–5, 93, 95 color velocity imaging 73–4, 74 power Doppler imaging 74 continuous wave (CW) 72 pulsed wave (PW) 72–3, 93, 95 pulse repetition frequency 73, 79, 80 strain rate imaging 95–6 pulse repetition frequency 73, 79, 80 recent developments 78–9 natural acoustic tagging 79, 79 system controls 78–9, 80 tissue Doppler 78–9, 93 two-dimensional strain (speckle tracking imaging) 78–9, 93–6, 96 ROI position/size 80 sample volume position/size 80 signal processing 74–8 demodulation 74 frequency content determination 75, 76 frequency shifts 74–5 pulsatility 77 schematic description 75 spectral analysis 72, 75 spectral ‘envelope’ 77, 77

765

spectral folding 73, 73 spectral results display 75, 76, 77, 77 umbilical–fetoplacental circulation 78 wave form analysis 77, 78 and Doppler indexes 77–8 spatiotemporal image correlation (STIC) 87, 88, 90–1, 91 total gain 80 in TTTS assessment 602 umbilical arteries 427–8, 428 umbilical cord blood flow 51 umbilicoplacental circulation 50–1 uteroplacental circulation 41, 44 wall motion filter 80 double aortic arch 331, 338 extracardiac anomalies 639 morphology 334 double-inlet left ventricle incidence 107 perinatal/neonatal management 669 postnatal physiology/ hemodynamics 669 double-outlet right ventricle (DORV) 321–2, 614 chromosomal anomalies 324, 622, 623 extracardiac anomalies 324, 636, 637, 638 frequency in CHD 115 incidence 107 outcome 325, 325 pathology diagrams 321 peak velocity index 432 perinatal/neonatal management 669–70 postnatal physiology/ hemodynamics 669–70 in right isomerism 358 double-vessel sign 242, 243, 245 Down syndrome 242 see also trisomies, trisomy 21 duct-dependent heart defects 663, 665 clinical presentation 664 ductal arches ‘digital casts’ 213 echocardiography 333 4D US SPIC imaging 203, 206, 210, 213, 214 longitudinal views 180, 180 ductal constriction/occlusion 568 ductus arteriosus 135 agenesis 496–7 anatomy 22 aortopulmonary transposition 592–3 blood flow flow velocity waveforms 164–5 regulation 144 velocity 140, 140 and cardiovascular anomalies 583 closure 590–3 coarctation of aorta 591–2

766

Index

ductus arteriosus (Continued) and hydrops fetalis 497 postnatal 148–9, 660–1 in preterm infant 149–50 pulmonary flow decrease 590 systemic flow decrease 590–1 constriction 497, 586 echocardiography 187 embryology 330, 414 normal 567 obstruction 583–4 opening with prostaglandin E 666, 674 in PAIVS 275 patent see patent ductus arteriosus in pericardial effusion 369 perinatal 146, 147 in preterm infants 592–3 restrictive 263–4, 264 in right heart assessment 251 size/orientation and blood flow/volume changes 582–3 stenting 682, 683 in tetralogy of Fallot 313–14, 315 in TGA 317, 319 ultrasound investigation 393 see also patent ductus arteriosus ductus venosus abnormality as CHD marker 628 agenesis 420 anatomy 414, 415 blood flow course 133–4, 134 early gestation 155 during gestation 539 growth-restricted fetus 543 postnatal 151 closure, postnatal 661 development 158, 414 Doppler screening 155 Doppler ultrasound 430–3, 431, 433, 565, 569–70 and CHD pathophysiology 431–3 flow velocity waveforms 533–4, 534 peak velocity index in right CHD 432 ductus venosus flow 553 pulsatility 548 velocity 553–5, 554 in FGR 554 ductus venosus shunting 547, 553 duodenal atresia 643, 643 dysrhythmias with cardiac tumors 405, 407, 408 early embryo loss placenta 46 placental disorders 46 vascular development 29 Ebstein’s anomaly 189, 253–5, 255, 741 cardiomegaly 491 in corrected TGA 320 Doppler ultrasound 433

erroneous diagnosis 253, 253 extracardiac anomalies 636, 640 flow velocity waveforms 433 frequency in CHD 115 in hydrops fetalis 491, 494–5 in neonates 665 outcome 495 pathophysiology/ultrasound findings 431, 432–3 peak velocity index 432 perinatal/neonatal management 671 postnatal physiology/ hemodynamics 671 tricuspid regurgitation 571 Ebstein’s malformation 272 echocardiography 96, 187 acquisition/processing/display 83–8 aortic arch 332 aortic arch anomalies 331–6 automated 3D/4D studies 96 in bradydysrhythmia 449–50 cardiac anomalies 610–12 correlation with outcome 615 first trimester screening 612 specific anomalies/ syndromes 612–16 cardiac mechanics 93, 95–6 challenges 197 contractility 93 diagnostic 449 early 187, 189, 189 five-chamber view 189 3D/4D ultrasound 227–33 dilated cardiomyopathy 364 early 185–96 accuracy factors 186 diagnostic 187, 189, 189 heart measurement 186 precautions/recommendations 191 studies 190 electronic focusing 83–4 extended examination 186–7, 197 fetal venous system 186–7 suggested improvements 186–7 fetal cardiac tumors 401–2, 402, 403, 404, 404 with coarctation of aorta 406 Doppler ultrasound 403 M-mode 403 postnatal 403, 404, 406 rhabdomyomas 404, 407 3D ultrasound 404 ventricular compression 406 five-chamber view 174, 179, 187, 188 approach 175 diagnostic 189 four-chamber view 121–4, 121, 173–8, 174 approaches 175, 176 checklist 175 ‘drop-out’ effect 177, 178

sensitivity 123, 186 in systole/diastole 178 tomographic ultrasound imaging 203 future developments 96–7, 184 ‘glass body’ 91 great arteries 179–80 heart rate 191 indications 112–17, 113 risk groups 112 lateral resolution 83, 83 longitudinal views of outflow tracts 178–80 matrix array transducers 64–5, 64, 65, 66, 92–3 full matrix 92 future developments 96–7 sparse matrix 92–3 2D 83–4, 84 modes B-mode/B-flow 87, 91, 191, 230, 611 color in 86–7 color inversion 92 M-mode 403, 449, 450 real-time compound imaging 84–5, 85 tissue Doppler 78–9, 93 tissue harmonic imaging 68–9, 70, 84, 87 motion detection 93 navigation/volume data display 88–92 operator skill 197 precautions 193 pulmonary view 174, 179 approach 175 quantification, semiautomatic 90–1 rendering 89–90, 91 risk groups 111–29 high risk population 111, 118–21 low risk population 111, 112–17 screening 107, 111–29 basic examination 197, 198 ductus venosus 155 first/early second trimester 185–96 benefits 185 four-chamber view 121–4, 121 high risk population 112 low risk population 118–21 scheduling 185 timing 112 ventricular outflow tracts 123–4 WHO criteria 118–20 yield by indication 124 short axis view 179 speckles 86 reduction 62, 85–6, 86 speckle tracking 78–9, 93–6 strain/strain rate 93–6 2D 78–9 technical advances 83–100 three-vessel and trachea view 187

Index

three vessel and trachea view 203 three-vessel view 180–2, 207 atrial isomerisms 245 upper thorax 181 3D/4D ultrasound 182, 190–1, 219–37, 611 accuracy factors 234–5 acoustic shadows 233 acquisition modalities 219–22 B-mode/B-flow 210–11, 219–20, 230 color/power Doppler 220–1 VOCAL analysis 224 automated 96 CHD diagnosis 227–33 arterial trunks 232 atrioventricular junction 229–30 functional evaluation 232–3 segmental approach 228–9 veins/atria 229 ventricles 230, 232–3 ventriculoarterial junctions 230–2 virtual planes 228 flow direction 233 interslice distance 183 insonation angle 233 planes of interest 182, 183, 184 post-processing modalities 219–22 inversion mode 210, 224, 611 MPR mode 222, 223 rendering 223–5, 233 techniques 208–10 tomographic ultrasound imaging 201, 201, 202, 203, 224 problems/artifacts 233, 234 real-time/live 92–3, 92, 93, 94 reconstructed 88–90, 88 screening examination 224–7 applications 225–6 functional heart evaluation 226–7 guidelines 224–5 spin technique 203, 207 STIC modality see under spatiotemporal image correlation (STIC) volume dataset 182–4 timing 193 tomographic imaging 89, 89, 90 tuberous sclerosis 407 2D ultrasound 78–9, 83–4, 84 normal heart 173–84 recommendations/guidelines 182 upper abdominal plane 173, 175 checklist 174 ventricular outflow tracts 178–80, 181, 203, 204 checklist 178 imaging planes 307 techniques 307–9 volume echocardiography 88 see also Doppler ultrasound; ultrasound

echogenic foci 410, 410 endocardial 368–9, 368 ectopic automatic rhythms 437 ectopic pacemakers 437 Edward syndrome see trisomies, trisomy 18 Ehlers–Danlos syndrome 104, 114 Eisenmenger syndrome 726 elective early (preterm premature) delivery 471, 520 electrocardiography (ECG) fetal 449 maternal surveillance 471 electronic fetal monitoring (EFM) 713, 719, 721–2 ancillary tests 719–20 electrophysiology, fetal 435–47 abnormal impulse conduction 436, 437, 441 impulse generation 436–7 catheter positions 685 investigation, intrauterine 437–40, 444–6 arrhythmias 438–9 atrioventricular conduction system 437–8 rates, abnormal 439–40, 444–6 rhythms, irregular 439, 441 normal impulse formation/ conduction 435–6 Ellis–van Creveld syndrome 104, 114 and complete AVSD 286 extracardiac anomalies 648, 650 and ostium primum ASD 284 embryology aortic arch 329–30 asymmetric development 239 atrioventricular canal 2, 3–4 atrium 2, 3 ductus arteriosus 330 looped heart 1, 2 primary fold 2, 4–5 septation/valve formation 6–7 sinoatrial transitional zone 2–3, 2 tissues of embryonic heart 5–6 venous sinus 1–2, 2 venous system 413–14 ventricles inlet segment 2, 4 outlet segment 2, 5 end-diastolic flow 50, 51 absence/reverse 52, 53 early detection 165 through gestation 167 end diastolic volume 156 end systolic volume 157 endocardial casts, virtual 94 endocardial cushions see atrioventricular (endocardial) cushions endocardial diseases 368 echogenic foci 368–9

767

endocardial fibroelastosis 297, 367–8 therapeutic considerations 369 endocardial echogenic foci 368–9, 368 endocarditis prophylaxis 730 endocardium development 5–6, 9 endothelin1 gene 12 environmental factors in CHD 105–6 EPDCs see epicardium-derived cells (EPDCs) ephrin receptors 31 ephrins 31–2, 31 epicardium 13 development 6 epicardium-derived cells (EPDCs) 13–14, 14 in cardiac development 15 migration 14 epigenetic factors 12 epithelial–mesenchymal transformation (EMT) 13 esophageal atresia 642 estradiol 44–5 Eurofetus Consortium Trial 604 Eustachian valve 3, 19 ex utero intrapartum treatment (EXIT) 520 examination of specimens 22–6 extracardiac Doppler ultrasound see under Doppler ultrasound extracardiac functionality 168 extracellular matrix degradation 29 extravillous trophoblasts (EVT) cytokine regulation of 34 early embryo loss 46 infiltration of decidua 42 maternal immune tolerance 32, 33, 34 see also trophoblastic shell FasL 35 FASTER (First and Second Trimester Evaluation of Risk) database 612 fetal alcohol syndrome (FAS) 609 fetal bypass 699 fetal distress 720–1 fetal hypoxia 51 fetal intra-abdominal umbilical vein (FIUV) varix 417 fetal (intrauterine) growth restriction (FGR) (IUGR) abnormal placentation 47–8 ‘brain sparing’ 751 coronary artery blood flow 395 gestational age 548 ‘heart sparing’ 390–2 hemodynamic modifications in fetus 538–40, 542–4 flow velocity waveforms aorta 541 ductus venosus 539, 543 inferior vena cava 542

768

Index

fetal (intrauterine) growth restriction (FGR) (IUGR) (Continued) middle cerebral artery 542 pulmonary arteries 541 umbilical vessels 540, 543 liver perfusion 547, 548 pathophysiology 543 physiology 547–8 right ventricular function 567 as screening indication 116 venous flow 547–59 fetal laryngeal atresia 519, 523 fetal pulse oximetry 720 fetal scalp sampling 720 fetal ST wave analysis 720 fetoscopic laser ablation 519, 522, 604 fetoscopy-assisted interventions 517, 518, 519–20, 523 limitations 519 FGR see fetal growth restriction (FGR) fibroblastic growth factor (FGF) in cardiogenesis 9 FGF10 in placental development 27–8 fibroblastic growth factor (FGF) receptors 27 fibromas 401, 408–9 fibrous annulus see atrioventricular fibrous annulus First and Second Trimester Evaluation of Risk (FASTER) database 612 five-chamber echocardiography 174, 179, 187, 188, 611 approach 175 diagnostic 189 flecainide 472, 474, 475, 477, 478 flow velocity contours arterial flow 139–41 venous flow 141–2 flow velocity waveforms (FVWs) 45, 161 abnormal placentation 47 aortic valve 536 arterial 162 time to peak velocity 164 atrioventricular valves 535 cardiac errors 531 parameters 531–2 coronary arteries 387, 389 coronary circulation 535 ductus arteriosus 164–5 ductus venosus 539 extracardiac venous 167–8 in hydrops fetalis 493 intracardiac 162–4 mitral valve 536 outflow tracts 164, 535 pulmonary circulation 537 pulmonary valve 536 pulmonary vessels 536 sequential changes 168–9

in tachyarrhythmias 463 transatrioventricular 163 umbilical 167 venous 161, 162, 533–5 foramen ovale in aortic atresia 294 in atrial flutter 466 in circulatory regulation 144, 159–60 closure 499 postnatal 661 in Ebstein’s anomaly 254, 255 echocardiography 175, 211 and hydrops fetalis 499 obstruction 584–5 perinatal 146, 147 restrictive 284–5, 499 in TGA 319 four-chamber echocardiography 121–4, 121, 611 sensitivity 186 screening 123 4D ultrasound (4DUS) see echocardiography, 3D/4D ultrasound fragile X syndrome 737–8 Frank–Starling mechanism 144, 157, 415, 562 onset 161 Fraser syndrome 520 Frynn syndrome 103 functional capacity 725 furosemide 456 gastroschisis 643 genetic abnormalities in congenital heart disease 102, 104, 105 genetic causes of cardiomyopathy dilated 379 hypertrophic 382 ‘genetic sonography’ 113, 117 genetics cardiac anomalies 609–19, 705–6 cardiogenesis 9, 101 central nervous system anomalies 750 gestational age and FGR 548 in hydrops fetalis 563 umbilical venous flow 550 and ventricular changes 566 gestational sac 41, 49 glucocorticoids 456 Goldenhar syndrome (hemifacial microsomia) 114 and complete AVSD 286 extracardiac anomalies 649, 650–1 ‘golf balls’ 368, 368, 369 great vessel visualization spatiotemporal image correlation 202–3, 209–10, 213 tomographic ultrasound imaging 227 group beating 452–3, 455

harmonic imaging see tissue harmonic imaging HbA1c levels 609 heart block in bradycardia 453 complete 443 and hydrops fetalis 484, 501–2 presentation/outcome 457 regular/slow heartbeat 453, 456 treatment 501–2 first degree 438 high degree 440 and isomerisms 246–7, 356, 359 prognoses 456 second degree 439, 440, 453 heart failure see cardiac failure; congestive heart failure heart fields 101 heart–hand syndrome see Holt–Oram syndrome heart position specimen examination 25–6 see also cardiac malpositions/ isomerisms; dextrocardia; levocardia heart rate bradycardia 451 bradydysrhythmia 450–1 in complete heart block 453, 456, 457 Doppler ultrasound 450 fetal heart rate calculation 198, 198 fetal heart rate monitoring efficacy 721–2 electronic fetal monitoring vs intermittent auscultation 713, 719, 721–2 interpretation/management fetal pulse oximetry 720 fetal ST wave analysis 720 future directions 720–1 unfulfilled expectations 722 fetal heart rate patterns 713–16 accelerations 714 baseline 713–14 decelerations 714–15 early 715 late 714–15 management 719 variable 715 interpretation/management 716–21 ancillary tests 719–20 sinusoidal pattern 715–16 variability 714 fetal heart rate tracings abnormal no accelerations/decelerations 717 severe variable decelerations 718 tachycardia 718 normal 716 during gestation 160–1, 436 sequential changes 161

Index

sinus tachycardia 465 STIC calculation 198, 198 supraventricular reentry tachycardia 468 supraventricular tachycardia 465, 467, 468, 469 in tachyarrhythmia 500, 501 tachyarrhythmias 461 ventricular tachycardia 465 heart size analysis 174 ‘heart sparing’ 390–2, 393, 394 heart transplantation infants 693 neonatal 294 heartbeat group beating 452–3, 455 onset 48, 155 hemangiomas 401, 409 and hydrops fetalis 499 hemifacial microsomia see Goldenhar syndrome (hemifacial microsomia) hemodynamics fetal modifications fetal growth restriction 538–40, 539, 540, 541–2, 542–4, 543 venous flow 547–59 ureteroplacental insufficiency 542 maternal changes 44–5, 725 umbilicoplacental circulation 49 Hemopump 699 Hensen’s node 101 heparin 732, 733, 733 hepatic flow, postnatal 151 hepatic sinusoids 413, 414 hepatic veins 413, 414, 415 flow velocity waveforms 534 velocimetry 555 heterotaxy of viscera 241, 347 with asplenia/polysplenia 350 with bronchopulmonary isomerism 349 and CHD 356–9 and chromosomal defects 350 color Doppler imaging 351–2 extracardiac anomalies 359–60, 651 familial 350 genes for 250 incidence 348, 350 left isomerism 351, 351 in right atrial isomerism 247 right isomerism 351 heterozygosity testing 709 His bundle tachycardia 463 His bundles 4, 435, 436 HNK-1 3, 4 Hofbauer cells 27–8 holoprosencephaly 642, 642 Holt–Oram syndrome 102, 104, 114 cardiac anomalies 614 extracardiac anomalies 647, 648

inheritance 708 horseshoe kidney 644 human chorionic gonadotropin (hCG) 45 human leukocyte antigen (HLA) maternal immune tolerance 33, 34, 35 preeclampsia 32 hydatid mole 46 hydralazine 733 hydrops fetalis 115, 489–93 associated disorders 483–514, 484–5 bradyarrhythmias 484 cardiac defects, structural 484, 494–9 cardiomegaly 490 cardiomyopathies 484 cardiovascular anomalies 494 twin pregnancies 485 chromosomal anomalies 506 congenital myotonic dystrophy 485 erythrocyte loss 484–5 erythrocyte underproduction 485 extracardiac associations 484–5 heart block, complete 501–2 idiopathic arterial calcification 484, 500 long QT syndrome 484 metabolic 485 myocardial diseases 484, 502–5 tachyarrhythmias 484, 499, 500–1 tumors 484 cardiac 484, 499–500, 503–5, 504 vascular 505 valvular incompetence 492 vascular disorders 484 cardiovascular profile score development 569 classification 563 definition 483–4 diagnosis, prenatal 483 considerations 489 diagnostic approach 487, 488, 489–93 karyotyping 487, 489 sonography 489 in dilated cardiomyopathy 375, 378 in early gestation 506 etiology 563–4 factors contributing 563–4 immune 483 incidence 483 as marker of chromosomal anomaly 627 of fetal mortality 564 myocardial response 392 pathophysiology 485–7 compensatory mechanisms 486–7 fluid imbalance 485–6 placental modification 487 prognostic indicators 489 in tachycardia 466, 469, 471 treatment 495, 505 postpartum 501

769

hydrothorax 524, 525, 642 bilateral 642 hyperglycemia 609 hyperinsulinism, fetal 365–6 hypertension 572 hyperthyroidism 740 hyper trabeculation see non-compacted cardiomyopathy hypertrophic cardiomyopathy 104, 364–6, 365, 740 causes 381–2, 382 diabetic 365–6, 381 echographic features 381 extracardiac anomalies 636 investigation 383 maternal, during pregnancy 731–2 in Noonan syndrome 381 outcome 382 with pericardial effusion 382 perinatal/neonatal management 668 postnatal physiology/ hemodynamics 668 recipient twin 365, 383 hypoplastic left heart and aorta/head diameters 751, 752 and aortic stenosis 296 and cerebral circulation in CHD 430, 430 cerebral vascular resistance 755 chromosomal anomalies 616, 622 frequency 710 extracardiac anomalies 636, 639 foramen ovale obstruction 584–5 frequency in CHD 115 and hydrops fetalis 497–8 incidence 107 in infants 693 inheritance 712 in neonates 663, 664 perinatal/neonatal management 668 physiology, preoperative 675 postnatal physiology/ hemodynamics 668 prevention 574–5 pulsatility index in 753–4, 754 rare variants 498–9 screening outcomes 120 hypoplastic right heart 269, 566 incidence 107 see also pulmonary atresia, with intact ventricular septum hypothetical double aortic arch model 329–30, 330, 332, 337 hypoxemia 386–7 hypoxia, fetal and cerebral circulation in CHD 429–30 and dilated cardiomyopathy 363 and fetal heart rate 719 prediction 51 and smooth muscle production 148 hypoxia-inducible factor (HIF) 29

770

Index

idiopathic arterial calcification 395, 484, 500 iliac arteries 166 immotile cilia syndrome see primary ciliary dyskinesia immune hydrops fetalis 483 immune tolerance, maternal 32, 34–5 implantation 27–40 blastocysts 30 requirements 41 trophoblasts 28 indomethacin 742 infants aortic commissurotomy 693 balloon aortoplasty 694 balloon valvuloplasty 692 with CHD 691–703 early intervention 691 intervention desirable 691, 692, 697 intervention essential 691, 692–7, 692 Norwood procedure 693 Ross–Konno aortoventriculoplasty 692, 693 infections 363, 364, 379 intrauterine 114 inferior vena cava flow velocity waveforms 533, 534 in congestive heart failure 570 in FGR 542 preload index (PLI) 537 venous flow distribution 548, 548 see also interrupted inferior vena cava infundibular (outlet) septum 25, 306 infundibulum 318 inheritance 705–6 multifactorial 706 inherited disease genetic counseling 705–12 heterozygosity testing 709 liability 705 modes of inheritance 706 predictive testing 709 prenatal diagnosis 709 inlet tract congenital malformations 14–15 development 2, 13–14 integrin-linked kinase 364 intensive therapy examples 676 poor prognostic indicators 676 principles 667, 674 interferon γ-inducible protein-10 (IP-10) 32 interleukins in angiogenesis 32 IL-2 35 IL-4 35 IL-6 35 IL-8 32 IL-10 35 in placental development 35

intermittent auscultation (IA) 713, 719, 721–2 International Society of Ultrasound in Obstetrics and Gynecology (ISUOG) fetal echocardiography guidelines 112, 182, 187, 224 interrupted aortic arch 300, 301 chromosomal anomalies 622 extracardiac anomalies 640 management infants 693 perinatal/neonatal 669 postnatal physiology/ hemodynamics 669 interrupted inferior vena cava with azygos continuation 214, 229, 230, 242, 245 and cardinal vein anomalies 417, 418 and isomerisms 356 interventricular septum 4D US SPIC imaging 211 3D rendering 226 virtual plane 231 intervillous circulation 42–4, 48 intracardiac flow velocity waveforms 162 intracardiac shunt malformations 281–90 intramural coronary artery 344, 345 intraplacental blood flow 52, 53 intrauterine growth restriction see fetal (intrauterine) growth restriction (FGR) (IUGR) intrauterine infections 114 invasion 28–9 blastocysts 30 defective 47 deficient 52–3 irregular cardiac rhythms 439, 441 isomerisms 349 extracardiac anomalies 651 gender differences 350 incidence 348 left vs right 248–9, 249, 356–7, 356 fetal death rates 360 prognosis 248–50 postnatal 360 see also atrial situs, isomerisms isoprenaline 457, 501 isotretinoin teratogenicity 609, 741 isovolumic contraction/relaxation 436 isoxsuprine 733 Jacobsen syndrome 103 Kabuki syndrome 708 Kartagener syndrome 347, 359, 708 Kasabach–Merritt sequence 504, 505 KLF2 gene 12 Klinefelter syndrome 710 Klippel–Feil syndrome 114 Klippel–Trenaunay syndrome 505

Kommerell diverticulum 337, 339, 341 Kruppel-like factor 2 gene 12 ladder diagrams 452 left aortic arch defect 330 aberrant right subclavian artery 334 left atrial isomerisms 245–7, 245, 246 and heart block 455–6 left atrium 19–20 left heart malformations 291–303 aortic valve anomalies 292, 294–7 aortic atresia 294 aortic–left ventricular tunnel 296–7 aortic stenosis 295–6 arch anomalies 297–300 associated extracardiac anomalies 300 management 300–2 mitral valve anomalies 291–2 mitral atresia 291 mitral regurgitation 292 mitral stenosis 291–2 outcome 302 prenatal counseling 300, 302 prognosis 301–2 left isomerism 356–7 fetal death rate 360 left ventricle diagram 307 diastolic function and cardiomyopathy 366 in dilated cardiomyopathy 375, 376, 377, 378 morphologically left ventricle 21, 21 outflow tracts 308 in TGA 318–19 primitive 9, 10 rudimentary 24 left ventricle non-compaction 104 left ventricular outflow tract obstruction 727 levocardia 25, 239, 240, 245 visceral/atrial situs 355 lidocaine 733 lithium 741 liver perfusion 547, 548 ‘liver sparing’ effect 547 long QT syndrome 104, 106, 462, 465 persistence 478 treatment 475, 686 long VA tachycardia 463 treatment 475 looped heart 101 Low–Ganong–Levine syndrome 463 lungs 181 lymphatic system 486 magnesium sulfate 742 magnetic resonance imaging (MRI) 404, 405 magnetocardiography 461–2, 463 Mahaim syndrome 463–4

Index

major histocompatibility complex (MHC) receptors 33 Marfan syndrome 104, 114, 708 maternal, during pregnancy 726, 730–1, 734, 737, 738 marijuana 742 maternal age 737 maternal cardiac disease see cardiac disease, maternal maternal diseases affecting fetal cardiovascular system 737–47 autoimmune disorders 742–3 drug exposure 741–2 congenital heart disease 737–8 infectious causes 738–9 metabolic/endocrine causes 739–40 genetic heart disease 737–8 as screening indication 113, 114 see also cardiac disease, maternal; diabetes, maternal maternofetal hyperoxygenation 519, 521 Meckel–Gruber syndrome 114, 645, 646, 649 membrane potentials 435 membranous atresia 272, 272 perinatal/neonatal management 670 postnatal physiology/ hemodynamics 670 mesocardia 241 visceral/atrial situs 355 mesoderm, precardiogenic 9, 10 metabolic causes of cardiomyopathy dilated 379, 380 hypertrophic 382 mexiletene 733 microdeletion 22q11 10, 103, 114, 117 with aortic arch anomalies 337, 339 cardiac anomalies 613, 622, 625 in CATCH-22 syndrome 181 in conotruncal anomalies 323, 324 diagnostic testing 613 extracardiac anomalies 646–7 frequency 613 in heterotaxy 350 inheritance 708 and interrupted aortic arch 300 and isomerisms 250 in ventricular septal defect 289 microdeletion syndromes 711 microseptostomy 604 middle cerebral artery 753 Doppler spectral tracing 753 Doppler ultrasound 428–30, 429 flow velocity waveforms in FGR 542 minimally-invasive surgery 517, 519–20, 523 complications 524, 527 mortality rates 525 miscarriage gestational sac 49

placental disorders 46, 47 spiral arteries 49 mitral atresia 291, 293 four-chamber view 292, 294 in neonate 591 prognosis 301 mitral regurgitation 292, 294, 377 with aortic stenosis 296, 296 congestive heart failure 572, 572 maternal, during pregnancy 730 prognosis 301 mitral stenosis 291–2, 294 management infants 693 perinatal/neonatal 669 maternal, during pregnancy 726, 729 postnatal physiology/ hemodynamics 669 prognosis 300–1 mitral valve anatomy 21, 291, 306, 306 flow velocity waveforms 536 4D US SPIC imaging 208 prolapse 106 3D surface rendering 209 mitral valve anomalies 291–2 insufficiency and hydrops fetalis 498 monosomy 103 partial 117 monosomy 1p36 708 monosomy X see Turner syndrome (monosomy X) morphogenesis 9–17 motion detection echocardiography 93 mucopolysaccharidoses 104 multifocal atrial tachycardia 463 multiplanar reconstruction (MPR) 222, 223 mumps 739 mutation analysis 629 myocardial biopsy 684–5 myocardial blood flow 467 myocardial diseases 363–7 extracardiac anomalies 636 and hydrops fetalis 484, 502–5 therapeutic considerations 369 myocardial flow reserve 390 myocardial function abnormal 571–3 estimation 490–1 myocardial heart tube 9 myocardial hypertrophy 492, 505, 572 myocardial insufficiency, global 363 myocardial level right heart anomalies 262–7 myocardial performance index 572–3 twin–twin transfusion syndrome 604 myocardial perfusion, regulation of 386–7 myocardium contractility 142, 157, 158

771

in hydrops fetalis 492 and maturation 159, 160 development 5, 11, 14 energy sources 562 fractional shortening 571 maturation, structural/ functional 159–60 postnatal changes, morphological 151 recruitment at arterial pole 10, 12 sinus venosus 14 myosin 160 National Institute for Clinical Excellence (NICE) 276 National Institute of Child Health and Human Development (NICHD) 713, 714 National Maternal PKU Collaborative Study 609 natural killer (NK) cells in angiogenesis/remodeling 32 decidual (dNK) 32, 33 KIR receptors 32 maternal immune tolerance 32 neonates aortic atresia 591 arrhythmia treatment 685–6 balloon aortic valvuloplasty 680–1, 680 balloon atrioseptostomy 677, 678 balloon dilatation aortic coarctation 678, 681 other vessels 682 pulmonary valve 679–80, 679 valves/blood vessels 677, 679 blood flow 591 with CHD 659–89 cardiac shunt defects 666–7 duct-dependent heart defects 663, 665 growth parameters 751 persistent fetal circulation 665–6 pulmonary hypertension 665–7 treatment for cardiac insufficiency 675 critical congenital heart defects 675–7 ductus arteriosus opening 666, 674 established procedures/ therapy 677–86 intensive therapy principles 667, 674 pulmonary perfusion regulation 674 pulmonary/systemic perfusion balance 674–5 without suspected problems 667 circulation establishment 660–2 circulation transition 659–67 in congenital heart disease 663–7 normal 659–62 ductal stenting 682–3, 682, 683 heart transplantation 659–89

772

Index

neonates (Continued) interventional occlusions atrial septal defects 684, 685 patent ductus arteriosus 683 ventricular septal defects 684 intravascular stents 682 mitral atresia 591 myocardial biopsy 684–5 pulmonary atresia 590, 659–89 nerve supply 48 placenta/umbilical cord 41 uterus 41 neural crest cells 10, 11, 13 neurofibromatosis 738 9q subtelomere deletion syndrome 708 nitrates 733 nitric oxide (NO) mechanisms 147–8 Nkx2.5 genes 13, 13 non-compacted cardiomyopathy 366–7, 367, 379, 380, 382–3 familial 367 and hydrops fetalis 503 Noonan syndrome 102, 104, 114, 381 cardiac anomalies 614 extracardiac anomalies 649, 650 inheritance 708 maternal, during pregnancy 738 umbilical veins in 418–19 norepinephrine (noradrenaline) 143 Norwood procedure 693 nuchal edema 627 nuchal translucency (NT) screening and cardiac anomalies 612 CHD detection, early 191, 506, 627–8 CHD prevalence 116, 116 echocardiography referrals 226 for trisomy 21 627 occlusion ductal constriction/occlusion 568 interventional atrial septal defects 684, 685 patent ductus arteriosus 683, 684 ventricular septal defects 684 umbilical cord, postnatal 146–7, 147 oligohydramnios 483, 597, 598 omphalocele 643–4, 644 Opitz syndrome 103 organogenesis 10, 11, 186, 413 Osler4Rendu4Weber syndrome 104 osteogenesis imperfecta 104 ostium primum ASD (partial AVSD) 281, 281, 282, 284 associated disorders 284 four-chamber view 282, 283, 284 ostium secundum ASD 281, 281, 282 outlet/outflow tracts anatomy 20, 21 congenital malformations 10, 12 development 2, 11 flow velocity waveforms 164, 535

4D US SPIC imaging 204, 212 in isomerism 248 peak flow velocities 164 right vs left 305–6, 306 tomographic ultrasound imaging 227 see also ventricular outflow tracts outlet septum see infundibular (outlet) septum oxidative stress 44 placental degeneration 46–7 trophoblast sensitivity 43 oxygen delivery, myocardial 386 oxygen free radicals (OFR) 43, 44 oxygen partial pressures see PO2 (partial pressure of oxygen) oxygen regulation 43 oxygen saturation 136 cardiac effects of changes 585–6 in cerebral circulation in CHD 429, 430 heart/great vessels 135 human fetal circulation 586 lamb fetal circulation 585 maternal oxygen administration 136–9 porta hepatis region 136 oxygenated blood streaming patterns 134, 159 venous blood admixture 135 pacing/pacemakers external 458 fetal cardiac pacing 700 and heart rate 142 implanted 686 latent 436–7 sinoatrial node 161 PAIVS see pulmonary atresia, with intact ventricular septum Pallister–Killian syndrome 103, 626 parasympathetic innervation 161 paroxysmal SVT intermittent 471 treatment 475 partial amniotic cavity insufflation (PACI) 519 partial anomalous pulmonary venous connection (PAPVC) 420–1, 421 partial (ostium primum ASD) 420–1 partial pressure of oxygen see PO2 (partial pressure of oxygen) parvovirus 739 Patau syndrome see trisomies, trisomy 13 patent (persistent) ductus arteriosus 106, 593 in aortic atresia 295 chromosomal anomalies 622 and hydrops fetalis 496 incidence 107

maternal, during pregnancy 726, 729, 729 perinatal/neonatal management 672 postnatal physiology/ hemodynamics 672 see also arterial duct peak flow velocities coronary arteries 388, 389, 395 outflow tracts 164 in severe anemia 393 systolic 164 peak systolic velocity (PSV) 45 peak velocity (PV) 532 normal ranges 537–8 percutaneous ultrasound-guided interventions 515–17 fetal balloon valvuloplasty 276, 515–17, 516, 574, 581 four-chamber view before/after 517 with maternal laparotomy 517 morbidity/mortality rates 516 selection criteria 517 thoracoamniotic shunt placement 524, 525 perfusion, fetoplacental 157 pericardial effusion 369, 369 in dilated cardiomyopathy 375 in hypertrophic cardiomyopathy 382 therapeutic considerations 369 peripartum cardiomyopathy 726, 731 periventricular leukomalacia 750, 750 permanent junctional reciprocating tachycardia (PJRT) 444, 445, 463 persistence 478 treatment 477 persistent ductus arteriosus see patent (persistent) ductus arteriosus persistent fetal circulation 665–6 persistent left superior vena cava 396, 418 persistent pulmonary hypertension of newborn 148 persistent right umbilical vein 419, 419 persistent truncus arteriosus frequency in CHD 115 incidence 107 phenylketonuria maternal 609, 740 as screening indication 114 phenytoin 741 Pitx2 gene 14 placenta abnormal placentation 47–8 blood supply during gestation 158 blood velocity 43, 43 hydatid mole 46 implantation/development 27–40 requirements 41 ‘jelly like’ US appearance 47 normal vs pregnancy failure 46 vasculogenesis/angiogenesis 29–32

Index

venous–placental–arterial interaction 415, 416 placental blood flow 169 placental blood flow resistance 167 placental circulations umbilicoplacental abnormal development 51–3 normal development 48–51 uteroplacental abnormal development 45–8 normal development 41–5 placental disorders 45–6, 47–8 placental insufficiency 169 placentomegaly 483, 503 PO2 (partial pressure of oxygen) cardiac effects of changes 585–6 and ductus arteriosus closure 148, 149 lamb fetus 136 birth-associated changes 145, 147, 148 in preterm infant 149 polyhydramnios 483, 597 polysplenia 347, 349, 350, 351, 351, 359, 360 portal system anomalies 420 portal veins 414, 415 velocimetry 555 portal venous blood course 133, 134 post-fontan operation 728 postnatal changes and cardiovascular anomalies 587–93 potassium ions 435 power Doppler imaging 74, 221–2 predictive testing 709 preeclampsia blood velocity 48 maternal immune tolerance 35 natural killer (NK) cells 32 normal/defective development 34 placental disorders 46, 47 pregnancy cardiac disease in 725–35 hemodynamic changes 725 preload 142–3, 143, 157, 158 and cardiac output 159 in hydrops fetalis 493 in twin–twin transfusion syndrome 599 preload index (PLI) 533 normal ranges 537–8 premature atrial contraction (PAC) 437, 439, 442 bigeminal 452, 453, 454 premature ventricular contraction 439, 442 prenatal counseling see counseling, prenatal prenatal diagnosis autosomal dominant inheritance 709 autosomal recessive inheritance 709 inherited disease 709

X-linked recessive inheritance 709 pressure waves and blood velocity 549–50 generation 548, 549 preterm premature (elective early) delivery 471, 520 primary ciliary dyskinesia 347, 359 primary fold development 4–5, 5 primary heart tube 9, 10, 101 procainamide 732 proepicardial organ (PEO) 13, 14 propafenone 476, 478, 732, 733 propranolol 473, 475, 478 prostacyclin (PGI2) 147 prostaglandins 591, 666, 667, 674, 694, 742 side effects 674 prosthetic heart valves 730 pseudoxanthoma elasticum 104 pulmonary arterial pressure changes, perinatal 663 pulmonary arteries blood pressure, perinatal 147, 148 blood velocity 140–1, 140 after vasodilator 140–1, 141 embryology 414 flow velocity waveforms 536 in FGR 541 4D US SPIC imaging 207, 210, 212 oxygen saturation 585–6 in PAIVS 275 in tetralogy of Fallot 313, 313, 314, 315, 316 see also branch pulmonary arteries pulmonary artery sling 340 pulmonary atresia 25, 258–60, 259, 583 in congestive heart failure 568 frequency in CHD 115 and hydrops fetalis 494–5 with intact ventricular septum (PAIVS) 266 anatomical findings/ultrasound correlates 269–75 associated disorders 275 biventricular repair 277 with cardiomegaly 273 diagnosis 269 fetal intervention 275–6 follow-up 278 four-chamber view 269, 270, 271 good-sized right ventricle 272 incidence 269 in infants 695 ‘one-and-a-half’ repair 277 outcome 277–8 fetal indicators 277 pathophysiology/ultrasound findings 275 postnatal diagnosis/ management 276–7, 276 prenatal natural history 275

773

presentation, neonatal 269 univentricular repair 276 and isomerisms 356 in neonate 590 in neonates 665, 665 peak velocity index 432 perinatal/neonatal management 670, 671 postnatal physiology/ hemodynamics 670, 671 with tetralogy of Fallot 260 with ventricular septal defect 316, 316 in infants 694 pulmonary banding, surgical 683, 695 pulmonary circulation 593–4 blood flow 138–9 and hydrops fetalis 494–5 perinatal 145, 145, 659 through gestation 166 development 148 Doppler indexes 165–6 ductus arteriosus obstruction 583–4 flow velocity waveforms 537 neonatal perfusion regulation 674 pulmonary/systemic perfusion balance 674–5 perinatal 147–8 postnatal adaptation 148, 579 vascular resistance, perinatal 145, 146, 146, 587–90 abnormal communications 587–9 communications in preterm infants 589–90 decreasing 589 and shunting 588 pulmonary hypertension in cardiac shunt defects 666–7 maternal, during pregnancy 725, 727 in total anomalous pulmonary venous return 665–6 pulmonary regurgitation 569 pulmonary stenosis 395 diagnosis, 3D/4D US 232 extracardiac anomalies 636 frequency in CHD 115 incidence 107 inheritance 712 maternal, during pregnancy 726 peak velocity index 432 perinatal/neonatal management 670 postnatal physiology/ hemodynamics 670 with VSD 260–2 without VSD 257 chromosomal anomalies 622 critical 258–60, 258 mild-to-moderate 257–8 pulmonary stenosis with VSD 260–2 pulmonary stenosis without VSD 257–60

774

Index

pulmonary trunk 22 blood velocity 139, 140, 140 and cardiovascular anomalies 583 echocardiography 178–9, 187 pulmonary valve anatomy 306, 306 balloon dilatation 679–80 flow velocity waveforms 536 normal 567 in tetralogy of Fallot 313, 313, 314, 315 see also absent pulmonary valve syndrome pulmonary valve anomalies and congestive heart failure 568–9 in trisomy 18 624 see also specific anomalies pulmonary valvuloplasty 276 pulmonary vascular obstructive disease 726 pulmonary vascular resistance 166 and alveolar fluid clearance 662 in AVSD 287 birth-related changes 145, 146–8, 146, 263 ductus arteriosus closure 582–3, 660 factors/therapies affecting 674, 676 flow velocity profiles 140, 141 in hypoplastic left heart syndrome 675 postnatal changes 583, 586, 587, 659, 661 adaptation to ‘adult’ situation 662–3, 663 in cardiac shunt defects 667, 667 interventions 667 prenatal increase 666 in transitional circulation 563, 659, 661 in VSD 289 pulmonary veins anomalous pulmonary venous drainage 266–7, 266, 417 partial (PAPVC) 420–1, 421 prognosis 422 total (TAPVC) 422 diagnosis, 3D/4D US 229 incidence 107 and isomerisms 248, 357, 357 STIC acquisition 229 embryology 413 flow velocity waveforms 534, 534 flow velocity waveforms (FVWs) 167–8 obstruction in infants 696 in right atrial isomerism 249 pulsatile umbilical venous flow velocity 551–2 pulsatility index (PI) 45, 50 branch pulmonary arteries 165–6 cardiovascular investigation 161 cerebral circulation 428–9, 429, 584 Doppler determination 78

ductus arteriosus 165 intracerebral vessels 165 renal/adrenal arteries 166 umbilical arteries 50, 53, 427–8, 428 pulse oximetry 720 pulsed wave (PW) Doppler 72–3, 532 Purkinje cells 14 Purkinje fibers 435, 436 quinidine 733 real-time compound imaging 84–5, 85 real-time fetal 3D echocardiography 93, 94 reciprocating tachycardia 464 reconstructed fetal 3D echocardiography 88–90, 88 reentry 437 reflection coefficient (RC) 548, 550 renal artery pulsatility index 166 renal disease in dilated cardiomyopathy 378, 379 in hypertrophic cardiomyopathy 382 renin–angiotensin system 599 restrictive atrial septum 294, 295 restrictive cardiomyopathy 366 restrictive ductus arteriosus 263–4, 264 restrictive foramen ovale 284–5, 499 rhabdomyomas 405–8, 738 echocardiography 402, 404 and hydrops fetalis 499 management/outcome 407–8 pathology 407 prevalence 401, 405 and tuberous sclerosis 407 Rhodes score 692 right aortic arch defect 232, 330–1 aberrant left subclavian artery 336, 337 B-mode/B-flow 233 extracardiac anomalies 639 mirror-image branching 335 right atrial isomerisms 246, 247–8, 247, 248 right atrium idiopathic enlargement 251–2, 252 isomerisms 246, 247–8, 247, 248 morphologically right atrium 19 right heart anomalies 251–68 common pathophysiology 251 inlet anomalies 251–7 atrium, idiopathic enlargement of 251–2, 252 tricuspid valve Ebstein’s anomaly 253–5, 254, 255 tricuspid atresia 255–7, 256 tricuspid dysplasia 252, 253 at myocardial level 262–7 diastolic overload of right ventricle 265–7

anomalous pulmonary venous drainage 266–7, 266, 417, 420–2, 421 intracerebral arteriovenous fistula 265–6, 265 primary anomalies 262–3 restrictive ductus arteriosus 263–4, 264 secondary anomalies 263–7 twin–twin transfusion syndrome 264–5, 264, 275 Uhl’s anomaly 262–3, 263 outlet anomalies 257–62 pulmonary atresia 258–60, 259 with tetralogy of Fallot 260 pulmonary stenosis with VSD 260–2 pulmonary stenosis without VSD 257–60 critical 258–60, 258 mild-to-moderate 257–8, 257 right isomerism 356–7, 357 fetal death rate 360 right ventricle diagram 307 diastolic overload 265–7 in dilated cardiomyopathy 375, 376, 377, 378 growth/function abnormal 567–8 normal 566 hypertrophy 566, 568 hypoplasia 396 with increased afterload 567–8 with increased preload 568 morphologically right ventricle 20–1 outflow tracts 308 in TGA 320 in PAIVS 271–3 rudimentary 24 in tricuspid atresia 256 uni-/bi-/tri-partite 271 right ventricle-dependent circulation 274 ritodrine 733 Ross–Konno aortoventriculoplasty 692, 693 rubella 710, 738–9 Rubinstein–Taybi syndrome 103, 114 salbutamol 457, 501, 733 sarcomere 160 sarcoplasmic reticulum 143 Schone syndrome 297 screening congenital heart disease 83, 106–7, 111 cost implications 119–21 high risk population 111, 112 impact 111 low risk population 111, 118–21 echocardiography 107 first/early second trimester 185–96 four-chamber view 121–4

Index

ISUOG guidelines 112 timing 112 ventricular outflow tracts 123–4 segmentation 1–5, 10 semilunar valve incompetence and hydrops fetalis 495–7 surgery 524 septal defects and chromosomal anomalies 622–3, 622 echocardiography 177 in trisomy 21 622–3, 622 see also atrial septal defects; atrioventricular septal defect; ventricular septal defects septation 6–7 septum formation 4–5 sequential segmental analysis 1, 22 sex polysomies 102 short rib–polydactyly syndromes 647, 648, 650 Shprintzen syndrome 103, 114 shunting in fetal heart 561 postnatal changes 659 shunts arterial level 696 atrial level 696 maternal, during pregnancy 726, 728–9 ventricular level 696 single gene defects 102 sinoatrial conduction system 14, 15 sinoatrial node 435, 436 as pacemaker 161 sinoatrial transitional zone 2–3, 2, 3 sinus bradycardia 440, 451–3 M-mode echocardiography 454 sinus node, automatic arrhythmias of 437 sinus tachycardia 465 sinus venosus embryology 413, 414 myocardium 14 sinus venosus atrial septal defect 281, 281, 284 sinusoids 258, 258, 273, 695 persistent myocardial see non-compacted cardiomyopathy situs ambiguus 241, 242 situs invertus 239, 249, 347, 349 and CHD 355 complete 239, 240 extracardiac anomalies 359 incidence 347 situs solitus 239, 240, 249, 347, 349, 350 incidence 347 Smith–Lemli–Opitz syndrome 648, 650, 708 sodium nitroprusside 733 sotalol 472, 474–5, 475, 477, 478 during pregnancy 732

spatiotemporal image correlation (STIC) 90–1, 91, 182, 182, 219, 230 acquisition plane/angle/time 199 acquisition quality 233 aortic/ductal arch visualization 203, 206 B-mode/B-flow 87, 202, 210–11, 212, 214 color Doppler 202, 212, 221–2 combination with other applications 219 coronal atrioventricular (CAV) plane 228, 230, 231, 232 fetal heart examination 199–211, 200, 201, 202 fetal heart rate calculation 198, 198 fetal/maternal movement 225 fetal position 199 five planes of fetal echocardiography 224–5, 225 in 4D US 190, 197–217, 532 ‘gradient light’ mode 210 great vessel visualization 202–3, 209–10, 213 interslice distance by gestation time 203 intracardiac structure visualization 203, 206, 208–9 inversion mode 209, 210, 213, 214 and VOCAL analysis 227, 228 limitations 211, 213–14 multiplanar slicing, automated 201–2, 206 overview image 205 power Doppler 224 region of interest (ROI) 199 rendering 88 inversion mode 213, 214 techniques 203, 206, 208–10 ‘thick slice’ 206, 206, 208, 209, 210 schematic demonstration 220–1 spin technique 203, 207 technology 197–8 TELE-STIC 200 valve visualization 206, 208–9 ventricular outflow tracts 203, 204 volume acquisition 198–9 optimization 199 volume data set 197 display options 197, 205 manipulations 203, 204–5 scrolling through 199–200, 224 specimen examination 22–6 speckle tracking imaging 93–6, 96 spectral analysis 72, 75–7 spiral arteries development 42 in placental development 28–9, 46 remodeling 42

775

spongy myocardium (spongiform cardiomyopathy) see non-compacted cardiomyopathy Spry2 27–8 SSA/Ro autoantibodies 454, 456, 457 SSB/LA autoantibodies 454, 457 ST wave analysis 720 STAN S31 fetal heart monitor 720 stenting, ductal 682–3, 682, 683 stents, intravascular 682 STIC see spatiotemporal image correlation (STIC) strain rate imaging 95–6 stroke volume 157, 532, 562 subaortic stenosis 297 superior vena cava bilateral 355, 356, 356 4D US SPIC imaging 207 persistent left 396, 418 supraventricular reentry tachycardia 468 supraventricular tachycardia (SVT) 461, 465, 468–9, 468 and hydrops fetalis 501 normalization after treatment 469 paroxysmal intermittent 471 treatment 475 treatment 475, 477, 685, 686 surgery 515–29 aortic commissurotomy 693 aortic stenosis repair 301 ASD repair 287 balloon aortic valvuloplasty 680–1, 680, 693 balloon aortoplasty 694 balloon atrioseptostomy 677, 678 balloon dilatation aortic coarctation 678, 681 other vessels 682 pulmonary valve 679–80, 679 valves/blood vessels 677, 679 ductal stenting 682–3, 682, 683 fetal cardiac 697–700 improving outcome 698 percutaneous vs open 699 preventing death 697 requirements 698–9 fetoscopy-assisted interventions 517, 519–20, 523 interventional occlusions atrial septal defects 684, 685 patent ductus arteriosus 683, 684 ventricular septal defects 684 intravascular stents 682 minimally-invasive 517, 519–20, 523 complications 524, 527 myocardial biopsy 684–5 neonatal 677–85 CHD 668–73 Norwood procedure 693

776

Index

surgery (Continued) percutaneous ultrasound-guided interventions 515–17 pulmonary banding 683, 695 radiofrequency ablation 686, 686 Ross–Konno aortoventriculoplasty 692, 693 two-ventricle repair in infants 692–3 vascular ring division 341 VSD repair 289 see also specific techniques SVT see supraventricular tachycardia (SVT) sympathetic innervation 161 and myocardial performance 144 syncytiotrophoblasts 28 systole 436 ventricular 155 ‘T’ artifact 287 T cells see also activated T cells; cytotoxic T lymphocytes (CTLs) tachyarrhythmias 444–6, 461–81 cardiac function assessment 469–70 classification 461 color Doppler imaging 461 definition 461 diagnosis, prenatal 461–2 differential diagnosis 461 echocardiography 461, 462 elective early delivery 471 electrophysiological mechanisms 463–5 follow-up, postnatal 477–8 and hydrops fetalis 484, 499, 500–1 M-mode echocardiography 461, 462, 463, 463 magnetocardiography 461–2, 463 maternal complications 461 pathophysiology 465–9, 500 persistence 478 pulsed wave Doppler imaging 461, 463 recurrence after treatment 477 surveillance fetal 469–70 maternal 471 sustained fetal 461 treatment 470–1, 474–7 antiarrhythmics 472–3 effectiveness 477 monitoring 470 placental effects 474 protocols 471, 474–7 risks 470 maternal surveillance 471 rationale 470–1 time to normalization 469 tachycardia 439 atrial ectopic (AET) 444–5 atrioventricular reentry 437, 444, 461

long VA 445, 445 myocardial involvement 467 permanent junctional reciprocating (PJRT) 444, 445 prenatal presentation 441 prenatal treatment 685 short VA 444, 445 Tbx1 gene 10–11 mutations 12 Tei index 364, 491–2 TELE-STIC 200 tendon of Toledo 19 teratogenic drugs 609, 710 as screening indication 113–14, 113 teratomas 401, 409 four-chamber view 500 and hydrops fetalis 499–500, 500, 503–5, 504 terbutaline 457, 574, 733 termination see abortion tetralogy of Fallot 189, 311, 313–14, 316 with absent/dysplastic pulmonary valve 260–2, 262 anatomical features 311, 311 with aortic arch anomalies 337 cardiac anomalies 614–15 chromosomal anomalies 324, 622, 623 frequency 710 classical form 260, 261 collateral arteries 316, 316 echocardiography basal short-axis view 315 color Doppler 315 four-chamber view 312 three-vessel view 313 extracardiac anomalies 636, 637, 638 extracardiac associations 324 4D US SPIC imaging 210 frequency in CHD 115, 115 and hydrops fetalis 496–7 incidence 107 maternal, during pregnancy 728 outcome 325, 325 peak velocity index 432 perinatal/neonatal management 670, 671 postnatal physiology/ hemodynamics 670, 671 prognosis 261–2 with pulmonary atresia 260, 262 severe, in infants 695 treatment 521 ventricular septal defect 311, 313 juxta-atrial 313 perimembranous 311, 312 with VSD 289 TGA see transposition of great arteries (TGA) TGFβ1 gene 12 thalassemia 503 thalidomide 742

Thebesian valve 3, 19 thrombocytopenia–absent radius (TAR) syndrome 114, 647, 648 thymus visualization 181, 181 thyroid hormone in contractility 160 in perinatal circulatory changes 150 Tie1/2 31, 31 time averaged maximum velocity (TAMX) 161 time averaged velocity (TAV) 161 time to peak velocity (TPV) 161, 532 arterial flow velocity waveform 164 normal ranges 538 time velocity integral (TVI) 532 normal ranges 537–8 tissue Doppler echocardiography 78–9, 83–4, 93 color 93, 95 pulsed wave 93, 95 tissue harmonic imaging 68–9, 70, 84, 87 tocolytic drugs 742 tomographic ultrasound imaging (TUI) 201, 201, 202, 203, 224, 227 total anomalous pulmonary venous connection (TAPVC) 422 diagnosis, 3D/4D US 229 incidence 107 and isomerisms 357, 357 in right atrial isomerism 248 STIC acquisition 229 total anomalous pulmonary venous return 665–6 perinatal/neonatal management 671 postnatal physiology/ hemodynamics 671 toxic cardiomyopathy 363 trabeculations 20, 21, 21 transabdominal sonography (TAS) 185 transdiaphragmatic cardiac access 519, 520 transesophageal echocardiography (TEE) 518, 519 transforming growth factor (TGF) 12, 29 transforming growth factor (TGF) genes 29 transitional zone development atrioventricular canal 3–4 primary fold 4–5, 5 sinoatrial 2–3 transposition of great arteries (TGA) with aortic arch anomalies 337 B-mode/B-flow 232 and cerebral circulation in CHD 429, 430 chromosomal anomalies 622 frequency 710 with coarctation of aorta 345 complete 317–19, 317 chromosomal anomalies 324

Index

echocardiography 317, 318 extracardiac associations 324 outcome 325, 325 with ventricular septal defect 318 congenitally corrected 319–21, 320 chromosomal anomalies 324 extracardiac anomalies 636 extracardiac associations 324 outcome 325, 325 visceral/atrial situs 355 in conotruncal anomalies 612–13 definition 343 diagnosis 343–5 diagnostic rates 343 extracardiac anomalies 636, 640 4D US SPIC imaging, rendered 213 frequency in CHD 115 incidence 107 in infants 696 and isomerisms 357, 359 maternal, during pregnancy 728 in neonates 665, 666 pathology diagram 317 perinatal/neonatal management 669 postnatal physiology/ hemodynamics 669 screening outcomes 120 3D/4D US diagnosis 230–2 with tricuspid atresia 255 with ventricular septal defect 343, 344 see also aortopulmonary transposition transvaginal sonography (TVS) 185, 186, 610 as screening method 611 tri-iodothyronine (T3) 150–1 tricuspid atresia 255–7 Doppler ultrasound 432 extracardiac anomalies 636, 639 flow changes 582 flow velocity waveforms 432 frequency in CHD 115 and hydrops fetalis 494–5 in infants 695 outcome 495 peak velocity index 432 perinatal/neonatal management 671 postnatal physiology/ hemodynamics 671 tricuspid dysplasia 252, 253 extracardiac anomalies 636, 640 tricuspid regurgitation 294, 377, 569 as chromosomal anomaly marker 628–9 congestive heart failure 567, 572, 572 ductus venosus blood flow 433 in Ebstein’s anomaly 571 in hydrops fetalis 492 pathophysiology/ultrasound findings 433 tricuspid valve 566–7 anatomy 19, 20, 306, 306

4D US SPIC imaging 208 in PAIVS 271, 272 3D surface rendering 209, 231 in VCAC 274 tricuspid valve anomalies in aortic atresia 294 and congestive heart failure 569 in hydrops fetalis 494–5 in trisomy 18 624 see also specific anomalies trimethoprim sulfonamide 741–2 triploidy 117, 646 trisomies 102, 103 and cardiac anomalies 610, 636 and isomerisms 249 in left heart malformations 300 partial (cat-eye syndrome) 103, 117 trisomy 13 (Patau syndrome) cardiac anomalies 621, 622, 624–5 and CHD 103, 117 and endocardial echogenic foci 368 extracardiac anomalies 645–6 frequency of heart disease 710 and hydrops fetalis 506 structural anomalies 625 in truncus arteriosus 323 trisomy 16 615 trisomy 18 (Edward syndrome) cardiac anomalies 621, 622, 624, 624 and CHD 103, 117 extracardiac anomalies 645 frequency of heart disease 710 in heterotaxy 350 and hydrops fetalis 506 prenatal detection rate 624 in truncus arteriosus 323 trisomy 21 (Down syndrome) with aortic arch anomalies 339 cardiac anomalies 610, 612, 615, 621, 622–4 associations 622, 636 studies 623 and CHD 103, 105, 117, 621 ‘double bubble’ sign 643 and endocardial echogenic foci 368 extracardiac anomalies 645 frequency of heart disease 710 and hydrops fetalis 506 and ostium primum ASD 284 and pericardial effusion 369 pulmonary vascular resistance, postnatal 667 in truncus arteriosus 323 trophoblastic plugs 41, 42, 44, 47, 49 trophoblastic shell 41, 42 early embryo loss 46 formation 42, 46 trophoblasts adhesion molecule expression 29 extracellular matrix digestion 29 implantation 28

777

invasion 28–9 deficient 52–3 maternal immune tolerance 32, 34–5 see also extravillous trophoblasts (EVT) troponins 160 truncoconal anomalies see ventricular outflow tracts, congenital malformations truncus arteriosus 322–3 chromosomal anomalies 324 echocardiography 323 embryology 414 extracardiac associations 324 outcome 325, 325 pathology diagram 323 perinatal/neonatal management 672 postnatal physiology/ hemodynamics 672 TTTS see twin–twin transfusion syndrome (TTTS) tuberous sclerosis 114, 407, 649 tumors see cardiac tumors; vascular tumors Turner syndrome (monosomy X) and cardiac anomalies 610, 621, 622, 626 and CHD 103, 117 extracardiac anomalies 646 frequency of heart disease 710 and hydrops fetalis 506 and left heart malformation 300 21q22 mutations 615 22q11 deletion syndrome see microdeletion 22q11 twin gestation 113 twin–twin transfusion syndrome ‘Barker hypothesis’ 607 CHOP Cardiovascular Score 603, 604, 606 pathophysiology 599, 606 twin–twin transfusion syndrome (TTTS) 264–5, 264, 597–608, 597 cardiovascular function 606 cardiovascular impact 599–600 assessment of burden 601–4 cartoon drawing 598 causes 598 CHOP Cardiovascular Score 605 diagnosis/findings 597–8 donor twin cardiovascular function 606 ductus venosus 602 mitral Doppler sampling 603 tricuspid Doppler sampling 603 umbilical Doppler sampling 602 echocardiographic findings 599–600 fetal death 598 and hypertrophic cardiomyopathy 365, 382, 383 mechanisms 599

778

Index

twin–twin transfusion syndrome (TTTS) (Continued) natural history 597 and PAIVS 275 perinatal/neonatal management 672 possible vascular connections 599 postnatal physiology/ hemodynamics 672 Quintero staging criteria 598 recipient twin ascites 599 cardiovascular function 606 dilated atria 600 four-chamber view of heart 601 mitral Doppler sampling 603 pulmonary insufficiency 602 tricuspid Doppler sampling 603 tricuspid/mitral regurgitation 601 umbilical Doppler sampling 602 right ventricular function 568 speculations/long-term issues 606–7 surgery 516, 519, 522 treatment strategies 604–5 umbilical flows 602 ventricular performance 604 Uhl’s anomaly 262–3, 263 ultrasound abnormal placentation 47–8 array transducers 64–5, 64, 65, 66 full matrix 92 future developments 96–7 sparse matrix 92–3 2D matrix 83–4 beam parameters 58 consistency 58 coronary circulation 385–99 applications in cardiac abnormalities 393–6 in normal fetuses 390–3 Doppler see Doppler ultrasound and ‘heart sparing’ 390–2 image formation in high resolution systems 63–71 dynamic range 66, 67, 69 compression 67–8, 67 post-processing 67–8, 68 time gain compensation 66 focusing 63–5 with array transducer 64–5, 64, 83–4, 84 beam-width artifacts 65 effects of array parameters 65, 66 fixed lens 63–4, 64 out-of-plane 65 grating lobes 65–6 effects of array parameters 66 grey scale 66, 66 systems control 70–1, 71 persistence 68 recent developments 68–71

coded excitation 69–70, 71 tissue harmonic imaging 68–9, 70, 84, 87 side lobes 65 effects of array parameters 65 signal-processing chain 63, 63 image generation 57 image quality 57, 70–1 parameters 58 noise 58, 60 physics 57–81 pulse length 58 sensitivity 58 sound propagation 57–63 attenuation 58–60 absorption 59, 59 acoustic impedance 59–60 enhancement 60, 60 scattering 59, 59 shadowing 60, 60 specular reflection 58–9, 59 beam formation 58 frequency/bandwidth 58, 59 interference, destructive/ constructive 57 resolution contrast 58 contrast (subject contrast) 61 spatial (detail) 58, 61 reverberations 62–3 speckles 61–2, 86 image compounding 62 reduction 62, 85–6, 86 tracking 93–6 under-sampling/aliasing 62 sound velocity in tissue 58 strain/strain rate 93–6 transabdominal sonography (TAS) 185 transvaginal sonography (TVS) 185, 186 operator skill 191 uteroplacental circulation 41, 44 wavelength/phase 57 see also echocardiography umbilical arteries 52 extracardiac Doppler ultrasound 427–8, 428 flow velocity waveforms 535 in FGR 540, 542 umbilical cord occlusion, postnatal 146–7, 147 placental insertion 52 umbilical cord blood flow 51 Doppler ultrasound 569–70 with supraventricular tachycardia 467 umbilical–fetoplacental circulation 78 umbilical flow velocity waveform 167 umbilical veins anatomy 414, 415 anomalies 417, 418–19 blood velocity 141–2, 551

embryology 413, 414 flow velocity waveforms 534–5, 535 in FGR 542 persistent right 419, 419 physiology 547 umbilical venous flow 550–1 calculation 550–1 during gestation 552 gestational age 550 improvement after atrial flutter 551 pulsatile velocity 551–2 atrial vs arterial origin 552 pulsatility 548 velocity 141–2, 551, 553 velocity waveforms 534–5, 535 in FGR 542 volume 167 umbilical venous pressure 135–6 umbilical venous system 133–4, 135 umbilicoplacental circulation abnormal development 51–3 hemodynamics 49 normal development 48–51 in vivo features 49–50 unifocalization 694 univentricular heart 622 ureteroplacental insufficiency 542 urogenital defects 644 US see ultrasound uterine artery ascending 41 resistance 44 uterine circulation flow impedance in pregnancy 45, 50 in placental development 42 uteroplacental blood flow 167 uteroplacental circulation abnormal development 45–8 normal development 41–5 ultrasound features 44 VACTERL (VATER) association 114, 644, 648 and complete AVSD 286 extracardiac anomalies 636, 647 inheritance 708 valves development 6–7, 14 incompetence in hydrops fetalis 492 insufficiency, maternal 726 prosthetic, maternal, during pregnancy 730 visualization 206, 208–9 see also specific valves vascular endothelial growth factor receptors (VEGFR) 31 vascular endothelial growth factor (VEGF) in placental development 29–31 schematic representation 31

Index

vascular endothelial growth factor (VEGF) genes 12 vascular pressures 135–6 perinatal 146, 147 vascular rings aortic arch anomalies 331, 334–5, 339–40 surgical division 341 in infants 696–7 vascular steal 524, 584 vascular tumors 505 vasculogenesis 29–32 VATER see VACTERL (VATER) association VCAC see ventriculocoronary arterial communication (VCAC) VEGF see vascular endothelial growth factor (VEGF) VEGF genes see vascular endothelial growth factor (VEGF) genes vein of Galen ‘aneurysm’ 505 velocardiofacial syndrome 103, 613, 711 inheritance 708 venae cavae blood velocity 141 velocimetry 555 venous Doppler 547 in congenital heart disease 570 using measurements 555–6 venous flow ductus venosus 553 velocity 553–5 in fetal growth restriction 547–59 inferior vena cava 548 pulsatility, determinants of 548, 550 umbilical veins 550–1 pulsatile velocity 551–2 velocity 551 velocimetry 555 velocity waveform 161 venous flow velocity waveform 162 venous pole cardiac cell recruitment 13 septation/valve formation 6 venous sinus development 1–2, 2 venous system, fetal 413–26 anatomy 414, 415 anomalies 415–22 associated malformations 416 classification 416, 417 and breathing movements 415 embryology 413–14 physiology 414–15 ultrasound examination 416 venous–placental–arterial interaction 415, 416 ventilation, postnatal 145–7 pulmonary circulation, effects on 147–8 ventricles 321 development

and aortic stenosis 580 and blood flow/volume changes 579–81 echocardiography 176–7 3D/4D 230, 232–3 inflow/outflow obstruction 579–80 premature ventricular contraction 439, 442 right vs left 566 systolic flow velocities 439 wall motion 438 see also double-outlet ventricle; left ventricle; right ventricle ventricular ejection force (VEF) calculation 532 congestive heart failure 574 evaluation 491 in FGR 542–3 normal ranges 538 ventricular filling in fetal growth restriction 539 ventricular function 565 diastolic 491 global 491–2 postnatal adaptation 661–2 systolic 490–1 ventricular inlet development 2, 4, 5 ventricular level shunts 696 ventricular outflow tracts anatomy, normal 305–6 congenital malformations (conotruncal anomalies) 12, 230–2, 305–27 with aortic arch anomalies 337, 339 chromosomal anomalies 323–4, 324, 646 distribution 305, 305 echocardiographic evaluation 307–9 and extracardiac anomalies 637–8 extracardiac associations 324, 324 frequency of different malformations 612 individual lesions 311–23 outcome 324–5, 325 postnatal course 325 development 2, 5 echocardiography imaging planes 307 screening 123–4, 186, 203, 204 techniques 307, 309 basal short-axis view 307, 309 Doppler ultrasound 309 four-chamber view 307, 310 long-axis view 310 scanning maneuver 310 three-vessel view 307–8, 309 3D with spin 308–9 extracardiac anomalies 640 and hydrops fetalis 497–9 left, obstruction 640 long-axis view 307, 310 obstruction, maternal 726

779

right obstruction extracardiac anomalies 640 in TTTS 599 vs left 306, 307, 308 in TGA complete 318–19 congenitally corrected 319, 320 ventricular outputs 157, 158 foramen ovale obstruction 584 lamb fetus 159 through gestation 168–9 ventricular septal defects (VSDs) 230, 287–9 anatomy 287 with aortic arch anomalies 337 and aortic atresia 292 associated lesions 288–9 extracardiac 289 chromosomal anomalies 615–16, 622, 623 frequency 710 color Doppler imaging 231 in double-outlet ventricles 321, 322 and extracardiac anomalies 636–7, 636 fetal diagnosis 287–8 flow across defect 287 four-chamber view 287 frequency in CHD 115 incidence 107 inheritance 712 maternal, during pregnancy 726, 728–9 and mitral atresia 291 natural history/outcome 289 in newborns 106–7 perimembranous 623 perinatal/neonatal management 672 postnatal changes 667 postnatal physiology/ hemodynamics 672 subaortic 322, 322 surgical repair 289 ‘T’ artifact 287 in tetralogy of Fallot 311, 313 in TGA complete 318 congenitally corrected 319 3D reconstruction 288 in transposition of great arteries 343 with tricuspid atresia 255, 256 in trisomy 13 625 in trisomy 18 624 in trisomy 21 622, 623 in truncus arteriosus 322 ventricular tachycardia 461, 464–5, 465 treatment 477 ventricular volumes 232–3 ventriculoarterial junctions concordant/discordant 25, 25 double/single outlet 25

780

Index

specimen examination 24–5 3D/4D ultrasound 230–2 ventriculocoronary arterial communication (VCAC) 15, 273–4, 274 ventriculocoronary connections 393–4, 396 ventriculoinfundibular fold 4 ventriculomegaly 641, 641 verapamil 732 very short VA tachycardia 463 villi development 27–8, 28 vascular, abnormal 51–3 differentiation 27 intervillous circulation 42–4 mesenchymal 27 villous circulation 48–9 virtual organ computer-aided analysis (VOCAL analysis) 224, 227 virtual planes 228, 231

visceral blood flow 166 visceral situs 347–62 determination 351–4 features 348 fetal/neonatal outcomes 360 vitelline veins anomalies 417 embryology 413, 414 VOCAL analysis see virtual organ computer-aided analysis (VOCAL analysis) volume echocardiography 88 volumetry, fetal cardiac 93, 94 volvulus 359 VSDs see ventricular septal defects (VSDs) wall motion 438 ‘wall-to-wall’ hearts 270 warfarin 732, 733 wave form analysis 77–8

advancing gestation 163–4 E/A diagram 163 E/A ratio 162–3 formulas 163 wave propagation 548, 549 wave reflection determination 548 well-being, fetal 713–24 Williams–Beuren syndrome 114, 711 Williams syndrome 103, 614, 626 inheritance 708 Wolf–Hirschhorn syndrome 103, 117, 627 frequency of heart disease 710 Wolff–Parkinson–White syndrome 407, 463, 464 treatment 685, 686 World Health Organization (WHO) CHD screening criteria 118–20 X-linked recessive inheritance 706 prenatal diagnosis 709