Environmental Toxicants: Human Exposures and Their Health Effects, 3rd Edition

ENVIRONMENTAL TOXICANTS

ENVIRONMENTAL TOXICANTS Human Exposures and Their Health Effects Third Edition

Edited by MORTON LIPPMANN

Copyright Ó 2009 by John Wiley & Sons, Inc. All rights reserved Published by John Wiley & Sons, Inc., Hoboken, New Jersey Published simultaneously in Canada No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning, or otherwise, except as permitted under Section 107 or 108 of the 1976 United States Copyright Act, without either the prior written permission of the Publisher, or authorization through payment of the appropriate per-copy fee to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, (978) 750-8400, fax (978) 750-4470, or on the web at www.copyright.com. Requests to the Publisher for permission should be addressed to the Permissions Department, John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, (201) 748-6011, fax (201) 748-6008, or online at http://www.wiley.com/go/permission. Limit of Liability/Disclaimer of Warranty: While the publisher and author have used their best efforts in preparing this book, they make no representations or warranties with respect to the accuracy or completeness of the contents of this book and specifically disclaim any implied warranties of merchantability or fitness for a particular purpose. No warranty may be created or extended by sales representatives or written sales materials. The advice and strategies contained herein may not be suitable for your situation. You should consult with a professional where appropriate. Neither the publisher nor author shall be liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages. For general information on our other products and services or for technical support, please contact our Customer Care Department within the United States at (800) 762-2974, outside the United States at (317) 572-3993 or fax (317) 572-4002. Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic formats. For more information about Wiley products, visit our web site at www.wiley.com. Library of Congress Cataloging-in-Publication Data: Environmental toxicants : human exposures and their health effects / [edited by] Morton Lippmann. – 3rd ed. p. ; cm. Includes bibliographical references and index. ISBN 978-0-471-79335-9 (cloth) 1. Environmental health. 2. Environmental technology. I. Lippmann, Morton. [DNLM: 1. Environmental Pollutants–adverse effects. 2. Environmental Exposure. 3. Environmental Health. 4. Environmental Pollutants–toxicity. WA 671 E615 2009] RA565.E58 2009 363.7–dc22 2008036266 Printed in the United States of America 10 9 8 7 6 5 4 3 2 1

CONTENTS

PREFACE CONTRIBUTORS

1 Introduction and Background 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 1.10 1.11 1.12 1.13

xvii

1

Characterization of Chemical Contaminants, 2 Human Exposures and Dosimetry, 7 Chemical Exposures and Dose to Target Tissues, 8 Concentration of Toxic Chemicals in Human Microenvironments, 9 Inhalation Exposures and Respiratory Tract Effects, 12 Ingestion Exposures and Gastrointestinal Tract Effects, 18 Skin Exposure and Dermal Effects, 19 Absorption through Membranes and Systemic Circulation, 20 Accumulation in Target Tissues and Dosimetric Models, 21 Indirect Measures of Past Exposures, 22 Characterization of Health, 23 Exposure–Response Relationships, 25 Study Options for Health Effects Studies, 31 References, 35

2 Perspectives on Individual and Community Risks 2.1 2.2 2.3 2.4

xv

39

Nature of Risk, 39 Identification and Quantification of Risks, 41 Risk Communication, 46 Risk Reduction, 49 References, 52 v

vi

CONTENTS

3 Reducing Risks—An Environmental Engineering Perspective 3.1 3.2 3.3 3.4 3.5 3.6

Introduction, 55 Environmental Risk-Based Decision Making, 56 Applications and Use, 60 Recent Information, 67 Integrated Assessments, 71 Summary, 72 References, 73

4 Clinical Perspective on Respiratory Toxicology 4.1 4.2 4.3 4.4 4.5 4.6 4.7

5.2 5.3 5.4 5.5 5.6

107

The Life Cycle of a Chemical: Many Points for Possible Intervention, 108 The Knowledge Base for the Identification of Hazard Control Strategies, 109 Industrial Hygiene and Occupational Health Programs: Implementing the Knowledge Base, 111 Product Stewardship, 114 Responsible CareÒ, 117 Concluding Perspective, 119

6 Drinking Water Disinfection By-Products 6.1 6.2 6.3

77

Concepts of Exposure, 78 Tools for Studying Individuals, 79 Tools for Studying Populations, 88 Cardiovascular Responses, 95 Limitations of Clinical and Epidemiological Assessments of the Effects of Inhaled Agents, 96 Advice and Counseling of Patients, 97 Summary, 99 References, 100

5 Industrial Perspectives: Translating the Knowledge Base into Corporate Policies, Programs, and Practices for Health Protection 5.1

55

Introduction, 121 Chemical Methods of Disinfection, 122 Chemical Nature and Occurrence of Disinfectant By-Products, 124 6.4 Associations of Human Disease with Drinking Water Disinfection, 132 6.5 General Toxicological Properties of Disinfectants, 144 6.6 General Toxicological Properties of Disinfectant By-Products, 145 6.7 Carcinogenic Properties of Disinfectants, 154 6.8 Carcinogenic By-Products of Disinfectants, 154 6.9 Effects of Disinfectants and Their By-Products on Reproduction, 165 6.10 Effects on Development, 168 6.11 By-Products of Potential Interest, 170

121

CONTENTS

vii

6.12 Summary and Conclusions, 172 Glossary, 174 References, 176 7 Food 7.1 7.2 7.3 7.4 7.5 7.6 7.7 7.8

197 Introduction, 197 Legal and Regulatory Framework in the United States, 201 Toxicity Test Requirements and Safety Criteria, 203 Substances Intentionally Added to Food, 208 Food Contaminants of Industrial Origin, 216 Constituents and Contaminants of Natural Origin, 219 Food Safety in the European Union, 229 Summary and Conclusion, 234 Acronyms, 235 References, 235

8 Volatile Organic Compounds and Sick Building Syndrome 8.1 8.2 8.3 8.4 8.5 8.6

Introduction, 241 Prevalence of Exposures to Volatile Organic Compounds, 242 Health and Volatile Organic Compounds, 245 Prevalence of the Sick Building Syndrome, 247 Dose–Response Relationships for Health Effects Caused by Low-Level VOC Exposure, 249 Guidelines for Volatile Organic Compounds in Nonindustrial Indoor Environments-Principles for Establishment of Guidelines, 251 References, 254

9 Formaldehyde and Other Aldehydes 9.1 9.2 9.3

257

Background, 257 Single-Exposure Health Effects, 269 Effects of Multiple Exposures, 281 References, 292

10 Ambient Air Particulate Matter 10.1 10.2 10.3 10.4

241

Sources and Pathways for Human Exposure, 318 Ambient Air PM Concentrations, 323 Extent of Population Exposures to Ambient Air PM, 326 Nature of the Evidence for Human Health Effects of Ambient Air PM, 328 10.5 Epidemiological Evidence for Human Health Effects of Ambient Air PM, 329 10.6 Discussion and Current Knowledge on the Health Effects of PM, 354

317

viii

CONTENTS

10.7 Standards and Exposure Guidelines, 356 References, 359 11 Arsenic

367

11.1 Introduction, 367 11.2 Physical and Chemical Properties of Environmental as and Its Compounds, 368 11.3 Environmental Exposures to the General Population: Sources and Standards, 371 11.4 Pathways and Kinetics for in vivo Uptake, Distribution, and Elimination, 374 11.5 As Essentiality, 375 11.6 Health Effects and Exposure–Response Relationships, 376 11.7 Biomarkers of Exposure, Susceptibility, and Effect, 380 11.8 Mitigating Effects and Controlling Exposures, 381 References, 383 12 Asbestos and Other Mineral and Vitreous Fibers

395

12.1 12.2 12.3 12.4 12.5 12.6 12.7

Important Special Properties of Fibers, 395 Exposures to Fibers, 399 Fiber Deposition in the Respiratory Tract, 402 Fiber Retention, Translocation, Disintegration, and Dissolution, 404 Properties of Fibers Relevant to Disease, 413 Fiber-Related Diseases/Processes, 413 Review of Biological Effects of Size-Classified Fibers in Animals and Humans, 415 12.8 Critical Fiber Parameters Affecting Disease Pathogenesis, 420 12.9 Exposure–Response Relationships for Asbestos-Related Lung Cancer and Mesothelioma: Human Experience, 429 12.10 Risk Assessment Issues, 438 12.11 Key Factors Affecting Fiber Dosimetry and Toxicity: Recapitulation and Synthesis, 443 Acknowledgments, 446 Acronyms, 446 References, 446 13 Benzene 13.1 13.2 13.3 13.4 13.5

Benzene Exposure, 460 Uptake, 462 Metabolism and Disposition, 462 Mechanisms of Toxicity, 471 Risk Assessment, 482 References, 486

14 Carbon Monoxide 14.1 14.2

459

Introduction, 499 CO Exposure and Dosimetry, 500

499

CONTENTS

14.3 14.4 14.5 14.6 14.7

Mechanisms of CO Toxicity, 501 Populations at Risk of Health Effects Due to CO Exposure, 502 Regulatory Background, 503 Health Effects of CO, 505 Summary and Conclusions, 515 Acknowledgments, 517 References, 517

15 Chromium 15.1 15.2 15.3 15.4 15.5

18.3 18.4 18.5 18.6 18.7

633

Introduction, 633 Sources, 634 Toxicological Effects and Mechanisms of Action, 640 Mechanisms of Action, 643 References, 651

18 Endocrine Active Chemicals: Broadening the Scope 18.1 18.2

551

Historical Overview, 551 Composition of Diesel Exhaust, 553 Exposures to Diesel Exhaust, 559 Health Effects, 561 Current Issues, 609 Acknowledgments, 613 References, 613

17 Dioxins and Dioxin-Like Chemicals 17.1 17.2 17.3 17.4

529

Introduction, 529 Essentiality, 529 Environmental Exposures, 530 Toxicological Effects, 535 Exposure Guidelines and Standards, 543 References, 544

16 Diesel Exhaust 16.1 16.2 16.3 16.4 16.5

ix

Introduction, 661 Biomarkers: Terminology from Various Disciplines, 664 End Points and Clinical Signs Associated with Endocrine Activity, 666 Environmental Chemicals and End Points: Case Examples, 675 Developmental Origins of Health and Disease, 681 Transgenerational Effects, 684 Conclusion, 686 References, 687

661

x

CONTENTS

19 Secondhand Smoke 19.1 19.2 19.3 19.4 19.5 19.6

703

Exposure to Secondhand Smoke, 705 Health Effects of Involuntary Smoking in Children, 711 Health Effects of Involuntary Smoking in Adults, 722 SHS and Coronary Heart Disease, 730 Respiratory Symptoms and Illnesses in Adults, 734 Summary, 740 References, 741

20 Lead and Compounds

757

20.1 20.2

Introduction, 757 Physical/Chemical Properties and Behavior of Lead and Its Compounds, 758 20.3 Lead in the Environment and Human Exposure, 761 20.4 Lead Absorption, 766 20.5 Distribution, 771 20.6 Kinetics, 774 20.7 Biomarkers, 781 20.8 Health Effects, 785 20.9 Mechanisms Underlying Lead Toxicity, 792 20.10 Treatment of Lead Toxicity, 796 References, 798 21 Mercury 21.1 21.2 21.3 21.4 21.5 21.6 21.7 21.8

Introduction, 811 Chemistry, 811 Sources, 812 Environmental Exposures, 813 Occupational Exposures, 815 Kinetics and Metabolism, 816 Health Effects, 818 Prevention, 820 References, 821

22 Nitrogen Oxides 22.1 22.2 22.3 22.4 22.5 22.6

823

Introduction, 823 Sources, 823 Nitrogen Dioxide, 824 Nitric Oxide, 845 Nitric/Nitrous Acid, 848 Inorganic Nitrates, 849 References, 851

23 Ozone 23.1 23.2

811

869 Introduction, 869 Background on Exposures and Health-Related Effects, 873

CONTENTS

xi

23.3 23.4 23.5 23.6 23.7

Effects of Short-Term Exposures to Ozone in Humans, 877 Factors Affecting the Variability of Responsiveness in Humans, 890 Studies of Populations Exposed to Ozone in Ambient Air, 892 Effects Observed in Studies in Laboratory Animals, 900 Determinants of Responsiveness to Ozone Exposures in Animal Studies, 901 23.8 Effects of Multiple Day and Ambient Episode Exposures, 908 23.9 Chronic Effects of Ambient Ozone Exposures, 910 23.10 Ambient Air Quality Standards and Guidelines, 917 23.11 Summary and Conclusions, 920 Acknowledgment, 922 References, 922 24 Pesticides 24.1 24.2 24.3 24.4 24.5 24.6 24.7 24.8 24.9

Evolving Patterns of Pesticide Use, 938 Export of Hazardous Pesticides, 939 Exposure to Pesticides, 939 Epidemiology of Acute Pesticide Poisoning, 942 Toxicity of Pesticides, 943 Pesticides and Endocrine/Reproductive Toxicity, 949 Pesticides and Childhood Cancer, 950 Legislative Framework, 950 Conclusion: Issues for the Future, 952 References, 953

25 Sulfur Oxides—SO2, H2SO4, NH4HSO4, and (NH4)2SO4 25.1 25.2 25.3

1001

Background, 1003 Philosophical Approaches, 1004 Standards Development, 1005 Current Developments, 1010 Protective Measures, 1012 Conclusions, 1014 Glossary, 1015 References, 1016

27 Sources, Levels and Effects of Manmade Ionizing Radiation and Radioactivity 27.1 27.2

957

Sources and Exposures, 957 Health Effects, 961 Ambient Air Quality Standard and Guidelines, 989 Acknowledgments, 991 References, 991

26 Microwaves and Electromagnetic Fields 26.1 26.2 26.3 26.4 26.5 26.6 26.7

937

Source Documents, 1021 Special Units, 1022

1021

xii

CONTENTS

27.3 27.4 27.5 27.6 27.7 27.8 27.9 27.10 27.11

Sources of Manmade Radioactivity and Radiation, 1024 Nuclear Fuel Cycle, 1025 Discussion of Radiation Doses from the Nuclear Fuel Cycle, 1038 Nuclear Weapons Complex, 1043 Local, Tropospheric, and Global Fallout, 1048 Medical Exposures, 1050 Industrial Uses (Other than the Nuclear Fuel Cycle), 1054 Consumer Products, 1055 Overview of Potential Health Impacts of Natural and Manmade Sources of Radioactivity, 1057 References, 1066

28 Noise: Its Effects and Control 28.1 28.2 28.3 28.4 28.5 28.6 28.7 28.8 28.9 28.10 28.11 28.12

Definitions of Sound and Noise, 1071 Noise Exposure is Widespread and Annoying, 1072 Effects of Loud Sounds and Noise on Hearing, 1075 Noise as a Stressor, 1076 Noise and Sleep Interference, 1077 Noise and Mental Health, 1077 Noise Affects Children’s Cognitive, Language and Learning Skills, 1078 Impacts of Low-Frequency Noise, 1079 Civility, Responsibility, and Noise, 1079 Controlling Noise, 1080 Education and Public Awareness, 1084 Summary, 1084 References, 1085

29 Radon and Lung Cancer 29.1 29.2 29.3 29.4 29.5 29.6 29.7 29.8 29.9 29.10 29.11 29.12

1089

Radon and Lung Cancer, 1089 Outdoor Radon, 1093 Indoor Radon, 1097 The Other Radon, 220Rn, Thoron, 1100 Radon Epidemiology in Underground Mines, 1100 Residential Epidemiology, 1102 Lung Dosimetry, 1104 Lung Cancer Models for Humans, 1107 Childhood Exposure, 1113 Animal Studies, 1114 Smoking and Radon, 1114 Summary, 1115 References, 1116

30 Ultraviolet Radiation 30.1 30.2 30.3

1071

Introduction, 1121 Pathways for Human Exposure, 1122 Sources of Ultraviolet Radiation, 1124

1121

CONTENTS

30.4 30.5 30.6 30.7 30.8 30.9 30.10 30.11 30.12 30.13 30.14

INDEX

xiii

Biological Mechanisms Leading to Health Effects, 1135 Ocular Effects, 1135 Nonmalignant Skin Effects, 1138 Skin Cancer, 1140 Malignant Melanoma, 1142 Immune System Effects, 1146 Populations at Special Risk: Ocular Damage, 1147 Populations at Special Risks: Skin Effects, 1148 Applicable Standards and Exposure Guidelines, 1150 Techniques for Evaluating Actual or Potential Exposures, 1152 Summary, 1156 References, 1157 1163

PREFACE

This is the third edition of Environmental Toxicants: Human Exposures and Their Health Effects. It provides updated versions of chapters that appeared in the first (1992) and second (2000) editions, and it broadens the coverage to include two new toxicant categories (one is arsenic and its compounds, and the other is endocrine disrupting chemicals). As before, it is focused on providing current knowledge on environmental health challenges to people in our homes and communities resulting from exposures to chemical and physical agents that they encounter in the course of their daily lives. This book remains unique in terms of its depth of coverage on a limited number of environmental agents that are known to have, or are highly likely to have, adverse health effects following exposures that are within the ranges that occur in contemporary populations in economically developed countries. Extrapolation of likely effects in developing countries, where toxicant exposures may be substantially higher, need to be made with caution, since susceptibility to adverse effects may differ as a result of differences in diet, pre-existing diseases, thermal stresses, and access to modern health care. Chapter 1 has been expanded to include discussions of study options for increasing our knowledge of biological responses to environmental toxicant exposures, as well as of new and developing methods for the elucidation of responses at the molecular level. I gratefully acknowledge the outstanding contributions of the other chapter authors who are my colleagues and peers. They are all outstanding and widely recognized professionals with many demands on their time, and this unique book would not have been possible without their generous commitment. Periodic revisions of the content of the chapters herein are necessary because of our everincreasing knowledge base, which has been facilitated by the development of new and improved measurement and modeling, and data management technologies. These technologies, and the growth of interdisciplinary investigations of complex phenomena, have enabled investigatory teams to go beyond the identification of statistically significant associations between environmental exposures and health-related responses in human populations, laboratory animal cohorts, and cell cultures in vitro, to the underlying biological pathways and mechanisms that are applicable to realistic exposure levels. While xv

xvi

PREFACE

notable progress has been made in environmental health sciences in recent years, significant challenges remain, not the least of which is access to research funding from government and private sources at a time when our collective capacities are increasing for characterizing (1) exposures and their geographic and temporal distributions; (2) biological mechanisms responsible for the adverse effects produced by environmental exposures; (3) susceptibility factors that account for the generally large interindividual variability in responses to exposure; and (4) exposure–response relationships for sensitive population segments. Another challenge is that the populations of both the general public and the environmental health research community are aging. Older people are clearly a susceptible population to many environmental toxicants, and the research needed to identify means of recognizing, evaluating, and controlling exposures to these toxicants will require both additional research funding and recruitment and training of young investigators who can carry out such research over at least several decades into the future. Some of these new trainees may well be the authors of chapters in future editions of this reference volume. I hereby recognize the contributions of those who, in addition to writing chapters, made substantial contributions to the completion of this edition. In particular, I want to recognize Toni Moore, Anita Parkhurst, and Angela Muniz for their diligent and effective management of the text preparation and presentation, and Gordon Cook for the preparation of many of the figures. Finally, my own contributions would not have been possible without the cooperation and patience of my wife, Janet. New York University School of Medicine

MORTON LIPPMANN

CONTRIBUTORS

Donald R. Bergfelt, Division of Exposure Assessment, Office of Science Coordination and Policy, Office of Prevention, Pesticides and Toxic Substances, United States Environmental Protection Agency, 1200 Pennsylvania Avenue, NW, Washington, DC 20460, USA Arline L. Bronzaft, 505 East 79th Street, 8B, New York, NY 10021, USA Richard J. Bull, MoBull Consulting, 1928 Meadows Drive North, Richland, WA 99352, USA James S. Bus, Toxicology Research Laboratory, Dow Chemical Co., 1803 Building, Midland, MI 48674, USA Luz Claudio, Department of Community and Preventive Medicine, Mount Sinai School of Medicine, New York, NY 10029, USA Mitchell D. Cohen, Nelson Institute of Environmental Medicine, New York University School of Medicine, 57 Old Forge Road, Tuxedo, NY 10987, USA Norman Cohen, New York University (Retired) Francis Colville, Radiofrequency Program, U.S. Army Center for Health Promotion and Preventive Medicine, Aberdeen Proving Ground, MD, 21010-5403, USA Nigel Cridland, National Radiological Protection Board, Chilton, Didcot, Oxon OX11 0RQ, UK Colin Driscoll, National Radiological Protection Board (Retired) Michael A. Gallo, Environmental & Occupational Health Science Institute, 681 Frelinghuysen Road, Piscataway, NJ 08855-1179, USA Eric Garshick, Pulmonary and Critical Care Medicine Section, VA Boston Healthcare System; Channing Laboratory, Brigham and Women’s Hospital; and Harvard Medical School, Boston, MA, USA xvii

xviii

CONTRIBUTORS

Bernard D. Goldstein, Graduate School of Public Health, University of Pittsburgh, Pittsburgh, PA, 15261, USA Philippe Grandjean, Department of Environmental Medicine, Odense University, Winslowparken 17, DK-5000 Odense, Denmark Lester D. Grant, 517 Colony Woods Drive, Chapel Hill, NC, USA K. Christiana Grim, Division of Exposure Assessment, Office of Science Coordination and Policy, Office of Prevention, Pesticides and Toxic Substances, United States Environmental Protection Agency, 1200 Pennsylvania Avenue, NW, Washington, DC 20460, USA Naomi H. Harley, Department of Environmental Medicine, New York University Medical School, 550 First Avenue, New York, NY 10016, USA Fred D. Hoerger, Dow Chemical (Retired) Michael T. Kleinman, Department of Community and Environmental Medicine, University of California at Irvine, Irvine, CA 92697-1825, USA Philip J. Landrigan, Department of Community Medicine, Mount Sinai Medical Center, Box 1057, New York, NY 10029-6574, USA George D. Leikauf, Department of Environmental and Occupational Health, Graduate School of Public Health, Bridgestone Point Bldg, Suite 359, 100 Technology Drive, Pittsburgh, PA 15219, USA Morton Lippmann, New York University School of Medicine, 21 Old Forge Lane, Tarrytown, NY 10591, USA Raymond C. Loehr, 19360 Magnolia Grove Square No. 405, Lansdowne, VA 20176, USA Kathryn R. Mahaffey, Office of Science Coordination and Policy, Office of Prevention, Pesticides and Toxic Substances, United States Environmental Protection Agency, 1200 Pennsylvania Avenue, NW, Washington, DC 20460, USA Joe L. Mauderly, Inhalation Toxicology Research Institute, P.O. Box 5890, Albuquerque, NM 87185, USA John J. Mauro, 209 Ueland Road, Red Bank, NJ 07701, USA Jessica C. Meiller, Division of Exposure Assessment, Office of Science Coordination and Policy, Office of Prevention, Pesticides and Toxic Substances, United States Environmental Protection Agency, 1200 Pennsylvania Avenue, NW, Washington, DC 20460, USA Lars Mølhave, Institute of Environmental and Occupational Medicine, University of Aarhus, DK-8000 Aarhus C, Denmark Gila I. Neta, Department of Epidemiology, Bloomberg School of Public Health, Johns Hopkins University, Baltimore, MD, USA Jesper B. Nielsen, Institute of Public Health, University of Southern Denmark, Winslowparken 17, DK-5000 Odense, Denmark

CONTRIBUTORS

xix

Larry W. Rampy, Dow Chemical (Retired) Douglas A. Rausch, Dow Chemical (Retired) Joseph V. Rodricks, The Life Sciences Consultancy LLC, 750 17th Street, NW, Suite 1000, Washington, DC 20006, USA Toby G. Rossman, Nelson Institute of Environmental Medicine, New York University School of Medicine, 57 Old Forge Road, Tuxedo, NY 10987, USA Jonathan M. Samet, Department of Epidemiology, The Johns Hopkins University, Suite 6039, 615 N. Wolfe Street, Baltimore, MD 21205-2179, USA Richard B. Schlesinger, Dyson College of Arts & Sciences, Pace University, 861 Bedford Road, Pleasantville, NY 10570, USA David H. Sliney, 406 Streamside Drive, Fallston, MD 21047-2806, USA Shirlee W. Tan, The Smithsonian Institution, National Zoological Park, 3000 Connecticut Avenue, NW, Washington, DC, USA Arthur C. Upton, 250 East Alameda, Apartment 636, Santa Fe, NM 87501, USA Mark J. Utell, Pulmonary Unit, University of Rochester Medical Center, Box 692, Rochester, NY 14642-8692, USA Sophia S. Wang, Division of Cancer Epidemiology and Genetics, National Cancer Institute, Washington, DC, USA Gisela Witz, Department of Environmental and Occupational Medicine, Robert Wood Johnson Medical School, Piscataway, NJ, USA

1 INTRODUCTION AND BACKGROUND Morton Lippmann and George D. Leikauf

This book identifies and critically reviews current knowledge on human exposure to selected chemical agents and physical factors in the ambient environment and the effects of such exposures on human health. It provides a state-of-the-art knowledge base essential for risk assessment for exposed individuals and populations to guide public health authorities, primary care physicians, and industrial managers having to deal with the consequences of environmental exposure. Aside from professionals in public health, medicine, and industry who may use this book to guide their management functions, the volume can also be used in graduate and postdoctoral training programs in universities and by toxicologists, clinicians, and epidemiologists in research as a resource for the preparation of research proposals and scientific papers. The subject is environmental toxicants, that is, chemical or physical agents released into the general environment that can produce adverse health effects among large numbers of people. Such effects are usually subclinical, except when cumulative changes lead to chronic effects after long exposure. Short-term responses following acute exposures are often manifest as transient alterations in physiological function that may, in some sensitive members of the population, be of sufficient magnitude to be considered adverse. Each of the specific topic chapters has a thorough discussion of the extent of human exposure as well as of toxic responses. The four chapters on the uses of the data for risk assessment, risk management, clinical applications, and industrial operations provide guidance for those performing individual and/or collective population hazard evaluations. The first provides individuals and public agency personnel with a basis for decisions on risk avoidance and relative risk assessment. The second outlines the operational philosophies and techniques used by environmental engineers in scoping and managing environmental risks. The third enables the primary care physician to recognize diseases and symptoms associated with exposures to environmental toxicants and to provide counsel to patients. The fourth assists decision makers in industry in evaluating the potential impacts of their plant operations and products on public health.

Environmental Toxicants, Third Edition Edited by Morton Lippmann Copyright
2009 John Wiley & Sons, Inc.

1

2

INTRODUCTION AND BACKGROUND

Although many books provide brief reviews of hundreds of chemicals encountered in the work environment at levels that can cause demonstrable health effects, both acute and chronic, they contain relatively little information on the effects of low-level exposures on large populations of primary interest in environmental health and risk assessment. This book has been designed to provide in-depth, critical reviews of the environmental toxicants of contemporary public health concern.

1.1 CHARACTERIZATION OF CHEMICAL CONTAMINANTS 1.1.1

Concentration Units

In environmental science, confusion often arises from the use of the same or similar sounding terms having different meanings in different contexts. This is especially true in describing the concentrations of air and water contaminants. Solutes are frequently expressed in parts per million (ppm) or parts per billion (ppb). However, when used for air contaminants, the units are molar or volume fractions, whereas when used for water contaminants, they are weight fractions. This problem can be avoided by expressing all fluid contaminant concentrations as the weight of contaminant per unit volume (e.g., m3 or L) of fluid. In air, the units generally used are mg/m3 or mg/m3, whereas in water they are most often mg/L or mg/L. 1.1.2

Air Contaminants

Chemical contaminants can be dispersed in air at normal ambient temperatures and pressures in gaseous, liquid, and solid forms. The latter two represent suspensions of particles in air and were given the generic term “aerosols” by Gibbs (1924) on the basis of analogy to the term “hydrosol,” used to describe disperse systems in water. On the contrary, gases and vapors, which are present as discrete molecules, form true solutions in air. Particles consisting of moderate- to high-vapor-pressure materials tend to evaporate rapidly, since those small enough to remain suspended in air for more than a few minutes (i.e., those smaller than about 10 mm) have large surface-to-volume ratios. Some materials with relatively low vapor pressures can have appreciable fractions in both the vapor and aerosol forms simultaneously. 1.1.2.1 Gases and Vapors Once dispersed in air, contaminant gases and vapors generally form mixtures so dilute that their physical properties, such as density, viscosity, enthalpy, and so on are indistinguishable from those of clean air. Such mixtures may be considered to follow ideal gas law relationships. There is no practical difference between a gas and a vapor except that the latter is generally considered to be the gaseous phase of a substance that is normally a solid or liquid at room temperature. While dispersed in the air, all molecules of a given compound are essentially equivalent in their size and probabilities of contact with ambient surfaces, respiratory tract surfaces, and contaminant collectors or samplers. 1.1.2.2 Aerosols Aerosols, being dispersions of solid or liquid particles in air, have the very significant additional variable of particle size. Size affects particle motion and, hence, the probabilities for physical phenomena such as coagulation, dispersion, sedimentation, impaction onto surfaces, interfacial phenomena, and light-scattering properties. It is not possible to fully characterize a given particle by a single size parameter. For example, a

CHARACTERIZATION OF CHEMICAL CONTAMINANTS

(a)

(b)

(c)

20

Frequency

Mean

Particle size (µm)

Median

Frequency

40 84.1% Value

Mode

0

3

20 10 Particle size (µm)

1

10 100 Particle size (µm)

10

50% Value

5

1

1

10 50 90 99 Cumulative (%)

FIGURE 1.1 Particle size distribution data. (a) Plotted on linear coordinates. (b) Plotted on a logarithmic size scale. (c) In practice, logarithmic probability coordinates are used to display the percentage of particles less than a specific size versus that size. The geometric standard deviation (sg) of the distribution is equal to the 84.1% size/50% size.

particle’s aerodynamic properties depend on density and shape as well as linear dimensions, and the effective size for light scattering is dependent on refractive index and shape. In some special cases, all of the particles are essentially the same in size. Such aerosols are considered to be monodisperse. Examples are natural pollens and some laboratorygenerated aerosols. More typically, aerosols are composed of particles of many different sizes and hence are called heterodisperse or polydisperse. Different aerosols have different degrees of size dispersion. It is, therefore, necessary to specify at least two parameters in characterizing aerosol size: a measure of central tendency, such as a mean or median, and a measure of dispersion, such as an arithmetic or geometric standard deviation. Particles generated by a single source or process generally have diameters following a lognormal distribution; that is, the logarithms of their individual diameters have a Gaussian distribution. In this case, the measure of dispersion is the geometric standard deviation, which is the ratio of the 84.16 percentile size to the 50th percentile size (Fig. 1.1). When more than one source of particles is significant, the resulting mixed aerosol will usually not follow a single lognormal distribution, and it may be necessary to describe it by the sum of several distributions. 1.1.3

Particle Characteristics

There are many properties of particles, other than their linear size, that can greatly influence their airborne behavior and their effects on the environment and health. These include Surface: For spherical particles, the surface varies as the square of the diameter. However, for an aerosol of given mass concentration, the total aerosol surface increases with decreasing particle size. Airborne particles have much greater ratios of external surface to volume than do bulk materials and, therefore, the particles can dissolve or participate in surface reactions to a much greater extent than would massive samples of the same materials. Furthermore, for nonspherical solid particles

4

INTRODUCTION AND BACKGROUND

or aggregate particles, the ratio of surface to volume is increased, and for particles with internal cracks or pores, the internal surface area can be much greater than the external area. Volume: Particle volume varies as the cube of diameter; therefore, the few largest particles in an aerosol tend to dominate its volume concentration. Shape: A particle’s shape affects its aerodynamic drag as well as its surface area and therefore its motion and deposition probabilities. Density: A particle’s velocity in response to gravitational or inertial forces increases as the square root of its density. Aerodynamic diameter: The diameter of a unit-density sphere having the same terminal settling velocity as the particle under consideration is equal to its aerodynamic diameter. Terminal settling velocity is the equilibrium velocity of a particle that is falling under the influence of gravity and fluid resistance. Aerodynamic diameter is determined by the actual particle size, the particle density, and an aerodynamic shape factor. 1.1.3.1 Types of Aerosols Aerosols are generally classified in terms of their processes of formation. Although the following classification is neither precise nor comprehensive, it is commonly used and accepted in the industrial hygiene and air pollution fields: Dust: An aerosol formed by mechanical subdivision of bulk material into airborne fines having the same chemical composition. A general term for the process of mechanical subdivision is comminution, and it occurs in operations such as abrasion, crushing, grinding, drilling, and blasting. Dust particles are generally solid and irregular in shape and have diameters greater than 1 mm. Fume: An aerosol of solid particles formed by condensation of vapors formed at elevated temperatures by combustion or sublimation. The primary particles are generally very small (less than 0.1 mm) and have spherical or characteristic crystalline shapes. They may be chemically identical to the parent material, or they may be composed of an oxidation product such as a metal oxide. Since they may be formed in high number concentration, they often rapidly coagulate, forming aggregate clusters of low overall density. Smoke: An aerosol formed by condensation of combustion products, generally of organic materials. The particles are generally liquid droplets with diameters of less than 0.5 mm. Mist: A droplet aerosol formed by mechanical shearing of a bulk liquid, for example, by atomization, nebulization, bubbling, or spraying. The initial droplet size can cover a very large range, usually from about 2 mm to greater than 50 mm. Fog: An aqueous aerosol formed by condensation of water vapor on atmospheric nuclei at high relative humidities. The droplet sizes are generally greater than 1 mm. Smog: A popular term for a pollution aerosol derived from a combination of smoke and fog. It is now commonly used for any atmospheric pollution mixture. Haze: A submicrometer-sized aerosol of hygroscopic particles that take up water vapor at relatively low relative humidities. Aitken or condensation nuclei (CN): Very small atmospheric particles (mostly smaller than 0.1 mm) formed by combustion processes and by chemical conversion from gaseous precursors.

CHARACTERIZATION OF CHEMICAL CONTAMINANTS

5

Accumulation mode: A term given to the particles in the ambient atmosphere ranging from 0.1 to about 1.0 mm, and extending up to 2.5 mm for hygroscopic particles in humid atmospheres. These particles generally are spherical, have liquid surfaces, and form by coagulation and condensation of smaller particles that derive from gaseous precursors. Being too few for rapid coagulation, and too small for effective sedimentation, they tend to accumulate in the ambient air. Coarse particle mode: Ambient air particles larger than about 2.5 mm and generally formed by mechanical processes and surface dust resuspension. 1.1.3.2 Aerosol Characteristics Aerosols have integral properties that depend upon the concentration and size distribution of the particles. In mathematical terms, these properties can be expressed in terms of certain constants or “moments” of the size distribution (Friedlander, 1977). Some integral properties such as light-scattering ability or electrical charge depend on other particle parameters as well. Some of the important integral properties are: Number concentration: The total number of airborne particles per unit volume of air, without distinction as to their sizes, is the zeroth moment of the size distribution. In current practice, instruments are available that count the numbers of particles of all sizes from about 0.005 to 50 mm. In many specific applications, such as fiber counting for airborne asbestos, a more restricted size range is specified. Surface concentration: The total external surface area of all the particles in the aerosol, which is the second moment of the size distribution, may be of interest when surface catalysis or gas adsorption processes are of concern. Aerosol surface is one factor affecting light-scatter and atmospheric-visibility reductions. Volume concentration: The total volume of all the particles, which is the third moment of the size distribution, is of little intrinsic interest in itself. However, it is closely related to the mass concentration, which for many environmental effects is the primary parameter of interest. Mass concentration: The total mass of all the particles in the aerosol is frequently of interest. The mass of a particle is the product of its volume and density. If all of the particles have the same density, the total mass concentration is simply the volume concentration times the density. In some cases, such as “respirable,” “thoracic,” and “inhalable” dust sampling (Vincent, 1999), the parameter of interest is the mass concentration over a restricted range of particle size. In these applications, particles outside the size range of interest are excluded from the integral. Dustfall: The mass of particles depositing from an aerosol onto a unit surface per unit time is proportional to the fifth moment of the size distribution. Dustfall has long been of interest in air pollution control because it provides an indication of the soiling properties of the aerosol. Light scatter: The ability of airborne particles to scatter light and cause a visibility reduction is well known. Total light scatter can be determined by integrating the aerosol surface distribution with the appropriate scattering coefficients. 1.1.4

Water Contaminants

Chemical contaminants can be found in water, in solution, or as hydrosols; the latter are immiscible solid or liquid particles in suspension. An aqueous suspension in liquid particles

6

INTRODUCTION AND BACKGROUND

is generally called an emulsion. Many materials with relatively low aqueous solubility will be found in both dissolved and suspended forms. 1.1.4.1 Dissolved Contaminants Water is known as the universal solvent. Although there are many compounds that are not completely soluble in water, there are a few that do not have some measurable solubility. In fact, the number of chemical contaminants in natural waters is primarily a function of the sensitivity of the analyses. For organic compounds in rivers and lakes, it has been observed that as the limits of detection decrease by an order of magnitude, the numbers of compounds detected increase by an order of magnitude, so that one might expect to find at least 10–12 g/L (approximately 1010 molecules/L) of each of the million organic compounds reported in the literature (NIEHS, 1977). Similar considerations undoubtedly apply to inorganic chemicals as well. 1.1.4.2 Dissolved Solids Water-quality criteria generally include a nonspecific parameter called “dissolved solids.” However, it is customary to exclude natural mineral salts such as sodium chloride from this classification. Also, water criteria for specific toxic chemicals dissolved in water are frequently exceeded without there being an excessive total dissolved-solids content. 1.1.4.3 Dissolved Gases Compounds dissolved in water may also exist in the gaseous phase at normal temperatures and pressures. Some of these, such as hydrogen sulfide (HS2), and ammonia (NH3), which are generated by decay processes, are toxicants. Oxygen (O2) is the most critical of the dissolved gases with respect to water quality. It is essential to most higher aquatic life forms and is needed for the oxidation of most of the organic chemical contaminants to more innocuous forms. Thus, a critical parameter of water quality is the concentration of dissolved oxygen (DO). Another important parameter is the extent of the oxygen “demand” associated with contaminants in the water. The most commonly used index of oxygen demand is the 5-day biochemical oxygen demand (BOD after 5 days of incubation). Another is the chemical oxygen demand (COD). 1.1.4.4 Suspended Particles A nonspecific water-quality parameter that is widely used is “suspended solids.” The stability of aqueous suspensions depends on particle size, density, and charge distributions. The fate of suspended particles depends on a number of factors, and particles can dissolve, grow, coagulate, or be ingested by various life forms in the water. They can become “floating solids” or part of an oil film, or they can fall to the bottom to become part of the sediments. There are many kinds of suspended particles in natural waters, and not all of them are contaminants. Any moving water will have currents that cause bottom sediments to become resuspended. Also, natural runoff will carry soil and organic debris into lakes and streams. In any industrialized area, such sediment and surface debris will always contain some chemicals considered to be contaminants. However, a large proportion of the mass of such suspended solids would usually be “natural,” and would not be considered as contaminants. The suspended particles can have densities that are less than, equal to, or greater than that of the water, so that the particles can rise as well as fall. Furthermore, the effective density of particles can be reduced by the attachment of gas bubbles. Gas bubbles form in water when the water becomes saturated and cannot hold any more of the gas in solution. The solubility of gases in water varies inversely with temperature. For example, oxygen saturation of fresh water is 14.2 ml/L at 0 C and 7.5 mg/L at 30 C, and in seawater the corresponding values are 11.2 and 6.1 mg/L.

HUMAN EXPOSURES AND DOSIMETRY

1.1.5

7

Food Contaminants

Chemical contaminants of almost every conceivable kind can be found in most types of human food. Food can acquire these contaminants at any of several stages in its production, harvesting, processing, packaging, transportation, storage, cooking, and serving. In addition, there are many naturally occurring toxicants in foods as well as compounds that can become toxicants upon conversion by chemical reactions with other constituents or additives or by thermal or microbiological conversion reactions during processing, storage, or handling. Each food product has its own natural history. Most foods are formed by selective metabolic processes of plants and animals. In forming tissue, these processes can act either to enrich or to discriminate against specific toxicants in the environment. For animal products, where the flesh of interest in foods was derived from the consumption of other life forms, there are likely to be several stages of biological discrimination and, therefore, large differences between contaminant concentrations in the ambient air and/or water and the concentrations within the animals.

1.2 HUMAN EXPOSURES AND DOSIMETRY People can be exposed to chemicals in the environment in numerous ways. The chemicals can be inhaled, ingested, or taken up by and through the skin. Effects of concern can take place at the initial epithelial barrier, that is, the respiratory tract, the gastrointestinal (GI) tract, or the skin, or can occur in other organ systems after penetration and translocation by diffusion or transport by blood, lymph, and so on. As illustrated in Fig. 1.2, exposure and dose factors are intermediate steps in a larger continuum ranging from release of chemicals into an environmental medium to an ultimate health effect. Exposure is a key step in this continuum and a complex one. The concept of total human exposure has developed in recent years as essential to the appreciation of the nature and extent of environmental health hazards associated with ubiquitous chemicals at low levels.

FIGURE 1.2

Environmental and biological modifiers of human exposure and health responses.

8

INTRODUCTION AND BACKGROUND

It provides a framework for considering and evaluating the contribution to the total insult from dermal uptake, ingestion of food and drinking water, and inhaled doses from potentially important microenvironments such as workplace, home, transportation, recreational sites, and so on. More thorough discussions of this key concept have been prepared by Sexton and Ryan (1988), Lioy (1990), and the National Research Council (NRC, 1991). Guidelines for Exposure Assessment have been formalized by the U.S. Environmental Protection Agency (U.S. EPA, 1992). 1.3 CHEMICAL EXPOSURES AND DOSE TO TARGET TISSUES Toxic chemicals in the environment that reach sensitive tissues in the human body can cause discomfort, loss of function, and changes in structure leading to disease. This section addresses the pathways and transport rates of chemicals from environmental media to critical tissue sites as well as retention times at those sites. It is designed to provide a conceptual framework as well as brief discussions of: (1) the mechanisms for—and some quantitative data on—uptake from the environment; (2) translocation within the body, retention at target sites, and the influence of the physicochemical properties of the chemicals on these factors; (3) the patterns and pathways for exposure of humans to chemicals in environmental media; (4) the effects of chemicals at the cellular and organ levels; and (5) the influence of age, sex, size, habits, health status, and so on. An agreed on terminology is critically important when discussing the relationships between toxic chemicals in the environment and human health. The terms used in this book are defined below: Exposure: Contact with external environmental media containing the chemical of interest. For fluid media in contact with the skin or respiratory tract, both concentration and contact time are critical. For ingested material, concentration and amount consumed are important. Deposition: Capture of the chemical at a body surface site on skin, respiratory tract, or GI tract. Clearance: Translocation from a deposition site to a storage site or depot within the body, or elimination from the body. Retention: Presence of residual material at a deposition site or along a clearance pathway. Dose: Amount of chemical deposited on or translocated to a site on or within the body where toxic effects take place. Target tissue: A site within the body where toxic effects lead to damage or disease. Depending on the toxic effects of concern, a target tissue can extend from whole organs down to specific cells to sub-cellular constituents. Exposure surrogates or indices: Indirect measures of exposure, such as: (1) concentra tions in environmental media at times or places other than those directly encountered; (2) concentrations of the chemical of interest, a metabolite of the chemical, or an enzy me induced by the chemical in circulating or excreted body fluids; and (3) elevations in body burden as measured by external probes. In summary, exposure represents contact between a concentration of an agent in air, water, food, or other material and the person or population of interest. The agent is the source

CONCENTRATION OF TOXIC CHEMICALS IN HUMAN MICROENVIRONMENTS

9

of an internal dose to a critical cell, organ, or tissue. The magnitude of the dose depends on a number of factors: (1) the volumes inhaled or ingested; (2) the fractions of the inhaled or ingested material transferred across epithelial membranes of the skin, the respiratory tract, and the GI tract; (3) the fractions transported via circulating fluids to target tissues; and (4) the fractional uptake by the target tissues. Each of these factors can have considerable intersubject variability. Sources of variability include activity level, age, sex, and health status as well as such inherent variabilities as race and size. With chronic or repetitive exposures, other factors affect the dose of interest. When the retention at, or effects on, the target tissues are cumulative and clearance or recovery is slow, the dose of interest can be represented by cumulative uptake. However, when the agent is rapidly eliminated, or when its effects are rapidly and completely reversible on removal from exposure, rate of delivery may be the dose parameter of primary interest.

1.4 CONCENTRATION OF TOXIC CHEMICALS IN HUMAN MICROENVIRONMENTS The technology for sampling air, water, and food is relatively well developed, as are the technologies for sample separation from copollutants, media, and interferences and for quantitative analyses of the components of interest. However, knowing when, where, how long, and at which rate and frequency to sample to collect data relevant to the exposures of interest is difficult, and requires knowledge of temporal and spatial variability of exposure concentrations. Unfortunately, we seldom have enough information of these kinds to guide our sample collections. Many of these factors are discussed in detail in the chapters that follow as they apply to the specific environmental toxicants being discussed. 1.4.1

Water and Foods

Concentrations of environmental chemicals in food and drinking water are extremely variable, and there are further variations in the amounts consumed because of the extreme variability in dietary preferences and food sources. The number of foods for which up-to-date concentration data for specific chemicals are available is extremely limited. Relevant human dietary exposure data are sometimes available in terms of market basket survey analyses. In this approach, foods for a mixed diet are purchased, cleaned, processed, and prepared as for consumption, and one set of specific chemical analyses is done for the composite mixture that is consumed. The concentrations of chemicals in potable piped water supplies depend greatly on the source of the water and its treatment history. Surface waters from protected watersheds generally have low concentrations of both dissolved minerals and environmental chemicals. Well waters usually have low concentrations of bacteria and environmental chemicals, but often have high mineral concentrations. Poor waste disposal practices may contribute to ground water contamination, especially in areas of high population density. Treated surface waters from lakes and rivers in densely populated and/or industrialized areas usually contain a wide variety of dissolved organics and trace metals, the concentrations of which vary greatly with season (because of variable surface runoff), with proximity to pollutant sources, with upstream usage, and with treatment efficacy. Uptake of environmental chemicals in bathing waters across intact skin is usually minimal in comparison to uptake via inhalation or ingestion. It depends on both the

10

INTRODUCTION AND BACKGROUND

concentration in the fluid surrounding the skin surface and the polarity of the chemical, with more polar chemicals having less ability to penetrate the intact skin. Uptake via skin can be significant for occupational exposures to concentrated liquids or solids. 1.4.2

Air

Although chemical uptake through ingestion and the skin surface is generally intermittent, inhalation provides a continuous means of exposure. The important variables affecting the uptake of inhaled chemicals are the depth and frequency of inhalation and the concentration and physicochemical properties of the chemicals in the air. Exposures to airborne chemicals vary widely among inhalation microenvironments, the categories of which include workplace, residence, outdoor ambient air, transportation, recreation, and public spaces. There are also wide variations in exposure within each category, depending on the number and strength of the sources of the airborne chemicals, the volume and mixing characteristics of the air within the defined microenvironment, the rate of air exchange with the outdoor air, and the rate of loss to surfaces within the microenvironment. 1.4.3

Workplace

Exposures to airborne chemicals at work are extremely variable in terms of composition and concentration, depending on the materials being handled, the process design and operation, the kinds and degree of engineering controls applied to minimize release to the air, work practices followed, and personal protection provided. Workplace air monitoring often involves breathing zone sampling, generally with passive samplers for gases and vapors or with personal battery-powered extraction samplers for both gases and particles; these operate over periods of 1–8 h. Analyses of the samples collected can provide accurate measures of individual exposures to specific air contaminants. Workplace air monitoring is also frequently done with fixed-site samplers or direct reading instruments. However, air concentrations at fixed sites may differ substantially from those in the breathing zones of individual workers. The fixed-site data may be relatable to the breathing zone when appropriate intercomparisons can be made, but otherwise they represent crude surrogates of exposure. The characteristics of equipment used for air sampling in industry are described in detail in Air Sampling Instruments (ACGIH, 2001). 1.4.4

Residential

Airborne chemicals in residential microenvironments are attributable to their presence in the air infiltrating from out-of-doors and to their release from indoor sources. The latter include unvented cooking stoves and space heaters, cigarettes, consumer products, and volatile emissions from wallboard, textiles, carpets, and so on. Personal exposures to chloroform, largely from indoor residential sources, are illustrated in Fig. 1.3, and the influence of smoking in the home on indoor exposures to respirable particulate matter is illustrated in Fig. 1.4. Indoor sources can release enough nitrogen dioxide (NO2), fine particle mass (FPM), and formaldehyde (HCHO) that indoor concentrations for these chemicals can be much higher than those in ambient outdoor air. Furthermore, their contributions to the total human exposure are usually even greater, since people usually spend much more time at home than in the outdoor ambient air.

CONCENTRATION OF TOXIC CHEMICALS IN HUMAN MICROENVIRONMENTS

11

FIGURE 1.3 Estimated frequency distributions of personal air exposures to chloroform: outdoor air concentrations, and exhaled breath values in Elizabeth-Bayonne, NJ area. Note: Air values are 12-h integrated samples. Breath value was taken following the daytime air sample (6:00 a.m. to 6:00 p.m.). Outdoor air samples were taken near participants’ homes. Source: Wallace et al. (1985).

1.4.5

Outdoor Ambient Air

For pollutants having national ambient air quality standards (NAAQS), there is an extensive network of fixed-site monitors, generally on rooftops. Although these devices generate large volumes of data, the concentrations at these sites may differ substantially from the concentrations that people breathe, especially for tailpipe pollutants such as carbon monoxide (CO), and reactive chemicals, such as ozone (O3) and sulfur dioxide (SO2). Data for other toxic pollutants in the outdoor ambient air are not generally collected on as routine a basis.

FIGURE 1.4 Respirable particle concentrations, six U.S. cities, November 1976 to April 1978. Source: National Academy of Science (1981).

12

1.4.6

INTRODUCTION AND BACKGROUND

Transportation

Many people spend from 1/2 to 3 h each day in autos or mass transport as they go to work, to school, or shopping. Inhalation exposures to CO in vehicles and garages can represent a significant fraction of total CO exposures. 1.4.7

Recreation and Public Spaces

Recreational exposure while exercising may be important to total daily exposure because the increased respiratory ventilation associated with exercise can produce much more than proportional increases in delivered dose and functional responses. Spectators and athletes in closed arenas can be exposed to high concentration of pollutants. For example, Spengler et al. (1978) documented high exposures to CO at ice rinks from exhaust discharges by the icescraping machinery.

1.5 INHALATION EXPOSURES AND RESPIRATORY TRACT EFFECTS 1.5.1

Deposition and Absorption

The surface and systemic uptake of chemicals from inhaled air depend on both their physical and chemical properties and on the anatomy and pattern of respiration within the respiratory airways. The basic structure of the respiratory tract is illustrated in Fig. 1.5. The following discussion outlines some of the primary factors affecting the deposition and retention of inhaled chemicals. More comprehensive discussions are available in recent reviews (ICRP, 1994; NCRP, 1997; U.S. EPA, 1996). Figure 1.6, from the 1994 ICRP Report, summarizes the morphometry, cytology, histology, function, and structure of the human respiratory tract, while Fig. 1.7 shows the compartmental model developed by ICRP (1994) to summarize particle transport from the deposition sites within the respiratory tract. Gases and vapors rapidly contact airway surfaces by molecular diffusion. Surface uptake is limited for compounds that are relatively insoluble in water, such as O3. For such chemicals, the greatest uptake can be in the lung periphery, where the residence time and surface areas are the greatest. For more water-soluble gases, dissolution and/or reaction with surface fluids on the airways facilitates removal from the airstream. Highly water-soluble vapors, such as SO2, are almost completely removed in the airways of the head, and very little of them penetrates into lung airways. For airborne particles, the most critical parameter affecting patterns and efficiencies of surface deposition is particle size. The mechanisms for particle deposition within respiratory airways are illustrated in Fig. 1.8. Almost all of the mass of airborne particulate matter is found in particles with diameters greater than 0.1 mm. Such particles have diffusional displacements many orders of magnitude smaller than those of gas molecules, and they are small in relation to the sizes of the airways in which they are suspended. Thus, the penetration of airborne particles into the lung airways is determined primarily by convective flow; that is, the motion of the air in which the particles are suspended. Some deposition by diffusion does occur for particles <0.5 mm in small airways, where it is favored by the small size of the airways and the low flow velocities in such airways. For particles >0.5 mm, deposition by sedimentation occurs in small to midsized airways. For particles with aerodynamic diameters >2 mm, particle inertia is sufficient to cause particle motion to deviate from the flow streamlines, resulting in deposition by impaction on surfaces

INHALATION EXPOSURES AND RESPIRATORY TRACT EFFECTS

FIGURE 1.5

13

Structure of the respiratory tract. Reproduced from National Research Council (1979).

downstream of changes in flow direction, primarily in mid- to large-sized airways, which have the highest flow velocities. The concentration of deposition on limited surface areas within the large airways is of special interest with respect to dosimetry and the pathogenesis of chronic lung diseases such as bronchial cancer and bronchitis. Although particle inertia accounts for much of the “hot-spot” deposition on the trachea below the laryngeal jet and at the bifurcations of large lung airways, some of the concentrated deposition is attributable to inertial airflow, which directs a disproportionately large fraction of the flow volume toward such surfaces and, at the same time, lessens the boundary layer thickness. Thus, there is some preferential deposition of submicrometer-sized particles and gas molecules at small airway bifurcations. Quantitative aspects of particle deposition are summarized in Figs. 1.9–1.12. It can be seen that deposition efficiencies in the major structural–functional regions of the human respiratory tract are both strongly particle size dependent and highly variable among normal humans. Additional variability results from structural changes in the airways associated with disease processes. Generally, these involve airway narrowing or localized constrictions, which act to increase deposition and concentrate it on limited surface areas. All of the preceding was based on the assumption that each particle has a specific size. For particles that are hygroscopic, there is considerable growth in size as they take up water vapor

Interalveolar septa covered by squamous epithelium, containing capillaries, surfactant

Cuboidal alveolar epithelial cells (Type il. Surfactant-producing), covering 7% of alveolar surface area

†

**

**

16 – 18

Lymphatics

Alveolar sacs

Alveolar ducts

Respiratory bronchioles

Terminal bronchioles

Bronchioles

Bronchi

Trachea Main bronchi

Larynx

Nose Mouth

FIGURE 1.6

ET1

New

Esophagus

† LNTH

L

P

(T-B)

511

140m2

7.5 m2

4.5 x 107

4.6 x 105

2.6 x 10–1 m2 6.5 x 104

3 x 10–2 m2

—

—

Morphometry, cytology, histology, function, and structure of the respiratory tract and regions used in the 1994 ICRP dosimetry model.

Al

bb

BB

4.5 x 10–2 m2

2 x 10–3 m2

Zones Airway Number of Old* (Air) Location Surface Airways

Pharynx ET2 LNET (N-P) posterior

Anterior nasal passages

Anatomy

Lymph nodes are located only in BB region but drain the bronchial and alveolar interstitial regions as well as the bronchial region.

* Previous ICRP model. ** Unnumbered because of imprecise information.

Alveolar macrophages

Wall consists of alveolar entrance rings, squamous epithelial layer, surfactant

Squamous alveolar epithelium cells (Type i), covering 93% of alveolar surface areas

Gas exchange; very slow particle clearance

Mucous membrane, single-layer respiratory epithelium of cuboidal cells, smooth muscle layers

15

Mucous membrane, single-layer respiratory epithelium, less ciliated, smooth muscle layer

Respiratory epithelium consisting mainly of clara cells (secretory) and few ciliated cells

9 – 14

2–8

1

0

Mucous membrane, respiratory epithelium, no cartilage, no glands, smooth muscle layer

Mucous membrane, respiratory epithelium, cartilage plates, smooth muscle layer, glands

Mucous membrane, respiratory epithelium, cartilage rings, glands

Mucous membrane, respiratory or stratified epithelium, glands

Mucous membrane, respiratory epithelium (pseudostratified, ciliated, mucous), glands

Histology (Walls)

Air conduction; gas exchange; slow particle clearance

Respiratory epithelium with clara cells (No goblet cells) Cell types: - Ciliated cells - Nonciliated cells • Clara (secretory) cells

Air conditioning; Respiratory epithelium with goblet cells: temperature and Cell types: humidily, and - Ciliated cells cleaning; fast - Nonciliated cells: particle clearance; • Goblet cells air conduction • Mucous (secretory) cells • Serous cells • Brush cells • Endocrine cells • Basal cells • Intermediate cells

Cylology (Epithelium)

Regions used in Model

Extrathoracic Thoracic

Functions

Generation Number

Conditioning Conduction Gas-exchange transitory

0.175 x 10–3 m3 (Anatomical Dead Space) 0.2 x 10–3 m3 4.5 x 10–3 m3

Extrapulmonary Pulmonary

14

INHALATION EXPOSURES AND RESPIRATORY TRACT EFFECTS

15

FIGURE 1.7 Compartment model to represent time-dependent particle transport from each region in 1994 ICRP model. Particle transport rate constants shown beside the arrows are reference values in d 1. Compartment numbers (shown in the lower right-hand corner of each compartment box) are used to define clearance pathways. Thus, the particle transport rate from bb1 to BB1 is denoted m4,7 and has the value 2 d 1.

in the airways. Some hygroscopic growth curves for acidic and ambient aerosols are illustrated in Fig. 1.13. Materials that dissolve into the mucus of the conductive airways or the surfactant layer of the alveolar region can rapidly diffuse into the underlying epithelia and the circulating blood, thereby gaining access to tissues throughout the body. Chemical reactions and metabolic processes may occur within the lung fluids and cells, limiting access of the inhaled material to the bloodstream and creating reaction products with either greater or lesser solubility and biological activity. Few generalizations about absorption rates are possible.

FIGURE 1.8 Schematic of mechanism for particle deposition in respiratory airways. Source: Lippmann and Schlesinger (1984).

16

INTRODUCTION AND BACKGROUND

1.0

0.8

ηn

Landahl & Tracewell Pattle Lippmann Hounam et al. Giacomelli-Maltoni et al. Martens & Jacobi Rudolf

1949 1961 1970 1971 1972 1973 1975

0.6 0.4 0.2

0 100

101

102

103

104

105

106

dae2 Q (µm2 cm3/s1)

FIGURE 1.9 Inspiratory deposition of the human nose as a function of particle aerodynamic 2 diameter and flow rate (dae Q). From: EPA (1997).

1.5.2

Translocation and Retention

Particles that do not dissolve at deposition sites can be translocated to remote retention sites by passive and active clearance processes. Passive transport depends on movement on or in surface fluids lining the airways. There is a continual proximal flow of lung surfactant from alveolar epithelial cells to and onto the mucociliary escalator, which begins at the terminal bronchioles, where it mixes with secretions from Clara and goblet cells in the airway epithelium. Within midsized and larger airways there are additional secretions from goblet cells and mucus glands, producing a thicker mucous layer having a serous subphase and an

FIGURE 1.10 Inspiratory extrathoracic deposition data in humans during mouth breathing as a 1=4 2 function of particle aerodynamic diameter, flow rate, and tidal volume (dae Q 2=3 VT ). From: EPA (1997).

INHALATION EXPOSURES AND RESPIRATORY TRACT EFFECTS

17

FIGURE 1.11 Tracheobronchial deposition data in humans at mouth breathing as a function of particle aerodynamic diameter (dae). The solid curve represents the approximate mean of all the experimental data; the broken curve represents the mean excluding the data of Stahlhofen et al. From: EPA (1997).

overlying more viscous gel layer. The gel layer, lying above the tips of the synchronously beating cilia, is found in discrete plaques in smaller airways and becomes more of a continuous layer in the larger airways. The mucus reaching the larynx and the particles carried by it are swallowed and enter the GI tract. The total transit time for particles depositing on terminal bronchioles varies from 2 to 24 h in healthy humans, accounting for the relatively rapid bronchial clearance phase. Macrophage-mediated particle clearance via the bronchial tree takes place over a period of

FIGURE 1.12 Alveolar deposition data in humans as a function of particle aerodynamic diameter (dae). The solid curve represents the mean of all the data; the broken curve is an estimate of deposition for nose breathing by Lippmann (1977). From: EPA (1997).

18

INTRODUCTION AND BACKGROUND

FIGURE 1.13 Tracheobronchial particle deposition as a function of particle size at various ages for both stable iron oxide particles and hygroscopic sulfuric acid droplets that grow in size in the warm moist respiratory airways. Source: Martonen (1990).

several weeks. The particles depositing in alveolar zone airways are ingested by alveolar macrophages within about 6 h, but the movement of the particle-laden macrophages depends on the several weeks that it takes for the normal turnover of the resident macrophage population. At the end of several weeks, the particles not cleared to the bronchial tree via macrophages have been incorporated into epithelial and interstitial cells, from which they are slowly cleared by dissolution and/or as particles via lymphatic drainage pathways, passing through pleural and eventually hilar and tracheal lymph nodes. Clearance times for these later phases depend strongly on the chemical nature of the particles and their sizes, with half-times ranging from about 30 to 1000 days, or more. All of the characteristic clearance times cited refer to inert, nontoxic particles in healthy lungs. Toxicants can drastically alter clearance times. Inhaled materials affecting mucociliary clearance rates include cigarette smoke (Albert et al., 1974, 1975), sulfuric acid (H2SO4) (Lippmann et al., 1982; Schlesinger et al., 1983), O3 (Phalen et al., 1980; Schlesinger and Driscoll, 1987), SO2 (Wolff et al., 1977), and formaldehyde (Morgan et al., 1984). Macrophage-mediated alveolar clearance is affected by SO2 (Ferin and Leach, 1973), NO2 and H2SO4 (Schlesinger et al., 1988), O3 (Phalen et al., 1980; Schlesinger et al., 1988), and silica dust (Jammet et al., 1970). Cigarette smoke is known to affect the later phases of alveolar zone clearance in a dose-dependent manner (Bohning et al., 1982). Clearance pathways as well as rates can be altered by these toxicants, affecting the distribution of retained particles and their dosimetry.

1.6 INGESTION EXPOSURES AND GASTROINTESTINAL TRACT EFFECTS Chemical contaminants in drinking water or food reach human tissues via the GI tract. Ingestion may also contribute to uptake of chemicals that were initially inhaled, since material deposited on or dissolved in the bronchial mucous blanket is eventually swallowed. The GI tract may be considered a tube running through the body, the contents of which are actually external to the body. Unless the ingested material affects the tract itself, any systemic response depends on absorption through the mucosal cells lining the lumen. Although

SKIN EXPOSURE AND DERMAL EFFECTS

19

absorption may occur anywhere along the length of the GI tract, the main region for effective translocation is the small intestine. The enormous absorptive capacity of this organ results from the presence in the intestinal mucosa of projections, termed villi, each of which contains a network of capillaries; the villi result in a large effective total surface area for absorption. Although passive diffusion is the main absorptive process, active transport systems also allow essential lipid-insoluble nutrients and inorganic ions to cross the intestinal epithelium and are responsible for uptake of some contaminants. For example, lead may be absorbed via the system that normally transports calcium ions (Sobel et al., 1938). Small quantities of particulate material and certain large macromolecules, such as intact proteins, may be absorbed directly by the intestinal epithelium. Materials absorbed from the GI tract enter either the lymphatic system or the portal blood circulation; the latter carries material to the liver, from which it may be actively excreted into the bile or diffuse into the bile from the blood. The bile is subsequently secreted into the intestines. Thus, a cycle of translocation of a chemical from the intestine to the liver to bile and back to the intestines, known as the enterohepatic circulation, may be established. Enterohepatic circulation usually involves contaminants that undergo metabolic degradation in the liver. For example, DDT undergoes enterohepatc circulation; a product of its metabolism in the liver is excreted into the bile, at least in experimental animals (Hayes, 1965). Various factors serve to modify absorption from the GI tract, enhancing or depressing its barrier function. A decrease in gastrointestinal mobility generally favors increased absorption. Specific stomach contents and secretions may react with the contaminant, possibly changing it to a form with different physicochemical properties (e.g., solubility), or they may absorb it, altering the available chemical and changing translocation rates. The size of ingested particulates also affects absorption. Since the rate of dissolution is inversely proportional to particle size, large particles are absorbed to a lesser degree, especially if they are of a fairly insoluble material in the first place. For example, arsenic trioxide is more hazardous when ingested as a finely divided powder than as a coarse powder (Schwartz, 1923). Certain chemicals, for example, chelating agents such as EDTA, also cause a nonspecific increase in absorption of many materials. As a defense, spastic contractions in the stomach and intestine may serve to eliminate noxious agents via vomiting or by acceleration of the transit of feces through the GI tract.

1.7 SKIN EXPOSURE AND DERMAL EFFECTS The skin is generally an effective barrier against the entry of environmental chemicals. In order to be absorbed via this route (percutaneous absorption), an agent must traverse a number of cellular layers before gaining access to the general circulation (Fig. 1.14). The skin consists of two structural regions, the epidermis and the dermis, which rest on connective tissue. The epidermis consists of a number of layers of cells and has varying thickness depending on the region of the body; the outermost layer is composed of keratinized cells. The dermis contains blood vessels, hair follicles, sebaceous and sweat glands, and nerve endings. The epidermis represents the primary barrier to percutaneous absorption, the dermis being freely permeable to many materials. Passage through the epidermis occurs by passive diffusion. The main factors that affect percutaneous absorption are degree of lipid solubility of the chemicals, site on the body, local blood flow, and skin temperature. Some environmental chemicals that are readily absorbed through the skin are phenol, carbon tetrachloride,

20

INTRODUCTION AND BACKGROUND

FIGURE 1.14

Idealized section of skin. Source: Birmingham (1973).

tetraethyl lead, and organophosphate pesticides. Certain chemicals, for example, dimethyl sulfoxide (DMSO) and formic acid, alter the integrity of skin and facilitate penetration of other materials by increasing the permeability of the stratum corneum. Moderate changes in permeability may also result following topical applications of acetone, methyl alcohol, and ethyl alcohol. In addition, cutaneous injury may enhance percutaneous absorption. Interspecies differences in percutaneous absorption are responsible for the selective toxicity of many insecticides. For example, chlorinated hydrocarbons (HC are about equally hazardous to insects and mammals if ingested but are much less hazardous to mammals when applied to the skin. This is because of their poor absorption through mammalian skin compared to their ready passage through the insect exoskeleton. Although the main route of percutaneous absorption is through the epidermal cells, some chemicals may follow an appendageal route, that is, entering through hair follicles, sweat glands, or sebaceous glands. Cuts and abrasions of the skin can provide additional pathways for penetration. 1.8 ABSORPTION THROUGH MEMBRANES AND SYSTEMIC CIRCULATION Depending upon its specific nature, a chemical contaminant may exert its toxic action at various sites in the body. At a portal of entry—the respiratory tract, GI tract, or skin—the chemical may have a topical effect. However, for actions at sites other than the portal, the agent must be absorbed through one or more body membranes and enter the general circulation, from which it may become available to affect cells and internal tissues (including the blood itself). The ultimate distribution of any chemical contaminant in the body is, therefore, highly dependent on its ability to traverse biological membranes. There are two main types of processes by which this occurs: passive transport and active transport. Passive transport is absorption according to purely physical processes, such as osmosis; the cell has no active role in transfer across the membrane. Since biological membranes contain lipids, they are highly permeable to lipid-soluble, nonpolar or

ACCUMULATION IN TARGET TISSUES AND DOSIMETRIC MODELS

21

nonionized agents and less so to lipid-insoluble, polar, or ionized materials. Many chemicals may exist in both lipid-soluble and -insoluble forms; the former is the prime determinant of the passive permeability properties for the specific agent. Active transport involves specialized mechanisms, with cells actively participating in transfer across membranes. These mechanisms include carrier systems within the membrane and active processes of cellular ingestion; that is, phagocytosis and pinocytosis. Phagocytosis is the ingestion of solid particles, whereas pinocytosis refers to the ingestion of fluid containing no visible solid material. Lipid-insoluble materials are often taken up by active-transport processes. Although some of these mechanisms are highly specific, if the chemical structure of a contaminant is similar to that of an endogeneous substrate, the former may be transported as well. In addition to its lipid-solubility characteristics, the distribution of a chemical contaminant is also dependent on its affinity for specific tissues or tissue components. Internal distribution may vary with time after exposure. For example, immediately following absorption into the blood, inorganic lead is found to localize in the liver, the kidney, and in red blood cells. Two hours later, about 50% is in the liver. A month later, approximately 90% of the remaining lead is localized in bone (Hammond, 1969). Once in the general circulation, a contaminent may be translocated throughout the body. In this process it may(1) become bound to macromolecules, (2) undergo metabolic transformation (biotransformation), (3) be deposited for storage in depots that may or may not be the sites of its toxic action, or (4) be excreted. Toxic effects may occur at any of several sites. The biological action of a contaminant may be terminated by storage, metabolic transformation, or excretion, the latter being the most permanent form of removal.

1.9 ACCUMULATION IN TARGET TISSUES AND DOSIMETRIC MODELS Some chemicals tend to concentrate in specific tissues because of physicochemial properties such as selective solubility or selective absorption on or combination with macromolecules such as proteins. Storage of a chemical often occurs when the rate of exposure is greater than the rate of metabolism and/or excretion. Storage or binding sites may not be the sites of toxic action. For example, CO produces its effects by binding with hemoglobin in red blood cells; on the contrary, inorganic Pb is stored primarily in bone but exerts it toxic effects mainly on the soft tissues of the body. If the storage site is not the site of toxic action, selective sequestration may be a protective mechanism, since only the freely circulating form of the contaminant produces harmful effects. Until the storage sites are saturated, a buildup of free chemical may be prevented. On the contrary, selective storage limits the amount of contaminant that is excreted. Since bound or stored toxicants are in equilibrium with their free form, as the contaminant is excreted or metabolized, it is released from the storage site. Contaminants that are stored (e.g., DDT) may remain in the body for years without effect. On the contrary, accumulation may produce illnesses that develop slowly, as occurs in chronic Cd poisoning. A number of descriptive and mathematical models have been developed to permit estimation from knowledge of exposure and one or more of the following factors: translocation, metabolism, and effects at the site of toxic action.

22

INTRODUCTION AND BACKGROUND

The use of these models for airborne particulate matter generally requires knowledge of the concentration within specific particle size intervals or of the particle size distribution of the compounds of interest. Simple deposition models break the respiratory tract into regions (summarized by Vincent, 1999): Head airways, nasopharynx, extrathoracic: nose, mouth, nasopharynx, oropharynx, laryngopharynx. Tracheobronchial: larynx, trachea, bronchi, bronchioles (to terminal bronchioles). Gas exchange, pulmonary, alveolar: respiratory bronchioles, alveolar ducts, alveolar sacs, alveoli. Size-selective aerosol sampling can mimic the head airways and tracheobronchial airway regions so that airborne particle collection can be limited to the size fraction directly related to the potential for disease. More complex models requiring data on translocation and metabolism have been developed for inhaled and ingested radionuclides by the International Commission on Radiological Protection (ICRP, 1966, 1979, 1981, 1994).

1.10 INDIRECT MEASURES OF PAST EXPOSURES Documented effects of environmental chemicals on humans seldom contain quantitative exposure data and only occasionally include more than crude exposure rankings based on known contact with or proximity to the materials believed to have caused the effects. Reasonable interpretation of the available human experience requires some appreciation of the uses and limitations of the data used to estimate the exposure side of the exposure– response relationship. The discussion that follows is an attempt to provide background for interpreting data, and for specifying the kinds of data needed for various analyses. Both direct and indirect exposure data can be used to rank exposed individuals by exposure intensity. External exposure can be measured directly by collection and analysis of environmental media. Internal exposure can be estimated from analyses of biological fluids and in vivo retention. Indirect measures generally rely on work or residential histories with some knowledge of exposure intensity at each exposure site and/or some enumeration of the frequency of process upsets and/or effluent discharges that result in high-intensity short-term exposures. 1.10.1

Concentrations in Air, Water, and Food

Historic data may occasionally be available on the concentrations of materials of interest in environmental media. However, they may or may not relate to the exposures of interest. Among the more important questions to be addressed in attempts to use much data are: (1) How accurate and reliable were the sampling and analytical techniques used in the collection of the data? Were they subjected to any quality assurance protocols? Were standardized and/or reliable techniques used? (2) When and where were the samples collected, and how did they relate to exposures at other sites? Air concentrations measured at fixed (area) sites in industry may be

CHARACTERIZATION OF HEALTH

23

much lower than those occurring in the breathing zone of workers close to the contaminant sources. Air concentrations at fixed (generally elevated) community air-sampling sites can be either much higher or much lower than those at street level and indoors as a result of strong gradients in source and sink strengths in indoor and outdoor air. (3) What is known or assumed about the ingestion of food and/or water containing the measured concentrations of the contaminants of interest? Time at home and dietary patterns are highly variable among populations at risk. 1.10.2

Biological Sampling Data

Many of the same questions that apply to the interpretation of environmental media concentration data also apply to biological samples, especially quality assurance. The time of sampling is especially critical in relation to the times of the exposures and to the metabolic rates and pathways. In most cases, it is quite difficult to separate the contributions to the concentrations in circulating fluids of levels from recent exposures and those from long-term reservoirs. 1.10.3

Exposure Histories

Exposure histories per se are generally unavailable, except in the sense that work histories or residential histories can be interpreted in terms of exposure histories. Job histories, as discussed below, are often available in company and/or union records and can be converted into relative rankings of exposure groups with the aid of long-term employees and managers familiar with the work processes, history of process changes, material handled, tasks performed, and the engineering controls of exposure. Routine, steady-state exposures may be the most important and dominant exposures of interest in many cases. On the contrary, for some health effects, the occasional or intermittent peak exposures may be of primary importance. In assessing or accumulating exposure histories or estimates, it is important to collect evidence for the frequency and magnitude of the occasional or intermittent releases associated with process upsets. 1.11 CHARACTERIZATION OF HEALTH 1.11.1

Definitions of Health

There is no universally accepted definition of health. Perhaps the most widely accepted one today is that of the World Health Organization, which describes health as a state of complete physical, mental, and social well-being, and not merely the absence of disease or infirmity. Unfortunately, by a strict interpretation of this rather idealistic definition, very few people could be considered healthy. The discussion to follow is limited largely to physical well-being. The health effects discussed are those that can be recognized by clinical signs, symptoms, or decrements in functional performance. Thus, for all practical purposes, in this volume we consider health to be the absence of measurable disease, disability, or dysfunction. 1.11.2

Health Effects

Recognizable health effects in populations are generally divided into two categories: mortality and morbidity. The former refers to the number of deaths per unit of population

24

INTRODUCTION AND BACKGROUND

per unit time, and to the ages at death. Morbidity refers to nonfatal cases of reportable disease. Accidents, infectious diseases, and massive overexposures to toxic chemicals can cause excess deaths to occur within a short time after the exposure to the hazard. They can also result in residual disease and/or dysfunction. In many cases, the causal relationships are well defined, and it may be possible to develop quantitative relationships between dose and subsequent response. The number of people exposed to chemical contaminants at low levels is, of course, much greater than the number exposed at levels high enough to produce overt responses. Furthermore, low-level exposures are often continuous or repetitive over periods of many years. The responses, if any, are likely to be nonspecific, for example, an increase in the frequency of chronic diseases that are also present in nonexposed populations. For example, any small increase in the incidence of heart disease or lung cancer attributable to a specific chemical exposure would be difficult to detect, since these diseases are present at high levels in nonexposed populations. In smokers they are likely to be influenced more by cigarette exposure than by the chemical in question. Increases in the incidence of diseases from low-level long-term exposure to environmental chemicals invariably occur among a very small percentage of the population and can only be determined by large-scale epidemiological studies (epidemiology is the study of the distribution and frequency of diseases in a specific population) involving thousands of person-years of exposure. The only exceptions are chemicals that produce very rare disease conditions, where the clustering of a relatively few cases may be sufficient to identify the causative agent. Notable examples of such special conditions are the industrial cases of chronic berylliosis caused by the inhalation of beryllium-containing dusts, a rare type of liver cancer that resulted from the inhalation of vinyl chloride vapors, and pleural cancers that resulted from the inhalation of asbestos fibers. If these exposures had produced more commonly seen diseases, the specific materials might never have been implicated as causative agents. Low-level chemical exposures may play contributory, rather than primary, roles in the causation of an increased disease incidence, or they may not express their effects without the co-action of other factors. For example, the excess incidence of lung cancer is very high in uranium miners and asbestos workers who smoke cigarettes but is only marginally elevated among nonsmoking workers with similar occupational exposures. For epidemiological studies to provide useful data, they must take appropriate account of smoking histories, age, and sex distributions, socioeconomic levels, and other factors that affect mortality rates and disease incidence. 1.11.2.1 Mortality In industrialized societies, there is generally good reporting of mortality and age at death but, with few exceptions, quite poor reporting of cause of death. In studies that are designed to determine associations between exposures and mortality rates, it is usually necessary to devote a major part of the effort to follow-up investigations of cause-of-death. The productivity of these follow-ups is often marginal, limiting the reliability of the overall study. 1.11.2.2 Morbidity Difficult as it may be to conduct good mortality studies, it is far more difficult, in most cases, to conduct studies involving other health effects. Although there is generally little significant variability in the definition of death, there is a great deal of variation in the diagnosis and reporting of many chronic diseases. There are variations

EXPOSURE–RESPONSE RELATIONSHIPS

25

between and within countries and states, and these are exacerbated by the differences in background and outlooks of the physicians making the individual diagnoses. Furthermore, there are some important chronic diseases that cannot be definitively diagnosed in vivo. Many epidemiological studies rely on standardized health status questionnaires, and the success of these studies depends heavily on the design of the questionnaires. Of equal importance in many studies are the training and motivation of the persons administering the questionnaires. Similar considerations apply to the measurement of functional impairment. The selection of the measurements to be used is very important; those functions measured should be capable of providing an index of the severity of the disease. Equally important here are the skills of the technicians administering these tests and their maintenance and periodic recalibration of the equipment. Some studies try to avoid bias from the administrators of the questionnaires and functional tests by having the selected population enter the desired information themselves. They may be asked to make appropriate notations in notebook diaries or to call a central station whenever they develop the symptoms of interest. Other investigations use nonsubjective indices such as hospital admissions, clinic visits, and industrial absenteeism as their indicators of the health effects to be associated with the environmental variables.

1.12 EXPOSURE–RESPONSE RELATIONSHIPS Exposure–response relationships can be developed from human experience, but there are many chemicals that are known to be toxic in animals for which the extent of human toxicity, if any, is unknown. In order to use animal bioassay data for the prediction of human responses to environmental exposures, it is necessary to make two major kinds of extrapolation. One is determine or estimate the relative responsiveness of humans and the animal species used in the bioassays. The second is to extrapolate from the observed effects resulting from relatively high administered doses to the much lower levels of effects still of concern at much lower levels of environmental exposure. To deal with interspecies extrapolation, estimates are made on the basis of whatever is known about differences in uptake from environmental media, metabolic rates and pathways, retention times in target tissues, and so on, and tissue sensitivities. As uncertain as these extrapolations are, they are more straightforward than the low-dose extrapolation. The goal of the dose–response assessment is to predict what response, if any, might occur, 10- to 1000-fold below the lowest dose tested in rodents (this is more representative of the range of doses to which humans are usually exposed). Because it would require the testing of thousands of animals to observe a response at low doses, mathematical models are used (Munro and Krewski, 1981). To appreciate the level of uncertainty in the dose extrapolation process and the typical regulatory use of low-dose models, it is useful to discuss the dose– response curve. However, reliance on the results of only one mathematical model is a potential pitfall in the dose–response assessment. There are at least six different modeling approaches that may need to be considered when estimating the risks at low doses. These models include the probit, multihit, multistage, Weibull, one-hit, and the Moolgavkar–Knudson–Venzon (MKV) biologically based approaches (Moolgavkar et al., 1988). Nearly all of them can yield results that are plausible. No single statistical model can be expected to predict accurately the low-dose response with greater certainty than another. As discussed by Paustenbach (1990),

26

INTRODUCTION AND BACKGROUND

FIGURE 1.15 The fit of most dose–response models to data in the range tested in animal studies is generally similar. However, because of the differences in the assumptions on which the equations are based, the risk estimates at low doses can vary dramatically between different models. Source: Paustenbach (1990).

one possible way to resolve this problem is to present the best estimate of the risk from the two or three models that are considered equally reasonable along with the upper- and lower-bound estimates. An alternate approach is to identify a single value based on the “weight of evidence,” as the EPA did for dioxin (U.S. EPA, 1988) Low-dose models usually fit the rodent data in the dose region used in the animal tests. However, they often predict quite different results in the unobserved low-dose region (Fig. 1.15). The results of the most commonly used low-dose models usually vary in a predictable manner because the models are based on different mathematical equations for describing the chemical’s likely behavior in the low-dose region. In general, the scientific underpinnings of the dose–response models are based on the present understanding of the cancer process caused by exposure to ionizing radiation and genotoxic chemicals (NRC, 1980). Both types of agents may well have a linear, or a nearly linear, response in the low-dose region. However, promoters and cytotoxicants (e.g., nongenotoxicants) would be expected to be very nonlinear at low doses and may have a genuine or practical threshold (a dose below which no response would be present) (Squire, 1987; Butterworth and Slaga, 1987). Thus, the linearized multistage model may be inappropriate for dioxin, thyroid-type carcinogens, nitrolotriacetic acid, and, presumably, similar nongenotoxic chemicals (Paynter et al., 1988; Andersen and Alden, 1989). For these types of chemicals, the MKV model, or one of the other biologically based models, should be more appropriate (Moolgavkar, 1978; Ellwein and Cohen, 1988). 1.12.1

Summary of Exposure- and Dose-Related Responses

Studies of the specific responses of biological systems to varying levels of exposure can provide a great deal of information on the nature of the responses, their underlying

EXPOSURE–RESPONSE RELATIONSHIPS

27

FIGURE 1.16 Dose (population) response relationship with suggested distinction between basic (toxicological) and practical (health) scales on the three axes. The illustrative curve on the horizontal plane portrays the dose–response relationship for the middle (50%) of the exposed population; the curve on the vertical plane shows the percentages of population response of the indicated degree over the whole range of doses. The vertical line from the dose scale indicates the magnitude of dose needed to produce the indicated degree of response at the 50% population level. Source: Hatch (1968).

causes, and the possible consequences of various levels of exposure. However, it must be remembered that the data are most reliable only for the conditions of the test and for the levels of exposure that produced clear-cut responses. Generally, in applying experimental data to low-level environmental exposure conditions, it is necessary to extrapolate to delivered doses that are orders of magnitude smaller than those that produced the effects in the test system. Since the slope of the curve becomes increasingly uncertain the further one extends it beyond the range of experimental data, the extrapolated effects estimate may be in error by a very large factor. The basic dimensions of the dose–response relationship for populations were described by Hatch (1968), as illustrated in Fig. 1.16. Many factors affect each of the basic dimensions. 1.12.1.1 Factors Affecting Dose The effective dose is the amount of toxicant reaching a critical site in the body. It is proportional to the concentrations available in the environment: in the air breathed, the water and food ingested, and so on. However, the uptake also depends on the route of entry into the body and the physical and chemical forms of the contaminant. For airborne contaminants, for example, the dose to the respiratory tract depends on whether they are present in a gaseous form or as an aerosol. For contaminants that are ingested, uptake depends on transport through the membranes lining the gastrointestinal tract and, in turn, is dependent on both aqueous and lipid solubilities.

28

INTRODUCTION AND BACKGROUND

For contaminants that penetrate membranes, reach the blood, and are transported systemically, subsequent retention in the body depends on their metabolism and toxicity in the various tissues in which they are deposited. In all of these factors, there are great variations within and between species, and therefore great variations in effective dose for a given environmental level of contamination. 1.12.1.2 Factors Affecting Response The response of an organism to a given environmental exposure can also be quite variable. It can be influenced by age, sex, the level of activity at the time of exposure, metabolism, and the competence of the various defense mechanisms of the body. The competence of the body’s defenses may, in turn, be influenced by the prior history of exposures to chemicals having similar effects, since those exposures may have reduced the reserve capacity of some important functions. The response may also depend on other environmental factors, such as heat stress and nutritional deficiencies. These must all be kept in mind in interpreting the outcomes of controlled exposures and epidemiological data and in extrapolating results to different species and across various age ranges, states of health, and so on. 1.12.1.3 Factors Affecting Individual Susceptibility The complete evaluation of the pathogenesis of human disease requires identification and assessment of the genetic, lifestyle, and environmental risk factors. Environmental factors are clearly critical to disease prevention because, at the societal level, they are determined by the most controllable processes. Thus, the overall purpose of environmental medicine is to improve our knowledge of the more commonly encountered environmental agents that present the greatest concern to human health. In the past, diseases were categorized by the main etiological factor into either (a) genetic, (b) lifestyle-induced, or (c) environmental-induced disorders. Illustrative examples for each category include (a) alpha-1 antitrypsin [serine (or cysteine) proteinase inhibitor, clade A, member1] deficiency-induced chronic obstructive pulmonary disease; (b) cigarette-induced lung cancer; and (c) asbestos-induced mesothelioma, respectively. Environmental medicine almost exclusively focused on the latter category, and has had its best impacts when armed with knowledge of quantifiable exposure (e.g., occupational diseases). Fortunately, additional cases of occupational diseases (e.g., mining-related coal workers pneumoconiosis or asbestosis) are becoming increasingly rare among the general population due to improved work practices. However, while categorizations based on etiological factors have allowed the development of therapeutic strategies to treat disease, they also may have limited past attempts to prevent disease. Disease prevention has a greater ability than most common treatment modalities to reduce incidence and to extend life expectancy. In the illustrative examples above, preventative approaches would include: (a) genetic counseling and possibly genereplacement therapy; (b) public health education and smoking cessation pharmaceuticals; and (c) reduction or elimination (banning widespread usage) of asbestos exposure. Each of theses approaches have had tremendous value in focusing efforts on disease causalities and thus has had dramatic impacts on disease prevention. Although the application of these simple approaches to disease prevention has been partially successful, further success will require more sophisticated approaches based on a deeper understanding of disease pathogenesis. With the widespread use of genetic screening, it soon became apparent that homozygotic recessive carriers (individuals inheriting both copies of the disease susceptibility alleles) often did not developed major signs and

EXPOSURE–RESPONSE RELATIONSHIPS

29

symptoms of the “genetic” disease. For example, homozygotic twins do not have 100% concordance in disease outcomes. Similarly, disease discordance is noted in lifestyleinduced diseases. For example, among individuals with exposures exceeding 50 pack-years, only four out five cigarette smokers develop a tobacco-related disease and these diseases vary in the population (i.e., multisite cancer, cardiovascular, and respiratory disease). Likewise, although diseases such as asbestosis, lung cancer, and mesthelioma are enriched in populations with excessive asbestos exposure, the incidence of affected workers given equivalent exposures in not 100%. What can explain a lack of penetrance (affected individuals/genotype positive individuals) in equally susceptible, equally risky, equally exposed populations? It is the interactions of additional factors within and among each etiological category. Common pathological conditions are complex and involve multiple rather than a single gene(s). Single gene diseases with strongly expressivity typically appear early in life, but due to a lack of complete penetrance affect only a few members of the population. Alternatively, most diseases involve multiple gene–gene interactions and are present in a large percentage of the population. Genetic polymorphisms that are strongly fixed in the genome probably did not arise from modern lifestyle or environmental factors (e.g., coal mining), but from ancient lifestyle factors or infectious agents. Another reason why most common diseases are complex is because the selective advantage to a population that stabilizes a genetic polymorphism is likely to be dependent on multiple alterations in multiple genes. Many polymorphisms also may have been acquired from phylogenetic ancestry (which are likely to be greater than those that uniquely arise within a given species). Thus, the mechanisms by which bacteria combat virus, or how drosophila resist bacteria (toll-receptors), or how mice resist influenza, may be conserved throughout species. This situation is further complicated by the fact that genes that impart sensitivity may also impart resistance to another disease. For example, the protection from tuberculosis may be advantageous to the population, but may impart increased susceptibility to chronic inflammatory diseases, like arthritis, to a portion of the population. From a solely genetic stand point, it would be advantageous if multiple genes contribute to a survival phenotype from a severe disease entity [i.e., the fully developed phenotype being dependent on gene–gene interactions (epistasis)]. Phenotypes with complex gene interaction require only a few members of an outbred population (humans) to carry the exact set of all the resistant alleles. Phenotypes dependent upon multiple genes thereby reduce the negative consequences of the combinatorial effects of multiple alleles to only a few individuals. Several other members of a population could inherit in partial combinations that would have little observable phenotypic expression. Thus, while only a few members of the population might survive an infectious epidemic, many members of population share the risky alleles. With the initial assembling of the entire human genome by the Human Genome Project, substantial research has been invested in the identification of all disease causing genetic polymorphisms. However, individual susceptibility to common diseases is not solely controlled by multiple gene–gene interactions. Rather disease penetrance is also influenced by multiple gene–lifestyle, gene–environment, lifestyle–environmental, and gene– lifestyle–environmental factors. Using our illustrative disease example, clear synergistic interactions occur among individuals who have: (a) alpha-1 antitrypsin deficiency and smoke cigarettes (gene–lifestyle interaction); (b) glutathione S-transferase pi 1 deficiency and are exposed to environmental tobacco smoke or to excessive air pollution (ozone and particulate matter) (gene-environment interaction); or (c) smoke and work with asbestos or radon (lifestyle–environment interactions).

30

INTRODUCTION AND BACKGROUND

1.12.2 Genomic Approaches to Understanding Gene-Lifestyle-Environmental Factors in Complex Disease Pathogenesis Environmental health sciences have developed a wonderful cadre of tools to obtain global (nearly complete) evaluations of the genetic variants, transcriptional profile, protein usage and activation state, and the metabolic capability of individuals and populations following environmental stress. Using high-throughput microarray or fluidic luminesence systems, genomics seeks to evaluate the entire genetic makeup of an individual and thereby identify candidate gene suspected to have a role in determining susceptibility. Genomics is based on known biological functional, cellular location, or pathophysiological roles of a gene, and seeks to identify the allelic variants that associate with increased risk. Moreover, the mapping of over 150,000 single nucleotide polymorphisms (SNPs) throughout the genome allows nearly complete coverage of all human variability and through linkage disequilibrium identification of small chromosomal region (areas of a few tens of thousands base pairs). Particularly attention is focused on nonsynonymous SNPs that result in alteration of RNA recognition codons and lead to amino acid changes in the predicted protein. Because regions of chromosomes are often inherited together, the number of independently inherited SNPs is reduced. Thus, the reduction to informative differences (tag-SNPs) make these analyses even more powerful and essentially entire genome coverage is possible. Supportive of genome wide linkage analysis is transcriptomics, proteomics, and metabonomics. Transcriptomics studies large sets of messenger ribonucleic acid (mRNA) molecules, or transcripts produced by cells in culture (revealing cell-type specificity), isolated tissue, or a whole organism. The transcriptome, unlike the genome, which is fixed (excluding acquired mutations), can be altered by the environment and reflects the cell’s attempt to acclimate to adverse conditions. The supportive high-throughput technology includes DNA microarrays that typically monitor steady-state transcript levels of 30,000 genes. This includes essentially all known genes and genes yet to be fully understood and annotated (e.g., predicted gene products and expressed sequence tags). Confirmational approaches to key genes identified by microarray include quantitative reverse transcription polymerase chain reaction, ribonuclease protection assay, and Northern blot that are conducted on single or small sets of transcripts. Like the transcriptome, the proteomics is a global approach to evaluate altered protein usage and activation state induced by environmental signaling. Indeed, it includes many signaling peptides (e.g., kinases) that generate amplifying cascades that alter cell functions including motility, transcription, cell–cell communication, proliferation, immunity, and apoptosis. The proteome includes nascent propeptides, mature inactive peptides, activated peptides, and peptide marked secretion or degradation. While the number of genes and possible transcripts are estimated to be less than 35,000 in humans, the proteome may have over 1 million members. Moreover, proteomics focuses on protein–protein, and protein–macromolecular interactions, and is thus closer to functional significance than genomics or transcriptomics. High-throughput technologies supporting proteomics include, gas (gas chromatography), fluidic (high-pressure liquid chromatography) or gel (electrophoresis) separation and large scale, mass spectrometric protein identification. Confirmational approaches include immuno (Western) blot, antibody arrays, and enzyme-linked immunosorbent assays (ELISA), and radioimmunoassay. Also under rapid development, metabonomics (or metabolomics) is the global approach to the assessment of the metabolic response of living systems to environmental stimuli. Typically the variety of small molecule metabolites generated by a living system is relatively

STUDY OPTIONS FOR HEALTH EFFECTS STUDIES

31

small (<5000 compounds) when compared to the genome, transcriptome, or proteome. Metabonomic profiles reflect the nutritional status of the organism, and thus this approach is useful in determining lifestyle-environmental interactions. In addition, the metabolic consequences of genetic manipulation can be assessed using this approach (bypassing assessment of the transcriptome or proteome). The focus is typically on body fluids (serum or urine) and minimally-invasive sampling (salvia or breath analysis). Another major advantage of metabonomics is safety. Although metabolic profiles may be shifted due to altered pathologies, metabolic capacities often can be predicted a very low doses of known toxicants (below the levels inducing any adverse response) or by surrogate that is handled much like a toxicant (e.g., caffeine). Supportive technologies include mass spectrometry and nuclear magnetic resonance (NMR). The latter is commonly preferred because does not require separation, does not destroy the sample, and can be performed with small volumes (0.01–0.1 mL).

1.13 STUDY OPTIONS FOR HEALTH EFFECTS STUDIES In the discussion of the health effects in the chapters that follow, it is important to appreciate the strengths and limitations of the various kinds of studies that generated the data. 1.13.1

Controlled Human Exposures

For O3, SO2, concentrated PM2.5, and CO, there are databases from studies in which selected human volunteers were exposed to the pollutant in purified lab air for specific time intervals ranging from minutes to eight hours. Most studies involve a series of such exposures in random order, in which there are exposures at one or more concentrations as well as a sham exposure to purified air. Many of them involved prescribed periods and intensities of exercise during the exposure interval. The most commonly measured pulmonary effects were changes in forced expiratory flow rates and volumes and/or changes in airway resistance and compliance. A broad variety of other pulmonary function tests require the inhalation of special breathing mixtures and hence more elaborate controls and protocol reviews. These include the inhalation of (1) a single breath of pure oxygen for the nitrogen washout test of small airway function (Buist and Ross, 1973); (2) 0.3% CO to determine diffusing capacity at the alveolo-capillary membrane (Crapo, 1986); and (3) a low density inert gas, such as helium (He), or a high density gas, such as sulfur hexafluoride (SF6), to measure inhomogeneities in flow distribution (Scott and Van Liew, 1983). Large-scale spatial inhomogeneities in ventilation can be detected using radioactive xenon (Xe) and external g-emission imaging equipment such as the Anger camera (Robertson et al., 1969). Functional tests made before and after administration of bronchoactive agents can also be of diagnostic value. These can include bronchodilators such as isoproterenol, epinephrine, and atropine to measure reversible bronchoconstriction. They also include bronchoconstrictors, such as histamine, carbachol, methacholine, cold air, and SO2, to detect airway hyperresponsiveness. Other tests of demonstrated utility and/or with potential for supplying important diagnostic information can also be applied. The permeability of the respiratory epithelium can be determined from the externally measured rate of clearance from the lung of g-emitting [99mTc] tagged diethylenetriaminepentaacetate (DTPA), inhaled as a droplet aerosol (Oberdorster et al., 1986).

32

INTRODUCTION AND BACKGROUND

Inert, insoluble, nonhygroscopic, g-tagged aerosols can be used to measure the regional deposition and clearance rates for inhaled particles. Thoracic retention of such aerosols after 1 day is considered to represent deposition in the nonciliated alveolar lung spaces, while the difference between the retention measured immediately after the particle inhalation and that at 1 day represents the aerosol that deposited on the conductive airways of the tracheobronchial tree (Lippmann, 1977). By appropriate control of particle size and respiratory parameters, the deposition efficiency data can be used to characterize airway obstruction (Chan and Lippmann, 1980). The rate and pattern of mucociliary particle clearance can be determined from serial measurements of thoracic retention during the first day, and the much slower rate of particle clearance from the gas-exchange region can be determined from serial retention measurements made after the first day (Albert et al., 1969). More recently there have been studies in which cardiac function measurements, such as heart rate and heart rate variability, have been made for O3 (Gong et al., 2003; Brook et al., 2002; Urch et al., 2004), SO2 (Tunnicliffe et al., 2001), and concentrated PM2.5 (Brook et al., 2002; Urch et al., 2004; Devlin et al., 2003). The advantages of controlled human exposure studies are: (1) the opportunity to carefully select and carefully characterize the subjects, whether they be healthy normals, atopics, asthmatics, smokers, and so on; (2) the willingness and ability of most volunteer subjects to perform various levels and durations of exercise during the exposures; (3) the ability to deliver and monitor the preselected challenge atmospheres during the exposure; (4) the ability of the subjects to reproducibly perform respiratory maneuvers required for some functional assays affected by the exposures and to provide information on mild symptomatic responses; and (5) avoidance of the need to make interspecies extrapolations in evaluating human exposure–response relationships. The limitations of controlled human exposures are: (1) that ethical constraints limit the challenges and effects assays that can be performed. In effect, we are limited to challenges that produce only transient functional changes. The most invasive assay that has been used involves analyses of the contents of lung lavage; (2) the numbers of repetitive challenges and assays are limited by subject tolerance and cooperation; and (3) the number of subjects that can be studied is limited by the generally large costs of performing the studies and/or by the availability of sufficient numbers of subjects with the desired characteristics. In summary, controlled human exposure studies are most useful for studying the nature and extent of transient functional changes resulting from one or a few brief controlled exposures. They can provide information on chronic pollutant effects only to the extent that prior exposures affect the transient response to single exposure challenges. Furthermore, interpretation of the results of such tests is limited by our generally inadequate ability to characterize the nature and/or magnitude of the prior chronic exposures. 1.13.2

Natural Human Exposures

There is a substantial database emerging from studies of the responses of natural populations to acute exposures to air pollutants (Lippmann, 1989b; Spektor et al., 1991; Thurston et al., 1997). Studying natural populations for evidence of acute health effects associated with exposures to ambient air pollutant is a challenging task. Among the more difficult challenges are: (1) identifying an accessible population at risk whose relevant exposures can be defined and adequately characterized; (2) specifying measurable indices of responses that may be

STUDY OPTIONS FOR HEALTH EFFECTS STUDIES

33

expected to occur as a result of the exposures of interest; (3) collecting an adequate amount of suitable quality-assured data on exposure and responses at times when exposures of magnitudes sufficient to elicit measurable responses actually occur; and (4) collecting sufficient data on identifiable host characteristics and environmental exposures to other agents that may influence the response variables and confound any of the hypothesized pollutant exposure–response relationships that may be present. In addition, one must also account for the usual operational problems encountered in performing population studies, especially studies in the field, such as maintaining (1) the motivation and skills of the field personnel for collecting reliable data; (2) the cooperation of the subjects in producing reliable data; and (3) access to sufficient numbers of subjects with the preselected characteristics in each category as may be needed. The basic design premise in field studies involving air pollutant exposures is to maximize the signal-to-noise ratio for the pollutant exposure versus response relationships. The noise on the response side of the relationships has been the focus of much work by others, and guidance on these aspects is available from the American Thoracic Society (1985). Focus is also needed on the reduction of the noise in the exposure variables. For example, the summer pollution haze is regional in scale and enriched in secondary air pollutants such as O3 and H2SO4, both of which form gradually during daylight hours in air masses containing diluted primary pollutants transported over long distances from industrial, power plant, and motor vehicle sources, especially SO2, NO2, and HC). For the NYU field studies (Lippmann et al., 1983; Lioy et al., 1985; Spektor et al., 1988a, 1991; Thurston et al., 1997), populations of children attending summer camp programs were selected for three main reasons: (1) cigarette smoking and occupational exposure to lung irritants would not be confounding factors; (2) the program of camp activities insured that they would be out of doors and physically active during the daytime periods when O3 and H2SO4 exposures are highest; and (3) the cooperation of the camp staff provided effective access to the children on a daily basis for the administration of functional tests and symptom questionnaires. Exposures to O3 and H2SO4 are almost always higher outdoors than indoors, and, as regional-scale secondary pollutants, their concentrations do not vary greatly from site to site within the camp’s activity areas or from those measured at nearby samplers or monitors. In addition, there was little variation of activity level among the children in the camp program. The 1985 summer study (Spektor et al., 1988b) on the effects of the summer haze pollutants on respiratory function in healthy nonsmoking adults engaged in a regular program of outdoor exercise had a similar absence of confounding exposure factors as well as similar exposures to the ambient secondary air pollutants. Each of the adult volunteers maintained a constant daily level and duration of exercise, but they differed widely from one to another in these important variables. This increased the variability of the response among the population but also provided a means of studying the influence of these variables on the responses. In summary, natural human exposure studies are most useful for studying the magnitude and extent of the acute responses to naturally occurring pollutants among people engaged in normal outdoor recreational activities. They provide little information on the possible influence of prior chronic exposures on acute responses to the exposure of the day or immediately preceding days. Also, since the ambient mixture contains varying amounts of a variety of pollutants, it may sometimes be difficult to apportion the responses to one or more of the pollutants or to other, uncontrolled variables such as temperature, humidity, and each individual’s precise level of exercise or ventilation.

34

1.13.3

INTRODUCTION AND BACKGROUND

Population-Based Studies of Chronic Health Effects of Air Pollution

Since neither controlled human exposure studies in the laboratory nor natural human exposure studies in the field can provide any direct information on chronic effects of prolonged human exposures to air pollutants, the only way to get such information is to use the conventional epidemiological approach of comparing data on: (1) reductions in lifespan and function; and (2) increases in symptom frequency, lost activity days, hospital admissions, clinic visits, medical diagnoses, and so on, in relation to estimates of chronic exposure intensity. There are many confounding factors affecting the effects indices of concern in such studies. The characteristics of the populations under study are highly variable in terms of age, sex, smoking history, cohabitation with smokers, health status, disease history, occupational exposures, hobby activities that generate air pollutants, use of unvented stoves and heaters at home, and so on. Also, their exposures to outdoor pollutants are difficult to quantitate, and are influenced by their proximity to the monitor that provides their exposure index, the time they spend outdoors, and whether this includes hours when the pollutant is high as well as the amount and duration of vigorous exercise during periods of high exposure. Because of the large number of possible confounders and the difficulty of properly classifying exposures, very large populations must be studied in order to find significant associations between exposures and effects. Any statistically significant effects that are attributed to air pollution would tend to be underestimated because of the influence of the confounders. Alternatively, they could be spurious if the effects are really caused by variables that are colinear with the pollutant being studied. In summary, epidemiological studies offer the prospect of establishing chronic health effects of long-term air pollution exposure in relevant populations and offer the possibility that the analyses can show the influence of other environmental factors on responses to exposure. On the contrary, the strengths of any of the associations may be difficult to establish because of the complications introduced by uncontrolled cofactors that may confound or obscure the underlying causal factors.

1.13.4

Controlled Exposures of Laboratory Animals

The most convenient and efficient way to study mechanisms and patterns of response to pollutants, and of the influence of other pollutants and stresses on these responses is by controlled exposures of laboratory animals. One can study the transient functional responses to acute exposures and establish the differences in response among different animal species and between them and humans similarly exposed. One can also look for responses that require highly invasive procedures or serial sacrifice and gain information that cannot be obtained from studies on human volunteers. One can expose the animals to a single pollutant, or to a complex mixture as found in the environment, or from a specific source. Finally, one can use long-term exposure protocols to study both short-term and cumulative responses, and the pathogenesis of chronic disease in animals. Other advantages of studies on animals are the ability to examine the presence of and basis for variations in response that are related to age, sex, species, strain, genetic variations, nutrition, the presence of other pollutants, and so on. As in controlled human exposure studies, the concentrations and duration of the exposure can be tightly controlled, as can the presence or absence of other pollutants and environmental variables. Another important advantage of controlled animal studies is that relatively large numbers of individuals can be simultaneously exposed, creating the possibility of detecting responses that only affect a limited fraction of the population.

REFERENCES

35

Among the significant limitations to the use of exposure–response data from animal studies in human risk assessments is our quite limited ability to interpret the animal responses in relation to likely responses in humans who might be exposed to the same or lower levels. Controlled chronic exposure protocols can be very labor-intensive and expensive, which tends to limit the number of variables that can effectively be examined in any given study. 1.13.5

Controlled Exposures In Vitro

For studies focused on the biochemical mechanisms of epithelial cells’ responses to O3, cells can be harvested from humans or animals and exposed to O3 in vitro. Techniques have been developed for reasonably realistic O3 exposures to cells and cell cultures in vitro (Valentine, 1985) for characterizing the release of eicosanoids from such cells (Leikauf et al., 1988), and for examining cell function (Driscoll et al., 1987). The main advantage of in vitro studies is their efficiency and relatively low cost. Interspecies comparisons of cellular response can often be made, and relatively few animals can provide much study material. However, our ability to interpret the results of in vitro assays in relation to likely effects in humans cells. In vivo are of limited value, even when the studies are done with human cells. The cellular response in vitro may differ from that of the same cells in vivo, and the in vivo controls on cellular metabolism and function, which may play a significant role in the overall response, are absent.

REFERENCES ACGIH (2001) Air Sampling Instruments,9th edn. Cincinnati: American Conference of Governmental Industrial Hygienists. Albert RE, Berger J, Sanborn K, Lippmann M (1974) Effects of cigarette smoke components on bronchial clearance in the donkey. Arch. Environ. Health 29:99–106. Albert RE, Lippmann M, Briscoe, W (1969) The characteristics of bronchial clearance in humans and the effects of cigarette smoking. Arch. Environ. Health 18:738-755. Albert RE, Peterson HT Jr, Bohning DE, Lippmann M (1975) Short-term effects of cigarette smoking on bronchial clearance in humans. Arch. Environ. Health 30:361–367 American Thoracic Society (1985) Guidelines as to what constitutes an adverse respiratory health effect, with special reference to epidemiologic studies of air pollution. Am. Rev.Respir. Dis. 131:666-668. Andersen RL, Alden CL (1989) Risk assessment for nitrilotriacetic acid (NTA). In: Paustenbach D, editor. The Risk Assessment of Environmental and Human Health Hazards: A Textbook of Case Studies. New York: John Wiley & Sons, pp. 390–426. Birmingham DJ (1973) Occupational dermatoses: their recognition and control In: The Industrial Environment: Its Evaluation and Control. Washington DC: U.S. Department of Health, Education and Welfare. Bohning DE, Atkins HL, Cohn SH (1982) Long-term particle clearance in man: normal and impaireditor. Ann. Occup. Hyg. 26:259–271. Brook RD, Brook JR, Urch B, Vincent R, Rajagopalan S, Silverman F (2002) Inhalation of fine particle air pollution and ozone causes acute arterial vasoconstriction in healthy adults. Circulation 105:1534–1536. Buist AS, Ross BB (1973) Quantitative analysis of the alveolar plateau in the diagnosis of early airway obstruction. Am. Rev. Respir. Dis. 108:1078–1085.

36

INTRODUCTION AND BACKGROUND

Butterworth BE, Slaga T (1987) Nongenotoxic Mechanisms in Carcinogenesis (Banbury Report 25). New York: Cold Spring Harbor Press, pp. 555–558. Chan TL, Lippmann M (1980) Experimental measurements and empirical modelling of the regional deposition of inhaled particles in humans. Am. Ind. Hyg. Assoc. J. 41:399–409. Crapo RO (1986) Single breath carbon monoxide diffusing capacity. ATS News, spring. Devlin RB, Ghio AJ, Kehrl H, Sanders G, Cascio W (2003) Elderly humans exposed to concentrated air pollution particles have decreased heart rate variability. Eur. Respir. J. 40 (Suppl.):76s–80s. Driscoll KE, Vollmuth TA, Schlesinger RB. (1987) Acute and subchronic ozone inhalation in the rabbit: response of alveolar macrophages. J. Toxicol. Environ. Health 21:27–43. Ellwein LB, Cohen SM (1988) A cellular dynamic model of experimental bladder cancer: analysis of the effect of sodium saccharin. Risk Anal. 8:215–221. Ferin J, Leach LJ (1973) The effect of SO2 on lung clearance of TiO2 particles in rats. Am. Ind. Hyg. Assoc. J. 34:260–263. Friedlander SD (1977) Smoke, Dust and Haze. New York: John Wiley & Sons. Gibbs WE (1924) Clouds and Smoke. New York: Blakiston. Gong H Jr, Linn WS, Sioutas C, Terrell SL, Clark KW, Anderson KR, Terrell LL (2003) Controlled exposures of healthy and asthmatic volunteers to concentrated ambient fine particles in Los Angeles. Inhal. Toxicol. 15:305–325. Hammond PB (1969) Lead poisoning: an old problem with a new dimension. In: Blood FR,editor. Essays in Toxicology. New York: Academic Press. Hatch TF (1968) Significant dimensions of the dose–response relationship. Arch. Environ. Health 16:571–578. Hayes WJ Jr (1965) Review of metabolism of chlorinated hydrocarbon insecticides especially in mammals. Ann. Rev. Pharmacol. 5:27–52. International Commission on Radiological Protection (1966) Task Group on Lung Dynamics. Deposition and retention models for internal dosimetry of the human respiratory tract. Health Phys. 12:173. International Commission on Radiological Protection (1979) Limits for Intakes of Radionuclides by Workers. Part 1. New York: Pergamon. International Commission on Radiological Protection (1981) Limits for Intakes of Radionuclides by Workers. Part 3. New York: Pergamon. ICRP (1994) Human Respiratory Tract Model for Radiological Protection. ICRP Publ. #66. Annals ICRP; 24(Nos. 1–3). Oxford, U.K.: Elsevier. Jammet H, LaFuma J, Nenot JC, Chameaud M, Perreau M, LeBouffant M, Lefevre M, Martin M (1970) Lung clearance: silicosis and anthracosis. In: Shapiro HA, editor. Pneumoconiosis: Proceedings of the International Conference, Johannesburg 1969. Capetown: Oxford Press. Leikauf GD, Driscoll KE, Wey HE (1988) Ozone-induced augmentation of eicosanoid metabolism in epithelial cells from bovine trachea. Am. Rev. Respir. Dis. 137:435–442. Lioy PJ (1990) Assessing total human exposure to contaminants. Environ. Sci. Technol. 24:938–945. Lioy PJ, Vollmuth TA, Lippmann M (1985) Persistence of peak flow decrement in children following ozone exposures exceding the National Ambient Air Quality Standard. J. Air Pollut. Control Assoc. 35:1068–1071. Lippmann M (1977) Regional deposition of particles in the human respiratory tract. In: Lee DHK, Falk HL, Murphy SD, editors. Handbook of Physiology, Section 9, Reactions to Environmental Agents. Bethesda: American Physiological Society. Lippmann M (1989) Effective strategies for population studies of acute air pollution health effects. Environ. Health Perspect. 81:115–119.

REFERENCES

37

Lippmann M, Schlesinger RB (1984) Interspecies comparison of particle deposition and mucociliary clearance in tracheobronchial airways. J. Toxicol. Environ. Health 13:441–469. Lippmann M, Schlesinger RB, Leikauf G, Spektor D, Albert RE (1982) Effects of sulphuric acid aerosols on the respiratory tract airways. Ann. Occup. Hyg. 26:677–690. Lippmann M, Lioy PJ, Leikauf G, Green KB, Baxter D, Morandi M, Pasternack B, Fife D, Speizer FE (1983). Effects of ozone on the pulmonary function of children. Adv. Modern Environ. Toxicol. 5:423–446. Martonen TB (1990) Acid aerosol deposition in the developing human lung In: Masuda S, Takahashi K, editors. Aerosols-Science,Industry,HealthandEnvironment,Vol.2.Oxford:Pergamon, pp.1287–1291. Moolgavkar SH (1978) The Multistage theory of carcinogenesis and the age distribution of cancer in man. J. Natl. Cancer Inst. 61:49–52. Moolgavkar SH, Dewanji A, Venzon DJ (1988) A stochastic two-stage model for cancer risk assessment: the hazard function and the probability of tumor. Risk Anal. 8(3):383–392. Morgan KT, Patterson DL, Gross EA (1984) Frog palate mucociliary apparatus: structure, function, and response to formaldehyde gas. Fundam. Appl. Toxicol. 4:58–68. Munro IC, Krewski DR (1981) Risk assessment and regulatory decision-making. Food Cosmet. Toxicol. 19:549–560. National Academy of Science (1981) Indoor Pollutants. Washington DC: National Academy Press. NCRP (1997) Deposition, retention and dosimetry of inhaled radioactive substances. NCRP Report No. 125, National Council on Radiation Protection and Measurements, Bethesda, MD 20814-3095. NIEHS (1977) Second task force for research planning in environmental health science. Health and the Environment-Some Research Needitors. DHEW Publ. #NIH77-1277. National Research Council (1979) Airborne Particles. Baltimore: University Park Press. National Research Council (1980) The Effects on Populations of Exposure to Low Levels of Ionizing Radiation. Washington, DC: National Academy Press. 21–23 National Research Council (1991) Human Exposure Assessment for Airborne Pollutants. Washington, DC: National Academy Press. Oberdorster G, Utell MJ, Morrow PE, Hyde RW, Weber DA (1986) Bronchial and alveolar absorption of inhaled 99mTc-DTPA. Am. Rev. Respir. Dis. 134944–950. Paustenbach DJ (1990) Health risk assessment and the practice of industrial hygiene. Am. Ind. Hyg. Assoc. J. 51:339–351. Paynter OE, Burin GJ, Gregorio CA (1988) Goitrogens and thyroid follicular cell neoplasia: evidence for a threshold process. Regul. Toxicol. Pharmacol. 8:102–119. Phalen RF, Kenoyer JL, Crocker TT, McClure TR (1980) Effects of sulfate aerosols in combination with ozone on elimination of tracer particles inhaled by rats. J. Toxicol. Environ. Health 6: 797–810. Robertson PC, Anthonisen NR, Ross D (1969) Effect of inspiratory flow rate on regional distribution of inspired gas. J. Appl. Physiol. 26:438–443. Schlesinger RB, Driscoll KE (1987) Mucociliary clearance from the lungs of rabbits following single and intermittent exposures to ozone. J. Toxicol. Environ. Health 20:120–134. Schlesinger RB, Naumann BD, Chen LC (1983) Physiological and histological alterations in the bronchial mucociliary clearance system of rabbits following intermittent oral or nasal inhalation of sulfuric acid mist. J. Toxicol. Environ. Health 12:441–465. Schlesinger RB, Driscoll KE, Naumann BD, Vollmuth TA (1988) Particle clearance from the lungs: assessment of effects due to inhaled irritants. Ann. Occup. Hyg. 32 (S1):113–123. Schwartz EW (1923) The so-called habituation to “arsenic”: variation in the toxicity of arsenious oxide. J. Pharmacol. Exp. Ther. 20:181–203.

38

INTRODUCTION AND BACKGROUND

Scott WR, Van Liew HD (1983) Measurement of lung emptying patterns during slow exhalations. J. Appl. Physiol. 55:1818–1824. Sexton K, Ryan PB (1988) Assessment of human exposures to air pollution: methods, measurements and models. In: The Automobile and Public Health. Washington, DC: National Academy Press. Sobel AE, Gawson O, Kramer B (1938) Influence of vitamin D in experimental lead poisoning. Proc. Soc. Exp. Biol. Med. 38:433–437. Spektor DM, Lippmann M, Lioy PJ, Thurston GD, Citak K, James DJ, Bock N, Speizer FE, Hayes, C. 1988a. Effects of ambient ozone on respiratory function in active normal children. Am. Rev. Respir. Dis. 137:313–320. Spektor DM, Lippmann M, Thurston GD, Lioy PJ, Stecko J, O’Connor G, Garshick E, Speizer FE, Hayes C (1988b) Effects of ambient ozone on respiratory function in healthy adults exercising outdoors. Am. Rev. Respir. Dis. 138:821–828. Spengler JD, Stone KR, Lilley FW (1978) High carbon monoxide levels measured in enclosed skating rinks. J. Air Pollut. Control Assoc. 28:776–779. Squire RA (1987) Ranking animal carcinogens: a proposed regulatory approach. Science 214: 877–880. Thurston GD, Lippmann M, Scott MB, Fine JM (1997) Summertime haze air pollution and children with asthma. Am. J. Respir. Crit. Care Med. 155:654–660. Tunnicliffe WS , et al. (2001) The effect of sulphur dioxide on indices of heart rate variability in normal and asthmatic adults. European Respiratory Journal 17:604–608. Urch B, Brook JR, Wasserstein D, Brook RD, Rajagopalan S, Corey P, Silverman F (2004) Relative contributions of PM2.5 chemical constituents to acute arterial vasoconstriction in humans. Inhal. Toxicol. 16:345–352. U.S. EPA (1988) A cancer risk-specific dose estimate for 2, 3, 7, 8 TCDD (EPA/600/6-88/007), U.S. Environmental Protection Agency, Washington, DC: Office of Health and Environmental Assessment, pp. 1–31. U.S. EPA (1992) Guidelines for exposure assessment. EPA/600/Z-92-001, Washington, DC: Risk Assessment Forum, U.S. Environmental Protection Agency, May 29, 1992. U.S. EPA (1996) Air quality criteria for particulate matter. EPA/600/P-95/001F. U.S. Environmental Protection Agency, Washington, DC 20460. U.S. EPA (2007) Air quality criteria for ozone and related photochemical oxidants. EPA/600/R-05/ 0042007. U.S. Environmental Protection Agency, Washington, DC. Valentine R (1985) An in vitro system for exposure of lung cells to gases: effects of ozone on rat macrophages. J. Toxicol. Environ. Health 16:115–126. Vincent JH (1999) Particle Size-Selective Sampling for Particulate Air Contaminants. Cincinnati: American Conference of Governmental Industrial Hygienists. Wallace LA, Pellizzari ED, Gordon SM (1985) Organic chemicals in indoor air: a review of human exposure studies and indoor air quality studies. In: Gammage RB, Kaye SV, editors. Indoor Air and Human Health. Chelsea, MI: Lewis. Wolff RK, Dolovich M, Obminski G, Newhouse MT (1977) Effect of sulfur dioxide on tracheobronchial clearance at rest and during exercise. In: Walton WH, editor. Inhaled Particles IV. Oxford: Pergamon.

2 PERSPECTIVES ON INDIVIDUAL AND COMMUNITY RISKS Arthur C. Upton

Humankind has always faced certain dangers. Risk is an inescapable fact of life. However, enlightened societies have traditionally sought to minimize avoidable risks. The greatly increased life expectancy now enjoyed by populations in the industrialized world attests to the success with which modern civilization has been able to reduce certain risks to human health and safety. This chapter reviews environmental risks to human health from two standpoints: the risk to the individual and the risk to the community. Considered in this context are scientific bases for assessing such risks, the relative magnitudes of different environmental risks as evaluated by knowledgeable experts, the contrasting perspectives in which risks may be perceived by different members of the public, difficulties in risk communication that complicate societal efforts to protect health and the environment, and options for reducing risks at the individual and community levels.

2.1 NATURE OF RISK 2.1.1

Probability of Effect

Risk is commonly defined as “hazard, peril, or exposure to loss or injury.” The term environmental risk to health is taken herein to mean the probability of an adverse effect on human health resulting from exposure to a particular environmental agent or combination of agents. Such a risk may be expressed in various ways, depending on the context in which it is considered, for example: (1) average annual risk per individual; (2) average lifetime risk per individual; (3) average number of individuals affected annually in a given population; or (4) average loss of life expectancy in affected individuals. Apart from risks

Environmental Toxicants, Third Edition Edited by Morton Lippmann Copyright
2009 John Wiley & Sons, Inc.

39

40

PERSPECTIVES ON INDIVIDUAL AND COMMUNITY RISKS

TABLE 2.1 Different Ways in Which the Carcinogenic Effects of 1 Sv Acute Whole Body Ionizing Radiation may be Expressed in the Individual and in the Community Type of Cancer Fatal Leukemia Risk to a population of 100,000 persons Annual number of excess deaths Lifetime number of excess deaths Cumulative person-years of life lost Risk to the individual Increase in relative risk (%) Excess annual risk of death (10 4) Excess lifetime risk of death (10 3) Attributable risk (%) Years of life lost per attributable death Years of life lost per exposed person

9 600 12,000 100 1 6 50 20 0.3

a

Other Fatal Cancera 70 5,000 75,000 25 7 12 20 15 0.8

Source: UNSCEAR (2000) and NAS (2005). a

Values rounded.

to human health and safety per se, such risks may also be expressed in terms of their economic impacts on the affected individuals, their families, their associates, and their communities, as noted below. For certain types of health effects, such as pollutant-induced cancers and mortality from cardiovascular disease, the risk may also be expressed either as an absolute risk (i.e., absolute increase in the number or probability of such disorders) or as a relative risk (i.e., a relative increase in the background frequency of such disorders). Depending on the baseline frequency, or background rate, a small increase in relative risk may be equivalent to a large increase in the number of individuals affected. Conversely, merely a few additional cases of an otherwise rare disorder may result in a large increase in the relative risk, and thus in a high attributable risk, of the disorder (Table 2.1). 2.1.2

Severity of Effect

The importance attached to a given risk depends on the severity as well as the frequency of the effect in question. Determinants of severity include such factors as the extent to which the effect is or is not symptomatic, painful, disfiguring, incapacitating, reversible, progressive, lethal, and so on, these being the properties that determine its impact on the affected individual and on his or her loved ones, descendants, coworkers, neighbors, and community. In the broadest context, the measures of severity therefore have many ramifications, including esthetic, psychosocial, ethical, and economic impacts, as well as impacts on health per se (Hammond and Coppock, 1990; Arrow et al., 1996).

2.1.3

Psychosocial and Cultural Factors Influencing the Perception of Risk

Apart from objective measures of the frequency and severity of environmental risks to health, other qualitative characteristics, such as those listed in Table 2.2, can be important in

IDENTIFICATION AND QUANTIFICATION OF RISKS

41

TABLE 2.2 Psychosocial and Cultural Characteristics Affecting the Perception of a Risk Characteristics of a Risk that Increase its Acceptability

Characteristics of a Risk that Decrease its Acceptability

Voluntary Familiar Immediate impact Detectable by individual Controllable by individual Fair Noncatastrophic Well understood Natural Trusted source Visible benefits Well defined Predictable

Involuntary Unfamiliar Remote impact Undetectable by individual Uncontrollable by individual Unfair Catastrophic Poorly understood Artificial Untrusted source No visible benefits Poorly defined Unpredictable

Source: Adapted from Slovic et al. (1979) and Plough and Krimsky (1987).

Technical Risk

=

risk

Nontechnical +

assessment

public concerns

or Risk

FIGURE 2.1

=

Hazard

+

Outrage

Holistic definition of risk. Adapted from Allen (1992) and Sandman (1985).

determining how the risks are perceived (Raynor and Cantor, 1987; Fischoff et al., 1997; Omenn and Faustman, 1997). Nonscientists often not only fail to understand the technical basis for evaluating a given risk but actually also distrust and reject it (Plough and Krimsky, 1987; Fischoff et al., 1997). It has been suggested, therefore, that the definition of an environmental risk needs to be broadened to include its nontechnical aspects that may be of concern to the public; that is, its “outrage” factors (Fig. 2.1). The importance of such nontechnical factors is illustrated by the marked degree to which public perceptions of a given risk may differ from those of informed experts (Table 2.3).

2.2 IDENTIFICATION AND QUANTIFICATION OF RISKS Assessment of an environmental risk to human health involves a sequence of interrelated steps (Omenn and Faustman, 1997), beginning with identification of the causative agent or exposure situation, and culminating in an evaluation of the number of persons affected and the severity of their effects (Fig. 2.2).

42

PERSPECTIVES ON INDIVIDUAL AND COMMUNITY RISKS

TABLE 2.3 “Perceived” Compared with “Real” Risks Associated with Various Widespread Activities and Technologies

Activity or Technology Smoking Alcoholic beverages Motor vehicles Handguns Electric power Motorcycles Swimming Surgery X-rays Railroads General (private) aviation Large construction Bicycles Hunting Home appliances Fire fighting Police work Contraceptives Commercial aviation Nuclear power Mountain climbing Power mowers School and college football Skiing Vaccinations

Geometric Mean Fatality Estimates, Average Year

Technical Estimates (deaths/year)a

LOWVb

Studentsc

150,000 100,000 50,000 17,000 14,000 3,000 3,000 2,800 2,300 1,950 1,300 1,000 1,000 800 200 195 160 150 130 100d 30 24 23 18 10

6,900 12,000 28,000 3,000 660 1,600 930 2,500 90 190 550 400 910 380 200 220 460 180 280 20 50 40 39 55 65

2,400 2,600 10,500 1,900 500 1,600 370 900 40 210 650 370 420 410 390 390 390 120 650 27 70 33 40 72 52

Source: Slovic et al. (1979). a

Based on assessments by technical experts. League of Women Voters. c College students. d Geometric mean of estimates, which ranged from 16 to 600 per year. b

2.2.1

Hazard Identification

The first step of identification and quantification of risks, hazard identification, consists of identifying potentiallyharmful agents towhich persons may be exposed, regardless of the level of exposure. For this purpose, reliance has traditionally been placed primarily on clinical and epidemiological evidence. For most environmental agents of interest, however, toxicity to humans cannot be evaluated adequately from the limited clinical and epidemiologicaldata that are available (NAS, 1984, 1994). Instead, the evaluation must depend on other approaches, including systematicanalysisofpertinentmolecularstructure–activityrelationships,resultsof in vitro short-term tests, and biological activity in short-term or long-term whole-animal bioassays (e.g., Tennant et al., 1987; Ashby and Tennant, 1988; ICPEMC, 1988; Omenn and Faustman, 1997). Principles and procedures for utilizing such methods in predicting toxicityto

IDENTIFICATION AND QUANTIFICATION OF RISKS

RISK ASSESSMENT

RESEARCH Liboratory and field abservations of adverse health effects and exposures to particular agents

Information on extrapolation methods for high to low dose and animal to human

Field measuremements, estlmatedexposures, characterization of populations

FIGURE 2.2 permission).

RISK MANAGEMENT Development of regulatory options

Hazard Identification (Does the agent cause the adverse effect?)

Evaluation of public health, economic, social, political consequences of regulatory options

Dose Response Assessment (What is the relationship between dose and inc idence in humans?)

Exposure Assessment (What exposures are currently expirienced of different conditions?)

43

Risk Characterlzation (What is the estimated incidence of the adverse effectin a given population?)

Agency decisions and actions

Risk assessment relies on evaluation techniques. From NAS (1994), reproduced with

humans have been developed, but the diversity of toxic reactions caused by different agents is so large, and the variations in reactivity among different species so great, that the reliability of this approach is limited (e.g., Lave et al., 1988; NAS, 1994). For most of the many thousands of chemicals in commercial production, moreover, the available toxicological data do not suffice to enable adequate evaluation (NAS, 1984, 1994; U.S. EPA, 1987, 1990). 2.2.2

Dose–Response Analysis

In the second step, dose–response analysis, the mathematical relationship between the dose of the agent of interest and any health effects that it may cause is evaluated in order to estimate the nature and magnitude of risks attributable to the agent at the levels of exposure encountered in practice. Since ambient exposure levels are typically many times lower than the levels at which any toxic effects may have been documented previously, formulation of the desired risk estimate often requires extrapolation over a broad range of doses and/or animal species, necessitating the use of a dose–response model that may be of uncertain validity (Omenn and Faustman, 1997). Although thresholds in individual humans and laboratory animals are known to exist for many, if not most, types of toxic reactions, no threshold is known or presumed to exist for the mutagenic and carcinogenic effects of ionizing radiation (UNSCEAR, 2000; NCRP, 2001; NAS, 2005), certain other toxicants (OSTP, 1985), and for populations exposed to lead, ozone, and fine particulate matter (PM) (see Chapters 23, 27, and 13). Assessing the risks of such agents, therefore, requires the use of an appropriate dose–response model, the selection of which is fraught with uncertainty (NAS, 1994). Again, the problem is complicated by a paucity of relevant dose–response data. Even in the relatively few instances where human data are available to provide anchor points from which to extrapolate, the data do not suffice

44

PERSPECTIVES ON INDIVIDUAL AND COMMUNITY RISKS

to define the dose–response relationship in the low-dose domain. In the use of dose–response data from laboratory animals, moreover, there is uncertainty both about the choice of the model for extrapolation to low doses and of the model for extrapolation to the human species (Zeise et al., 1987; Lave et al., 1988; NAS, 2005). The problem is also complicated by the fact that the dose–response relationships for many toxicants are biphasic, characterized by effects that are inhibitory or toxic at intermediate-to-high dose levels but stimulatory at lower dose levels. This phenomenon, known as “hormesis,” has been interpreted by some observers to imply that exposure to such toxicants at low dose levels may be beneficial rather than harmful (Calabrese et al., 1999); the suggestion that this may be the case with ionizing radiation (Cohen, 2002), however, has met with controversy (NAS, 2005). Another major source of uncertainty stems from the fact that in the human environment, agents are usually, if not always, encountered in combination with untold numbers of other agents, rather than in the pure form in which they have been studied in most clinical or toxicological experiments. Because synergistic or other complex interactions among agents may occur under such conditions, the combined effects of mixtures of agents can seldom be confidently predicted from what is known about the toxicological effects of any given agent acting alone (NAS, 1988; Mauderly, 1993). Furthermore, because individuals vary in susceptibility on the basis of differences in genetic background, age, gender, physiological state, diet, lifestyle, health habits, histories of smoking and occupational exposures, and other variables, a risk assessment that is based on the distribution of a risk within a particular population cannot be assumed to apply to any given individual in the population or to any other population as a whole (NAS, 1994, 2005; Aardema and MacGregor, 2002). 2.2.3

Exposure Assessment

The third step, exposure assessment, consists in evaluating the extent to which persons are, or are likely to be, exposed to a particular environmental agent or combination of agents. For the most part, assessments of exposure have relied thus far largely on data from nonvalidated exposure models or from the monitoring of the regulated exposure media (air, water, soil, food, etc.). Monitoring of human beings themselves has been limited, in part because of the lack of suitably sensitive, reliable, and practicable methods and measures of exposure. Recent advances in analytical techniques, however, and in the development of molecular biomarkers, such as DNA adducts (NAS, 1987), give promise of future improvements in this area (Brandt-Rauf, 1997). 2.2.4

Risk Characterization

In the fourth step, risk characterization, the information generated in the first three steps is integrated to derive an estimate of the numbers of persons who may be affected and the types and severities of their effects. To the extent that the information obtained in each of the preceding steps is constrained by uncertainty, the final characterization of risk derived in the fourth step will, of course, be constrained correspondingly. Because of the complexity, data requirements, and cost of each step in the process, as well as the uncertainties inherent therein, detailed and comprehensive attempts at risk characterization have been made for relatively few environmental problems to date (e.g., Ames et al, 1987; U.S. EPA, 1987, 1996, 2005a; NAS, 1994). Examples illustrating some of the uncertainties involved in such assessments are shown in Tables 2.4 and 2.5.

TABLE 2.4 Cancer Risk Rankings Assigned to Various Environmental Sources in EPA’s “Unfinished Business” Report Environmental Source

Rank Order

Category 1 (high risk) Exposure of workers to chemicals Indoor radon

1 1 (tied)

Pesticide residues in foods

3

Indoor air pollution (non-radon)

4

Exposure to consumer products

4 (tied)

Other hazardous air pollutants

6

Category 2 (medium to high risk) Depletion of stratospheric ozone Hazardous waste sites (inactive) Pollutants in drinking water Application of pesticides

7 8 9 10

Ionizing radiation

11

Other exposures to pesticides Hazardous waste sites (active)

12 13

New toxic chemicals

15

Category 3 (low to medium risk) Municipal waste

16

Contaminated sludge Mining waste

17 18

Storage tank releases Nonpoint source discharges to surface water Other groundwater contamination Criteria air pollutants

19 20

Category 4 (low risk) Direct point discharges to surface water Indirect point discharges to surface water Accidental release––toxicants Accidental release––oil spills Category 5 (unranked) Biotechnology CO2 and global warming Other air pollutants

21 22

23

Estimated Yearly Number of Cases of Cancer in the U.S. Population 250, from only four of the many carcinogens in question; risks to individuals may be high 5,000–20,000 (of lung); risks to individuals may be high 6,000, based on assessment of only seven of 200 potentially carcinogenic pesticides 3,500–6,500 (primarily from tobacco smoke); risks to individuals may be high 100–135, from only four of the more than 10,000 chemicals in consumer products 2,000, from only 20 of the many pollutants in air; risks to individuals may be high Possibly 10,000 annually by the year 2100 More than 1,000 400–1,000 100 in the small population exposed; risk to individuals may be high 360 (largely from building materials); risks to individuals may be high 150 (estimate highly uncertain) Probably fewer than 100; risks to individuals can be high No quantitative possible; risks judged to be moderate 40 (excluding municipal surface impoundments) 40 (mostly from incineration and sludge) 10–20 (largely from arsenic); risks to individuals can be high Less than one No quantitative estimate, but judged to be the most serious surface water category Less than one Carcinogenicity questionable but exposure extensive

24 25 26

No quantitative estimate (excluding drinking water) No quantitative estimate No quantitative estimate No quantitative estimate

– – –

No estimate No estimate No estimate

Source: U.S. EPA (1987).

45

46

PERSPECTIVES ON INDIVIDUAL AND COMMUNITY RISKS

TABLE 2.5 Estimated Risks of Human Bladder Cancer from Daily Ingestion of 0.12 g Saccharin, Based on Extrapolation from Oncogenic Effects Observed in the Rat Method of High-to-Low-Dose Extrapolation

Lifetime Cases per Million Exposed

Rat dose adjusted to human dose by surface area rule Single-hit model Multistage model (with quadratic term) Multihit model Mantel-Bryan probit model

1200 5 0.001 450

Rat dose adjusted to human dose by milligram chemical per kilogram body weight per day equivalence Single-hit model Multihit model Mantel-Bryan probit model

200 0.001 21

Rat dose adjusted to human dose by milligram chemical per kilogram Body weight per lifetime equivalence Single-hit model Multihit model Mantel-Bryan probit model

5200 0.001 4200

Source: NAS (1978).

2.3 RISK COMMUNICATION 2.3.1

Bridging Different Cultures

People respond to risks as they perceive them (Fischoff et al., 1997). Therefore, since experts trained and experienced in evaluating risks often fail to communicate their assessments adequately to the public, efforts to protect health and the environment are sometimes misdirected (NAS, 1996). As noted earlier, perceptions of risk involve nontechnical considerations (“cultural rationality”) as well as technical considerations (“technical rationality”). Hence, in order to communicate effectively about the nature and magnitude of a given risk, both types of considerations (Table 2.6) must be taken into account. Because the nontechnical considerations are rooted in cultural, anthropological, and ethical traditions, they vary among different groups in society (Fischoff et al., 1997; NAS, 1989). For communicating a given risk to all audiences, no single message is therefore likely to be adequate. Furthermore, for optimal effectiveness in risk communication, the process should involve a two-way, iterative exchange of information between the technical risk assessor and any stakeholders who may be directly or indirectly affected. Ideally such an exchange should begin as early as possible in the process of risk assessment so that those who may bear the risks can participate fully in the derivation of the assessment itself. Trust between all who are involved is also critical to the success of risk communication, and it is best fostered through the mutual exchange of information in an open, participatory, consensual process (Sandman, 1985; NAS, 1996; Slovic, 1998).

RISK COMMUNICATION

47

TABLE 2.6 Comparison of Factors Relevant to the Cultural Rationality, as Opposed to the Technical Rationality, of Risk Technical Rationality Trust in scientific methods, explanations, democratic process Appeal to authority and expertise Boundaries of analysis are narrow and reductionist Risks are depersonalized Emphasis on statistical variation and probability Appeal to consistency and universality Where there is controversy in science, resolution follows expertise, status Those impacts that cannot be measured are less relevant

Cultural Rationality Trust in political culture and evidence Appeal to folk wisdom, peer groups, and traditions Boundaries of analysis are broad and include the use of analogy and historical precedent Risks are personalized Emphasis on the impacts of risk on the family and community Focus on particularity; less concerned about consistency of approach Popular responses to science; the differences do not follow the prestige principle Unanticipated or unarticulated risks are relevant

Source: Plough and Krimsky (1987).

2.3.2

Treatment of Uncertainty

In contrast to other risks in daily life (e.g., various types of accidents whose frequencies are well documented in recorded statistics), many environmental risks to health are not known precisely and can be estimated only on the basis of unproved assumptions and extrapolations. As noted earlier, such assessments are complicated at virtually every step by large uncertainties in (1) the numerical values of measurements or other quantities affecting the risks; (2) the modeling of exposure and/or toxic responses; (3) temporal, spatial, and interindividual differences in exposure and/or susceptibility; and (4) the comparison of societal and personal measures of risk. To the extent that these sources of uncertainty limit the reliability of a risk assessment, each must be made explicit if the comprehensibility, credibility, and utility of the assessment are not to be jeopardized (Finkel, 1990; Morgan and Hendon, 1990; NAS, 1994). Exemplifying the kinds of difficulties that can complicate the treatment of uncertainty in a risk assessment are the adverse reactions (e.g., New York Times, 2005) encountered by the Environmental Protection Agency in its efforts to regulate the risks that a projected high-level nuclear waste repository might pose to generations ten thousand to one million years in the future (U.S. EPA, 2005b). 2.3.3

Placing Risks in Proper Perspective

The perception of risk is a complex process with the result that risk is difficult to communicate in a way that places it in proper perspective (e.g., NAS, 1989; Zeckhauser and Viscusi, 1990; Fischoff et al., 1997). Comparisons of quantitative risk estimates, such as have often been presented for the purpose (e.g., Peto et al, 1984; Tables 2.7 and 2.8), or attempts to weight risks solely on the basis of their impacts on life expectancy (e.g., ICRP, 1977, 1991), on the quality of life (e.g., ICRP, 1991), or on their economic costs (e.g., Arrow et al., 1996) are likely to be

48

PERSPECTIVES ON INDIVIDUAL AND COMMUNITY RISKS

TABLE 2.7 of Death

Situations and Activities Involving an Estimated One-in-a-Million Risk

Exposure or Activity

Cause of Death

Smoking 1.4 cigarettes Drinking 1.5 L of wine Spending 1 h in a coal mine Spending 3 h in a coal mine Living 2 days in New York or Boston Traveling 6 min by canoe Traveling 10 miles by bicycle Traveling 300 miles by car Flying 1000 miles by jet Flying 6000 miles by jet Living 2 months in Denver Living 2 months in an average masonry building One chest X-ray Living 2 months with a cigarette smoker Eating 40 tablespoons of peanut butter Drinking Miami drinking water for 1 year Drinking thirty 12-oz cans of diet soda Living 5 years at the boundary of a nuclear plant Eating 100 charcoal-broiled steaks

Cancer, heart disease Cirrhosis of the liver Black lung disease Accident Air pollution Accident Accident Accident Accident Cancer caused by cosmic radiation Cancer caused by cosmic radiation Cancer caused by natural radioactivity Cancer caused by radiation Cancer, heart disease Liver cancer caused by aflatoxin B1 Cancer caused by chloroform Cancer caused by saccharin Cancer caused by radiation Cancer caused by benzopyrene

Source: Wilson (1979).

TABLE 2.8 Risks of Some Scenarios Involving Exposure to Possible Carcinogens Ranked on the Basis of Their Potencies as Estimated from Experimental Data Hazard Index: HERP (%)a Environmental pollution 2.1 0.004 0.001 Pesticide and other residues 0.0004 0.0003 Natural pesticides and dietary 4.7 0.1 0.3

Daily Exposure Level

Carcinogen Dose per 70-kg Person

Mobile home air (14 h/day) Well water, 1 L contaminated Tap water, 1 L

Formaldehyde, 2.2 mg Trichloroethylene 280 mg Chloroform 83 mg (U.S. Avg.)

EDB: daily dietary intake (grains) DDE/DT daily dietary intake

Ethylene dibromide 0.42 mg (U.S. Avg.) DDE 2.2 mg (U.S. Avg.)

toxins Wine (250 mL) Basil (1 g of dried leaf) Bacon, cooked (100 g)

Ethyl alcohol 30 mL Estragole 3.8 mg Dimethylnitrosamine 0.3 mg

Food additives 0.06

Diet cola

Saccharin 95 mg

Drugs 17 16

Clofibrate (Avg. daily dose) Phenobarbital, one sleeping pill

Clofibrate 2000 mg Phenobarbital 60 mg

Source: Ames et al. (1987). a

HERP stands for the “human exposure/rodent potency” index, a measure of the possible cancer risk to humans from a given exposure scenario, based on data derived from toxicological experiments with rodents.

RISK REDUCTION

49

inadequate by themselves (Slovic et al., 1981; NAS, 1989). Instead, the strategy for risk communication must take into account the known dynamics of risk perception, which involve the following principles: (1) unfamiliar risks tend to be less acceptable than familiar risks, (2) involuntary risks are less acceptable than voluntary risks, (3) risks controlled by others are less acceptable than risks that are under one’s own control, (4) nonapparent and undetectable risks are less acceptable than risks that are apparent and detectable, (5) risks that are perceived to be unfair are less acceptable than risks that are perceived to be fair, (6) risks that do not permit individual protective action are less acceptable than risks that do, (7) dramatic and dreadful risks are less acceptable than nondramatic and commonplace risks, (8) unpredictable risks are less acceptable than predictable risks, (9) cross-hazard comparisons tend to be unacceptable, and (10) risk estimation is inherently of less interest to people than risk reduction, and neither is likely to be of interest in the absence of real concern about the risk, or risks, in question (Sandman, 1985; Fischoff et al., 1997).

2.4 RISK REDUCTION While the acceptability of a given risk may vary widely among different individuals, for the reasons stated above, regulatory agencies are guided by the prevailing views of society at large in setting standards to protect human health (Rodricks, 1992). Under the Clean Air Act, for example, EPA sought to limit the permissible levels of carcinogenic contaminants in air sufficiently to prevent the attributable lifetime risks of cancer in members of the public from exceeding 10 6, a level of risk that was judged to be acceptably protective for the purpose (Travis, 1989). It is clear, however, that the level of risk that is judged to be acceptable has varied among different carcinogens, depending on specific circumstances (Fig. 2.3). In the case of residential radon, for example, the EPA has recommended remedial measures when its concentration in indoor air exceeds 4pCi/L, a level which is about three-times higher than the average in U.S. homes, and which is estimated to pose a lifetime risk of lung cancer of about 1.4  10 2 (U.S. EPA, 2005). For occupationally exposed workers, moreover, whose choice of employment is more or less voluntary, the attributable risks implied by existing regulatory limits for carcinogens are appreciably higher than those which are considered acceptable for members of the public at large (e.g., Fig. 2.3). Measures for reducing environmental risks may involve substantial economic costs and/ or the substitution of other undesirable risks. For this reason, as discussed in the next chapter, the relative costs and benefits of alternate risk management strategies must be compared with one another to decide on the most appropriate approach for reducing a given risk (Davies, 1996). Included among the criteria to be considered in such comparative risk assessments are both the risk to society and the risk to the individual (Evans and Verlander, 1997), since the acceptability of a given risk generally varies inversely with the number of individuals who may be affected (e.g., Fig. 2.4). 2.4.1

Options for Risk Reduction at the Individual and Community Levels

Options for reducing environmental risks to human health may include measures for intervening at any point in the sequence of steps typically involved in the process by which a potentially hazardous agent is produced, released, transported through the environment, reaches a susceptible individual, taken up by the individual, and subsequently gives rise to a reaction adversely affecting the health of the individual and/or his/her offspring (Fig. 2.5).

50

PERSPECTIVES ON INDIVIDUAL AND COMMUNITY RISKS

FIGURE 2.3 The level of risk judged to be acceptable varies among carcinogens. From Travis et al. (1989), reproduced with permission).

Some options are not possible without action at the national or community level, whereas others lie within the power of the individual acting alone. All options, however, depend to varying degrees on understanding each of the risks in question and on having the skills needed to reduce them. Research and education in the relevant aspects of environmental health and safety are, therefore, essential for arriving at sound policies for risk reduction at the individual and the community levels (Tones, 1997). Also basic to sound policies is the precautionary principle, which holds that if there are serious doubts about the safety of a particular item or activity, it should not be accepted unless it can be expected to provide a benefit that greatly exceeds any reasonable estimate of the associated risks of harm. At the community level, the options for risk reduction encompass a broad range of activities including (1) support of research for identifying potentially hazardous agents, elucidating their toxicity and modes of action, and defining their relevant dose–effect relationships; (2) systematic monitoring of the extent to which individuals or populations may be exposed to harmful agents via air, water, soil, food, or other media, and proper assessment of the magnitude of any risks that may result from such exposures; (3) identification of individuals or groups at unusually increased risk because of heightened susceptibility and/or level of exposure; (4) formulation and enforcement of standards and regulations for limiting the exposure of individuals or populations to potentially harmful agents, along with engineering measures for controlling the production and/or release of potentially

RISK REDUCTION

51

FIGURE 2.4 The level of risk judged to be acceptable varies inversely with the number of victims affected. From Channel Tunnel Safety Study, Eurotunnel, (1994).

toxic agents, as discussed in the next chapter; (5) planning ahead to cope with emergencies that may result from the accidental release of hazardous agents; (6) maintaining in readiness the organizational capability needed to cope with environmental emergencies; (7) mounting programs of public and professional education in environmental risk reduction (including information clearinghouses, workshops, telephone hot lines, internet web pages, etc.)

FIGURE 2.5 Options for reducing environmental risks to human health. From Andrews and Turner, (1984), reproduced with permission.

52

PERSPECTIVES ON INDIVIDUAL AND COMMUNITY RISKS

(U.S. EPA, 1990; Griffith and Saunders, 1997; Omenn and Faustman, 1997; Presidential Commission, 1997). At the individual level, the options for risk reduction include (1) staying abreast of relevant information received via the media and/or communications from health authorities, environmental protection agencies, and other sources; (2) modifying one’s own behavior, diet, and lifestyle to minimize risks to oneself and to others; (3) carefully observing any special precautions that may be called for to protect oneself and one’s fellow workers against hazardous agents in the workplace; and (4) joining with others in efforts to promote collective awareness of environmental risks and to reduce such risks (Griffith and Saunders, 1997).

REFERENCES Aardema MJ, MacGregor JT (2002) Toxicology and genetic toxicology in the new era of “toxicogenomics”: Impact of “-omics” technologies. Mutat. Res. 499:13–25. Allen FW (1992) Differing views of risk: The challenge for decision makers in a democracy. In:Levine DC, Upton AC, editors. Management of Hazardous Agents: Social. Political, and Policy Aspects, Vol. 2. Westport, CT: Praeger, pp.81–94. Ames BN, Magaw R, Gold LS (1987) Ranking possible carcinogenic hazards. Science 236:271–280. Andrews RNL, Turner AG (1987) Controlling toxic chemicals in the environment. In: Lave LB, Upton AC, editors. Toxic Chemicals, Health, and the Environment. Baltimore, MD: Johns Hopkins University Press, pp.5–37. Arrow KJ, Cropper ML,Eads GC, HahnRW,Lave LB, Noll RG,PortneyPR, Russell M,Schmalensee R, Smith VK, Stavins RN (1996) Is there a role for benefit-cost analysis in environmental, health, and safety regulation. A statement of principles. Sponsored by the Annapolis Center, the American Enterprise Institute, and Resources for the Future. Science 272:221–222. Ashby J, Tennant R (1988) Chemical structure, Salmonella mutagenicity, and extent of carcinogenicity as indicators of genotoxic carcinogens among 222 chemical tests in rodents by the U.S. NCl/ NTP. Mutat. Res. 204:17–115. Calabrese EJ, Baldwin LA, Holland CD (1999) Hormesis. A highly generalizable and reproducible phenomenon with important implications for risk assessment. Risk. Anal. 19:261–281. Cohen BL (2002) Cancer risk from low-level radiation. Am. J. Roentgenol. 179:1137–1143. Davies JC (1996) Comparing Environmental Risks.Washington, DC: Resources for the Future. Eurotunnel(1994) The Channel Tunnel: A Safety Case. Folkestone, Eurotunnel. Evans AW, Verlander NQ (1997) What is wrong with criterion FN-lines for judging the tolerability of risk. Risk. Anal. 17:157–168. Finkel AM (1990) Confronting Uncertainty in Risk Management.Washington, DC: Center for Risk Management, Resources for the Future. Fischoff B, Bostrom A, Quadrel MJ (1997) Risk perception and communication. In: Detels R, Holland W, McEwen J, Omenn GS, editors. Oxford Textbook of Public Health,New York: Oxford University Press, pp.987–1002. Griffith R, Saunders P (1997)Reducing environmental risk. In: Detels R, Holland W, McEwen J, Omenn GS,editors. Oxford Textbook of Public Health.New York: Oxford University Press, pp.1601–1620 Hammond PB, Coppock R (1990) Valuing Health Risks, Costs, and Benefits for Environmental Decision Making.Washington, DC: National Academy Press. International Commission on Protection against Environmental Mutagens and Carcinogens (ICPEMC) (1988) Testing for mutagens and carcinogens: The role of short-term genotoxicity assays. Mutat. Res. 205:3–12.

REFERENCES

53

International Commission on Radiological Protection (ICRP) (1997) Problems involved in developing an index of harm. ICRP Publication 27. Ann. ICRP 1 (4). International Commission on Radiological Protection (ICRP) (1991) 1990 Recommendations of the International Commission on Radiological Protection. ICRP Publication 60. Ann. ICRP 21(1–3) Lave LB, Ennever FK, Rosenkranz HS, Omenn GS (1988) Information value of the rodent bioassay. Nature 336:631–633. Mauderly JL (1993) Toxicological approaches to complex mixtures. Environ. Health Perspect. 101 (Suppl.):155–164. Morgan MG, Hendon M (1990) Uncertainty: A Guide to Dealing with Uncertainty in Quantitative Risk and Policy Analysis.New York: Cambridge University Press. National Academy of Sciences/National Research Council (NAS) (1978) Saccharin: Technical Assessment of Risks and Benefits. Part 1 of a 2-part Study of the Committee for a Study of Saccharin and Food Safety Policy. Panel 1: Saccharin and Its Impurities.Washington, DC: National Academy Press. National Academy of Sciences/National Research Council (1983) Risk Assessment in the Federal Government: Managing the Process. Washington, DC: National Academy Press. National Academy of Sciences/National Research Council (1984) Toxicity Testing: Strategies to Deter-mine Needitors and Priorities. Washington, DC: National Academy Press. National Academy of Sciences/National Research Council (1987) Biological markers in environmental health research. Report of the National Research Council Committee on Biological Markers. Environ. Health Perspect. 74:39. National Academy of Sciences/National Research Council (NAS) (1988) Complex Mixtures: Method for In Vivo Toxicity Testing. Washington, DC: National Academy Press. National Academy of Sciences/National Research Council (NAS) (1989) Improving Risk Communication. Washington, DC: National Academy Press. National Academy of Sciences/National Research Council (NAS) (1994) Science and Judgment in Risk Assessment. Washington, DC: National Academy Press. National Academy of Sciences/National Research Council (NAS) (1996) Understanding Risk: Informing Decisions in a Democratic Society. Washington, DC: National Academy Press. National Academy of Sciences/National Research Council (NAS) (2005) Health Risks from Exposure to Low Levels of Ionizing Radiation. BEIR VII-Phase 2. Washington, DC: National Academy Press. New York Times Editorial: The Million-Year Health Standard. New York Times, November 25, 2005. Office of Science and Technology Policy (1985) Chemical carcinogens: A review of the science and its associated principles. Feditor. Register 50:10371. Omenn GS, Faustman EM (1997) Risk Assessment, risk communication, and risk management. In: Detels R, Holland W, McEwen J, Omenn GS, editors. Oxford Textbook of Public Health.New York: Oxford University Press, pp.969–986. Peto R, Pike MC, Bernstein L, Gold LS, Ames BN (1984) A proposed general convention for the numerical description of the carcinogenic potency of chemicals in chronic-exposure animal experiments. Environ. Health Perspect. 58:1–8. Plough A, Krimsky S (1987) The emergence of risk communication studies social and political context. Sci. Technol. Hum. Values 12:4–10. Presidential/Congressional Commission on Risk Assessment and Risk Management (1997) Risk Assessment and Risk Management in Regulatory Decision Making. Washington. DC. Raynor S, Cantor R (1987) How fair is safe enough? The cultured approach to societal technology choice. Risk. Anal. 7:3–9.

54

PERSPECTIVES ON INDIVIDUAL AND COMMUNITY RISKS

Rodricks JV (1992) Calculated Risks: Understanding the Toxicity and Human Health Risks of Chemicals in Our Environment.Cambridge: University of Cambridge Press. Sandman PM (1985) Getting to maybe: Some communication aspects of siting hazardous waste facilities. Seton Hall Legis. J. 9:442–465. Slovic PB, Fischoff B, Lichtenstein S (1979) Rating the risks. Environment 21:14–20, 36–39. Slovic PB, Fischoff B, Lichtenstein S (1981) Perceived risk: Psychological factors and social implications. Proc. R. Soc. Land. A376:17–34. Slovic P. (1998) trust, emotion, sex, politics, and science: surveying the risk assessment battlefield. In: Tinker, TL, Pavlova, Gotsch, AR, Arkin EB, editors. Communicating Risk in a Changing World. Solomons Island, MD; Beverly Farms, MA: OEM Press. Tennant RW, Margolin BH, Shelby MD, Zeiger E, Hascman LK, Spalding J, Caspary W, Resnick M, Staciewicz S, Anderson B, Minor R (1987) Prediction of chemical carcinogenicity in rodents from in vitro genetic toxicity assays. Science 236:933–941. Tones K (1997) Health education, behaviour change, and the public health. In: Detels R, Holland W, McEwen J, Omenn GS,editors. Oxford Textbook of Public Health.New York: Oxford University Press, pp.783–814. Travis CC, Pack SR, Hattmer-Frey HA (1989) Is ionizing radiation regulated more stringently than chemical carcinogens?Health Phys. 56:527–531. United Nations Scientific Committee on the Effects of Atomic Radiation (UNSCEAR) (2000) Sources. and Effects of Ionizing Radiation. Report to the General Assembly. Official Records. New York: United Nations. United States Environmental Protection Agency (1987) Unfinished Business: A Comparative Assessment of Environmental Problems. Washington, DC: U.S. EPA. United States Environmental Protection Agency (1990) Reducing Risk: Setting Priorities and Strategies for Environmental Protection. Washington, DC: U.S. EPA. United States Environmental Protection Agency (1996) Review of the National Ambient Air Quality Standards for Particulate Matter: Policy Assessment of Scientific and Technical Information–– OAQPS Staff Paper, (EPA/452/R-96-013), Office of Air Quality Planning and Standards, Research Triangle Park, NC 27711. United States Environmental Protection Agency (2005a) Review of the National Ambient Air Quality Standards for Particulate Matter: Policy Assessment of Scientific and Technical Information–– OAQPS Staff Paper, (EPA/452/R-05-005a), U.S. EPA Office of Air Quality Planning and Standards, Research Triangle Park, NC 27711. United States Environmental Protection Agency 40 CFR Part 197(August 22, 2005b) Public Health and Environmental Radiation Protection Standards for Yucca Mountain, Nevada: Proposed Rule. Federal Register, Vol. 70(161), pp. 49014–49065. United States Environmental Protection Agency (2005c) A Citizen’s Guide to Radon. The Guide to Protecting Yourself and Your Family from Radon. U.S. EPA 402-K02-006. Washington, DC: U.S. EPA. Wilson R (1979) Analyzing the risks of daily life. Technol. Rev. 81:41–46. Zeckhauser RJ, Viscusi WK (1990) Risk within reason. Science 248:559–564. Zeise LR, Wilson R, Crouch AC (1987) The dose–response relationship for carcinogens: a review. Environ Health Perspect. 73:259–306.

3 REDUCING RISKS––AN ENVIRONMENTAL ENGINEERING PERSPECTIVE Raymond C. Loehr

3.1 INTRODUCTION Concern about environmental risks, particularly those affecting human health, can be traced to the earliest human records (Graham, 1994; Paustenbach, 1989, 2002). However, it was not until the twentieth century that there was a concerted focus on the protection of humans from the adverse effects of chemicals at work, from industrial emissions, in commerce, and in the environment. The purpose of this chapter is to provide a perspective of the use of risk assessment and risk-based decision making by environmental engineers to protect human health and the environment. The topics covered include . . . .

.

risk-based decision making in the Superfund process; the risk-based corrective action (RBCA) framework; factors that affect site-specific health and environmental risk; recent scientific and research knowledge that can be used for site-specific risk evaluations; the value of site-specific parameters, rather than default assumptions, for risk-based management decisions.

Environmental Toxicants, Third Edition Edited by Morton Lippmann Copyright Ó 2009 John Wiley & Sons, Inc.

55

56

REDUCING RISKS––AN ENVIRONMENTAL ENGINEERING PERSPECTIVE

3.2 ENVIRONMENTAL RISK-BASED DECISION MAKING 3.2.1

Overview

Environmental risks have changed over time. In the early 1900s, lead arsenate was the principal insecticide sprayed on fruits and potatoes to control insects; medicinals were unreliable; industrial towns were black with soot as were people’s lungs; and workers labored at their own peril (Lowrance, 1976). The principal fatal diseases were pneumonia, influenza, and tuberculosis. More than 13% of all American children died before their first birthday. Paustenbach (2002) (Chapter 1) provides a detailed historical timeline of events that relate to human health risk assessment issues. Many steps have been taken to reduce the major earlier human health and environmental risks. As a result, other risks, which at one time were low in priority, are now of concern. Both personal and societal risks are inherent in human activity and it is not possible to reduce all risks to zero. Risk is an inescapable fact of life. However, it is not risk per se that generally is of concern to the public. Involuntary risk is the major concern. Opinions of “acceptable” risk frequently depend on the degree of choice to decline a specific risk. Site-specific risk assessments are used to illuminate possible degree of impact, options, priorities, and frequently relative costs. A risk assessment provides information to policy makers, regulatory agencies, other risk managers, and to the public so that the most appropriate decisions can be made. A risk assessment provides the bridge between scientific information and risk management and puts the words toxicity, hazard, and risk into perspective (Paustenbach, 1989, 2002). The real world of risk assessment is based on sound scientific and empirical knowledge. The commonly accepted definitions of risk assessment and risk management are those suggested by the National Research Council (NRC, 1983): Risk Assessment—the characterization of potential adverse health effects of human exposures to environmental hazards; the assessment includes characterization of the uncertainties inherent in the process of inferring risk. Risk Management—the process of evaluating alternative regulatory actions and selecting among them; the selection process requires the use of value judgments on such issues as the acceptability of risk and the reasonableness of the costs of control. Until the early 1990s, the concepts of the risk assessment paradigm and the use of risk assessment as part of the environmental decision-making process were viewed by environmental engineers as an esoteric practice. While possibly valuable, it was considered on the fringes of environmental engineering and hazardous waste management practice. Emphasis was placed on meeting specific regulatory requirements, which commonly were based on worst-case assumptions used to assure maximum protection of human health. Today, risk assessment is an important component of many environmental engineering and waste management projects, specific state and federal guidance is available and it is an integral part of federal and state environmental legislation. 3.2.2

Risk Assessment Process

The risk assessment process can be divided into four major steps: hazard identification, exposure assessment, dose–response assessment, and risk characterization (NAS, 1994).

ENVIRONMENTAL RISK-BASED DECISION MAKING

57

Risk assessment is then followed by risk management if the estimated risk is indicated to be unacceptable. Risk assessment considers—how risky is the situation or exposure? Risk management considers what to do about the risk, that is, how to reduce or remove it, if it is deemed unacceptable. The values of the risk assessment process are many. An important value is that it provides a consistent, disciplined approach to organize scientific information so that the relevant items are considered and used for subsequent risk management decisions. The risk assessment process helps identify the uncertainties involved in the data and the assumptions that are involved. It helps indicate the time frames involved, and who or what is affected. It also helps identify strategies and priorities such as where in the process risk management decisions can be most effective. Overall, it helps educate all involved about the factors, exposures, effects, and relative risks that exist at a site or for a particular situation. 3.2.3

Risk Management from an Environmental Engineering Perspective

The concepts of risk-based corrective action and risk-based decision making are being widely applied to identify appropriate risk management decisions that should be considered at specific sites to protect human health and the environment. The following example may help put the use of the risk assessment process for environmental decision making into perspective. A simplistic risk assessment paradigm is shown in Fig. 3.1. For a situation to pose a threat to human health and the environment, there must be a source of contamination. Then, one or more chemicals must be released from that source. Once released, the chemicals must be transported to a receptor of concern such as a human, plant, or fish. During the transport, there may be transformations that can change the form or speciation of the chemical, which in turn may make the chemical more or less mobile or toxic. If the chemical is able to contact a sensitive area of a receptor, that is, if exposure occurs, there could be a negative impact. The environmental risk associated with that negative impact can be identified and may be considered acceptable or unacceptable. The risk assessment process has been formalized and is used increasingly by the U.S. Environmental Protection Agency (EPA) and state regulatory agencies as part of environmental decision making. An environmental engineer uses the knowledge from the risk assessment process to identify locations and situations by which the identified risks can be reduced. For instance, in terms of the paradigm in Fig. 3.1, possible control or remediation options are noted in Table 3.1. Schematically, these options can be included in the risk assessment paradigm in Fig. 3.2. The interaction between risk assessment and risk management is shown in Fig. 3.3 which also indicates some of the nonengineering factors that are involved in a risk management decision.

Emission/ release

Transformations and transport

Source of chemicals

Receptor

Risk - acceptable - unacceptable

FIGURE 3.1 Conceptual risk assessment paradigm.

58

REDUCING RISKS––AN ENVIRONMENTAL ENGINEERING PERSPECTIVE

TABLE 3.1 .

.

.

.

.

.

Example Environmental Engineering Options to Reduce Environmental Risks

Removing the source of contamination in soils and allowing natural environmental assimilative processes to control any remaining chemicals and those that have been released. Changing the chemical or manufacturing process that is emitting chemicals of concern, thus decreasing or eliminating the chemical causing the risk. This is the cornerstone of the pollution prevention approach now widely used by industry. Changing the form of the chemical, either at the source or after release. This can be done by modifying the pH of the media containing the chemical, or adding other chemicals that will immobilize the chemical of concern, that is, solidification–stabilization. Intercepting and treating emitted and released chemicals of concern before they reach a receptor. Examples include in-situ and ex-situ biological and chemical treatment processes. Preventing released chemicals of concern from reaching a receptor. Capturing a chemical of concern at the source and in-situ slurry walls are two such possibilities. Separating the receptor from the chemical of concern. Providing alternative drinking water and food supplies and purchasing property and moving residents are possibilities.

For a real-world situation, there are many risk management options that can be considered. For a specific problem, the challenge for the environmental engineer is to identify, develop, and implement cost-effective technical solutions that will protect human health and the environment. The cost implications associated with protective risk-based management decisions are not trivial. Consider the following example that resulted at an actual site where alternatives for remediation were evaluated. The detailed site remedial investigation studies indicated that the life cycle remediation costs were related to the required clean-up levels (Table 3.2). A site-specific risk-based evaluation determined that a clean-up level of 400 ppm for the chemical of concern was protective of human health. In the absence of the risk-based evaluation, a default and lower clean-up level would have been required, resulting in a considerably larger site remediation cost. Paustenbach (1989, p. 92) also provided a pertinent example of the relationship between the cost of remediation and clean-up levels. In that example, the estimated costs of soil removal and destruction for various soil clean-up levels of dioxin were shown. The cost was indicated to be about $17 million for a clean-up level of about 1 ppb dioxin to less than $1 million for a clean-up level of 100 ppb dioxin in the soil. Source - emissions - transport - fate

Exposure evaluation - human - environmental

Risk reduction techniques - waste minimization - source reduction or elimination - treatment and control techniques - exposure reduction - site remediation

FIGURE 3.2

Effects -

human environmental

Unacceptable risk

Acceptable (no action)

Risk assessment and illustrative engineering risk management approaches.

ENVIRONMENTAL RISK-BASED DECISION MAKING

59

Statutory and legal considerations Dose–response assessment

Hazard identification

Public health considerations Risk characterization

Exposure assessment

Social factors

Risk management decision

Risk management options

Economic factors

Political considerations

Risk assessment

FIGURE 3.3 NRC, 1983).

Risk management

Conceptual interactions between risk assessment and risk management (adapted from

Evaluations of the risk associated with specific chemicals at a site can help identify the site-specific clean-up levels that are protective of human health and the environment. A sound site-specific risk assessment is important to assure that human health and the environment are protected and to assure that resources (time, energy, funds) are used effectively. The resource considerations related to reducing environmental risks were stated clearly by EPA in 1984 (U.S. EPA, 1984, p. 23): One can argue about how much should be spent on environmental protection, but at some point everyone must accept that the commitment of resources for any social purpose has a finite limit. If the number of potential risk targets is very large in comparison to the number we can realistically pursue, which seems now to be the case, then some rational method of choosing which risks to reduce and deciding how far we should try to reduce them is indispensable. It is important to keep in mind that while individual risk management decisions may be seen as balancing risk reduction against resources, the system as a whole is designed to balance risk against risk. In other words, it is essential that we address the worst and most controllable risks first; failure to do so means that the total amount of harm that we prevent is smaller than the

TABLE 3.2 Example Situations––Relationship of Clean-Up Action Levels to Remediation Life Cycle Costs Assumed Clean-Up Action Level (ppm of Chemical in Soil) 10 50 180 400

Total Estimated Remediation Site-Specific Life Cycle Costs ($ Millions) 1375 163 69 30

60

REDUCING RISKS––AN ENVIRONMENTAL ENGINEERING PERSPECTIVE

amount we might have prevented. Making incorrect priority choices, saving one where we might have saved two, represents a profound failure of the Agency’s basic mission.

The ideas and concepts in the above statement have been recognized broadly, and have stimulated many of the approaches now being used to assure that (a) the focus is on important environmental risks; and (b) cost-effective environmental risk reduction strategies are utilized. U.S. EPA continued to explore approaches to use the concept of relative risk for environmental decision making. In 1990, the EPA Administrator called for a national debate on environmental directions and policies (Reilly, 1990). What stimulated that call and the debate was the U.S. EPA Science Advisory Board Report, Reducing Risk (U.S. EPA, 1990). The 10 summary recommendations of the report are noted in Table 3.3. The recommendations emphasized targeting environmental protection efforts on the basis of environmental risk and risk reduction opportunities. They called for risk-based priorities in national planning and budgeting and for emphasis on pollution prevention rather than on end-of-pipe treatment. Efforts to better educate the public about actual risks also were recommended. 3.3 APPLICATIONS AND USE Risk assessment evaluations have been applied in many situations for environmental management decisions. This is a result of the following: (a) environmental engineers and regulators now better understand the limits of existing technologies; (b) the sciences of risk assessment and modeling now can be used to measure and compare the benefits of possible options; and (c) many studies have documented that all sites and situations do not require treatment or removal to the same generic standard. The following sections indicate the use of risk-based decision making for different environmental situations and decisions. These include an overview of the risk assessment process as used to evaluate site-specific risks, Superfund remediations, a discussion of the risk-based corrective action approach now used widely for environmental management decisions, and the type of recent information that has become available for improved sitespecific risk-based management decisions. 3.3.1

Overview

There are many uses of a site-specific risk assessment (Table 3.4). From an engineering and environmental decision standpoint, an environmental risk assessment can be conducted in a forward or a reverse mode. In the forward mode, using the items in Fig. 3.1, data about the chemical at the source and transformation and transport knowledge is used to assess the risk to a site-specific human or ecological receptor from a defined source, such as a leaking fuel tank, a spill, or a gaseous emission. In the reverse mode, one would work backward from the exposure to a chemical that a known receptor may have. With that exposure information and, using protective health or ecological criteria, the soil, groundwater, or emission levels that would have to exist to be protective of human health and the environment can be calculated. This reverse approach can be used to determine site remediation or chemical emission levels that would have to be achieved so that there is no adverse effect to the receptor.

APPLICATIONS AND USE

TABLE 3.3

61

Recommendations from the EPA Reducing Risk Report

1. EPA should target its environmental protection efforts on the basis of opportunities for the greatest risk reduction. The United States already has taken actions to address the most obvious environmental problems. EPA needs to set priorities for future actions so the nation takes advantage of the best opportunities for reducing the most serious remaining risks. 2. EPA should attach as much importance to reducing ecological risk as it does to reducing human health risk. Productive natural ecosystems are essential to human health and to sustainable, long-term economic growth. 3. EPA should improve the data and analytical methodologies that support the assessment, comparison, and reduction of different environmental risks. Setting priorities for national environmental protection efforts always will involve subjective judgments and uncertainty. The scientific data and analytical methodologies that underpin those judgments and reduce their uncertainty should be improved. 4. EPA should reflect risk-based priorities in its strategic planning processes. The Agency’s long-range plans should be driven not so much by past risk reduction efforts or by existing programmatic structures, but by assessments of remaining environmental risks, and the analysis of opportunities available for reducing risks. 5. EPA should reflect risk-based priorities in its budget process. Although EPA’s budget priorities are determined to a large extent by the different environmental laws that the Agency implements, it should use whatever discretion it has to focus budget resources at those environmental problems that pose the most serious risks.

Source: (U.S. EPA, 1990).

6. EPA–and the nation as a whole–should make greater use of all the tools available to reduce risk. The extent and complexity of future risks will necessitate the use of a much broader array of tools, including market incentives and information. 7. EPA should emphasize pollution prevention as the preferred option for reducing risk. By encouraging actions that prevent pollution from being generated in the first place, EPA will help reduce the costs, intermedia transfers of pollution, and residual risks associated with end-ofpipe controls. 8. EPA should increase its efforts to integrate environmental considerations into broader aspects of public policy in as fundamental a manner as are economic concerns. Other federal agencies often affect the quality of the environment, e.g., through the implementation of tax, energy, agricultural, and international policy. EPA should work to ensure that environmental considerations are integrated, where appropriate, into the policy deliberations of such agencies. 9. EPA should work to improve public understanding of environmental risks and train a professional workforce to help reduce them. The improved environmental literacy of the general public, together with an expanded and better-trained technical workforce, will be essential to the nation’s success at reducing environmental risks in the future. 10. EPA should develop improved analytical methods to value natural resources and to account for long-term environmental effects in its economic analyses. Traditional methods of economic analysis tend to undervalue ecological resources and fail to treat adequately questions of intergenerational equity. EPA should develop and implement innovative approaches to economic analysis that will address these shortcomings.

62

REDUCING RISKS––AN ENVIRONMENTAL ENGINEERING PERSPECTIVE

TABLE 3.4 Example Uses of a Risk Assessment for Environmental Engineering Purposes Determines the need for remedial action at contaminated sites Helps set priorities and establish the urgency to remediate sites . Determines site-specific, health-based clean-up levels . Provides a position from which the concerned parties can negotiate clean-up levels . Allows the use of less expensive remedial alternatives, while still being protective of human health and the environment . Helps identify suitable locations for waste management facilities, such as municipal incinerators and landfills . Supports litigation involving chemical emissions and exposures . Helps meet mandates of federal or state environmental regulations . .

Site-specific risk assessments that are part of risk-based management decisions commonly are conducted at several levels or tiers, with a screening level assessment being done first. The basic steps of a site-specific risk assessment are indicated in Table 3.5. The detail and accuracy of the information needed, the time needed to complete the risk assessment evaluation, and the cost of the evaluation are different in each tier of the evaluation. The screening level evaluation requires the least amount of information, since many conservative assumptions are involved. The cost of a risk assessment evaluation can increase by a factor of 5–10 when the evaluation moves from a screening level to a detailed evaluation. The specifics of a risk assessment evaluation will be different for each site. The general characteristics of a risk assessment process for environmental engineering purposes are indicated in Table 3.6. 3.3.2

Superfund

From an environmental engineering standpoint, the risk assessments and the site-specific risk management decisions for Superfund sites represent some of the more detailed and comprehensive evaluations. The details of a Superfund risk assessment are well developed and standardized (U.S. EPA, 1988, 1989a, 1989b, 1989c, 1991a, 1991b). The following is a summary of the Superfund site risk assessment steps and process. The major decisions made at a Superfund site are based on answers to questions such as (a) how serious are the problems at the site; (b) should something be done at this site; (c) what should be done; and (d) when has enough been done? TABLE 3.5 . . . . . . . . . . .

Basis Steps in a Site-Specific Risk Assessment

Prepare a risk assessment plan for the site or situation Identify available data, data needs, and data gaps Define data quality objectives Prepare a sampling plan Identify chemicals of concern and assess toxicology Identify current and potential future receptor(s) Define exposure pathways Estimate exposure point concentrations Estimate applied or absorbed dose Characterize health risks and uncertainties Identify and state key assumptions

APPLICATIONS AND USE

TABLE 3.6 Purposes

63

Characteristics of the Risk Assessment Process for Environmental Engineering

Develop reasonable exposure scenarios Current site and surrounding land use Future reasonable site and surrounding land use Real and possible hypothetical pathways from source to receptor Evaluate multiple exposure pathways Air Soil Groundwater Surface water Evaluate multiple exposure routes Ingestion Inhalation Dermal Involve multidisciplinary expertise Quantify environmental multimedia fate and transport Understand and apply toxicological principles Understand site-specific environmental chemistry Interpret multimedia field data Involve statistical analysis of the data Risk communication skills needed

The Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA, also known as Superfund) was passed by Congress in 1980 and amended in 1986. Superfund requires EPA to identify and, if needed, to reduce the risks from past inadequate waste management and disposal approaches. Different risk assessment approaches are applied in evaluating possible effects of Superfund sites on human health and the environment. Neither EPA nor the Superfund program has one standardized environmental evaluation methodology because there are so many different aspects of the environment––individuals, species, ecosystems, natural resources, endangered species–– which may need to be evaluated. As a result, the Superfund program has identified an orderly process for environmental risk evaluation. Risk assessment for a Superfund site is a four-step process. The first step, data collection and evaluation, identifies contaminants present in the environmental media––soil, groundwater, surface water, air, fish––of the site. The second step, toxicity assessment, uses the results of prior research and testing to decide which of the contaminants found on site might pose a health threat. The third step, exposure assessment, defines which exposure pathways might bring the contaminants in contact with people. The final step, risk characterization, brings information from the first three steps together to determine the potential severity of health threats from the site. Figure 3.4 indicates the basic steps of the Superfund site evaluation process and the role of risk assessment in that process. As indicated, the detailed risk assessment evaluation occurs at the remedial investigation/feasibility study (RI/FS) stage in the process. However, a screening level risk assessment occurs earlier in the hazard ranking system (HRS) scoring.

64

REDUCING RISKS––AN ENVIRONMENTAL ENGINEERING PERSPECTIVE

FIGURE 3.4

Major steps in the Superfund site evaluation process (adapted from U.S. EPA 1991b).

The HRS analysis uses specific assessment principles to determine whether conditions at a site warrant placing the site on the National Priorities List (NPL) and using federal funds to continue with the rest of the site evaluation process. Most of the formal risk management activities occur after a site is placed on the NPL. These activities .

. .

help develop preliminary remediation goals during project scoping and their modification during the feasibility study (FS); develop the baseline risk assessment during the remediation investigation (RI); evaluate the effectiveness of remedial alternatives in the FS report; in the Record of Decision (ROD) to relate target clean-up concentrations to health risks; and during remedial action to monitor progress toward “acceptable risk.”

The baseline risk assessment of the RI is the central risk evaluation activity in the Superfund program and is a four-step risk assessment paradigm involving data evaluation, exposure assessment, toxicity assessment, and risk characterization. To initiate the baseline risk assessment of the RI, information on site history and data gathered during the preremedial program or a recent site visit are assembled and used to guide the remedial investigation. A critical step is identifying contaminants of significant toxicity and all exposure pathways of concern. In addition to the media paths––soil, air, and drinking water––other paths such as the eating of contaminated food or recreation may be important. Knowing toxicity and exposures to be evaluated leads directly to a sampling strategy (e.g., identifying “hot spots,” gathering sufficient data for reasonable maximum exposure (RME) determinations), to appropriate analytical methods (e.g., requesting special analytical services for detection at low concentration), and to related data quality objectives (DQOs) for sampling and analysis. The exposure assessment looks for pathways of exposure to particular individuals on or near the site, since the activities of the people determine the exposure. The baseline risk

APPLICATIONS AND USE

65

assessment considers both present exposures and those that might result from current or probable future land use. What is determined is the reasonable maximum exposure (RME). This is to be the highest exposure that is reasonably expected to occur at a site, considering land use, intake variables, and pathway combinations. The intent is to estimate a conservative exposure case that is still within the range of possible exposures. The result is that people at or near sites will be protected, but cleanups will not be driven by assumed exposures outside the range of possibility. Toxicity assessments are the next step in the process and are done using available information and the expertise of toxicologists. These are weight-of-evidence classifications. EPA has standardized this approach in a five-class grouping. In general, multiple welldesigned studies, studies showing adverse effects in several species of animals, and evidence of adverse effects in humans provide greater weight-of-evidence of information. Once the exposure and toxicity assessments are complete, the risk assessor must characterize the risks. This step is crucial for communicating both to individuals who must decide what actions to take at a site, and to those who may live at or near the site. Excess lifetime carcinogen risks to individuals of less than 10 4 (1 in 10,000) do not require remedial actions at Superfund sites. However, actions may be taken to reduce risks below 10 4. If the baseline risk assessment shows risks greater than 10 4 to individuals, then initial clean-up target concentrations corresponding to 10 6 risk are chosen. Remedial alternatives are evaluated against nine criteria (Table 3.7). The remedy selected in the Record of Decision (ROD) is the one that best satisfies the criteria. Every ROD should, at a minimum, identify the contaminants posing risks, target concentrations for cleanup, points of compliance for cleanup in each medium, and the risks that will remain after completion of the remedy if clean-up goals are achieved. Every attempt is made to have a Superfund risk assessment and the remediation decisionmaking process be straightforward and progress logically. In practice, each site is unique, data are incomplete, uncertainties can be large, and professional judgment and interpretation are involved. However, the risk assessment evaluation is an extremely important part of the decision process and is the key to environmentally sound and protective site-specific decisions and subsequent management decisions. The incorporation of risk assessment in Superfund decisions represents the most standardized environmental engineering application of the process. However, in the past decade, there have been an increasing number of real-world situations in which environmental engineers have been involved in risk-based management decisions to protect human TABLE 3.7 . . . . . . . . . a

Criteria Used to Determine Superfund Site Remediation Alternativesa

Overall protection of human health and the environment Compliance with other regulations Long-term effectiveness and permanence Reduction of toxicity, mobility, or volume Short-term effectiveness Implementability Cost State Acceptance Community Acceptance Adapted from U.S. EPA (1991b).

66

REDUCING RISKS––AN ENVIRONMENTAL ENGINEERING PERSPECTIVE

health and the environment. In doing so, increasing use has been made of the risk-based corrective action approach and methodology. 3.3.3

Risk-Based Corrective Action

The risk-based corrective action (RBCA) approach takes the risk assessment methodology and applies it to situations needing evaluation. RBCA is a standardized approach for developing remediation strategies and has gained wide acceptance from regulatory agencies. It identifies the types of risks to human health and the environment at the sites and allows the types of corrective actions considered and implemented to be commensurate with those risks. The RBCA process integrates components of site assessment, risk assessment, risk management, and remediation into a holistic site-specific approach that is consistent and technically defensible while still being practical and cost-effective. The details of the process and examples are available in several ASTM guides (ASTM, 1995, 1999). There are some who view RBCA as an approach to avoid remediating a contaminated site. Based on experience with the use of RBCA and with the overview provided by regulatory agencies, that is not the case. Rather, RBCA provides the following to determine site-specific effective and protective risk-based management approaches: (a) flexible framework; (b) procedure consistency; (c) a classification scheme that helps focus efforts and direct the type and urgency of response; and (d) a tiered approach that provides an increasingly sitespecific assessment where site conditions warrant. It also is resource effective and focuses resources toward assessment and remedial measures on sites, exposure pathways, and substances of significant concern, with remedial goals based on reducing risk to acceptable levels. Sites with surface and subsurface contamination vary greatly in terms of complexity and physical and chemical characteristics, and in the risk that they may pose to human health and environmental resources. The RBCA process recognizes this diversity, and utilizes a tiered approach involving increasingly sophisticated levels of data collection and analysis. As needed, the conservative assumptions of an earlier tier are replaced with site-specific assumptions in later tiers. Upon completion of each tier, the user reviews the results and recommendations, and decides if more site-specific analysis is required. In Tier 1, sites are classified by the urgency of need for initial corrective action, based on information collected from historical records, a visual inspection, and minimal site assessment data. The user is required to identify contaminant sources, existing environmental impacts (if any), the presence of potentially impacted humans and environmental resources (e.g., workers, residents, water bodies), and potential significant transport pathways (e.g., groundwater flow, atmospheric dispersion). Associated with site classifications are prescribed initial response actions, if such are needed, that are to be implemented prior to proceeding further with the RBCA process. In addition, as part of Tier 1, conservative corrective action goals are based on a list of nonsite-specific, risk-based screening levels (RBSLs), aesthetic criteria, and other appropriate standards such as Maximum Contaminant Levels (MCLs) for potable groundwater use. Tier 1 RBSLs are typically derived for standard exposure scenarios using current reasonable maximum exposure (RME) and toxicological parameters as recommended by the U.S. EPA, and using conservative contaminant migration models. Tier 2 provides the user with an option for determining site-specific target levels (SSTLs) and appropriate points of compliance when it is judged that Tier 1 corrective action goals are not adequately protective of human health and the environment or if such goals appear overly

RECENT INFORMATION

67

conservative for the site conditions. Both Tier 1 and Tier 2 screening levels are based on achieving similar levels of human health and environmental resource protection (e.g., 10 4 to 10 6 risk levels). In moving to higher tiers, the user is able to develop more cost-effective corrective action plans because the conservative assumptions of earlier tiers are replaced with more realistic site-specific assumptions. In some cases, the Tier 2 SSTLs are derived from the same equations used to calculate Tier 1 RBSLs, except that site-specific parameters are used in the calculations rather than default values. At other sites, the Tier 2 analysis may involve applying Tier 1 RBSLs at more probable points of exposure, such as property boundaries and negotiated points of compliance, and then deriving Tier 2 corrective action goals for the contaminant source areas based on demonstrated and predicted attenuation of contaminants of concern with distance. Tier 3 provides the user with an option for determining SSTLs and appropriate points of compliance when it is judged that Tier 2 corrective action goals are not appropriate or appear overly conservative. The major distinction between Tier 2 and Tier 3 analyses is that a Tier 3 analysis is generally a substantial incremental effort relative to Tier 2, as the analysis is much more complex and may include increased site sampling and analyses, detailed site assessments, probabilistic evaluations, and more sophisticated chemical fate/transport models. If the selected target levels are exceeded and corrective action is necessary, the user develops a corrective action plan in order to reduce the potential for adverse impacts. One option is to utilize traditional remediation processes to reduce contaminant concentrations below the target levels. Another option is to achieve exposure reduction (or elimination) through use of institutional controls or through the use of containment measures, such as capping and hydraulic control. In the RBCA process, remedial action is determined to be appropriate, based on the comparison of representative concentrations to the target levels determined under the tier evaluation. This allows the project to focus only on those areas or media posing a potential threat to human health or the environment. Monitoring is conducted following a remedial action to demonstrate that target levels are met and continue to be met, and to verify the assumptions and predictions that were used in any Tier 2 or Tier 3 evaluations. The original RBCA guide (ASTM, 1995) focused on sites containing chemicals at petroleum release sites. Shortly after, additional guides that addressed collecting site characterization data and considering natural attenuation as a safe remedial alternative were prepared (ASTM, 1999). Subsequently, RBCA type efforts have been developed for exploration and production sites (McMillen et al., 2001) and for sites with residues from manufactured gas plant (MGP) operations (Linz and Nakles, 1997). The RBCA approach to contaminated site remediation has been adopted by many states. Such rules or guidance are an improvement from the previously overly conservative approach of requiring removal of all contaminants. Such rules and guidelines also achieve protection of human health and the environment using remedies that have a high degree of long-term effectiveness.

3.4 RECENT INFORMATION In the past 20 years, the ability to identify site-specific human health and ecological risks has increased considerably. In the context of the concepts being discussed (Fig. 3.1), this has increased the ability to use site-specific risk-based evaluations for remediation and management decisions.

68

REDUCING RISKS––AN ENVIRONMENTAL ENGINEERING PERSPECTIVE

In the same period, there has been a considerable body of new knowledge developed about the factors that affect the release, emission, transformation and transport of chemicals that may be found at a particular site. This has been the result of research using actual field soil and sediment samples and has provided specific information about actual chemical release and about the actual environmental availability and bioavailability of chemicals in a specific soil or sediment. This increased knowledge is particularly relevant to the human and environmental risk associated with anthropogenic hydrophobic organics such as petroleum hydrocarbons, polyaromatic hydrocarbons (PAH) and polychlorinated biphenyl compounds (PCB). The weight of evidence information that has resulted from such research and that now is in the peer-reviewed literature indicates the following: .

.

.

.

.

.

.

Results obtained from methods used for the analysis of organic compounds in a soil or sediment do not measure the chemical availability or bioavailability of chemicals such as PAH or petroleum hydrocarbons that are in such media. Assuming that all extractable organic chemicals in field soils and sediments are chemically available or bioavailable overestimates the risk of such chemicals. Desorption of sorbed hydrocarbons is not immediate. The actual desorption that occurs can be represented by a two-step pattern that consists of an initially relatively fast release followed by slow release of the remaining hydrocarbon. Neglect of such desorption (release) kinetics can lead to erroneous conclusions when estimating the movement and risk of a hydrophobic hydrocarbon in a soil or sediment. Two analytical methods, supercritical carbon dioxide extraction (SFE) and water desorption, are available to determine the fraction of a hydrophobic organic chemical that can be desorbed (released) relatively rapidly. That “fast” fraction is known as the F value for a particular organic chemical in a field soil or sediment sample. The rapidly released fractions (F values) of PAH and petroleum hydrocarbons in weathered field soils can be low, with many in the 0.1–0.4 range. The type of organic carbon in a field sample and the weathering of spilled or released hydrocarbons affects the rapidly released fraction of such chemicals.

Relevant information about these items is included in the following sections. 3.4.1

Analytical Measurements

Methods to analyze organic chemicals have been developed to extract and measure the total concentration of an organic chemical in a sample of soil or sediment. Regulatory agencies often conservatively assume that the amount of chemical measured by common and exhaustive extraction methods is 100% chemically and bioavailable. Such is not the case. Assuming that all extractable chemicals are fully available can considerably overestimate the risk of such chemicals. 3.4.2

Weathering

Unless the chemicals in a soil or sediment have resulted from a fresh spill, the chemicals at a site of concern have undergone considerable weathering. The impact of such weathering reduces the actual risk of the chemicals at a site. Weathering refers to the result of normal

RECENT INFORMATION

69

biological, chemical and physical processes that can affect the chemicals that remain at a site over time. When a spill or release of organic chemical occurs, weathering begins to alter the composition of the released material. These weathering processes include (a) biodegradation of the most readily degradable organic chemicals, (b) volatilization of the lower boiling point organics, (c) greater sequestration or sorption of residual organics to the soil particles and (d) leaching of water soluble low boiling point fractions. Thus, weathering affects the presence of industrial organic chemicals that may be at an actual site over time. As a result, the chemicals that remain at a weathered spill site will have a different distribution than those that originally entered the soil. The weathering processes also will result in a lower chemical availability and bioavailability of the remaining soilbound organic chemicals. 3.4.3

Parameter Determination

To have consistent baseline data for important chemical availability parameters (i.e., solubility, volatility, Kow and Koc), such as those in references (Lyman et al., 1982; Howard, 1990; Mackay et al., 1992; Verschueren, 1996), laboratory studies have been conducted with single, pure chemicals under controlled conditions. Where it was not possible to conduct controlled laboratory experiments, estimation methods have been used to determine reasonable values for organic compounds of interest. However, such parameters are not the best measure of what happens to organic chemicals at an actual contaminated site. Current data indicate that such parameters do not represent the actual solubility and partitioning that exist at an actual site. 3.4.4

Sorption/Desorption Kinetics

Sorption and desorption of an organic chemical in a soil or sediment are fundamental to the fate of chemicals in the environment. The parameters Kow and Koc are measures of the partitioning (sorption) of an organic chemical to soil or sediment particles. Numerous studies now have shown that the desorption (release) of sorbed organics in actual soils and sediments is not immediate but is represented by a two-stage pattern that consists of a relatively initial fast step followed by a slow one. In addition, such desorption occurs over many days. Neglect of such sorption kinetics can lead to erroneous conclusions in estimating chemical movement and risk at a site. Many studies have been undertaken to identify and quantify the rates of organic chemical desorption that do occur in actual soils. Two parameters have been determined as important. One is the fraction that is desorbed (released) rapidly––the F value. The other is the rate, K2, at which the remaining organic chemical is slowly released. F values for an organic chemical in a soil provide an estimate of the amount of an organic chemical that is likely to have the more immediate impact on the risk of the chemical at a specific site. The K2 value represents the rate at which the residual organic chemicals may be released and be chemically and biologically available over time. The K2 values can be important because they can be compared to biodegradation and volatilization rates for the organic chemical in the media. If the rate of volatilization, biodegradation or subsequent sorption is faster that the initial desorption (release) rate (K2), the slowly released chemicals may not be transported to a receptor and may not cause an adverse health or environmental impact.

70

REDUCING RISKS––AN ENVIRONMENTAL ENGINEERING PERSPECTIVE

TABLE 3.8

Illustrative Desorption Kinetic Data––F Values––Petroleum Hydrocarbons

Source Refinery site Refinery site Refinery site Crude oil Storage site Industrial Industrial Industrial

Soil

Hydrocarbon

F Value

Sandy silt Sandy silt Sandy silt Clayey

C20 C32 MRO MRO

0.11 0.02 0.09 0.04

Sandy silt Sandy silt Sandy silt

C10 C20 C25

0.40 0.45 0.05

MRO ¼ mineral range organic petroleum hydrocarbons.

As a result of extensive studies using field soils, there are now two peer-reviewed methods to determine F and K2 values. These methods are supercritical carbon dioxide extraction (SFE) and water desorption (Berg et al., 1998; Cornelissen et al., 2001; Hawthorne et al., 2001). Both methods produce comparable F values (Hawthorne et al., 2001; Hawthorne et al., 2002). In addition, a protocol based on F values has been developed (Loehr et al., 2003) that can be used for a relative risk evaluation of a site that has PAH or petroleum hydrocarbon impacted soils. Extensive desorption kinetic studies indicate that the kinetics and F values can be different for different organic chemicals depending on the conditions at a spill or chemical release site. Table 3.8 provides illustrative F values for petroleum hydrocarbons at different sites. As noted (Table 3.8), F values and hence rapid hydrocarbon desorption from a soil can be very low. For such situations, the data indicate that only a small fraction (some less than 10%) of a specific petroleum hydrocarbon at a site is likely to be desorbed rapidly and have an impact on human health. An evaluation of harbor sediments (Oen et al., 2006) provided data indicating that the actual rapidly desorbing fractions (F values) for several PAH in these sediments can be very low. The F values that were determined were less than 6% for phenanthrene, 3–19% for pyrene and 1–12% for benzo(a)pyrene. Both laboratory and field experience indicate that the following factors influence actual F values for PAH and other hydrophobic hydrocarbons in soils and sediments: Spill or Release Conditions––hydrocarbons at a fresh spill or release site have had less opportunity to sorb to soil particles and will have higher F values. Weathering––residual hydrocarbons at weathered hydrocarbon spill or release sites are more tightly sorbed to site organic matter and will have lower F values. Available Carbon––hydrocarbons sorbed to anthropogenic carbon––soot, combustion products, charcoal, coal, or coke products––have lower F values than do hydrocarbons sorbed to carbon in natural organic matter. 3.4.5

Type of Carbon

The various types of carbon, particularly anthropogenic carbon, that can be in a soil or sediment at a site play an important role in determining the chemical and biological

INTEGRATED ASSESSMENTS

71

availability of hydrophobic organic compounds of concern. Such hydrocarbons are more tightly sorbed to anthropogenic carbon, such as charred wood particles and combustion products such as soot, than to natural plant based organic carbon (Bucheli and Gustafsson, 2000; Jonker and Koelmans, 2002; Hawthorne and Miller, 2003; Hong et al., 2003). Jonker et al., 2005, studied the release of PAH from anthropogenic carbon and found very slow PAH desorption. Their results implied reduced environmental risk for PAH associated with such carbon. These noted differences can be quite important to site-specific risk-based decisions since anthropogenic carbon can be a predominant form of carbon at industrial and other sites of concern where risk-based site management decisions may be needed. These differences also cast doubt on current (2007) risk assessment procedures and environmental quality standards for PAH and other hydrophobic hydrocarbons. Recognition of such differences has resulted in a remedial approach that includes the possible addition of activated carbon to a site of concern (Zimmerman et al., 2005). The added anthropogenic carbon, when mixed with the soil or sediment needing attention, sorbs the hydrocarbons of concern, making them less easily released thus reducing site risks to humans and ecosystems. 3.4.6

Partition Coefficients

Studies with field soils have developed experimentally measured Koc values and soil–water partitioning coefficients, Kd (Kd ¼ Koc  foc), where foc is the organic carbon fraction in the soil. Results have indicated that the experimentally measured Koc and Kd values in field soil samples can be significantly greater than such values noted in reference books or estimated from chemical relationships noted in such references. In one study, the Koc values determined from analyzing field soil samples were 35–250 times higher than the respective reference predictions (Bucheli and Gustafsson, 2000). In another study (Dondelle and Loehr, 2002), Koc values determined from field soil samples were from 2 to over 200 times higher than the respective reference predictions. Such results were confirmed (Hawthorne et al., 2006) in an extensive evaluation of PAH partition coefficients in 114 contaminated sediments. The Koc values for the PAH in these sediments were considerably greater than typical literature Koc values. The authors concluded that using PAH Koc values derived from spiked sediments and from existing predictive models can greatly over estimate the actual PAH partitioning from sediments into the overlying water. These large differences indicate that PAH and petroleum hydrocarbons in actual field soils are more tightly sorbed than would be expected from Koc and Kow found in references. Thus, soil and sediment site-specific Kd or Koc values should be obtained and used to identify actual human and ecological risks at a site and to identify the need for remediation actions at field sites of concern.

3.5 INTEGRATED ASSESSMENTS Several evaluations have attempted to assess comprehensive aspects of chemical availability and bioavailability using field samples. In one study (Salanitro et al., 1997), the extent of hydrocarbon degradation, earthworm and plant toxicity and chemical availability of soils containing crude oil was evaluated. The samples were silt loam soils containing 0.3% and

72

REDUCING RISKS––AN ENVIRONMENTAL ENGINEERING PERSPECTIVE

4.7% petroleum hydrocarbon (4000–27,000 mg/kg measured as total petroleum hydrocarbons (TPH). Bioremediation after 3–4 months resulted in 50–75% and 10–90% loss of the initial TPH for the two samples respectively. Even though considerable TPH remained in the bioremediated soils, the bioremediated soils were not toxic to earthworms nor inhibited seed germination. The results indicated that the remaining petroleum hydrocarbons, which ranged from 1000 to 8600 mg/kg as TPH, were tightly bound to the soils. In addition, these residual hydrocarbons were not able to be degraded further and were not susceptible to leaching. A comprehensive study of the availability of PAH in soils at a site in California has occurred (Stroo et al., 2000). Site samples were subjected to a battery of physical and biological tests that focused on determining the availability of soil-bound PAH. The results demonstrated that sorption of the PAH to the soil particles, weathering, and biological treatment significantly reduced PAH chemical availability and bioavailability. When the reduced site-specific availability results were included in risk assessment calculations, the potential site risk was considerably less than that determined using California risk assessment default assumptions. Additional evaluation of field samples from the above California site (Stroo et al., 2005b) demonstrated that the dermal and ingestion absorption factors for the PAH in those samples were far lower than factors that resulted from using default assumptions in common risk assessments. Using sample-specific absorption factors in standard risk assessment equations increased the required risk-based clean-up levels by an average factor of 72 with a range of from 23 to 142. These results indicated that, at this site, higher concentrations of the PAH could be left in the site soil. It was stated that a tiered evaluation protocol using site-specific chemical analyses, chemical release data and in vitro bioassays should be used to establish realistic site-specific risk management criteria. A more specific example of the value of site-specific data, rather than default assumptions, was obtained when the chemical bioavailability of benzo(a)pyrene (BaP) in industrial site soil was evaluated (Stroo et al., 2005a). Results indicated that the site-specific BaP dermal absorption factors were 14 to 107 times lower than the regulatory default assumption factors. In terms of a site-specific risk assessment, this resulted in the calculated excess cancer risk being reduced by 97% on average. From the ecological standpoint, field data have indicated that, at sediment sites, toxicity to aquatic organisms is not related to the concentration of total measured PAH in the sediment. Instead, such toxicity is correlated to the concentration of the easily released and bioavailable PAH. It also has been noted (Cornelissen and Gustafsson, 2005) that the strong sorption of hydrophobic organic compounds to anthropogenic black carbon is the cause of the widely different biota to sediment accumulation factors (BASFs) that occur at different sites. Such recent information reinforces the desirability of obtaining and using site-specific chemical availability data at a soil or sediment site of concern to identify human health and ecological risks.

3.6 SUMMARY The concept of environmental risk and its use for environmental decision making and site remediation selection is now an important aspect of environmental engineering activities. The use of environmental risk for decision making and treatment or remedy selection has resulted from the recognition that (a) not all environmental problems are equally serious;

REFERENCES

73

(b) the resources for environmental protection are not limitless; and (c) the focus should be on sites and situations that pose the greatest risk to human health and the environment. In the past several decades, there has been considerable progress improving the knowledge related to all parts of the risk assessment process. There now is better scientific knowledge about pharmacological bioavailability that can be used for human health and ecological assessments. Equally important is the increased knowledge that has been developed about chemical release from contaminated sites, about chemical environmental availability, about chemical environmental bioavailability and about chemical transport and transformations at real-world sites. Such information, when used with a tiered RBCA type framework, will allow environmental engineers, regulators, and the concerned public to determine environmentally sound and protective site management and remediation decisions. In addition, scientific and field information has made it increasingly clear that site-specific data should be used for risk-based decision making rather than depending solely on default assumptions and parameters.

REFERENCES American Society for Testing Materials (ASTM) (1995) Guide for Risk-Based Corrective Action Applied at Petroleum Release Sites. Philadelphia, PA. E-1739–1795. American Society for Testing Materials (ASTM) (1999) ASTM Standards on Assessment and Remediation of Petroleum Release Sites. West Conshohocken, PA. Berg MS, Loehr RC, Webster MT (1998) Release of petroleum hydrocarbons from contaminated soils. J. Soil Contam.7:675–695. Bucheli TD, Gustafsson O (2000) Quantification of the soot water distribution coefficient of PAHs provides mechanistic basis for enhanced sorption observations. Environ. Sci. Technol. 34:5144–5151. Cornelissen G, Gustafsson O (2005) Prediction of large variation in biota to sediment accumulation factors due to concentration-dependent black carbon adsorption of planar hydrophobic compounds. Environ. Toxicol. Chem. 24:495–498. Cornelissen G, Rigterink H, Ten Hulscher TEM, Vrind BA, van Noort PCM (2001) A simple Tenax extraction method to determine the availability of sediment-sorbed organic compounds. Environ. Toxicol. Chem. 20:706–711. Dondelle M, Loehr RC (2002) Comparison of estimated and experimentally obtained soil water distribution coefficients. Pract. Periodical Hazard. Toxic Radioactive Waste Manage. 6:218–226. Graham JD (1994) A Historical Perspective on Risk Assessment in the Federal Government. Boston, MA: Center for Risk Analysis, Harvard School of Public Health. Hawthorne SB, Miller DJ (2003) Evidence for very tight sequestration of BTEX compounds in manufactured gas plant soils based on supercritical fluid extraction and soil/water partitioning. Environ. Sci. Technol. 37:3587–3594. Hawthorne SB, Poppendieck DG, Grabanski CB, Loehr RC (2001) PAH release during water desorption, supercritical carbon dioxide extraction and field bioremediation. Environ. Sci. Technol. 35:4577–4583. Hawthorne SB, Poppendieck DG, Grabanski CB, Loehr RC (2002) Comparing PAH availability from manufactured gas plant soils and sediments with chemical and biological tests. Part 1. PAH release during water desorption and supercritical carbon dioxide extraction. Environ. Sci. Technol. 36:4795–4803.

74

REDUCING RISKS––AN ENVIRONMENTAL ENGINEERING PERSPECTIVE

Hawthorne SB, Grabanski CB, Miller DJ (2006) Measured partitioning coefficients for parent and alkyl polycyclic aromatic hydrocarbons in 114 historically contaminated sediments. Part 1. Koc values. Environ. Toxicol. Chem. 25:2901–2911. Hong L, Ghosh U, Mahajan T, Zare RN, Luthy RG (2003) PAH sorption mechanism and partitioning behavior in lampblack-impacted soils from former oil–gas plant sites. Environ. Sci. Technol. 37:3625–3634. Howard PH (1990) Handbook of Environmental Fate and Exposure Data for Organic Chemicals.Ann Arbor, MI: Lewis Publishers. Jonker MTO, Koelmans AA (2002) Sorption of polycyclic aromatic hydrocarbons and polychlorinated biphenyls to soot and soot-like materials in the aqueous environment. Environ. Sci. Technol. 36:3725–3734. Jonker MTO, Hawthorne SB, Koelmans AA (2005) Extremely slowly desorbing polycyclic hydrocarbons from soot and soot-like materials: Evidence by supercritical fluid extraction. Environ. Sci. Technol. 39:7889–7895. Linz DG, Nakles DV, editors (1997) Environmentally Acceptable Endpoints in Soil. Annapolis, MD: American Academy of Environmental Engineers. Loehr RC, Lamar MR, Poppendieck DG (2003) A protocol to estimate the release of anthropogenic hydrocarbons from contaminated soils. Environ. Toxicol. Chem. 22:2202–2208. Lowrance MW (1976) Of Acceptable Risk: Science and the Determination of Safety. Los Altos, CA: W. Kaufmann, Inc. Lyman WJ, Reehl WF, Rosenblat DH, editors (1982) Handbook of Chemical Property Estimation Methods. Washington, DC: American Chemical Society. Mackay D, Shiu WY, Ma KC (1992) Illustrated Handbook of Physical–Chemical Properties and Environmental Fate for Organic Chemicals, Volume II: Polynuclear Aromatic Hydrocarbons, Polychlorinated Dioxins and Dibenzofurans. Ann Arbor, MI: Lewis Publishers. McMillen SJ, Magaw RJ, Carovillano RL, editors (2001) Risk-Based Decision-Making for Assessing Petroleum Impacts at Exploration and Production Sites Tulsa, OK: U.S. Department of Energy, National Energy Technology Laboratory. National Academy of Sciences (NAS) (1994) Science and Judgment in Risk Assessment. Washington, DC: National Academy Press. National Research Council (NRC) (1983) Risk Assessment in the Federal Government: Managing the Process.Washington, DC: National Academy Press. Oen AMP, Breedveld GD, Kalaitzidis S, Christanis K, Cornelissen G (2006) How quality and quantity of organic matter affect polycyclic hydrocarbon desorption from Norwegian Harbor sediments. Environ. Toxicol. Chem. 25:1258–1267. Paustenbach DJ (1989) The Risk Assessment of Environmental and Human Health Hazards: A Textbook of Case Studies. New York: John Wiley & Sons. Paustenbach DJ, editor (2002) Human and Ecological Risk Assessment: Theory and Practice.New York: John Wiley & Sons. Reilly WK (1990) Aiming before we shoot: the quiet revolution in environmental policy. Presentation at the National Press Club, September 26, 1990, Washington, DC. Washington, DC: U.S. Environmental Protection Agency, pp.207–1011. Salanitro JP, Dorn PB, Huesemann MH, Moore KO, Rhodes IA, Rice Jackson LM, Vipind TE, Western MM, Wisniewski HL (1997) Crude oil hydrocarbon bioremediation and soil ecotoxicity assessment. Environ. Sci. Technol. 31:1769–1776. Stroo H, Jensen R, Loehr RC, Nakles DV, Fairbrother A, Liban CB (2000) Environmentally acceptable endpoints for PAHs at a manufactured gas plant site. Environ. Sci. Technol. 34:3831– 3836.

REFERENCES

75

Stroo H, Roy TA, Liban CB, Kreitinger JP (2005a) Dermal bioavailability of benzo(a)pyrene on lampblack: Implications for risk assessment. Environ. Toxicol. Chem. 24:1568–1601. Stroo H, Nakles DV, Kreitinger JP, Loehr RC, Hawthorne SB, Holman H-Y, Lapierre A (2005b) Improving risk assessments for manufactured gas plant soils by measuring PAH availability. Integr. Environ. Assess. Manage. 1:259–266. U.S. Environmental Protection Agency (1984) Risk Assessment and Management: Framework for Decision Making. Washington, DC. EPA/600/9-85-002. U.S. Environmental Protection Agency (1988) Guidance for Conducting Remedial Investigations and Feasibility Studies Under CERCLA. Washington, DC: Office of Solid Waste and Emergency Response. EPA/540/6-89/004. U.S. Environmental Protection Agency (1989a) Risk Assessment Guidance for Superfund, Volume 1: Human Health Evaluation Manual.Washington, DC: Office of Emergency and Remedial Response. EPA/540/1-89/002. U.S. Environmental Protection Agency (1989b) Risk Assessment Guidance for Superfund, Volume 2: Environmental Evaluation Manual. Washington, DC: Office of Emergency and Remedial Response. EPA/540/1-89/001. U.S. Environmental Protection Agency (1989c) Ecological Assessments of Hazardous Waste Sites: A Field and Laboratory Reference Manual. Washington, DC: Office of Research and Development. EPA/600/3-89/013. U.S. Environmental Protection Agency (1990) Reducing Risk: Setting Priorities and Strategies for Environmental Protection. Washington, DC: Science Advisory Board. SAB-EC-90-021. U.S. Environmental Protection Agency (1991a) Risk Assessment Guidance for Superfund, Volume 1: Human Health Evaluation Manual (Part B, Development of Risk-Based Preliminary Remediation Goals). Washington, DC. EPA/540/R-92/003. U.S. Environmental Protection Agency (1991b) Risk Assessment in Superfund: A Primer. Washington, DC: Office of Emergency and Remedial Response. EPA/540/X91/002. Verschueren K (1996) Handbook of Environmental Data on Organic Chemicals, 3rd edn. New York: Van Nostrand Reinhold. Zimmerman J, Werner D, Ghosh U, Millward RN, Bridges TS, Luthy RG (2005) Effects of dose and particle size on activated carbon treatment to sequester polychlorinated biphenyls and polycyclic aromatic hydrocarbons in marine sediments. Environ. Toxicol. Chem. 24:1595–1601.

4 CLINICAL PERSPECTIVE ON RESPIRATORY TOXICOLOGY Mark J. Utell and Jonathan M. Samet

The chapters in this volume have described the adverse effects of diverse agents on the lung and, to a lesser extent, on other target organs. These effects have been characterized through multidisciplinary approaches, typically involving in vitro and in vivo laboratory studies, and often controlled human exposures and epidemiological investigations as well. The resulting evidence on adverse effects may have substantial societal impact through regulatory and nonregulatory pathways. However, exposed individuals sustain the associated risk and disease burden. In this chapter, we address the extension of this research to the diagnosis and management of environmentally related disease in individual patients and in exposed groups. The care of individuals falls to the practicing physician or other health care providers. Most often, the physician addresses the fears a patient raises about the consequences of exposure or manages the disease that may have resulted from or been exacerbated by exposure. The physician may also become enmeshed in legal questions as to the causation of disease and to compensation. In this chapter we consider the clinical approach to the individual who has been exposed to a respiratory hazard. We review the tools available to the health care provider with the purpose of describing clinical methods and their strengths and limits in answering the frequently complex questions that arise around environmental exposures. The evaluation of exposed populations, for example, specific worker groups, may also fall to health care providers. However, to address the effects of exposure on populations, the approach must extend beyond simply evaluating individuals to provide measures of impact on groups. Epidemiological studies, typically cross-sectional surveys, are often conducted to evaluate potential adverse respiratory or other effects. In this review, we consider the tools used to conduct such surveys. Although a survey might be performed in response to recognition of a perceived hazard, persons exposed to potential or known hazards may be subject to ongoing surveillance, often required in the workplace. We also review surveillance

Environmental Toxicants, Third Edition Edited by Morton Lippmann Copyright
2009 John Wiley & Sons, Inc.

77

78

CLINICAL PERSPECTIVE ON RESPIRATORY TOXICOLOGY

methodology. Of necessity, our treatment of these subjects is limited. Textbooks on environmental medicine are available (Rom, 2007; Rosenstock et al., 2005). Additionally a series of case studies has been published by The Institute of Medicine (1995).

4.1 CONCEPTS OF EXPOSURE Patterns of time use and activity place people in diverse indoor and outdoor environments throughout the day. Personal exposure to air pollutants represents the time-weighted average of pollutant concentrations in microenvironments, environments having relatively homogeneous air quality (Sexton and Ryan, 1988), with each microenvironment possibly having its own unique set of air contaminants. The characteristics of these microenvironments, and the time spent in them, vary with age, sex, and other sociodemographic factors. Diverse microenvironments that may be relevant for health include the home, the workplace, schools, outdoors, transportation environments, and locations where recreational time is spent. The clinician’s approach to assessing exposure in these microenvironments primarily involves interviewing the patient. Standardized instruments for collecting information on environmental and occupational exposures have been published (Occupational and Environmental Health Committee, 1983), but clinicians generally take exposure histories in idiosyncratic ways, the completeness of the history reflecting the clinician’s training, fund of knowledge, and familiarity with the environments of concern to specific patients. The clinical history of exposures may touch on a few widely know hazards, for example, asbestos, but rarely inventories the duties of specific jobs, the materials handled, or the use of respiratory protection. Most physicians have limited knowledge of the exposures associated with specific occupations. Biomarkers of exposure allow assessment of exposure to a chemical on the basis of its measurement in a biologic fluid or tissue. Biological markers of exposure are available on a routine clinical basis for only a few inhaled pollutants, including, for example, carboxyhemoglobin for carbon monoxide and blood or urine lead level for inhaled lead (National Research Council, 1989). Levels of other toxicants can be measured in blood or other samples, if needed, for example, measurement of dioxins in blood or blood lipids, pesticides in blood, mercury in hair, benzene in exhaled breath, cotinine in saliva, and cadmium in urine. In itself, quantification of a specific biomarker in a biologic material establishes only that the chemical is present in the body. If the substance is not otherwise known to be endogenous, an exposure from the outside environment can be inferred. The clinical significance of the exposure depends on factors such as reversibility, dose, individual susceptibility, and the health status of the individual (National Research Council, 2006). Serum antibodies or skin test reactivity to intradermal antigen injection can be measured to provide an indication of past exposure and the development of sensitivity to selected antigens; tests are available for some common biological antigens, such as the house dust mite, but for only a few inhaled chemicals. Routine clinical tests for adducts and antibodies to adducts are not yet available, nor are tests readily available for most intermediate endpoints that may be relevant from a toxicological perspective. Thus, the routine clinical evaluation does not usually provide comprehensive information concerning environmental exposures. In fact, the routine history taken by a primary care provider usually addresses only tobacco smoking and the current or usual employment; some common exposures with well-known consequences, such as asbestos, may also be covered.

TOOLS FOR STUDYING INDIVIDUALS

79

Often, however, it is the identification of a disease known to be caused by a specific agent that prompts full questioning of the patient concerning relevant exposures and thus, routine medical records do not usually offer any more than a superficial assessment of inhalation exposures. Alternatively, the occurrence of an uncommon syndrome or a clustering of cases may result in more complete evaluation. A physician trained in environmental or occupational medicine routinely obtains more detailed and disease-relevant information; this type of physician should be consulted in cases involving possible effects of complex environmental exposures. Physicians trained in pulmonary medicine may also have special expertise related to environmental lung disease, and allergists may be appropriate for addressing workplace-related allergic disorders. In evaluating the chest X-ray for the pneumonoconioses, fibrotic disorders of the lung, the chest X-ray may be interpreted according to a standard system, the International Labour Office (ILO) classification (International Labour Office, 1980). The “B Reader” certification uses this system, and is given by the National Institute for Occupational Safety and Health (NIOSH) to persons who have taken a 2-day course and successfully passed an examination involving standardized interpretation of chest roentgenographs showing various occupational lung diseases. This certification demonstrates competency in interpreting patterns of radiographic abnormality, but does not specifically establish expertise in occupational lung disease.

4.2 TOOLS FOR STUDYING INDIVIDUALS 4.2.1

Overview

Environmental diseases frequently masquerade as common medical disorders. From a clinical perspective, primary care physicians underdiagnose environmentally and occupationally related disorders and far too infrequently enter them into their differential diagnosis. For example, more than 80% of occupational or environmental disease diagnoses had not been correctly recognized prior to evaluation in an occupational medicine clinic, although most patients had consulted one or more physicians (Cullen and Cherniack, 1989). The inadequate knowledge base of primary care physicians in environmental medicine (Institute of Medicine, 1988) and the physician shortages in these specialized disciplines have been emphasized (Castorina and Rosenstock, 1990). Unfortunately, there has been little progress since the 1988 Institute of Medicine report on environmental and occupational medicine. However, with increased State and Federal Support, Children’s Centers of Excellence in Environmental Health have been established in a number of communities. These facilities serve as resources both for more comprehensive investigation of potentially exposed children and for enhanced physician training resulting in earlier recognition of environmentally mediated diseases. In considering environmentally induced lung diseases, it is important to recognize that symptoms and signs of lung damage are seldom indicative of the specific injuring agent. For the clinician, the recognition of occupational or environmental respiratory illness may be obscured by the nonspecificity of symptoms, findings on physical examination, alterations of pulmonary function, or radiographic changes. Lung biopsy is rarely indicated to confirm a diagnosis of environmentally related disease, and even when tissue is available, pathological findings may be nonspecific and not indicate a specific etiologic agent. For example, airways inflammation may reflect the consequence of a myriad of inhaled agents. Specific links

80

CLINICAL PERSPECTIVE ON RESPIRATORY TOXICOLOGY

between exposure and disease, such as beryllium exposure and noncaseating granulomas, are few. Rarely, specific diagnostic fibers or dusts may be identified in lung tissue. In addition, exposures are often multiple or mixed, symptoms may be minimal until disease is advanced or more severe than anticipated for the extent of physiological and radiographic abnormality, and latency periods between onset of exposure and disease development may be long. Identifying occupationally related lung disease is further complicated by the frequent overlay of cigarette smoking and the possibility of additive or even synergistic effects acting in combination with other agents (e.g., asbestos-related lung tumors); thus, the association between lung disease and occupational exposure is often neither obvious nor simple in a particular patient. The clinician is now called on to deal not only with the more classical occupational diseases, like the pneumoconioses, but also the more problematic “environmental” illnesses of the past three decades. For example, as new construction techniques and ventilation practices were directed at conserving energy, and sealed buildings began to age, outbreaks of nonspecific complaints characterized by persistent respiratory symptoms, headaches, and lassitude began to occur among workers, who attributed their symptoms to the indoor environments where they worked (Samet et al., 1988). Now referred to as “sick building syndrome,” these outbreaks continue—but seemingly in smaller numbers than 10 years ago. Even more perplexing is the appearance of illness in previously healthy individuals who, following accidental exposure to solvents, gases, or irritating fumes, develop unrelenting respiratory complaints, lethargy, and central nervous system dysfunction when subsequently exposed to trace amounts of the material or other irritants such as fumes or cigarette smoke. This symptom complex has been labeled “multiple chemical sensitivities” (MCS)” (Cullen, 1987); it poses substantial diagnostic and therapeutic difficulty. Perhaps, the most recent environmental concern is disease resulting from potential exposure to mold in homes, particularly related to water damage. The flooding in New Orleans and Houston as a result of catastrophic hurricanes has served to reinforce the potential for chronic exposure to mold. In this section, we focus on the tools available to the clinician for evaluating the individual patient suspected of having an environmentally related pulmonary disorder. The emphasis on respiratory disorders is highly relevant: respiratory diseases are the most frequently diagnosed work-related conditions in industrial regions, and the best-established medical consequences of mining and farming involve the lower respiratory tract. Also, the airway is most important portal of entry for toxic agents from the environment. New technologies, such as nanotechnology, will introduce new agents into the environment and workplace with potential respiratory threats to the individual as well as larger populations. The tools commonly available for evaluating the individual are shown in Table 4.1; the general principles also apply to the evaluation of environmentally related nonrespiratory disorders. 4.2.2

Clinical Approaches

For the physician, the central problem in environmental lung medicine is determining that a respiratory symptom or particular structural or functional derangement of the lung is caused by a certain inhaled agent. Table 4.2 lists examples of the pathophysiological responses of the respiratory tract to environmental particles and gases. As with all disorders associated with the environment, a careful and thorough history is mandatory. Both the work and home environments need to be considered for exposure to known allergens, irritants, chemicals or organic dusts. Careful inquiry is necessary concerning not only the materials the individual is

TOOLS FOR STUDYING INDIVIDUALS

81

TABLE 4.1 Tools for Evaluating Individuals with Suspected Occupational or Environmental Lung Disease General evaluation Medical and respiratory history Detailed occupational history Identification of materials Identification of toxic responses Chest X-ray Special studies Immunologic and skin tests Airway hyperreactivity testing CAT scan Fiberoptic bronchoscopy with bronchoalveolar lavage Industrial hygiene information Site or plant visit Epidemiological investigation

working with, but also those being used by coworkers. The occurrence of similar problems in coworkers also should be assessed. Other clues in the history are useful. With occupational asthma, there is often a latent period between the first exposure to the offending agent and the onset of asthma (Chan-Yeung, 1990). This period may vary from a few weeks to over 20 years. Therefore, the physician attempts to correlate temporally respiratory and/or systemic symptoms with exposure in a particular environment, although varying temporal relationships between exposure and outcome may cloud interpretation of the clinical history. Thus, in some instances, the causal relationship between exposure and symptoms may not be readily recognized. Correlation of symptom occurrence with cumulative exposure may be difficult in cases of late-onset, delayed, or repetitive patterns of response, or in those responses in which cough, chest tightness, or malaise predominates. This contrasts with the clinical picture of a worker who develops symptoms immediately and repeatedly when working with a particular substance. Rarely, following a catastrophic event such as occurred with the attack on the World Trade Center on September 11, 2001, an inhalation exposure to a

TABLE 4.2 Pathophysiological Responses of Respiratory Tract to Occupational and Environmental Particles and Gases Site Response Nose Airways

Parenchyma

Agent

Potential

Pollen Formaldehyde Sulfur dioxide, nitrogen dioxide High- and low-molecular weight chemicals Aeroallergens Formaldehyde, wood smoke Radon, asbestos Inorganic dusts Thermophilic actinomycetes, fungi

Hay fever, rhinitis Nasal cancer Bronchoconstriction Asthma Asthma Irritation, cough Cancer Pneumoconiosis Hypersensitivity pneumonitis

82

CLINICAL PERSPECTIVE ON RESPIRATORY TOXICOLOGY

“pollutant-mix” may cause immediate symptoms and potentially more chronic respiratory disease; the evaluation of both the acute and more chronic symptoms in an individual will likely be initiated by a primary care practitioner; if chronic symptoms are recognized in a significant subgroup, this subpopulation may undergo more intensive surveillance (Herbert et al., 2006). The initial care of individual patients exposed to potentially toxic materials usually falls to the practicing primary care physician or other health care providers. Typically, the exposure history obtained by such practitioners is insufficient, and only characterizes the workplace qualitatively, as “dusty” or “extensive smoke and fumes.” Major effort should be extended to accurately identify the specific toxic agent or agents present in each work site. This may require direct contact with employers and/or assistance from governmental agencies such as NIOSH or the Occupational Safety and Health Administration (OSHA) for information. Some efforts have been focused on development of occupational/ environmental history forms to assist the clinician in collecting the data base (Occupational and Environmental Health Committee, 1983; Kilbourne and Weiner, 1990; Agency for Toxic Substances and Disease Registry, 1993). The newer problems of sick building syndrome and multiple chemical sensitivity may be readily recognized if the patient presents with an unmistakable clinical picture (Spengler et al., 2000; Brightman, 2000.) The clinician is faced with an individual with a myriad of complaints, often including intractable upper respiratory symptoms, ill-defined central nervous system dysfunction, headache, fatigue, and low productivity that are attributed to odors, poor ventilation or other aspects of a poorly-ventilated building. Despite a careful physical evaluation, there may be no physical signs revealed on examination. In outbreaks with an identified etiology, a spectrum of causative agents has been identified: infectious agents, specific air contaminants, and environmental conditions such as temperature and humidity (Finnegan and Pickering, 1986; American Thoracic Society, 1997). Outbreaks without an identifiable etiology have frequently occurred in tight buildings and have given rise to the “sick building syndrome.” However, these outbreaks are by no means limited to tight buildings; they may occur because of maintenance problems, changing uses, and other factors. An even more puzzling syndrome is the so-called “MCS” in which the individual becomes “sensitized” to almost all organic and synthetic chemicals (Cullen, 1987; Ashford and Miller, 1998). The history is typically that of a healthy individual who after accidental exposure to a solvent or irritant gas or fumes develops progressive symptoms with exposure to traces of the original agent or a variety of nonspecific agents. The assessment lies in the historical data, as the remainder of the evaluation is quite unremarkable. Many hypotheses have now been examined, and there are few known, extrinsic causes of MCS supported by research. Immunologic abnormalities now appear to be a less plausible cause (Simon et al., 1993). It has been suggested that the clinician is dealing with a variant of a post-traumatic stress disorder (Schottenfeld and Cullen, 1986) and case reports have revealed a probable role of conditioned response, especially to odor. Although no well-controlled studies establish a clear mechanism, the MCS patient often requires comprehensive evaluation and compassionate support. Several major medical societies including the American College of Physicians (1989), the American Medical Association (1992), and the American College of Occupational and Environmental Medicine (1999) have concluded that the existence of MCS is an unproven hypothesis, and that credible scientific research is needed. Physical examination is often unrevealing in cases of environmental lung disease. There should be a full evaluation of the nose, oropharynx, and chest and lungs. Abnormal lung

TOOLS FOR STUDYING INDIVIDUALS

83

sounds such as bibasilar rales (cracking sounds) are heard with interstitial lung disease, as in asbestosis, which results from asbestos exposure. Signs of airway disease, such as wheezing and cough, are typically associated with environmental asthma. However, neither a normal examination, nor one in which wheezing persists even after avoidance of the work environment, excludes environmental airway disease from the diagnostic possibilities. There may be a need to systematically correlate symptoms and lung function levels with activities and exposure in the work environment and at home. There are no pathognomonic findings with sick building syndrome or multiple chemical sensitivities. 4.2.3

Imaging Studies

Since the clinician is likely to obtain a chest X-ray in the evaluation of any individual with respiratory complaints, it is worth emphasizing that radiological examination remains a major diagnostic tool for revealing occupationally induced interstitial lung disease. The principle disorders amenable to radiological diagnosis are the pneumoconioses, which include asbestosis, coal workers’ pneumoconiosis, silicosis, and berylliosis. The chest X-ray in asbestos exposure may serve to identify biological markers of exposure or evidence of disease providing documentation of fiber exposure of the parenchyma and/or pleural reactions. Workers exposed to relatively low concentrations of asbestos fibers may develop a variety of pleural processes including effusions, plaques, or diffuse thickening; the pleural changes serve as markers of asbestos exposure and rarely are associated with symptoms or functional changes (Craighead and Mossman, 1982). In contrast, workers exposed to high concentrations of asbestos fibers may develop interstitial infiltrates, primarily in the lower lung fields, characterized by an irregular, linear pattern. Persons with this pattern typically have reduced lung function. Workers exposed to other dusts such as silica and/or coal mine dust typically demonstrate bilateral upper lobe interstitial infiltrates (Fraser et al., 1990). These infiltrates characteristically are nodular in appearance; the nodules range in size from 1 to 10 mm, may coalesce, and eventually distort or cause retraction in the hilar region. It should be emphasized that the diagnosis of pneumoconiosis in the absence of an abnormal chest radiograph is highly unusual. The rare patient in this category, symptomatic but with a normal chest X-ray, would require additional supportive data to substantiate the diagnosis. For the clinician evaluating the patient with suspected environmental or occupational asthma, the chest X-ray may not be of great diagnostic help. Often it is normal, or reveals only minimal hyperinflation. The patient with hypersensitivity pneumonitis will often demonstrate diffuse infiltrates on the radiograph. Finally, occupationally related pulmonary malignancies present with the various X-ray patterns of lung cancers associated with smoking. The radiographic changes in various occupational lung diseases have been described in detail (Fraser et al., 1990). But even with a radiographic pattern specific for an occupational lung disease, there continues to be debate regarding the radiographic criteria for diagnosis. For example, an American Thoracic Society (ATS) committee in 1986 concluded chest roengenographic evidence of small irregular opacifications of a profusion of 1/1 or greater was required to diagnose asbestosis (American Thoracic Society, 1986), whereas the 2004 ATS committee concluded that profusion 1/0 was adequate (American Thoracic Society, 2005). More sophisticated imaging techniques used in clinical practice, such as computed tomography (CT), are not currently applicable in large surveys, but may be useful in the evaluation of specific chest X-ray abnormalities. In environmental medicine, the CT scan is

84

CLINICAL PERSPECTIVE ON RESPIRATORY TOXICOLOGY

an important modality in locating and characterizing small lesions that could represent early lung tumors. The CT scan has been useful in assessing the location, extent, and potential for surgical resection of small lung lesions and masses. In addition, the CT scan has been useful in confirming the presence of pleural plaques noted on chest X-ray; it also raises the possibility that there are large groups of persons with pleural disease undetectable on standard radiographs (Friedman et al., 1988). High resolution chest CT (HRCT) can detect pneumoconiotic small opacities or abnormal lung tissue associated with diffuse forms of interstitial fibrosis not visible on plain chest X-ray (Aberle et al., 1988). Schwartz et al. (1990) observed that among asbestos-exposed patients with normal parenchyma on plain chest X-ray, patients with pleural disease are more likely than those with normal pleura to have interstitial abnormalities on the HRCT. Clearly, prospective, controlled evaluations are required to determine the prognostic significance of these abnormalities on HRCT among exposed persons with normal-looking parenchyma by chest X-ray. 4.2.4

Pulmonary Function Testing

Tests of pulmonary function are useful tools for evaluating the physiologic consequences of lung damage caused by inhaled materials. In evaluating individual patients with respiratory complaints or with potentially toxic exposures, physicians obtain pulmonary function tests. In fact, failure to obtain such studies almost invariably represents inadequate clinical evaluation. Guidelines for characteristics of pulmonary function equipment, test interpretation, and quality assurance are available (Clausen, 1982; American Thoracic Society, 1995; American Thoracic Society/European Respiratory Society, 2005). The clinical utility of physiological function testing in individual patients lies in finding evidence of response to inhaled toxic agents. The tests can show the functional manifestations of structural changes in the respiratory system, whether the changes are transient (e.g., reflex bronchoconstriction) or permanent (e.g., fibrosis). However, these tests have little specificity for effects of specific environmental agents. The lung responds to injury in a limited number of ways, and the physiological manifestation of various types of injury is often the same, regardless of the causative agent. Thus, abnormal lung function tests are a general, rather than a specific, indication of injury. One type of testing, inhalation challenge with specific agents may lead to more specific findings. The real workhorse in the evaluation of the individual patient is simple spirometry, a test that can be readily performed in the field or office setting. The spirometric tracing, which describes the rate of forced exhalation, distinguishes the obstructive pattern from restrictive patterns of physiologic impairment. Detailed reviews of the use of pulmonary function testing in evaluation of the patient with occupational lung disease are available (Townsend, 2000; McKay and Lockey, 1991). In brief, the restrictive pattern, typical of interstitial lung disease, is suggested by a forced vital capacity (FVC) <80% predicted and FEV1/FVC >75%. In contrast, the obstructive pattern, typical of airways disease, includes an FVC >80% predicted, FEV1 <80% predicted, and FEV1/FVC <75%. In addition, the maximum voluntary ventilation (MVV) is usually reduced below 80% predicted. Predicted values take into account patient age, height, and sex. Although the MVV is a valuable test when properly performed, it requires considerable patient cooperation and effort. A reduction in the forced expiratory flow from 25 to 75% of FVC, FEF 25–75%, below 70% predicted with otherwise normal flow rates is a marker of early but mild airway obstruction. The restrictive pattern is the hallmark of interstitial lung diseases, such as asbestosis, but is not specific and can result from such diverse processes as large pleural effusions, muscle

TOOLS FOR STUDYING INDIVIDUALS

85

weakness, and even marked obesity. In the presence of significant obstructive airway disease, as in smoking-caused chronic obstructive pulmonary disease, it is difficult to interpret the spirogram for restrictive changes unless these are extremely severe, since obstructive disease may also reduce the forced vital capacity. In this case, more sophisticated measurements of lung volumes using body plethysmography or helium dilution techniques may be necessary. Additionally, imaging can be informative. Restrictive disease is classically defined by a reduced total lung capacity below 80% predicted. A reduction in the diffusing capacity for carbon monoxide may be valuable in supporting a diagnosis of interstitial lung disease. Although cigarette smoking is the most common cause of the obstructive pattern, a history of smoking should not preclude careful investigation into occupational exposures as contributing if not causal factors for physiological impairment. All too often an abnormality in lung function is attributed to cigarettes to the exclusion of other potential agents. On the contrary, some of the impairment in persons with occupational lung disease may be due to smoking. Smokers are usually diagnosed with chronic obstructive respiratory diseases when the level of airflow limitation interferes with activities that they would otherwise be able to perform (Speizer and Tager, 1979). There is a continuum between health and respiratory disease. Disability typically occurs when the FEV1 approaches 40–50% of the predicted value. Pulmonary clinicians consider the reduction of FEV1 below 80% of the predicted value as indicative of obstructive disease; as the reduction in the FEV1 progresses, the severity of disease increases. In the evaluation of potential occupational or environmental asthma, it is important to objectively establish the relationship of exposure to symptom occurrence and reduction of lung function. The serial recording of peak expiratory flow rates (PEFR) with a mini-peakflow meter by the patient at work or at home may be a valuable method of establishing a causal relationship. These devices are inexpensive and sufficiently accurate if used properly. The serial measurement of lung function generally provides stronger information on the effect of exposure than one or two spirometric measurements spaced over time. In following the PEFR, the individual typically measures and records the value every 2–4 h from waking to sleeping (Chan-Yeung, 1990). A record is maintained for several work weeks, often followed by at least 1 week away from any work exposure. Although criteria for a positive response have not been well established, the clinician is seeking to find a pattern of exposure followed by symptoms and a change in PEFR. Improvement in lung function and reduction in symptoms during the week away from work are also helpful in confirming a diagnosis. A major concern with PEFR measurement is the effort dependence of the test, such that inadequate respiratory effort can lead to misleading data. Specialized testing, including an assessment of airway hyperreactivity, may be needed to evaluate environmental asthma. Nonspecific airway hyperreactivity is defined as an exaggerated bronchoconstrictor response to a variety of chemical, physical, and pharmacological stimuli. There is now nearly a consensus that nonspecific airway hyperreactivity is a characteristic shared by virtually all asthmatics (Boushey et al., 1980; National Institutes of Health, 1997); that is, the asthmatic develops bronchoconstriction after inhaling a lower concentration of a provoking agent than is needed to cause a similar degree of change in airway tone in a healthy subject. In the laboratory, airway reactivity testing is divided into two general categories, depending on the choice of nonspecific versus specific agents. In both, the increased bronchoconstrictor response is assessed with pulmonary function tests. Nonspecific stimuli include pharmacological agents such as methacholine, carbachol, and histamine; exercise; hyperpnea with cold or dry air; and inhalation of hypertonic or hypotonic aerosols. Although

86

CLINICAL PERSPECTIVE ON RESPIRATORY TOXICOLOGY

pharmacological challenge is used most often in the clinical laboratory, it is less suitable for population studies, especially those involving children because of safety considerations. Cold-air challenge with hyperventilation has been used effectively, and response to cold air is generally correlated closely with methacholine responses (O’Byrne et al., 1982). Challenge with specific agents, common antigens, chemicals such as isocyanates, and organic materials such as plicatic acid (from western red cedar) can be used to identify specific sensitizing agents. These approaches can be particularly powerful in incriminating occupational chemicals and confirming the diagnosis of occupation-related airway disease; challenge may provoke immediate and/or late pulmonary responses that do not resolve spontaneously. Even with specific agents, the interpretation of responses can be difficult and be confounded by a variety of factors, such as dose and irritant effects (McKay, 1986). In the assessment of asthma induced by occupational agents, airway reactivity testing with nonspecific and specific agents serves a diagnostic function (Eisen et al., 1993). ChanYeung (1995) and Chan-Yeung and Lam (1986) have published comprehensive reviews on the subject of occupational asthma and the role of airway reactivity testing. Nonspecific airway hyperreactivity occurs in most workers with occupationally induced asthma despite their not having the usual factors such as atopy. Furthermore, Lam et al. (1983) found a good correlation between the degree of nonspecific bronchial hyperreactivity and the severity of response to the provoking agent, plicatic acid in workers with red cedar asthma. Measurement of hyperreactivity also assists in providing objective evidence of sensitization (ChanYeung and Lam, 1986). The demonstration of an increase in bronchial reactivity on returning to the workplace and a decrease away from work, with appropriate changes in lung function, is evidence for a causal relationship between symptoms and the work environment. To pinpoint the etiological agent in the workplace responsible for asthma, specific challenge may be necessary. Chan-Yeung and Lam (1986) emphasize that such testing can be dangerous, and should be performed only by experienced persons in a hospital setting for the following conditions: studying previously unrecognized occupational asthma, determining the precise etiological agent in a complex industrial environment, and confirming a diagnosis for medicolegal purposes. Detailed guidelines and testing procedures have been developed and published (Pepys and Hutchcroft, 1975; Salvaggio and Hendrick, 1989). 4.2.5

Special Laboratory Investigations

In a vast majority of patients, a diagnosis of environmental or occupationally related lung disease will be made through careful history and physical exam performed in conjunction with pulmonary function tests and a chest X-ray. The clinician rarely proceeds to specialized laboratory testing. In selected cases, however, additional supportive information can be obtained from skin testing with appropriate extracts, particularly of high-molecular-weight compounds for specific agents responsible for occupational asthma. Skin tests are generally accepted for IgE-mediated protein allergens. For example, extracts from flour (Block et al., 1983) and animal products produce immediate positive reaction on skin testing in sensitized subjects. Unfortunately, these skin tests are not always available. There is no established value for skin tests against chemicals such as formaldehyde, or cigarette smoke. This contrasts with more routine allergy skin tests with common inhalants and food allergens that are used to define the atopic status of the individual. Antigen-specific IgE antibody testing can be done with RAST or by ELISA for occupational asthma. Specific IgE antibodies against low-molecular-weight compounds conjugated to a protein (e.g., isocyanate) have been demonstrated in some exposed

TOOLS FOR STUDYING INDIVIDUALS

87

individuals (Butcher et al., 1980). Again, such tests are not readily available, and a positive response may occur in exposed workers without asthma. It is not appropriate to obtain antibodies to evaluate symptoms resulting from a single exposure. Serum precipitin testing can be done for hypersensitivity pneumonitis. The precipitins can be detected in farmer’s lung disease with antibodies to the thermophilic actinomyces, especially Micropolyspora faeni, or in pigeon breeders with antibodies to various proteins contained in pigeon serum or pigeon-droppings extracts. Office workers may develop hypersensitivity from fungal contaminants in ventilation systems; exposures have been reported to thermophilic actinomyces and penicillium species. Several studies have examined the role of bronchoalveolar lavage (BAL) in occupational lung diseases (Rom et al., 1987). Workers with farmer’s lung demonstrate a predominantly suppressor T-cell lymphocytic alveolitis (Cormier et al., 1986), and workers with chronic beryllium disease show a T-helper-cell lymphocytosis (Epstein et al., 1982), whereas workers with asbestosis demonstrate an increase in alveolar macrophages with or without a modest increase in neutrophils (depending on the smoking status) (Rom et al., 1987). Not only can the cellular differentials provide a clue to the underlying disorder, but also the asbestos fibers in the lavage fluid can be quantified. Although the utility of these research techniques in patient evaluation and diagnosis remains largely investigational, BAL may have diagnostic and prognostic implications in chronic beryllium disease. The extent of BAL cellularity, lymphocytosis, and beryllium lymphocyte proliferation test response correlates with disease severity (Newman et al., 1994), suggesting that the magnitude of the inflammatory and antigenic response in the lung may help predict disease progression or response to therapy. The application of molecular biology tools to examine susceptibility to toxicants is happening at an accelerated rate. Molecular approaches using arrays to analyze changes in gene expression are being applied to cells (e.g., macrophages and blood monocytes) removed from humans in exposed populations, and following controlled exposure to pollutants. Clinical studies are also incorporating analyses to identify genetic polymorphisms that make a person particularly susceptible to air pollutants. In humans, one of the candidate genes most implicated in air pollution responses is GSTM1, an important enzyme in the glutathione pathway for protection against oxidant injury (Peden, 2005). GSTM1 has a null allele with no protein expression, which leads to reduction in antioxidant protection and is present in 40% of the population of the United States. Children with the GSTM1 null allele have reduced lung function growth and increased susceptibility to ambient ozone (Romieu et al., 2004), while in adults the GSTM1 polymorphism may play a role in enhancing the nasal IgE response to diesel exhaust particle exposure (Gilliland et al., 2004). With the rapid advances in genomics and proteomics, the possibility of unraveling susceptibility at the population level becomes increasingly realistic. 4.2.6

Exposure Assessments

For the clinician, insufficient information on exposure data often limits establishing a causal relationship between exposure and disease. Especially in large industries, industrial hygiene information may be available that provides quantitative estimates of dusts, vapors, or aerosols in the work environment. An understanding of the pathophysiology of environmentally induced lung disorders requires a working knowledge of the mechanisms involved in the uptake of gases, the deposition of dusts, and their subsequent retention. In addition, in relating clinical disease to specific gaseous exposures, the clinician must recognize that

88

CLINICAL PERSPECTIVE ON RESPIRATORY TOXICOLOGY

penetration into and retention within the respiratory tract of toxic gases can vary widely depend on a number of factors: the physical properties of the gas (e.g., solubility), the concentration of the gas in the expired air, the rate of depth of ventilation, and the extent to which the material is reactive (Utell and Samet, 1990). Likewise, the aerodynamic properties of particles, airway anatomy, and breathing pattern largely determine the sites of deposition of particles and fibers in the airways. Thus, the identification and characterization of the inhaled materials are essential in linking inhaled materials with specific types of lung damage. Patterns of time use and activity place individuals in diverse indoor and outdoor microenvironments throughout the day, each microenvironment having its own unique set of air contaminants. Perhaps because of the distinct sources contaminating outdoor air and indoor air and the separate regulatory mechanisms for outdoor air, the workplace, and the home, the health effects of inhaled toxicants have often been addressed separately for outdoor and indoor air. However, the concept of total personal exposure is most relevant for health; personal exposure to air pollution represents the time-weighted average of pollutant concentrations in microenvironments that are environments having relatively homogeneous air quality. Thus, for an office worker, relevant microenvironments might include home, office, car, outdoors at work, and a movie theater. For some chemically reactive pollutants, such as ozone or acid aerosols, outdoor environments make the predominant contribution to total personal exposure; for others, such as radon and formaldehyde, indoor locations are most important. In considering the health consequences of environmental toxicants, the physician should recognize the potential contributions of various pollution sources. In the workplace, the safety officer or industrial hygienist can be very helpful in reviewing exposures and assisting in interpreting the results. In the investigation of building-related illnesses, indoor contaminants such as formaldehyde, environmental tobacco smoke, and microorganisms as well as the adequacy of the ventilation system need to be assessed. In special situations, a visit to the work or environmental site may be crucial in assisting the physician to better understand the potential exposure or even the adequacy of the ventilation system. Occasionally intensive investigation of the individual will lead to a recommendation for a population study as a result of introduction of new materials or issues of “safe” levels in the workplace.

4.3 TOOLS FOR STUDYING POPULATIONS 4.3.1

Overview

Persons may be exposed to inhaled pollutants in a variety of locations, including outdoors, workplaces, and other indoor microenvironments. Typically, epidemiological investigations of health effects are initiated because of concern about the consequences of exposure in a particular location, for example, outdoors or the workplace. For example, many studies of asbestos-exposed workers were undertaken after adverse consequences of exposure to this agent first became apparent. Studies of outdoor air pollution have been implemented in areas with unusual localized patterns of exposure, for example, effluents from petrochemical or other manufacturing plants, or with regional patterns of high outdoor pollution levels, for example, photochemical or acid aerosol pollution. Similarly, groups have been targeted because of particular exposures indoors.

TOOLS FOR STUDYING POPULATIONS

89

Epidemiological studies may also be implemented because of concern about the contribution of inhaled pollutants to particular diseases, both malignant and nonmalignant. For example, lung cancer increased in epidemic fashion among males in many countries beginning in the 1940s. Many of the initial investigations focused on exposure to outdoor air pollution in addition to cigarette smoking, which was subsequently found to be the cause of most cases. In the 1980s and 1990s, the contribution of indoor air pollutants, particularly secondhand tobacco smoke and radon, to lung cancer was an area of active investigation. Now, secondhand smoke is established as a cause of lung cancer (USDHHS, 2006) and the risks of radon have been well-characterized (Darby et al., 2005; Krewski et al., 2006). In most places in the developed world, although some individuals may be sufficiently impacted by inhaled pollutants to have clinically evident effects, the anticipated effects of most current exposures of concern on the population as a whole are likely to be subtler and not detectable by routine clinical assessment. However, epidemiological assessment of populations may provide evidence for adverse effects and the finding of such effects may become a basis for regulations or legal actions. Comparisons of exposed with nonexposed populations or of more highly exposed with less-exposed populations are made to find evidence of adverse effects that may not be manifest on evaluation of single individuals. Epidemiology comprises the methods used to study the effect of exposure on the population. Of the complementary approaches used to study inhaled pollutants, laboratory studies, clinical studies, and epidemiological studies, only the results of epidemiological studies provide evidence of the effects of agents as exposures occur in the community. Epidemiologic data may span the full range of susceptibility and inherently incorporate the interaction among agents. Surveillance refers to ongoing, systematic collection of data on health and disease (Baker et al., 1989). Programs for health surveillance may be implemented to monitor the consequences of intervention programs or to identify “sentinel” cases signaling unacceptable exposures (Rutstein et al., 1983). Exposures to hazards can also be monitored (Froines et al., 1989). Surveillance approaches have been most widely used for monitoring the occurrence of occupational diseases; surveillance may be implemented within populations or within exposed work forces. Data sources on populations include reporting by clinicians, death certificates, cancer registries, workers’ compensation systems, hospital discharge records, employer reports, and special surveys (Baker et al., 1989; Melius et al., 1989; Freund et al., 1989). Large industries have implemented their own surveillance systems. A report on silicosis illustrates the application of a surveillance system in a defined population. Valiante and Rosenman (1989) implemented a surveillance system for silicosis in the state of New Jersey; death certificates, hospital discharge data, and physicians were the principal sources for case ascertainment. For the period 1979 through 1987, 401 cases were identified, primarily by screening of hospital discharge data. Only one case was reported by a physician, and most hospitals did not voluntarily report their cases in spite of state requirements. Follow-up inspections at industries where cases had occurred identified inadequate control of exposure to crystalline silica at most. This system had the purpose of finding all cases of this fully preventable disease. There is also now a call for “accountability research,” referring to studies that track the consequences of regulations and other measures intended to protect public health (Health Effects Institute, 2003). Such research might be considered as a form of surveillance with focus on tracking such indicators as reductions in emissions or gains in health indicators.

90

CLINICAL PERSPECTIVE ON RESPIRATORY TOXICOLOGY

TABLE 4.3 Epidemiological Study Designs Used to Investigate the Effects of Inhaled Pollutants Case–control study: An analytical design involving selection of diseased cases and nondiseased controls followed by assessment of past exposures Cohort study: An analytical design involving selection of exposed and nonexposed subjects with subsequent observation for disease occurrence. Short-term cohort studies of the health status of susceptible groups are often called “panel studies” Cross-sectional study: Subjects are identified, and exposure and disease status determined, at one point in time

4.3.2

Epidemiological Approaches

Conventional epidemiological approaches used to study the adverse effects of inhaled pollutants on the lung are the cross-sectional study, the cohort study, and the case–control study (Table 4.3). More recently, the case-crossover design has been increasingly used for assessing the short-term risk of an exposure in relation to the occurrence of discrete events. These designs have the exposed individual as the unit of observation. Ecological designs, also used to study the environment and health, have groups of individuals, for example, communities or even countries, as one unit of observation. Multi-level designs incorporate elements of both the ecologic-studies and the individual-level designs. Each design has advantages and disadvantages for examining the effects of environmental exposures. Ecological study designs have long been used to investigate the health effects of air pollution. Cross-sectional studies have compared the health characteristics or rates of disease occurrence in communities having differing levels of exposure to air pollution (Lave and Seskin, 1977). Time-series designs have also received widespread application (Bell et al., 2004). In these ecological studies, temporal associations between air pollution levels and disease measures are evaluated. For example, a series of studies have assessed daily concentrations of air pollution levels with mortality counts and morbidity measures (U.S. EPA, 2004; U.S. EPA, 2006). Initially, these studies involved data from single cities, but advances in computing and statistical analysis methods now make possible the analysis of data from multiple cities. In the National Morbidity, Mortality and Air Pollution Study (NMMAPS), the investigators used data from up to 100 cities to assess the relationship between airborne particles and mortality (Dominici et al., 2005; Samet et al., 2000a, 2000b). A principal limitation of the ecologic design is the assumption that associations observed at the group level reflect effects at the individual level, the so-called “ecologic fallacy.” Nonetheless, the time-series studies have raised concern that current levels of air pollution in the U.S. and other developed countries may be adversely affecting public health. The cross-sectional study is a generally economical and feasible approach, often used to investigate indoor and outdoor air pollution and occupational lung diseases. When this design is used to study indoor or outdoor air pollution, populations of children or adults are typically surveyed, and health status is assessed. Exposure to outdoor air pollution may be inferred from geographic location or the results of area and limited personal monitoring; in studies of indoor air pollution, exposure may be categorized by the presence of sources or by monitoring the indoors. For example, in the Harvard Six Cities Study of air pollution, outdoor exposures were inferred from the city of residence and centrally sited pollutant monitors, and indoor exposures have been classified by sources and monitoring for some specific pollutants (Ware et al., 1984; Dockery et al., 1989; Neas et al., 1991). Personal

TOOLS FOR STUDYING POPULATIONS

91

monitoring was used to assess the contributions of indoor and outdoor pollutants to total personal exposures of the subjects (Spengler et al., 1985). The cross-sectional design is also widely used to evaluate the effects of occupational agents on the lung. In a typical cross-sectional study or survey, employed workers receive an assessment that includes a standardized respiratory symptom questionnaire, spirometry, chest radiograph, and often a limited physical examination. Exposure classification may be based on job title or duties, length of employment, reported duration of exposure to materials of interest, or a cumulative exposure measure calculated from measured or presumed concentrations of an agent and length of time at each level of exposure. For example, Samet et al. (1984) conducted a respiratory survey of long-term underground uranium miners. The study population was recruited by sampling from employed miners at two large companies. Increasing duration of underground employment, used as an index of exposure to potentially hazardous agents underground, for example, silica, radon, and blasting fumes, was associated with lower level of midmaximal expiratory flow. The prevalence of an abnormal chest radiograph compatible with silicosis increased with longer duration worked underground. The cross-sectional study, although one of the most economical and feasible designs, is subject to potentially significant methodological limitations. Estimates of the effects of exposure may be biased by the tendency of more susceptible or more affected persons to reduce their level of exposure, for example, by leaving an industry or polluted area. For example, in the survey of uranium miners reported by Samet et al. (1984), long-term miners who had already developed disease may have retired or even died, leaving a population less susceptible to develop silicosis or lung cancer. This type of selection bias is likely to be most prominent for agents with immediate effects on susceptible person; thus, asthmatic persons are likely to leave jobs involving exposures that worsen their disease. The temporal relationship between exposure and disease may be obscured or misrepresented in crosssectional data because exposure and disease are assessed at only one point in time. Cohort and case–control studies, which establish the proper sequence between exposure and disease, are also used to investigate the effects of inhaled pollutants on the lung. Cohort studies represent the optimal approach for assessing the effects of rare and special exposures, such as inhalation of toxic gases or exposure to asbestos. Cohort studies are termed “prospective” if the disease events will occur in the future, after the study population is established, and “retrospective” if disease events have already occurred when the cohort is assembled. The cohort design has the advantages of permitting direct estimation of disease incidence or mortality rates for exposed and nonexposed persons and of prospectively accumulating comprehensive data on exposure. The retrospective cohort design, often applied to occupational groups, can possibly be used to quickly assess the effects of a pollutant, since exposure and disease have already taken place when the investigation is initiated. The principal disadvantages of the cohort design include potentially high costs and losses to follow-up. Use of the cohort design is well illustrated by the many studies of mortality from cancer, asbestosis, and other causes of death in asbestos-exposed workers. For example, Selikoff and colleagues have described the mortality experience of 17,800 members of the insulation workers’ union in the U.S. (Selikoff et al., 1979). The cohort members were active in the union on January 1, 1967; follow-up of mortality was described through 1976 in the 1979 publication. Mortality in the cohort was compared to U.S. death rates for white males. Overall, 2271 deaths were observed with 1659 expected; cancer of the lung occurred with a five-fold excess, and 49 deaths from mesothelioma were identified. In this study, comparison of the mortality of the exposed population (asbestos workers) with the unexposed population

92

CLINICAL PERSPECTIVE ON RESPIRATORY TOXICOLOGY

(U.S. white males) provided strong evidence for associations of asbestos exposure with excess mortality from several causes of death. Other notable cohort studies have assessed air pollution and mortality: the Harvard Six Cities Study (Dockery et al., 1993; Laden et al., 2006) and the American Cancer Society’s Cancer Prevention II Study (Pope et al., 1995; Pope et al., 2002). The case–control study, like the cohort study, provides a measure of association between exposure and disease. This design has been widely used for studying lung cancer and occupational and environmental agents but infrequently for nonmalignant respiratory diseases. In comparison with the cohort study, the case–control study has the advantages of generally lower cost, greater feasibility, and usually a shorter time frame. The case– control study is the optimum approach for studying uncommon diseases. The results of this design may be limited by bias in assessing exposure or by bias in the method used to select cases and controls. The application of the case–control approach can also be illustrated by studies of the effects of asbestos. Analyses of geographic patterns of lung cancer mortality for the U.S. showed the highest rates along the coastal regions, particularly in the Southeast (Blot and Fraumeni, 1976). A series of case–control studies were conducted to assess occupational and other factorspotentiallycontributingto theexcesslungcancer(Blottet al., 1978;Vineiset al.,1988). Cases wereidentified throughhospitalsand deathcertificates, and controls were sampled from persons admitted to the same hospitals with diseases other than lung cancer, from death certificate files, or from the general population. Significant associations were found between employment in coastal shipyards, where asbestos was used for insulating ships, and lung cancer risk. For example, in a study conducted in coastal Georgia, employment in area shipyards was associated with a 60% increase in lung cancer risk after adjustment for smoking (Blott et al., 1978). These case–control studies provided a relatively rapid and informative evaluation of the excess mortality from lung cancer documented by mortality rates. The results of each type of study may be affected by biases, which alter the relationship between exposure to an inhaled agent and the health effect of concern; bias may increase or decrease the strength of an association. The three principal types of bias are selection bias, misclassification bias, and confounding bias. Selection bias refers to distortion of the exposure–outcome relationship by differential patterns of subject participation depending on exposure and disease status. For example, subjects with airway hyperresponsiveness might be more likely to withdraw from occupational cohorts exposed to respiratory irritants. Error in measuring either pollutant exposure or the health outcome results in misclassification. If the error equally affects cases and controls in a case–control study or exposed and nonexposed subjects in a cohort study, the bias reduces associations toward the null value, that is, no effect of exposure. Such nondifferential or random misclassification is of concern in most studies of inhaled pollutants and the lung; pollutant exposures are generally estimated using limited measurements or surrogates, such as presence of sources or duration of employment. Statistical power, the ability of a study to detect exposure-disease associations, declines as the degree of random misclassification increases (Gladen and Rogan, 1979; Shy et al., 1978). For example, Lubin et al. (1990) estimated sample sizes needed for case– control studies of indoor radon and lung cancer. As the degree of measurement error increased from the implausible level of 0% to more plausible levels above 50%, statistical power declined well below the desired level of 80–90%. If misclassification is differential (varying with case or control status in a case–control study or with exposure status in a cohort study), then the bias may increase or decrease the

TOOLS FOR STUDYING POPULATIONS

93

strength of association. Differential misclassification is of particular concern in case–control studies using interview to assess exposure. Information obtained from persons with and without a disease may not have comparable validity. For example, in comparison with controls, persons with lung cancer might minimize the extent of prior smoking or better recall occupational exposures such as asbestos. Bias from confounding results when the effect of the exposure of interest is altered by another risk factor. For example, confounding by cigarette smoking would occur in a cohort study of asbestos workers if smoking differed between exposed and nonexposed subjects. In fact, in studies of inhaled pollutants, particularly those with weak effects, confounding can be controlled through matching exposed and nonexposed subjects on potential confounding factors or through collection of data on potential confounding factors and use of appropriate data analysis methods. 4.3.3

Exposure Assessment

Awide range of approaches are used in epidemiological studies to assess exposure to inhaled agents (Table 4.4). Some approaches can be used feasibly and at low cost in large populations, whereas others require intensive and costly personal or biological monitoring. The simplest approaches, such as using job title in an occupational study or residence location in a study of outdoor air pollution, are likely to be subject to the greatest degree of misclassification. Misclassification is least if personal exposures are directly monitored. However, personal monitoring is not possible for all pollutants, and small passive monitors are only available for a few pollutants of concern (Wallace and Ott, 1982; McCarthy et al., 1991). Moreover, current exposures may not be relevant to many chronic diseases with risks reflecting cumulative rather than immediate exposures. Biomonitoring or biologic monitoring of tissues such as blood, urine, and breast milk is an extremely valuable tool for identifying population exposures to a variety of chemicals. Biological monitoring with exposure markers that can be measured inexpensively and noninvasively is feasible in large populations for only a few pollutants, for example, carbon monoxide, lead, and nicotine. Recently, the CDC has completed three reports, findings summarized in the “National Report on Human Exposure to Environmental Chemicals,” giving concentrations of chemicals and metabolites in blood and urine of a representative sample of the U.S. population. The third report (CDC, 2005) includes exposure information on 148 chemicals in a cohort of 2400 individuals. Although the absence of health based reference values and exposure assessments limit the interpretation of the data, the report provides the most comprehensive summary on biomonitoring data currently available.

TABLE 4.4

Assessment of Inhaled Pollutant Exposure in Epidemiological Studies

Occupational Agents

Outdoor Air Pollutants

Indoor Air Pollutants

Job title and industry Length of employment Self-reported exposure Job-exposure matrices Area monitoring Personal monitoring Biological monitoring

Residence location Proximity to sources Self-reported exposure Central site monitoring Small-area monitoring Personal monitoring Biological monitoring

Source inventory Indoor monitoring Personal monitoring Biological monitoring

94

CLINICAL PERSPECTIVE ON RESPIRATORY TOXICOLOGY

Tiered approaches for exposure assessment have been proposed (Leaderer et al., 1986). Nesting more intensive monitoring approaches for small numbers of subjects selected from a larger study population permits estimation of the misclassification resulting from use of less intense but feasible approaches. For air pollution, modeling strategies may be used that incorporate information on air pollution sources and traffic exposure. Models may also incorporate measured pollution data. Jerrett et al. (2005) provide a review of these methods. 4.3.4

Outcome Assessment

The range of adverse health effects caused by inhaled pollutants is wide, extending from excess mortality to subtle effects on function or symptom occurrence (Table 4.5) (American Thoracic Society, 2000). Studies of morbidity most often examine symptom and disease occurrence and level of lung function. Standardized methods have been developed for collecting data on symptoms and for assessing lung function and airway responsiveness (Ferris, 1978; Samet, 1989; Sparrow and Weiss, 1989). The American Thoracic Society has published standardized respiratory symptoms questionnaires, which are currently under revision, for children and for adults (Ferris, 1978). The questionnaires emphasize chronic respiratory symptoms and conditions, and are not appropriate instruments for investigations of acute responses to inhaled pollutants. Questionnaires specific to the investigation of asthma are available (Burney and Chinn, 1987; Asher et al., 1995; Weiland et al., 2004). Recommendations for spirometric testing cover the characteristics of the equipment, procedures for testing, and data acceptability and interpretation (American Thoracic Society, 1994). In studying the pneumoconioses, occupational lung diseases caused by inhalation of inorganic dusts of low solubility, chest radiographs are usually classified according to the system of the International Labour Office, most recently revised in 1980 (International

TABLE 4.5

Health Outcomes in Epidemiological Studies of Inhaled Pollutants

Occupational agents Mortality, specific or all causes Occurrence of specific diseases Respiratory symptoms Reduced lung function Abnormality on chest radiograph Increased airway responsiveness Immunologic sensitization Outdoor and indoor air pollutants Mortality, specific or all causes Hospitalization Emergency room or other outpatient visit Absenteeism Disease occurrence Respiratory symptoms Reduced lung function Increased airway responsiveness Immunologic sensitization

CARDIOVASCULAR RESPONSES

95

Labour Office, 1980). This system classifies abnormalities of the lung parenchyma on the basis of size and extent and changes in the pleura. The system was designed to provide a uniform method for coding changes on chest radiographs and thereby to facilitate comparisons across regions and over time. In the U.S., training is offered in the use of this system by the American College of Radiology; those successfully passing an examination in its use are designated as “B readers” by the NIOSH. Even the use of a standardized classification and trained film readers may not eliminate strong observer bias in X-ray interpretation (Ducatman et al., 1988; Parker et al., 1989).

4.4 CARDIOVASCULAR RESPONSES Although the respiratory system generally has been considered the major target of inhaled materials, recent evidence has demonstrated that for several inhaled agents the lung serves primarily as a conduit. This has especially been the case for inhaled particulate matter (PM), where epidemiological studies provide convincing evidence that exposure to increasing concentrations of ambient PM trigger adverse cardiac events (Samet et al., 2000a, 2000b). Literally hundreds of epidemiological studies linking PM with adverse cardiovascular outcomes have been published over the past decade. Clinical studies on the cardiovascular effects of air pollution and PM were the basis for a recent workshop (Utell et al., 2002). The American Heart Association has acknowledged the role of air pollution in cardiovascular disease in a recently published document reviewing the proposed mechanistic pathways and providing advice for clinicians (Brook et al., 2004). Exposure to fine PM (PM2.5) in ambient air is associated with increased mortality and morbidity related to cardiovascular disease. An increase in PM10 concentration of 50 mg/m3 was associated with a 3–8% increase in relative risk of death (U.S. Environmental Protection Agency, 2006). The strongest associations were seen for respiratory and cardiac deaths, particularly among the elderly. Because deaths from cardiovascular causes outnumber those from respiratory causes, the majority of deaths are attributable to cardiovascular causes. Increased PM levels have been linked with death from myocardial infarction, arrhythmia (sudden death) and congestive heart failure. Novel, noninvasive diagnostic techniques, such as electrophysiological monitoring, echocardiography, scintigraphic scanning and impedance cardiography, and implantable defibrillators with recorders, available in clinical cardiology practice have been incorporated into epidemiological, panel and human clinical air pollution studies. For example, PM2.5 exposure was associated with increased frequency of cardiac arrhythmias in patients with implantable defibrillators (Peters et al., 2000) and with reductions in heart rate variability in panel studies of elderly residents (Gold et al., 2000). These studies suggest that exposure to ambient PM2.5 influence autonomic regulation of the heart, and are associated with increased arrhythmias and myocardial infarction in susceptible patients with heart disease. In addition, changes in blood viscosity, ST-segment depression, increased blood pressure, and increased circulating markers of inflammation and thrombosis have been linked with increases in PM2.5 air pollution (Godleski, 2006). Exposure to PM2.5 air pollution may also have acute effects on vascular function in humans. Investigators exposed subjects to concentrated ambient PM2.5 plus ozone, and demonstrated constriction of the brachial artery of the forearm immediately after pollutant exposure (Brook et al., 2002). Inhalation of ultrafine carbon particles alters blood leukocyte expression of adhesion molecules and the lung diffusing capacity in humans

96

CLINICAL PERSPECTIVE ON RESPIRATORY TOXICOLOGY

(Frampton et al., 2006; Pietropaoli et al., 2004). There is recent additional evidence, from both animal (Chen and Nadziejko, 2005; Sun et al., 2005) and human studies (K€ unzli et al., 2005), indicating potential mechanisms by which PM may worsen or even induce atherosclerosis. These findings indicate that there are acute vascular effects of exposure to air pollution, providing a mechanistic link between PM exposure and cardiovascular effects in humans. Although the cardiovascular consequences of PM2.5 have been the major focus of recent research, the effects of exposure to carbon monoxide as a result of its affinity for hemoglobin has been recognized for many decades. More recently, several reports have suggested hat ozone and perhaps sulfur dioxide exposures could also be associated with adverse cardiovascular outcomes. Future research may identify other extra-pulmonary effects of inhaled toxicants as novel nontraditional approaches are applied to inhaled materials.

4.5 LIMITATIONS OF CLINICAL AND EPIDEMIOLOGICAL ASSESSMENTS OF THE EFFECTS OF INHALED AGENTS As the 21st century begins, many respiratory hazards have been recognized and controlled in the U.S. and other developed countries. Concern remains, however, about the safety provided by standards for workplace and environmental exposures and the risks of new and unevaluated agents. Large segments of the population are exposed to indoor and outdoor pollutants with adverse effects, and workers expect that their jobs will not carry unacceptable risks. In response to individual and societal fears, clinicians and epidemiologists are asked to assess the effects of inhaled pollutants: the clinician to evaluate the health of exposed individuals and the epidemiologist to address the effects on exposed groups. Both types of assessments have limitations, particularly for answering concerns about safety. Few clinicians have the proper training to evaluate patients with toxic occupational and environmental exposures. More than 15 years ago, an assessment identified a supply of 1200–1500 appropriately trained physicians while there was a need for 4600–6700 (Castorina and Rosenstock, 1990). This gap remains. Most primary care providers (internists and family practitioners) lack training in toxicology and in occupational and environmental medicine. They do not have skills for characterizing exposures and making links between exposures and health outcomes; they are also unlikely to have an understanding of such key concepts as individual susceptibility, interactions among agents, and exposure–response relationships. Their skills in counseling patients concerning risks may also be limited. Furthermore, clinical methods for assessing respiratory effects may be insensitive if applied in the conventional, clinical fashion. For example, pulmonary function testing may be unstandardized, and the chest radiograph may only be interpreted for gross clinical abnormalities, and not for subtle changes reflective of early disease. Epidemiology has been an extremely effective tool for investigating inhaled pollutants with either very strong (e.g., cigarette smoking and lung cancer) or very specific (e.g., asbestos and mesothelioma) effects. In investigating agents that may have effects that are not strong but of public health concern, the epidemiological studies may be limited by misclassification of exposure and disease, confounding, and other methodological problems. Large sample sizes may be needed to test for the anticipated effects, particularly if exposure estimates are subject to substantial misclassification (Lubin et al., 1990). The results of an epidemiological study cannot provide sufficient precision to exclude the possibility of some increased risk of exposure to an agent. Thus, the findings of epidemiological studies of

ADVICE AND COUNSELING OF PATIENTS

97

inhaled pollutants often leave uncertainty concerning the risks posed and do not fully address the concerns and questions of exposed persons and those involved in policy and regulation.

4.6 ADVICE AND COUNSELING OF PATIENTS 4.6.1

Patient-Oriented

Although often called on to advise on control strategies for minimizing exposure to occupational and environmental exposure, the primary care physician may have little training or experience in this arena. Approaches for limiting the health risks of breathing polluted ambient air have received little investigation and dissemination to primary care practitioners. Present understanding of the determinants of exposure suggests that modifying time-activity patterns to limit time outside during episodes of pollution represents the most effective strategy for dealing with acute effects. The levels of some reactive pollutants tend to be lower indoors than outdoors. Ozone levels in buildings are lower than outdoor levels but can be drivenupward by increasing the rate of exchange of indoor with outdoor air. Fine acid aerosols can penetrate indoors, but neutralization by ammonia produced by occupants, household products, and pets may reduce concentrations. Other types of particles in outdoor air may also enter indoor air. Nevertheless, health care providers can reasonably advise patients to stay indoors during pollution episodes. Vigorous exercise outdoors, which increases the dose of pollution delivered to the respiratory tract, should also be avoided at such times. Susceptible patients should be counseled concerning the nature and degree of their susceptibility. The use of medications should follow the usual clinical indications, and therapeutic regimens should not be adjusted because of the occurrence of a pollution episode without evidence of an adverse effect on symptoms or function. In the laboratory, inhalation of cromolyn sodium and bronchodilating agents may block the response to some pollutants, but use of these drugs solely because of exposure to air pollution cannot be advised. Respiratory protective equipment has been developed for use in the workplace in order to minimize exposure to toxic gases and particles. Many of these devices, particularly those likely to be most effective, add to the work of breathing and cannot be tolerated by persons with respiratory disease. Respirators can provide effective personal protection only when they are properly selected and when they are used in the context of a comprehensive respiratory protection program. OSHA has specified the minimum requirements for an acceptable program. In order for respirators to provide adequate worker protection, both the proper selection and the correct use of respirators are essential. Respirators are the least preferred method of protection from respiratory hazards, and they should be used only when engineering controls are not technically feasible, while controls are being installed or repaired, or in emergency or other temporary situations. Under most circumstances, health care providers should not suggest respiratory protection as a method of reducing the risks of ambient air pollution. Similarly, household air cleaners have not been shown to have health benefits, whether directed at pollutants generated by indoor sources or those brought in with outside air. A variety of federal agencies are active in the area of environmental health and have services of some degree of value to practicing physicians. Unfortunately, the responsibilities among the agencies are often fragmented, and no one source is targeted for providing assistance to the practicing physician. If the clinician wishes further information or consultation, the following agencies may be of value:

98

CLINICAL PERSPECTIVE ON RESPIRATORY TOXICOLOGY

1. Agency for Toxic Substances and Disease Registry (ATSDR): Funded via Superfund legislation, this agency is part of the U.S. Public Health Service and affiliated with the Centers for Disease Control and Prevention (CDC). Among its mandates are the educations of physicians concerning toxic hazards associated with hazardous wastes. 2. Centers for Disease Control and Prevention: This agency is available to investigate illnesses that may be linked to environmental hazards. 3. Environmental Protection Agency (EPA): This federal agency has the responsibility of protecting humans and the environment from the unwanted effects of chemicals and physical agents, and sets for pollutant concentration limits for ambient air and drinking water. It has focused primarily on providing the public with information on environmental risks and has not been a major resource for the practicing physician. 4. National Institute for Occupational Safety and Health: This agency, a component of CDC, has responsibilities for research and professional training issues related to worker health. It can provide information about specific workplace hazards and has the authority to enter workplaces to evaluate health and safety problems. 5. National Institute for Environmental Health Sciences (NIEHS): This federal agency is a component of the National Institutes of Health (NIH), and has responsibility for research related for understanding the health hazards related to environmental chemicals. It can provide information about the known toxicity, mechanisms of toxicity, and potential carcinogenicity of specific environmental chemicals 6. Other important resources may include medical school faculty with expertise in areas of environmental and occupational medicine, poison control centers, and occupational and environmental medicine clinics. As a result of the OSHA Hazard Communication Standard, industries that use chemicals in the workplace have a Material Safety Data Sheet (MSDS) for each hazardous chemical. The companies should have these available on request of the physician or the worker. Goldstein and Gochfeld (1990) present additional discussion of resource agencies for clinicians and mechanisms for making professional contacts.

4.6.2

Community-Oriented

Communities frequently become concerned about the impact of particular local sources, perhaps a power plant or manufacturing facility. Concern about the health risks may quickly lead to controversy and litigation. Thus, understanding the health risks posed by local sources may be difficult and require skills in community health as well as in epidemiology and toxicology. Local physicians may become involved through concerns about the health of their patients, or as advocates for the community’s environment or for the polluting facility. Most often the dimensions of such complex problems exceed the skills of local physicians. Nevertheless, involvement may be appropriate, but guidance should be obtained from appropriate public health and environmental agencies. In the United States, no regulatory agency has authority over indoor air quality. Workplaces with more than 25 people are regulated by OSHA, although standards do not directly address the new problem of building-related illnesses. Good practice guidelines that are often adopted by States and localities as standards for ventilation in nonresidential buildings are set by the American Society of Heating, Refrigerating and Air Conditioning Engineers (ASHRAE).

SUMMARY

99

The interplay of factors that must be manipulated to prevent environmentally induced disease is complex, but standards are considered to be objective indicators of success preventive measures or strategies. In 1976, the EPA proposed cautionary statements for public reporting of outdoor air quality, the Pollutant Standards Index (PSI) for criteria pollutants; in 1999, the PSI was replaced with the Air Quality Index (AQI) to incorporate new PM2.5 and ozone standards. The actions taken when “alert levels” are reached or expected to be reached include the issuance of health advisories (or cautionary statements) to the public. The EPA’s advice is intended to be applied by local air pollution agencies in preparing daily air quality summaries, which are disseminated to the media. Although the cautionary statements require some revisions, especially as related to ozone exposures, useful guidelines are offered for physicians and public health officials. Finally there are important impacts to the community and workplace resulting from changes in health care insurance and the delivery of health care. Nearly 75% of employees in the United States are enrolled in health plans through health maintenance organizations (HMOs) and their provider networks. Both physicians and employees are subject to managed care’s efforts to hold health care professionals more accountable for the medical and financial aspects of care. This often results in managing occupational health care demands and workers’ compensation according to established guidelines. Although the original goal of all of these programs was to provide high quality care, the focus of many stakeholders has been cost containment. With the introduction of workers’ compensation managed care, many businesses have developed and implemented transitional work programs, developed alliances with Occupational Medicine clinics, retained the services of case managers, requested more independent medical examinations, and contested more claims, all in an effort to contain costs. Although it is premature to assess the impact of managed care, it certainly has the potential to negatively impact health care of the injured worker. However, the Managed Care Pilot Project in Washington State (Sparks and Feldstein, 1997) studied the effect of experienced-based capitation on medical and disability costs, quality of care, worker satisfaction with medical care, and employer satisfaction. Much of the care, and all of the treatment coordination, was provided by physicians specialized in occupational medicine and oriented towards timely return to work. The study revealed that “medical costs were reduced by approximately 27%, functional outcomes remained the same, workers were less satisfied with their treatment and access to care initially, and employers were much more satisfied with the quality and speed of the information received from the providers.” As the effort to contain costs extend into occupational health, it will be important to carefully track injury outcomes and worker satisfaction.

4.7 SUMMARY Environmental medicine very likely requires a greater interdisciplinary approach than any other medical specialty. For the primary care physician tackling an occupational or environmental medicine problem, there is a necessity not only to work closely with nonmedical personnel such as industrial hygienists, ergonomists, toxicologists, epidemiologists, lawyers, regulators, and union representatives, but also to become knowledgeable about these various disciplines. The diseases of individuals may provide indications of unacceptable exposures in the environment or workplace; health care providers should be able to recognize such “sentinel” disease and respond appropriately by contacting employers

100

CLINICAL PERSPECTIVE ON RESPIRATORY TOXICOLOGY

and regulatory or public health agencies. Ultimately, it may be necessary to visit a workplace or environmental site, request industrial hygiene data, and consider screening the exposed population of individuals in order to determine whether an environmental pulmonary hazard exists. In the final analysis, an inquisitive mind and a bit of detective work are often prerequisites for establishing causation between an environmental exposure and a pulmonary disorder.

REFERENCES Aberle DR, Gamsu G, Ray CS, Feuerstein IM (1988) Asbestos-related parenchymal fibrosis: detection with high-resolution CT. Radiology 166:729–734. Agency for Toxic Substances and Disease Agency Registry (1993) Obtaining an exposure history. United States Department of Health and Human Services, Public Health Service, Atlanta, Georgia. Am. Fam. Physician 48:483–491. American College of Occupational and Environmental Medicine (1999) ACOEM Statement on multiple chemical hypersensitivity syndrome, multiple chemical sensitivities, environmental tobacco smoke, and indoor air quality, May 2, 1991. Multiple Chemical Sensitivities (Addendum, 1993, 1999). (Available from the College.) American College of Physicians (1989) American College of Physicians position statement: clinical ecology. Ann. Intern. Med. 111:168–178. American Medical Association, Council on Scientific Affairs (1992) Clinical ecology. JAMA 268:3465–3467. American Thoracic Society (1986) The diagnosis of nonmalignant diseases related to asbestos. Am. Rev. Respir. Dis. 134:363–368. American Thoracic Society(1995) Standardization of spirometry—1994 update. Am. J. Respir. Crit. Care Med. 152:1107–1136. American Thoracic Society (1997) Achieving healthy indoor air. Am. J. Respir. Crit. Care Med. (Suppl.) 156 (3):533–544. American Thoracic Society (2000) What constitutes an adverse health effect of air pollution? Official statement of the American Thoracic Society. Am. J. Respir. Crit. Care Med. 161:665–673. American Thoracic Society (2004) Diagnoses and initial management of nonmalignant diseases related to asbestos. Am. J. Respir. Crit. Care Med. 170:691–715. American Thoracic Society/European Respiratory Society (2005) General consideration for lung function testing. Eur. Respir. J. 26:153–161. Asher MI, Keil U, Anderson HR, Beasley R, Crane J, Martinez F, Mitchell EA, Pearch N, Sibbald B, Stewart AW, Strachan D, Weiland SK, Williams HC (1995) International study of asthma and allergies in childhood (ISAAC): rationale and methods. Eur. Respir. J. 8:483–491. Ashford NA, Miller CS (1998) Chemical exposures: low levels and high states,2nd edn. New York, NY:Van Nostrand Reinhold/Wiley. Baker EL, Honcha EA, Fine LJ (1989) I. Surveillance in occupational illness and injury: concepts and content. Am. J. Public Health 79 (Dec. Suppl.):9–11. Bell ML, Samet JM, Dominici F (2004) Time-series studies of particulate matter. Annu. Rev. Public Health 25:247–280. Block G, Tse KS, Kijek K, Chan H, Chan-Yeung M (1983) Baker’s asthma: clinical and immunological studies. Clin. Allergy 13:359–370. Blot WJ, Fraumeni JE Jr (1976) Geographical patterns of lung cancer: industrial correlations. Am. J. Epidemiol. 103:539–550.

REFERENCES

101

Blott WJ, Harrington JH, Tolego A, Hoover R, Health CW, Fraumeni JE Jr (1978) Lung cancer after employment in shipyards during World War II. N. Engl. J. Med. 299:620–624. Boushey HA, Holtzman MJ, Sheller JR, Nadel JA (1980) Bronchial hyperreactivity. Am. Rev. Respir. Dis. 121:389–413. Brightman HS (2000)Sick building syndrome studies and the compilation of normative and comparative values. In:Spengler JD, Samet JM, McCarthy JF, Indoor Air Quality Handbook. New York: McGraw-Hill, pp.3.1–3.32. Brook RD, Brook JR, Urch B, Vincent R, Rajagoplan S, Silverman F (2002) Inhalation of fine particulate air pollution and ozone causes acute arterial vasoconstriction in healthy adults. Circulation 105:1534–1536. Brook RD, Franklin B, Cascio W, Hong YL, Howard G, Lipsett M, Luepker R, Mittleman M, Samet J, Smith SC Jr, Tager I (2004) Air pollution and cardiovascular disease—a statement for healthcare professionals from the expert panel on population and prevention science of the American Heart Association. Circulation 109 (21):2655–2671. Burney P, Chinn S (1987) Developing a new questionnaire for measuring the prevalence and distribution of asthma. Chest 91(Suppl.):79S–83S. Butcher BT, O’Neil CE, Reed MA, Salvaggio JE (1980) Radioallergosorbent testing of toluene diisocyanate-reactive individuals using p-tolyl isocyanate antigen. J. Allergy Clin. Immunol. 66:213–216. Castorina JS, Rosenstock L (1990) Physician shortages in occupational and environmental medicine. Ann. Intern. Med. 113:983–986. CDC(2005) Third National Report on human exposure to environmental chemicals. U.S. Department of Health and Human Services, Center for Disease Control and Prevention, Atlanta, GA [online]. Available:https://www.cdc.gov/exposure report/3rd/ [accessed November 16, 2005]. Chan-Yeung M (1986) Occupational asthma. Chest 98:148S–161S. Chan-Yeung M (1995) American College of Physicians consensus statement: assessment of asthma in the workplace. Chest 108:1084–1117. Chan-Yeung M, Lam S (1986) Occupational asthma. Am. Rev. Respir. Dis. 133:686–703. Chen LC, Nadziejko C (2005) Effects of subchronic exposures to concentrated ambient particles (CAPs) in mice: VCAPs exacerbate aortic plaque development in hyperlipidemic mice. Inhal. Toxicol. 17:217–224. Clausen JL (editor)(1982) Pulmonary Function Testing Guidelines and Controversies: Equipment, Methods, and Normal Values.Academic Press:New York. Cormier YJ, Belanger LeBlanc P, Laviolette M (1986) Bronchoalveolar lavage in farmer’s lung disease: diagnostic and physiologic significance. Br. J. Ind. Med. 43:401–405. Craighead JE, Mossman BT (1982) The pathogenesis of asbestos-associated diseases. N. Engl. J. Med. 306:1446–1455. Cullen MR (1987) The worker with multiple chemical sensitivities: an overview. State Art Rev. Occup. Med. 2:655–661. Cullen MR, Cherniack MG (1989) Spectrum of occupational disease in an academic hospital-based referral center in Connecticut from 1979 to 1987. Arch. Intern. Med. 149:1621–1626. Darby S, Hill D, Auvinen A, Barros-Dios JM, Baysson H, Bochicchio F, Deo H, Falk R, Forastiere F, Hakama M, Heid I, Kreienbrock L, Kreuzer M, Lagarde F, Makelainen I, Muirhead C, Oberaigner W, Pershagen G, Ruano-Ravina A, Ruosteenoja E, Rosario AS, Tirmarche M, Tomasek L, Whitley E, Wichmann HE, Doll R (2005) Radon in homes and risk of lung cancer: collaborative analysis of individual data from 13 European case-control studies. BMJ 330 (7485):223. Dockery DW, Speizer FE, Stram DO, Ware JH, Spengler JD, Ferris BG Jr (1989) Effects of inhalable particles on respiratory health of children. Am. Rev. Respir. Dis. 139:587–594.

102

CLINICAL PERSPECTIVE ON RESPIRATORY TOXICOLOGY

Dockery DW, Pope CA 3rd, Xu X, Spengler JD, Ware JH, Fay ME, Ferris BG Jr, Speizer FE (1993) An association between air pollution and mortality in six U.S. cities. N. Engl. J. Med. 329:1753–1759. Dominici F, McDermott A, Daniel M, Zeger SL, Samet JM (2005) Revised analyses of the National Morbidity, Mortality, and Air Pollution Study: mortality among residents of 90 cities. J. Toxicol. Environ. Health 68:1071–1092. Ducatman AM, Yang WN, Forman SA (1988) “B-readers” and asbestos medical surveillance. J. Occup. Med. 30:644–647. Eisen EA, Wegman DH, Kriebel D (1993) Application of peak expiratory flow in epidemiologic studies of occupation. Occup. Med. 8(2):265–277. Epstein PE, Dauber JH, Rossman MD, Danick RP (1982) Bronchoalveolar lavage in a patient with chronic berylliosis: evidence for hypersensitivity pneumonitis. Ann. Intern. Med. 97:213–216. Ferris BG (1978) Epidemiology standardization project. Am. Rev. Respir. Dis. 118:1–120. Finnegan MJ, Pickering CAC (1986) Building related illness. Clin. Allergy 16:389–405. Frampton MW, Stewart J, Oberd€orster G, Morrow PE, Chalupa D, Frasier LM, Speers DM, Cox C, Huang L-S, Utell MJ ,(2006) Inhalation of carbon ultrafine particles alters blood leukocyte expression profile in humans. Environ. Health Perspect. 114:51–58. Fraser RG, Pare JA, Pare PD, Fraser RS, Genereux GP, <Ed (editors) (1990) Pleuropulmonary disease caused by inhalation of inorganic dust. In: Diagnosis of Diseases of the Chest, 3rd edn. Philadelphia:W. B. Saunders, pp. 3:2276–3:2282. Freund E, Seligman PJ, Chorba TL, Safford SK, Drachman JG, Hull HF (1989) Mandatory reporting of occupational diseases by clinicians. JAMA 262:3041–3044. Friedman AC, Fiel SB, Fisher MS, Radecki PD, Ler-Toaft AS, Caroline DF (1988) Asbestos-related pleural diseases and asbestosis: a comparison of CT and chest radiography. Am. J. Roentgenol. 150:269–275. Froines J, Wegman D, Eisen E (1989) VI. Hazard surveillance in occupational disease. Am. J. Public Health 79(Dec. Suppl.):26–31. Gilliland FD, Li YF, Saxon A, Diaz-Sanchez D , (2004) Effect of glutathione-S-transferase M1 and G1 genotypes on xenobiotic enhancement of allergic responses: randomized, placebo-controlled crossover study. Lancet 363:119–125. Gladen B, Rogan WJ (1979) Misclassification and the design of environmental studies. Am. J. Epidemiol. 109:607–616. Godleski JJ (2006) Responses of the heart to ambient particle inhalation Frampton MW, Utell MJ, editors. Clinics in Occupational and Environmental Medicine, Vol.5, No. 4.Philadelphia:Elsevier Saunders, pp. 849–864. Gold DR, Litonjua A, Schwartz J, Lovett E, Larson A, Nearing B , et al. (2000) Ambient pollution and heart rate variability. Circulation 101:1267–1273. Goldstein BD, Gochfeld M (1990) Role of the physician in environmental medicine. Med. Clin. North Am. 74:245–261. Health Effects Institute (2003) Assessing the health impact of air quality regulations: concepts and methods for accountability research. Report # Communication 11. Boston: Health Effects Institute. Herbert R, Moline J, Skloot G, Metzger K, Baron S, Luft B, Markowitz S, Usadin I, Harrison D, Stein D, Todd A, Enright P, Stellman JM, Landrigan PJ, Levin SM (2006) The World Trade Center disaster and the health of workers: five-year assessment of a unique medical screening program. Environ. Health Perspect. 114:1853–1858. Institute of Medicine (1988) Role of the Primary Care Physician in Occupational and Environmental Medicine.Washington, DC:National Academy Press. Institute of Medicine (1995) In:Pope AM, Rall DR, editors Environmental Medicine.Washington, D.C.:National Academy Press.

REFERENCES

103

International Labour Office (1980) Guidelines for the Use of International Labour Office Classification of Radiographs of Pneumoconioses. Occupational Safety and Health Sciences 22.Geneva: International Labour Office. Jerrett M, Arain A, Kanaroglou P, Beckerman B, Potoglou D, Sahsuvaroglu T, Morrison J, Giovis C (2005) A review and evaluation of intraurban air pollution exposure models. J. Expo. Anal. Environ. Epidemiol. 15:185–204. Kilbourne EM, Weiner J (1990) Occupational and environmental medicine: the internist’s role. Ann. Intern. Med. 113:974–982. Krewski D, Lubin JH, Zielinski JM, Alavanja M, Catalan VS, Field RW, Klotz JB, Letourneau EG, Lynch CF, Lyon JL, Sandler DP, Schoenberg JB, Steck DJ, Stolwijk JA, Weinberg C, Wilcox HB (2006) A combined analysis of North American case-control studies of residential radon and lung cancer. J. Toxicol. Environ. Health A. 69:533–597. K€unzli N, Jerret M, Mack WJ, Beckerman B, LaBree L, Gilliland F, Thomas D, Peters J, Hodis HN (2005) Ambient air pollution and atherosclerosis in Los Angeles. Environ. Health Perspect. 113:201–206. Laden F, Schwartz J, Speizer FE, Dockery DW (2006) Reduction in fine particulate air pollution and mortality: extended follow-up of the Harvard Six Cities study. Am. J. Respir. Crit. Care Med. 173:667–72. Lam S, Tan F, Chan H, Chan-Yeung M (1983) Relationship between types of asthmatic reaction, nonspecific bronchial reactivity, and specific IgE Antibodies in patients with red cedar asthma. J. Allergy Clin. Immunol. 72:134–139. Lave LB, Seskin EP (1977) Air Pollution and Human Health. Baltimore, MD:Johns Hopkins University Press. Leaderer BP, Zagraniski RT, Berwick M, Stolwijk JAJ (1986) Assessment of exposure to indoor air contaminants from combustion sources: methodology and application. Am. J. Epidemiol. 124:275–289. Lubin JH, Samet JM, Weinberg C (1990) Design issues in epidemiologic studies of indoor exposure to Rn and risk of lung cancer. Health Phys . 59:807–817. McCarthy JF, Bearg DW, Spengler JD (1991) Assessment of indoor quality. In:Samet JM, Spengler JD, editors. Indoor Air Pollution: A Health Perspective. Baltimore:Johns Hopkins University Press. McKay RT (1986) Bronchoprovocation challenge testing in occupational airways disorders. Semin. Respir. Med. 7:297–306. McKay RT, Lockey JE (1991) Pulmonary function testing: guidelines for medical surveillance and epidemiological studies. Occup. Med. 6:43–57. Melius JM, Sestito JP, Seligman PJ (1989) Occupational disease surveillance with existing data sources. Am. J. Public Health 79(Dec. Suppl.):46–52. National Institutes of Health: National Heart, Lung, and Blood Institute (1997) National Asthma Education and Prevention Program: Guidelines for the Diagnosis and Management of Asthma. NIH Publication #97-4051, Bethesda, MD, pp. 6–11. National Research Council, Committee on Human Biomonitoring for Environmental Toxicants (2006) Human Biomonitoring for Environmental Chemicals.Washington, DC:National Academy Press. Neas LM, Ware JG, Dockery DW, Spengler JD, Speizer FE, Ferris BG Jr (1991) Association of indoor nitrogen dioxide with respiratory symptoms and pulmonary function in children. Am. J. Epidemiol. 134:204–219. Newman LS, Bobka C, Schumacher B, Daniloff R, Zhen B, Mroz MM, King TE Jr (1994) Compartmentalized immune response reflects clinical severity of beryllium disease. Am. J. Respir. Crit. Care Med. 150:135–142. O’Byrne PM, Ryan G, Morris M, McCormack D, Jones NL, Morse JLC, Hargreave FE (1982) Asthma induced by cold air and its relationship to nonspecific bronchial responsiveness to methacholine. Am. Rev. Respir. Dis. 125:281–285.

104

CLINICAL PERSPECTIVE ON RESPIRATORY TOXICOLOGY

Occupational and Environmental Health Committee of the American Lung Association of San Diego and Imperial Counties (1983) Taking the occupational history. Ann. Intern. Med. 99:641–651. Parker DL, Bender AP, Hankinson S, Aeppli D (1989) Public health implication of the variability in the interpretation of “B” readings for pleural changes. J. Occup. Med. 31:775–780. Pepys J, Hutchcroft BJ (1975) Bronchial provocation tests in etiologic diagnosis and analysis of asthma. Am. Rev. Respir. Dis. 112:829–859. Peden DB (2005) The epidemiology and genetics of asthma risk associated with air pollution. J. Allergy Clin. Immunol. 115:213–219. Peters A, Liu E, Verrier RL, Schwartz J, Gold DR, Mittleman M, Baliff F, Oh JA, Allen G, Monahan K, Dockery DW (2000) Air pollution and incidence of cardiac arrhythmia. Epidemiology 11:11–17. Pietropaoli AP, Frampton MW, Hyde RW, Morrow PE, Oberd€orster G, Cox C, Speers DM, Frasier LM, Chalupa DC, Huang LS, Utell MJ (2004) Pulmonary function, diffusing capacity and inflammation in healthy and asthmatic subjects exposed to ultrafine particles. Inhal. Toxicol. 16(Suppl. 1):59– 72. Pope CA 3rd Thun MJ, Namboodiri MM, Dockery DW, Evans JS, Speizer FE, Heath CW Jr (1995) Particulate air pollution as a predictor of mortality in a prospective study of U.S. adults. Am. J. Respir. Crit. Care Med. 15:669–674. Pope CA, 3rd, Burnett RT, Thun MJ, Calle EE, Krewski D, Ito K, Thurston GD (2002) Lung cancer, cardiopulmonary mortality, and long-term exposure to fine particulate air pollution. JAMA 287:1132–1141. Rom WN (2007) Environmental and Occupational Medicine, 4th edn. Philadelphia: Lippincott-Raven. Rom W, Bitterman PB, Rennard SI, Catin A, Crystal RG (1987) Characterization of the lower respiratory tract inflammation of non-smoking individuals with interstitial lung disease associated with chronic inhalation of inorganic dusts. Am. Rev. Respir. Dis. 136:1429–1434. Romieu I, Sienra-Monge JJ, Ramirez-Aguilar M, Moreno-Macias H, Reyes-Ruiz NI, Estela del RioNavarro B, Hernandez-Avilla M, London SJ (2004) Genetic polymorphism of GSTM1 and antioxidant supplementation influence lung function in relation to ozone exposure in asthmatic children in Mexico City. Thorax 59:8–10. Rosenstock L, Cullen MR, Brodkin CA, Redlich CA (editors) (2005) Textbook of Clinical Occupational and Environmental Medicine, 2nd edn. Philadelphia:Elsevier Saunders. Rutstein DD, Mullan RJ, Todd M, Frazier TM, Halperin WE, Melius JM, Sestito JP (1983) Sentinel health events (occupational): a basis for physician recognition and public health surveillance. Am. J. Public Health 73:1054–1062. Salvaggio JE, Hendrick DJ (1989) The use of bronchial inhalation challenge in the investigation of occupational diseases. In:Spector S,editor. Provocative Challenge Procedures: Background and Methodology.Mount Kisco, NY:Futura. Samet JM, Young RA, Morgan MV, Humble CG, Epler GR, McCloud TC (1984) Prevalence survey of respiratory abnormalities in New Mexico uranium miners. Health Phys .46:361–370. Samet JM, Marbury MD, Spengler JD (1988) Health effects and sources of indoor air pollution. Am. Rev. Respir. Dis. 137:221–242. Samet JM (1989) Definitions and methodology in COPD research. In:Hensley MJ, Saunders NA, editors. Clinical Epidemiology of Chronic Obstructive Pulmonary Disease. New York:Marcel Dekker. Samet JM, Dominici F, Curriero FC, Coursac I, Zeger SL (2000a) Fine particulate air pollution and mortality in 20 U.S. cities. 1987–1994. N. Engl. J. Med. 343:1742–1749. Samet JM, Zeger SL, Dominici F , et al. (2000b) The National Morbidity, Mortality, and Air Pollution Study (NMMAPS). Part 2. Morbidity and Mortality from Air Pollution in the United States. Boston, MA:Health Effects Institute.

REFERENCES

105

Schottenfeld RW, Cullen MR (1986) Recognition of occupational-induced post traumatic stress disorders. J. Occup. Med. 28:365–369. Schwartz DA, Galvin JR, Dayton CS (1990) Determinants of restrictive lung function in asbestosinduced pleural fibrosis. J. Appl. Physiol. 68:1932–1937. Selikoff IJ, Hammond EC, Seidman H (1979) Mortality experience of insulation workers in the United States and Canada, 1943–1976. Ann. N. Y. Acad. Sci. 330:91–116. Sexton K, Ryan PB (1988) Assessment of human exposure to air pollution: methods, measurements, and models. In:Watson, Bates RR, Kennedy D, editors. Air Pollution, the Automobile, and Public Health.Washington, DC:National Academy Press. Shy CM, Kleinbaum DG, aMorgenstern H (1978) The effect of misclassification of exposure status in epidemiological studies of air pollution health effects. Bull. N. Y. Acad. Med. 54:1155–1165. Simon G, Daniell W, Stockbridge H, Claypool K, Rosenstock L (1993) Immunologic, psychological, and neuropsychological factors in multiple chemical sensitivity: a controlled study. Ann. Intern. Med. 119:97–103. Sparks PJ, Feldstein A (1997) The success of the Washington Department of Labor and industries managed care pilot project: the occupational medicine-based delivery model. J. Occup. Environ. Med. 39:1068–1073. Sparrow D, Weiss ST (1989) Methodological issues in airway responsiveness testing. In:Weiss ST, Sparrow D, editors. Airway Responsiveness and Atopy in the Development of Chronic Lung Disease. New York:Raven Press. Speizer FE, Tager IB (1979) Epidemiology of chronic mucus hypersecretion and obstructive airways disease. Epidmiol. Rev. 1:124–142. Spengler JD, Treitmen RD, Tosteson TD, Mage DT, Soczek ML (1985) Personal exposure to respirable particulate and implications for air pollution epidemiology. Environ. Sci. Technol. 19:700–706. Spengler JD, Samet JM, McCarthy JF,(editors) (2000) Indoor Air Quality Handbook.New York: McGraw-Hill. Sun Q, Wang A, Jin X, Natanson A, Duquaine D, Brook RD, Aguinaldo JS, Fayad ZA, Fuster V, Lippmann M, Chen LC, Rajagopalan S (2005) Long-term air pollution exposure and acceleration of atherosclerosis and vascular inflammation in an animal model. JAMA 204:3003–3010. Townsend MC (2000) ACOEM position statement. Spirometry in the occupational setting. American College of Occupational and Environmental Medicine. J. Occup. Environ. Med. 42 (3):228–245. USDHHS (2006) The Health Consequences of Involuntary Exposure to Tobacco Smoke. A Report of the Surgeon General.Atlanta, GA:USDHHS, CDC. U.S. EPA (2004) The Particle Pollution Report: Current Understanding of Air Quality and Emissions through 2003. U.S. Environmental Protection Agency Office of Air Quality Planning and Standards. U.S. EPA (2006) Air Quality Criteria for Ozone and Related Photochemical Oxidants (Final). U.S. Environmental Protection Agency, Washington, DC, EPA/600/R-05/004aF-cF. Utell MJ, Samet JM (1990) Environmentally mediated disorders of the respiratory tract. Med. Clin. North Am. 74:291–306. Utell MJ, Frampton MW, Zareba W, Devlin RB, Cascio WE (2002) Cardiovascular effects associated with air pollution: potential mechanisms and methods of testing. Inhal. Toxicol. 14:1231–1247. Valiante DJ, Rosenman KD (1989) Does silicosis still occur? JAMA 262:3003–3007. Vineis P, Thomas T, Hayes RB, Blot MW, Mason TJ, Pickle LW, Corvia P, Fontham ETH, Schoenberg J (1988) Proportion of lung cancers in males, due to occupation, in different areas of the USA. Int. J. Cancer 42:851–856. Wallace LA, Ott WR (1982) Personal monitors: a state-of-the-art survey. J. Air Pollut. Control Assoc. 32:601–610.

106

CLINICAL PERSPECTIVE ON RESPIRATORY TOXICOLOGY

Ware JH, Dockery DW, Spiro A III, Speizer FE, aFerris BG Jr (1984) Passive smoking, gas cooking, and respiratory health of children living in six cities. Am. Rev. Respir. Dis. 129:366–374. Weiland SK, Bjorksten B, Brunekreef B, Cookson WO, von Mutius E, Strachan DP (2004) Phase II of the International Study of Asthma and Allergies in Childhood (ISAAC II): rationale and methods. Eur. Respir. J. 24:406–412.

5 INDUSTRIAL PERSPECTIVES: TRANSLATING THE KNOWLEDGE BASE INTO CORPORATE POLICIES, PROGRAMS, AND PRACTICES FOR HEALTH PROTECTION Fred D. Hoerger, Larry W. Rampy, Douglas A. Rausch and James S. Bus

The purpose of this chapter is to present an overview of the important considerations involved in establishing policies, programs, and practices for manufacturing, use, and disposal of commercial chemicals in ways that are reflective of the available knowledge of potential hazards. From the standpoint of a manufacturer of a broad and diversified product line of basic chemicals, plastics, and specialty products, such as The Dow Chemical Company, extensive industrial hygiene, occupational health, and product stewardship programs are essential. From the standpoint of the chemical industry, increasing attention is being focused on a program of “Responsible Care” embodying a set of codes relating to such topics as process safety, emissions reduction, and product stewardship. Knowledge of the health effects of environmental toxicants is an essential background component for establishing corporate or industrywide programs to provide adequate protection for workers and the community. From an industrial perspective, health-protective operating and marketing practices must be established before critical reviews are published in compendia such as this book. However, such critical reviews are extremely useful to industrial scientists, physicians, and engineers as a basis for reviewing the literature and current practices on specific chemicals. Equally important, since critical reviews are published on chemicals that have received much investigation, the reviews are useful background education for these disciplines. By analogy, critical reviews provide insight for the design of operating practices and research and monitoring programs for other chemicals.

Environmental Toxicants, Third Edition Edited by Morton Lippmann Copyright Ó 2009 John Wiley & Sons, Inc.

107

108

INDUSTRIAL PERSPECTIVES

Also, many corporations recruiting new industrial hygienists, occupational physicians, and toxicologists seek candidates who have a thorough background in the health effects of the environmental toxicants covered in this book. Consistent with the scope of this volume, this chapter focuses on industry practices that provide human health protection. However, we wish to note that the policies and programs that we describe embody a stewardship for environmental concerns and for other safety concerns, such as process safety and chemical reactivity.

5.1 THE LIFE CYCLE OF A CHEMICAL: MANY POINTS FOR POSSIBLE INTERVENTION Design of appropriate health protection policies, programs, and practices for environmental toxicants requires consideration of the entire life cycle of the substance. A generalized description of a typical life cycle for a commercial chemical is shown in Fig. 5.1 and involves the sequence of manufacture, transportation, use, and ultimate disposition. The life cycle for a specific chemical may take divergent patterns. For example, benzene as a component for gasoline will become widely dispersed in the environment, whereas when benzene is used as a chemical intermediate in the manufacture of styrene, the benzene is consumed, and only relatively small quantities are released to the environment. As another example, formaldehyde has many uses; some uses may involve the potential for exposure at low concentration levels to large numbers of people in the population while some chemical intermediate uses may involve only limited potential for exposure, and only to the workforce involved in the processing. Many of the environmental toxicants reviewed in this book are exclusively or largely unintentional by-products of manufacturing, processing, or energy production, for example, carbon monoxide, diesel exhaust, dioxin, disinfection by-products, nitrogen oxides,

Manufacturing

Knowledge base —Intrinsic properties —Exposure control technology —Educational/training approaches —Regulations/liability —Feedback loops

Packaging, transportation

Use

Recovery, recycle, disposal

FIGURE 5.1

The knowledge base is relevant to all phases of the life cycle of a chemical.

THE KNOWLEDGE BASE FOR THE IDENTIFICATION OF HAZARD CONTROL STRATEGIES

109

ozone, and sulfur oxides. The theoretician may wish to consider the life cycle sequence of environmental toxicant generation, environmental transport, and environmental fate. However, even for such unintentional by-products, the life cycle sequence shown in Fig. 5.1 is useful in design of control and stewardship practices, since it highlights the points at which intervention practices can be considered.

5.2 THE KNOWLEDGE BASE FOR THE IDENTIFICATION OF HAZARD CONTROL STRATEGIES The previous chapters have indicated the complexity of the health effects data base available for many environmental toxicants and the extensive judgmental considerations involved in interpreting the information to determine the toxicological end point(s) to be used in establishing regulatory or risk-reduction controls. Strategies for protection of human health become chemical specific because of the multiplicity of types of health effects that might be of concern from one substance to another, such as pulmonary effects from asbestos, neurotoxicity from lead, and leukemia from benzene. Added to these complexities in designing appropriate controls is the fact that environmental and process safety considerations need to be integrated into the overall risk control strategies. Thus, it seems important to highlight briefly the additional features of the knowledge base that are essential parts of the designing and planning of protective measures during the life cycle of a chemical. Important components of the overall knowledge base are shown in Table 5.1 and include the intrinsic properties of the substance, the current knowledge of exposure control

TABLE 5.1 Components of the Knowledge Base for Establishing Hazard Control Strategies Intrinsic Properties Toxicological and clinical information Flammability and chemical reactivity measurements Other physical and chemical property information Ecological and environmental fate characteristics Product use characteristics Exposure control technology Restriction of use Engineering controls Process efficiency, waste reduction Emission/discharge treatment or destruction Educational and training approaches Labels Material safety data sheets Technical brochures Workforce training Public availability Regulatory requirements’ liability potential Feedback loops Industrial hygiene and environmental monitoring Health surveillance programs

110

INDUSTRIAL PERSPECTIVES

technology, the educational and training approaches that are available, and information feedback loops. The general knowledge of options for exposure control and education/ training can be tailored to the specific substance and to specific points in the life cycle. Several intrinsic properties of a substance may be critical in considering hazard control strategies. Certain flammability and chemical reactivity properties may be critical for process safety and selection of containers and packaging. Knowledge of downstream product use and disposal practices is essential for evaluating points of environmental release. Basic ecological properties such as phytotoxicity and information on toxicity for typical aquatic and mammalian species and environmental bioconcentration potential, fate, and transport may be essential to evaluate potential ecological hazards. Once one has identified the intrinsic properties that are of potential concern at various stages in the life cycle of a chemical, the array of possible exposure control and monitoring technologies can be considered for application. In a few instances, decisions may be made to restrict uses of a substance, such as has been done by manufacturers or regulatory agencies in the cases of asbestos, PCBs, DDT, flame retardants in children’s sleepwear, or cyclamates as a food additive. More significantly, a wide array of engineering controls and operating practices have evolved over the years to minimize environmental release—a few examples include vapor recovery systems for tanks and vents, enclosed systems, designs to improve processing efficiency, waste reduction programs, and biological or chemical treatment or incineration of emissions and discharges. Education and training, that is, approaches to information transfer, are important aspects of control strategies. Over the years, there has been an evolution of information transfer approaches, including product labeling, use of material safety data sheets (MSDSs), product brochures containing summaries of hazard information, precautionary approaches, and frequent recommendations for precautionary design of aspects of storage or operation. Increasingly, there has been attention to the implementation of training programs for the workforce; these have evolved to include much emphasis on leak control, clean up of spills, and on operating sophisticated technology to minimize process perturbations leading to unanticipated releases. The fields of industrial hygiene and environmental monitoring have evolved many analytical methods and techniques for detecting environmental toxicants. These approaches provide a feedback loop for evaluating the effectiveness of control strategies. Another feedback loop to both exposure control and to knowledge of the intrinsic properties is provided by occupational health programs. If health surveillance or epidemiological programs identify a cause–effect relationship between an adverse health effect and an exposure, adjustments in exposure and health protection strategies must obviously be made. Superimposed on the pragmatic considerations outlined above is a fabric of regulations, public perceptions, and potentials for liability that influence corporate policies and programs. Many regulations require extensive record keeping and prescribe emission control standards. The risk of litigation and public perceptions of hazard and risk also influence the types of controls that are implemented. The need for feedback loops from manufacturing operations to the knowledge base and the need to assure compliance with company or regulatory emission or ambient standards have prompted the evolution of industrial hygiene and occupational health programs in many companies. An industrial perspective of the considerations for establishing these programs seems relevant to this review.

INDUSTRIAL HYGIENE AND OCCUPATIONAL HEALTH PROGRAMS

111

5.3 INDUSTRIAL HYGIENE AND OCCUPATIONAL HEALTH PROGRAMS: IMPLEMENTING THE KNOWLEDGE BASE 5.3.1

Exposure Estimation

Many scholarly articles have been written on the subject of measuring exposure to chemicals and other stresses in the occupational setting. The broad subject of how to do this from a sampling and analytical standpoint is beyond the scope of this chapter, but we offer a few observations about the strategies that lead to good estimates of actual exposures while providing a reasonable basis for evaluating and controlling hazards. There are three main objectives of monitoring: 1. to assure that untoward health effects are not likely to be encountered, 2. to describe what the exposure actually was for future use in exposure–response (epidemiology) studies, and 3. to assure that exposures are within legal requirements (e.g., OSHA permissible exposure limits) or consistent with other guidelines (e.g., ACGIH TLVs). These three objectives are not mutually exclusive, but may require somewhat different approaches. Some knowledge of the type of effect a chemical may potentially exhibit and an estimate of its toxicokinetic properties are important to meet the first two monitoring objectives. For example, for gases or vapors that are believed to cause only respiratory irritation, a method should be chosen that will measure peak exposures, as the toxic effect of such materials is likely to be linked more closely to the highest concentration reached than to the total dose received. Hydrogen chloride and other acid gases are examples of chemicals in this category. In contrast, the toxic effects of some materials seem to be linked more closely to the total dose received than to the concentration seen over any short period of time. Vinyl chloride is a good example of this type. Vinyl chloride has been linked to hemangiosarcomas of the liver, and it appears that exposures over relatively long periods (years) are required to produce this effect. This suggests that exposure during a given short period is relatively unimportant compared with cumulative exposures over time. Thus, an exposure measurement strategy that gives an estimate of the time-weighted average over lengthy periods is most appropriate. This aspect has been recognized to some extent, although not strictly based on health considerations, in a European Economic Community (EEC) directive for vinyl chloride. The United Kingdom has already adopted a standard with the same requirements as the EEC directive. It requires that exposure to vinyl chloride be controlled to 3 ppm averaged over a 1year period. In addition, exposures must be controlled to no more than 5 ppm for 1 month, 6 ppm for 1 week, 7 ppm for 8 h and 8 ppm for 1 h. Although this philosophy has not been widely adopted, it does appear to be particularly appropriate for vinyl chloride, and, with sufficient knowledge of their properties, other chemicals should, no doubt, be considered for this type of monitoring and control philosophy. For the second objective, measurements taken to be useful for epidemiology studies, the main need is to know as much as possible about the exposure. The end objective in this case is to estimate the actual dose to people as accurately as possible. When much is known about the biological properties of a chemical, it is easier to design a sampling strategy, but it should be borne in mind that new effects may be discovered in the future. Thus, as complete a description of exposure should be obtained as is feasible. For example, even though it may

112

INDUSTRIAL PERSPECTIVES

be thought that exposure over relatively long periods of time is the most important concern for a material, it is best to estimate both long-term and excursion (short-term) exposures. How the chemical is being used should also be a factor in deciding when, where, and how long to take samples. It is necessary to know what the probability will be of a person actually encountering a given concentration before deciding whether to take a measurement under those particular conditions. For example, if an in-line filter in a chemical process is periodically changed, it would be most sensible to measure the level of chemical encountered over the range(s) of time it takes to change the filter. On the other hand, if a chemical is likely to be present in the general work area either continually or for major portions of a work shift, a full shift-length sample would probably be the best type of sample to take. Other sampling strategies should be considered depending on the way people will encounter the exposure. In designing sampling approaches, it is necessary to consider each situation in the light of the chemical or chemicals to be encountered, their toxic properties, and how they are used. In judging the seriousness or acceptability of an exposure, permissible exposure limits or guidelines would certainly be taken into account, but the toxicokinetic properties that might be dependent on the pattern of exposure, as well as the potential future use of the data for epidemiological studies, should also be considered. This is in contrast to merely characterizing full-shift time-weighted averages, just because the exposure standard or guidelines happens to be written in the most common form, and 8-h time-weighted average. 5.3.2

Biomonitoring

Biomonitoring is defined, for the purposes of this discussion, as the evaluation of the internal exposure of a human to a chemical agent as determined by the analysis of biological specimens. Depending on the chemical agent and biological specimen collected, the internal exposure reflects the amount of the chemical recently absorbed, the amount of chemical in the body, or, less frequently, the amount of the chemical at its action site(s). Biological monitoring techniques potentially offer better estimates of internal exposure than environmental monitoring because biological parameters related to internal exposure are more directly associated with potential systemic adverse health effects than environmental measurements. Biological monitoring takes into consideration absorption by all routes of exposure and measures occupational as well as nonoccupational exposures. Biological monitoring may also allow a better estimate of an individual worker’s exposure than environmental measurements because of individual differences in work habits or because of individual differences in the way the chemical is handled in the body. However, it is important to distinguish between measuring the amount of material and understanding the relationship between that amount and actual exposure or the risk the person is at because of that amount. For example, it is often possible to measure quantitatively the amount of a chemical present in a urine sample. If the exposure that produced that amount cannot be inferred from the urine concentration, then correlations to acceptable or unacceptable limits may be impossible. To make the best use of biological monitoring, it is necessary to know how the level to be measured is related to exposure, so that judgments can be made about risk. Ideally, controlled experiments in humans must be done to determine the quantitative relationship between exposure and concentration of the chemical or metabolite in the medium to be measured. When this is not possible, it may be possible to arrive at concentration ranges of acceptability by referring to animal studies if a model can be constructed with confidence that relates how

INDUSTRIAL HYGIENE AND OCCUPATIONAL HEALTH PROGRAMS

113

the chemical is handled in the animal species with how the chemical is handled by humans. Similarly, biological monitoring techniques may not be appropriate if the primary adverse response involves acute exposure to the respiratory tract, skin, or eyes. Biological monitoring has little value in detecting peak chemical exposure levels unless one knows the time interval between an acute exposure and sample collection as well as the metabolic fate of the chemical in the body. Over the past decade, there has been much research interest in identifying biological markers of exposure. Systematic monitoring of population groups, however, has been somewhat limited. In the design of applied biomonitoring programs, a number of points need to be considered. Sampling times must be carefully selected and controlled with respect to when exposure occurred, when a chemical is rapidly excreted, and when individual variability is great. Programs of biological monitoring should be undertaken only when the relationship between the exposure and the parameter to be measured is known. Sufficient pharmacokinetic information should be available when a chemical or metabolite is to be measured. Finally, appropriate evaluation criteria such as biological limit values should be established before these ongoing programs are undertaken. 5.3.3

Health Surveillance

The goal of all effective monitoring and exposure control strategies should be to achieve exposure results that do not result in adverse changes in health parameters. Health surveillance programs can provide insight on the effectiveness of control strategies in addition to their other purposes, such as the early detection of nonoccupationally related disease. At Dow and a number of other manufacturing and processing industries, a standard physical examination is offered to employees on a periodic basis. This is a general screening approach and is followed up with physician–employee consultation and further diagnosis and treatment as might be indicated (Table 5.2). Some might consider the measurement of such health screening parameters as enzymes related to liver and pulmonary function testing and chest X-rays to be types of biomonitoring. However, it should be kept in mind that these programs are not a substitute for environmental and/or biological monitoring as discussed above.

TABLE 5.2 Product Stewardship: Types of Information and Consultation Provided to Customers Material safety data sheets Technical literature and summary brochures Label instructions Posters and video or audio tapes for training a Phone consultation a Presentations at safety meetings by experts a Seminars on regulatory compliance a Plant visitations and information exchange a Industrial hygiene surveys at customer’s site a Vent stack monitoring a “How-to” presentations at professional and trade association meetings a

These depend on the nature of the product, its intended use, and the resources of the customer.

114

INDUSTRIAL PERSPECTIVES

To be most useful in epidemiological studies, health surveillance programs should be designed to gather information from related groups in a fashion that will permit valid statistical analysis of the data and comparison with appropriate control groups. Epidemiological experts should be consulted when data of this type are to be used in evaluating population or subpopulation groups. In a sense the health surveillance program may be regarded as a feedback loop to the monitoring/exposure control strategy and the toxicological/epidemiological knowledge base. Under ideal circumstances, it provides a degree of support and “comfort” that the exposure control strategy and the knowledge base on potential health effects are appropriate. Infrequently, the feedback may suggest modification in the control strategy or suggest further toxicological or epidemiological studies.

5.4 PRODUCT STEWARDSHIP Knowledge of the potential health and environmental effects of chemicals has been evolutionary, whether indicated by increases in the amount and sophistication of toxicological and epidemiological literature, the number of professionals in the field, or the proliferation of industrial hygiene, environmental health, and environmental medicine curricula and postgraduate research programs. Industrial approaches to expanding the knowledge base and its application to product stewardship have also been evolutionary. The Dow Chemical Company, a pioneer and leader in many aspects of environmental health and product safety, established one of the first toxicology research laboratories for the study of industrial chemicals in the early 1930s. Industrial hygiene and occupational medicine departments were established in the 1940s and epidemiology and environmental science groups in the early 1970s. In this evolutionary process, it was increasingly recognized that the knowledge base had to be transferred to customers, many of whom further processed or formulated the chemicals into other products. For example, safety and material-handling information began to be supplied to customers some 40 years ago. This recognition led to the evolutionary development of what at Dow is called product stewardship—a philosophy or ethic with associated practices to promote a synergistic knowledge partnership between manufacturer and customer. The philosophy of product stewardship, or product safety as it is referred to in some companies, is now practiced by a number of chemical companies. The remainder of this chapter reflects the relatively advanced product stewardship programs of Dow and a few other major chemical suppliers and then briefly describes the current development of an industrywide program called Responsible Care, which is based on these advanced company programs. In 1972, Dow adopted the following product stewardship philosophy: The Dow Chemical Company has a fundamental concern for all who make, distribute, and use its products, and for the environment in which we live. This concern is the basis for our product stewardship philosophy by which we assess the safety, health, and environmental information on our products and then take appropriate steps to protect employee and public health and the environment. Our product stewardship program rests with each and every individual involved with Dow products—from the initial concept and research to the manufacture, sale, distribution, use, and disposal of each product.

In short, product stewardship is a commitment to action—to “do the right things.” This commitment must permeate all levels and functions of the organization, from top management policy to planning to day-to-day decisions and practices.

PRODUCT STEWARDSHIP

115

The purpose of product stewardship is fivefold: 1. 2. 3. 4. 5.

protects employees, public health, and the environment, protects products from the environment, helps reduce liability, helps prevent adverse publicity, and builds trust with your own employees and your customers.

Each of these purposes is self-explanatory. Although the protection of people and the environment is most important, it is extremely important that our products be protected from the environment—an environment in which our materials could be misused, abused, and disposed of improperly. All of which could lead to calls for a product ban, product restrictions, and tighter regulations. Product stewardship must be the responsibility of all employees. It cannot just be the responsibility of a safety coordinator or a single department. It ranges from the chemist experimenting on a product for the future to the salesman taking orders for large quantities of an established commodity such as caustic soda. Industrial health and environmental scientists have the responsibility for generating safety data evaluating its impact on potential or existing products applications, and ultimately communicating this information in appropriate terms to all concerned. Manufacturing people must inform all employees of the potential health effects, proper safety practices, and other product facts and assure that necessary equipment is available and used. It is also their responsibility to adhere to pollution control and industrial hygiene standards and practices. Stewardship does not end when a product leaves the plant, however. Distribution people must select carriers, warehouses, and terminals that will perform consistently within guidelines and assure that products reach the customer in a safe manner. Marketing people must furnish customers and distributors with appropriate handling and application information, and be on the lookout for potential misuse, mishandling, or improper disposal of products. They should form a supplier–customer partnership that promotes safe uses and applications of chemicals. Another key person to a product safety program is what at Dow Chemical is called the product steward—the person responsible for recommendation of needed health and environmental studies and safety assessments for each product. In addition, these stewards help evaluate the appropriateness of a customer’s use and disposal of a product and facilitate preparation of safety training tools. Each Dow product has an assigned steward who maintains contact with sales representatives, customers, health and environmental scientists, and government regulators to assure that appropriate health and environmental information and concerns are considered in all aspects of product development and use. A number of considerations are central to a product evaluation. How is the product manufactured? What are its raw materials? What is the manufacturing process? What impurities does a product contain, and what problems can be expected if changes are made in the process? In addition, product distribution must be evaluated. Is the product transported by road, rail, or water? Does it go through terminals? Is it handled by distributors? How is it packaged? Other considerations include how the product will be used and how, potentially, it can be misused. Can it be ingested? Inhaled? Touched? What are recommended disposal practices? Will it be burned? Land filled? What would happen if the product is spilled? Additional considerations include evaluation of the probability and extent of human exposure. Two or three decades ago, only worker exposures of healthy males between the

116

INDUSTRIAL PERSPECTIVES

ages of 18 and 65 were considered. Today the unborn fetus of a pregnant employee must be considered. And if the product leaves a plant site, the potential for adverse effects on sick people, children, the elderly—every demographic group—must be considered. Furthermore, the probability and amount of release to the air, water, and land must be evaluated along with the toxicity and persistence of the product in the environment. In more recent years, health and environmental concerns have often dominated product stewardship evaluations. However, safety concerns involving fire, explosion, and reactivity cannot be underestimated or overlooked. It should never be forgotten that the Bhopal, India, incident in 1984 was really caused by a reaction of methyl isocyanate with water, and the tragic accident earlier the same year in Mexico City was the result of a fire and explosion of a hydrocarbon. How much health and environmental data are needed on a new product is a question always asked during the developmental cycle. For early stage research work, range-finding toxicological data are obtained. These studies include acute oral, eye irritation, and skin irritation studies. In addition, it is at this point that reactive chemical data should be obtained to determine if a product is shock sensitive, is flammable, undergoes exothermic decomposition, and so on. Its reactivity with other chemicals and construction materials should also be determined or anticipated. When a potential new product reaches pilot plant stage, range-finding studies may still suffice, depending on the results and the applications considered for the material. Finally, when a material is sent to potential customers for applications development, a review must be conducted to determine the need for more acute and subchronic toxicological and environmental testing. It is at this point that long-range studies must be considered and, if needed, planned for the future. An important link in the product stewardship philosophy is to establish a supplier– customer partnership in health, safety, and environmental matters. Experience has shown that in addition to having a high-quality product at a competitive price, the supplier must be the buyer’s expert in avoiding injury or environmental damage from downstream use of its products—a trend that is in the supplier’s own interest. The time when a company could sell a product merely on the basis of a competitive price is past. Today, the key to successful marketing is to reduce the knowledge gap between seller and buyer. The supplier has to focus attention on the buyer’s needs and experience in handling chemicals. Regulation is only one factor influencing these needs; the most essential factors are product knowledge and an attitude of commitment to developing and carrying out sound practices. Knowledge about health, safety, and environment is a specialty the supplier must offer with its products. This knowledge is essential for customer success. Only if customers are successful will the supplier be successful. The amount of time and effort required for product stewardship depends on both the properties of the product and the resources and expertise of the customer. Some of the approaches to supplier–customer interchange are shown in Table 5.1. Obviously, more supplier resources are required for a highly toxic product such as chlorine being used by a customer with limited resources than for a polystyrene resin sold to another large manufacturing company. In both cases, however, it is important that a safety partnership be developed between the supplier and the customer. The challenge for the future is to build supplier–customer relationships on an industrywide basis with emphasis on all areas of stewardship—manufacturing, transportation, storage, use, and disposal. This challenge is being addressed by the American Chemistry Council (ACC) renamed from Chemical Manufacturers Association (CMA) in 2000), a trade association representing companies that produce approximately 90% of U.S. industrial

RESPONSIBLE CAREÒ

117

chemicals. In 1988, the Association began an industrywide initiative to enhance operational performance on all these matters. Adherence to Responsible CareÒ is condition for membership in the ACC. 5.5 RESPONSIBLE CAREÒ Responsible CareÒ has two critical program elements. The first is intended to guide companies toward continually improving their health, safety, and environmental performance; the second is designed to assist companies to do a better job of understanding and responding to public concerns about safely managing the use of chemicals. The framework to accomplish these objectives consists of 10 Guiding Principles, which serve as the operational philosophy of the program, and 6 Codes of Management Practices, which describe the key activities companies must undertake to manage chemicals as safely as possible while constantly improving performance. The purpose of the Guiding Principles is to provide the foundation for instituting cultural change within the chemical industry, change resulting in improved openness with the public accompanied by a commitment to continued improvement in performance. The Guiding Principles are described as the following: 1. Recognize and respond to community concerns about chemicals and production operations. 2. Develop and produce chemicals that can be manufactured, transported, used, and disposed of safely. 3. Make health, safety, and environment considerations a priority in planning for all existing and new products and processes. 4. Report promptly to officials, employees, customers, and the public information on chemical-related health or environmental hazards and to recommend protective measures. 5. Counsel customers on the safe use, transportation, and disposal of chemical products. 6. Operate plants and facilities in a manner that protects the environment and the health and safety of employees and the public. 7. Extend knowledge by conducting or supporting research on the health, safety, and environmental effects of products processes and waste materials. 8. Work with others to resolve problems created by past handling and disposal of hazardous substances. 9. Participate with government and others in creating responsible laws, regulations, and standards to safeguard the community, workplace, and environment. 10. Promote the principles and practices of Responsible CareÒ by sharing experiences and offering assistance to others who produce, handle, use, transport, or dispose of chemicals. The first code of management or practice, community awareness and emergency response (CAER), was approved by CMA in 1989. This program element had its beginnings prior to 1989, however, when the CMA realized that following the 1984

118

INDUSTRIAL PERSPECTIVES

chemical release incident in Bhopal, India, few of its members had emergency plans that addressed what responses needed to be implemented and coordinated with local emergency services should an accidental release occur. The CAER program action steps, many of which are incorporated into elements of the 1986 Superfund Amendment Reauthorization Act (SARA), required formation of State Emergency Response Commissions and Local Emergency Planning Committees (LEPCs). Importantly, this code element also mandated companies to implement community outreach through formation of Community Advisory Panels (CAPs). CAP members are citizens who provide perspectives of issues and questions arising from chemical manufacturing facilities located within their communities. As of 1998, over 300 CAPs have been established in communities where chemicals are manufactured across the United States. The CAER program also provides an important mechanism for the chemical industry to facilitate the development of community communication plans consistent with the EPA-mandated Risk Management Planning (RMP) rule, which requires disclosure of worst-case release scenarios to public emergency response teams. The second management practice code, pollution prevention, was adopted in 1990 and requires both reporting and a commitment to reduction of emissions. EPA annually publishes emission information as the Toxics Release Inventory (TRI). The combination of these programs has had significant impact on pollution prevention; ACC members achieved over a 77% reduction in chemical emissions during 1988–2004, despite a 47% increase in chemical production. Public reporting of emission data is intended to encourage continued improvement, coupled with trackable public progress, in future emission reduction programs. For example, the Dow Chemical Company announced 2005 goals of achieving reductions in air and water emissions of 75% for priority chemicals (e.g., carcinogens, persistent, toxic, and bioaccumulative chemicals) and 50% for other chemical emissions. These targeted reductions are to be accompanied by a 50% reduction in waste and wastewater, and a 20% reduction in energy, generated per pound of chemical production. Progress against the goals can be viewed by the public on the Dow Web site www.dow.com. The employee health and safety management code requires Occupational Injury and Illness Reports (OIIR) be submitted semiannually to the ACC/CMA. These are the same data that must be recorded under the OSHA record-keeping standards and provide a publicly visible measurement of industry progress; in 2004, ACC members improved performance by 10% over 2003 and the overall incidence rate was 4.5-fold less than that of the overall U.S. manufacturing sector. The fourth and fifth safety management codes are distribution and process safety. Beginning in 1995, the CMA developed a DOT Hazardous Materials Incident Report database for measuring performance against distribution-related incidents. By 2005, reportable distribution accidents were reduced by 49% despite a 10% increase in overall volume of chemicals transported (approximately 778 million pounds transported by truck, rail, and other modes in 2005). The reports provide improved reporting and investigation of incidents and strengthen implementation of preventative measures. The process safety code identifies improved processes for prevention of fires, explosions, and accidental releases, and is aligned with regulatory mandates such as the OSHA process safety rule and the EPA risk management planning rule. The sixth and last management code is product stewardship. This process, as described earlier in this chapter, supports management of environmental, health, and safety risks through the entire life cycle of a product design, manufacture, marketing, distribution, use, recycling, and disposal. An important element of this code is collection and communication

CONCLUDING PERSPECTIVE

119

of safety data to customers. Another critical code element is a commitment to conduct of research that contributes to knowledge necessary for evaluating chemical risks. In 1998, the CMA announced a plan, in cooperation with the EPA, to initiate a voluntary testing program to develop expanded health and environmental safety information on a spectrum of high production volume (HPV) chemicals. The approximate 2000 HPV chemicals (each produced in volumes greater than 1 million pounds per year) account for approximately 95% of total chemical production volume in the United States. The information obtained from these studies will be shared with the EPA and the public through a readily available database at www.epa.gov/chemrtk. In response to concerns about health effects of chemicals in children, the ACC launched a pilot program, the Voluntary Children’s Chemical Evaluation Program (VCCEP) in 2000. Developed in concert with the U.S. EPA, the VCCEP pilot was designed to characterize the exposure and health risks of 23 chemicals believed to present potential exposures to children. The results of the program can be tracked on an EPA Web site, www.epa.gov/oppt/vccep/index.htm. In 1999, the ACC also initiated a research program, the Long Range Research Initiative ((LRI) www.uslri.org) intended to improve the generic science-based understanding of potential chemical risks to humans and the environment. This program included assuming governance and partial financial support for the Chemical Industry Institute of Toxicology (CIIT), a not-for-profit research institute committed to independent research into mechanisms of chemical toxicity. In 2007, the CIIT became a standing institute within the Hamner Institutes. This new Hamner Institutes organizational structure is designed to tap the expertise and technologies available with the broader biological research community, for example, pharmaceuticals and academia, for application to the research needs of the chemical industry. A key issue fundamental to the public credibility of Responsible CareÒ is external verification of performance. To facilitate this objective, CMA has implemented a management systems verification (MSV) program. This process includes reviews conducted by CMA member company peers and community representatives and individual company compliance with the management codes, and supports the integrity of the initiative to key audiences including employees, local communities, public officials, and others. The Responsible CareÒ activities and progress of the members of the ACC can be viewed on their intranet Web site www.americanchemistry.org.

5.6 CONCLUDING PERSPECTIVE From an industry perspective, meeting health, safety, and environmental concerns is and will remain a top priority. However, there is often a gap between public perceptions of risk and realities of risk as evaluated by scientific methods—sometimes perceived concerns have been without scientific foundation; in other instances, there has been a failure to recognize hazards or to respond in a timely manner. Critical reviews of the literature such as those in this book assist professionals in a position to advise on these gaps and thereby focus resources on the more important priorities.

6 DRINKING WATER DISINFECTION BY-PRODUCTS Richard J. Bull

6.1 INTRODUCTION The disinfection of drinking water brings competing health concerns more sharply into focus than most other environmental problems. What is the contribution of drinking water disinfection to the control of infectious disease? What is the magnitude of the risk of cancer or other health hazards that are associated with the by-products of disinfection reactions? Are there alternative methods that reduce one of these risks without exacerbating the other to unacceptable levels? The reciprocity implied by these questions should not actually be thought of as trade-offs because preventing waterborne infectious disease is the primary responsibility of public water suppliers. The questions do illustrate the need for systematic evaluation of each new approach to treatment to ensure that decisions are not made that put the public health at greater risk. Waterborne infectious disease was common in the United States before the introduction of disinfection (Akin et al., 1982; Bull et al., 1990a). The introduction of chlorination and other improvements in drinking water treatment, particularly filtration, contributed to a very sharp decline in deaths from cholera, typhoid and a variety of other infectious diseases (Akin et al., 1982). Waterborne infectious disease does still occur and its occurrence is most frequently associated with inadequate disinfection (e.g., equipment failure or water plant operator error) (Blackburn et al., 2004). There is even some suggestion that endemic infectious diseases in large cities might well be attributed to well designed and maintained urban water systems (Payment et al., 1990a, 1990b). Although it is difficult to quantitatively estimate the impact of suspending the use of disinfectants in drinking water, recent history (ILSI, 1993) allows little doubt that the problem of waterborne infectious disease is a real, not an abstract problem. Thus, the disinfection of drinking water is a well-established and

Environmental Toxicants, Third Edition Edited by Morton Lippmann Copyright
2009 John Wiley & Sons, Inc.

121

122

DRINKING WATER DISINFECTION BY-PRODUCTS

effective means of preventing pathogenic organisms from being transmitted through public water supplies. Since the 1970s the evidence has steadily increased to show that(1) by-products are generated in the use of disinfectants; and (2) these compounds have toxicological properties that are of potential concern to human health (IARC, 1991, 1999; Bull et al., 1995). These data have been paralleled by epidemiological associations between chlorinated water and an increased incidence of cancer. There are epidemiological data to suggest disinfected drinking water might contribute to reproductive/developmental toxicities, but these findings are less consistent. There are usually two points where disinfectants are added. The first is made early in the treatment process, with the intent of inactivating all pathogenic microbes and viruses that might contaminate the source water. This is referred to as primary disinfection (Bull et al., 1990a). In most water supplies there is a second addition of disinfectant as the water leaves the treatment plant and occasionally at distal sites in the distribution system. These additions are referred to as post- or secondary disinfection. Their intent is to reduce colonization of the water distribution system and to minimize the impact of sources of contamination in the system (e.g., cross-connections, main breaks, etc.). Thus, the overall process may involve successive additions of the same disinfectant or utilize different disinfectants in each of the locations to minimize operational issues as well as by-product formation. Because the number of disinfectant by-products (DBPs) identified continues to expand, it is not possible to comprehensively review the field in the small space available. Consequently, this chapter begins with a short general introduction to commonly used disinfectants and is followed by a description of many, but not all DBPs known to be produced. An overview of epidemiological data is included to provide some perspective on the risks that have been associated with disinfection of drinking water. A review of the toxicological properties of this relatively small number of DBPs will focus on whether these compounds contribute significantly to the risks that have been suggested based upon in epidemiological studies of disinfected water. Toxicological data provide the only means available for estimating the potential risks of methods of disinfection other than chlorine.

6.2 CHEMICAL METHODS OF DISINFECTION The chemical disinfectants most commonly utilized in public water supplies are chlorine, chloramine (chlorine þ ammonia), ozone, and chlorine dioxide in approximate order of popularity. Iodine is also used, but it is limited to short-term emergency use (e.g., backpackers or the field situations by the military). Recent studies have shown that relatively small intakes of iodine lead to elevation of thyroid stimulating hormone (Robison et al., 1998). There are significant numbers of individuals in the population that may react adversely to levels of iodine needed to ensure disinfection (Carswell et al., 1970; Clark et al., 1990; Woeber, 1991). Thus, iodine is not likely to be utilized on a continuous basis in municipal systems. The use of membranes in drinking water treatment is becoming more popular. Membranes are configured in microfiltration, ultrafiltration, or reverse osmosis systems. Microfiltration and ultrafiltration are included primarily to prevent fouling of the reverse osmosis membranes that are generally responsible for removing chemical pollutants, but they contribute to the removal of microorganisms at the treatment plant as well. Reverse osmosis removes much of the natural organic matter (NOM) and inorganic chemicals present in source water that react with the disinfectants to form DBPs. Consequently, the options for

CHEMICAL METHODS OF DISINFECTION

123

minimizing by-product formation are increasing as new technology is introduced. To date, the expense of such systems has limited to their use to indirect potable reuse of municipal wastewater. The four commonly used disinfectants vary in their ability to achieve the goals of primary and secondary disinfection (Bull et al., 1990a). The high reactivity of ozone makes it an excellent primary disinfectant, but has insufficient stability to be useful in secondary disinfection. Chloramine is weak as a primary disinfectant, requiring very long contact times for it to be effective. However, its stability makes it an excellent choice as a bacteriostatic agent in the distribution system. Chlorine and chlorine dioxide are effective in both situations if sufficient quantities are added to satisfy the disinfectant demand. Once the demand has been satisfied, reasonably stable residuals can be maintained in the distribution system. When chlorine is used as the disinfectant, it is added either as chlorine gas that forms hypochlorous acid in water, or it is added as a solution of the hypochlorite salts. Hypochlorous acid dissociates to hypochlorite and hydrogen ions with a pKa of 7.5. At equivalent pH in the finished water, these processes yield the same reactive chemical species. The efficacy of OCl hypochlorite as a disinfectant is less than that of HOCl (Akin et al., 1982). Chlorine dioxide can be prepared in situ from chlorite or chlorate. Generation from the acidification of chlorite is generally favored in drinking water applications, whereas generation from chlorate is favored in bleaching of pulp and other industrial processes. Acidification of sodium chlorite is accomplished by the addition of mineral acid or chlorine (Masschelein, 1989). As a consequence of how it is prepared and reactions of chlorine dioxide in solution, chlorine dioxide disinfected water will contain varying amounts of chlorite, chlorate and free chlorine. Chlorine dioxide is as effective as chlorine as a disinfectant, but it has the advantage that its effectiveness is less sensitive to changes in pH (Akin et al., 1982). Chloramine is prepared by the addition of ammonia to chlorine, which can be done simultaneously or sequentially. The type of chloramine formed depends upon the relative ratios of chlorine to ammonia and the pH. Generally speaking, every effort is made to optimize for formation of monochloramine, since dichloramine and trichloramine create severe taste and odor problems. Because chloramine is less effective as a disinfectant than chlorine (Akin et al., 1982), the addition is often made sequentially to allow for exposure to free chlorine because of its greater disinfecting power. Once formed, however, chloramine is much more stable than chlorine and, for this reason, it is preferred as a means of maintaining a disinfectant residual in the water distribution system (Olivieri et al., 1986). Ozone is a very reactive chemical. For this reason it is always generated on site (Fahroog et al., 1977). Ozone is frequently used in combination with other disinfectants or oxidants (AWWA, 1982). Hydrogen peroxide used in conjunction with ozone facilitates formation of hydroxyl radical (Wolfe et al., 1989), increasing its effectiveness in breaking down various organic materials in the source waters (Hoigne and Bader, 1975). A similar enhancement of hydroxyl radical can be accomplished by combining ozone with ultraviolet radiation (UV) (Rice and Gomez-Taylor, 1986). As a primary disinfectant, ozone is the most effective chemical disinfectant currently in use. For example, it is quite effective in inactivating oocysts of Giarida and Cryptosporidium, two organisms that are known for their resistance to chlorine. However, as noted earlier, its high reactivity with water makes ozone ineffective for postdisinfection. As ozone use increases in the United States, a second disinfectant, commonly chlorine or chloramine, is used to prevent the growth of organisms in the distribution system.

124

DRINKING WATER DISINFECTION BY-PRODUCTS

Physical methods of disinfecting drinking water are becoming more common in the United States largely due to the discovery that UVis very efficient in killing Cryptosporidium spores. A number of treatment systems based on electrolytic production of oxidants have also been introduced. The reaction products from such systems have been very poorly described in the open scientific literature. A general problem is that many of electrolytic systems have not been well characterized either in terms of the nature of the oxidants that are produced or the potential for unusual by-product formation. Consequently, they will not be further considered in this review.

6.3 CHEMICAL NATURE AND OCCURRENCE OF DISINFECTANT BY-PRODUCTS The universal property of chemical disinfectants is that they are oxidants. Rice and GomezTaylor (1986) compiled the oxidation potentials and the relative disinfection power of each. These are displayed in Table 6.1. Hydroxyl radical is included in the list because it is one of the reactive species produced with ozone, particularly in the presence of hydrogen peroxide. The most prominent difference among the disinfectants is the degree to which they halogenate NOM. This is illustrated in Fig. 6.1 where the formation of total organic halogen (TOX), as a fraction of the total organic carbon (TOC) with the different disinfectants, is compared. These experiments were conducted using a fulvic acid isolated from a natural water and reconstituted in purified water at 3 mg TOC/L, a reasonable approximation of the TOC found in most surface waters (Zhang et al., 2000). The experiment selected for the comparison included bromide in the water at 0.1 mg/L to illustrate that halogenation reactions do occur with ozone and chlorine dioxide. Clearly, chlorine produces more TOX. Chloramine produces 25%, and chlorine dioxide and ozone about 10% of that produced by chlorine. The levels observed in the absence of bromide would be much lower with ozone, whereas levels might vary somewhat with chlorine dioxide depending upon the conditions of its application (i.e., it is sometimes generated by the addition of free chlorine to sodium chlorite solutions). TABLE 6.1

Power of Alternative Drinking Water Disinfectantsa

Disinfectant

Oxidation Potential, at 25 C, V b

Relative Oxidation Powerc

Effectiveness as a Disinfectant

2.80 2.07 1.77 1.49 1.36 1.33 1.275 1.16 0.99

2.05 1.52 1.30 1.10 1.00 0.98 0.94 0.85 0.73

High Moderate High High High High Low High

Hydroxyl radical Ozone Hydrogen peroxide Hypochlorous acid Chlorine Hypobromous acid (bromine) Chlorine dioxide Monochloramine Hypoiodous acid (iodine) a

From Rice and Gomez-Taylor (1986). Relative to the hydrogen electrode. c Relative to chlorine ¼ 1.00. b

CHEMICAL NATURE AND OCCURRENCE OF DISINFECTANT BY-PRODUCTS

125

FIGURE 6.1 Formation of total organic halogen (TOX) as a function of total organic carbon (TOC), and fraction of TOC unidentified, for different disinfectants, that is, hypochlorous acid (HOCl), chloramine (NH2Cl), chlorine dioxide (ClO2), and ozone (O3).

The halogenated by-products that are produced by each of the disinfectants are somewhat different. In the case of chlorine, about 50% of the TOX can be accounted for by known DBPs (e.g., trihalomethanes [THMs], haloacetates [HAs], haloacetonitriles [HANs], haloacetaldehyes, and some minor products). The unknown fraction of TOX with other disinfectants is much larger, approaching 90%. While halogenation is lower, the compounds produced are different, at least in part, from those produced by chlorine. It should also be noted that there is more nonhalogenated carbon than halogenated carbon in disinfected water. The nature of these chemicals has had only superficial study. However, it is probable that many of the products in this mixture have carbons that are variously oxidized, such as carboxylic acids, aldehydes, alcohols, and ketones. Chemicals identified as DBPs have been reviewed by Richardson (1998). Oxidation reactions as well as substitution reactions form DBPs. Table 6.2 is a partial list of compounds, by general chemical class that are found as DBPs. The more representative concentrations were drawn from the survey conducted under the auspices of the Information Collection Rule (ICR) (McGuire et al., 2002). To provide perspective on concentrations of DBPs not measured in that survey, some earlier surveys are quoted that may no longer be representative. There are three important points: (1) the most frequently identified by-products are halogenated, but these are most frequently associated with the use of chlorine gas or hypochlorite; (2) there are nonhalogenated by-products associated with all disinfectants; and (3) there are by-products that are uniquely associated with a particular disinfectant. The recognition of halogenated by-products is due in part to the fact that halogen substitution is easily identified by mass spectrometry. Oxygen substations that arise from oxidation reactions of NOM are not so unique. NOM is a mixture of natural substrates (e.g., amino acids), humic and fulvic acids are the major precursors for disinfectant by-products in most water supplies (Thurman, 1986). The humic and fulvic acids are a mixture of partially polymerized decayed biological matter that characterized only in a general way (Hwang et al., 2001). The reaction of disinfectants with NOM can be extensive, and the resulting products have not been fully characterized. Therefore, the

126

DRINKING WATER DISINFECTION BY-PRODUCTS

TABLE 6.2

Organic By-Products of Drinking Water Disinfection

Chemical Class

Median Concentrations Ranges (mg/L) Found (mg/L)

References

Trihalomethanes Chloroform Bromodichloromethane Dibromochloromethane Bromoform Dichloroiodomethane

12 8 4 0.6 0.3

0.4–120 1–8 ND–40 ND–19 ND–3

McGuire McGuire McGuire McGuire McGuire

Haloacids Chloroacetate Bromoacetate Dichloroacetate Bromochloracetate Dibromoacetate Trichloroacetate Bromodichloroacetate Dibromochloroacetate Tribromoacetate 2-Chloropropanoate 2,2-Dichloropropanoate 3,3-Dichloropropanoate Chlorobutanedioic acid 2,2-Dichlorobutanoate 3,3-Dichloropropenoic

ND ND 14 4.3 1.3 5.0 3.9 1.7 ND Detected Detected Detected Detected Detected ND

ND–8.2 ND–2.1 1.3–40 1.2–19 ND–18 ND–80 ND–17 ND–15 ND–3.6 Not quantified

ND–4.7

McGuire et al., 2002 McGuire et al., 2002 McGuire et al., 2002 McGuire et al., 2002 McGuire et al., 2002 McGuire et al., 2002 McGuire et al., 2002 McGuire et al., 2002 McGuire et al., 2002 Krasner et al., 1989 Krasner et al., 1989 Krasner et al., 1989 Krasner et al., 1989 Krasner et al., 1989 Weinberg et al., 2002

Haloacetamides Chloroacetamide Bromoacetamide 2,2-Dichloroacetamide Dibromoacetamide Trichloroacetamide

ND ND 1.3 0.6 0.3

ND–0.5 ND–1.1 ND–5.6 ND–2.8 ND–1.1

Weinberg et Weinberg et Weinberg et Weinberg et Weinberg et

al., al., al., al., al.,

2002 2002 2002 2002 2002

Haloacetonitriles Chloroacetonitrile Bromoacetonitrile Dichloroacetonitrile Bromochloroacetonitrile Dibromoacetonitrile Trichloroacetonitrile

ND ND 1 0.6 0.3 ND

ND–0.9 ND–0.2 ND–12 ND–3 ND–3 ND–0.4

Weinberg et Weinberg et Weinberg et Weinberg et Weinberg et Weinberg et

al., al., al., al., al., al.,

2002 2002 2002 2002 2002 2002

Haloaldehydes Chloroacetaldehyde Dichloroacetaldehyde Bromochoroacetaldehyde Trichloroacetaldehdye Tribromoacetaldehyde

ND 1 0.3 1 ND

ND–2.4 ND–14 ND–4 ND–16 ND–3

Weinberg et al., 2002 McGuire et al., 2002 McGuire et al., 2002 McGuire et al., 2002 McGuire et al., 2002

et et et et et

al., al., al., al., al.,

2002 2002 2002 2002 2002

CHEMICAL NATURE AND OCCURRENCE OF DISINFECTANT BY-PRODUCTS

TABLE 6.2

127

(Continued)

Chemical Class

Median Concentrations (mg/L)

Haloketones Chloropropanone 1,1-Dichloropropanone 1,3-Dichloropropanone 1,1-Dibromopropanone 1,1,1-Trichloropropanone 1,1,1-Tribromopropanone 1,1,3-Trichloropropanone 1,1,3-Tribromopropanone 1,1,1,3-Tetrachloropropanone 1,1,3,3-Tetrachloropropanone 1,1,3,3-Tetrabromopropanone

Ranges Found (mg/L)

References

ND 0.5 ND ND 0.8 ND ND ND ND ND ND

ND–2 ND–2 ND ND–0.4 ND–7 ND ND–0.3 ND–0.1 ND– 51 ND–0.6 ND–0.6–2

McGuire McGuire McGuire McGuire McGuire McGuire McGuire McGuire McGuire McGuire McGuire

ND

ND–0.31

Weinberg et al., 2002

ND

ND–0.17

Weinberg et al., 2002

ND

ND–0.03

Weinberg et al., 2002

ND

ND–0.04

Weinberg et al., 2002

ND

ND–0.10

Weinberg et al., 2002

ND

ND–0.72

Weinberg et al., 2002

ND

ND–0.81

Weinberg et al., 2002

ND

ND–0.41

Weinberg et al., 2002

0.01

0.01–0.58

Weinberg et al., 2002

ND

ND–0.23

Weinberg et al., 2002

ND

ND

Weinberg et al., 2002

ND ND

ND–0.71 ND–0.31

Weinberg et al., 2002 Weinberg et al., 2002

Halophenols 2,4-Dichlorophenol 2,4,6-Trichlorophenol

Detected

0.01–1.4 Not quantified

Aldehydes (nonhalogenated) Formaldehyde (ozone) Acetaldehyde (ozone)

13 5

4–27 2.0–20

Halofuranones and related compounds 3-Chloro-4-(dichloromethyl)-5hydroxy-2(5H)-furanone (MX) 3-Chloro-4-(bromochloromethyl)-5hydroxy-2(5H)-furanone (BMX-1) 3-Chloro-4-(dibromomethyl)-5hydroxy-2(5H)-furanone (BMX-2) 3-Bromo-4-(dibromomethyl)-5hydroxy-2(5H)-furanone (BMX-3) (E)-2-Chloro-3-(dichloromethyl)-4oxobutenoic acid (EMX) (E)-2-Chloro-3-(bromochloromethyl)4- oxobutenoic acid (BEMX-1) (E)-2-Chloro-3-(dibromomethyl)-4oxobutenoic acid (BEMX-2) (E)-2-bromo-3-(dibromomethyl)-4oxobutenoic acid (BEMX-3) 3-Chloro-4-(dichloromethyl)-2(5H)furanone (Red-MX) (Z)-2-Chloro-3-(dichloromethyl)-4oxobutenoic acid (ZMX) (Z)-2-Chloro-3-(dichloromethyl)butenedioc acid (Ox-MX) Mucochloric acid (ring) Mucochloric acid (open)

et et et et et et et et et et et

al., 2002 al., 2002 al., 2002 al., 2002 al., 2002 al., 2002 al., 2002 al., 2002 al., 2002 al., 2002 al., 2002

Symons et al., 1975 Symons et al., 1975

Glaze and Weinberg, 1993 Glaze and Weinberg, 1993 (continued)

128

DRINKING WATER DISINFECTION BY-PRODUCTS

TABLE 6.2

(Continued)

Chemical Class Glyoxal (ozone) Methylglyoxal (ozone) Benzaldehyde 2-Methyl propanal 2-Methyl butanal Phenylacetaldehyde Decanal Methoxydimethyl octanal C2-Benzaldehydes Halonitromethanes Chloronitromethane Bromonitromethane Dichloronitromethane Bromochloronitromethane Dibromonitromethane Chloropicrin Bromodichloromethane Dibromochloromethane Bromopicrin Miscellaneous Cyanogen chloride

Median Concentrations (mg/L)

Ranges Found (mg/L)

8.3 6 Detected Detected Detected Detected Detected Detected Detected

5–15 2–11 Not quantifed

ND ND ND ND ND ND ND ND ND

ND–0.8 ND–0.3 ND–51 ND–53 ND–0.6 ND–2 ND–3 ND–3 ND–5

50.2–4.5

References Glaze and Weinberg, 1993 Glaze and Weinberg, 1993 Stevens et al., 1990 Hrudey et al., 1988 Hrudey et al., 1988 Hrudey et al., 1988 Daignault et al., 1987 Daignault et al., 1987 Daignault et al., 1987

McGuire McGuire McGuire McGuire McGuire McGuire McGuire McGuire McGuire

et et et et et et et et et

al., al., al., al., al., al., al., al., al.,

2002 2002 2002 2002 2002 2002 2002 2002 2002

Krasner et al., 1989

simple listing of chemicals and/or classes that are identified in Table 6.2 is likely to be a significant under-representation of the variety of chemicals that are produced. Exposure to DBPs is not a simple function of the disinfectant that is used in a particular water supply. The amount of by-products produced and their chemical character depends upon the characteristics of the water that is being treated. There are four major water variables that are important: TOC concentration, the chemical nature of the TOC, bromide concentration, and pH. The relationship between TOC and total DBP formation (measured as TOX, for example) may be linear within a given water supply; however, the yield of individual by-products will vary depending upon the nature of NOM present at the time and whether specific precursors are depleted at low concentrations of disinfectant (Reckhow et al., 1990). In many climates, the character of the NOM varies seasonally. Relative yields of THMs and haloacetates from particular fractions of NOM can vary by more than an order of magnitude. The proportion these fractions are of the total NOM varies seasonally (Hwang et al., 2001). Of course, there are also geographical differences in the character of NOM. pH affects the type of DBP that is formed. Fig. 6.2 provides one example of how pH can influence net formation of a given set of DBPs. Compounds from different classes behave quite differently (Stevens et al., 1990). Chloroform, an example of the behavior of the THM class, increases substantially in concentration as pH increases. Conversely, trichloroacetate (TCA) concentrations decrease over the same pH range. The concentrations of chloral hydrate also decrease with pH. At alkaline pH, trihaloacetates and trihaloacetaldehydes

CHEMICAL NATURE AND OCCURRENCE OF DISINFECTANT BY-PRODUCTS

129

FIGURE 6.2 The effect of pH on the yield of several representative by-products of the chlorination of a concentrated solution of humic substances. TCA, trichloroacetate; DCA, dichloroacetate; TCM, chloroform, and CH, chloral hydrate (Miller and Uden, 1983).

decompose, in part, to THMs. On the contrary, dichloroacetate (DCA) concentrations do not vary significantly with pH. The influence of bromide on by-product formation is more complicated. First, bromide is oxidized in the presence of at least two disinfectants, chlorine and ozone to HOBr/OBr . HOBr and OBr are in equilibrium that is controlled by pH. Acid pH favors the bromination of organic substrates (Stevens et al., 1989), while alkaline pH favors the formation of bromate is favored when ozone is the disinfectant (Haag and Hoigne, 1983; Krasner et al., 1993). In Table 6.3, the mean concentrations of various members of the THM, HA, and HAN classes of by-products in the EPA/CDHS survey (McGuire et al., 1989) of 35 water supplies are compared to their concentrations in the water supply with the highest bromide concentrations (ca. 3 mg/L). Under such conditions, the brominated by-products predominate. Bromoform was present at 66 times the concentration of chloroform; while the more general pattern is that chloroform is found in about 30-fold excess over bromoform. Very similar shifts in distribution between chlorinated and brominated by-products are observed with the HAs and HANs (Cowman and Singer, 1996). This distribution of chlorine and bromine substitution extends to the remainder of the TOX that is produced. The THMs and HAs are the two classes of DBPs observed at the highest concentrations in chlorinated water (McGuire et al., 2002). The regulated THMs are chloroform, bromodichloromethane (BDCM), dibromochloromethane (DBCM), and bromoform. These are generally what is referred to as THMs, although there are a number of iodinated THMs that are produced in trace amounts, which are rarely measured. To avoid confusion, the regulated THMs will be referred to as THM4. Similarly, there are large number of HAs. The regulated HAs include monochloroacetate (MCA), DCA, bromochloroacetate (BCA), monobromoacetate (MBA), and dibromoacetate (DBA), and are collectively referred to as the HAA5. This ignores the fact that the mixed bromochloro trihaloacetates are present at some level in virtually all chlorinated waters. Tribomoacetate (TBA) is relatively unstable and may or may not be present at the tap. Again, there are traces of iodinated analogs of the HAs, but these are rarely measured. Mean or median concentrations of THM4 range from 20 to 40 mg/L and 15 to 25 mg/L for the HAs (Table 6.2). Under unusual circumstances (high precursor

130

DRINKING WATER DISINFECTION BY-PRODUCTS

TABLE 6.3

Impact of Bromide Concentration of Nature of Chlorination By-Productsa

By-Product Trihalomethanes (THM) Chloroform Bromodichloromethane Dibromochloromethane Bromoform Total THM Haloacetates (HA) Chloroacetate Dichloroacetate Trichloroacetate Bromochloroacetate Dibromochloroacetate Bromoacetate Dibromoacetate Total HA Haloacetonitriles (HAN) Dichloroacetonitrile Trichloroacetonitrile Bromochloroacetonitrile Dibromoacetonitrile Total HN

Mean Concentrations in 35 Utilities

Utility with Highest Br Conc. (2.8–3.0 mg/L)

15b 10 4.5 0.57

0.59 2.9 9.2 40

44

53

1.2 6.8 5.8 NQc NQ 50.5 1.5

51.0 0.9 50.6 NQ NQ 1.2 21

20

21

1.1 50.012 0.58 0.48

0.34 50.012 1.2 5.9

2.5

7.4

a

Table adapted from Krasner et al. (1989). Concentrations are all in mg/L. c NQ ¼ not quantitated. These by-products were detected but no standard was available to quantitate. b

concentrations, high temperatures) as illustrated by experience in South Australia with THMs, these concentrations can be an order of magnitude higher (Baker and Bursill, 1990). Related to the HAs are a group of haloacetaldehydes, which are generally found at concentrations that are about an order of magnitude lower than the HAs. A variety of halogenated propanones and HANs have been found to occur at concentrations in the same range as the halogenated aldehydes. HANs, generally dominated by the dihaloacetonitriles, are also found in this range of concentrations. This is because the more heavily substituted compounds have limited stability (Stevens et al., 1989). Formaldehyde and acetaldehyde are the most commonly measured nonhalogenated aldehydes in chlorinated water, but it is probable that other aldehydes are also present at appreciable concentrations. Glyoxal and methyl glyoxal are cases in point (Glaze and Weinberg, 1993). Because of analytical difficulties, other aldehydes have not been quantified in drinking water, but a number of those which have been detected are included in Table 6.1 as examples. Aldehydes are generally found at higher concentrations in water treated with chlorine dioxide or ozone than the same water treated with chlorine.

CHEMICAL NATURE AND OCCURRENCE OF DISINFECTANT BY-PRODUCTS

131

Halofuranones and some related compounds (i.e., open-ring forms, oxidized and reduced forms) received attention because 2-chloro-4-(dichloromethyl)-5-hydroxy-2(5H)-furanone (MX, i.e., mutagen X) was found to account for a substantial fraction of the mutagenic activity that results from the chlorination of drinking water (Meier et al., 1987; Kronberg and Vartiainen, 1988; Romero et al., 1997). A recent survey of these compounds, including the brominated analogs of MX, has shown these compounds to be present at higher concentrations than previously thought (Weinberg et al., 2002). A number of other halogenated by-products have also been observed, and a selection of these compounds have been included in Table 6.2 to provide a broader perspective for the type of products that are observed. This list is far from complete, and the reader is referred to a review by Richardson (1998) and subsequent publications from this group (Richardson et al., 1999a, 1999b, 2003; Weinberg et al., 2002) for a more complete listing of DBPs that have been identified. There is also a series of nonchlorinated acids whose concentrations in finished drinking water have been measured, but infrequently quantified. This group of compounds contains many nutrients (e.g., fatty acids). At the low concentrations that can be anticipated (i.e., limited by the TOC content of the water) these compounds are not of toxicological significance. Many halogenated products seen with chlorine are also observed with chloramination, although generally at lower concentrations. However, the concentrations of all classes of halogenated by-products are not uniformly reduced. As illustrated in Fig. 6.1, the fraction of the TOX composed of unknown compounds is greater with chloramine, ozone, or chlorine dioxide than with chlorine. The amounts of halogenated by-products formed depends upon how the disinfectant is introduced into the system, for example, if free chlorine is allowed to exist for some minimum contact time before the addition of ammonia versus the simultaneous addition of chlorine and ammonia. Generally, water supplies that exceed the THM4 standard using chlorine find that the least expensive alternative is to switch to chloramination (Bull et al., 1990a). The total THM and HAN concentrations will be significantly decreased with a switch to chloramine (Jacangelo et al., 1989). Total concentrations of HAs are less readily predicted because the dihaloacetates are not reduced to the same extent that the trihaloacetates are (Qi et al., 2004). Cyanogen chloride is actually formed at higher concentrations in chloraminated versus chlorinated water (Krasner et al., 1989). However, the median concentration of cyanogen chloride observed was only 2.2 mg/L in plant effluents, and the compound appears to be relatively unstable in the distribution system. In the presence of secondary amines, chloramine will form nitrosamines at generally higher concentrations than are observed in the same water treated with chlorine. This requires a source of a secondary amine precursor, but the nitroso-group is derived from chloramine (Choi and Valentine, 2002; Mitch and Sedlak, 2002a). In most water supplies the levels of N-nitroso-N-dimethylamine (NDMA) observed are 510 ng/L, but in some chloraminated wastewaters it has been found in concentrations in excess of 1 mg/L. The major by-products associated with the use of chlorine dioxide are chlorite and chlorate. Chlorite is the major by-product arising as the reduction product of chlorine dioxide when it oxidizes organic material in the water. Chlorate is produced by either disproportionation or by photodecomposition of chlorine dioxide (Masschelein, 1989). Somewhere between 40 and 80% of the applied dose of chlorine dioxide is ordinarily converted to chlorite (Lykins and Griese, 1986; Howe et al., 1989; Masschelein, 1989; Gordon et al., 1990). With concentrations of ClO2 used in disinfection ranging as high as 4–6 mg/L (Gordon et al., 1990), several mg/L of chlorite could be expected in tapwater. Chlorate is

132

DRINKING WATER DISINFECTION BY-PRODUCTS

generally not formed from chlorine dioxide unless the water is exposed to UV radiation (Gordon et al., 1990; Masschelein, 1989). Awide variety of aldehydes are formed by ozone (Glaze et al., 1989a, 1989b), but of those identified, formaldehyde and acetaldehyde are generally found at the highest concentrations (Table 6.2). Glyoxal and methylglyoxal are also universally found with ozonation. As with chlorine, ozonating waters containing significant amounts of bromide results in formation of brominated organic by-products. The by-products include bromoform, bromoacetates, bromoacetones and bromoacetonitriles under simulated conditions in the laboratory (Haag and Hoigne, 1983). Bromate is the by-product of ozonation that raises most concern (McGuire et al., 1989). Its formation appears to be facilitated by the addition of hydrogen peroxide and/or alkaline pH (Krasner et al., 1993; Von Gunten et al., 1996; Song et al., 1997). Krasner et al. (1993) found concentrations ranging from 55 (limit of detection) to 58 m/L of bromate under conditions that reasonably approximated actual disinfection conditions. The concept that more than one disinfectant can be used in water treatment to negate some of the disadvantages associated individual disinfectants has been utilized in recent years. For example, water utilities that utilize ozone as a primary disinfectant usually use chlorine or chloramine to maintain microbiological water quality in the distribution system. In Europe, chlorine dioxide and hypochlorite are frequently employed at different points in the same water system (Aggazzotti et al., 2004). The use of mixed disinfectants introduces a degree of complexity that is not easily captured by the measurement of traditional DBPs (e.g., THMs, HAs, selected oxyhalide anions). For this reason assessing the health risks that might arise from switching from one disinfection practice to another is difficult, if not impossible without a much more complete description of DBPs that are produced and knowledge of their toxicological properties. 6.4 ASSOCIATIONS OF HUMAN DISEASE WITH DRINKING WATER DISINFECTION 6.4.1

Cancer

The relationships between human cancer and the chlorination of drinking water have received extensive epidemiological study. Many of the early studies were ecological and were meant to be preliminary studies. They did not take into account many potentially confounding factors (Craun, 1988). An association between the use of chlorine and cancers of the rectum (Gottlieb et al., 1982), colon (De Rouen and Diem, 1977; Kuzma et al., 1977; Young et al., 1981), bladder (De Rouen and Diem, 1977; Kuzma et al., 1977; Cantor et al., 1978), and lung (Bean et al., 1982) were found with some consistency across geographical regions and study populations. Because of design difficulties, these initial studies were dismissed as being inconclusive (NRC, 1980). Case control and cohort studies conducted in the past two decades have tended to support the relationship between the consumption of chlorinated drinking water and cancers at selected sites. These more recent studies are summarized in Table 6.4. An early case control study (Alavanja et al., 1978) identified several tumor sites, the colon, bladder, liver and kidney, esophagus, pancreas, stomach and brain. The most consistent association has been with bladder cancer. Studies of intestinal cancers have been less consistent, with some studies identifying cancers of the colon and others cancers of the rectum and anus.

133

3028 GI and urinary tract cancer deaths 10,205 kidney, bladder, stomach, liver, and colorectal cancer deaths Diagnosis of cancer over 12-year period

395 colorectal cancer deaths

200 new cases of colon cancer

51,645 kidney, bladder, stomach, pancreas, colon, lung, breast cancer deaths 347 new cases of colon cancer 2982 new cases of bladder cancer

Brenniman et al., 1980

Wilkins and Comstock, 1981

Lawrence et al., 1984

Cragle et al., 1985

Zierler, et al., 1986, 1988

Ijesselmuiden et al., 1992

Young et al., 1987 Cantor et al., 1987

101pancreatic cancer cases

8029 cancer deaths

Young et al., 1981

Gottlieb et al., 1982

3446 GI and urinary tract cancer deaths

Population (Cases)

Chlorinated versus unchlorinated water at residence Chlorinated versus unchlorinated water Chlorinated versus unchlorinated water at residence Chlorinated water in Hagerstown versus deep wells in County Modeled chloroform exposure in chlorinated surface versus ground water Years of exposure to chlorinated water Chlorinated versus chloraminated drinking water from common source Water consumption, chloroform Residence, beverage consumption, and water samples Chlorinated versus unchlorinated water

Chlorinated versus unchlorinated water at residence

Exposure Variables

Case–Control and Cohort Studies of Cancer and Chlorinated Drinking Water

Alavanja et al., 1978

References

TABLE 6.4

Odds ratio

Logistic regression Logistic regression

Odds ratios with Mantel– Haenszel adjustment

(continued)

2.23 (1.24–4.10)

Colon, 1.57 (1.04–2.37) Bladder, 1.43 (1.23–1.67)

No statistically significant association with chloroform Colorectal, 1.4 (1.1–1.7) to 3.4 (2.4–4.6) ages 6–89 Bladder, 1.7 (1.3–2.2)

Logistic regression

Logistic regression

No statistically significant associations

Logistic regression

Odds ratios with Mantel– Haenzel adjustment Stratified odds ratio with c2 test

Rectal, 1.68; breast, 1.58

Colon, 1.99 men; bladder, 2.02 men; liver and kidney, 2.76; esophagus, 2.39; pancreas, 2.23; stomach, 2.39 and 2.23 Association of colon cancer with average daily chlorine dosage Rectal, 1.35 women

Stratified odds ratio with c2 test

Logistic regression

Results

Analysis

134

696 bladder cancer cases

375 brain cancer patients

Prospective cohort study 41,836 postmenopausal women in Iowa

1123 cases and 1983 controls

Cantor et al., 1996

Doyle et al., 1997

Cantor et al., 1998

Vena et al., 1993

King and Marrett, 1996

Population (Cases)

327 verified bladder cancer cases 351 bladder cancer cases

(Continued)

McGeehin et al., 1993

References

TABLE 6.4

Duration of chlorinated drinking water usage; lifetime THM intake

Chlorinated versus unchlorinated water Duration of exposure to chlorinated water Chloroform in drinking water

Years of exposure to chlorinated drinking water Tap water consumption— typically chlorinated

Exposure Variables

0 years 1–19 years 20–39 years 40–59 years 560 years

Odds ratio

Age-adjusted risk ratios; 1– 2 mg/L, 3–13 mg/L, 14– 287 mg/L

Odds ratio

Odds ratio

Odds ratio in quartiles of water intake

Odds ratio

Analysis

1.0 1.1 (0.8–1.3) 1.3 (0.9–1.8) 1.5 (0.95–2.3) 1.9 (1.1–3.6) enhanced by smoking

Bladder negative Colon cancer: 1.09 (0.70–1.69) 1.41 (0.91–2.19) 1.72 (1.14–2.59) Lung cancer: 1.38 (0.82–2.25) 1.96 (1.20–3.21) 1.85 (1.13–3.01) Bladder cancer;

Brain cancer, 1.8, 20–39 years; 2.4, 440 years Rectal negative

1.8 (1.1–2.9) for more than 30 years exposure 2.62 (1.53–4.47) in individuals 565 that drank 10–39 cups of water per day 1.41 (1.09–1.81)

Results

135

732 bladder cancer cases; 703 kidney; 914 controls

291 brain cancer cases; 1983 controls

Cantor et al., 1999

655 rectal; 2434 controls

685 colon cancer cases;

Koivusalo et al., 1998

Hildesheim et al., 1998

Years of chlorinated drinking water usage and lifetime average THM concentration

Net TA100 revertants 5 3000/L 1950–1987

Duration of chlorinated drinking water usage; lifetime THM intake

1.0 1.3 (0.8–2.1) 1.7 (0.9–3.3) 2.5 (1.2–5.0) Increased with tap water consumption; poorly related to THM dose Brain (female); NS

0 year 1–19 years 20–39 years 540 years

(continued)

Brain (male);

Kidney (men); 1.49 (1.05– 2.13) Bladder (both sexes); 1.22 (0.92–1.62) Bladder (male nonsmokers); 2.59 (1.13–5.94)

1.0 1.08 (0.8–1.4) 1.55 (1.1–2–2) 1.63 (1.0–2.6) 2.61 (1.5–5.0) Also related to THM intake; colon negative

Rectal cancer;

Odds ratio

Odds ratio

0 year 1–19 years 20–39 years 40–59 years 560 years

Odds ratio

136

767 colon cancer cases; 661 rectal; 1545 controls

1068 incident leukemia cases, 5039 controls in Canada

476 incident cases of pancreatic cancer, 3596 controls

King et al., 2000a

Kasim et al., 2005

Do et al., 2005

Population (Cases)

Cohort study, micronuclei in exfoliated urinary bladder cells 615 cases in Australia

(Continued)

Ranmuthugala et al., 2003

References

TABLE 6.4

Odds ratios 450 mg/L 45 mg/L 430 mg/L 3 years

THM4 BDCM Chloroform Latency weighted by water consumption

420 mg/L 440 mg/L 45 mg/L

THM4 BDCM

0.91 (0.67–1.25) 1.08 (0.78–1.50) All odds ratios NS

1.70 (1.00–3.03) 1.72 (1.01–3.08) 1.63 (1.00–3.10) All other forms of leukemia NS 0.86 (0.58–1.78)

CM leukemia

1.00 1.70 (1.07–2.68) 1.33 (0.96–1.86) 1.53 (1.13–2.09) cumulative THM exposure similar results; rectal NS; Females NS

0–9 years 10–19 years 20–34 years 535 years

Odds ratios

Colon (males);

1.00 (0.97–1.03) 1.04 (0.84–1.29) 1.00 (0.98–1.02) Smokers also NS

Nonsmokers

Results

Odds ratio

Relative risk per 10 mg/kg/ day

Analysis

THM4

Duration of exposure to chlorinated water, cumulative THM exposure

Bromoform THM4

Chloroform

Exposure Variables

ASSOCIATIONS OF HUMAN DISEASE WITH DRINKING WATER DISINFECTION

137

Nine, more or less independent, analytical studies have found significant associations between bladder cancer and chlorinated drinking water consumption in males (Alavanja et al., 1978; Zierler et al., 1986, 1988; Cantor et al., 1987; McGeehin et al., 1993; Vena et al., 1993; King and Marrett, 1996; Cantor et al., 1998; Koivusalo et al., 1998). In general chlorinated drinking water has not been strongly associated with bladder cancer in women including a recent large prospective cohort study (Doyle et al., 1997). Several of these studies have identified dose–response relationships, primarily related to the duration of exposure to chlorinated water and/or the volume of water consumed (Cantor et al., 1987; Vena et al., 1993; McGeehin et al., 1993; King and Marrett, 1996; Cantor et al., 1998). A different approach to dosimetry was utilized in the study of Koivusalo et al. (1998). The authors associated mutagenic activity of water as measured in Salmonella strain TA100 with increased odds ratios for kidney cancer in males and bladder cancer in nonsmokers. It is important to recognize that the odds ratios reported with THM4, mutagenic activity, or other specific parameters do not show these parameters to be causally involved in producing bladder or other cancers. Essentially, there is no evidence that the odds ratios seen with these parameters are any higher than those associated with chlorinated versus nonchlorinated water. As the chemical composition of chlorinated water varies by location, the THM4 provides a very limited insight into the large number of potentially carcinogenic by-products that could be present. Three groups of investigators have found evidence that the introduction of other forms of drinking water disinfection reduces bladder cancer risk. Specifically, a comparison was made of cases of bladder cancer that were drawn from communities that used of chloramine versus chlorine to disinfect water drawn from the same source. The relationship with chlorinated drinking water with bladder cancer was observed, but there was no significant increase in bladder cancer in the Massachusetts communities that employed chloramines (Zierler et al., 1988). McGeehin et al. (1993) also reported a lower risk for bladder cancer in communities utilizing chloramines in Colorado. Although the common DBPs are also produced with chloramine, adding ammonia does suppress the reactions of chlorine with NOM (see Fig. 6.1). More recently, Chevrier et al. (2004) recently found that the risk from bladder cancer decreased with duration of exposure to ozonated drinking water of 10 years or more. Of particular interest was the finding that a reduction of bladder cancer risk was observed with ozonation in systems that continued to use chlorine for secondary disinfection. This implies that prior ozonation destroyed the precursors of the putative bladder carcinogen produced with chlorine alone. Colon and colorectal cancer have been associated with chlorinated drinking water. An estimated odds ratio of 1.99 for chlorinated water relative to nonchlorinated water was reported by Alavanja et al. (1978) in male subjects. An association was also reported by Young et al. (1981) for colon cancer. On the contrary, Brenniman et al. (1980) reported an odds ratio of 1.35 for rectal cancer in women. The inconsistencies between the colon and rectum among studies continue; Gottlieb et al. (1982) identified the rectum and the study of Cragle et al. (1985) implicated the colorectal area. Young et al. (1987) and, more recently King et al. (2000a, 2000b), both found associations with colon cancer but nonsignificant relationships with the cancers of the rectum. Finally, a recent study by Hildesheim et al. (1998) indicates little risk for colon cancer and significant risks for rectal cancer. The odds ratios of chlorinated to nonchlorinated water in these studies ranged 1.35–2.61. In addition, several earlier studies failed to identify an association at all (Lawrence et al., 1984; Zierler et al., 1986, 1988) or found very inconsistent relationships (Bean et al., 1982).

138

DRINKING WATER DISINFECTION BY-PRODUCTS

In what appears to be the first attempt at a biomarker study of DBP effects, the potential relationship between micronuclei in exfoliated urinary bladder epithelial cells and single THMs and THM4 were studied (Ranmuthugala et al., 2003). No association was observed between micronuclei exposure to THM4, chloroform, or BDCM. Additional target organs have been the subject of case–control studies. Ijesselmuiden et al. (1992) found a significant association with pancreatic cancer in Maryland (OR 2.23). A more recent study by Do et al. (2005) also conducted in Maryland, found no association between pancreatic cancer and THM4. Brain cancer was found to be associated with the duration of exposure to chlorinated water (Cantor et al., 1996). In this case, Cantor et al. (1999) confirmed the earlier association and found that the odds ratio with brain cancer increased with increasing consumption of chlorinated tap water, implying a dose–response relationship. Both the pancreas and the brain were identified as cancer sites in the early study of Alavanja et al. (1978). There have been a limited number of ecological studies of associations of individual cancers with chlorinated drinking water in recent years. Marcus et al. (1998) found no association between female breast cancer and THM4 levels in drinking water. A broad investigation of cancers associated with different measures of drinking water quality was conducted by Foster et al. (1997). THM4 were found to correlate with myelodysplasias, but no other leukemia-related disorders. A recent case–control study did find an association between THM4 and chronic myelogenous leukemias, but there appeared to be protective effect for other forms of leukemia (Kasim et al., 2005). There have been several attempts to estimate the population attributable risk (PAR) of cancer to chlorinated drinking water (Morris et al., 1992; Poole, 1997; King, 1999; Villanueva et al., 2003, 2004). These analyses have focused on bladder cancer and colon and rectal cancer. In the development of the Stage II Disinfection By-Products Rule, the USEPA (2003a, 2003b) utilized the first three of these studies to estimate the PAR of chlorinated drinking water to bladder cancer as ranging between 2 and 17% of the incidence observed in the United States (Table 6.5). To place this conclusion in perspective, these data can be translated into lifetime risks of 7/10,000–6/1,000 to a lifetime risk in men and about one-third this risk in women. The most recent analysis of the bladder cancer risk associated with chlorinated drinking water pooled six independent studies from around the world (Villanueva et al., 2004). An odds ration of 1.24 (CI 1.09–1.41) for bladder cancer in males with THM4 4 1 mg/L resulted

TABLE 6.5

Risks of Bladder Cancer Attributed to Chlorination of Drinking Watera Males

Bladder cancer incidence Lifetime probability for developing bladder cancer

39/100,000 0.0356c

Population attributable risksd 2% 17%

0.0007 0.006

a

Females b

10.1/100,000 0.0113c 0.0002 0.002

Reproduced with permission from Bull et al. (2006). Age-adjusted incidence for years 1997–2001, Cancer Facts, 2005; American Cancer Society (http://www.cancer. org). c Years 1999–2001, Cancer Facts, 2005; American Cancer Society (http://www.cancer.org). d USEPA (2003a). b

ASSOCIATIONS OF HUMAN DISEASE WITH DRINKING WATER DISINFECTION

139

from the analysis. No association was found for females (OR ¼ 0.95, CI 0.76–1.20). The authors pointed out that the THM4 concentration is a surrogate measure for chlorination byproducts and that this analysis does not imply a causal relationship. Nevertheless, the consistency of the relationship makes it difficult to dismiss chlorinated water as a cancer risk, albeit a rather small one. 6.4.2

Reproductive and Developmental Effects

There have been a number of epidemiological investigations that have associated various parameters related to drinking water disinfection with adverse reproductive or developmental outcomes in humans. Much of this work has focused upon organic DBPs in chlorinated drinking water, but several have evaluated inorganic by-products that are related to the use of chlorine dioxide as well as with chlorine. Studies that have been conducted focusing on by-products of disinfection are summarized in Table 6.6. In most cases, negative findings have been omitted from the table for the sake of simplicity, except when the findings are inconsistent with findings of other studies. To form firm conclusions from epidemiological data on reproductive parameters, there a number of criteria that should be applied. One is the strength of the association. In general the odds ratios or risk ratios reported with adverse reproductive outcomes are small, generally below 1.5. Second, a certain level of consistency needs to be observed among studies. The outcomes examined have focused primarily on low birth weight (LBW), small for gestational age (SGA), short body length, head circumference, stillbirths, miscarriage, or spontaneous abortions. While there are undoubtedly some correlations among these parameters, they have been used as distinct metrics. Thus, there is a high probability that classification error might contribute to some associations observed in one study, but not another. In the case of chlorine and its by-products, studies have found associations with one or more measures of reduced in utero growth, such as LBW, intrauterine growth retardation (IUGR), SGA, body length, head circumference, and preterm delivery (Kramer et al., 1992; Bove et al., 1995; Kanitz et al., 1996; Kallen and Robert, 2000; Yang, 2004; Porter et al., 2005; Hinckley et al., 2005; Lewis et al., 2006; Wright et al., 2004). There are inconsistencies among the significant associations identified within this group of studies. Studies of comparable quality have found no significant association for the same outcomes (Savitz et al., 1995; Gallagher et al., 1998; Jaakkola et al., 2001; Yang, 2004; Aggazzotti et al., 2004; Toledano et al., 2005). Four recent studies examined the associations between decreased IUGR or LBW and exposures during different stages of pregnancy (Porter et al., 2005; Hinckley et al., 2005; Savitz et al., 2005). THM4 or HAA5 associations with these outcomes were not significant when examined over the entire pregnancy. Some associations were observed with exposure levels during the third trimester. However, the details of these associations varied among the three studies. Porter et al. (2005) found a significant association of lower IUGR and exposure to the regulated HAA5 in the third trimester, but found no association with any specific HA. In contrast Hinckley et al. (2005) found that DCA had the strongest relationship with lowered IUGR, but there was no relationship with HAA5. The latter study associated low term birth weight (TBW) with DBA exposure during weeks 33–36 of gestational age. Wright et al. (2004) found significant associations of SGA with THM4, chloroform, and BDCM. In the case of BDCM, however, the effect disappeared at the higher concentrations (415 mg/L). Wright et al. (2004) found no relationship to the HAA5, TCA or DCA, but these waters did not contain measurable amounts of DBA. Wright et al. (2004) was the only group to explore

140 Low birth weight, term low birth weight, preterm delivery

Gallagher et al., 1998

1893 births, Denver, CO

Spontaneous abortion

Small for gestational age, all birth defects Miscarriage, low birth weight 548 births in Genoa, IT; Small body length, small 128 births in Chiavari, IT cranial circumference, neonatal jaundice

1039 congenital anomaly cases; 77 stillbirth; 55 neonatal deaths; 1177 controls 59,141 births in New Jersey 126/122

Prematurity, postnatal weight loss Low birth weight, prematurity, intrauterine growth retardation Stillbirth

Outcomes

Swan et al., 1998; 5144 pregnant women in Waller et al., 1998 prepaid health plan

Kanitz et al., 1996

Savitz et al., 1995

Bove et al., 1995

Aschengrau et al., 1993

903 births ClO2; 1112 births OCl Kramer et al., 1992 159/795 342/1710 187/035

Population (Cases/Controls)

THMs 5 75 mg/L and 55 glasses/day; BDCM 5 20 mg/L TTHM 5 50 mg/L

No disinfection versus ClO2, OCl , or combined disinfection

TTHM 4 100 mg/L, TTHM 4 80 mg/L THMs 40–168

Chlorinated versus nonchlorinated drinking water

ClO2-disinfected water versus OCl -disinfected communties Chloroform, 410 mg/L, odds Ratio

Exposure Variables; Analysis

3.0 (1.4–6.6) 1.5 (0.8–3.0), 2.6 (1.1–6.1), 0.9 (0.4–2.0)

1.3 (1.0–1.6)

ClO2, 2.0 (1.2–3.3); OCl , 2.3 (1.3–4.2); both, 1.4 (0.8–2.5) ClO2, 2.2 (1.4–3.9); OCl , 3.5 (2.1–8.5); both, 2.4 (1.6–5.3) ClO2, 1.7 (1.1–3.1)

1.2 (0.6–2.4), 1.3 (0.8–2.1)

1.5 (1.04–2.09), 2.59 (1.04–5.55)

2.6 (0.9–7.5)

1.3 (0.8–2.2), 1.1 (0.7–1.6), 1.8 (1.1–2.9)

ClO2 4 OCl , ClO2 4 OCl

Findings Odds Ratio or Risk Ratio ¼ (RR)

Epidemiological Studies of Relationships Between Drinking Water Disinfection and Adverse Reproductive Outcomes

Tuthill et al. 1982

References

TABLE 6.6

141

58,699 women in Sweden with 753 cases of congenital cardiac defects; 80 water supplies, control ground water 182,796 women in Preterm delivery, low birth 128 municipalities weight 1194 mothers who Small for gestational age supplied water for analysis 1,000,000 births in 3 water Stillbirth, low birth districts in England weight, very low birth weight

Cedergren et al., 2002

Toledano et al., 2005

Aggazzotti et al., 2004

Yang, 2004

50,000 births in Nova Scotia 49,842 births in Nova Scotia 137,145 births in Norway

King et al., 2000a, 2000b Dodds and King, 2001 Jaakkola et al., 2001 Low birth weight, small for gestational age, duration of gestation Cardiac defects

Neural tube defects

Stillbirths

Body length 78,237 infants no disinfection; 24,731 OCl ; 15,429 ClO2 Body mass index Head circumference

0.97 (0.89–1.06), 1.00 (0.91–1.10), 0.91 (0.84–0.99)

High color and chlorination versus no chlorination

TTHM 5 60 mg/L

Chlorinated versus nonchlorinated drinking water supply THMs 5 30 mg/L, ClO2 5 200 mg/L, ClO3 5 200 mg/L

(continued)

1.11 (1.0–1.23), 1.09 (0.93–1.27), 1.05 (0.82–1.34)

NS, 1.70a (0.97–3/00), NS

1.37 (1.20–1.56), NS

Groundwater, OCl , ClO2 þ OCl , 1.31 (1.09–1.57), 0.85 (0.60–1.21), THM 4 10 mg/L 1.61 (1.00–2.59), c2 ¼ 7.87, P ¼ 0.0005

1.66 (1.09–2.54), 1.56 (1.04–2.34), 1.98 (1.23–3.49) (RR) 2.5 (1.2–5.1)

OCl , 1.27 (1.19–1.37); ClO2, NS OCl , 1.46 (1.07–1.98); ClO2, NS

OCl , 1.97 (1.30–1.37); ClO2, NS

1.66 (1.09–2.53)

1.9 (1.0–4.0)

TTHM 5 100 mg/L, chloroform 5 100 mg/L, BDCM 5 20 mg/L BDCM 5 20 mg/L

OCl or ClO2 versus no disinfection

TTHM 5 100 mg/L

Stillbirths

Kallen and Robert, 2000

TTHMs 5 40 mg/L

Neural tube defects

All NJ births, 248 random controls 50,000 births in Nova Scotia

Klotz and Pyrch, 1999 Dodds et al., 1999

142

112 still birth and 398 live births in Eastern Ontario 2413 pregnancies in North Carolina

King et al., 2005

39,954 live births and fetal deaths in Arizona

27 MA communities served by same utility

Hinckley et al., 2005

Lewis et al., 2006

Coupled with higher inhalation exposure in shower.

a

15,315 births in a Maryland County

Porter et al., 2005

Savitz et al., 2005

194,827 singleton births in MA

Population (Cases/Controls)

Wright et al., 2004

References

TABLE 6.6 (Continued)

Low birth weight, versus weekly THM measurements

IUGR, term low birth weight

Pregnancy loss across gestation (PL), preterm birth (PB), small for gestational age, term birth weight Intrauterine growth rate (IUGR)

Stillbirth

SGA, Gestational age

Outcomes

NS

1.06 (1.02–1.10), 1.05 (1.02–1.09), 1.1 (1.07–1.14), NS, NS, NS, 1.14 (0.95–1.37), 1.25 (1.04–1.51), Older at birth

Findings Odds Ratio or Risk Ratio ¼ (RR)

THM4, chloroform, BDCM, DBCM, THM-Br, HAA9, HAA5, HAA-Br, TOX

SGA 2.1 (1.1–3.8), NS all variables, PL 1.6 (1.0–2.4), NS, PL-late 2.22 (1.02–4.87), NS all variables, NS all variables, NS all variables, NS all variables, PL 1.66 (1–2.77) 0.98 (0.81–1.19), 1.04 (0.85–1.27), Fifth quintile OR entire 0.97 (0.80–1.18), 0.96 (0.79–1.17), pregnancy, THM4, chloroform, 1.16 (0.94–1.41), BDCM, DBCM, bromoform, 0.94 (0.73–1.20), 0.94 (0.74–1.20), HAA5, MCA, DCA, TCA, 0.96 (0.75–1.24), MBA, DBA, third trimester, fifth quintile, HAA5, DCA, others 0.97 (0.76–1.24), 0.92 (0.72–1.18), 0.84 (0.66–1.07), 1.34 (1.04–1.71), 1.27 (0.99–1.61), NS IUGR, NS, NS, 1.28 (1.08–1.51), 3-week intervals, THM4, HAA5, 1.49 (1.10–2.02) DCA, weeks 33–36, DBA 4 5, TCA 4 6 TTHM 5 70 mg/L during third 1.5 (1.07–2.10) trimester

Third trimester, THM4 4 30 mg/L, chloroform 4 26, BDCM 4 5, HAA5, TCA, DCA, MX 4 46 ng/L, mutagenicity 4 2250 Haloacetic acids

Exposure Variables; Analysis

ASSOCIATIONS OF HUMAN DISEASE WITH DRINKING WATER DISINFECTION

143

associations of SGA with mutagenic activity in chlorinated drinking water. It was of interest that this association had the highest odds ratio of the exposure variables examined, including the concentrations of the strong mutagen, MX. The THMs and HAs are weakly mutagenic at best and, thus, contribute virtually nothing to the mutagenic activity of drinking water. In the most detailed analysis of exposure to DBPs and adverse reproductive outcomes and to different DBP concentrations to date, Savitz et al. (2005) found an association between THM4 and SGA, pregnancy loss with BDCM and TOX, and late pregnancy loss with the brominated THMs. Considering the many statistical analysis among variables in this study, the authors concluded that the net results of the study indicated a weakening of the relationships previously reported between measures of DBP exposure and adverse reproductive outcomes. These studies hint at the complex approach that will be necessary to clarify relationships these reproductive outcomes and DBP exposure using epidemiological methods. Collectively, the data available do not support the conclusion that reproductive and developmental effects are associated with specific windows of exposure to chlorinated water or with specific DBPs. These same parameters have been variously associated with the use of chlorine dioxide or in the presence of chlorite, which is produced as a reduction product of chlorine dioxide (Kanitz et al., 1996; Aggazzotti et al., 2004). Again these findings are apparently contradicted in another study (Kallen and Robert, 2000). There have been efforts to associate more severe reproductive outcomes with chlorinated drinking water, including stillbirth, miscarriages, or spontaneous abortion in Massachusetts (Aschengrau et al., 1993) in California (Swan et al., 1998; Waller et al., 1998), in Nova Scotia (Dodds et al., 1999; King et al., 2000a, 2000b; Dodds and King, 2001) and England (Toledano et al., 2005). Savitz et al. (1995) did not find a significant association in North Carolina. Two of the positive studies (Waller et al., 1998; King et al., 2000b) found the strongest relationship to be with BDCM concentrations. A small number of studies have associated birth defects with chlorinated drinking water (Bove et al., 1995; Klotz and Pyrch, 1999; Dodds and King, 2001). The latter two studies focused specifically on neural tube defects; the Dodds and King (2001) study again found a stronger relationship with BDCM concentrations (420 mg/L). A complex result was obtained by Cedergren et al. (2002) in Sweden, where a significant association between cardiac defects and the combined use of chlorine dioxide and hypochlorite was identified, but there was no association of cardiac defects with use of hypochlorite alone. Despite the lack of association with hypochlorite use, there was apparently an association with THM4 concentrations above 10 mg/L. On the surface, these associations appear to be contradictory, in that THM4 concentrations would be expected to be higher with hypochlorite alone. Further, the THM4 reported in this study are quite low (highest observed was 41 mg/L) relative to those found in other studies raising further questions about the adequacy of the THM4 as a surrogate for the responsible DBPs. Recently, studies have explored potential associations of chlorinated drinking water with menstrual cycle perturbations (Windham et al., 2003). The length of the follicular phase of the menstrual cycle was shorter in communities where drinking water had concentrations of THM4 4 40 mg/L. Menstrual cycle phases were identified by daily measurement of hormone concentrations in the urine. Because of the intensive participation required, only 41% of the initially eligible women completed the study. BDCM and DBCM displayed the strongest association among the THM4. The high THM4 group was found to also smoke and consume alcohol. It is not clear that the authors adjusted for these factors.

144

DRINKING WATER DISINFECTION BY-PRODUCTS

Exposure assessment issues with reproductive outcomes are even more demanding than those for cancer because effects are likely to be mediated by specific exposures in time and these exposures are poorly reflected in the quarterly sampling data usually required by statute. Some of the studies that have been reported involve a variety of exposures unrelated to disinfection. In these cases, correlations were observed with THMs, but the number of comparisons that were pursued lessens the confidence that can be held in individual outcomes. Fenster et al. (2003) attempted to relate parameters of semen quality in 157 men with THM4 and individual THM concentrations. No statistically significant associations were found. It is also important to recognize that it is sometimes very difficult to tease out the important exposure variables as disinfection practice varies significantly around the world and combined treatments are not uncommon, especially in Europe. In future work, some issues related to exposure assessment may be mitigated by pursuing treatment-related effects on endocrine, paracrine, or autocrine factors that allow parallels with animal studies to be more specifically drawn. These physiological systems are known to control modifications in their function without producing overt signs of toxicity in the mother. It is very difficult to argue that a chemical that affects reproduction or development only at doses that are toxic to the mother is truly a reproductive toxicant (e.g., individual THMs). In the population, signs of the maternal toxicity should be more easily identified than adverse reproductive outcomes. As will be pointed out below, some DBPs have been shown to modify hormonal systems and have the potential to be reproductive and developmental toxicants at less than maternally toxic doses. However, the effective doses of these chemicals as endocrine system modifiers tend to be much higher than would be expected from drinking water.

6.5 GENERAL TOXICOLOGICAL PROPERTIES OF DISINFECTANTS 6.5.1

Chlorine/Hypochlorite

Prior to 1980, little information existed concerning the toxicology of chlorine as it is consumed in water. In the past two decades a series of studies in experimental animals have been published in which the general toxicology, carcinogenicity and reproductive effects of chlorine/hypochlorite have investigated by the oral route of exposure. There has been some variation in the protocols utilized, making comparison of results between studies difficult. In some cases chlorine gas has been bubbled into water until the desired concentration has been achieved. In others sodium hypochlorite has been utilized. At equivalent pH, these methods of preparation yield identical chemical forms, but if the dosing solution is prepared from hypochlorite solutions, there are significant amounts of chlorate as a result of the strongly alkaline pH that are introduced into the water that can approach 1 mg/L (Bolyard et al., 1993). Animals tolerate concentrations of chlorine in their drinking water up to 625 mg/L with little indication of adverse effect beyond those that are secondary to reduced water consumption (Druckrey 1968, using Cl2 bubbled into water; Furukawa et al., 1980; Hasegawa et al., 1986; Kurokawa et al., 1986b all using sodium hypochlorite). Daniel et al. (1990) compared chlorine (pH 9.4) with chloramine and chlorine dioxide in rats. A second study compared chlorine and monochloramine in B6C3F1 mice (Daniel et al., 1991). A series of clinical and hematological parameters were examined in addition to observations of gross pathology and effects on the weights of organs. Of these three disinfectants, chlorine was clearly the best tolerated and the highest dose examined (250 mg/L in rats and 200 mg/L in mice) produced no adverse effects.

GENERAL TOXICOLOGICAL PROPERTIES OF DISINFECTANT BY-PRODUCTS

145

Chlorine may have subtle effects on immune function. Administration of sodium hypochlorite in drinking water at levels of 25–30 mg/L was shown to reduce recovery of peritoneal exudate cells after 1–4 weeks of treatment (Fidler, 1977; Fidler et al., 1982). The treatment led to an increase in the number of pulmonary metastases of injected B16-BL6 cells. In a subsequent study, Exon et al. (1987) did not observe a decrement in the in vitro phagocytic activity of peritoneal macrophages recovered by lavage. However, Exon et al. (1987) did observe impaired TPA-induced oxidative metabolism and an increase in the production of prostaglandin E2 in cells from animals treated with sodium hypochlorite. Whether these subtle effects would result in an increased disease burden in the population is uncertain. 6.5.2

Chloramine

As indicated above, the toxicology of chloramine was investigated by Daniel et al. (1990, 1991). The effects of chloramine were nonspecific, with a lowest-observed-adverseeffect-level (LOAEL) identified in rats of 200 mg/L and no-observed-adverse-effect-level (NOAEL) of 100 mg/L and 100 and 50 mg/L in mice, respectively. Poon et al. (1997) found that the minor biochemical, hematological, immunological, and histopathological changes observed in rats with 200 mg chloramine/L in their drinking water were related to reduced water intake and food consumption, consistent with the interpretation of Daniel et al. (1990). While chloramine, itself, is reasonably stable in the mouth, it is rapidly degraded in stomach fluid (Kotiaho et al., 1992). Thus, if chloramine has systemic effects, they would almost certainly involve reaction products formed in the stomach with secreted materials and/or food stuffs. However, the Daniel et al. (1990, 1991) and Poon et al. (1997) studies suggest that there is little reason for concern. 6.5.3

Chlorine Dioxide

Studies on the general toxicology of chlorine dioxide have been conducted much less frequently than studies that have concentrated one specific effect. In this respect, the comparison of the subchronic toxicities of chlorine, chloramine, and chlorine dioxide conducted by Daniel et al. (1990) provides an important benchmark. This study found goblet cell hyperplasia in the nasal turbinates of rats at 100 and 200 mg/L, and inflammation in the nasal cavity in males at 25 mg/L. Thus, vapors inhaled while consuming water are irritating to the nasal mucosa. Other effects attributed to chlorine dioxide appear to be related to decreased water and food consumption. In addition, studies with more limited objectives were conducted in primates, and the only effects observed were depressed serum T4 levels (Bercz et al., 1982).

6.6 GENERAL TOXICOLOGICAL PROPERTIES OF DISINFECTANT BY-PRODUCTS 6.6.1

Inorganic By-Products

6.6.1.1 Modes of Action of Oxyhalide Anions The use of chlorine, chlorine dioxide, chloramine, or ozone as drinking water disinfectants inevitably introduces a variety of oxyhalide anions into the water. In the broadest sense, oxyhalides are of interest toxicologically for two reasons, for their ability to interact with various anion transporters in the

146

DRINKING WATER DISINFECTION BY-PRODUCTS

body (Wolff, 1964; Wright and Diamond, 1977) and as oxidants. Oxyhalides may be disinfectants themselves (e.g., hypochlorite), disinfectant precursors (e.g., chlorite), reduction products (e.g., chlorite), oxidation products (e.g., bromate) or contaminants of the disinfectant (e.g., chlorate, bromate). In the latter case, bromate contaminates hypochlorite solutions prepared from seawater, whereas significant amounts of chlorate are formed from hypochlorite, itself (Bolyard et al., 1993). In many waters, these anions are the predominant by-products of disinfection. Chlorite occurrence appears to be specific to the use of chlorine dioxide. Bromate occurs at the highest concentration when ozone is used to disinfect water that contains high concentrations of bromide. However, if chlorine dioxide treatment is used simultaneously with ozone to reduce bromate formation, very high levels of chlorate (41 mg/L) can be produced (Masschelein, 1989). The affinity of oxyhalide anions for the sodium iodide symporter (NIS) has not been established for all members of the group. A sequence of the ions TcO4 4 ClO4 4 ReO4 4 BF4 4 SeCN  SO3 4 SCN 4 I 4 NO3 4 OCN 4 NO2 4 Br 4 Cl 4 F relates to their relative affinity for transport by sheep thyroid slices which, in turn, relates to their partial molal ionic volumes (between 25 and 46 cm3/mol)(Wolff, 1964). The locations of other oxyhalide ions of concern in drinking water (BrO3 , ClO3 , IO4 ) are predicted to have a higher affinity than all of the monoatomic halide ions based upon thermodynamic properties (Wright and Diamond, 1977). Experimentally they appear to be weaker than I and perhaps even Br . The main reason for the uncertainty is that these ions are less stable in biological systems because they are rapidly degraded (Wolff, 1964). No predictions are available for ClO2 , but it is also reactive in biological systems. ClO2 would not be expected to significantly impact iodide transport because its ionic volume would be much less than chlorate, bromate, or iodate. The toxicological effects of oxyhalide anions that are unrelated to the thyroid appear to arise from their oxidant properties. However, there are several distinct toxicological effects displayed by the different oxyhalide anions, so their effects are not a simple expression of their redox potentials. In some cases, this may be attributed to participation of some of the anions in competing reactions. For example, some of the oxyhalides (e.g., HOCl/OCl ; HOBr/OBr ; ClN ) participate in halogenation reactions, which may be viewed as competitive with oxidation reactions. On the contrary, the specific nature of the effects of other oxyhalides appears dependent upon how selective their interactions are with biological antioxidants, particularly thiols. The consequences that arise from the interaction of oxyhalides with thiols are different from and appear to set the conditions that affect where each oxyhalide anion produces its effects. Ingram et al. (2003) found that the stoichiometries and kinetics displayed by HOCl/ OCl and ClO2 are quite different in cellular systems. ClO2 reacts rapidly with sulfhydryls with a fixed stoichiometry of 1:4, with little evidence of lipid peroxidation and no further degradation of glutathione. HOCl/OCl was found to react less rapidly, but more extensively. This was because HOCl/OCl further degraded glutathione and also produced extensive lipid peroxidation. Interactions of HOCl/OCl and HOBr/OBr with red blood cells also differ significantly (Vissers et al., 1998; Hawkins et al., 2001). HOBr/OBr is about 10 times as potent in lysing red blood cells than is HOCl/OCl . This difference appears to be related to a more specific attack of HOBr/OBr on membrane proteins at sublytic concentrations. The lesser specificity of HOCl/OCl is demonstrated by a more extensive formation of halohydrin derivatives of phospholipids and cholesterol. Experience with other direct acting hemolytic agents indicate that oxidative attack on proteins is more important to hemolytic effects of oxidants than reactions with membrane lipids (McMillan et al., 2005).

GENERAL TOXICOLOGICAL PROPERTIES OF DISINFECTANT BY-PRODUCTS

147

HOCl/OCl reactions with monocytes and macrophages result in the formation of much higher levels of N-centered free radicals than HOBr/OBr at lysing doses, but there is no difference in the concentrations of the two agents to induce lysis (Hawkins et al., 2001). One might speculate that these results relate to the fact that HOCl more readily forms N-haloamines than HOBr. The specificity of the reactions of oxyhalide anions are further illustrated by their reactions with the purine and pyrimidine bases in DNA. Parsons and Chipman (2000) demonstrated that the oxidation of guanine in calf thymus DNA by BrO3 was thioldependent. In the presence of glutathione this was associated with the formation of oxidized glutathione (GSSG). Neither ClO3 nor IO3 participated in the reaction with guanine. It has subsequently been shown that the thiol-dependent oxidation of DNA by BrO3 is specific to guanine residues, affecting other purines and pyrmidines to a much lesser extent than other oxidants (e.g., hydroxyl radical) (Ballmaier and Epe, 1995; Murata et al., 2001). The formation of 8-OH-dG by OBr is also thiol-dependent, but formation of 8-OH-dG with OCl is independent of thiols (Ohnishi et al., 2002). OCl–induced oxidation of DNAwas also much less specific as reflected in the fact that this damage was piperidine-labile while that formed with BrO3 is not piperidine-labile. A different circumstance exists when high concentrations/doses of BrO3 are used, as increased cellular thiol concentrations will decrease the 8-OH-dG formation, while depletion of glutathione increases its formation (Sai et al., 1992; Parsons and Chipman, 2000; Murata et al., 2001). The specificity of the BrO3 interaction with thiols is important because the oxidation of guanine occurs much more extensively in the kidney, a target organ for BrO3-induced cancer, than in the liver, which is not a target organ (Kasai et al., 1987; Lee et al., 1996). The main conclusion is that there is a great deal of specificity in the individual reactions of oxyhalide anions with biological systems, and that this specificity has differing toxicological implications. 6.6.1.2 Descriptive Toxicological Data for Oxyhalide Anions Early investigations of chlorite’s toxic properties focused almost entirely on it potential ability to produce methemoglobin and hemolysis. Heffernan et al. (1979a, 1979b) examined the ability of chlorite to induce methemoglobin in cats and Sprague–Dawley rats. When administered as an oral bolus dose, as little as 20 mg/kg resulted in formation of significant amounts of methemoglobin. Intraperitoneal doses of 20 mg/kg in rats also induced methemoglobin. However, when administered in drinking water, no significant elevation in methemoglobin was observed in cats (up to 1000 mg/L as sodium chlorite) or rats (up to 500 mg/L). Thus, chlorite must enter into the systemic circulation at a rapid rate to induce methemoglobin and studies utilizing bolus doses are not representative of an exposure spread over the waking hours of the day. Treatment of both cats and rats with chlorite in drinking water for extended periods (up to 90 days) did result in decreases in red blood cell counts, hemoglobin concentrations and packed cell volume. These effects were observed with 500 mg sodium chlorite/L in cats and with as little as 100 mg/L in rats (Heffernan et al., 1979b). The changes in these blood parameters appeared to generally decrease in severity as the treatment was extended from 30 to 90 days indicating that adaptation to the treatment occurred. In rats, decreases in the RBC content of glutathione and elevation of 2,3-diphosphoglycerate were maintained through 90 days of treatment. In the cat, increased turnover of erythrocytes was detectable at concentrations of 100 mg/L with no significant effect being observed at 10 mg/L (daily dose calculated to be 0.6 mg/kg/day). Thus, the anemia caused by hemolysis was largely

148

DRINKING WATER DISINFECTION BY-PRODUCTS

compensated for in subchronic treatment of healthy rats, but there was still evidence of oxidative stress being exerted by the chlorite treatment. The results of Heffernan et al. (1979b) have been generally confirmed by subsequent studies in a variety of species (AbdelRahman et al., 1980; Couri and Abdel-Rahman, 1980; Moore and Calabrese, 1980; Bercz et al., 1982; Moore et al., 1984). An important study that treated male and female Crl: CD (SD) BR rats (Harrington et al., 1995a) with sodium chlorite at doses of 0, 10, 25, or 80 mg/kg/day by gavage for 13 weeks extended to include many of the standard parameters of subchronic toxicological studies, while previous studies had focused almost entirely on hematological parameters. Gavage doses of 80 mg/kg produced death in a number animals. RBC counts were reduced slightly at doses of 10 mg/kg in male rats with further decreases being observed at 80 mg/kg. RBC counts were significantly depressed in female rats at doses of 25 mg/kg and above. Histopathological examination of necropsied tissues revealed squamous cell epithelial hyperplasia, hyperkeratosis, ulceration, chronic inflammation, and edema in the stomach of 7/15 males and 8/15 females given 80 mg/kg doses. This effect was observed in only 2/15 animals at the 25 mg/kg dose and was not observed at all at 10 mg/kg. It is not known whether or not these local lesions in the stomach would be observed if chlorite were to be administered in drinking water. A two-generation reproduction and developmental toxicity study has been conducted in rats that included measures of hematological parameters in 25-day-old animals (Gill et al., 2000). Reductions in hemoglobin, hematocrit, mean cell volume, and a very small increase in methemoglobin concentrations were observed at 300 mg/L of drinking water. Only minor changes were observed at the next lower dose (70 mg/L). Therefore, young animals are not substantively more sensitive to chlorite-induced hematological effects than adult animals. A clinical study of the effects of chlorite, chlorate, and chlorine dioxide was conducted in three parts, a rising dose tolerance study in normal male volunteers (Lubbers and Bianchine, 1984), a 12-week subchronic study (Lubbers et al., 1981, 1982, 1984a) and a limited study in glucose-6-phosphate dehydrogenase deficient male volunteers (Lubbers et al., 1984b). The 12-week study required subjects to drink 500 mL of water containing 5 mg/L chlorine dioxide, chlorite or chlorate within a 15 min period each day. Although there were significant, but small trends in certain parameters, no physiological significance could be attributed to any of the results (Lubbers et al., 1984a). Three glucose-6-phosphate dehydrogenase-deficient individuals were studied under similar circumstances, but were followed for an additional 8 weeks post-treatment. Again no significant trends could be identified in this small group of individuals (Lubbers et al., 1984b). Very high doses (e.g., 2.2 g/kg) of KBrO3 oxidize hemoglobin when given parenterally (Watanabe et al., 2002). Both methemoglobin formation and lipid peroxidation in the kidney produced by this high dose are blocked by administration of thiols. The relevance of these findings to exposure to bromate at the concentrations found in drinking water has to be questioned. This effect has not been observed with treatment with as much as 600 mg/L in drinking water (Kurokawa et al., 1987). Much more relevant to the toxicity of bromate at low doses is the oxidative damage that is dependent upon thiols in the tissues at lower dose rates. The dependence of damage to DNA on glutathione or other thiols has been independently demonstrated in several laboratories both in vivo and in vitro (e.g., Ballmaier and Epe, 2006; Chipman et al., 2006; Kawanishi and Murata, 2006). Increases in 8-OH-dG levels in DNA are observed with treatment with as little as 250 mg/L in drinking water, which corresponds to 12.5 mg/kg/day (Umemura et al., 2004). These doses are only 20% of those needed to produce lipid peroxidation and less than 1% of those that produced methemoglobinemia.

GENERAL TOXICOLOGICAL PROPERTIES OF DISINFECTANT BY-PRODUCTS

149

Bromate, given orally, oxidatively damages DNA in the kidney, but neither chlorite nor hypochlorite produced this damage (Kasai et al., 1987). The chronic studies of bromateinduced damage to the kidney have been variously interpreted. Wolf et al. (1998) concluded that proximal tubule nephropathy did not occur, but there was some accumulation of hyaline droplets in the kidney. On the contrary, Umemura et al. (2004) found extensive damage to proximal tubular cells and cell proliferation at all carcinogenic doses in male rats, extending to doses much lower than those that increased the levels of 8-OH-dG in renal DNA. In part, renal damage by bromate in the rat appears to be related to its much greater potency in this species. Female rats were also observed to have an increased rate of cell division at carcinogenic doses, but the rate of cell division was more modest. At the high doses, the levels of 8-OH-dG were equivalent in the two sexes. Thus, it appears that proximal tubule damage may contribute to the carcinogenic response in both sexes. The lesser sensitivity of species other than the rat to bromate may relate more closely to less overt renal toxicity. This may explain why Ogg1-knockout mice treated with bromate did not develop tumors despite a substantial increase in the 8-OH-dG levels in renal DNA (Arai et al., 2006). Thekidneyexpressesthesodiumiodidesymporter(Spitzwegetal.,1998,2001)andbromate induces tumorsinthisorgan.Other tissues thatare target organsforbromatecarcinogenesisalso express the NIS. Bromate appears to be transported by the NIS (Dohan et al., 2003). This raises the possibility that pharmacokinetic as well as metabolic factors can play an important role in determining the nature and extent of adverse health outcomes by bromate. Cytotoxicity in the kidney of the rat also appears to distinguish bromate from other oxyhalide anions. KBrO3 produces clear evidence of renal toxicity when administered chronically in drinking water (Umemura and Kurokawa, 2006). Eosinophilic droplets were observed in the cytoplasm of the proximal renal tubule cells in male F344 rats and alpha2u-accumulation is observed at concentrations as low as 100 mg/L. In contrast, earlier studies with sodium chlorite at concentrations of up to 500 mg sodium chlorite/L of drinking water of rats for up to 180 days, found no evidence of renal toxicity (Moore et al., 1984). 6.6.1.3 Reproductive and Developmental Effects of Oxyhalides Chlorine dioxide treatment affects thyroid function at lower doses than either chlorite or chlorate. Bercz et al. (1982) found that administration of chlorine dioxide to primates at doses of 100 mg/L for 6 weeks produced a depression of serum thyroxine (T4) concentrations. This effect was not observed with up to 8 weeks treatment with 400 mg/L sodium chlorite or sodium chlorate. A similar effect was observed when chlorine dioxide was administered to rats at levels of 100 mg/L or more (Orme et al., 1985; Harrington et al., 1986; Carlton et al., 1987). These results could be explained by a more rapid absorption of [36Cl]-chlorine dioxide than the ions, chlorite and chlorate (Abdel-Rahman, 1985). Chlorite does not appear to interfere significantly with iodide uptake in intact animals. Bercz et al. (1982) found a 10% decrease in serum T4 in African green monkeys at a concentration of 400 mg/L for 8 weeks, but this was not statistically significant. Subsequent studies in rats have not found such depressions in rats treated with up to 60 mg/L during gestation (Mobley et al., 1990) or at concentrations of 300 mg/L (as sodium chlorite) at the 25th postnatal day (Gill et al., 2000). Chlorate does appear to have effects on the thyroid in rats. In the initial stages (days 4 and 21 of treatment) of a chronic study, dose-dependent decreases of triiodothyronine (T3) and T4 were observed at concentrations of sodium chlorate 41 g/L (Hooth et al., 2003). These effects were accompanied by a compensatory increase in serum TSH concentrations and the T3 and T4 returned to control levels by day 90. TSH remained significantly elevated at the highest dose

150

DRINKING WATER DISINFECTION BY-PRODUCTS

(2 g/L) at 90 days. Bercz et al. (1982) did not observe decreases in serum T4 in African green monkeys treated with 400 mg/L sodium chlorate, however, the measurements were only made at 8 weeks of treatment. In any case, thyroid function in primates (including humans) are much less sensitive to minor perturbations in iodide uptake than rats (NRC, 2005). Chronic KBrO3 treatment was found produce a depression in serum T3 levels, but this did not display a clear dose–response (Wolf et al., 1998). T4 levels were not significantly altered. Although BrO3 may be transported by the NIS, its affinity is apparently not high enough to interfere significantly with the organification of iodide in the thyroid at doses that can be tolerated chronically. Therefore, there appears no reason to conclude that bromate might contribute to thyroid-mediated effects on development at doses that would be obtained from drinking water. 6.6.2

Organic By-Products

The long list of organic by-products from disinfection makes it difficult develop a broad perspective of likely toxicities that might occur as a result of drinking waters treated with different disinfectants. Most effects of by-products are observed in the range of 1–100 mg/ kg/day. The highest dose that could be anticipated for any one by-product in water is about 1–2 mg/kg/day. As long as margins of exposure of this magnitude remain, it is unlikely that health effects are likely to result, with the possible exception of carcinogenesis. Attempts to define the toxicity of chlorination by-products collectively have been made by testing concentrated organic material from chlorinated water (Miller et al., 1986) or humic substances that have been chlorinated at high concentrations in the laboratory (Condie et al., 1985). The only clearly positive findings from these studies of concentrates has been increased mutagenic activity in bacterial and/or in vitro systems (Bull et al., 1982; Meier et al., 1983, 1985; Meier and Bull, 1985). Such studies continue to appear in the literature, but provide little basis for assessing risk, in vivo. While the negative data obtained with in vivo studies is reassuring, there are methodological difficulties with such studies that are difficult to overcome. Experiments involving concentrates of organic chemicals are plagued by questions of selective recovery, the potential for introduction of artifacts by the concentration technique, the possibility of increased reaction rates between components at the higher concentrations, the overwhelming concentrations of inorganic constituents of drinking water, and practical limits on the amount of organic material that can be recovered from drinking water. As a consequence of the above, the assessment of toxicological hazards associated with chlorination still must depend on what is known about the toxicology of individual byproducts or what can be learned from the study of simple binary or tertiary mixtures of these chemicals. The five classes of chemicals that have received the most attention are the THMs, selected haloacids, haloacetonitrile and some halogenated aldehydes and ketones. 6.6.2.1 Trihalomethanes Largely because of its long use as an anesthetic, chloroform is known for its ability to produce hepatic and renal damage in humans (Davidson et al., 1982). The cytotoxic effects of chloroform in the liver and kidney of rodents have received thorough study with short and long-term treatments utilizing oral and inhalation exposure. In addition, the nasal turbinates of the rat have been identified as a particularly sensitive target organ for chloroform (Larson et al., 1995b). The toxicity of other THMs also primarily appears to involve cytotoxic effects to the liver and or kidneys (Moore et al., 1982; Chu et al., 1982a, 1982b; Munson et al., 1982; Thornton-Manning et al., 1994; Lilly et al., 1997).

GENERAL TOXICOLOGICAL PROPERTIES OF DISINFECTANT BY-PRODUCTS

151

The toxicity of chloroform is dramatically affected by the vehicle in which it is administered, being significantly augmented when given in an oil vehicle (Bull et al., 1986; Larson et al., 1994a, 1995a, 1995b, 1996). It has been suggested that these effects are largely the result of altered kinetics of absorption and subsequent metabolism of chloroform, although this has not been rigorously demonstrated. Similar differences have been noted with short-term treatment with BDCM (Lilly et al., 1994), but that separate treatments with corn oil with BDCM in an aqueous vehicle had no effect on its toxicity to the liver. These data seem to suggest that the different responses that have been extensively documented relate more to the method of administration than the vehicle (i.e., gavage versus drinking water). The toxicity of chloroform and the other THMs depends on their metabolism to reactive intermediates. This activation can occur by several mechanisms, oxidative metabolism to phosgene (Pohl and Krishna, 1978), reductive metabolism to free radical intermediates (Testai et al., 1986), or as a result of a reactive metabolite formed from a glutathione conjugate (Ross and Pegram, 2003). The general toxicological properties of the THM4 are quite similar, so the other members of the class will not be reviewed here. The more critical issue relates to their carcinogenic activities that are discussed in a subsequent section. 6.6.2.2 Haloacetates There are significant differences in the toxicities produced by various members of the HA class of DBPs. The monohaloacetates are different both in terms of mechanisms and effects from the dihaloacetates, which in turn differ from the trihaloacetates. However, there are parallels within each of these groups. All of the members of the dihaloacetate class are rapidly metabolized, to a large extent through a bifunctional cytosolic enzyme, cytosolic glutathione transferase zeta (Lipscomb et al., 1995; Tong et al., 1998). However, the brominated analogs are increasingly involved in processes, presumably metabolic, that induce lipid peroxidation and oxidative damage to nuclear DNA of the hepatocyte (Austin et al., 1996). Prior treatment of either humans (Curry et al., 1985) or rodents (Gonzalez-Leon et al., 1997) inhibits the metabolism of the dihaloacetates that occurs in the cytosol. Inhibition of dibromochloroacetate (BDCA) metabolism by DCA suggests that the initial dehalogenation of this trihaloacetate is also mediated by glutathione transferase zeta. As bromine substitution increases, the metabolism of the dihaloacetates appears to shift from the cytosol to microsomal fractions. Perhaps this shift accounts for the increasing evidence of formation of reactive oxygen species (Austin et al., 1996). There is human experience with DCA because of early investigations of its potential as an oral hypoglycemic agent (Crabb et al., 1976). The association of neurological deficits characterized by weakness of muscles in the lower extremities, face and fingers, diminished tendon reflexes; and slowed nerve conduction velocity in individuals treated with doses of DCA as low as 50 mg/kg/day for 16 weeks (Moore et al., 1979; Stacpoole et al., 1979) has limited its use in medicine. Subsequent to this work, DCA was shown to cause hind limb weakness and/or paralysis in dogs and rats (Katz et al., 1981; Yount et al., 1982; Mather et al., 1990; Cicmanec et al., 1991). In addition to peripheral neuropathy, DCA administered sub-chronically produces testicular abnormalities in rats, prostate glandular atrophy, cystic mucosal hyperplasia in the gall bladder and ocular lesions in dogs (Katz et al., 1981; Yount et al., 1982; Mather et al., 1990). The testicular effects of DCA have been examined with the other dihaloacetic acids. The implications of these studies are dealt with more extensively in the section addressing reproductive and developmental toxicities of the HAs. The most dramatic effect of the

152

DRINKING WATER DISINFECTION BY-PRODUCTS

dihaloacetates in rodents is the production of a severe hepatomegaly and cytomegaly that is associated with large accumulations of glycogen in mice (Sanchez and Bull, 1990; Carter et al., 1995). Similar, but somewhat less severe hepatomegaly and glycogen accumulation has been observed in rats (Bull et al., 1990b; Mather et al., 1990). Accumulation of glycogen in heptatocytes of mice is characteristic of the dihaloacetates, with DCA producing a more rapid accumulation, reaching a maximum level within 1 week of treatment. Significant increases of glycogen are observed with concentrations of DCA in drinking water as low as 0.2 g/L (Kato-Weinstein et al., 1998). The glycogen becomes progressively more resistant to mobilization by fasting. Substantial decreases in serum insulin are observed with treatments of 14 days or longer in duration (Kato-Weinstein et al., 1998). Since this effect develops with a latency of about 14 days, it is probably a result of down-regulating some of the insulin-mimetic effects of DCA. The effects on glycogen deposition appear to result from an activation of a phosphotidyl 3-kinase (PI3K) dependent pathway in murine hepatocytes (Lingohr et al., 2002). This effect parallels effects of insulin, but the effect of insulin is only partially blocked by PI3K-inhibitors, while the effect of DCA on glycogen deposition is completely ablated by these compounds. It is postulated that these effects are responsible for the down-regulation of serum insulin concentrations and insulin receptor expression in the liver (Lingohr et al., 2001). The dose–response for glycogen disposition relates closely to that observed for the hepatomegaly and induction of distortions of cell size seen with dihaloacetate-treatment in vivo. The glycogen accumulation induced by DCA in the liver occurs at much lower doses than shown to activate the pyruvate dehydrogenase complex through inhibition of the pyruvate dehydrogenase kinase (Blackshear et al., 1974; Whitehouse et al., 1974; Crabb et al., 1976; Stacpoole, 1989). This classical effect of DCA occurs only after sufficient amounts have been administered to inhibit glutathione transferase zeta, because blood levels do not approach the Ki for the pyruvate dehydrogenase kinase (ca. 100 mM)(Whitehouse et al., 1974) unless its metabolism is inhibited (discussed in more detail later). Analytical methods currently available are insufficient to measure the minimum systemic concentrations of DCA that are necessary to produce toxicity or carcinogenicity. Methods that are sufficiently sensitive to measure blood concentrations of 51 mM in rodents show that systemic levels below this are sufficient to produce hepatomegaly and liver cancer (Kato-Weinstein et al., 1998; Merdink et al., 1998b) and require high concentrations of DCA in drinking water of 0.5 g/L to approach this systemic level. The trihaloacetates produce a less marked hepatomegaly than the dihaloacetates (Sanchez and Bull, 1990; Kato-Weinstein et al., 1998; Stauber et al., 1998). Within the group, TCA is effective in the range of 0.5–2 g/L as a peroxisome proliferator, but it is not observed with BDCA treatment in the same concentration range (Kato-Weinstein et al., 1998). The ineffectiveness of bromine-substituted trihaloacetates as peroxisome proliferators may be related to their more rapid and complete metabolism (Austin and Bull, 1997). In contrast to DCA, TCA reduces liver glycogen (Kato-Weinstein et al., 2001). At low doses BDCA does produce a distribution pattern of glycogen deposition more similar to that of DCA, but it requires considerably more time to develop. It is likely that this is explained by the metabolism of BDCA to DCA (Xu et al., 1995; Austin and Bull, 1997), but its rate of conversion is less than prior studies suggested (Merdink et al., 2001). As with the dihaloacetates, the most prominent effects of the trihaloacetates are in the liver (Goldsworthy and Popp, 1987; Mather et al., 1990; Bull et al., 1990). TCA is without apparent cytotoxic effects in liver cells in vivo at dose rates of up to 300 mg/kg/day (Sanchez and Bull, 1990; Acharya et al., 1997) or in vitro at concentrations as high as 5 mM

GENERAL TOXICOLOGICAL PROPERTIES OF DISINFECTANT BY-PRODUCTS

153

(Bruschi and Bull, 1993). Similar liver enlargement has been observed with BDCA. Trihaloacetates with greater bromine substitution have not been studied. 6.6.2.3 Haloacetonitriles The relationship between the metabolism of the HANs and their toxic effects is less well established. Pereira et al. (1984) found that from 2.3 to 14% of doses of chloroacetonitrile, dichloroacetonitrile, bromodichloroacetonitrile, dibromoacetonitrile and trichloroacetonitrile were converted to thiocyanate in the urine. Chronically, production of cyanide followed by conversion to thiocyanate is associated with adverse effects on the thyroid gland (Olusi et al., 1979). In addition to cyanide, Pereira et al. (1984) postulated that metabolism of the HANs would yield the alkylating agents, formaldehyde with chloroacetonitrile, formyl chloride or formylhalide from the dihaloacetonitriles and phosgene or cyanoformyl chloride from trichloroacetonitrile. These reactive metabolites may be associated with the mutagenic effects of these chemicals (Bull et al., 1985). The HANs do not produce distinctive toxicological effects with treatments of 14 or 90 days in duration (Hayes et al., 1986). Dichloroacetonitrile is tolerated at daily doses as high as 8 mg/kg for 90 days in male and female CD rats. Dibromoacetonitrile was tolerated at doses of 23 mg/kg/day for 90 days. 6.6.2.4 3-Chloro-4-(dichloromethyl)-5-Hydroxy-2[5H]-Furanone (MX) The toxicology of MX, other than those effects associated with its carcinogenicity, has received very limited attention. Most research has focused on testing whether its potent mutagenic effects in bacterial systems and in vitro can be duplicated in vivo. It would appear from these studies that animals can tolerate dose rates of 32 mg/kg for up to 14 days (Meier et al., 1996). A dose rate of 64 mg/kg was overtly toxic. Little evidence of toxicity to specific organs was observed in the chronic bioassay of MX in Wistar rats (Komulainen et al., 1997). The mutagenic activity of MX is effectively quenched by glutathione and significantly decreases the systemic absorption of the chemical (Clark and Chipman, 1995). This may account for the fact that MX appears to be less potent as a carcinogen than would be suggested by its mutagenic potency. 6.6.2.5 Miscellaneous Several groups of DBPs include mutagenic members. The more frequently identified compounds fall into the broad classes of haloaldehydes, haloketones and halogenated hydroxy-furanones (Meier, 1988). With the exception of chloral hydrate, the metabolism of these compounds has received very little systematic study. Chloral hydrate is largely metabolized to trichloroethanol and TCA (Merdink et al., 1998a, 1998b). Lesser amounts may be converted to DCA, but the concentrations achieved in blood are quite low (Merdink et al., 1998b). Chloral hydrate has been used in high doses (500 to more than 2000 mg) as a sedative/hypnotic. This effect has been largely attributed to its conversion to trichloroethanol (Gessner and Cabana, 1970). The central nervous system depressant effects require doses of chloral hydrate much higher (e.g., 500 mg) than could be obtained from drinking chlorinated water. Therefore, most emphasis has been placed on the conversion of chloral hydrate to DCA and TCA, two established liver carcinogens in rodents. Some work has been conducted examining the toxic properties of a limited number of the halogenated ketones that have been identified as chlorination by-products. Daniel et al. (1993) examined the effects of 1,1,1-trichloro-2-propanone. The only significant pathology was observed as hyperkeratosis in the stomach of rats treated by intubation at doses of 48 mg/kg and higher for up to 90 days. No effects were seen at 16 or 30 mg/kg/day.

154

DRINKING WATER DISINFECTION BY-PRODUCTS

6.7 CARCINOGENIC PROPERTIES OF DISINFECTANTS 6.7.1

Chlorine and Chloramines

Administration of chlorine to rats or mice in drinking water at concentrations up to 250 mg/L failed to produce any specific toxicological effects (Druckrey, 1968; Furukawa et al., 1980; Hasegawa et al., 1986; Kurokawa et al., 1986b). The National Toxicology Program (NTP, 1990b) completed a carcinogenesis bioassay of sodium hypochlorite and chloramine in drinking water at concentrations up to 275 mg/L. There was little indication of pathology in either mice (B6C3F1) or rats (F344/N) with either disinfectant. There was a significant increase in mononuclear cell leukemia in female rats, which was discounted because the incidence in the concurrent control group was significantly lower than that of the historical controls (ca. 25%). Compared against historical controls, there was no significant increase in leukemia in either sex. Soffritti et al. (1997) conducted carcinogenesis studies in Sprague– Dawley rats and observed marginal increases in the combined incidence of lymphomas and leukemias at 500 and 750 mg/L (6/group versus 4/group in controls). These data were not appropriately reported, no statistical analysis was provided, and the doses administered were above the maximally tolerated doses of other studies. None of the prior studies found evidence that chlorine in drinking water was carcinogenic (Druckrey, 1968; Furukawa et al., 1980; Hasegawa et al., 1986; Kurokawa et al., 1986b). Collectively, these data suggest that neither chlorine nor chloramine is carcinogenic in experimental animals. Hypochlorite does possess weak mutagenic activity in bacterial systems (Rosenkranz, 1973; Wlodkowski and Rosenkranz, 1975; Rosenkranz et al., 1976). Such responses are observed only at cytotoxic doses, however. The absence of other evidence of genotoxic activity, particularly in vivo, makes it difficult to assign much importance to these responses. Shih and Lederberg (1976) and Thomas et al. (1987) both reported that chloramine is also mutagenic in bacterial systems. 6.7.2

Chlorine Dioxide

There are no modern chronic toxicological studies of chlorine dioxide. 6.7.3

Ozone

As ozone off-gases rapidly from water and any remaining amount reacts with water constituents with a short half-life, no residual ozone reaches the tap. As a consequence, a chronic study is not warranted. 6.8 CARCINOGENIC BY-PRODUCTS OF DISINFECTANTS Those DBPs that have been studied as potential carcinogens are listed in Table 6.7. To facilitate comparisons of the potential hazards these chemical might pose at the concentrations found in drinking water, it lists the dose of the chemical that would produce an increase of cancer in 50% of the animals in a lifetime (TD50). In this chapter, readers will find judgments made relative to the ability of identified compounds to make a significant contribution to cancer risks of the magnitude that have been associated with chlorinated drinking water. These judgments are made considering the concentrations that are listed in Table 6.2 and the TD50 in Table 6.7.

155

CARCINOGENIC BY-PRODUCTS OF DISINFECTANTS

TABLE 6.7 in Animals

By-Products of Disinfection That Have Been Shown to Produce Cancer

By-Product

TD50 (mg/kg/day)

Species/Strain/Sex/Site

Vehicle Route

Result

Mice/Strain A/F/livera Mice/B6C3F1/M/liverb Mice/B6C3F1/F/liverb Mice/B6C3F1/F/liverc Mice/ICI/M/kidneyd Mice/ICI/M/kidneyd Mice/CBA/Md Mice/CF1/Md Rats/Osb./M/kidney, thyroidb Rats/Osb.-M/M/kidneyc Rats/Osb.-M/M/thyroidc

Olive oil/gavage Corn oil/gavage Corn oil/gavage Drinking water Toothpaste/gavage Toothpaste/gavage Toothpaste/gavage Toothpaste/gavage Corn oil/gavage

Positive Positive Positive Negative Positive Negative Negative Negative Positive

90.3

Drinking water Drinking water

Positive Negative

262

Mice/B6C3F1/M/kidneye Mice/B6C3F1/F/livere Rats/F344/M&F/kidney, intestinee Mice/B6C3F1/M&Ff Rats/F344/M&Ff

Corn oil/gavage Corn oil/gavage Corn oil/gavage

Positive Positive Positive

47.7

Drinking water Drinking water

Negative Negative

CHBr2Cl

Mice/B6C3F1/M/liverg Rats/F344/M&Fg

Corn oil/gavage Corn oil/gavage

Positive Negative

CHBr3

Rats/F344/M/intestineh Rats/F344/F/intestineh

Corn oil/gavage Corn oil/gavage

Positive Positive

CHI3

Mice/B6C3F1/M&Fi Rats/F344/M&Fi

Corn oil/gavage Corn oil/gavage

Negative Negative

Mice/B6C3F1/M&Fj Rats/F344/M&Fj Rats/F344/Mk

Water/gavage Water/gavage Drinking water

Negative Negative Negative

Mice/B6C3F1/M/liverl Mice/B6C3F1/M/liverm Mice/B6C3F1/M/livern Mice/B6C3F1/F/livero Mice/B6C3F1/M/liverp Rats/F-344/M/liverq

Drinking Drinking Drinking Drinking Drinking Drinking

water water water water water water

Positive Positive Positive Positive Positive Positive

102

Mice/B6C3F1/M/liverl Mice/B6C3F1/M/liverm Mice/B6C3F1/F/livern Mice/B6C3F1/M/liverp Rats/F344/M/liverk

Drinking Drinking Drinking Drinking Drinking

water water water water water

Positive Positive Positive Positve Negative

584

Trihalomethanes CHCl3

CHBrCl2

Haloacetates ClCH2COOH

Cl2CHOOH

Cl3COOH

72.5

648

NA

(continued)

156

DRINKING WATER DISINFECTION BY-PRODUCTS

TABLE 6.7

(Continued)

By-Product 3-Chloro-4 (dichloromethyl)5-hydroxyl 2(5H)furanone (MX)

Species/Strain/Sex/Site Rats/Wistar/M&F/ thyroid, liver, adrenal, lungs, mammary, lymph/ leukemiar

Vehicle Route

Result

TD50 (mg/kg/day)

Drinking water

Positive

1

Diet

Positive

405

Diet

Positive

1070

Mice/B6C3F1/Neo./livert Mice/B6C3F1/M/livern Rats/F344/Mu Mice/B6C3F1/M&F/liverv MiceB6C3F1/F/pituitaryw

Single oral Drinking water Drinking water Gavage Gavage

Positive Positive Negative Positive Equivocal

Formaldehyde

Rats/F344/M&F/nasalx Rats/Wistar/M/nasaly Rats/S-D/M&F/leukemiaz

Inhalation Inhalation Drinking water

Positive Positive Positive

2.19

Acetaldehyde

Rats/Wistar/M&F/nasalaa

Inhalation

Positive

153

Rats/F344/M&F/kidneybb Rats/F344/M/kidneycc Rats/F344/M/thyroiddd Rats/F344/M/peritoniumbb Hamsters/Syr/M/kidneycc Mice/B6C3F1dd

Drinking Drinking Drinking Drinking Drinking Drinking

Positive Positive Positive Positive Positive Positive

7.5

Chlorate

Rats/F344/M/thyroidee Mice/B6C3F1/F/pancreasee

Drinking water Drinking water

Positive Equivocal

Chlorite

Rats/F344/M&Fff Mice/B6C3F1/M/liverff

Drinking water Drinking water

Inadequate Positive

Mice/C57BL/M&F/ intestinegg

Drinking water

Positive

Rats/several/M&F/liver, kidney, lung, testicle, vasculaturehh Mice/several/M&F/liver, nervous system, lungii

Drinking water

Positive

0.096

Drinking water

Positive

0.189

2,4,6-Trichlorophenol Mice/B6C3F1/M&F/ Livers Rats/F344/M&F/ leukemias Aldehydes Chloral hydrate

Oxyhalides Bromate

Hydrogen peroxide

Nitrosamines N-Nitroso-Ndimethylamine (NDMA)

water water water water water water

106

405 41

NE

157

CARCINOGENIC BY-PRODUCTS OF DISINFECTANTS

TABLE 6.7

(Continued)

By-Product N-Nitroso-Ndiethylamine (NDEA)

N-Nitroso-mopholine

N-Nitroso-pyrrolidine

Species/Strain/Sex/Site

Vehicle Route

Result

TD50 (mg/kg/day)

Rats/several/M&F/liver, eso, kidney, vasculature, stomachii Monkeys/Rhesus/liverii Monkeys/Cynomulgusii

Drinking water

Positive

0.0265

Food Food

Positive Positive

0.0536 0.00725

Rats/F344/F/liver, vasculatureii Hamsters/Syr/M&F/liver, Nasal, oral cavityii

Drinking Water

Positive

0.109

Drinking Water

Positive

3.57

Rats/several/kidney, liver, vasculatureii Hamster/Syrii Mice/Swiss/M/all tumorii

Drinking Water

Positive

0.799

Drinking water Drinking water

Positive Positive

14.2 0.679 Bearing animals

N-Nitroso-piperidine

Rats/SD/M&F/liver, lung, stomachii Mice/icm/M&F/ esophagus, liver, nasal Hamster/Syr/M&F/liver, Nasal, oral cavityii Monkeys/Rhesus/liverii Monkeys/Cynomulgus/ liverii

NE: not estimated. a Eschenbrenner and Miller (1945). b NCI (1976). c Jorgenson et al. (1985). d Roe et al. (1979). e NTP (1986). f NTP (2005b). g Dunnick et al. (1985). h NTP (1989). i NCI (1978). j NTP (1990a, 1990b). k DeAngelo et al., 1997. l Herren-Freund et al. (1987). m Bull et al. (1990a, 1990b). n Daniel et al. (1992). o Pereira (1996). p Stauber and Bull, 1997. q DeAngelo et al. (1996). r Komulainen et al. (1997).

s

Drinking water

Positive

1.43

Food

Positive

1.3

Drinking Water

Positive

83.3

Food Food

Positive Positive

2.76 12.1

NCI (1979). Rijhsinghani et al. (1986). u George et al. (2000a). v NTP (2000a). w NTP (2002). x Kerns et al. (1983). y Johannsen et al. (1986). z Soffritti et al. (1989). aa Woutersen et al. (1984). bb Kurokawa et al. (1983). cc Kurokawa et al. (1986). dd Takamura et al., 1985. ee NTP (2005). ff Kurokawa et al. (1986b). gg Ito et al. (1982). hh Peto et al. (1991). ii Gold et al. (2006). t

158

DRINKING WATER DISINFECTION BY-PRODUCTS

Because of the generally low concentrations of DBPs in drinking water, it is necessary to consider dose–response relationships, mechanisms by which the cancer is produced, and the intrinsic human susceptibility to the effects of these compounds before concluding that they represent a risk to humans. There have been substantial efforts invested in understanding the modes of action of THMs, HAs, and nitrosamine by-product groups. 6.8.1

Carcinogenic Activity of Oxyhalide Anions

6.8.1.1 Descriptive Data—Oxyhalide Anions The carcinogenicity of bromate poses a serious problem for utilities that use ozone as a disinfectant in water supplies that contain significant quantities of bromide. Of all DBPs studied to date, bromate occurs at concentrations that are closer to those that produce overt effects in animals. It can be found at concentrations of 5–50 mg/L in waters containing substantial amounts of bromide. Bromate has been shown to induce renal tumors in two species, male and female F344 rats (Kurokawa et al., 1983, 1986a) and male Syrian Golden hamsters (Takamura et al., 1985). In addition, mesothelioma and thyroid tumors were induced in male rats (Kurokawa et al., 1983). The minimally effective dose rate of bromate as a carcinogen in the rat is 2.7 mg/kg/ day (Kurokawa et al., 1986a). The species differences in sensitivity to bromate are substantial, with TD50s of 7.5, 41, and 408 mg/kg/day in rats, mice, and hamsters, respectively (Gold et al., 2006). Clearly, there is evidence that bromate is carcinogenic in animals. Chlorate and chlorite are the other inorganic by-products of concern, which can occur at reasonably high concentrations in water disinfected by hypochlorite solutions (chlorate) or chlorine dioxide (primarily chlorite). The chronic toxicity of sodium chlorate has been studied in by the US Environmental Protection Agency (Hooth et al., 2001) and by the National Toxicology Program (NTP, 2005a). Chlorate treatment at concentrations in drinking water 41 g/L (435 mg/kg/day) resulted in thyroid tumors in male rats and a marginal increase pancreatic tumors in female mice. It appears that the thyroid tumors result from inhibition of iodide uptake which decreases thyroid hormone synthesis and subsequent release of TSH (Hooth et al., 2001). Rats are known to be extraordinarily sensitive to the development of thyroid tumors by this mechanism (NRC, 2005), there fore, chlorate at the levels found in drinking water is unlikely to represent a carcinogenic hazard to humans. Sodium chlorite has been studied as a carcinogen at concentrations of 0, 300, and 600 mg/L of drinking water in F344 rats and 750 mg/L in B6C3F1 mice (Kurokawa et al., 1986b). The rats became infected with a Sendai virus in all groups that resulted in the termination of the study after only 85weeks.In male mice therewas a statistically significant increase in the incidence of hyperplastic nodules in male mice treated with 250 mg/L, but not in females. The incidence of these lesions did not increase when the dose of chlorite was increased to 500 mg/L. Hepatocellular carcinomasweretoo few toadd anythingsubstantivetotheevaluation. Therewerenoother treatment-related changes in the incidence of other tumors in either male or female mice. Yokose et al. (1987) examined the carcinogenicity of sodium chlorite in B6C3F1 mice in a chronic bioassay. The concentrations utilized in this study were 0.025% (250 mg/L) and 0.05% (500 mg/L) as sodium chlorite. A small, but statistically significant (p 5 0.05) increase in the incidence of lung adenomas was observed at 500 mg/L. The authors noted that this was not accompanied by the appearance of lung adenocarcinomas. The chronic toxicological studies of chlorite are insufficient to determine whether it presents a carcinogenic hazard or not. This is primarily because of the inadequacy of the study in rats. On the contrary, the concentrations used in the studies in mice were at 250 times those found in drinking water without producing any striking evidence of carcinogenic activity.

CARCINOGENIC BY-PRODUCTS OF DISINFECTANTS

159

6.8.1.2 Mode of Action—Bromate The involvement of DNA damage in bromateinduced cancer was discussed previously. There is other evidence to support this mode of action as an important aspect of its effects. There are also other data to suggest that the dose–response for bromate may not be linear at low doses. Delker et al. (2006) examined changes in gene expression that occur in the kidney of male rats subjected to prolonged treatment with bromate in drinking water at concentrations that were noncarcinogenic and carcinogenic, 20 and 400 mg/L, respectively. Genes that metabolize glutathione were up-regulated at the high dose, but not at the low doses. Ogg1, part of the repair process for 8-OH-dG lesions in DNA, was also up-regulated. These changes are consistent with the oxidative stress mechanism postulated to be responsible for the tumorigenic effects of bromate. The data also suggested a sharp increase in 18O from labeled KBrO3 into macromolecules of the kidney at doses of bromate 420 mg/L. Collectively, these data suggest a transition in the degree of accumulating damage in DNA and the induction of compensatory responses as treatment levels approach doses that induce cancer in the rat. These studies are being pursued further and may provide information that will allow a more precise definition of the risks from bromate at the much lower concentrations that are found in drinking water treated with ozone. Bromate is unstable at low pH and undoubtedly reacts with thiols in nontarget tissues with distribution in the body (Keith et al., 2006). Therefore, it is probable that delivery of bromate to tissues is nonlinear at low doses (Fisher and Bull, 2006). It is probable that linear extrapolation from high doses to low doses overestimates its carcinogenic risk. 6.8.2

Organic By-Products

6.8.2.1 Descriptive Data—Trihalomethanes The THMs have received the most study as carcinogens. All four compounds have been shown to produce cancer in either rats or mice, but only chloroform and BDCM produced tumors in both species. However, the induction of tumors by these two compounds has been shown to be dependent upon the vehicle that is used and/or the method of administration. Jorgenson et al. (1985) found no liver tumors in female B6C3F1 mice treated with chloroform in drinking water at doses that were similar to those required to produce an 80% incidence of tumors in the NCI (1976) study, where it had been administered by corn oil gavage. A low incidence of renal tumors was still observed in the rat with the drinking water exposure. Similarly, a recent NTP (2005b) bioassay of BDCM using drinking water as the vehicle found no tumors at any site in either mice or rats, despite the fact that the compound produced tumors of the liver, kidney and large intestine when administered by stomach tube in corn oil (NTP, 1986). DBCM was shown to induce liver cancer in mice (NTP, 1986) and bromoform induced colon tumors in rats (NTP, 1989). Both experiments were conducted by administering the chemical in corn oil by intubation. The BDCM bioassay in the rat was considered inadequate, but it has not been repeated. No tumors were observed in mice treated with bromoform. In light of the lack of carcinogenic effect of BDCM when administered in drinking water, the likelihood that DBCM and bromoform are carcinogens in drinking water also has to be questioned. 6.8.2.2 Mode of Action—Trihalomethanes Chloroform induces tumors by cytotoxicity in the target organs that is followed by reparative hyperplasia (USEPA, 2005). This conclusion is supported by a wide variety of data. First, evidence that chloroform is

160

DRINKING WATER DISINFECTION BY-PRODUCTS

mutagenic is very weak (Bull et al., 2001) and the few positive results are counterbalanced by the inability to demonstrate covalent interactions with DNA (Diaz-Gomez and Castro, 1980; Reitz et al., 1982) in vivo or to induce unscheduled DNA synthesis in the liver of mice (Larson et al., 1994b). Second, it has been shown that the doses that produced tumors in the rat kidney or mouse liver also produced significant cytotoxicity and evidence of increased cell division (Larson et al., 1994a, 1995a). Moreover, these latter data are consistent with a variety of other findings by different investigators. For example, studies have demonstrated that the corn oil vehicle significantly enhanced the liver damage produced by chloroform (Bull et al., 1986; Larson et al., 1995a). Evidence from initiation/promotion studies has shown that a corn oil vehicle can render chloroform as a hepatic tumor promoter in diethylnitrosamine-initiated rats (Deml and Oesterle, 1985), but chloroform in drinking water was inactive in this regard (Herren-Freund and Pereira, 1986). Subsequent studies have confirmed the lack of tumor promoting activity of chloroform given in drinking water (Klaunig et al., 1986). In point of fact, chloroform in drinking water has been shown to inhibit diethylnitrosamine- and ethylnitrosourea-induced liver tumors in mice (Klaunig et al., 1986; Pereira et al., 1985) and to reduce yields of dimethylhydrazine-induced gastrointestinal tract tumors (Daniel et al., 1989). To further complicate the use of studies that utilized corn oil gavage to administer THMs in risk assessment is the finding that chloroform administered in drinking water could abolish the cell killing and cellular proliferation produced by chloroform administered by corn oil gavage in a dose-dependent manner (Pereira and Grothaus, 1997). The lowest effective concentration of chloroform in drinking water to inhibit the induction of hepatic necrosis by a high dose of chloroform in corn oil was 120 mg/L, or about 12 mg/kg/day. These doses are much below those required to produce liver cancer, even when administered in corn oil. The finding of parallel route- and vehicle-specific responses to BDCM raise considerable doubt as to whether THMs at the low doses in drinking water should be considered as a carcinogenic threat to humans at all. Some credence has been given to the finding of colon cancer in rats treated with BDCM and bromoform (NTP, 1986, 1989) because of the occasional association of colon and rectal cancer with chlorinated drinking water. High doses were required to induce such tumors in rats, 450 and 200 mg/kg/day, for BDCM and bromoform, respectively, administered by gavage in a corn oil vehicle. As indicated, concentrations of 700 mg/L BDCM in drinking water did not induce colon tumors in rats (NTP, 2005b). BDCM, DBCM, and bromoform appear capable of inducing point mutations in Salmonella (Zeiger et al., 1990). All of the brominated THMs appear to be capable of inducing mutations in the mouse lymphoma assay. BDCM has been shown to produce adducts to DNA when incubated with tissue homogenates of kidney and intestine of the rat (Ross and Pegram, 2004), which were target tissues when BDCM was administered in corn oil. However, no evidence of in vivo genotoxicity was observed with BDCM even though it was administered intraperitoneally (NTP, 2005b). At the very least, cancer risk estimates based on corn oil gavages studies of the THMs appear to be substantial overestimates. Mode of action studies further reduce the likelihood that THM4 in drinking water contribute to the development of cancer in humans. 6.8.2.3 Descriptive Data—Haloacetates DCA and TCA have been shown to induce hepatic tumors in B6C3F1 mice (Herren-Freund et al., 1987; Bull et al., 1990a; DeAngelo et al., 1991, 1997; Daniel et al., 1992; Pereira, 1996; Stauber and Bull, 1997). Unlike the THMs, these data have all been collected using drinking water as the vehicle. There is a

CARCINOGENIC BY-PRODUCTS OF DISINFECTANTS

161

remarkable similarity in the results obtained among laboratories. DCA also induces liver tumors in F344 rats (DeAngelo et al., 1996), but TCA does not (DeAngelo et al., 1997). As was the case with DCA, DBA was shown to induce hepatic tumors in mice administered 50, 500, and 1000 mg/L of drinking water. In addition, a small increase in alveolar/bronchiolar adenomas was observed in the lungs of mice (NTP, 2005c). Rats were found to have a dose-related increase in the incidence of mononuclear cell leukemia in female rats and an increase in malignant mesothelioma at the high dose in male rats. Preliminary data from the author’s laboratory also indicate that BDCA, BCA, and DBA produce dose-related increases in liver cancer in male B6C3F1 mice. Lymphoma and lung tumor incidence are significantly elevated in B6C3F1 mice treated with BDCA. In separate experiments, F344 rats developed aberrant crypt foci in the distal colon as a result of treatment with DBA (So and Bull, 1995). These data support the hypothesis that all of the dihaloacetates and the brominated trihaloacetates are carcinogenic. MCA is not carcinogenic. It produces no increases in tumor incidence in either B6C3F1 mice or F344 rats (NTP, 1990a; DeAngelo et al., 1997). 6.8.2.4 Modes of Action—Haloacetates Suicide inhibition plays an important role in the pharmacokinetics of the dihaloacetates in both rodents and humans (Anderson et al., 2002). If repeated doses of DCA 4 10 mg/kg/day are administered, its clearance becomes progressively prolonged and is accompanied by increases in peak concentrations by two orders of magnitude in rodents (Gonzalez-Leon et al., 1997; Schultz et al., 1999; Schultz et al., 2002) and humans (Curry et al., 1985) relative to these same variables with the administration of single doses. In reality DCA is effective as a liver carcinogen below concentrations that produce significant inhibition of the glutathione transferase zeta, which is responsible for its rapid metabolism (Tong et al., 1998; Tzeng et al., 2000). These data imply that DCA is active at systemic concentrations as low as 1 mM (Kato-Weinstein et al., 1998; Barton et al., 1999; Merdink et al., 1998b). However, to approach the limit of detection in blood requires that animals drink 0.5 g/L of DCA. These results are important in considering the extent to which in vitro and in vivo toxicological studies that have only examined the effects of very high concentrations or doses (e.g., 4100 mM in vitro, or 40 mg/kg/day in vivo) are relevant to doses obtained from drinking water. Acinar necrosis and reparative hyperplasia are observed with high doses of DCA 4 2 g/L (Bull et al., 1990a; Sanchez and Bull, 1990). However, these lesions are not observed at lower doses that still result in a high incidence of hepatic tumors (Stauber and Bull, 1997; KatoWeinstein et al., 1998). Considering the pharmacokinetics of DCA, it appears that the carcinogenic effects of DCA in its lower effective dose range are largely, if not entirely, attributable to modification of clonal growth rates at the expense of depressed rates of cell division of normal hepatocytes (Stauber and Bull, 1997). The lower rate of cell division in normal hepatocytes apparently arises from suppressed apoptosis (Snyder et al., 1995), whereas cell replication rates within responsive clones remains the same and then becomes greatly elevated at high doses (Miller et al., 2000). These effects of DCA can be replicated in anchorage-independent colonies of primary hepatocytes from B6C3F1 mice (Stauber et al., 1998). The colonies match the phenotype of tumors produced by DCA in vivo. The concentrations shown to stimulate colony growth were as low as 10 mM. These data strongly suggest that DCA is stimulating the growth of spontaneously initiated cells in the mouse liver. The carcinogenic effects of DCA may be indirectly related to its insulin-mimetic effects on the liver, which were discussed earlier. Glycogen accumulation in the liver is a hallmark of

162

DRINKING WATER DISINFECTION BY-PRODUCTS

carcinogenic dose rates of DCA in mice (Kato-Weinstein et al., 1998). Tumors induced by DCA express higher levels of the insulin receptor than surrounding normal tissue in which the receptor has been down-regulated in vivo (Lingohr et al., 2001). Increases in glycogen storage are associated with pathology similar to individuals that suffer from glycogen storage diseases. Such individuals are at very high risk for liver cancer at young ages (Limmer et al., 1988; Narita et al., 1994). Individuals with diabetes mellitus are also at greater risk for liver cancer than the general population (Wideroff et al., 1997). Other findings support the hypothesis that effects of DCA in the liver are not related to direct interactions with DNA. Prominent among these is the suppression of DNA methylation that occurs with DCA treatment that can be reversed by increasing methionine in the diet (Pereira et al., 2004). Similar effects on DNA methylation have been demonstrated for DBA (Tao et al., 2004). Suppressed DNA methylation is passed to daughter cells during mitosis. Elcombe, 1985 first suggested that the carcinogenic activity of TCA was dependent upon its ability to induce peroxisome synthesis. Indeed a close correlation does exist between this response and hepatic carcinogenesis in mice (DeAngelo et al., 1989; Bull et al., 1990a; DeAngelo et al., 1991) and is consistent with the binding of TCA to the PPAR-a receptor (Maloney and Waxman, 1999). Tumors induced by TCA carry a phenotype typical of peroxisome proliferators (Pereira, 1996; Stauber and Bull, 1997; Bannasch et al., 1997). Avariety of other differences has been noted in the phenotypes of the tumors that are induced by DCA and TCA (Latendresse and Pereira, 1997). Stauber et al. (1998) were able to show that the anchorage independent colonies resulting form treating primary B6C3F1 hepatocytes displayed the c-Jun negative phenotype typical of peroxisome proliferators. The media concentrations required for this effect fall into the same range as blood concentrations that are required for the induction of peroxisome synthesis in vivo (DeAngelo et al., 1989). Both DCA and TCA have been shown to be very effective tumor promoters in the liver of B6C3F1 mice (Pereira and Phelps, 1996; Pereira et al., 1997; Bull et al., 2004). The phenotype of tumors in female mice whose growth was promoted by DCA was overwhelming eosinophilic, whereas those promoted by TCA were basophilic. These data provide strong evidence that much, if not all of DCA and TCA’s activities as carcinogens are as a tumor promoters. However, the available evidence indicates that the two chemicals select different phenotypes of initiated cells. This is supported by the observation that DCA and TCA chemicals appear to be additive at low doses, but are mutually antagonistic as primary carcinogens (Bull et al., 2002) and as promoters (Bull et al., 2004) at higher doses. There are several studies that suggest that the DCA is mutagenic (reviewed by Moore and Harrington-Brock, 2000; IARC, 2004). Some of the positive data available are difficult to interpret. At the same time there is a growing body of negative data. The most serious difficulty is the excessive concentrations that have used for in vitro or bacterial assays and the high doses that are required to observe positive responses in vivo. Finally, the design of in vivo studies that ostensibly provide evidence of in vivo mutagenic effects have been designed in such a way to make it very difficult to differentiate between mutagenesis and selection. For example increased recovery of mutant cells from the lacI transgenic mouse was increased when mice had been treated with 1 or 3.5 g/L for 60 weeks, but not at shorter time intervals or at lower dose rates (Leavitt et al., 1997). While the authors took care to be sure that nodules and tumors were excluded from the sampling, Stauber and Bull (1997) demonstrated that there are numerous lesions that are smaller than nodules in B6C3F1 mice maintained on 2 g/L DCA for only 40 weeks. It was inevitable that these smaller lesions were included within the tissue samples described. Given the marked stimulation of cell replication that occurs within lesions in mice, it is not possible to determine if the effect reported by Leavitt et al. (1997) is due to weak mutagenic effects of DCA or its demonstrated ability to

CARCINOGENIC BY-PRODUCTS OF DISINFECTANTS

163

selectively stimulate the growth of tumor phenotype. A review of these data and other mode of action data, have concluded that DCA is likely to be acting primarily, if not entirely, by a nongenotoxic mechanism (Moore and Harrington-Brock, 2000; Bull, 2000; IARC, 2004). Austin et al. (1996) did find that DCA, when administered acutely in high doses, could increase the proportions of 8-OH-dG to 2-deoxyguanosine in nuclear DNA of the mouse liver. However, significant elevations could not be detected in the liver of mice maintained on drinking water containing 2 g DCA/L (Parrish et al., 1996). The ability to induce oxidative damage to nuclear DNA acutely in the liver of mice was also higher with BCA and DBA than DCA. BDCA was also more effective than TCA in producing this damage (Austin et al., 1996). Bromine substitution on the dihaloacetates appears to increase the probability that they possess mutagenic effects. In the Salmonella fluctuation assay and the SOS chromotest (Giller et al., 1997), DBA was 10–100 times as potent as DCA. MCA was negative but MBA was positive. The small amount of oxidative damage to DNA does not appear to play a big role in the induction of liver cancer based on the fact that none of the brominated analogs are more potent than DCA as carcinogens. However, the oxidative stress induced by the brominated haloacetates may contribute to carcinogenic effects in other target organs. Despite suggestions of weak mutagenic activity, the hepatocarcinogenic activity of the dihaloacetates seems to be associated with the disturbances in insulin signaling. Although other aspects of the mode of action of brominated dihaloacetates have not been studied explicitly, the accumulation of glycogen with all members of this class at least suggests a similar mode of action in the liver (Kato-Weinstein et al., 1998). They are approximately equivalent in potency at low doses, but the short latency of liver tumor induction by concentrations of DCA in water of 2 g/L and above appear to be specific to DCA. As indicated earlier, these high doses of DCA also cause glycogen to be deposited in the liver much more rapidly (Stauber et al., 1998; Kato-Weinstein et al., 1998). Thus, the dihaloacetates probably all act by providing a selective advantage to colonies of altered cells present within the liver. BDCA also induces glycogen accumulation in the liver of mice, but this is delayed for 3–4 weeks. It is possible that the slower accumulation of glycogen might be attributed to low concentrations of DCA produced from its metabolism with prolonged treatment (Merdink et al., 2001). TCA actually reduces hepatic glycogen (Kato-Weinstein et al., 2001), another reflection of its differing mode of action. It is also consistent with the much slower rates of TCA metabolism relative to that of BDCA (Merdink et al., 2001). 6.8.2.5 MX The strong mutagen, 3-chloro-4-(dichloromethyl)-5-hydroxy-2(5H)furanone (MX), was shown to be carcinogenic in male and female Wistar rats when administered in drinking water at concentrations that resulted in average daily doses of 0.4, 1.3, 5.0 mg/kg/day (Komulainen et al., 1997). The most prominent tumor site was follicular adenoma and carcinoma of the thyroid at all doses in both sexes. At the highest dose, both males and females were also found to have cortical adenomas of the adrenal gland. Adenocarcinoma and fibroadenoma of the breast, cholangiomas and adenomas of the liver and lymphoma and leukemia were also found in female rats. Males were found to have an increased incidence of alveolar and bronchiolar adenomas and adenomas of the Langerhans cells of the pancreas only at the high dose. These data identify MX as the most potent carcinogen identified to date in drinking water. Its concentrations were originally estimated to be quite low, but more recent surveys have found concentrations at concentrations as high as 0.31 mg/L (Weinberg et al., 2002). Moreover, a large number of related compounds have been observed in the same range (Table 6.2). Combined with its apparent potency, the halofuranone class of DBPs appears more important than it has in the past.

164

DRINKING WATER DISINFECTION BY-PRODUCTS

6.8.2.6 Nitrosamines As can be seen in Table 6.7, DBPs of the nitrosamine class are orders of magnitude more potent than other DBPs that have been shown to be carcinogenic. There is a massive literature based upon studies of the mechanisms by which lower molecular weight nitrosamines produce cancer. Metabolites of all the nitrosamines listed interact directly with DNA and result in mutations in daughter cells. As a consequence, the nitrosamines identified as DBPs are considered “genotoxic” carcinogens and their dose response is considered linear with dose (IARC, 1978, 1985). As DBPs, nitrosamines are generally found at very low concentrations (510 ng/L) in systems that employ chloramination. Therefore, the likelihood that the identified nitrosamines contribute significantly to widespread cancer risks of the magnitude suggested by epidemiological studies of chlorinated water is low. There have been unusual cases that have been documented, so it is important that the occurrence of this group of DBPs should not be ignored. A main concern is whether higher molecular weight secondary amines in NOM could form nitrosamines that have yet to be detected. 6.8.2.7 Miscellaneous By-Products Members of other classes of chlorination byproduct have only received cursory study as carcinogens. Several by-products of chlorination that have been shown to be carcinogenic are produced in such small concentrations to be of no concern (e.g., 2,4,6-trichlorophenol) or are of questionable importance as carcinogens by the oral route of exposure (formaldehyde and acetaldehyde). There is little question that inhalation of formaldehyde is a carcinogenic hazard at high dose levels (Kerns et al., 1983; Johannsen et al., 1986) In addition to being carcinogenic, formaldehyde is clearly genotoxic (Ma and Harris, 1988). Although one study which exposed rats to formaldehyde in drinking water indicated an increased incidence of leukemias (Soffritti et al., 1989), two more complete studies failed to produce such a response in other strains of rat (Til et al., 1989; Tobe et al., 1989). Even if the Soffriti studies correctly reflect sensitivities of humans to formaldehyde, the small concentrations of formaldehyde contribute very little to calculated risks of cancer attributed to identified chlorination by-products (Bull and Kopfler, 1991). There is no evidence that acetaldehyde is a carcinogen by the oral route of administration despite some evidence that it produces nasal tumors when inhaled (Woutersen et al., 1984; Woutersen and Feron, 1987). There are incomplete datasets for many DBPs, making evaluation of their potential importance difficult. As with TCA, the data implicating chloral hydrate as a carcinogen is not convincing (IARC, 2004). Several other chemicals that have been identified in drinking water have been shown to act as tumor initiators in the mouse skin when applied topically. Among these are chloroacetonitrile, bromochloroacetonitrile, dibromoacetonitrile (Bull et al., 1985), 2-chloropropenal, 2-bromopropenal and 1,3-dichloropropanone (Robinson et al., 1989). Trichloroacetonitrile has been reported to induce micronuclei in the gastrointestinal tract of mice (Lin et al., 1992). A number of these chemicals are generally directacting mutagens in Salmonella or they induce sister chromatid exchange in cultured mammalian cells (Bull et al., 1985; Meier et al., 1985). Dichloroacetonitrile has been shown capable of binding to nucleic acids in vitro; the other HANs are capable of alkylating nucleophiles (Daniel et al., 1986). The chemicals listed are but a small fraction of chemicals within some of these classes that are produced with chlorination (Richardson et al., 1999b). However, their concentrations in drinking water are generally 55 mg/L, with many of them occurring at even lower concentrations (Table 6.2). More recently, Plewa et al. (2004) reported formation of mixed bromo- and chloro-nitromethanes in chlorinated drinking water. Some of these compounds are strong mutagens, but also occur at concentrations below

EFFECTS OF DISINFECTANTS AND THEIR BY-PRODUCTS ON REPRODUCTION

165

1 mg/L. In the absence of in vivo data, it is difficult to assess the probability that these compounds are carcinogens and impossible to assess their potency. The limited data in animals do support epidemiological findings that the use of reactive chemicals to disinfect drinking water results in the formation of chemicals that are carcinogenic. However, as specifics of the toxicological results are brought to bear, it is difficult to conclude that these data provide strong support for the epidemiological findings. There are two problems. The first is that the target organs identified with most of the by-products that have been studied in animals are predominately the liver and kidney, whereas bladder cancer is most consistently seen in epidemiological studies. The second and more important problem is that those DBPs whose carcinogenic activity has been established in animals occur at concentrations that are two orders of magnitude too low to contribute to a risk as large as seems apparent in epidemiological studies, based upon their carcinogenic potency in animals. Even accepting linear extrapolation of the animal data for low-dose assessments for humans, these compounds would not significantly contribute to a cancer risk of the magnitude estimated from epidemiological data. Studies of the modes of action of DBPs in the limited number of classes that have been studied argue for a lower, rather than a higher, probability that these compounds are causally involved in risk of the magnitude observed epidemiologically. Therefore, if chlorinated water poses a carcinogenic risk, its cause is unlikely to be related to the effects of DBPs that have been studied in animals to date. 6.9 EFFECTS OF DISINFECTANTS AND THEIR BY-PRODUCTS ON REPRODUCTION 6.9.1

Inorganic By-Products

6.9.1.1 Chlorine Dioxide Development of the central nervous system is the key concern with chemicals that modify thyroid function (NRC, 2005). Treatment with chlorine dioxide during the immediate postnatal period produced delayed exploratory behavior in 16- to 20-day-old rat pups (Orme et al., 1985) and decreased cell numbers in the cerebellum and forebrain in 11-day-old animals (Taylor et al., 1985). It was assumed that these effects were secondary to the depressed serum thyroxine levels produced at the same doses (14 mg/kg/ day). In a second study, chlorine dioxide treatment was found to decrease forebrain weights, forebrain protein to weight ratio, and dendritic spine counts in Krieg’s area 18 in rat pups from postnatal days 21–35 (Toth et al., 1990). However, the effects in the latter study were not associated with changes in the serum concentrations of the thyroid hormones. 6.9.1.2 Chlorite A number of animal studies have been conducted to determine whether various disinfectants of oxyhalide by-products adversely affect reproduction. Carlton et al. (1987) examined the effects of 0, 1, 10, and 100 mg of sodium chlorite/L of drinking water administered for 14 days prior to mating and throughout the 10-day breeding period that was allowed to evaluate reproductive function in male Long-Evans rats. There was a significant increase in abnormal sperm morphology and a decrease in the progressive movement of sperm at 100 and 500 mg/L. Female rats were treated with concentrations of 0, 1, 10, and 100 mg/L throughout gestation and lactation. There was no sign of impaired reproductive function when males (treated as described above) were mated with these females. Decreases in the concentrations of triiodothyronine and thyroxine in blood were noted in both male and female rats treated with 100 mg/L. Depressions of thyroid hormone levels have not been associated with chlorite treatment in primates (Bercz et al., 1982).

166

DRINKING WATER DISINFECTION BY-PRODUCTS

Moore et al. (1980) reported that chlorite at a concentration of 100 mg/L reduced the conception rate, live births, litter weight, and the number of pups alive at weaning in A/J mice. A significantly reduced pup weight at weaning was interpreted as indicating that chlorite retarded growth rate. These changes were not related to decreased water consumption. A two-generation reproduction and developmental neurotoxicity study was conducted with sodium chlorite in drinking water at concentrations up to 300 mg/L (Gill et al., 2000). There was only mild evidence of effects on brain development. The authors indicated a NOAEL of 300 mg/L, but this included a small decrease in brain weight in postnatal development day (PND) 11 pups. Brain weight is generally well conserved even with decreased water consumption, and measurably smaller brain weights could be indicative of a delayed brain development. Although effects on brain weight disappeared at later times, measures of developmental delay that are reversible have been shown to result in deficits in CNS function later in life with other chemicals such as lead. In any case, it would appear that there is a NOAEL of 70 mg/L (10 mg of sodium chlorite/kg/day). There were no effects on thyroid hormone levels in rat pups at the 25th PND. Collectively, the available data in animals suggest that there are no significant reproductive hazards associated with chlorite as long as doses are kept well below concentrations that produce oxidative damage to erythrocytes. 6.9.1.3 Chlorate There appear to be no reproductive and developmental toxicological studies involving chlorate. Such a study could be of interesting considering the role that inhibition of iodide uptake appears to play in the induction of thyroid tumors that result from chronic chlorate treatment (Hooth et al., 2001; NTP, 2005a). 6.9.1.4 Bromate A one-generation study (Wolfe and Kaiser, 1996) comprehensively examined a number of reproductive endpoints of both male and female reproductive toxicity as well as early (PND 5) in the offspring. Aside from mild effect on epididymal sperm density in rats treated with 250 mg/L of drinking water, no effects were observed that appeared treatment related. Therefore, the existing data suggest bromate is not a reproductive or developmental toxicant. 6.9.2

Organic By-Products

6.9.2.1 Trihalomethanes The NTP conducted short-term (i.e., single generation) reproductive and developmental toxicity studies of BDCM (NTP, 1998b) and DBCM (NTP, 1996). The design of these studies allowed separate evaluation of exposure during the peri-conception period and gestation at concentrations of the two THMs ranging from 50 to 1300 mg/L in the drinking water of Sprague–Dawley rats. Some liver damage was observed in males treated with BDCM at 700 and 1300 mg/L and some indication of increased rates of cell division in females treated with 1300 mg/L. DBCM produced mild liver damage at doses of 50 mg/L and higher in males. Neither compound induced any adverse effects on male or female reproductive function. Studies in which BDCM was administered to F344 rats by aqueous gavage at 75 mg/kg/ day between gestational days (GD) 6–10 resulted in a high incidence of full litter absorption, but Sprague–Dawley rats administered the doses of up to 100 mg/kg/day maintained their litters (Bielmeier et al., 2001). In a different study, BDCM was administered in drinking water from GD 6–21 to ClrSD rats (up to 82 mg/kg/day) without significant effect (Christian et al., 2001) aside from reduced body weight due to decreased water consumption and some

EFFECTS OF DISINFECTANTS AND THEIR BY-PRODUCTS ON REPRODUCTION

167

reversible developmental delays. The former study was performed in context of the epidemiological studies that associated BDCM with increased incidence of spontaneous abortion (Waller et al., 1998). It is clear that there are substantial strain differences in responses that are of importance. However, the doses in the F344 rats exceeded the doses that would be calculated in the epidemiological study by a factor of 100,000 (i.e., a fraction of a mg/kg/day versus 75 mg/kg/day). Therefore, even the positive data do not strongly support the epidemiological findings. The effects of BDCM have been explored in human placental trophoblast cultures. Decreases of chorionic gonadatropin (CG) secretion were observed at concentrations as low as 20 nM in the absence of evidence of cytotoxicity (Chen et al., 2003). In a subsequent paper (Chen et al., 2004) effects of BDCM on trophoblast differentiation was examined and approximately one thousand times the concentration was required to produce effects. Effects on trophoblast differentiation would be considered an adverse effect, whereas small decrements in CG secretion would not be considered adverse without evidence of harmful effects. While the concentrations used in these studies are low compared to other in vitro studies, they are much higher than the highest concentrations obtained from showering with chlorinated water (20 mM versus 0.5 nM). 6.9.2.2 Haloacetates DCA does produce testicular toxicity when administered at high doses in drinking water. These effects were first noted in studies of the general toxicity of DCA (Katz et al., 1981) and were discussed earlier. Cicmanec et al. (1991) followed up on these original observations in dogs and detected degeneration of testicular epithelium and syncytial giant cell formation at doses as low as 12.5 mg/kg. Toth et al. (1992) conducted an examination of DCA’s ability to modify male reproductive function in Long-Evans rats. Reduced weights of accessory organ (epididymus, cauda epididymus and preputial gland) weights were observed at doses as low as 31 mg/kg/day for 10 weeks. Epididymal sperm counts were found to be depressed and sperm morphology increasingly abnormal at doses of 62 mg/kg and above. These latter effects were accompanied by changes in sperm motion analysis. Fertility was tested in overnight matings and was found to be depressed only at the highest dose evaluated 125 mg/kg/day. Spermatotoxicity of DCA has been studied further by Linder et al. (1997b) who found effects at doses of 160 mg/kg/day. The authors found that the testicular toxicity of DCA in Sprague–Dawley rats increased significantly as the duration of exposure was extended in time. This progressive increase in testicular toxicity may be related to the reduced metabolic clearance of DCA that occurs as the result of pretreatment (Gonzalez-Leon et al., 1997). A series of papers establish that DBA is significantly more potent than DCA as a testicular toxicant in Sprague–Dawley rats (Linder et al., 1994a, 1994b, 1995). DBA produced mild effects on spermiation at doses as low as 10 mg/kg with repeated doses (Linder et al., 1994b). Effects were progressively more severe as doses increased to 30, 90, and 270 mg/kg (Linder et al., 1997a). With prolonged exposure, the highest dose becomes severely toxic. At lower doses (50 mg/kg) there were no significant effects on male fertility, although mating behavior may have been altered (Linder et al., 1995). The NTP conducted short-term (i.e., single generation) reproductive and developmental toxicity studies of BCA (NTP, 1998a), DBCA (NTP, 2000b), and TBA (NTP, 1998c) in Sprague–Dawley rats. Animals were treated with concentrations of the haloacetic acids that ranged from 30 to 1500 mg/L during the peri-conception and gestation periods. Hepatomegaly was observed with BCA at treatments of 600 mg/L. BCA also caused a decrease in the total number of implants and a 50% decrease in the numbers of live fetuses/litter at the

168

DRINKING WATER DISINFECTION BY-PRODUCTS

same treatment level. No soft tissue malformations were observed. DBCA produced no effects on reproductive function in females at treatments up to 1500 mg/L and no malformations were observed. DBCA did produce a 11% decrease in sperm velocity with this treatment level. TBA did not affect reproductive toxicity or produce general toxicity at the highest concentration employed, 400 mg/L. Christian et al. (2002) conducted a two-generation reproductive toxicity study of DBA in Crl:SD rats at concentrations of 50, 250, and 650 mg/L. Some testicular hypoplasia and epididymal malformations were observed in the high dose group and evidence of retained step 19 spermatids was observed at both 250 and 650 mg/L. The 50 mg/L had effects on liver size of the parenteral generation as well as in the offspring. Thus, reproductive and developmental effects appeared at doses higher than those that produced the recognized toxicity of the dihaloacetates to the liver. The data that are available indicate that the dihaloacetates at least appear to be both reproductive and developmental toxicants. However, the doses required to produce such effects are more than 50,000 times those that would be anticipated from chlorinated drinking water. 6.9.2.3 Haloacetonitriles The HANs appear to be the most specific reproductive toxins among chlorination by-products. Smith et al. (1987) found that dichloroacetonitrile and trichloroacetonitrile reduced fertility and early implantation loss in Long-Evans rats. Subsequent study demonstrated that trichloroacetonitrile, the most toxic member of this group, produced embryolethality and resorptions at doses of 7.5 mg/kg/day (Smith et al., 1988) A significant number of malformations was detected in the cardiovascular system in the same dose range and at doses considerably below those found to be maternally toxic. Dichloroacetonitrile did produce embryolethality at doses that were not maternally toxic, but produced soft tissue malformations only at doses that were maternally toxic (Smith et al., 1989a) Thus, trichloroacetonitrile may well be teratogenic, but the doses expected from drinking water are still less than 1/1000 of those required to produce these effects in rats. Short-term (i.e., single generation) reproductive and developmental toxicity studies were conducted with bromoacetonitrile (NTP, 1996) and dibromoacetonitrile (NTP, 1997b) by the NTP in Sprague–Dawley rats. Concentrations of bromoacetonitrile in drinking water of the animals were 5, 30, and 100 mg/L. Bromoacetonitrile produced effects suggestive of a reproductive toxicant at 100 mg/L (i.e., preimplantation losses and resorptions), but these were not statistically significant. It also produced evidence of mild renal damage in males at this same treatment level. Dibromoacetonitrile was taste aversive at 50 mg/L, but produced no evidence of reproductive or developmental toxicity even at the highest treatment concentration. While there is some evidence that the HANs are reproductive toxicants at high doses, they appear to pose no risk at the concentrations found in drinking water. Therefore, the available data do not support a hypothesis that the HANs are likely to account for associations of chlorinated water with adverse reproductive outcomes in epidemiological studies.

6.10 EFFECTS ON DEVELOPMENT 6.10.1

Inorganic By-Products

6.10.1.1 Chlorite Harrington et al. (1995b) examined the developmental toxicity of chlorite in New Zealand white rabbits. The rabbits were treated with 0, 200, 600, or 1200 mg sodium chlorite/L in their drinking water from GD 7 to 19 of pregnancy. The

EFFECTS ON DEVELOPMENT

169

animals were necropsied on GD 28. There were no dose-related increases in defects identified. Minor skeletal anomalies increased with increasing concentration of chlorite in water and food consumption was depressed. A two-generation reproductive and developmental toxicity study was described in the section dealing with reproductive effects of chlorite (Gill et al., 2000). The only developmental effect observed in the study was a small decrease in brain weight on PND 11 in animals consuming 300 mg sodium chlorite/L. There was no evidence of developmental toxicity due to chlorite at doses less than 10 mg/kg/day. Concentrations of chlorite in drinking water can approach the maximum contaminant limit (MCL) of 0.8 mg/L in systems utilizing chlorine dioxide as the disinfectant. This would contribute as much as 10 mg/kg/day to a child, a factor of more than one thousand times relative to the doses that produced the effect on brain weight. 6.10.2

Organic By-Products

6.10.2.1 Trihalomethanes Schwetz et al. (1974) noted some minor malformations that are associated with nonspecific delays in development in Sprague–Dawley rats exposed to high levels as chloroform vapor (30, 100 and 300 ppm). At the highest two doses there was clear evidence of embyrolethality. In terms of systemic dose, these levels are orders of magnitude higher than exposures that are encountered in drinking water. Therefore, reproductive and developmental effects by chloroform at doses that are not overtly toxic to the mother are unlikely. Reproductive and developmental toxicity studies conducted on BDCM and DBCM by the NTP were discussed in the section on reproductive effects. These studies found no evidence of malformations. 6.10.2.2 Haloacetates Chlorinated acetates appear to be true developmental toxicants. TCA produced malformations in the cardiovascular system of Long-Evans rats at doses of 330 mg/kg and higher. Skeletal malformations were observed at doses of 1200 mg/kg, but these doses were maternally toxic (Smith et al., 1989b). DCA treatment also produced levocardia and malformations between the ascending aorta and right ventricle at doses of 140 mg/kg and above (Smith et al., 1992). Urogenital defects (bilateral hydronephrosis and renal papilla) and defects of the orbit were observed. While not grossly toxic, 140 mg/kg did produce the enlarged liver typical of DCA. Higher doses (400 mg/kg) resulted in weight loss and progressively more severe toxicity. Epstein et al. (1992) attempted to more finely define the period of susceptibility to DCA-induced aorta/ ventricular septal defects. Very high single doses, 2400 and 3500 mg/kg, were required to produce effects. An increasing severity of developmental effects was observed when lower doses were administered for several days, suggesting that this progressive increase in toxicity is probably related to the inhibition of metabolism of DCA that occurs with prior treatment (Gonzalez-Leon et al., 1997). The evidence indicates that DCA is teratogenic in rats, but the doses required to produce these effects were five orders of magnitude higher than those encountered in chlorinated drinking water. The short-term reproductive and developmental toxicity studies conducted of selected HAs conducted by the NTP were discussed in the section on reproductive toxicity. Although BCA appears to be a reproductive toxicant, no malformations were noted. The trihaloacetates examined were much less active as reproductive toxicants and the available data suggest that they even less likely to be developmental toxicants.

170

DRINKING WATER DISINFECTION BY-PRODUCTS

The quantitative structure–activity relationships (QSAR) involved in the developmental toxicology of the HAs were studied in whole embyro culture (Richard and Hunter, 1996). Strong associations of malformations with the lowest unoccupied molecular orbital and acid dissociation constants were found. While these relationships may not hold for in vivo potency of the class because of intervening mechanisms involved in their metabolism and disposition, they do suggest some commonality in the mechanisms involved in developmental toxicities. Hunter et al. (2006) extended observations of developmental toxicity in whole mouse embryo culture to include BCA, BDCA, and BDCA. Benchmark concentrations (concentrations that were estimated from the data to produce a 5% increase in neural tube dysmorphogenesis were 63, 500, and 536 mM for the three compounds, respectively. These data fit well with QSAR analysis in this and the prior study. The results also appear to be consistent with the relative doses of these compounds that produced effects in vivo (NTP, 1998a, 1998c, 2000b). These data all support the conclusion that the dihaloacetates are much more potent as reproductive and developmental toxicants than the trihaloacetates. However, the systemic concentrations that are produced by carcinogenic doses of DCA are 51 mM (Merdink et al., 1998b) and require concentrations of 500 mg/L, raising questions about how meaningful these results are at concentrations of the HAs that are encountered in chlorinated drinking water. A study of mixtures of the dihaloacetates indicates that their effects are additive in whole rat embryo culture (Andrews et al., 2004). However, fewer neural tube defects were observed at equivalent concentrations in rat embryo culture than in the mouse embryo culture. While the dihaloacetates appear to be both reproductive and developmental toxicants, the concentrations and/or doses required to produce such effects suggest that they are not likely to be contributing to the adverse reproductive outcomes observed in epidemiological studies. 6.10.2.3 Haloacetonitriles Descriptive data on the representatives of the HAN class that have been evaluated for developmental toxicities are provided in the section dealing with reproductive toxicities. Mechanisms that are involved in the effects of this class of DBPs have not received significant study. As a consequence, it is not possible to refine the evaluation of whether these chemicals are likely to pose a significant hazard at concentrations found in chlorinated drinking water. Based on the dose–response information available from the descriptive data, mentioned previously, it is improbable that the HANs contribute to adverse reproductive or developmental outcomes that have been inconsistently associated with chlorinated drinking water in epidemiological studies.

6.11 BY-PRODUCTS OF POTENTIAL INTEREST Risks that appear to be associated with chlorinated drinking water are not readily accounted for by the toxicological data on regulated DBPs. To address this issue, efforts have been expended to predict the toxicological properties of the large number of DBPs for which directly applicable experimental toxicological data are lacking. The first of these efforts utilized the TOPKAT (Accelrys, 2001) quantitative structure–toxicity relationships (QSTR) program to identify likely carcinogenic and developmental toxicants among recognized DBPs (Moudgal et al., 2000). Among the 252 compounds evaluated, the carcinogenicity submodels identified aliphatic aldehydes and acids as being of potential importance. Developmental toxicity was predicted to be associated with most carboxylic acids, aliphatic halogenated ketones, and HANs. This study made no effort to prioritize these

BY-PRODUCTS OF POTENTIAL INTEREST

171

compounds by potency and/or doses that might be expected to arise from consuming disinfected drinking water. Woo et al. (2002) used a mechanism-based structure–activity approach for prioritizing over 200 DBPs based upon their carcinogenic potential. DBPs were ranked as highly likely to be potent multispecies, multitarget carcinogens at low doses (H), likely to be an active multispecies/target carcinogen at moderate doses (HM) and several other categories of lower priority, including unlikely to be carcinogenic (L). As doses of DBPs obtained from drinking water are generally quite low, only predictions in the first two categories were considered relevant for this review. Three compounds related to MX were identified in the HM category. These were 3-bromo-4-(dibromomethyl)-5-hydroxy-2(5H)-furanone, 3-chloro-4-(bromochloromethyl)-5-hydroxy-2(5H)-furanone, and 3-chloro-4-(dibromomethyl)-5-hydroxy2(5H)-furanone. A third effort included analyses of DBPs that could be predicted to occur based upon likely reactions between chlorine and organic substructures known to occur in NOM found in water (Bull et al., 2006). These chemicals were evaluated, along with other established DBPs, with an updated version of TOPKAT (Accelrys, 2001). Additional evaluations of related compounds from the literature that were not within the optimal predictive space of the models or lacked related chemicals in the program’s training sets. A set of decision criteria were established to prioritize chemicals based on predicted chronic LOAELs, probability that they were carcinogenic, developmental toxicants, or Ames’ test mutagens. The most significant of these additional criteria was a predicted chronic LOAEL of 51 mg/kg/day and a positive prediction of carcinogenic activity in at least one sex of two species for a compound to be rated as high priority for additional research. Several compounds were identified with predicted chronic LOAELs as low as 1 mg/kg/day and among these were chemicals that were predicted to be carcinogens. Prominent among these were a series of halogenated quinone derivatives, some of which might be plausible causes of bladder cancer based on their probable metabolism (e.g., 2,3,5-trichloro-3-methyl-1,4-dibenzoquinone and quinones with fewer halogens) (Bolton et al., 2000; Lau et al., 2001). A number of cyclopentene derivatives with some resemblance to MX were identified (e.g., 3,5-dichloro-1-hydroxycyclopent-2enoic acid). The identification of this group reinforces the assignment of a high priority to assigned to brominated halofuranones related to MX by Woo et al. (2002) and the finding of higher concentrations of these chemicals in chlorinated drinking water than had been previously documented (Weinberg et al., 2002). While levels of common nitrosamines have been found to be in the low ng/L range in most chloraminated waters (Charrois et al., 2004), little attention has been paid to formation of these compounds with alkaloidal precursors that possess secondary amine substructures such as a pyrrolidine ring (nicotine-like compounds) that might arise from the vegetable matter that makes up NOM. Mitch and Sedlak (2002b) have shown that chloramine does nitrosate pyrrolidine. Finally, there are several well established DBP classes for which no toxicological information has been generated other than bacterial mutagenicity assays. These include organic N-haloamines and halonitriles other than the HANs. The latter compounds were routinely predicted to be developmental toxicants, as well as being carcinogenic. The absence of data to establish an approximate order of potency for these halonitriles, made it difficult to determine their priority. These efforts to identify more probable causes of some of the health risks that have been associated with the chlorination of drinking water epidemiologically do not establish that such compounds exist, especially at concentrations sufficient to be of concern. They do clearly demonstrate that there are compounds with predicted potencies sufficient to account for some of the adverse effects that have been observed, if they are present in disinfected

172

DRINKING WATER DISINFECTION BY-PRODUCTS

drinking water. The results suggest that broadening of the research agenda beyond the regulated DBPs will be important to resolving whether or not chlorinated drinking water presents risks as high as have been suggested by some epidemiological studies.

6.12 SUMMARY AND CONCLUSIONS The volume of health research that has focused on DBPs has expanded exponentially in the past decade. Only a smattering of that research could be touched upon, so the author apologizes for omitting many important papers in making editorial decisions needed to develop a cohesive analysis of where the issue now stands. There have been more than 25 analytical epidemiological studies examining associations of chlorinated water with various cancers. The association with bladder cancer has been strengthened with time. It is perhaps too early to conclude that this apparent increase in risk has been established without question, in part because of the relatively low odds ratios that are commonly found. Nevertheless, bladder cancer risk appears to be the major risk associated with chlorinated water, approximately 1/1000 in a lifetime. Based upon laboratory-based research there appears little to distinguish risks from chlorination of water with treatment with ozone, chlorine dioxide or chloramine. A number of disinfectant by-products are capable of inducing cancer in animals with each of these treatments. The potencies of these compounds are such that in combination they present 1/10–100th of the risk from chlorinated water that is suggested by epidemiological data, but only if one accepts that the toxicological data can be linearly extrapolated to low doses. Current regulations appear to adequately protect against chemical hazards with disinfection, although there is a need to develop an MCL for chlorate based on recent studies. In considering the regulated chlorination by-products, it is becoming increasingly clear that the two major classes of DBPs act largely, if not entirely, by mechanisms for which linear extrapolation to low doses is inappropriate. The concentration of animal research on these few products as opposed to attempting to identify DBPs that plausibly might be more responsible, has had the effect of broadening rather than narrowing the differences in risk estimates from epidemiological and toxicological studies. There has been even greater effort devoted to determining whether various forms of drinking water disinfection may be associated with reproductive and developmental toxicities. Collectively, it is difficult to assemble a coherent view of these risks. In large part this is because the results of these studies are contradictory in their specific findings. In other cases, it is difficult to place much weight on outcomes that have only been observed in a single study. The reader is referred to Graves et al. (2001) for a more detailed critique of the earlier studies. In this area, it seems odd that efforts have not focused more clearly on the oxyhalide ions because of their potential effects on the thyroid and because these are by-products that usually occur at much higher concentrations than organic by-products. Those studies that have addressed the exposure to the oxyhalide anions have been conducted chiefly in Europe where mixed use of disinfectants is more common than in the U.S. The formation of compounds that are truly reproductive and developmental toxicants has also been amply demonstrated in animal studies of individual DBPs. It appears that such effects can be produced, especially with the dihaloacetate group of DBPs which do appear to affect sperm maturation at relatively low doses. The HANs appear to have such effects. The doses that have beenrequiredtoproducethese effectsinrodentaremuchtoohightoaccountforthelevelof risks that have been suggested by some epidemiological studies (and denied by others).

SUMMARY AND CONCLUSIONS

173

There have been studies of the reproductive and developmental toxicities of chlorite that indicate some effects that have been variously associated with effects on the thyroid. Bromate has been studied and found to be without significant reproductive and developmental effects. Neither of these anions would be expected to be as effective as an antithyroid agent as chlorate, based simply on the size of the chlorite anion and the reactivity of bromate. However, based on the chronic study of chlorate, the minor depression of thyroid hormones and perturbations in TSH, the doses required to produce these effects are much too high to account for epidemiological findings. There are issues of human sensitivity and interactions that should be considered. This usually triggers concern over the very young or very old, but genetic differences are much more likely to be of substantive importance. By definition, epidemiological studies are going to identify sensitive members of the population. Epidemiological research also studies the intact mixture in species of most concern and must continue to play a role in prompting technological shifts to reduce harm of current practice. However, epidemiological studies are necessarily conducted after the fact. For this reason they cannot contribute to identifying risks with untried disinfection strategies. As a consequence, predictive chemical and toxicological methods need to assume an important role in providing direction for the future. Preliminary efforts using these methods have identified 23 chemicals that are likely to be formed with chlorine that would be 4100 times the potency of regulated DBPs. These methods are still crude, especially if attempting to quantify risk. However, if they are properly coupled to appropriately directed chemical and toxicological testing, their utility will evolve quickly. Movement away from the use of chlorine as a disinfectant cannot be treated simply. First, the simplicity of chlorination as a process allows it to be employed in systems of all sizes with minimal training of plant operators. Second, toxicological data indicate the possibility of cancer risks with alternate methods of disinfection. Finally, the fact that ozone and physical methods (e.g., ultraviolet light) do not leave stable residuals needed to maintain the microbial quality of water during distribution, means that ozone treatment will commonly be followed by the addition of chlorine or chloramine. Studies do indicate that bladder cancer risk appears to be decreased with the use of chloramine or ozone. On the contrary, animal studies identify carcinogens produced by these disinfectants that are more likely to cause cancers at sites other than the bladder (low molecular weight nitrosamines and bromate, respectively). In addition, structure–activity analyses of predicted by-products suggest the possibility that some additional potent carcinogens (haloquinones, higher molecular weight nitrosamines) may be produced with chloramination. The study of Chevrier et al. (2004) deserves some special attention because of an apparent reduction of bladder cancer risk that was seen when municipal supplies converted to ozone disinfection. Remarkably this reduction was observed even though chlorine continued to be used as a secondary disinfectant. Although firm conclusions should not be made based upon a single epidemiological study, these data suggest that the precursor of the putative bladder carcinogen was destroyed by ozone. Clearly, this provides direction for future research and could potentially enhance the benefits that could accrue from switching to ozone for primary disinfection. It is fortunate that the revival of concerns over the microbiological contaminants of drinking water has made it apparent why decisions should not be without a full evaluation of all the risks that are associated with changes in treatment processes. The critical role that disinfection plays in the delivery of safe water makes it necessary to understand the health risks involved with more clarity than is usually required for regulatory action. Any changes

174

DRINKING WATER DISINFECTION BY-PRODUCTS

should be on a path that clearly reduces health risks, both microbial and/or chemical, before the large investments that are involved in the continual process of expansion and upgrading of municipal and other public water systems can move forward with confidence. As long as drinking water requires disinfection, it is important that a solid scientific basis is developed to identify both the risks that may be decreased and those that may increase by using one reactive chemical in the place or in addition to another. In conclusion, evidence that there may be some adverse health impacts associated with the use of chemical disinfectants continues to grow. At this point in time, the only substantive risk that appears to emerge as being associated with chlorination of drinking water is an increased risk of bladder cancer in men. It is important that the chemicals causing this effect be identified, not simply to satisfy scientific curiosity, but to define alternatives that reduce or eliminate the risk. Second, there needs to be a more organized effort for identifying the potential hazards that might be uniquely associated with each disinfectant considered for broad application. Disinfection strategies are increasingly broadening to include the use of mixed disinfectants. In one case, ozonation followed by chlorination appears to reduce bladder cancer risk, but that evaluation did not examine the possibility of other risks that might be introduced. The scientific approach to this problem must supplant much of the rather narrowly focused testing of individual by-products and mixtures in a plethora of shortterm assays that has characterized this field in the past. The problem might be more efficiently approached using predictive chemical and toxicological methods coupled with experimental studies that provide a basis for estimating in vivo toxicological potency, not simply toxicological potential.

GLOSSARY Chloramine. A chemical or group of chemicals formed from mixing chlorine and ammonia in water. Depending upon the relative amounts of chlorine and ammonia and pH, the actual chemical form can be monochloramine (conditions optimized for this form), dichloramine or trichloramine. Chloramine is frequently used as a residual disinfectant as it is more stable than hypochlorous acid or hypochlorite. Also referred to as combined chlorine. Chloramination. Disinfection of water with monochloramine. Chlorination. Disinfection of water by adding chlorine gas or solutions of hypochlorite salts. Chlorine. In water this refers to the products of hydration following the introduction of chlorine into water. Depending upon pH, the major forms are hypochlorite and hypochlorous acid which are in an equilibrium controlled by the pH of the water. Chlorine dioxide. A disinfectant based on a dioxide of chlorine (i.e., ClO2). It possesses an unpaired electron and is unstable as a gas, so it must be generated on site in water. Cross-connections. Inadvertent connections of service pipes and mains intended to carry drinking water with counterparts intended for other purposes (i.e., conveyance of sewage). Disinfection. A single or set of processes employed to eliminate or control the growth of organisms in drinking water. Primary disinfection includes processes employed early in the treatment of drinking water that is intended to inactivate all infectious agents that might have contaminated a

GLOSSARY

175

water that is utilized as a source of drinking water. Disinfectants that are rapidly lethal (i. e., with the shortest contact time) are most desirable (ozone, chlorine, or chlorine dioxide). A secondary intent of these processes is to control microbial fowling of equipment in the treatment plant. Secondary disinfection is intended to control the regrowth of organisms within the drinking water distribution system. Disinfectant is added as it leaves the treatment plant or in booster stations in the distribution system. The emphasis is to maintain effective concentrations of disinfectant throughout the distribution system. Within the distribution system there are inevitably areas of low flow or actual dead ends in where a residual disinfectant may need to be present for many days. Postdisinfection is another term for secondary disinfection. Disinfection by-products (DBPs) are reaction products of disinfectants with organic and inorganic chemicals that are present in the water being treated. In most cases the bulk of these chemicals are of natural origin (see natural organic material) or co-occur with chloride in brackish waters (e.g., bromide). However, manmade chemicals also react with disinfectants to produce DBPs. Distribution systems for drinking water includes the mains and pipes that deliver water from its source, with or without intervening treatment, to the consumers of the water. Finished water is a term that is applied to water that has been treated to drinking water specifications. HAA5 is a legal term that applies to the five haloacetates that are regulated under the Safe Drinking Water Act, monochloroacetate, monobromoacetate, dichloroacetate, dibromoacetate, and trichloroacetate. Haloacetates (Haloacetic acids, HAs) are a group of important disinfectant by-products in which halogens (most frequently chlorine or bromine) are substituted on the alpha carbon of acetic acid. Dihalo- and trihalo-forms are the most common. Monochloro- and monobromo-acetates are generally found at significantly lower concentrations. These compounds are more commonly referred to as haloacetic acids, but they exist as salts at pHs above 2, so it is more accurate to refer to them as haloacetates. Haloacetonitrile are the third most prevalent by-products of chlorination, with halogen substitution occurring on the alpha carbon of acetonitrile. Chlorine and bromine are the most commons substituents. Hypochlorous acid is the hydrated form of chlorine gas in water (HOCl). Hypochlorite is the salt form of hypochlorous acid. The pKa of hypochlorous acid is approximately 7.5, so the predominate form of chlorine in water is a mixture of hypochlorous acid and hypochlorite. Interuterine growth retardation (IUGR) is determined by weight of full-term infants. It is most frequently used as a means of segregating growth retardation from low weight as a result of premature birth. The specific definition will vary among epidemiological studies because it is necessary to determine the time of conception and a criteria for measuring size. Low birth weight is operationally defined in an epidemiological study. In general, the definition will specify criteria for minimal gestational age and a weight that demarcates the boundary between normal and low birth weights. For example, one study identified infants 437 weeks that weighed 52500 g as having low birth weight.

176

DRINKING WATER DISINFECTION BY-PRODUCTS

Natural organic matter Natural organic matter is found in all waters. It is a complex mixture of nutrients and biological decay products. A prominent portion of the organic matter in surface waters is made up of humic and fulvic acids. MX is a commonly used abbreviation for a highly mutagenic chemical that is produced in the chlorination of drinking water. The proper chemical name is 3-chloro-4-(dichloromethyl)-5-hydroxy-2(5H)-furanone. Population attributable risk is the risk in a population with an exposure minus the risk of the unexposed members of the same population. Postnatal development is the age of an animal after birth in days. Preoxidation is the application of a oxidant chemical early in treatment to achieve a goal, one of which can be disinfection. Preoxidation is also used to control certain types of taste and odor or other aesthetic problems. Oxidants include chemicals that are used in disinfection, such as chlorine, ozone, or chlorine dioxide, as well as chemicals like potassium permanganate or hydrogen peroxide. Ozonation. A form of disinfection where ozone is hydrolytically on site to disinfect drinking water. Raw water is water as it is extracted from a source prior to treatment. Residual disinfectant is that disinfectant that remains in the water after “demand” has been satisfied. Demand consists of natural organic matter, ammonia and other reactants with a disinfectant. Some of the demand, usually a small fraction is accounted for by reactions with microorganisms that are present in the source water (see secondary disinfection). Source water is simply a body of surface water (rivers, lakes or reservoirs) or groundwater that serves as a source of drinking water. THM4 is a term used in this monograph to identify those THMs that are officially regulated as a sum of their individual concentrations. The regulated THMs are chloroform, bromodichloromethane, dibromochloromethane, and bromoform. Other THMs also are produced in the chlorination of drinking water and include all possible substitutions of iodine for the halogens identified in the THM4. The iodinated forms are generally found at much lower concentrations than the THM4. Total organic carbon is determined by the oxidation of organic material present in a water sample and measuring the amount of CO2. It is usually expressed as mg of carbon equivalents per liter of water. The CO2 that is present as carbonates is subtracted from the total analysis. This and related measures are simply ways of characterizing the amount of organic matter that might serve as precursors to disinfectant by-product formation. An alternative, but not precisely identical terms is dissolved organic carbon (DOC). Total organic halogen is a measure amount of halogen present in water that is substituted with a halogen on carbon. In general it is expressed in chlorine equivalents. Trihalomethane are methanes substituted in three of four possible positions on a single carbon. The halogens are primarily chlorine and bromine, although iodine can also be substituted if iodide and related salts are present in the treated water.

REFERENCES Abdel-Rahman MS (1985) Pharmacokinetics of chlorine obtained from chlorine dioxide, chlorine, chloramine, and chloride. In: Jolley RL, et al. , editor. Water Chlorination: Environmental Impact and Health Effects, Vol. 5. Chelsea, MI: Lewis Publishers, Inc., pp. 281–293.

REFERENCES

177

Abdel-Rahman MS, Couri D, Bull RJ (1980) Kinetics of ClO2 and effect of ClO2 and ClO3 in drinking water on blood glutathione and hemolysis in rat and chicken. J. Environ. Pathol. Toxicol. 3: 431–449. Accelrys (2001) TOPKAT User Guide. Version 6.1. San Diego, California: Accelrys, Inc. Acharya S, Mehta K, Rodriguez S, Pereira J, Krishnan S, Rao CV (1997) A histopathological study of liver and kidney in male Wistar rats treated with subtoxic doses of t-butyl alcohol and trichloroacetic acid. Exp. Toxic Pathol. 49:369–373. Aggazzotti G, Righi E, Fantuzzi G, Biasotti B, Rivera G, Kanitz S, Barbone F, Sansebastiano G, Battaglia MA, Leoni V, Fabiani L, Triassi M, Sciacca S (2004) Chlorination by-products (CBPs) in drinking water and adverse pregnancy outcomes in Italy. J. Water Health 2:233–247. Akin EW, Hoff JC, Lippy EC (1982) Waterborne outbreak control: which disinfectant? Environ. Health Perspect. 46:7–12. Alavanja M, Goldstein I, Susser M (1978) Case–control study of gastrointestinal and urinary tract cancer mortality and drinking water chlorination. In: Jolley RL, et al. , editor. Water Chlorination: Environmental Impact and Health Effects, Vol. 2. Ann Arbor, MI:Ann Arbor Science Publishers, pp. 395–409. American Cancer Society (2005) Cancer Facts [Online]. Available: http://www.cancer.org. [cited December 15, 2005]. Anderson WB, Liebler DC, Board PG, Anders MW (2002) Mass spectral characterization of dichloroacetic acid-modified human glutathione transferase zeta. Chem. Res. Toxicol. 15:1387–1397. Andrews JE, Nichols HP, Schmid JE, Mole LM, Hunter ES III, Klinefelter GR (2004) Developmental toxicity of mixtures: the water disinfection by-products, dichloro-, dibromo-, and bromochloro acetic acids in rat embryo culture. Rep. Toxicol. 199:111–116. Arai T, Kelly VP, Minowa O, Noda T, Nishimura S (2006) The study using wild-type and Ogg 1 knockout mice exposed to potassium bromate shows no tumor induction despite an extensive accumulation of 8-hydroxydeoxyguanosine in kidney DNA. Toxicology. 221:179–185. Aschengrau A, Zierler S, Cohen A (1993) Quality of community drinking water and the occurrence of late adverse pregnancy outcomes. Arch. Environ. Health 48:105–113. Austin EW, Bull RJ (1997) Effect of pretreatment with dichloroacetate or trichloroacetate on the metabolism of bromodichloroacetate. J. Toxicol. Environ. Health 52:367–383. Austin EW, Parrish JM, Kinder DH, Bull RJ (1996) Lipid peroxidation and formation of 8hydroxydeoxyguanosine from acute doses of halogenated acetic acids. Fundam. Appl. Toxicol. 31:77–82. AWWA (1982) Disinfection Committee Report. J. Am. Water Works Assoc. 74:376–380. Baker CC, Bursill DB (1990) South Australian water: a disinfection problem. In: Jolly RL, et al. , editor. Water Chlorination; Chemistry, Environmental Impact and Health Effects, Vol. 6. Chelsea, MI:Lewis Publishers, pp. 47–59. Ballmaier D, Epe B (1995) Oxidative DNA damage induced by potassium bromate under cell-free conditions and in mammalian cells. Carcinogenesis 16:335–342. Ballmaier D, Epe B (2006) DNA damage by bromate: mechanisms and consequences. Toxicology 221:166–171. Bannasch P, Klimek F, Mayer D (1997) Early bioenergetic changes in hepatocarcinogenesis: preneoplastic phenotypes mimic responses to insulin and thyroid hormones. J. Bioenerg. Biomembr. 29:303–313. Barton HA, Bull R, Schultz I, Anderson ME (1999) Dichloroacetate (DCA) dosimetry: interpreting DCA-induced liver cancer dose response and the potential for DCA to contribute to trichloroethylene-induced liver cancer. Toxicol. Lett. 106:9–21. Bean JA, Isacson P, Hausler WJ Jr, Kohler J (1982) Drinking water and cancer incidence in Iowa. I. Trends and incidence by source of drinking water and size of municipality. Am. J. Epidemiol. 116:912–923.

178

DRINKING WATER DISINFECTION BY-PRODUCTS

Bercz JP, Jones L, Garner L, Murray D, Ludwig DA, Boston J (1982) Subchronic toxicity of chlorine dioxide and related compounds in drinking water in the nonhuman primate. Environ. Health Perspect. 46:47–55. Bielmeier SR, Best DS, Guidici DL, Narotsky MG (2001) Pregnancy loss in the rat caused by bromodichloromethane. Toxicol. Sci. 59:309–315. Blackburn BG, Craun GF, Yoder JS, Hill V, Calderon RL, Chen N, Lee SH, Levy DA, Beach MJ (2004) Surveillance for waterborne-disease outbreaks associated with drinking water—United States 2001–2002. MMWR Surveill. Summ. 53 (SS08):23–45. Blackshear PJ, Holloway PA, Alberti KG (1974) The metabolic effects of sodium dichloroacetate in the starved rat. Biochem. J. 142:279–286. Bolton JL, Trush MA, Penning TM, Dryhurst G, Monks TJ (2000) Role of quinones in toxicology. Chem. Res. Toxicol. 13 (3):135–160. Bolyard M, Fair PS, Hautman DP (1993) Sources of chlorate ion in U.S. drinking water. J. Am. Water Works Assoc. 85:81–88. Bove FJ, Fulcomer MC, Klotz JB, Esmart J, Dufficy EM, Savrin JE (1995) Public drinking water contamination and birth outcomes. Am. J. Epidemiol. 141:850–862. Brenniman GR, Vasilomanolakis-Lagos J, Amsel J (1980) Case–control study of cacner deaths in Illinois communities served by chlorinated or nonchlorinated water. In: Jolley RL, et al., editor. Water Chlorination: Environmental Impact and Health Effects, Vol. 3. Ann Arbor, MI: Ann Arbor Science Publishers, Inc., pp. 1043–1057. Bruschi SA, Bull RJ (1993) In vitro cytotoxicity of mono-, di- and trichloroacetate and its modulation by peroxisome proliferation. Fundam. Appl. Toxicol. 21:366–375. Bull RJ (2000) Mode of action of liver tumor induction by trichloroethylene and its metabolites, trichloroacetate and dichloroacetate. Environ. Health Perspect. 108 (Suppl. 2):241–259. Bull RJ, Kopfler FC (1991) Health Effects of Disinfectants and Disinfectant By-Products. Denver, CO:AWWA Research Foundation and American Water Works Association. Bull RJ, Robinson M, Meier JR, Stober JA (1982) Use of biological assay systems to assess the relative carcinogenic hazards of disinfection by-products. Environ. Health Perspect. 46:215– 228. Bull RJ, Meier JR, Robinson M, Ringhand HP, Laurie RP, Stober JA (1985) Evaluation of mutagenic and carcinogenic properties of brominated and chlorinated acetonitriles: by-products of chlorination. Fundam. Appl. Toxicol. 5:1065–1074. Bull RJ, Brown JM, Meierhenry EA, Jorgenson TA, Robinson M, Stober JA (1986) Enhancement of the hepatotoxicity of chloroform in B6C3F1 mice by corn oil: Implications for chloroform carcinogenesis. Environ. Health Perspect. 69:49–58. Bull RJ, Gerba C, Trussell RR (1990a) Evaluation of the health risks associated with disinfection. Crit. Rev. Environ. Control 20:77–113. Bull RJ, Sanchez IM, Nelson MA, Larson JL, Lansing AJ (1990b) Liver tumor induction in B6C3F1 mice by dichloroacetate and trichloroacetate. Toxicology 63:341–359. Bull RJ, Birnbaum LS, Cantor KP, Rose JB, Butterworth BE, Pegram R, Tuomisto J (1995) Water chlorination: essential process or cancer hazard? Fundam. Appl. Toxicol. 28:155–166. Bull RJ, Krasner SW, Daniel PA, Bull RD (2001) Health Effects and Occurrence of Disinfection ByProducts. Denver, Colo: AWWA Research Foundation. Bull RJ, Orner GA, Cheng RS, Stillwell L, Stauber AJ, Sasser LB, Lingohr MK, Thrall BD (2002) The contribution of dichloroacetate and trichloroacetate to liver tumor induction in mice by trichloroethylene. Toxicol. Appl. Pharmacol. 182:55–65. Bull RJ, Sasser LB, Lei XC (2004) Interactions in the tumor promoting activity of carbon tetrachloride, trichloroacetate, and dichloroacetate in the liver of male B6C3F1 mice. Toxicology 199:169–183.

REFERENCES

179

Bull RJ, Reckhow DA, Rotello V, Bull OM, Kim J (2006) Use of toxicological and chemical models to prioritize DBP Research. Final Report AwwaRF 2867. Denver, CO: AWWA Research Foundation and American Water Works Association. Cantor KP, Hoover R, Mason TJ, McCabe LJ (1978) Associations of cancer mortality with trihalomethanes in drinking water. J. Natl. Cancer Inst. 61:979–985. Cantor KP, Hoover R, Hartge P, Mason TJ, Silverman DT, Altman R, Austin DF, Child MA, Key CR, Marrett LD, Myers MH, Narayana AS, Levin LI, Sullivan JW, Swanson GM, Thomas DB, West DW (1987) Bladder cancer, drinking water source, and tap water consumption: a case–control study. J. Natl. Cancer Inst. 79:1269–1279. Cantor KP, Lynch CF, Hildeshiem M (1996) Chlorinated drinking water and risk of glioma: a case control study in Iowa, USA. Epidemiology 7:S83. Cantor KP, Lynch CF, Hildesheim ME, Dosemeci M, Lubin J, Alavanja M, Craun G (1998) Drinking water source and chlorination byproducts. I. Risk of bladder cancer. Epidemiology 9:211–228. Cantor KP, Lynch CF, Hildesheim ME, Dosemeci M, Lubin J, Alavanja M, Craun G (1999) Drinking water source and chlorination by-products in Iowa. III. Risk of brain cancer. Am. J. Epidemiol. 150:552–560. Carlton BD, Habash DL, Basaran AH, George EL, Smith MK (1987) Sodium chlorite administration in Long-Evans rats: reproductive and endocrine effects. Environ. Res. 42:238–245. Carr AC, Hawkins CL, Thomas SR, Stocker R, Frei B (2001) Relative reactivities of N-chloramines and hypochlorous acid with human plasma constituents. Free Radic. Biol. Med. 30 (5):526–536. Carter JH, Carter HW, DeAngelo AB (1995) Biochemical, pathologic and morphometric alterations induced in male B6C3F1 mouse liver by short-term exposure to dichloroacetic acid. Toxicol. Lett. 81:55–71. Carswell F, Kerr MM, Hutchison JH (1970) Congenital goitre and hypothyroidism produced by maternal ingestion of iodides. Lancet 1:1241–1243. Cedergren MI, Selbing AJ, L€ofman O, K€allen BAJ (2002) Chlorination by-products and nitrate in drinking water and risk for congenital cardiac defects. Environ. Res. 89:124–130. Charrois JW, Arend MW, Froese KL, Hrudey SE (2004) Detecting N-nitrosamines in drinking water at nanogram per liter levels using ammonia positive chemical ionization. Environ. Sci. Technol. 38: (18):4835–4841. Chen J, Douglas GC, Thirkill TL, Lohstroh PN, Bielmeier SR, Narotsky MG, Best DS, Harrison RA, Natargajan K, Pegram RA, Overstreet JW, Lasley BL (2003) Effect of bromodichloromethane on chorionic gonadotropin secretion by human placental trophoblast cultures. Toxicol. Sci. 76:75– 82. Chen J, Thirkill TL, Lohstroh PN, Bielmeier SR, Narotsky MG, Best DS, Harrison RA, Natarajan K, Pegram RA, Overstreet JW, Lasley BL, Douglas GC (2004) Bromodichloromethane inhibits human placental trophoblast differentiation. Toxicol. Sci. 78:166–174. Chevrier C, Junod B, Cordier S (2004) Does ozonation of drinking water reduce the risk of bladder cancer? Epidemiology 15:605–614. Chipman JK, Parsons JL, Beddowes EJ (2006) The multiple influences of glutathione on bromate genotoxicity: implications for the dose–response relationship. Toxicology 221:187–189. Choi J, Valentine RL (2002) Formation of N-nitrosodimethylamine (NDMA) from reaction of monochloramine: a new disinfection by-product. Water Res. 36 (4):817–824. Christian MS, York RG, Huberman AM, Diener RM, Fisher LC (2001) Oral (drinking water) developmental toxicity studies of bromodichloromethane (BDCM) in rats and rabbits. Int. J. Toxicol. 20:225–237. Christian MS, York RG, Hoberman AM, Frazee J, Fisher LC, Brown WR, Creasy DM (2002) Oral (drinking water) two-generation reproductive toxicity study of dibromoacetic acid (DBA) in rats. Int. J. Toxicol. 21:237–276.

180

DRINKING WATER DISINFECTION BY-PRODUCTS

Chu I, Villeneuve DC, Secours VE, Becking GC, Valli VE (1982a) Toxicity of Trihalomethanes: I. The acute and subacute toxicity of chloroform, bromodichloromethane, chlorodibromomethane, and bromoform in rats. J. Environ. Sci. Health B17(3):205–224. Chu I, Villeneuve DC, Secours VE, Becking GC, Valli VE (1982b) Trihalomethanes: II. Reversibility of toxicological changes produced by chloroform, bromodichloromethane, chlorodibromomethane, and bromoform in rats. J. Environ. Sci. Health B17(3):225–240. Cicmanec JL, Condie LW, Olson GR, Wang S-R (1991) 90-day toxicity study of dichloracetate in dogs. Fundam. Appl. Toxicol. 17:376–389. Clark NEW, Chipman JK (1995) Absorption of 3-chloro-4-(dichloromethyl)-5-hydroxy-2-[5H]furanone (MX) through rat small intestine in vitro. Toxicol. Lett. 81:33–38. Clark OH, Cavalieri RR, Moser C, Ingbar SH (1990) Iodide-induced hypothyroidism in patients after thyroid resection. Eur. J. Clin. Invest. 20:573–580. Condie LW, Laurie RD, Bercz JP (1985) Subchronic toxicology of humic acid following chlorination in the rat. J. Toxicol. Environ. Health 15:305–314. Couri D, Abdel-Rahman MS (1980) Effect of chlorine dioxide and metabolites on glutathione dependent systems in rat, mouse and chicken blood. J. Environ. Pathol. Toxicol. 3:451–460. Cowman GA, Singer PC (1996) Effect of bromide ion on haloacetic acid speciation resulting from chlorination and chloramination of aquatic humic substances. Environ. Sci. Technol. 30:16–24. Crabb DW, Mapes JP, Boersma RW, Harris RA (1976) Effect of dichloroacetate on carbohydrate and lipid metabolism of isolated hepatocytes. Arch. Biochem. Biophys. 173:658–665. Cragle DL, Shy CM, Strubaand RJ, Siff EF (1985) Case–control study of colon cancer and water chlorination in North Carolina. In: Jolley RL, et al., editor. Water Chlorination: Chemistry, Environmental Impact and Health Effects, Vol. 5. Chelsea, MI: Lewis Publishers, Inc., pp. 153–159. Craun GF (1988) Surface water supplies and health. J. Am. Water Works Assoc 80:40–52. Curry SH, Pei-I Chu BS, Baumgartner TG, Stacpoole PW (1985) Plasma concentrations and metabolic effects of intravenous sodium dichloroacetate. Clin. Pharmacol. Ther. 37:89–93. Daignault SA, Huck PM, Anderson WB, von Borstel RC, Savage E, Irvine GA, Rector DW, Williams DT (1987) Drinking water treatment alternatives as evaluated chemically and biologically at pilot scale. (1987) Annual Conference Proceedings. Part 2. Denver, CO: American Water Works Association, pp. 1031–1056. Daniel FB, Schenck KM, Mattox JK, Lin ELC, Haas DL, Pereira MA (1986) Genotoxic properties of haloacetonitriles: drinking water by-products of chlorine disinfection. Fundam. Appl. Toxicol. 6:447–453. Daniel FB, DeAngelo AB, Stober JA, Pereira MA, Olson GR (1989) Chloroform inhibition of 1,2dimethylhydrazine-induced gastrointestinal tract tumors in the Fisher 344 rat. Fundam. Appl. Toxicol. 13:40–45. Daniel FB, Condie LW, Robinson M, Stober JA, York RG, Olson GR, Wang S-R (1990) Comparative 90-day subchronic toxicity studies on three drinking water disinfectants, chlorine, monochloramine and chlorine dioxide, in the Sprague–Dawley rats. J. Am. Water Works Assoc 82:61–69. Daniel FB, Ringhand HA, Robinson M, Stober JA, Olson GR, Page NP (1991) Comparative subchronic toxicity of chlorine and monochloramine in the B6C3F1 mouse. J. Am Water Works Assoc. 82:68–75. Daniel FB, DeAngelo AB, Stober JA, Olson GR, Page NP (1992) Hepatocarcinogenicity of chloral hydrate, 2-chloroacetaldehyde, and dichloroacetic acid in male B6C3F1 mouse. Fundam. Appl. Toxicol. 19:159–168. Daniel FB, Robinson M, Stober JA, Olson GR, Page NP (1993) Toxicity of 1,1,1-trihchloro2-propanone in Sprague–Dawley rats. J. Toxicol. Environ. Health 39:383–393.

REFERENCES

181

Davidson IWF, Sumner DD, Parker JC (1982) Chloroform: a review of its metabolism, teratogenic, mutagenic and carcinogenic potential. Drug Chem. Toxicol. 5:1–87. DeAngelo AB, Daniel FB, McMillan L, Wernsing P, Savage RE Jr (1989) Species and strain sensitivity to the induction of peroxisome proliferation by chloroacetic acids. Toxicol. Appl. Pharmacol. 101:285–298. DeAngelo AB, Daniel FB, Stober JA, Olson GR (1991) The carcinogenicity of dichloroacetic acid in the male B6C3F1 mouse. Fundam. Appl. Toxicol. 16:337–347. DeAngelo AB, Daniel FB, Most BM, Olson GR (1996) The carcinogenicity of dichloroacetic acid in the male Fischer 344 rat. Toxicology 114:207–221. DeAngelo AB, Daniel FB, Most BM, Olson GR (1997) Two-year bioassay of monochloroacetic acid and trichloroacetic acid administered in the drinking water of male F344/N rats. Toxicology 114:207–221. Delker D, Hatch G, Allen J, Crissman B, George M, Geter D, Kilburn S, Moore T, Nelson G, Roop B, Slade R, Swank A, Ward W, DeAngelo AB (2006) Molecular biomarkers of oxidative stress associated with bromate carcinogenicity. Toxicology 221:158–165. Deml E, Oesterle D (1985) Dose-dependent promoting activity of chloroform in rat liver foci assay. Cancer Lett. 29:59–63. De Rouen TA, Diem JE (1977) Relationships between cancer mortality in Louisiana: drinking water source and other possible causative agents. In: Hiatt HH, editor. Origins of Human Cancer, Book A.Cold Spring Harbor, ME: Cold Spring Harbor Symposium. Diaz-Gomez MI, Castro JA (1980) Covalent binding of chloroform metabolites to nuclear proteins-no evidence for binding to nuclear acids. Cancer Lett. 9:213–218. Do MT, Birkett NJ, Johnson KC, Krewski D, Villeneuve P. The Canadian Cancer Registries Epidemiology Research Group (2005) Chlorination disinfection by-products and pancreatic cancer risk. Environ. Health Perspect. 113:418–424. Dodds L, King WD (2001) Relationship between trihalomethane compounds and birth defects. Occup. Environ. Med. 58:443–446. Dodds L, King W, Woolcott C, Pole J (1999) Trihalomethanes in public water supplies and adverse birth outcomes. Epidemiology 10:233–237. Dohan O, de la Vieja A, Paroder V, Riedel C, Hartan M, Reed M, Ginter CS, Carrasco N (2003) The sodium/iodide symporter (NIS): characterization, regulation, and medical significance. Endocr. Rev. 24:48–77. Doyle TJ, Zheng W, Cerhan JR, Hong C-P, Sellers TA, Kushi LH, Folsom AR (1997) The association of drinking water source and chlorination by-products with cancer incidence among postmenopausal women in Iowa: a prospective cohort study. Am. J. Public Health 87:1168–1176. Druckrey H (1968) Chlorinated drinking water, toxicity tests, involving seven generations of rats. Food Cosmetics Toxicol. 6:147–154. Dunnick JK, Haseman JK, Lilja HS, Wyand S (1985) Toxicity and carcinogenicity of chlorodibromomethane in Fischer 344/N rats and B6C3F1 mice. Fundam. Appl. Toxicol. 5:1128–1136. Elcombe CR (1985) Species difference in carcinogenicity and peroxisome proliferation due to trichloroethylene. A biochemical human hazard assessment. Arch. Toxicol. Suppl. 8:6–17. Epstein DL, Nolen GA, Randall JL, Christ SA, Read EJ, Stober JA, Smith MK (1992) Cardiopathic effects of dichloroacetate in the fetal Long-Evans rat. Teratology 46:225–235. Eschenbrenner AB, Miller E (1945) Induction of hepatomas in mice by repeated administration of chloroform, with observations on sex differences. J. Natl. Cancer Inst. 5:251–255. Exon JH, Koller KD, O’Reilly CA, Bercz JP (1987) Immunologic evaluation of chlorine-based drinking water disinfectants, sodium hypochlorite and monochloramine. Toxicology 44:257–269. Fahroog S, Chian ESK, Engelbrecht RS (1977) Basic concepts in disinfection with ozone. J. Water Pollut. Control Fed. 49:1818–1831.

182

DRINKING WATER DISINFECTION BY-PRODUCTS

Fenster L, Waller K, Windham G, Henneman T, Anderson M, Mendola P, Overstreet JW, Swan SH (2003) Trihalomethane levels in home tap water and semen quality. Epidemiology 14:650–658. Fidler IJ (1977) Depression of macrophages in mice drinking hyperchlorinated water. Nature 270:735–736. Fidler IJ, Kirsh R, Bugelski P, Poste G (1982) Involvement of macrophages in the eradication of established metastases following intravenous injection of liposomes containing macrophage activators. Cancer Res. 42:496–501. Fisher J, Bull RJ (2006) Development of a rat dosimetry model for bromate. Toxicology 221:235–240. Foster AM, Prentice AG, Copplestone JA, Cartwright RA, Ricketts C (1997) The distribution of leukemia in association with domestic water quality in South West England. Eur. J. Cancer Prev. 6:11–19. Furukawa F, Kurata Y, Kokubo T, Takahashi M, Nakadate M (1980) Oral acute and subchronic toxicity studies for sodium hypochlorite in F-344 rat. Bull. Nat. Inst. Hyg. Sci. 98:62–69. Gallagher MD, Nuckols JR, Stallones L, Savitz DA (1998) Exposure to trihalomethanes and adverse pregnancy outcomes. Epidemiology 9:484–489. George MH, Moore T, Kilburn S, DeAngelo AB (2000) Carcinogenicity of chloral hydrate administered in drinking water to the male F344/N rat and B6C3F1 mouse. Toxicol. Pathol. 28:610–618. Gessner PK, Cabana BE (1970) A study of the interaction of the hypnotic effects and of the toxic effects of chloral hydrate and ethanol. J. Pharmacol. Exp. Ther. 174:247–259. Gill MW, Swanson MS, Murphy SR, Bailey GP (2000) Two-generation reproduction and developmental neurotoxicity study with sodium chlorite in the rat. J. Appl. Toxicol. 20:291–303. Giller S, Le Curieux F, Erb F, Marzin D (1997) Comparative genotoxicity of halogenated acetic acids found in drinking water. Mutagenesis 12:321–328. Glaze WH, Koga M, Cancilla D, Wang K, McGuire MJ, Liang S, Davis MK, Tate CH, Aieta EM (1989a) Evaluation of ozonation by-products from two California surface waters. J. Am. Water Works Assoc. 81:66–73. Glaze WH, Koga M, Cancilla D (1989b) Ozonation by-products. 2. Improvement of an aqueous-phase derivatization method for the detection of formaldehyde and other carbonyl compounds formed by the ozonation of drinking water. Environ. Sci. Technol. 23:838–847. Glaze WH, Weinberg ES (1993) Identification and Occurrence of Ozonation By-products in Drinking Water. Denver, CO: Awwa Research Foundation and American Water Works Association. Gold LS, Slone TH, Manley MB, Garfinkel GB, Ames BN (2006) Carcinogen Potency Database (CPDB) [Online]. Available:http://potency.berkeley.edu/pub/tables/chemicalsummary April 7, 2006. Goldsworthy TL, Popp JA (1987) Chlorinated hydrocarbon-induced peroxisomal enzyme activity in relation to species and organ carcinogenicity. Toxicol. Appl. Pharmacol. 88:225–233. Gonzalez-Leon A, Schultz IR, Bull RJ (1997) Pharmacokinetics and metabolism of dichloroacetate in the F344 rat after prolonged administration in drinking water. Toxicol. Appl. Pharmacol. 146:189– 195. Gordon G, Slootmaekers B, Tachiyashiki S, Wood DW III (1990) Minimizing chlorite ion and chlorate ion in water treated with chlorine dioxide. J. Am. Water Works Assoc. 82:160–165. Gottlieb MS, Carr JK, Clarkson JR (1982) Drinking water and cancer in Louisiana. A retrospective mortality study. Am. J. Epidemiol. 116:652–667. Graves GG, Matanoski GM, Tardiff RG (2001) Weight of evidence for an association between adverse reproductive and developmental effects and exposure to disinfection by-products: a critical review. Regul. Toxicol. Pharmacol. 34:103–124. Haag HB, Hoigne J (1983) Ozonation of bromide-containing waters: kinetics of formation of hypobromous acid and bromate. Environ. Sci. Technol. 17:261–267.

REFERENCES

183

Harrington RM, Shertzer HG, Bercz JP (1986) Effects of chlorine dioxide on thyroid function in the African green monkey and the rat. J. Toxicol. Environ. Health 19:235–242. Harrington RM, Romano RR, Gates D, Ridgway P (1995a) Subchronic toxicity of sodium chlorite in the rat. J. Am. Coll. Toxicol. 14:21–33. Harrington RM, Ramano RR, Irvine L (1995b) Developmental toxicity of sodium chlorite in the rabbit. J. Am. Coll. Toxicol. 14:108–118. Hasegawa R, Takahashi M, Kokubo T, Furukawa F, Toyoda K, Sato H, Kurokawa Y, Hayashi Y (1986) Carcinogenicity study of sodium hypochlorite in F344 rats. Food Chem. Toxicol. 24 (12):1295– 1302. Hawkins CL, Brown BE, Davies MJ (2001) Hypochlorite- and hypobromite-mediated radical formation and its role in cell lysis. Arch. Biochem. Biophys. 395:137–145. Hayes JR, Condie LW Jr, Borzelleca JF (1986) Toxicology of haloacetonitriles. Environ. Health Perspect. 69:183–202. Heffernan WP, Guion C, Bull RJ (1979a) Oxidative damage to the erythrocyte induced by sodium chlorite, in vitro. J. Environ. Path. Toxicol. 2:1501–1510. Heffernan WP, Guion C, Bull RJ (1979b) Oxidative damage to the erythrocyte induced by sodium chlorite, in vivo. J. Environ. Path. Toxicol. 2:1487–1499. Herren-Freund SL, Pereira MA (1986) Carcinogenicity of by-products of disinfection in mouse and rat liver. Environ. Health Perspect. 69:59–65. Herren-Freund SL, Pereira MA, Khoury MD, Olson G (1987) The carcinogenicity of trichloroethylene and its metabolites, trichloroacetic acid and dichloroacetic acid, in mouse liver. Toxicol. Appl. Pharmacol. 90:183–189. Hildesheim ME, Cantor KP, Lynch CF, Dosemeci M, Lubin J, Alavanja M, Craun G (1998) Drinking water source and chlorination by-products. II. Risk of colon and rectal cancers. Epidemiology 9:29–35. Hinckley AF, Bachand AM, Reif JS (2005) Late pregnancy exposures to disinfection by-products and growth-related birth outcomes. Environ. Health Perspect. 113:1808–1813. Hoigne J, Bader H (1975) Ozonation of water: role of hydroxyl radicals as oxidizing intermediates. Science 190:782–784. Hoigne J, Bader H (1988) The formation of trichloronitromethane (chloropicrin) and chloroform in a combined ozonation/chlorination treatment of drinking water. Water Res. 22:313–319. Hooth MJ, DeAngelo AB, George MH, Gaillard ET, Travlos GS, Boorman GA, Wolf DC (2001) Subchronic sodium chlorate exposure in drinking water results in a concentration-dependent increase in rat thyroid follicular cell hyperplasia. Toxicol. Pathol. 29:250–259. Howe EW, Aieta EM, Liang S, McGuire MJ (1989) Removal of chlorite residuals with granular activated carbon; a case study. In: Chlorine Dioxide: Scientific, Regulatory and Application Issues (A Workbook for the International Chlorine Dioxide Symposium).Washington, DC: Chemical Manufacturers Assoc., pp. 126–133. Hrudey SE, Gac A, Daignault SA (1988) Potent odour-causing chemicals arising from drinking water disinfection. Water Sci. Tech 20:55–61. Hunter ES III, Rogers E, Blanton M, Richard A, Chernoff N (2006) Bromochloro-haloacetic acids: effects on mouse embryos in vitro and QSAR considerations. Rep. Toxicol. 21:260–266. Hwang CJ, Krasner SW, Sclimenti MJ, Amy GL, Dickerson E, Bruchet A, Prompsy C, Filippi G, Croue J-P, Violleau D, Leenheer JA (2001) Polar NOM: Characterization, DBPs, Treatment. Denver, CO: Awwa Research Foundation. IARC(1978) Monographs on the Evaluation of Carcinogenic Risks to Humans. Some N-Nitro Compounds, Vol. 7. Lyon, France:International Agency for Research on Cancer. IARC (1985) Monographs on the Evaluation of Carcinogenic Risks to Humans. Tobacco habits other than smoking; and some related nitrosamines, Vol. 37. Lyon, France: International Agency for Research on Cancer.

184

DRINKING WATER DISINFECTION BY-PRODUCTS

IARC (1991) Monographs on the Evaluation of Carcinogenic Risks to Humans. Chlorinated Drinking Water; Chlorination By-Products; some other Halogenated Compounds; Cobalt and Cobalt Compounds. Lyon, France: International Agency for Research on Cancer. pp. 45–296. IARC (1999) In: Capen CC, Dybing E, Rice JM, Wilbourn JD,editors Species Differences in Thyroid, Kidney and Urinary Bladder Carcinogenesis. Appendix 1 and 2. Lyon, France: IARC Press, pp. 191–225. IARC (2004) Some Drinking Water Disinfectants and Contaminants, Including Arsenic, Vol. 84. Lyon, France: IARC Press. Ijesselmuiden CB, Gaydos C, Feighner B, Novakoski WL, Serwadda D, Caris LH, Vlahov D, Comstock GW (1992) Cancer of the pancreas and drinking water: a population-based case–control study in Washington County. Maryland. Am. J. Epidemiol. 136:836–842. ILSI (1993) In: Graun G, editor. Part 1. Global Perspectives on Drinking Water: Developed and Developing Countries, Safety of Water Disinfection: Balancing Chemical and Microbial Risks. Washington, DC: ILSI Press, pp. 1–30. Ingram PR, Homer NZM, Smith RA, Pitt AR, Wilson CG, Olejnik O, Spickett CM (2003) The interaction of sodium chlorite with phospholipids and glutathione: a comparison of effects in vitro, in mammalian and in microbial cells. Arch. Biochem. Biophys. 410:121–133. Ito A, Naito M, Naito Y, Watanabe H (1982) Induction and characterization of gastro-duodenal lesions in mice given continuous oral administration of hydrogen peroxide. Gann 73:315–322. Jaakkola JJK, Magnus P, Skrondal A, Hwang B-F, Becheer G, Dybing E (2001) Foetal growth and duration of gestation relative to water chlorination. Occup. Environ. Med. 58:437–442. Jacangelo JG, Patania NL, Reagan KM, Aieta EM, Krasner SW, McGuire MJ (1989) Ozonation: assessing its role in the formation and control of disinfection by-products. J. Am. Water Works Assoc. 81:74–84. Johannsen FR, Levinskas GJ, Tegeris AS (1986) Effects of formaldehyde in the rat and dog following oral exposure. Toxicol. Lett. 30:1–6. Jorgenson TA, Meierhenry EF, Rushbrook CJ, Bull RJ, Robinson M (1985) Carcinogenicity of chloroform in drinking water to male Osborne-Mendel rats and female B6C3F1 mice. Fundam. Appl. Toxicol. 5:760–769. Kasai H, Nishimura S, Kurokawa Y, Hayashi Y (1987) Oral administration of the renal carcinogen, potassium bromate, specifically produces 8-hydroxydeoxyquanosine in rat target organ DNA. Carcinogenesis 8 (12):1959–1961. Kallen BAJ, Robert E (2000) Drinking water chlorination and delivery outcome—a registry-based study in Sweden. Rep. Toxicol. 14:303–309. Kanitz S, Franco Y, Patrone V, Caltabellotta M, Raffo E, Riggi C, Timitilli D, Ravera G (1996) Association between drinking water disinfection and somatic parameters at birth. Environ. Health Perspect. 104:516–520. Kasim K, Levallois P, Johnson KC, Abdous B, Auger P , The Canadian Cancer Registries Epidemiology Research Group (2005) Chlorination disinfection by-products in drinking water and the risk of adult leukemia in Canada. Am. J. Epidemiol. 163:116–126. Kato-Weinstein J, Lingohr MK, Thrall BD, Bull RJ (1998) Effects of dichloroacetate on carbohydrate metabolism in B6C3F1 mice. Toxicology 130:141–154. Kato-Weinstein J, Stauber AJ, Orner GA, Thrall BD, Bull RJ (2001) Differential effects of dihalogenated and trihalogenated acetates in the liver of B6C3F1 mice. J. Appl. Toxicol. 21:81–89. Katz R, Tai CN, Diener RM, McConnell RF, Semonick DE (1981) Dichloroacetate, sodium: 3-month oral toxicity studies in rats and dogs. Toxicol. Appl. Pharmacol. 57:273–287. Kawanishi S, Murata M (2006) Mechanism of DNA damage induced by bromate differs from general types of oxidative stress. Toxicology 221:172–178.

REFERENCES

185

Keith JD, Pacey GE, Cotruvo JA, Gordon G (2006) Experimental results from the reaction of bromate ion with synthetic and real gastric juice. Toxicology 221:225–228. Kerns WD, Pavkov KL, Donofrio DJ, Gralla EJ (1983) Carcinogenicity of formaldehyde in rats and mice after long-term inhalation exposure. Cancer Res. 43:4382–4392. King WD (1999) Chlorination by-products and cancer: relative risk, population attributable risk, and individual risk. Contract #1021-8-0282 Report to Health Canada, Ottawa. King WD, Marrett LD (1996) Case–control study of water source and bladder cancer. Cancer Causes Control (November, 1996). King WD, Marrett LD, Woolcott CG (2000a) Case–control study of colon and rectal cancers and chlorination by-products in treated water. Cancer Epidemiol, Biomarkers Prev. 9:813–818. King WD, Dodds L, Allen A (2000b) Relation between stillbirth and specific chlorination by-products in public water supplies. Environ. Health Perspect. 108:883–886. King WD, Dodds L, Allen AC, Armson BA, Fell D, Nimrod C (2005) Haloacetic acids in drinking water and risk for stillbirth. Occup. Environ. Med. 62:124–127. Klaunig JE, Ruch RS, Pereira MA (1986) Carcinogenicity of chlorinated methane and ethane compounds administered in drinking water to mice. Environ. Health Perspect. 69:89–95. Klotz JB, Pyrch LA (1999) Neural tube defects and drinking water disinfection by-products. Epidemiology 10:383–390. Koivusalo M, Hakulinen T, Vartiainen T, Pukkala E, Jaakkola JJK, Tuomisto J (1998) Drinking water mutagenicity and urinary tract cancers: a population-based case–control study in Finland. Am. J. Epidemiol. 148:704–712. Komulainen H, Kosma V-M, Vaittinen S-L, Vartiainen T, Kaliste-Korhonen E, Lotjonen S, Tuominen RK, Tuomisto J (1997) Carcinogenicity of the drinking water mutagen 3-chloro-4-(dichloromethyl)-5-hydroxy-2(5H)-furanone. J. Natl. Cancer Inst. 89:848–856. Kotiaho T, Wood JM, Wick J, Dejarme LE, Ranasinghe A, Cooks RG, Ringhand HP (1992) Time persistence of monochloramine in human saliva and stomach fluid. Environ. Sci. Technol. 26:302– 306. Kramer MD, Lynch CF, Isacson P, Hanson JW (1992) The association of waterborne chloroform with interuterine growth retardation. Epidemiology 3:407–413. Krasner SW, McGuire MJ, Jacangelo JG, Patania NJ, Reagen KM, Aieta EM (1989) The occurrence of disinfection by-products in US drinking water. J. Am. Water Works Assoc. 81:41–53. Krasner SW, Glaze WH, Weinburg HS, Daniel PA, Najm IN (1993) Formation and control of bromate during ozonation of waters containing bromide. J. Am. Water Works Assoc. 85:73–81. Kronberg L, Vartiainen T (1988) Ames mutagenicity and concentration of the strong mutagen 3-chloro4-(dichloromethyl)-5-hydroxy-2(5H)-furanone and of its geometric isomer (E)-2-Chloro-3-(dichloromethyl)-4-oxobutenoic acid in chlorine-treated tap waters. Mutat. Res. 206:177–182. Kurokawa Y, Hayashi Y, Maekawa A, Takahashi M, Kokubo T, Odashima S (1983) Carcinogenicity of potassium bromate administered orally to F344 rats. J. Natl. Cancer Inst. 71:965–972. Kurokawa Y, Aoki S, Matusushima Y, Takamura N, Imazawa T, Hayashi Y (1986a) Dose–response studies on the carcinogenicity of potassium bromate in F344 rats after long-term oral administration. J. Natl. Cancer Inst. 77:977–982. Kurokawa Y, Takayama S, Konishi Y, Hiasa Y, Asahina S, Takahashi M, Maekawa A, Hayashi Y (1986b) Long-term in vivo carcinogenicity tests of potassium bromate, sodium hypochlorite, and sodium chlorite conducted in Japan. Environ. Health Perspect. 69:221–235. Kurokawa Y, Matusushima Y, Takamura N, Imazawa T, Hayashi Y (1987) Relationship between the duration of treatment and the incidence of renal cell tumors in male F344 rats administered potassium bromate. Jpn. J. Cancer Res. 78:358–364. Kuzuma RJ, Kuzuma CM, Buncher (1977) Ohio drinking water source and cancer rates. Am. J. Publ. Health 67:725–729.

186

DRINKING WATER DISINFECTION BY-PRODUCTS

Larson JL, Wolf DC, Butterworth BE (1994a) Induced cytotoxicity and cell proliferation in the hepatocarcinogenicity of chloroform in female B6C3F1 mice: comparision of administration by gavage in corn oil vs. ad libitum in drinking water. Fundam. Appl. Toxicol. 22:90–102. Larson JL, Sprankle CS, Butterworth BE (1994b) Lack of chloroform-induced DNA repair in vitro and in vivo in hepatocytes of female B6C3F1 mice. Environ. Mol. Mutagen. 23:132–136. Larson JL, Wolf DC, Butterworth BE (1995a) Induced regenerative cell proliferation in livers and kidneys of male F3-44 rats given chloroform in corn oil by gavage or ad libitum in drinking water. Toxicology 95:73–86. Larson JL, Wolf DC, Mery S, Morgan KT, Butterworth BE (1995b) Toxicity and cell proliferation in the liver, kidneys and nasal passages of female F-344 rats induced by chloroform administered by gavage. Food Chem. Toxicol. 33:443–456. Larson JL, Templin MV, Wolf DC, Jamison KC, Leininger JR, Mery S, Morgan KT, Wong BA, Conolly RB, Butterworth BE (1996) A 90-day chloroform inhalation study in female and male B6C3F1 mice: implications for risk assessment. Fundam. Appl. Toxicol. 30:118–137. Latendresse JR, Pereira MA (1997) Dissimilar characteristics of N-methyl-N-nitrosourea-initiated foci and tumors promoted by dichloroacetic acid or trichloroacetic acid in the liver of female B6C3F1 mice. Toxicol. Pathol. 25:433–440. Lau SS, Monks TJ, Everitt JI, Kleymenova E, Walker CL (2001) Carcinogenicity of a nephrotoxic metabolite of the “nongenotoxic” carcinogen hydroquinone. Chem. Res. Toxicol. 14 (1):25–33. Lawrence CE, Taylor PR, Trock BJ, Reilly AA (1984) Trihalomethanes in drinking water and human colorectal cancer. J. Natl. Cancer Inst. 72:563–568. Leavitt SA, DeAngelo AB, George MH, Ross JA (1997) Assessment of the mutagenicity of dichloroacetic acid in lacI transgenic B6C3F1 mouse liver. Carcinogenesis 18:2101–2106. Lee Y-S, Choi J-Y, Park M-K, Choi E-M, Kasai H, Chung M-H (1996) Induction of oh8Gua glycosylase in rat kidneys by potassium bromate (KBrO3), a renal oxidative carcinogen. Mutat. Res. 364:227–233. Lewis C, Suffet IH, Ritz D (2006) Estimated effects of disinfection by-products on birth weight in a population served by a single water utility. Am. J. Epidemiol. 163:38–47. Lilly PD, Simmons JE, Pegram RA (1994) Dose-dependent vehicle differences in the acute toxicity of bromodichloromethane. Fundam. Appl. Toxicol. 23:132–140. Lilly PD, Ross TM, Pegram RA (1997) Trihalomethane comparative toxicity: acute renal and hepatic toxicity of chloroform and bromodichloromethane following aqueous gavage. Toxicology 40:101– 110. Limmer J, Fleig WE, Leupold D, Bittner R, Dischuneit H, Berger H-G (1988) Hepatocellular carcinoma in Type 1 glycogen storage disease. Hepatology 8:531–537. Lin LCL, Reddy TV, Daniel FB (1992) Macromolecular adduction by trichloroacetonitrile in the Fischer 344 rat following oral gavage. Cancer Lett. 62:1–9. Linder RE, Klinefelter GR, Strader LF, Suarez JD, Dyer CJ (1994a) Acute spermatogenic effects of bromoacetic acids. Fundam. Appl. Toxicol. 22:422–430. Linder RE, Klinefelter GR, Strader LF, Suarez JD, Roberts NL, Dyer CJ (1994b) Spermatotoxicity of dibromoacetic acid in rats after 14 daily exposures. Rep. Toxicol. 8:251–259. Linder RE, Klinefelter GR, Strader LF, Narotosky MG, Suarez JD, Roberts NL, Perreault SD (1995) Dibromoacetic acid affects reproductive competence and sperm quality in the male rat. Fundam. Appl. Toxicol. 28:9–17. Linder RE, Klinefelter GR, Strader LF, Veeramachaneni DNR, Roberts NL, Suarez JD (1997a) Histopathologic changes in the testes of rats exposed to dibromoacetic acid. Rep. Toxicol. 11:47– 56. Linder RE, Klinefelter GR, Strader LF, Suarez JD, Roberts NL (1997b) Spermatotoxicity of dichloroacetic acid. Rep. Toxicol. 11:681–688.

REFERENCES

187

Lingohr MK, Thrall BD, Bull RJ (2001) Effects of dichloroacetate (DCA) on serum insulin levels and insulin-controlled signaling proteins in livers of male B6C3F1 mice. Toxicol. Sci. 59:178–184. Lingohr MK, Bull RJ, Kato-Weinstein J, Thrall BD (2002) Dichloroacetate stimulates glycogen accumulation in primary hepatocytes through an insulin-independent mechanism. Toxicol. Sci. 68:508–515. Lipscomb JC, Mahle DA, Brashear WT, Barton HA (1995) Dichloroacetic acid: metabolism in cytosol. Drug Metab. Dispos. 23:1202–1205. Lubbers JR, Chauhan S, Bianchine JR (1981) Controlled clinical evaluations of chlorine dioxide, chlorite and chlorate in man. Fundam. Appl. Toxicol. 1:334–338. Lubbers JR, Bianchine JR (1984) Effects of acute rising dose administration of chlorine dioxide, chlorate and chlorite to normal healthy male volunteers. J. Environ. Pathol. Toxicol. Oncol. 5:215–228. Lubbers JR, Chauhan S, Bianchine JR (1982) Controlled clinical evaluations of chlorine dioxide, chlorite and chlorate in man. Environ. Health Perspect. 46:57–62. Lubbers JR, Chauhan S, Miller JK, Bianchine JR (1984a) The effects of chronic administration of chlorine dioxide, chlorite and chlorate to normal healthy adult male volunteers. J. Environ. Pathol. Toxicol. Oncol. 5:229–238. Lubbers JR, Chauhan S, Miller JK, Bianchine JR (1984b) The effects of chronic administration of chlorite to glucose-6-phosphate dehydrogenase deficient healthy adult male volunteers. J. Environ. Pathol. Toxicol. Oncol. 5:239–242. Lykins BW, Griese MH (1986) Using chlorine dioxide for trihalomethane control. J. Am. Water Works Assoc. 78:88–93. Ma TH, Harris MM (1988) Review of genotoxicity of formaldehyde. Mutat. Res. 196:37–59. Maloney EK, Waxman DJ (1999) trans-Activation of PPARa and PPARg by structurally diverse environmental chemicals. Toxicol. Appl. Pharmacol. 161:209–218. Marcus PM, Savitz DA, Millikan RC, Morgenstern H (1998) Female breast cancer and trihalomethane levels in drinking water in North Carolina. Epidemiology 9:156–160. Masschelein WJ (1989) Historic and current overview of chlorine dioxide. In: Chlorine Dioxide: Scientific, Regulatory and Application Issues.Washington, DC: Chemical manufacturers Assoc., pp. 4–40. Mather GG, Exon JH, Parnell JJ, Koller LD (1990) Subchronic 90 day toxicity of dichloroacetic acid and trichloroacetic acid in rats. Toxicology 64:71–80. McGeehin MA, Reif JS, Becher JC, Mangione EJ (1993) Case–control study of bladder cancer and water disinfection methods in Colorado. Am. J. Epidemiol. 138:492–501. McGuire MJ, Krasner SW, Reagan KM, Aieta EM, Jacangelo JG, Patania NL, Gramith KM (1989) Final report for disinfection by-products in United States drinking waters, Vol. 1. Report. US EPA & California Department of Health Services. McGuire MJ, McLain JL, Obolensky A (2002) Infomation Collection Rule Data Analysis. Denver: American Water Works Association Research Foundation. McMillan DC, Powell CL, Bowman ZS, Morrow JD, Jollow JD (2005) Lipids versus proteins as major targets of pro-oxidant, direct acting hemolytic agents. Toxicol. Sci. 88:274–283. Meier JR (1988) Genotoxic activity of organic chemicals in drinking water. Mutat. Res. 196:211–245. Meier JR, Bull RJ (1985) Mutagenic properties of drinking water disinfectants and by-products. In: Jolley RL, et al., editor. Water Chlorination: Chemistry, Environmental Impact and Health Effects, Vol. 5. Chelsea, MI: Lewis Publishers, Inc., pp. 207–220. Meier JR, Lingg RD, Bull RJ (1983) Formation of mutagens following chlorination of humic acid, a model for mutagen formation during drinking water treatment. Mutat. Res. 118:25–41. Meier JR, Bull RJ, Stober JA, Cimino MC (1985) Evaluation of chemicals used for drinking water disinfection for production of chromosomal damage and sperm-head abnormalities in mice. Environ. Mutagen. 7:201–211.

188

DRINKING WATER DISINFECTION BY-PRODUCTS

Meier JR, Knohl RB, Coleman WE, Ringhand HP, Munch JW, Kaylor WH, Streicher RP, Kopfler FC (1987) Studies on the potent bacterial mutagen, 3-chloro-4-(dichloro-methyl)-5-hydroxy-2(5H)furanone: aqueous stability, XAD recovery and analytical determination in drinking water and in chlorinated humic acid solution. Mutat. Res. 189:363–373. Meier JR, Monarca S, Patterson KS, Villarini M, Daniel FB, Moretti K, Pasquini R (1996) Urine mutagenicity and biochemical effects of the drinking water mutagen, 3-chloro-4-(dichloromethyl)-5-hydroxy-2[5H]-furanone (9MX), following repeated oral administration to mice and rats. Toxicology 110:59–70. Merdink JL, Stevens DK, Stenner RD, Bull RJ (1998a) Effect of enterohepatic circulation on the disposition of chloral hydrate and its metabolites in F-344 rats. J. Toxicol. Environ. Health 56:357–368. Merdink JL, Gonzalez-Leon A, Bull RJ, Schultz IR (1998b) Dichloroacetic acid formation in B6C3F1 mice following administration of trichloroethylene or its metabolites. Toxicol. Sci. 45:33–41. Merdink JL, Bull RJ, Schultz IR (2001) Toxicokinetics of bromodichloroacetate in B6C3F1 mice. J. Appl. Toxicol. 21:53–57. Miller JW, Uden PC (1983) Characterization of non-volatile aqueous chlorination products of humic substances. Environ. Sci. Technol. 17:150–157. Miller RG, Kopfler FC, Condie LW, Pereira MA, Meier JR, Ringhand HP, Robinson M, Casto B (1986) Results of toxicological testing of Jefferson Parish pilot plant samples. Environ. Health Perspect. 69:129–140. Miller JH, Minard KM, Wind RA, Orner GA, Bull RJ (2000) In vivo MRI measurements of tumor growth induced by dichloroacetate: implications for mode of action. Toxicology 145: 115–125. Mitch WA, Sedlak DL (2002a) Formation of N-nitrosodimethylamine (NDMA) from dimethylamine during chlorination. Environ. Sci. Technol. 36: (4):588–595. Mitch WA, Sedlak DL (2002b) Factors controlling nitrosamine formation during wastewater chlorination. Water Sci. Technol.: Water Supply 2:191–198. Mobley SA, Taylor DH, Laurie RD, Pfohl RJ (1990) Chlorine dioxide depresses T3 uptake and delays development of locomotor activity in young rats. In: Jolley RL, Brungs WA, Cotruvo JA, Cumming PB, Mattice JS, Jacobs VA, editors. Water Chlorination: Chemistry, Environmental Impact and Health Effects. Vol. 6 Chelsea, MI: Lewis Publishers, pp. 342–358. Moore GS, Calabrese EJ (1980) Effect of chlorine dioxide and sodium chlorite on erythrocytes of A/J and C57L/J mice. J. Environ. Pathol. Toxicol. 4 (23):513–524. Moore MM, Harrington-Brock K (2000) Mutagenicity of trichloroethylene and its metabolites: implications for the risk assessment of trichloroethylene. Environ. Health Perspect. 108 (Suppl 2):215–223. Moore GW, Swift LL, Rabinowitz DR, Crofford OB, Oates JA, Stacpoole PW (1979) Reduction of serum cholesterol in patients with homozygous familial hypercholesterolemia by dichloroacetate. Atherosclerosis 33:285–293. Moore GS, Calabrese EJ, Leonard DA (1980) Effects of chlorite exposure on conception rate and litters of A/J strain mice. Bull. Environ. Contam. Toxicol. 25:689–696. Moore DH, Chasseaud LF, Majeed SK, Prentice DE, Roe FJC (1982) The effect of dose and vehicle on early tissue damage and regenerative activity after chloroform administration to mice. Food Chem. Toxicol. 20:951–954. Moore GS, Calabrese EJ, Forti A (1984) The lack of nephrotoxicity in the rat by sodium chlorite, a possible by-product of chlorine dioxide disinfection in drinking water. J. Environ. Sci. Health A19:643–661. Morris RD, Audet A-M, Angelillo IF, Chalmers TC, Mosteller F (1992) Chlorination, chlorination by-products, and cancer: a meta-analysis. Am. J. Public Health 82:955–963.

REFERENCES

189

Moudgal CJ, Lipscomb JC, Bruce RM (2000) Potential health effects of drinking water disinfection by-products using quantitative structure toxicity relationship. Toxicology 147 (2):109–131. Munson A, Sain LE, Sanders VM, Kauffman BM, White KL Jr, Page DG, Barnes DW, Borzelleca JF (1982) Toxicology of organic drinking water contaminants: Trichloromethane, bromodichloromethane and tribromomethane. Environ. Health Perspects. 46:117–120. Murata M, Bansho Y, Inoue S, Ito K, Ohnishi S, Midorikawa K, Kawanishi S (2001) Requirement of glutathione and cysteine in guanine-specific oxidation of DNA by carcinogenic potassium bromate. Chem. Res. Toxicol. 14:678–685. Narita T, Nakazawa H, Hizawa Y, Kudo H (1994) Glycogen storage disease associated with NiemannPick disease: histochemical, enzymatic and lipid analysis. Modern Pathol. 7:416–421. NCI (1976) Report on the carcinogenesis bioassay of chloroform. Bethesda, MD: National Cancer Institute, NTIS PB-264-018. NCI (1978) Bioassay of iodoform for possible carcinogenicity. Bethesda, MD: National Cancer Institute. Technical Report Series No. NCI-CG-TR-110. NCI (1979) Bioassay of 2,4,6-trichlorophenol for possible carcinogenicity. Bethesda, MD: National Cancer Institute. NCI-CG-TR-155. NRC (1980) National Academy of Sciences, Drinking Water and Health, Vol. 3. Washington, DC: National Academy Press. NRC (2005) Health Effects of Perchlorate Ingestion.Washington DC:National Academy Press. NTP (1986) Toxicology and carcinogenesis studies of bromodichloromethane in F344/N rats and B6C3F1 mice. Research Triangle Park, NC: National Toxicology Program. Technical Report Series No. TRS No. 321, NIH No. 88-2537, CAS. NO. 75-27-4. NTP (1989) Toxicology and carcinogenesis studies of tribromomethane (bromoform) in F344/N rats and B6C3F1 mice. Research Triangle Park, NC: National Toxicology Program. Technical Report Series No. 350, NIH No. 89-2805, CAS NO. 75-25-2. NTP (1990a) Technical report on the toxicology and carcinogenesis studies of monochloroacetic acid in F344/N rats and B6C3F1 mice. Research Triangle Park, NC: National Toxicology Program. Technical Report Series 396, NIH No. 90-2851. NTP (1990b) Draft Technical Report on the toxicology and carcinogenesis studies of chlorinated and chloraminated water in F344/N rats and B6C3F1 mice. Research Triangle Park, NC: National Toxicology Program. Technical Report Series 392, NIH No. 91-2847. NTP (1996) Short-term reproductive & developmental toxicity of chlorodibromomethane (CAS No. 124-48-1) administered in drinking water to Sprague–Dawley rats (NTP Study Number: RDGT94003) [Online]. Available:http://ntp.niehs.nih.gov/ntpweb/index.cfm?objectid ¼ 4B4F9501-F1F6-975E-70DDA4D64E8A65AF [cited January 17, 2005 1996a]. NTP (1997b) Reproductive toxicity of dibromoacetonitrile (CAS No. 3252-43-5) administered to Sprague–Dawley rats in the drinking water (NTP Study Number: RDGT94014) [Online]. Available: http://ntp.niehs.nih.gov/ntpweb/index.cfm?objectid ¼ 4B4F9501-F1F6-975E-70DDA 4D64E8A65AF [cited January 17, 2005 1997b]. NTP (1998a) Short-term reproductive and developmental toxicity of bromochloroacetic acid (CAS No. 5589-96-8) administered in the drinking water to Sprague–Dawley rats (NTP Study No. RDGT96001) [Online]. Available:http://ntp.niehs.nih.gov/ntpweb/index.cfm?objectid ¼ 4B4F9501-F1F6-975E-7099A4D64E8A65AF [cited January 15, 2005 1998b]. NTP (1998b) Short-term reproductive and developmental toxicity of bromodichloromethane (CAS No. 75-27-4) administered in the drinking water to Sprague–Dawley Rats (NTP Study Number: RDGT94017) [Online]. Available:http://ntp.niehs.nih.gov/ntpweb/index.cfm?objectid ¼ 4B4F9501-F1F6-975E-70DDA4D64E8A65AF [cited January 17, 2005 1998c]. NTP (1998c) Short-term reproductive and developmental toxicity of tribromoacetic acid (CAS No. 75-96-7) administered in drinking water to Sprague–Dawley rats (NTP Study Number: RDGT94009)

190

DRINKING WATER DISINFECTION BY-PRODUCTS

[Online]. Available:http://ntp.niehs.nih.gov/ntpweb/index.cfm?objectid ¼ 4B4F9501-F1F6-975E70DDA4D64E8A65AF [cited January 17, 2005 1998c]. NTP (2000a) Toxicology and carcinogenesis studies of chloral hydrate (ad libitum and dietary controlled) in male B6C3F1 mice (gavage study). NTP Technical Report No. 503, NIH Publication No. 00-4437. Research Triangle Park, NC: NIEHS; National Toxicology Program. NTP (2000b) Short-term reproductive and developmental toxicity study of dibromochloroacetic acid (CAS No. 5278-95-5) administered to Sprague–Dawley rats in the drinking water (NTP Study Number: RDGT98001) [Online]. Available:http://ntp.niehs.nih.gov/ntpweb/index.cfm?objectid ¼ 4B4F9501-F1F6-975E-70DDA4D64E8A65AF [cited January 17, 2005 2000b]. NTP (2002) Toxicology and carcinogenesis studies of chloral hydrate in B6C3F1 mice (gavage studies). NTP Technical Report No. 502. NIH Publication No. 02-4436. Research Triangle Park, NC: NIEHS; National Toxicology Program. NTP (2005a) Toxicology and carcinogenesis studies of sodium chlorate (CAS No. 7775-09-9) in F344/N rates and B6C3F1 mice [Online]. Available: http://ntp.niehs.nih.gov/index.cfm? objectid ¼ 00132319-F1F6-975E-778A4E6504EB919 [cited January 17, 2005 2005a]. NTP (2005b) Toxicology and carcinogenesis studies of bromodichloromethane (CAS No. 75-27-4) in male F344/N rats and female B6C3F1 mice (drinking water studies. TR-532) [Online]. Available: http://ntp.niehs.nih.gov/index.cfm?objectid ¼ 00271EF5-F1F6-975E-73E6FE7AEE1A1A31 [cited January 17, 2005 2005b]. NTP (2005c) Toxicology and carcinogenesis studies of dibromoacetic acid (CAS No. 631-64-1) in F3/ N rats and B6C3F1 mice (drinking water studies). NTP Technical Report TR-537, NIH Publication No. Research Triangle Park, NC: NIEHS; National Toxicology Program. Ohnishi S, Murata M, Kawanishi, S (2002) DNA damage induced by hypochlorite and hypobromite with reference to inflammation-associated carcinogenesis. Cancer Lett. 178:37–42. Olivieri VP, Snead MC, Kruse CW, Kuwata K (1986) Stability and effectiveness of chlorine disinfectants in water distribution systems. Environ. Health Perspect. 69:15–29. Olusi SO, Oke OL, Odusote A (1979) Effects of cyanogenic agents on reproduction and neonatal development in rats. Biol. Neonat. 36:233–243. Orme J, Taylor DH, Laurie RD, Bull RJ (1985) Effects of chlorine dioxide on thyroid function in neonatal rats. J. Toxicol. Environ. Health 15:315–322. Parrish JM, Austin EW, Stevens DK, Kinder DH, Bull RJ (1996) Haloacetate-induced oxidative damage to DNA in the liver of male B6C3F1 mice. Toxicology 110:103–111. Parsons JL, Chipman JK (2000) The role of glutathione in DNA damage by potassium bromate in vitro. Carcinogenesis 15:311–316. Payment P, Richardson L, Edwardes M, Franco E, Siemiatycki J (1990a) A prospective epidemiological study of drinking water related gastrointestinal illnesses. International Assoc. Water Poll. Res. & Contr. Symp. on Health Related Microbiology. April 1–6, 1990, Eberhard-Kals-Universitat, Tubingen, Germany, p. 44. Payment P, Richardson L, Edwardes M, Franco E, Siemiatycki J (1990b) Drinking water-related gastrointestinal illness. Abstracts of the 90th Ann Mtg Am. Soc. Microbiol. Abstract Q-ll, p. 290. Pereira MA, Lin L-H, Mattox JK (1984) Haloacetonitrile excretion as thiocyanate and inhibition of dimethylnitrosamine demethylase: a proposed metabolic scheme. J. Toxicol. Environ. Health 13:633–641. Pereira MA, Knutsen GL, Herren-Freund SL (1985) Effect of subsequent treatment of chloroform or phenobarbital on the incidence of lung and liver tumors initiated by ethylnitrosourea in 15 day old mice. Carcinogenesis 6:203–207. Pereira MA (1996) Carcinogenic activity of dichloroacetic acid and trichloroacetic acid in the liver of female B6C3F1 mice. Fundam. Appl. Toxicol. 31:192–199.

REFERENCES

191

Pereira MA, Phelps JB (1996) Promotion by dichloroacetic acid and trichloroacetic acid of Nmethyl-N-nitrosourea-initiated cancer in the liver of female B6C3F1 mice. Cancer Lett. 102:133–141. Pereira MA, Grothaus MA (1997) Chloroform in drinking water prevents hepatic cell proliferation induced by chloroform administered by gavage in corn oil to mice. Fundam. Appl. Toxicol. 37:82–87. Pereira MA, Li K, Kraemer PM (1997) Promotion by mixtures of dichloroacetic acid and trichloracetic acid of N-methyl-N-nitrosourea-initiated cancer in the liver of female B6C3F1 mice. Cancer Lett. 115:15–23. Pereira MA, Weng W, Kraemer PM, Tao L (2004) Prevention by methionine of dichloroacetic-acid induced liver cancer and DNA hypomethylation in mice. Toxicol. Sci. 77:243–248. Peto R, Gray R, Brantom P, Grasso P (1991) Dose and time relationships for tumor induction in the liver and esophagus of 4080 inbred rats by chronic ingestion of N-nitrosodiethylemine or Nnitrosodimethylamine. Cancer Res. 51:6452–6469. Plewa MJ, Wagner ED, Jazwierska P, Richardson SD, Chen PH, McKague AB (2004) Halonitromethane drinking water disinfection byproducts: chemical characterization and mammalian cell cytotoxicity and genotoxicity. Environ. Sci. Technol. 38:62–68. Pohl LR, Krishna K (1978) Deuterium isotope effect in bioactivation and hepatotoxicity of chloroform. Life Sci. 23:1067–1072. Poole C (1997) Analytical meta-analysis of epidemiologic studies of chlorinated drinking water and cancer: Quantitative review and reanalysis of the work published by Morris et al., Am. J. Public Health 1992;82:955–963. Report the USEPA National Center for Environmental Assessment, Cincinnati, OH. Poon R, Lecavalier P, Tryphonas H, Bondy G, Chen M, Chu I, Yagminas A, Valli VE, D’Amour M, Thomas B (1997) Effects of subchronic exposure of monochloramine in drinking water. Regul. Toxicol. Pharmacol. 25:166–175. Porter CK, Putnam SD, Hunting KL, Riddle MR (2005) The effect of trihalomethane and haloacetic acid exposure on fetal growth in a Maryland county. Am. J. Epidemiol. 162:334–344. Qi Y, Shang C, Lo IMC (2004) Formation of haloacetic acids during monochloramination. Water Res. 38:2375–2383. Ranmuthugala G, Pilotto L, Smith W, Vimalasiri T, Dear K, Douglas R (2003) Chlorinated drinking water and micronuclei in urinary bladder epithelial cells. Epidemiology 14:617–622. Reckhow DA, Singer PC, Malcolm RL (1990) Chlorination of humic materials: byproduct formation and chemical interpretations. Environ. Sci. Technol. 24:1655–1664. Reitz RH, Fox TR, Quast JF (1982) Mechanistic considerations for carcinogenic risk assessment: chloroform. Environ. Health Perspect. 46:163–168. Rice RG, Gomez-Taylor M (1986) Occurrence of by-products of strong oxidants reacting with drinking water contaminants—scope of problem. Environ. Health Perspect. 69:31–44. Richard AM, Hunter ES III (1996) Quantitative structure–activity relationships for the developmental toxicity of haloacetic acids in mammalian whole embryo culture. Teratology 53:352–360. Richardson SD (1998) Drinking water disinfection by-products. In: Meyers RA, editor. Encyclopedia of Environmental Analysis and Remediation New York: John Wiley & Sons, Inc. Richardson SD, Thruston AD Jr, Caughran TV, Chen PH, Collette TW, Floyd TL, Schenck KM, Lykins BW Jr, Sun G-R, Majetich G (1999a) Identification of new ozone disinfection byproducts in drinking water. Environ. Sci. Technol. 33:3368–3377. Richardson SD, Thruston AD Jr, Caughran TW, Chen PH, Collette TW, Floyd TL, Schenck KM, Lykins BW Jr, Sun G-R, Majetich G (1999b) Identification of new drinking water disinfection byproducts formed in the presence of bromide. Environ. Sci. Technol. 33:3378– 3383.

192

DRINKING WATER DISINFECTION BY-PRODUCTS

Richardson SD, Thruston AD Jr, Rav-Acha C, Groisman L, Popilevsky I, Juraev O, Glezer V, McKague AB, Plewa MJ, Wagner ED (2003) Tribromopyrrole, brominated acids, and other disinfection byproducts produced by disinfection of drinking water rich in bromide. Environ. Sci. Technol. 37 (17):3782–3793. Rijhsinghani KS, Swerdlow MA, Ghose T, Abrahams C, Rao KVN (1986) Induction of neoplastic lesions in the livers of C57BLXC3HF1 mice by chloral hydrate. Cancer Detect. Prev. 9:279–288. Robinson M, Bull RJ, Olson GR, Stober J (1989) Carcinogenic activity associated with halogenated acetones and acroleins in the mouse skin assay. Cancer Lett. 48:197–203. Robison LM, Sylvester PW, Lang JP, Bull RJ (1998) Comparison of the effects of iodine and iodide on thyroid function in humans. J. Toxicol. Environ. Health, submitted for publication. Romero J, Ventura F, Caizach J, Rivera J, Guerrero R (1997) Identification and quantification of the mutagenic compound 3-chloro-4-(dichloromethyl)-5-hydroxy-2(5H)-furanone (MX) in chlorinetreated water. Environ. Contam. Toxicol. 59:715–722. Rosenkranz HS (1973) Sodium hypochlorite and sodium perborate: preferential inhibitors of DNA polymerase-deficient bacteria. Mutat. Res. 21:171–174. Rosenkranz HS, Gutter B, Speck WT (1976) Mutagenicity and DNA-modifying activity: a comparison of two microbial assays. Mutat. Res. 41:61–70. Ross MK, Pegram RA (2003) Glutathione transferase theta 1-1-dependent metabolism of the water disinfection by-product bromodichloromethane. Chem. Res. Toxicol. 16:216–226. Ross MK, Pegram RA (2004) In vitro biotransformation and genotoxicity of the drinking water disinfection by-product: DNA-binding mediated by glutathione transferase theta 1-1. Toxicol. Appl. Pharmacol. 195:166–181. Sai K, Uchiyama S, Ohno Y, Hasegawa R, Kurokawa Y (1992) Generation of active oxygen species in vitro by the interaction of potassium bromate with rat kidney cell. Carcinogenesis 13:333–339. Sanchez IM, Bull RJ (1990) Early induction of reparative hyperplasia in the liver of B6C3F1 mice treated with dichloroacetate and trichloroacetate. Toxicology 64:33–46. Savitz DA, Andrews KW, Pastore LM (1995) Drinking water and pregnancy outcome in central North Carolina: source, amount, and trihalomethane levels. Environ. Health Perspect. 103:592–593. Savitz DA, Singer PC, Hartmann KE, Herring AH, Weinberg HS, Makarushka C, Hoffman C, Chan R, Maclehose R (2005) Drinking Water Disinfection By-Products and Pregnancy Outcomes. Denver, CO: Awwa Research Foundation, 212 pp. Schultz IR, Merdink JL, Gonzalez-Leon A, Bull RJ (1999) Comparative toxicokinetics of chlorinated and brominated haloacetates in F344 rats. Toxicol. Appl. Pharmacol. 158:103–114. Schultz IR, Merdink JL, Gonzalez-Leon A, Bull RJ (2002) Dichloroacetate toxicokinetics and disruption of tyrosine catabolism in B6C3F1 mice: dose–response relationships and age as a modifying factor. Toxicology 173:229–247. Schwetz BA, Leong BKJ, Gehring PJ (1974) Embryo- and fetotoxicity of inhaled chloroform in rats. Toxicol. Appl. Pharmacol. 28:442–451. Shih KL, Lederberg J (1976) Chloramine mutagenesis in Bacillus subtilis. Science 192:1141–1143. Smith MK, George EL, Manson JM, Stober JA (1987) Developmental toxicity of halogenated acetonitriles: drinking water by-products of chlorine disinfection. Toxicology 46:83–93. Smith MK, Randall JL, Tocco DR, York RG, Stober JA, Read EJ (1988) Teratogenic effects of trichloroacetonitrile in the Long-Evans rat. Teratology 38:113–120. Smith MK, Randall JL, Stober JA, Read EJ (1989a) Developmental toxicity of dichloroacetonitrile, a by-product of drinking water disinfection. Fundam. Appl. Toxicol. 12:765–772. Smith MK, Randall JL, Read EJ, Stober JA (1989b) Teratogenic activity of trichloroacetic acid in the rat. Teratology 40:445–451.

REFERENCES

193

Smith MK, Randall JL, Read EJ, Stober JA (1992) Developmental toxicity of dichloroacetate in the rat. Teratology 46:217–223. Snyder RD, Pullman J, Carter JH, Carter HW, DeAngelo AB (1995) In vivo administration of dichloroacetic acid suppresses spontaneous apoptosis in murine hepatocytes. Cancer Res. 55:3702–3705. So B-J, Bull RJ (1995) Dibromoacetate (DBA) acts as a promoter of abnormal crypt foci in the colon of F344 rats. Toxicologist 15:1242. Soffritti M, Maltoni C, Maffei F, Biagi R (1989) Formaldehyde: an experimental mulitpotential carcinogen. Toxicol. Ind. Health 5:699–730. Soffritti M, Belpoggi F, Lenzi A, Maltoni C (1997) Results of long-term carcinogenicity studies of chlorine in rats. Ann. NY Acad. Sci. 637:189–208. Song R, Westerhoff P, Minear R, Amy G (1997) Bromate minimization during ozonation. J. Am. Water Works Assoc. 89:69–78. Spitzweg C, Joba W, Eisenmenger W, Heufelder AE (1998) Analysis of human sodium iodide symporter gene expression in extrathyroidal tissues and cloning of its complementary deoxyribonucleic acids from salivary gland, mammary gland, and gastric mucosa. J. Clin. Endocrinol. Metab. 83:1746–1751. Spitzweg C, Dutton CM, Castro MR, Bergert ER, Goellner JR, Heufelder AE, Morris JC (2001) Expression of the sodium iodide symporter in human kidney. Kidney Int. 59:1013–1023. Stacpoole PW, Moore GW, Kornhauser DM (1979) Toxicity of chronic dichloroacetate. New Eng. J. Med. 300:372. Stacpoole PW (1989) The pharmacology of dichloroacetate. Metabolism 38:1124–1144. Stauber AJ, Bull RJ (1997) Differences in phenotype and cell replicative behavior of hepatic tumors inducted by dichloroacetate (DCA) and trichloroacetate (TCA). Toxicol. Appl. Pharmacol. 144:235–246. Stauber AJ, Bull RJ, Thrall BD (1998) Dichloroacetate and trichloroacetate promote clonal expansion of anchorage-independent hepatocytes in vivo and in vitro. Toxicol. Appl. Pharmacol. 150:287–294. Stevens AA, Moore LA, Miltner RJ (1989) Formation and control of non-trihalomethane disinfection by-products. J. Am. Water Works Assoc. 81:54–60. Stevens AA, Moore LA, Slocum CJ, Smith BL, Seeger DR, Ireland JC (1990) By-products of chlorination at ten operating utilities. In: Jolley RL, et al. , editor. Water Chlorination, Environmental Impact and Health Effects, Vol. 6. Chelsea, MI: Lewis Publishers, Inc., pp. 579–604. Swan SH, Waller K, Hopkins B, Windham G, Fenster L, Schaefer C, Neutra RR (1998) A prospective study of spontaneous abortion: relation to amount and source of drinking water consumed in early pregnancy. Epidemiology 9:126–133. Symons JM, Bellar TA, Carswell JK, DeMarco J, Kropp KL, Robeck GG, Seeger DR, Slocum CJ, Smith BL, Stevens AA (1975) National organics reconnaissance survey for halogenated organics. J. Am. Water Works Assoc. 67:634–647. Takamura N, Kurokawa Y, Matsushima Y, Imazawa T, Onodera H, Hayashi Y (1985) Long-term oral administration of potassium bromate in male Syrian golden hamsters. The Report of the Research Institute for Tuberculosis and Cancer, Tohoku University Series C (Medicine) 32: 43–46. Tao L, Wang W, Li L, Kramer PM, Pereira MA (2004) Effect of dibromoacetic acid on DNA methylation, glycogen accumulation, and peroxisome proliferation in mouse and rat liver. Toxicol. Sci. 82:62–69. Taylor DH, Pfoh RJL (1985) Effects of chlorine dioxide on neurobehavioral development of rats. In: Jolley RL, et al., editor. Water Chlorination: Chemistry, Environmental Impact and Health Effects, Vol. 5. Chelsea, MI:Lewis Publishers, Inc., pp. 355–364.

194

DRINKING WATER DISINFECTION BY-PRODUCTS

Testai E, Vittozzi L (1986) Biochemical alterations elicited in rat liver microsomes by oxidation and reduction products of chloroform metabolism. Chem. Biol. Interact. 59:157–171. Thomas EL, Jefferson MM, Bennett JJ, Learn DB (1987) Mutagenic activity of chloramines. Mutat. Res. 188:35–43. Thornton-Manning JR, Seely JR, Pegram RA (1994) Toxicity of bromodichloromethane in female rats pretreated with cytochrome P450 inducers and inhibitors. Toxicology 94:3–18. Thurman EM (1986) Dissolved organic compounds in natural waters. In: Ram N, et al., editor. Organic Carcinogens in Drinking Water. New York: John Wiley and Sons, Inc., pp. 55–92. Til HP, Woutersen RA, Feron VJ, Hollanders HM, Falke HE (1989) Two-year drinking-water study of formaldehyde in rats. Food Chem. Toxicol. 27:77–87. Tobe M, Naito K, Kurokawa Y (1989) Chronic toxicity study on formaldehyde administered orally to rats. Toxicology 56:79–86. Toledano M, Nieuwenhuijsen MJ, Best N, Whitaker H, Hambly P, de Hoogh D, Fawell J, Jarup L, Elliott P (2005) Relationship of trihalomethane concentrations in public water supplies to stillbirth and birth weight in three water regions in England. Environ. Health Perspect. 113:225– 232. Tong Z, Board PG, Anders MW (1998) Glutathione S-transferase zeta-catalyzed biotransformation of dichloroacetic acid and other a-haloacids. Chem. Res. Toxicol. 11:1332–1338. Toth GP, Long RE, Mills TS, Smith MK (1990) Effects of chlorine dioxide on the developing rat brain. J. Toxicol. Environ. Health 31:29–44. Toth GP, Kelty KC, George EL, Read EJ, Smith MK (1992) Adverse male reproductive effects following subchronic exposure of rats to sodium dichloroacetate. Fundam. Appl. Toxicol. 19:57–63. Tuthill RW, Giusti RA, Moore GS, Calabrese EJ (1982) Health effects among newborns after prenatal exposure to ClO2-disinfected drinking water. Environ. Health Perspect. 46:39–43. Tzeng H-F, Blackburn AC, Board PG, Anders MW (2000) Polymorphism and species-dependent inactivation of glutathione transferase zeta by dichloroacetate. Chem. Res. Toxicol. 13:231–236. Umemura T, Kurokawa Y (2006) Etiology of bromate-induced cancer and possible modes of action— studies in Japan. Toxicology 221:154–157. Umemura T, Kitamura Y, Kanki K, Marumyama S, Okazaki K, Imazawa T, Nishimura T, Hasegawa R, Nishikawa A, Hirose M (2004) Dose-related changes of oxidative stress and proliferation in kidneys of male and female F344 rats exposed to potassium bromate. Cancer Sci. 95:393–398. USEPA (2003a) National primary drinking water regulations: Stage 2 disinfectants and disinfection byproducts rule. National Primary and Secondary Drinking Water Regulations: Approval of Analytical Methods for Chemical Contaminants. Proposed Rule. Parts 141, 142, and 143, Vol. 68, pp. 49548–49681. USEPA (2003b) Health risks to fetuses, infants, and children: proposed Stage 2 disinfectant/ disinfection byproducts: a review. EPA-822-R-03-010. USEPA (2005) IRIS substance list. http://www.epa.gov/iris/subst/index.html Toxicological Reviews in Adobe PDF*. EPA 635/R-01/001. Vena JE, Graham S, Freudenheim J, Marshall J, Zielezny M, Swanson M, Sufrin G (1993) Drinking water, fluid intake, and bladder cancer in Western New York. Arch. Environ. Health 48:191– 198. Villanueva CM, Fernandez F, Malats N, Grimalt JO, Kogevinas M (2003) Meta-analysis of studies on individual consumption of chlorinated drinking water and bladder cancer. J. Epidemiol. Commun. Health 57:166–173. Villanueva CM, Cantor KP, Cordier S, Jaakola JJ, King WD, Lynch CF, Porru S, Kogevinas M (2004) Disinfection by-products and bladder cancer: a pooled analysis. Epidemiology 15:357–367.

REFERENCES

195

Vissers MCM, Carr AC, Chapman ALP (1998) Comparison of human red blood cell lysis by hypochlorous and hypobromous acids: insights into the mechanism of lysis. Biochem. J. 330:131–138. Von Gunten U, Bruchet A, Costentin E (1996) Bromate formation in advanced oxidation processes. J. Am. Water Works Assoc. 88:53–65. Waller K, Swan SH, DeLorenze G, Hopkins B (1998) Trihalomethanes in drinking water and spontaneous abortion. Epidemiology 9:134–140. Watanabe S, Togashi S-I, Fukui T (2002) Contribution of nitric oxide to potassium bromate-induced elevation of methaemoglobin concentration in mouse blood. Biol. Pharm. Bull. 25:1315–1319. Weinberg HS (2000) Bromate Occurrence in Chlorinated Waters. Denver, Colo.: American Water Works Association—Government Affairs Office. Weinberg HS, Krasner SW, Richardson SD, Thruston AD Jr (2002) The Occurrence of Disinfection By-Products (DBPs) of Health Concern in Drinking Water: Results of a Nationwide DBP Occurrence Study. Athens, Ga.: U.S. Environmental Protection Agency; National Exposure Research Laboratory. Whitehouse S, Cooper RH, Randle PJ (1974) Mechanism of activation of pyruvate dehydrogenase by dichloroacetate and other halogenated carboxylic acids. Biochem. J. 141:761–764. Wideroff L, Gridley G, Mellemkjaer L, Chow W-H, Linet M, Keehn S, Borch-Johnsen K, Olsen JH (1997) Cancer incidence in a population-based cohort of patients hospitalized with diabetes mellitus in Denmark. J. Natl. Cancer Inst. 89:1360–1365. Wilkins JR III, Comstock GW (1981) Source of drinking water at home and site-specific cancer incidence in Washington County, Maryland. Am. J. Epidemiol. 114:178–190. Windham GC, Waller K, Anderson M, Fenster L, Medola P, Swan S (2003) Chlorination by-products in drinking water and menstrual cycle. Environ. Health Perspect. 111:935–941. Wlodkowski TJ, Rosenkranz HS (1975) Mutagenicity of sodium hypochlorite for Salmonella typhumurium. Mutat. Res. 31:39–42. Woeber KA (1991) Iodine and thyroid disease. Med. Clin. North Am. 75:169–178. Wolf DC, Crosby LM, George MH, Kilburn SR, Moore TM, Miller RT, DeAngelo AB (1998) Time and dose-dependent development of potassium bromate-induced tumors in male Fischer 344 rats. Toxicol. Pathol. 26:724–729. Wolfe GW, Kaiser L (1996) Sodium bromate: short-term reproductive and developmental toxicity study when administered to Sprague–Dawley rats in the drinking water. Final Report on NIEHS Contract NIEHS-NO1-ES-15323. NTIS PB96-190640, Springfield, VA 22161, 313 pp. Wolfe RL, Stewart MH, Liang S, McGuire MJ (1989) Disinfection of model indicator organisms in a drinking water pilot plant by using PEROXONE. Appl. Environ. Microbiol. 55:2230–2241. Wolff J (1964) Transport of iodide and other anions in the thyroid gland. Physiol. Rev. 44:45–90. Woo YT, Lai D, McLain JL, Manibusan MK, Dellarco V (2002) Use of mechanism-based structure– activity relationships analysis in carcinogenic potential ranking for drinking water disinfection by-products. Environ. Health Perspect. 110 (Suppl 1):75–87. Woutersen RA, Feron VJ (1987) Inhalation toxicology of acetaldehyde in rats. IV. Progression and regression of nasal lesions after discontinuation of exposure. Toxicology 47:295–305. Woutersen RA, Appelman LM, Feron VJ, Vander Heijden CA (1984) Inhalation toxicity of acetaldehyde in rats. II. Carcinogenicity study interim results after 15 months. Toxicology 31:123–133. Wright EM, Diamond JM (1977) Anion selectivity in biological systems. Physiol. Rev. 57:109– 156. Wright JM, Schwartz J, Dockery DW (2004) The effect of disinfection by-products and mutagenic activity on birth weight and gestational duration. Environ. Health Perspect. 112:920–925.

196

DRINKING WATER DISINFECTION BY-PRODUCTS

Xu G, Stevens DK, Bull RJ (1995) Metabolism of bromodichloroacetate in B6C3F1 mice. Drug Metab. Dispos. 23:1412–1416. Yang C-H (2004) Drinking water chlorination and adverse birth outcomes in Taiwan. Toxicology 198:249–254. Yokose Y, Uchida K, Nakae D, Shiraiwa K, Yamamoto K, Konishi Y (1987) Studies of the carcinogenicity of sodium chlorite in B6C3F1 mice. Environ. Health Perspect. 76:205–210. Young TB, Kanarek MS, Tsiatis AA (1981) Epidemiologic study of drinking water chlorination and Wisconsin female cancer mortality. J. Natl. Cancer Inst. 67:1191–1198. Young TB, Wolf DA, Kanarek MS (1987) Case control study of colon cancer and drinking water trihalomethanes in Wisconsin. Int. J. Epidemiol. 16:90. Yount EA, Felten SY, O’Connor BL, Peterson RG, Powell RS, Yum MN, Harris RA (1982) Comparison of the metabolic and toxic effects of 2-chloropropionate and dichloroacetate. J. Pharmacol. Exp. Ther. 222:501–508. Zeiger E, Haseman JK, Shelby MD, Margolin BH, Tennant RW (1990) Evaluation of four in vitro genetic toxicity tests for predicting rodent carcinogenicity: confirmation of earlier results with 41 additional chemicals. Environ. Mol. Mutagen. 16 (Suppl 18):1–14. Zhang X, Echigo S, Minear RA, Plewa MJ (2000) Characterization and comparison of disinfection byproducts of four major disinfectants. In: Barrett SE, Krasner SW, Amy GL, editors. Natural Organic Matter and Disinfection By-products: Characerization and Control in Drinking Water. Washington, DC: American Chemical Society. Zierler S, Danley RA, Feingold L (1986) Type of disinfectant in drinking water and patterns of mortality in Massachusetts. Environ. Health Perspect. 69:275–279. Zierler S, Feingold L, Danley RA, Craun G (1988) Bladder cancer in Massachusetts related to chlorinated and chloraminated drinking water: A case–control study. Arch. Environ. Health 43:195–200.

7 FOOD Joseph V. Rodricks

7.1 INTRODUCTION 7.1.1

Composition of Food

Food is by far the most chemically complex part of the environment to which humans are directly exposed. We have no reliable estimate of the number of distinct chemical compounds in the different items of food and drink we select for nourishment and pleasure, but it is surely in the hundreds of thousands. The chemical structures of most of these are unknown, but the known constituents display immense variety. To make matters more complex, the chemical composition of the human diet varies from culture to culture and over time within cultures. Food chemists will probably never know more than a small fraction of the chemicals we deliberately put into our mouths every day of our lives (NRC, 1996). The natural constituents of foods and beverages represent the major share of dietary chemicals. In addition to the hundreds of distinct compounds that supply nutritional requirements, there are thousands more that impart flavor and color. Food plants contain large numbers of natural constituents that contribute neither nutritional nor esthetic properties, but are present because they play some role in the lives of these plants. It has been estimated, for one small example, that a freshly brewed cup of coffee contains more than 600 distinct (and mostly unidentified) compounds (Smith, 1991). We also need to note the additional burden of natural products from the hundreds of herbs and spices used in food preparation. Also among the constituents of the human diet are substances that arise during food and beverage preparation. Fermentation, for example, produces numerous chemical alterations of organic compounds, yielding products bearing little chemical similarity to the starting materials. Little scientific skill, and not much gastronomic skill, is needed to recognize that each variety of wine and cheese possesses a unique chemical composition, and that none of these bears much resemblance to grape juice and milk. Roasting, broiling, baking, smoking,

Environmental Toxicants, Third Edition Edited by Morton Lippmann Copyright
2009 John Wiley & Sons, Inc.

197

198

FOOD

and other means of preparing and processing foods each sets off dozens of chemical reactions. Because most methods of food preparation have been in use for centuries, people by now think of the products of preparation as “natural.” This is perhaps appropriate, but, strictly speaking, they actually result from human manipulation of raw food products. Human beings have of course never been satisfied to leave nature as it is and have added substances to food to achieve any number of desirable technical effects. Food preservation using various inorganic salts was probably one of the earliest examples of this practice, but adding substances to color, to sweeten, to emulsify, to flavor, and to alter taste perception is also a fairly ancient practice that continues to this day. Many chemicals not directly added to food but that are intentionally used in food production, processing, and storage, actually end up in the diet, although usually in very small concentrations. Among these indirectly added substances are residues of drugs and feed additives used in animal production, crop-use pesticides, and their metabolites and degradation products, and migrants from materials used in food processing and packaging. Several thousand direct and indirect additives that may be present in foods add a significant increment to the uncounted numbers of natural substances resulting from food preparation. Some foods may also contain contaminants––unwanted by-products of nature or human industry that somehow come to be present in food. Included are bacterial and fungal metabolites resulting from the growth on food of species of these organisms, organic chemicals of industrial origin, and various metals and other inorganic species that arise either because of their natural presence in soils and water used for food production or because they have accumulated to unusually high environmental levels as a result of mining, industrial, or other human activities. For completeness we need to include bacterial metabolites that are not produced directly in food but rather in the intestines following ingestion of foods contaminated with the offending organisms. Some contaminants are fairly regularly occurring constituents of certain foods, whereas others arise only occasionally (and unpredictably) because of a human or natural mishap. As with the other dietary constituents, the total number of possible dietary contaminants is unknown, although the most important ones are fairly well documented National Academy of Sciences (NAS) (1987). The major categories of the constituents of food are summarized in Table 7.1. 7.1.2

The Problem of Understanding Food-Related Health Risks

Given the complexity of food, it is no surprise that we find little uniformity in the study of the health risks associated with its constituents. One approach is that of clinicians and epidemiologists interested in diet and health. Their studies of health trends in populations with different intakes of certain food constituents have revealed such significant associations as those between high intakes of calories and of animal fats and low intakes of fiber on the one hand and cardiovascular disease and certain forms of cancer on the other (NRC, , 1996). Epidemiologists have also uncovered associations between excessive intakes of specific dietary constituents, such as salt, nitrates, methyl mercury, and aflatoxins, and specific human diseases, although epidemiological science is usually working close to its limits in many such situations. Nutritionists rely on the tools of epidemiology, but also turn to clinical studies and studies in experimental animals to learn about the risks and benefits of nutrients. Most of their efforts have focused on the major constituents and nutrients (Reddy and Cohen, 1986; NRC, 1996; IOM, 1997–2002). The contributions of toxicologists, whose main tools are experimental investigations, have generally been limited to the study of individual constituents. The efforts of toxicologists are

199

Largest and chemically most diverse class.

Food additives, GRAS substances.

Natural, nonnutritive constituents

Intentionally introduced substances, direct and indirect

Substances permitted under law to be sold as supplements.

Dietary supplements

Ethanol plus many fermentation products.

Substances not expected to be present. Industrial and natural chemicals, biological organisms.

Contaminants

Alcoholic beverages

Reaction products from heating, irradiation, etc. Chemically very diverse.

Chemicals produced during processing, preparation

Pesticides, veterinary drugs, feed additives. Indirect additives.

Macronutrients (supplying energy) and micronutrients.

Description

Epidemiological studies extensive.

Little systematic study. Epidemiological/clinical, experimental all applicable.

Several contaminants very well studied, many not. Epidemiological and experimental.

Little systematic study, mostly experimental.

Experimental studies well developed. Little epidemiological study.

Major Abusive intakes associated with much mortality and morbidity.

Insufficient knowledge Reasonable conjecture not possible. Little systematic study.

Moderate Pathogens produce largest number of “countable” cases of food-related illnesses; most not serious or irreversible. Persistent, bioaccumulative chemicals are or concern.

Insufficient knowledge A reasonable conjecture is that they are of minor but not negligible importance.

Minor/moderate Many substances introduced decades ago have not been studied using current experimental methods. Cumulative effects unknown

Insufficient knowledge A reasonable conjecture is that they are at least moderately important.

Major Excessive energy, saturated fat intake; abusive intakes. Inadequate or excessive intakes may be significant for some micronutrients.

Clinical/epidemiological. Major challenge to develop experimental models to study excessive intakes.

Little systematic study, mostly experimental.

Importance and Source of Health Risks

Methods Used to Study Risks and Benefits

Classes of Food Constituents and Contaminants and Their Known and Potential Health Impacts

Nutrients

Class

TABLE 7.1

Moderate Range of low intakes almost certain to decrease risk of CHD.

Insufficient knowledge Many perceived as beneficial. Little systematic study, but increasing.

Insufficient knowledge Not expected to confer benefits.

Insufficient knowledge Not expected to confer significant benefits

Minor Some may provide benefits through pathogen reduction, preservation of foods.

Insufficient knowledge Some appear to offer benefits, and many may prove to be beneficial.

Major Essential for life. Levels in excess of recommended intake levels may prove to be beneficial.

Known and Possible Health Benefits

200

FOOD

primarily driven by regulatory requirements to establish limits on human exposure to additives and certain well-recognized contaminants. Up to the present, relatively little systematic study of the thousands of natural constituents of food, except for those that have made themselves known by high and easily detectable (usually acute) toxicity, has been carried out by epidemiologists, nutritionists, and toxicologists (Doull, 1981; Kotsonis et al., 2001). The tools available to study risks of food origin are seriously limited (Table 7.1). Certainly, epidemiologists have been able to provide some highly important clues about health benefits and risks associated with certain nutritional characteristics of the diet, but so far have been able to say little about the possible risks of the many thousands of individual constituents and contaminants. The clinician primarily studies health benefits. Clinical studies may detect unexpected side effects, but such studies are not carried out until preclinical toxicology data have been collected showing that, at the most, only minor and readily reversible adverse effects are to be expected under the proposed conditions of human dosing. Toxicology studies suffer from the obvious limitation that experimental animals are not the species of interest. They are also limited because they typically involve study of individual constituents (although there have been some efforts at wider application of this tool). It is almost certain that the collection of toxicology data on each of the individual constituents and contaminants of food (a clearly impossible task) would still not provide a thorough picture of food-related risks. The total health risk associated with food is surely not simply the sum of the risks associated with each of its individual constituents and contaminants. Moreover, the picture is becoming increasingly complicated by the fact that many constituents of food, in addition to the nutrients, may confer substantial health benefits (NRC, 1996; Broch et al., 2003). Indeed, most of the major influences of food on health, both positive and negative, have more to do with overall dietary patterns than with the effects of individual constituents. Table 7.1 provides a glimpse of what is now understood about the health impacts of the diet and its many constituents and contaminates, and of the many significant gaps in understanding. 7.1.3

Scope and Limitations of this Chapter

Rather than attempting to provide a complete evaluation of the role of food in human health, we focus on individual constituents, additives, and contaminants and on the methods by which they are evaluated. Unlike most of the chapters in this volume, which deal with one or a few chemicals, we are forced somehow to consider thousands of individual substances. It seemed to make little sense to provide detailed exposure and toxicology reviews for a few important food substances, because little of general value can be learned by such an approach. Instead the choice has been made to emphasize the principles and methods for evaluating individual constituents, for assessing their health risks, and for establishing limits on human exposure to them. Broad surveys of the major categories of food constituents and contaminants are presented, and examples are drawn from several of these categories to illustrate certain principles and methods. Because it is the subject of Chapter 25, the matter of pesticide residues in food is omitted here. Because much of what has been learned about food constituents and contaminants resulted from the scientific investigations that have been conducted because of legal requirements, we begin with a discussion of the regulatory framework under which these substances are treated. Following this, we proceed to surveys and examples of substances directly and indirectly added to food, contaminants of industrial origin, and constituents and

LEGAL AND REGULATORY FRAMEWORK IN THE UNITED STATES

201

contaminants of natural origin. Several new additions to this edition appear in these sections. Also new to this edition are sections on developments in the European Union (EU) and on the issue of dietary supplements; here the evolving regulatory frameworks are emphasized because there is as yet little to report regarding purely scientific matters. The closing section deals with gaps in understanding and some suggestions regarding possible avenues toward improvements in knowledge.

7.2 LEGAL AND REGULATORY FRAMEWORK IN THE UNITED STATES Although this chapter emphasizes the scientific evaluation of risks from food constituents and contaminants, it is necessary to include some background on the legal and regulatory contexts. We do not propose to discuss the intricacies of the Federal Food, Drug and Cosmetic Act, the law governing food safety in the United States, but only to summarize certain broad features of it. Legal experts will recognize this summary as inadequate (but, we hope, not misleading). It is intended to provide scientists with some understanding of why certain categories of food constituents have received more extensive study than others, and about the role of risk information in decision making. In connection with the risks of food constituents and contaminants, the Federal Food, Drug and Cosmetic Act recognizes and distinguishes among at least three categories (Roberts, 1981; IOM, 1998b): (1) substances that are intentionally added to food, both directly and indirectly; (2) substances considered to be unavoidable contaminants of food; and (3) substances that are natural components of food. Substances that have come to be called dietary supplements are subject to legal requirements that were enacted in 1994 (Table 7.1). The major federal agency responsible for enforcing the Act is the Food and Drug Administration (FDA). The Department of Agriculture’s Food Safety and Inspection Service (FSIS) has enforcement responsibility for the Federal Meat and Poultry Inspection Act, and most of its provisions dealing with potential food risks match those of the Food, Drug and Cosmetic Act. Although the law provides these agencies with broad authority to act to ensure the safety of the food supply, it places different burdens on the agency and on the regulated industries for the different categories of food constituents and contaminants. For the first group listed above––substances intentionally added to food1––the FDA has been given power to prevent their addition unless certain safety criteria are met. The agency has responded to this legal mandate by specifying the types of toxicity studies that must be undertaken prior to the introduction of the added substance and the criteria by which safety is to be judged. In essence, such substances can be introduced only if they are “shown to be safe”; the sections of food law governing intentionally introduced substances do not permit other considerations (for example, any possible benefits conferred) to influence the decision about the acceptability of these substances. “Safe” is defined as the “practical certainty of no harm” under proposed conditions of use. The Delaney Amendment, introduced in 1958, further specified that no additive found to induce cancer could be judged safe at any level of addition (clarification of the limits of applicability of the Delaney clause is offered later in this chapter). The burden of demonstrating safety falls primarily on those who would seek to add substances to food (the “petitioner”); FDA can prevent such actions simply by showing 1

This group consists of several subgroups, and distinctions are made among these; they are discussed later in this chapter.

202

FOOD

that the burden has not been met. Except for substances generally recognized as safe (GRAS), substances can be intentionally introduced into food only in conformance with written regulations, the content of which in large part depends on the toxicology and human intake data supplied to FDA by the petitioner (Merrill, 1996). Somewhat different burdens and criteria apply to contaminants. Clearly, if certain lots of food are deliberately or accidentally contaminated, or if the contamination can readily be avoided by good manufacturing practice (botulism for example), FDA and FSIS have substantial authority to ban or otherwise limit human exposure to the contaminated food and can take powerful emergency actions if the risks are judged substantial or imminent. The more difficult problem concerns contaminants the agency considers unavoidable but that may be present at levels that pose significant (though perhaps not imminent) health risks. There are, for example, certain substances that may enter the food chain because of their widespread presence in the environment. Chemical pollutants such as mercury, lead, polychlorinated biphenyls (PCBs), and several chlorinated hydrocarbon pesticides, for example, can be found in certain foods on a fairly regular basis, mostly at low (but not always insignificant) levels. These substances are not present because of any deliberate act of food adulteration. They are present because of industrial practices that led to widespread environmental contamination and, with the metals at least, because there is also a certain level of natural occurrence. Obviously, with enough foresight, significant PCB and pesticide contamination of the environment might have been prevented by the institution of strict controls from the first days of their commercial production. But this was not done in a way that would be considered appropriate by today’s standards, and regulators are now faced with problems of food chain contamination that can be controlled in the short term only by banning or limiting consumption of the affected food itself. This situation is clearly different from that of the deliberately added substances, and under the law the FDA has authority to balance health risks from the contaminant against certain costs––notably the loss of portions of the food supply––in setting limits on human exposure. The FDA has also applied these criteria to certain contaminants of natural origin, such as the aflatoxins (Merrill, 1996; IOM, 1996). Contaminants differ from intentionally added substances in another significant respect: no specific responsibility for developing the data necessary to characterize health effects, human exposures, and risks, is assigned under the law. Generally, the FDA, relying on its own or other governmental testing programs, or on data appearing in the scientific literature, has the burden of demonstrating contaminant risks prior to devising programs to control them. Thus, although some contaminants, such as lead and mercury, have received considerable scrutiny from toxicologists and epidemiologists because of their widespread environmental occurrence, many have been only poorly characterized as to the risks they pose, certainly far less than intentional additives. Nonnutritive, naturally occurring constituents (not contaminants) of food have received relatively little attention, in part because the law prefers not to tamper with food itself unless the risks are clearly substantial. Thus, for example, certain plants that might otherwise be considered suitable for food historically have been excluded, not by the FDA, but because they contain levels of toxicants sufficiently high to cause immediate adverse or even deadly affects. However, we readily accept many natural substances (caffeine, for example) that produce subclinical effects under normal rates of intake, or substances that produce serious, chronic toxicity at high doses in animal tests, when it is clear their intentional addition to the diet would be prohibited because they fail to meet the safety criteria applied to additives. And, as with contaminants, no special responsibility is assigned under the law for the

TOXICITY TEST REQUIREMENTS AND SAFETY CRITERIA

203

development of risk-related data on these substances. Not surprisingly, most investigations into the toxicity and risks of natural constituents of the diet (and they are very few in number relative to the size of this class of substances) tend to be of limited scope (Rodricks, 1981; Kotsonis et al., 2001). These various legal criteria (and others not mentioned here) help to explain why the extent of our knowledge of the classes of substances to be discussed later varies so greatly among them. They also reveal why different approaches to risk management have been taken for different constituents of the diet.

7.3 TOXICITY TEST REQUIREMENTS AND SAFETY CRITERIA 7.3.1

The Acceptable Daily Intake

One risk criterion for judging the acceptability of substances intentionally added to food is the acceptable daily intake (ADI). The ADI is a level of daily intake that is not expected to cause adverse health effects when maintained over a full lifetime (Joint Codex Alimentarius Commission, 1979; Food Protection Committee, 1970; Dourson and Stara, 1983; Tennant, 1997). Arnold Lehman and O. Garth Fitzhugh of the FDA introduced the ADI approach in the early 1950s. This was to assist regulation of food additives and pesticides. The ADI has been widely used in all areas of regulation; the FAO/WHO Joint Expert Committee on Food Additives has published ADIs for many food additives (some conditionally because of data gaps or uncertainties), and they are generally recognized as authoritative in member countries. The theoretical basis for the ADI is that, for all toxic effects with the possible exception of carcinogenicity (see later), a threshold dose must be exceeded before any toxic response is produced. Experimentally measured thresholds, or no observed adverse effect levels (NOAELs), are divided by various “uncertainty factors” to estimate the corresponding threshold dose for the general human population. The Environmental Protection Agency (EPA) now uses the term Toxicity Reference Dose (RfD) for what is the practical equivalent of the ADI. Uncertainty factors are used to account for uncertainties regarding variability in susceptibility between animals and humans, and among members of the human population (see Table 7.2; Dourson et al., 1996). TABLE 7.2 Typical Uncertainty Factors used to Derive ADIs from Animal Toxicity Data Source of Uncertainty or Variability Extrapolation from animal NOAEL to estimate NOAEL for “average” humana Variability within human population; “average” to most susceptible Chronic NOAEL from subchronic NOAEL Chronic LOAEL to chronic NOAEL Limited database (e.g., data available from a single species only)

Factors Typically Applied 10 10 10 2–10 2–10

NOAEL: No observed adverse effect level. LOAEL: Lowest Observed Adverse Effect Level a Most sensitive species. Data available on specific compounds may allow departures from these standard “default” factors.

204

FOOD

Typically, a 100-fold uncertainty factor is applied to NOAELs from chronic studies to derive a chronically applicable ADI. Exceptions to the use of a 100-fold factor are made when data are available to reduce uncertainties regarding inter- or intraspecies extrapolation or when certain data are lacking or are inadequate in some way, in which case larger factors are used (Table 7.2). A substance is eligible for addition to foods as long as the total daily intake from all sources does not exceed its ADI (Kokoski and Flamm, 1984). An ADI is generally derived only when sufficient toxicity data have been developed, according to regulatory testing requirements. An RfD may be developed based on whatever data happen to be available; for many substances that do not require premarket testing, data gaps may be substantial. The ADI is not a sharp dividing line between “safe” and “unsafe” intakes. It is by no means certain that intakes at or below the ADI are “risk-free” or that intakes above it pose significant risks. The ADI provides no insight into the question of risk because the generic uncertainty factors used, and even those that are in part based on chemical-specific data, provide no information on the fraction of the population whose thresholds are below or above the ADI. Risk might be expressed in terms of those fractions (on the theory that thresholds are distributed in some regular fashion among members of the population), and policy decisions could then be made to ensure that no more than the tiniest fraction would, in theory, be exposed at intakes exceeding their thresholds; this is similar conceptually to the approach taken for carcinogens (see later). No standardized methodology is available to estimate risks for threshold agents in this fashion; indeed, toxicology data are typically reported in insufficiently quantitative terms to permit risk assessors to move toward more quantitative evaluations of risk for these classes of agents. 7.3.1.1 Toxicity Testing for Additives The types and numbers of toxicity tests specified by the FDAwere first described in detail in a publication entitled Toxicological Principles for the Safety Assessment of Direct Food Additives and Color Additives Used in Food. This document is commonly known as the “Red Book” because of the color of its cover when it was originally published in 1982 (FDA, 1982a, 1993). The Red Book sets forth practices that have evolved over the years based on knowledge of toxicological properties associated with certain types of chemical compounds. It reflects an awareness of the need for toxicological information commensurate with the potential of an additive to cause safety concerns. Thus, the extent of toxicological testing required for a food additive or a color additive used in food is determined on a chemicalspecific basis, and relates to its chemical structure (insofar as chemical structure reflects toxicological potential) and the extent of expected human exposure. The Red Book also provides guidance regarding the criteria used for food and color additive safety evaluations. The Red Book invokes the term “level of concern” to establish a system for gathering necessary safety information. Table 7.3 presents the levels of concern for various anticipated exposure levels for direct additives classified according to their structural similarity to compounds of known biological activity. Compounds in category C represent those potential additives that may exhibit significant toxicological activity. For example, organic halides, compounds with highly reactive heterocyclic ring systems such as epoxides or unsaturated lactones, are structure C compounds. At the other extreme, substances such as simple aliphatic and noncyclic hydrocarbons, saturated noncyclic straight-chain alcohols and carboxylic acids, and compounds identical to normal human metabolites are placed in structure A category. Based on

205

TOXICITY TEST REQUIREMENTS AND SAFETY CRITERIA

TABLE 7.3 Levels of Concern for Direct Food Additives at Different Concentrations in the Dieta and Expected Toxicity Categories Based on Chemical Structures

Expected Toxicity (Structure) III A. Low

B. Moderate

C. High

a b

Anticipated Human Exposure (ppm in Total Diet) >1.0 0.05–1.0 <0.05 >0.50 0.025–0.50 <0.025 >0.25 0.0125–0.25 <0.0125

Concern Levelb Level I (Least)

Level II (Moderate)

Level III (Most) þ

þ þ þ þ þ þ þ þ

Adapted from the Red Book (FDA, 1982a). See proposed revisions in the 1993 Red Book (FDA, 1993). Extent of Toxicity testing needed increases with increasing concern level.

anticipated human exposure level, the FDA derives a level of concern for a potential additive once it is classified according to chemical structure. Once a level of concern is determined, it then becomes possible to identify the studies required by the FDA to support safety. Table 7.4 lists the studies required for a direct food additive with the highest level of concern. The list of studies does not include acute or subacute toxicity tests, but it is hard to imagine appropriate design and conduct of the studies listed here without this information. The example is, of course, illustrative and requirements for specific additives may vary significantly. Moreover, the need for additional information may arise when the results of testing become known (FDA, 1982b; Frankos and Rodricks, 2001). 7.3.2

Information from Clinical Studies in Humans

Under the Food, Drug, and Cosmetic Act, substantial clinical data are necessary to gain approval for a human drug. No such requirement exists for food additives; the basis for introducing such materials into the human diet can rest entirely on the results of animal studies. This may seem odd, in that human exposure to some additives will ordinarily be much more widespread than drug exposure, and, moreover, additives are not introduced

TABLE 7.4 Minimum Required Studies for Direct Food Additive of Level III (Highest Level of Concern)a Two-generation reproduction study with a teratololgy phase Rodent chronic feeding study of at least 1 year Carcinogenicity bioassay in two rodent species (in utero exposure in rat) Nonrodent long-term feeding study Short-term test for carcinogenic potential Pharmacokinetics studies a

From FDA (1982a). Revisions proposed in the 1993 Red Book include requirements for additional neurotoxicity and immunotoxicity studies, and clinical investigations.

206

FOOD

with the expectation that they directly confer health benefits. Neither is there a general requirement for postapproval monitoring of health effects, as there is for some drugs. Some manufacturers conduct some form of clinical investigation for additives when human exposure is expected to be relatively high, but this is not a legal requirement (Tollefson, 1988). Because of major advances in food science and technology, the past decade has witnessed an enormous surge of interest in additives, such as noncaloric fat substances, that may be added to food in very large amounts. It seems clear that evaluation of the safety of such “macro ingredients” cannot be resolved entirely through the use of animal studies, and it is now recognized that clinical data are necessary to assess adequately human risks. A fuller discussion of the problem of macro ingredients is presented later.

7.3.3

Carcinogens and the Use of Risk Assessment

A demonstration of carcinogenicity by “appropriate” tests (interpreted by the FDA to mean tests by the oral route) invokes the Delaney Amendment; no “safe” intake level (no ADI) can be legally assigned to a substance having this property (Merrill, 1996). The FDA apparently has little scientific discretion to discount any positive finding because of possible irrelevance to humans or because the carcinogenic effect is likely secondary to toxic phenomena for which a threshold of action might be applicable, although there are cases (selenium, melamine, BHA) in which the FDA has found reasons to discount such data. In 1973, the FDA proposed the use of risk assessment coupled with the notion of insignificant risk to deal with a certain class of intentionally added substances. The Delaney Amendment, as we have noted, applies to intentionally introduced substances, but in 1968, Congress modified the law to deal with carcinogenic animal drugs that were used in foodproducing animals. Such drugs, Congress allowed, could be used as long as “no residue” of the drug could be found in food (meat, milk, eggs) from treated animals. One supposes that, in the Congressional mind, “no residue” was not different from “no addition,” as applied to directly introduced additives, but Congress modified “no residue” by adding “by a method of analysis approved by the Secretary” (read FDA). The FDA was then charged with the decision of whether any given “method of analysis” was adequate to show the absence of residues (Rodricks, 1988). The critical question facing the FDA, of course, was the detection limit such methods should possess. In scientific terms, “no residue” meant only that the carcinogenic drug was not present above the detection limit of the analytical method used. In fact, it was likely to be present in food at some nonzero level once an animal was treated and could be present at any level up to the detection limit. The FDA introduced risk assessment as a regulatory tool to deal with this class of agents in the following way (FDA, 1979a): (i) The risks (or upper bounds thereon) of the particular animal drug would be quantified based on rodent bioassay data and the application of a linear, no-threshold model of the dose–response curve. (ii) The level of daily intake of drug residue corresponding to a (upper bound) lifetime cancer risk of 10 6 would be estimated. (iii)An estimate would be made of the concentration of the drug in food (meat, milk, or eggs) that would yield, when the food is consumed, the level of daily intake estimated in step (ii).

TOXICITY TEST REQUIREMENTS AND SAFETY CRITERIA

207

(iv) The petitioner would be required to demonstrate that a reliable analytical method is available that has a detection limit at least as low as the concentration estimated in step iii. (v) The petitioner would be required to show that, under the proposed conditions of drug use, “no residue” remains in food when the analytical method of step (iv) is applied. This “sensitivity of the method” approach, as FDA has dubbed it, constituted the first regulatory use of quantitative risk assessment for carcinogens and the first time regulations defined safety explicitly in terms of risk. The FDA noted that, if the 10 6 level of lifetime (70 years) risk were accurate, and if every member of the U.S. population were exposed to it, the annual number of extra cancer cases associated with it would be ð240  106 persons  10

6

lifetime risk per personÞ

¼ 240 affected persons per 70 years; or 3 to 4 extra cases per year The agency then went on to qualify this estimate as follows: (i) The risk estimate is most likely an upper bound because of the conservative biological and statistical assumptions underlying it. (ii) It is extremely unlikely that all food from treated animals would always contain the residue of carcinogen at a concentration just at (or slightly below) the detection limit of the analytical method. (iii)It is extremely unlikely that every member of the population will consume meat (or milk or eggs) only from animals treated with the drug, and that they would do so on every day of their lives. For these and several other similar reasons, the FDA concluded that the actual number of extra cancer cases associated with the 10 6 lifetime risk criterion, applied in this way, was likely to be far fewer than three to four per year, but that the actual number could not be quantified. The risk was an insignificant burden on the public health (FDA, 1979a). The presence in food of carcinogenic animal drugs can be detected only by the application of an analytical method. In this respect, they are different from directly added substances but similar to migrants from packaging and other food contact materials (typically monomeric forms of polymers). Migrants become “additives” only if they can be detected in food. How far should we search? What analytical method, with what detection limit, should be used? The FDA seems satisfied if the methods that are used are shown capable of detecting residues at concentrations sufficient to create daily intakes corresponding to lifetime risks no greater than 10 6. In the absence of a finding that a carcinogen has migrated from food contact materials, using an analytical method of sufficient detection power, there is no need to apply the Delaney restriction, because no “additive” is introduced (Rodricks and Taylor, 1989). Logic would have it that if a lifetime risk level of 10 6 is of insignificant public health concern for indirectly introduced food additives, it ought to hold that directly introduced substances that are carcinogenic but pose similarly small risks would be of no concern. Indeed, the FDA attempted to apply this logic to some color additives (some of which showed lifetime risks of the order of 10 8 or less), but its efforts were thwarted by the U.S. Court of Appeals on the ground that the Delaney Amendment does not exempt substances because they cause negligible health risks: it is clearly a zero-risk (no addition) law

208

FOOD

(Merrill, 1979, 1996). So, although risk assessment has utility in helping to define the required characteristics of analytical methods used to search for indirectly introduced substances, the “no-risk” requirement of Delaney holds for directly introduced food and color additives, the presence of which in food does not depend on a search with analytical methods. As we see later, the FDA has also used risk assessment as a tool in the regulation of certain carcinogenic food contaminants (Food Chemical News, 1970). 7.3.4

Human Exposure Assessment–– The Estimated Daily Intake

The evaluation of carcinogenic and noncarcinogenic risks for food constituents depends not only upon toxicity criteria (ADIs or carcinogenicity potencies), but also upon estimates of human exposures to those constituents. The typical term used in food constituent risk assessment is the estimated daily intake (EDI). Generally, for intentionally introduced food constituents that are not carcinogenic, FDA’s safety criteria (risk management goals) are satisfied if the constituent’s EDI is less than its ADI. Carcinogenic risks are estimated by multiplying potencies (upper bounds on estimated lifetime risks per unit of daily intake) by EDIs. EDIs are estimated for directly introduced constituents by multiplying the concentration of the constituent in the food by the weight of that item of food consumed per day (Pao et al., 1982; Brock et al., 2003). High-end consumers, at the 90th or 95th percentile of food consumption rates, are the usual targets for the risk assessment; if such users are protected than it can be concluded that the safety standards set forth in law are satisfied (FDA, 1988). Estimating intakes for indirect additives (e.g., substances migrating from food contact surfaces and packaging) is more complex, because it depends upon analytical data regarding the amount of migration into food and the chemical identities of migrants. FDA’s Recommendations for Chemistry Data for Indirect Food Additive Petitions (FDA, 1988) sets forth procedures to be followed to collect relevant data. Various food-simulating solvents are used and extraction studies performed to estimate food concentrations; these data are combined with information on expected food contact surface areas and the daily intake of food to estimate EDIs. The vinyl chloride example, presented below, illustrates the evaluation procedure (Frankos and Rodricks, 2001). By no means does the above summary of regulatory approaches do justice to the FDA’s implementation of federal food laws. Many subtle and not so subtle points have been omitted. But it does set the stage for our discussion of various categories of food constituents and contaminants. As noted earlier, in addition to describing these categories, we also provide examples to illustrate the general principles and methods discussed in the preceding sections.

7.4 SUBSTANCES INTENTIONALLY ADDED TO FOOD Under the law, the term “food additive” has a specific meaning (Federal Food Drug and Cosmetic Act, 21 United States Code): . . . any substance the intended use of which results or may reasonably be expected to result, directly or indirectly, in its becoming a component of or otherwise affecting the characteristics of any food (including any substance intended for use in producing, manufacturing, packing, processing, preparing, treating, packaging, transporting, or holding food; and including any source of radiation intended for any such use) (emphasis added).

SUBSTANCES INTENTIONALLY ADDED TO FOOD

209

The statute goes on to exclude the following: . . . .

Substances generally recognized as safe (GRAS). Pesticide chemicals in or on raw agricultural commodities. Color additives. New animal drugs.

These four categories of excluded substances (and some others as well) are certainly “intentionally introduced additives” in the popular and even technical use of the term, but legally they are not “food additives.” For our purposes, many of the legal distinctions among these groups of additives are not important: safety must be demonstrated for all of these various groups of additives, although data requirements and criteria for acceptability vary among them (Merrill, 1996; IOM, 1998). 7.4.1

GRAS Substances

When the food additive amendments to the Federal Food, Drug, and Cosmetic Act were enacted in 1958, certain food ingredients that had long been in use were exempted from the premarket testing and approval process required for food additives. Ingredients in use prior to January 1, 1958, could be considered GRAS based on a common use in food or through scientific evaluation procedures. Any food ingredient can be classified as GRAS if it is “generally recognized, among scientific experts qualified by scientific training and experience to evaluate its safety . . . to be safe under the conditions of its intended use.” These criteria are quite general and basically leave the decision about GRAS status to scientific experts. After public review of its proposals, the FDA published a GRAS list with the following commentary (Frankos and Rodricks, 2001): It is impracticable to list all substances that are generally recognized as safe for their intended use. However, by way of illustration, the Commissioner regards such common food ingredients as salt, pepper, sugar, vinegar, baking powder, and monosodium glutamate as safe for their intended use.

The FDA’s published GRAS list includes more than 600 substances, but this list by no means includes every substance that is or could be considered GRAS. Indeed, classification of substances as GRAS can be made by any group of qualified experts. The Flavor and Extract Manufacturers Association (FEMA), for example, convened such a group in 1960, and the group listed more than 1000 flavoring ingredients and their levels of addition to food that could be considered GRAS; FDA has generally considered the FEMA process and lists as meeting the criteria set forth in the Act (Oser and Hall, 1977). A selected list of GRAS substances is presented in Table 7.5. The levels of addition of these substances to food are specified for some, but for most usage levels are simply defined according to the amounts consistent with “good manufacturing practice.” For those having specified limits, new uses that would result in increased exposures have to be justified based on “scientific procedures” recognized by experts. The FDA may also “affirm” the GRAS status of substances by regulation (Merrill, 1979). Most chemicals directly added to food are GRAS. Scrutiny of the highly abbreviated list presented in Table 7.5 reveals the presence of agents of diverse toxicological characteristics. Most are present because they have had a long history of use in food with no significant

210

FOOD

TABLE 7.5

Selected GRAS Substances Listed by the FDA

Spices and other natural flavors Anise Geranium Basil Ginger Capsicum Glycyrrhiza Elder flowers Licorice Multipurpose substances Acetic acid Aluminum sulfate Caffeine Calcium carbonate Caramel Carbon dioxide

Parsley Spearmint Vanilla

Hydrogen peroxide Lecithin Methylcellulose Papain Propane Rennet

Affirmed as GRAS by FDA regulations Benzoic acid Potassium iodide Clove Propyl gallate Ethyl alcohol Sorbitol Garlic Dextrans Guar gum Gum tragacanth

reports of adverse health effects. (FDA cannot remove a substance from GRAS status unless evidence appears showing that it can no longer be considered safe for its intended use.) The quantity and quality of available toxicology data vary greatly among these substances, and decisions about the adequacy of these databases to judge risk and safety have been in the hands of experts, both within and outside of FDA (Select Committee on GRAS substances, 1981). It is difficult to generalize about the criteria used to affirm or judge GRAS status, because expert judgment is such an important part of the process. A comprehensive review of the scientific bases of the GRAS reviews that have been undertaken by SCOGS, FEMA, FDA, and others would be needed before generalizations could be developed regarding the specific risk and safety criteria employed; no such review has been conducted (Hall, 1979). 7.4.2

Direct and Indirect Food Additives

As noted, there are hundreds of direct and indirect food ingredients listed as GRAS. There are also more than 100 direct and indirect food additives regulated as “food additives” in the legal sense of the term; that is, they have been the subject of a petition submitted to FDA since 1958, containing toxicity and human intake data adequate to meet FDA’s safety criteria for food additives (as discussed earlier). Some major use categories for direct and indirect additives, along with specific examples, are given in Table 7.6. 7.4.2.1 Toxicity Evaluation of a Direct Food Additive: Aspartame The FDA’s ban on cyclamate and its proposed ban on saccharin (overridden by a special act of the Congress) were highly controversial actions, for the most part because the demand for nonnutritive sweeteners is very high among consumers. The value of such agents for diabetics is fairly clear, but whether they play a significant role in weight control is debatable. Whatever the actual benefits of such agents, they certainly fill some need in many individuals.

SUBSTANCES INTENTIONALLY ADDED TO FOOD

TABLE 7.6

211

Selected Examples of Direct and Indirect Food Additives Regulated by the FDA

Major Categories of Direct Additives

Example

Notes

Primary direct additives Food preservatives

BHA, BHT nitrites, nitrates

Anticaking agents

Silicon dioxide

Flavoring agents

Elder tree leaves

Coating agents, films Gums, chewing gum bases

Coumarone-indene resin Arabinogalactan

Special dietary, nutritional agents

Nicotinamide–ascorbic acid complex

Multipurpose Agents Emulsifiers Dough conditions

Sodium laurel sulfate Azodicarbonamide

Dispersing agent

Polysorbate 60

Stabilizing agents Nonnutritive sweeteners

Propylene glycol alginate Aspartame

GRAS for some uses Antimicrobial actions, color fixation in meat, poultry, smoked fish: up to 2% in various foods Amount not to exceed 2% by weight of food Alcoholic beverages, 25 ppm limit on HCN Citrus fruits Used in minimum quantity required to produce intended effect as an emulsifier, stabilizer, binder, or bodying agent Source of ascorbic acid and Nicotinamide in multivitamin preparations Egg white solids, etc. Also used for aging, bleaching of flour Also used as emulsifier, dough conditioner Emulsifier, baked goods, cheeses, etc. See text

Secondary direct additives: Processing aids Polymers, resins Acrylate–acrylamide resins Clarification of juices Enzyme preparations Catalase derived from Cheese production micrococcus lysodeikticus Microbial agents Candida lipolytica Foods resulting from fermentation Solvents, related agents 1,3-Butylene glycol Extraction of flavors from spices, etc. Indirect food additives Adhesives and components of coatings Paper and paperboard components Polymers Adjuvants, production aids, and sanitizers

Acrylate ester copolymer coating Acrylamide–acrylic acid resins n-Butyl methacrylate Hydrogen peroxide

Approved for food contact Specifications for monomer Approved for food contact For sterilizing food-contact surfaces

212

FOOD

In 1981 the FDA approved certain food uses of 1-methyl-N-L-aspartyl-L-phenylalanine, the sweetener otherwise known as aspartame, and has since permitted additional uses. The compound is a simple monomethylester of a dipeptide; per unit weight it yields about the same number of calories as sucrose (so it is not a nonnutritive agent, like saccharin or cyclamate), but it is 150–200 times sweeter than sucrose (FDA, 1983). In aqueous solution, especially at elevated temperatures or at low pH, aspartame releases methanol when it hydrolyzes to the dipeptide. Because the dipeptide is bitter to the taste, aspartame cannot be used in baked goods or in highly acidic products that require prolonged storage. The dipeptide may also cyclize to a diketopiperazine (FDA, 1983). The toxicity of the degradation products, most especially the diketopiperazine but also the methanol, was of substantial concern in the FDA review process. Beginning in the 1960s, the petitioner, G.D. Searle Co., sponsored or conducted extensive toxicological studies on aspartame. All of the types of studies called for in the “Red Book” for a major-use food additive were conducted, and many more as well. The diketopiperazine degradation product was also subjected to extensive study. Metabolism studies in several species, including humans, were conducted to understand, among other things, the blood levels of phenylalanine and methanol resulting from single and repeated exposures (the former compound because of concern about infant phenylketonurics). Although there is little reason to suspect that aspartame is carcinogenic, the diketopiperazine degradation product might undergo mono- or dinitrososation in the gastrointestinal (GI) tract, yielding carcinogenic N-nitrosamines (both mono- and dinitrosopiperazine are carcinogenic in rats). Excess nonmalignant, uterine polyps result from long-term feeding of diketopiperazine to rats. (Note that these last references to diketopiperazine concern that specific compound, not the diketopiperazine derivative associated with aspartame.) Two-year feeding studies of aspartame in mice, rats, and dogs yielded no evidence of carcinogenicity. The possibility of transplacental carcinogenicity has also been investigated in rats; animals were exposed beginning in utero and throughout their lifetimes to aspartame doses of 2 or 4 g/kg. No significant tumor excesses were observed. Bladder implantation studies involving both aspartame and the diketopiperazine degradation product also produced no evidence of carcinogenicity. Chronic administration of an aspartame/diketopiperazine mixture to rats revealed no evidence of carcinogenicity (FDA, 1983). The potential neurotoxicity of aspartic acid also came under special scrutiny. Hypothalamic lesions, similar to but less severe than those produced by monosodium glutamate, were observed in neonatal mice orally administered aspartame at doses of 1 or 2 g/kg. Infant macaques administered 2 g/kg aspartame by gavage showed no evidence of brain damage. Feeding of aspartame at doses of 100 or 200 mg/kg to normal adults, 1-year-old infants, and individuals heterozygous for phenylketonuria revealed rapid metabolism and only small increases in blood level of aspartate. The rapid rise necessary to induce brain lesions was not observed. Methanol levels in human subjects receiving these doses were also evaluated (Stagink et al., 1979). Although several sources indicate an ADI for aspartame as 50 mg/kg per day, the source of this number is difficult to establish. Instead of an ADI, the Commissioner’s Final Decision on aspartame refers to a toxic threshold of 100 mmol/dL for plasma phenylalanine. (FDA, 1981). The Decision of the Public Board of Inquiry also focuses on a figure of 100 mmol/dL as the limit for the plasma level in order to protect against impaired brain development. In pregnant women, a maternal plasma phenylalanine concentration of 50 mmol/dL can produce a fetal plasma concentration of 100 mmol/dL. To achieve a blood concentration no greater than

SUBSTANCES INTENTIONALLY ADDED TO FOOD

213

50 mmol/dL requires a loading dose of 200 mg/kg, or a daily intake of about 12 g for an adult. Usage of aspartame by the 99th percentile consumer yields an EDI of only about one-fifth this level (FDA, 1983). 7.4.2.2 When is a Food Contact Material a Food Additive? Vinyl Chloride and Bisphenol A In 1975 FDA proposed to prohibit some uses of vinyl chloride polymers (homo- and copolymers), including their use in semirigid and rigid food-contact articles such as bottles and sheet. The FDA withdrew this proposal in 1986 and in its place proposed to amend its regulations to provide for the use of vinyl chloride polymers. This change of position resulted from (1) improved production technology that reduced the level of residual vinyl chloride monomer by a large factor; and (2) new agency policy, as described earlier, concerning the regulation of food and color additives that may contain carcinogenic impurities (FDA, 1986). The FDA concluded that vinyl chloride polymer will become a component of food, and the extent to which this occurs depends, at least in part, on the amount of monomer in the polymer. The FDA decided to regulate the use of vinyl chloride polymers to ensure that the polymer that is marketed does not contain unsafe levels of the monomer. The agency proposed that the Delaney clause is not triggered unless the additive as a whole (polymer) is found to induce cancer. An additive that has not been shown to induce cancer but that contains a carcinogenic impurity (vinyl chloride) is evaluated under the general safety clause of the statute, using risk assessment procedures to determine whether there is a reasonable certainty that no harm will result from the proposed use of the additive. The FDA has evaluated the safety of this additive under the general safety clause, using risk assessment procedures to estimate the upper-bound limit of risk presented by the carcinogenic chemical present as an impurity in the additive (as a by-product of polymer synthesis). The FDA conservatively estimated that the lifetime average individual exposure to vinyl chloride monomer from the probable food-contact uses of vinyl chloride polymers (e.g., liquor bottles, wine bottles, oil bottles, vinyl chloride homopolymer film, vinyl chloride– vinylidene chloride copolymers [films and fresh citrus fruit coatings], and miscellaneous uses) will not exceed 25 ng per day. The agency used a quantitative risk assessment procedure (linear, no-threshold model) to extrapolate from the doses in animal carcinogenicity studies of vinyl chloride to the very low doses of possible human exposure. The FDA estimated that the individual lifetime risk of cancer from exposure to vinyl chloride monomer at 25 ng per day is less than one in ten million. The agency concluded that there is a reasonable certainty of no harm from these exposures. The monomer, though carcinogenic, need not be treated as an additive subject to the Delaney Clause (FDA, 1986). Other monomeric residuals in food-contact materials have been subjected to similarly close scrutiny. Of particular current interest is bisphenol A. Bisphenol A is present in certain plastics used as food and beverage containers. The EPA has established an RfD for this important product at 50 mg/kg per day, based on data from toxicity studies published in the 1980s. Current human exposures through food seem to be well below the RfD. But in the time since the RfD was derived many investigators have pursued research into the compound’s reproductive and developmental effects, and some report biological activities said to be related to endocrine system disruption at doses below the RfD (which was derived by the application of large uncertainty factors to the earlier toxicity studies). Either the newer test systems are yielding irrelevant toxicological findings, or they are telling us that traditional

214

FOOD

study protocols fail to uncover many important endocrine system effects. Many investigators are in pursuit of answers (see Chapter 13). 7.4.3

Color Additives

Color additives have wide use in foods, drugs, cosmetics, other consumer products, and medical devices. Specific colors are listed or approved for food uses under FDA color additive regulations. Synthetic colors have to be certified by FDA on a batch-by-batch basis to assure that each new batch of the color additive meets certain standards of purity representative of the material tested for safety. Colors exempt from certification mainly include those of natural origin, such as beet powder, grape skin extract, titanium dioxide, and various fruit and vegetable juices. Of particular concern among the certified colors are trace constituents, typically starting materials used in color synthesis that are carcinogenic, including certain azo compounds and aromatic amines. Although there are some legal distinctions that result in different scientific requirements, safety criteria for color additives to be added to food are generally similar to those applicable to food additives. The Delaney Amendment also applies. An issue of some importance, similar to the vinyl chloride polymer matter, concerns the presence of trace amounts of known carcinogens in some colors that, when tested in cancer bioassays, do not themselves provoke a detectable carcinogenic response. The color additive, including its trace constituents, is thus not carcinogenic, but it is known from other data that the trace constituent(s) is carcinogenic. Such outcomes are not surprising, given the limited detection power of cancer tests and the typically low level of the contaminant in the color additive. FDA’s “constituents” policy allows the use of such colors as long as application of risk assessment to the trace constituent reveals that carcinogenic risks associated with it are negligible. Thus, for example, FDA found that the excess risks from p-toluidine, a trace contaminant of FD&C Green No. 6, a well-tested and noncarcinogenic color, did not exceed 10 8 and permitted use of the color to continue (FDA, 1982b). A particularly interesting example of the scientific dilemma created when emerging science encounters inflexible regulation (the Delaney Amendment) is presented by the events concerning the color additive, FD&C Red No. 3. This color is carcinogenic, producing excess thyroid tumors in rats. A substantial scientific basis exists for supposing that the increased incidence of rats with thyroid tumors resulting from chronic administration of FD&C Red No. 3 results from excessive stimulation of thyroid cells by thyroidstimulating hormone (TSH), a hormone released from the hypothalamus–pituitary axis. It appears that the normal feedback mechanism by which TSH release is controlled, the elevation of blood levels of the hormone triiodothyronine (T3) secreted from the TSHstimulated thyroid, is interfered with by FD&C Red No. 3, perhaps because its structure mimics that of thyroxine (T4) and allows it to compete for enzyme sites necessary to convert T4 to T3. Blockage of the T4–T3 pathway results in decreased control over TSH release, which in turn places thyroid cells at increased risk of neoplastic transformation. This phenomenon has been recognized for other agents that reduce controls on TSH release, such as the herbicide amitrole, several ethylene thiourea derivatives, and bridged-ring aromatic amines (Paynter et al., 1988; Hill et al., 1989). The important consequence of this hormonally mediated mechanism is the strong likelihood that no carcinogenic risk would exist unless levels of the color sufficient to interfere significantly with T4–T3 conversion were reached; this indirect mechanism of carcinogenesis seems to require a threshold dose to be exceeded. Although manufacturers of the color conducted studies that appeared to support this

SUBSTANCES INTENTIONALLY ADDED TO FOOD

215

hypothesis, the crucial experiment of modulating the end point, follicular cell neoplasia, in a long-term bioassay was not undertaken. The regulatory situation became complex because this color additive had been permanently listed for many years for use in food. The provisional uses were for cosmetics and externally applied drugs. Termination of the provisional listing and denial of a petition for these uses (FDA, 1990a, 1990b) are tantamount to banning the color additive for all uses, but this must be accomplished separately. Although the Delaney Amendment may have been invoked to support these actions, it is not clear what would have occurred had the secondary mechanism been established. 7.4.4

Animal Drug Residues

The issue of carcinogenic animal drugs, as we have already seen, gave rise to the incorporation of the risk assessment concept and the notion of negligible, or insignificant, risk into the FDA’s regulatory process. Noncarcinogenic drugs are evaluated, like food and color additives, under the ADI approach; human intakes of drug residues in meat, milk, and eggs must be shown not to exceed the ADI for all uses of the drug combined. A particularly difficult issue that arises in connection with the administration of drugs to food-producing animals concerns the fact that drug metabolism or degradation may lead to residues in edible food products of compounds other than the parent drug. A question arises whether toxicity studies on the parent drug alone are sufficient to characterize the toxicity of the total food residue. Although there are uncertainties, it may be argued that if the same profile of metabolites and degradation products is produced in the animal species used in the toxicity testing, then the resulting test data are reasonably representative of the toxicity of the total drug residue; indeed, this approach has been taken for some drugs. Alternatively, separate toxicity testing of metabolites or degradation products can be contemplated, but this may become impractical when many products are involved. The issue of so-called “bound residues” has also caused difficulties. Not infrequently some drug metabolites covalently bind to tissues in treated animals and might be released when these tissues are consumed by people. How is the potential toxicity of such “bound residues” (which many times are found actually to represent incorporation of drug moieties into natural macromolecules) to be measured? Both this problem and the broader problem of “total residue” are still subjects of considerable debate, and completely satisfactory solutions have not been found (Guest and Fitzpatrick, 1990). 7.4.5

Macroingredients: The New Class of Food Additives

Over the past decade intense interest in a variety of nontraditional additives has arisen. Among these are noncaloric substances designed to replace normal fats, novel sources and types of fiber, and other products designed to confer health benefits (so-called “nutriceuticals”). Advances in chemical technology, including the use of the tools of modern biotechnology, have made practical the industrial synthesis of substances that may occur naturally in food only at very low levels. If such substances confer some benefits to consumers, then there exists the impetus and the wherewithal to add them to foods in greatly increased amounts. The term “macroingredients” has come into use to describe substances intended for addition to food at or near levels formerly associated only with macronutrients (fats, carbohydrates, proteins). In Table 7.7 are compared some of the characteristics of “traditional” and “new-generation” food additives.

216

FOOD

TABLE 7.7 Comparison of “Traditional” and “New-Generation” Food Additives (See Text) Characteristics of many “traditional” food activities No natural occurrence in diet. Low molecular weight; typically metabolized as foreign compound Information on significant adverse effects obtainable only through animal studies. Little potential for identifying the “most relevant” animal model. Little potential for obtaining significant data concerning biological behavior in humans. Little potential to determine whether there are significant subpopulations of humans at especially high risk. Many likely to produce at least moderately serious toxicity at high doses. Useful animal testing possible without the confounding effects of dietary imbalances. Human intakes very low. Trace constituents unlikely to present risk. Characteristics of many “newgeneration” food activities Occur naturally in diet or closely related to compounds that do. Human intakes high. Metabolized to normal body constituents. Very low potential for toxicity Traditional animal testing not possible without confounding effects of dietary imbalances. Substantial data can be obtained in humans (ADME, tolerance, allergic potential, effect on nutritional status). Intakes of trace contaminants may not be negligible, especially problematic when ingredient derived from a novel source.

The traditional methods for understanding the toxicity, intakes, and risks of food additives would appear to offer limited opportunities to understand these aspects of the “newgeneration” additives. The traditional model of ensuring at least a 100-fold safety factor for chronic toxicity, which involves conducting toxicity studies at doses at least 100 times the human EDI, would not serve well, for example, to evaluate macroingredients intended for use at dietary levels of 0.1–1%. Exaggerating doses to test animals to 100 times these levels makes animal diets completely unsuitable for understanding the toxicity of the additive, because of the substantial confounding effects on the animals of dietary imbalances created by administering such diets (the practical limit of dietary concentration for a test substance in animal experiments is usually put at 5%). Thus, while some information of value may be gleaned from animal studies, they may be most useful for guiding clinical investigations. The latter seem necessary to gauge the potential effects of macroingredients in regard to possible consequences not expected of most “traditional” additives: allergenicity; GI tolerance; alterations of intestinal microflora; modification of nutritional status (International Food Biotechnology Council, 1990; Munro, 1990; FDA, 1993).

7.5 FOOD CONTAMINANTS OF INDUSTRIAL ORIGIN 7.5.1

Classes and Sources

Two broad classes of food contaminants have been identified. The first are certain metals and some of their organic derivatives. It is misleading, however, to place all occurrences of metals

FOOD CONTAMINANTS OF INDUSTRIAL ORIGIN

217

and organometallics in the class of contaminants, in that a certain level of most of these substances can be found in food because of natural occurrence. Moreover, because background levels of substances such as lead, arsenic, mercury, and cadmium vary widely in soils and water from different geographic regions, it is not surprising that background levels in the diet depend heavily on the geographic sources of the various components of the diet (Goyer, 1986). See Chapters 9, 17, and 19 for discussions of three important heavy metals, chromium, lead, and mercury, Chapter 5 for a review of arsenic, and Chapter 28 for more abbreviated discussions of aluminum, cadmium, and nickel. Dietary levels of these same metals can, of course, be affected by industrial contamination. The nature and extent of this contamination depend on regional sources of the metals. A major element of EPA’s characterization of risks arising from Superfund and other hazardous waste sites concerns the migration of metals from such sites through air, soils, ground water, and surface water into crops and livestock, with subsequent accumulation in food (EPA, 1988). Some organic chemicals of strictly industrial origin have been found on a fairly regular basis in foods. Of particular concern are various chlorinated hydrocarbons that exhibit high environmental stability and that bioaccumulate in fatty tissue. Table 7.8 contains a list of industrial compounds that have been of most concern as food contaminants. 7.5.2

Limiting Contaminant Exposure: PCBs and Polychlorinated Dioxins

Food contaminants of the sort listed in Table 7.8 are unavoidable in the sense discussed in the earlier section on Legal and Regulatory Framework. Decisions about limits on exposures to these materials are somewhat different in character from those we have described for the

TABLE 7.8

Some Food Contaminants of Industrial Origina

Chemical

Major Sources

Arsenic Cadmium Lead

Smelting, mining Smelting, sewage sludge Smelting, mining, lead glazed pottery and ceramic ware, automobile exhaust

Mercury, alkyl mercurials

Chlorine, lye manufacturing, incineration, power plants Pesticide usagec Electrical Industry Impurities in certain chemicals; incineration; bleached paper manufacturing

Aldrin, dieldrin, DDT, mirex Polychlorinated biphenyls Polychlorinated benzodioxins a

Foods Subject to Contamination Many including fishb Grains, vegetables, meat Several, including acidic foods coming into contact with lead ceramic ware and pottery Fish Fish, milk, eggs Fish, human milk Fish, milk, beef fat

Note that the metals listed are also present in foods because of natural occurrence. From Munro and Charbonneau (1981). b Most of the arsenic in food appears to be organically bound. These forms are substantially less toxic than inorganic arsenic. EPA considers the latter form to be carcinogenic by ingestion. c Pesticide residue in foods for which no official tolerance was ever granted or was rescinded, are considered contaminants.

218

FOOD

various classes of intentional additives. In some instances imposition of an ADI or negligible cancer risk standard, derived using conventional methods, would require banning of certain foods altogether or very severe restrictions on the fraction of the available supply that could be marketed. Using the carcinogenic potency derived by EPA for PCBs, for example, and a 10 6 upper limit on lifetime risk as the health protection standard, would require maximum fish residue levels of approximately 1–5 ppb (the limit depends heavily on data and assumptions concerning fish consumption rates). At the present time PCB residues in commercial fish in several areas of the United States contain PCBs in the range of 1–5 ppm (FDA, 1979b; Maxim, 1989). Clearly, imposition of a 5 ppb limit would mean the end of much commercial fishing until environmental levels of PCBs declined to sufficiently low levels. The regulatory approach at FDA has been to specify limits on PCBs in fish (commercial products moving in interstate commerce) at 2 ppm in edible portions. This tolerance level was chosen both to minimize health risk and to avoid an intolerable level of economic disruption in the affected industry. The risk level found tolerable by FDA exceeded by more than 100 times the 10 6 lifetime risk level used as a guide in the various “insignificant risk” decisions discussed earlier. Many state agencies have sought much stricter limits (Rodricks, 1994). The PCBs represent a particularly difficult problem for the risk assessor. Most pertinent toxicology data have been collected on the commercial products: Aroclors in the United States, Clophens in West Germany, and Kaneclors in Japan. The difficulty for the risk assessor increases when it is realized that PCB mixtures are modified as they move through the environment. The mixtures of chlorinated biphenyl found in fish, for example, vary widely in composition according to source of PCB, length of time the chemicals have been in the environment, and the species and age of the fish that has accumulated the compounds. And none of these matches in composition the commercial products for which cancer bioassay are available. The risk assessor can do no better than to use dose–response data from the commercial product that most closely matches in composition the mixture found in fish. Of increasing interest is the potential for certain PCBs to display estrogen-like properties, and to affect reproductive functions in animals. Certain isomers of PCBs bear structural similarities to certain chlorinated dibenzodioxins, and may act through common, and apparently receptor-mediated, mechanisms. The EPA has suggested, in its current and continuing review of chlorinated dioxins, that normal human background levels of the combination of structurally related, chlorinated biphenyls, dibenzo-p-dioxins, and dibenzofurans, are within an order of magnitude of the level at which they may cause adverse effects on endocrine systems (EPA, 1997). Fatty foods are thought to constitute the major media through which human exposures are created. This issue remains one of intense debate (see Chapter 11). A new study on the health affects of this whole class of persistent contaminants is about to be released by the NRC. The European Commission has recently adopted new legislation setting maximum levels for dioxins and dioxin-like PCBs in food and feed. Maximum levels for dioxins in food of animal origin and all animal feed have been applicable since July 2002, but beginning in November 2006, any food or feed in which the sum of dioxins and dioxin-like PCBs exceeds these maximum levels will not be allowed to be marketed in the EU. The effects upon commerce of this type of restriction are unpredictable. In the past several years much international attention has been directed to the family of compounds known as polybrominated diphenyl ethers (PBDEs). Some members of the family have been used widely as flame-retardants in computers and other electronic devices. The substances are persistent in the environment and much evidence regarding their

CONSTITUENTS AND CONTAMINANTS OF NATURAL ORIGIN

219

presence in food fats has accumulated in the worldwide literature. Regulations on their presence in food have not yet been issued. 7.5.2.1 Methylmercury Methylmercury occurs in fish and shellfish found in both the ocean and fresh water systems. The mercury that is the source of methylmercury arises from power plant emissions and industrial processes. Some even comes from dental amalgam wastes and from natural sources in the ocean bed. It is a developmental toxicant, causing severe defects. It causes behavioral and learning impairments in the offspring of women exposed during pregnancy (Chapter 19). Several recent epidemiological studies have involved examination on populations that consume unusually high quantity of fish. One of these, conducted in the Seychelles Islands, has so far not revealed these behavioral and learning impairments in children whose mothers exhibited mercury levels (measured in hair) higher than those typically seen in the United States and European countries. But another study, conducted in the Faroes Islands, turned up evidence of cognitive and behavioral impairments in children. Scientists have struggled to understand why two well-done studies have turned up with such different outcomes, and some possible reasons have been suggested. The EPA and public health officials have acted on the basis of the Faroes data, out of both caution and also because they seem to be supported by other, more limited data, and by experimental studies. The debate is not so much over whether methylmercury is a developmental toxicant, but rather over the dose required (NRC, 2002). This particular issue of developmental toxicity is complicated in a most interesting way. Fish are a critical source of certain fatty acids that are important contributors to normal neurological development. So, should we ask women who are pregnant to reduce their intakes of a food that is important to successful neurological development so as to reduce their intakes of a chemical that may adversely affect those same developmental processes? This is a risk management problem with no easy solution. The FDA and EPA have jointly issued guidelines on fish consumption during pregnancy, and they seem to reflect this concern for adequate nutrition during this critical period.

7.6 CONSTITUENTS AND CONTAMINANTS OF NATURAL ORIGIN 7.6.1

Categories of Natural Constituents and Contaminants

It is convenient to group food constituents and contaminants of natural origin into five broad categories as shown in Table 7.9 (Rodricks and Pohland, 1981). These categories include the substances that are natural components of the food itself or that have become incorporated from the plant’s environment or from the animal’s food supply. Other substances become a part of the food after the food is collected and stored or during preparation. Investigations into the possible risks to health of the enormous number of chemical compounds in these groups are limited to a relatively few members of each, particularly of the first. A brief survey of some selected examples of each group reveals that the range of health risks associated with these natural substances is wide and as potentially serious as that associated with additives and industrial contaminants. 7.6.1.1 Intrinsic Components of Foods: Oxalate In this category are organic compounds that are natural constituents of the plants and animals we have chosen for

220

FOOD

TABLE 7.9

Toxic Food Constituents and Contaminants of Natural Origina

Categories

Sources

Intrinsic food components Soil and water constituents

Natural constituents of plants and animals Natural mineral sources

Microbial metabolites Contaminants of natural origin Products of storage or preparation

Toxins from bacteria and fungi Toxins accumulating in marine organisms and forage plants Toxins arising in aged foods or in food preparation

a

Examples Natural pesticides in plants, puffer fish (fugu) toxin Nitrate; metals such as mercury, arsenic Aflatoxin B1, botulinum toxins Ciguatera, paralytic shellfish poison Oxidized fats, polycyclic aromatic hydrocarbons

From Rodricks and Pohland (1981). In some cases human activities can affect presence of toxicant.

our diets. Nutrients are included, but most of the many thousands of compounds in this group have no known nutritive value. Relatively few members of the group have been toxicologically investigated, although it seems clear that, because of the enormous chemical diversity exhibited by these compounds, thorough toxicological study would reveal that they would display the full spectrum of toxic effects observed with synthetic chemicals. Some of those that have been studied are listed in Table 7.10 (Committee on Food Protection, 1973; Rodricks and Pohland, 1981; Ames, 1983a, 1983b). A fuller discussion of one of these, TABLE 7.10

Some Intrinsic Components of Food of Known Toxicity

Compounds Salts of oxalic acid Solanine, chaconine HCN

Vasoactive amines Xanthines (caffeine, theophylline, theobromine Myristicin Carotatoxin Lathyrus toxins Tannins Safrole and other methylenedioxy benzenes 5- and 8- Methoxypsoralen (light activated) Ethyl acrylate Estragole a

Food Sources See text White potatoa Many plants, as adducts, released when plant tissue is damaged Pineapple, banana, plum Coffeeb

Nutmeg, mace Carrots, celery Legumes of genus Lathyrus Tea, coffee, cocoa Oil of sassafras, cinnamon, nutmeg, anise, parsley, celery, black pepper Parsley, parsnip, celery Pineapple Basil, fennel

Suspected or Known Toxic End Pointsc See text Nervous system Hemoglobin, cyanosis

Cardiovascular system CNS stimulation, other biochemical changes, cardiac effects Nervous system Nervous system Lathyrism (neurological disease) Carcinogenic in animals Carcinogenic in animals (not all members of the class) Carcinogenic in animals, with UV light Carcinogenic in animals Carcinogenic in animals

Solanaceous glycoalkaloids are present in other Solanaceae, including eggplant and tomato. Coffee contains more than 600 compounds in addition to caffeine. This is typical of natural foods. Included are many different classes of organic compounds. c Relevance of animal carcinogenicity to humans is doubtful in some cases. b

CONSTITUENTS AND CONTAMINANTS OF NATURAL ORIGIN

221

oxalate, is presented below to illustrate both typical and unique problems associated with this huge and largely unstudied class. Oxalic acid (HOOC COOH) is the simplest of the dicarboxylic acids. Salts of oxalic acid are found in most plant and animal species, but certain common plants such as spinach, rhubarb, beet leaves, red beets, tea, cocoa, and cereal grains have especially high concentrations. Both soluble and insoluble oxalates occur in plants and animals, but free oxalic acid generally is not present in plants. Early interest in the toxicity of dietary oxalates centered about reported poisoning in humans and animals ingesting high-oxalate-containing plants. The occurrence in kidney stones of calcium oxalate also raised concern about the role of dietary oxalate in their formation (Committee on Food Protection, 1973). By applying commonly used uncertainty factors, ADIs for oxalates can be calculated based on three animal studies. These studies are listed in Table 7.11. For chronic, reproductive, and developmental toxicity studies, a 100-fold uncertainty factor is usually applied. This uncertainty factor can be further modified to reflect the seriousness of the effect, the duration of the study (e.g., subchronic rather than chronic data), and the quality of the data (e.g., small number of animals; Kotsonis et al., 2001). Examples of possible uncertainty and modifying factors are shown in Table 7.11, along with the resulting ADIs. The ADIs shown in Table 7.11 range from 0.2 to 3 mg/kg per day, compared with an average oxalate consumption of 3.2 mg/kg per day from oxalate-containing food. More specifically, the average U.S. intake of naturally occurring dietary oxalate is 4 to 17 times greater than the ADI for developmental effects and 1.6 times greater than the ADI for reproductive effects. Thus the background levels of oxalate exceed the ADI levels that are suggested from the animal toxicity data available on this substance. It is not at all unusual for intakes of naturally occurring components of food, such as oxalate, to exceed the ADI that could be derived for them using methods applicable to food additives. Such outcomes can be interpreted in at least two ways. According to one interpretation, oxalate and other such intrinsic food components actually create a risk of chronic toxicity. Although there are no data to suggest that this is the case for oxalate, this possibility appears not to have been thoroughly examined, either clinically or epidemiologically. Some evidence suggests, however, that oxalate is only poorly bioavailable, although it is not clear that it is greatly less bioavailable in people than it is in the animal species from which the toxicity data were derived. A second interpretation is that the methods for deriving ADIs are excessively conservative and lead to acceptable intakes that are lower than they need to be. There seems to be no way to determine definitively which of these interpretations is correct; indeed, perhaps both TABLE 7.11

Animal Studies for Oxalate Suitable for Establishing an ADI

Study Type (Species) Chronic (2-years) (rat) Reproduction (rat) Developmental toxicity a

NOAEL (mg/kg per day)

Uncertainty Factor

Modifying Factora

ADI (mg/kg per day)

600

100

2

3

2000

100

10

2

175

100

2–10

0.2–0.9

Introduced to compensate for data limitations and severity of toxicity.

Reference Fitzhugh and Nelson (1947) Goldman et al. (1977) Sheikh-Omar and Schiefer, 1980

222

FOOD

are to a degree. But consideration of the toxicity data on many intrinsic components of food creates dilemmas not unlike that exhibited in the case of oxalate (Committee on Food Protection, 1973; Brock et al.,2003). 7.6.1.2 Naturally Occurring Pesticides Substances with pesticidal properties that are found in plants are a particularly interesting subgroup of the toxic intrinsic constituents of food. For about two decades now, several biologists, including Bruce Ames, have been acquiring, organizing, and evaluating information on food plant metabolites displaying pesticidal activity. Ames and colleagues (Ames, 1983a, 1983b; Ames and Gold, 1989) estimate daily intake of natural pesticides to be about 1.5 g, which they note is about 10,000 times the level of daily intake of man-made pesticides (Gartell et al., 1986a,b). Thousands and probably tens of thousands of compounds having insecticidal and fungicidal activity and other types of toxicity toward predators are present in plants we use as food, frequently at concentrations in the parts-per-million to parts-per-thousand range. Moreover, plant stress resulting from predator attack often induces biosynthesis of greater-than-normal concentrations. Most of these substances have not been evaluated toxicologically. Some of those that have been evaluated display the same range of toxic properties associated with synthetic chemicals. Ames and Gold (1989) list natural carcinogenic pesticides present in anise, apples, bananas, basil, broccoli, and about 40 other plant foods, herbs, and spices. For reasons already stated, none of these findings has yet had a significant impact on regulation, either direct or indirect. Certainly there has been no attempt to regulate any of these natural carcinogens, and, until recently (see below), little effort has been devoted to evaluating the risks they may pose. Although regulation in any traditional sense would seem improbable, it may be possible to control the levels of these natural toxicants by breeding them out of plants. There seems to be little sign of interest in such an activity. Contrariwise, there are at least two instances in which plant breeders have inadvertently increased the level of natural toxicants––psoralens in celery and solanine in white potatoes––to levels sufficient to cause acute toxicity (rashes in the former case and cholinesterase inhibition in the latter). The dangerous variety of celery actually made it to market and had to be withdrawn; the problem in the new variety of white potato was discovered before marketing occurred (Ames and Gold, 1989). Surely one of the major concerns of regulators and public health officials with newer “bioengineered” foods is the potential for inadvertent introduction of dangerous levels of natural toxicants (many of which now are present at levels uncomfortably near the minimum toxic level). Although the work of Ames and his associates and others as well has not yet had a major impact on the way health risks associated with environmental chemicals are viewed, the issue cannot forever be ignored. On the one hand, it may point to a significant role for natural food toxicants in chronic human diseases, including cancer. At the other extreme, it may suggest that our concerns over synthetic chemicals are wrongheaded, because it is clear that our risk assessment methodologies are inappropriate and greatly overstate low-dose risks (on the assumption that even the natural toxicants are not a significant health risk). It will require considerable research to untangle this issue, involving close collaboration between toxicologists and epidemiologists. To confuse matters further, it is clear, as Ames himself has pointed out, that there are many naturally occurring dietary constituents that, probably by several mechanisms, protect against or reduce the risk of cancer. These dietary “anticarcinogens” are probably as prevalent as the carcinogens (Davis, 1989). In any event, risk evaluation requires consideration of both sets of naturally occurring food constituents.

CONSTITUENTS AND CONTAMINANTS OF NATURAL ORIGIN

223

A committee of the National Research Council (NRC) has reviewed the state of knowledge regarding dietary carcinogens and anticarcinogens (NRC, 1996). The committee concluded that, though our understanding is relatively poor, current evidence suggests that carcinogens naturally present in food far exceed those present because of human actions. Understanding the risks associated with such agents, and the benefits associated with naturally occurring anticarcinogens, is a formidable task; the committee nevertheless stressed the need to acquire that understanding, and set forth a program toward that end. Although the evidence for a significant role for natural dietary constituents in other chronic diseases was not reviewed, it seems likely that carcinogenesis is not the only disease process in which these substances play a significant role. 7.6.1.3 Accumulation of Chemicals from Water and Soils: Nitrate The major members of this group are the metals and other elements that were discussed earlier in the section on industrial contaminants. A large number of metals and other elements beyond those discussed earlier are also naturally present in foods. Some are essential nutrients (copper, chromium, calcium, magnesium, iron, zinc), but many more that are present have no established nutritional value. Like lead, cadmium, arsenic, and mercury, some of these substances can, in limited geographic areas, accumulate to excessive levels because of industrial pollution, but for most of these, natural occurrence appears to be the dominant source (Munro and Charbonneau, 1981). The health risks from nitrates pose an interesting aspect, not only because the natural occurrence of nitrate is by far the greatest source but also because nitrate itself is not the material of toxicological interest. Rather, derivatives of nitrate––nitrite and N-nitroso compounds––are formed as a result of chemical reactions and are responsible for harmful effects (NRC/NAS, 1981). Several kinds of food plants accumulate naturally occurring nitrates from water and soil. The location of the plant and its variety, state of health, and moisture content determine how much nitrate a plant accumulates. The water and soil in the environment are the source of nitrate, and they, in turn, accumulate nitrate from nitrogen-based fertilizers as well as from nitrogenous wastes from humans and livestock. Certain food vegetables such as spinach, beets, cauliflower, lettuce, celery, radishes, kale, and mustard often contain high concentrations of nitrates (Committee on Food Protection, 1973). Poisonings of human infants have been reported from high nitrate levels in well water used to prepare infant formulas, mostly in rural areas, and from high concentrations of nitrate in baby foods. The total daily intake of nitrate from all sources has been estimated as 75 mg/ person, but in areas of high nitrate in water this can be doubled. Although these sound like large amounts, the real concern is with the conversion products of nitrates. Nitrates are converted into nitrites by microorganisms in the mouth and gut through reactions with ammonia and other organic nitrogen-containing compounds. Nitrites, in turn, react with hemoglobin, the iron-containing respiratory protein in red blood cells, and convert it to methemoglobin. Methemoglobin is unable to combine with oxygen. Thus, the blood of persons with too much methemoglobin, a condition known as methemoglobinemia, has a reduced oxygen-carrying capacity as well as a decreased capacity of residual oxyhemoglobin to dissociate and release oxygen to the tissues where it is needed. Infants are at a special risk from methemoglobinemia because their hemoglobin occurs in a form that is more easily oxidized. Furthermore, they are developmentally deficient in methemoglobin reductase, and the lower gastric acidity of infants permits nitrate-reducing microorganisms to thrive. Most episodes of methemoglobinemia in infants have been the result of high concentrations in well water (Menzer and Nelson, 1986).

224

FOOD

Long-term carcinogenic effects of compounds formed from nitrate with other nitrogencontaining compounds are also of great concern. The reaction of nitrite with secondary amines to form nitrosamines in foods has been modeled in animal studies that show that feeding nitrate together with certain amines, for example, produces tumors in rats or mice. The nitrosamines are among the most potent animal carcinogens; single doses of certain nitrosamines are sufficient to cause cancer in experimental animals. The nitrosamines are discussed further in a later section of this chapter. 7.6.1.4 Microbial Metabolites: Aflatoxins Mycotoxicoses––poisonings resulting from ingestion through foods and feeds of toxic fungal metabolites––have been and continue to be widely reported in the veterinary and medical literature. The best-known case is that of ergotism, which has been reported throughout the world periodically since the Middle Ages. Less widespread but equally serious human mycotoxicoses that have been associated with consumption of moldy foods include alimentary toxic aleukia, a hemorrhagic disease that occurred in the Soviet Union several times in the first half of this century; the related hemorrhagic disease, stachybotryotoxicosis; and yellow-rice disease, a neurotoxin disease reported from Japan. Veterinary outbreaks have been more numerous than human outbreaks because animals are more likely to receive moldy feeds. Fungi are enormously productive manufacturers of organic chemicals of immense and bewilderingly complex structural variety, and it is not surprising that many can produce serious forms of toxicity. Examples of some mycotoxins of potential concern as contaminants of food and feed are shown in Table 7.12 (Hayes and Campbell, 1986). The discovery of the, by now, well-known group of mycotoxins called aflatoxins (metabolites of Aspergillus flavus and A. parasiticus) in the early 1960s focused attention on several concerns that previously had not been considered. Most notably, scientists discovered that these mycotoxins could be found in foods not obviously moldy, indeed in foods showing no living mold on microscopic examination. Although partial destruction and elimination occur, the toxins can survive food-processing conditions that eliminate mold growth. The aflatoxins are detected in foods by chemical, not microbiological, analysis (Stoloff, 1977). Aflatoxins are potent hepatic toxicants, but production of frank liver toxicity in either livestock or people is usually associated with the relatively high levels associated with heavily molded foods. The discovery during the 1963–1975 period that aflatoxins could induce liver and other types of malignancies in rats, mice, monkeys, ferrets, guinea pigs, trout, and possibly humans provided the first suggestion that fungal toxins might present

TABLE 7.12

Some Mycotoxins of Potential Concern as Food and Feed Contaminantsa

Mycotoxin b

Aflatoxins Ergot alkaloids Ochratoxins Trichlothecense Zearalenone a b

From Hayes and Campbell (1986). See text for discussion.

Health Concern

Source

Carcinogenicity Neurotoxicity Nephrotoxicity GI, blood, and neurotoxicity Interference with reproduction

Peanuts, corn, milk Grains Grains Grains Grains, corn

CONSTITUENTS AND CONTAMINANTS OF NATURAL ORIGIN

225

health risks on more than an intermittent basis––the aflatoxins turn out to be present in certain foods, albeit at very low levels, on a fairly regular basis (Stoloff, 1977). In the Fisher strain of male rat, aflatoxin B1 is a very potent carcinogen, exceeded only by 2,3,7,8-tetrachlorodibenzo-p-dioxin. Unlike dioxin, aflatoxin also displays genotoxicity. Application of a linear extrapolation model to the male rat data yields a low-dose potency of 2.5  10þ3 per mg/kg per day (lifetime average daily dose). Potency does, however, vary greatly among species (Rodricks and Park, 1983). Based on epidemiological investigations reported recently from Southern Guangxi, China, Bechtel and Wilcock (1990) estimated the potency in humans to be 16.9–20 per mg/kg per day, significantly less than that observed for the male rat. These investigators were also able to estimate quantitatively the influence on hepatic cancer risk of hepatitis B infection and found it to have a potentially effect on aflatoxin of about 33-fold. The studies from China and other locales point convincingly to a significant role for hepatitis B infection in aflatoxin-induced liver cancer in humans (Bechtel and Wilcock, 1990). Given the relatively low level of aflatoxin intake in the United States, and the low rate of occurrence of hepatitis B infection, the compound is probably not a major contributor to liver cancer rates (Park and Stoloff, 1989). In some years, however, especially in corn grown in the Southern states, aflatoxin levels increase dramatically, and contamination becomes widespread. This phenomenon is dependent on weather conditions and is, fortunately, intermittent. Livestock appear to experience the greatest threat, but some above-average increase in human health risk probably results as well. In areas of the world where aflatoxin contamination of food and human intake is substantial, and where hepatitis B virus infection rates are high, the contribution of these mycotoxins to human liver cancer is probably far from negligible (Van Rensburg et al., 1985). Regulatory and manufacturing controls in the United States have greatly reduced aflatoxin intake over the past 25 years, but no way has been found to eliminate the problem. 7.6.2 Compounds Contaminating Edible Animal Products: Some Naturally Occurring Marine Toxicants Fish and shellfish can accumulate some highly toxic substances from the natural marine environment. Paralytic shellfish poisoning, ciguatera intoxication, tetrodotoxin poisoning, and scombroid poisoning have been known for many years and continue to be significant problems in many areas of this world. Less well recognized than the marine toxins are those toxins that can be found in the edible tissues of animals grazing on plants containing them. The oceans would be an inexhaustible source of human food but for the fact that thousands of species of marine organisms are too poisonous to consume. Only a small fraction of the compounds responsible for the toxicity of fish, molluscs, and other forms of ocean life have been identified, yet it is apparent that toxicants of marine origin display some of the most chemically and biologically complex properties of all known toxic chemicals. Many of these toxicants are endogenous constituents of seafood, but some of the important ones are actually metabolic products of marine organisms that are used as food by fish and shellfish. Ingestion of the organisms results in accumulation of their toxic metabolites in edible tissues, rendering them potentially hazardous. The paralytic shellfish toxins (PSTs) are the toxic contaminants of seafood of this group that have received the greatest public health attention and the greatest scientific investigation (Hayes and Campbell, 1986). The PSTs are metabolites of several varieties of plankton or dinoflagellates. The principal organisms producing these toxins are Gonyaulax catenella, G. tamarensis, G. acatenella,

226

FOOD

and several species of Gymnodinium. Under certain conditions, these organisms undergo rapid growth and color the ocean where this is occurring in various shades of red or yellow, creating the so-called “red-tide.” The concentrations of PSTs in shellfish growing under these conditions may rise to levels that are seriously hazardous for those who might consume the shellfish. Many species of mussels, clams, cockles, oysters, and scallops are susceptible in dozens of regions of the earth (Halsted, 1978). The PSTs are low-molecular-weight compounds (around 300) having extraordinarily potent neurotoxic properties. Seven such toxins have been identified and are simple chemical derivatives of saxitoxin, the principal member of the group. Saxitoxin is a tricyclic compound with a molecular weight of 283 containing two guanide functional groups. The estimated human LD50 for saxitoxin is 10–20 mg/kg. Although this value is not as low as some protein or polypeptide toxins, such as botulinum or cobra venom toxins, few compounds with molecular weights this low display such an extreme lethality. Unlike so many protein or polypeptide neurotoxicants, PSTs are chemically stable to a wide variety of conditions (Concon, 1988). Respiration may be turned off almost immediately following ingestion of PSTs, depending on the dose received, or may be only partially depressed. Victims who recover from the initial central nervous system (CNS) effects within 12–24 h generally do not suffer any further manifestation of poisoning. The PSTs, or at least saxitoxin, act as sodium channel blockers and appear to prevent entry of Naþ ions into the nervous system. The initial increase in sodium permeability associated with excitation is impaired, and nerve impulse transmission is blocked without depolarization. Paralysis of diaphragmatic muscles ensues. Hypotension is also frequently associated with PST poisoning (Concon, 1988). In the United States, and in most areas of the world, PST poisoning is now quite uncommon, but this is only because the Public Health Service and similar agencies worldwide have maintained programs of shellfish monitoring and quarantine of affected beds. The cost of PST is thus not so much in human lives but rather an economic loss of potential sources of food (Halsted, 1978). Ciguatera is a group of marine toxicants that are probably of algal origin and move up the marine food chain from smaller to larger fish. Their neurotoxicity appears to be based on anticholinesterase effects, although there are surely more complex pharmacological activities involved. Indeed, ciguatera poisoning is manifested in several ways, with GI distress, leg cramps, burning sensations on the tongue and skin, metallic taste, dental pain, myalgia, headache, vertigo, and chills among the most common symptoms. Recovery from severe poisoning can be very prolonged (Concon, 1988). Dozens of common fish species, particularly those of tropical origin, are reported as ciguatoxic every year (Halsted, 1978; Lee, 1980). As in the case of PSTs, close monitoring is essential in preventing human poisonings. The chemistry of ciguatera poisons remains obscure, and analytic methods are accordingly less than ideal (Concon, 1988). Many other examples of toxic contaminants of animal foods could be cited, including substances present in range plants that may remain as residues in tissues of grazing cattle or sheep. This is a little-explored area, in part because natural chemicals of marine or plant origin tend to be highly complex substances, and this creates difficulties for the analytical chemists. The importance of studies in animals, such as those reported by Tryphonas et al. (1990), cannot be overemphasized. These investigators found that the clinical picture in monkeys reflected that observed in patients who have recovered from toxic-mussel-induced brain damage from domoic acid. The establishment of this model makes it possible to clarify the role of certain complicating factors in fatal human cases.

CONSTITUENTS AND CONTAMINANTS OF NATURAL ORIGIN

7.6.3

227

Compounds Produced During Food Storage or Preparation

N-nitrosamines and acrylamide: Substances such as oxidized fats, the polycyclic aromatic hydrocarbons, several types of heterocyclic compounds resulting from reactions taking place during cooking, the N-nitrosamines, and a variety of vasoactive amines are produced during food storage or preparation. The compounds that arise in processes that have been in use by man for a long time are considered here to be of natural origin, although in theory some might legally be considered food additives or contaminants of industrial origin (Roberts, 1981). Oxidized fats, which include a variety of epoxides and peroxides, pyrolysis products of L-tryptophan and other amino acids, and polycyclic aromatic hydrocarbons resulting from heating of foods, are all well-known and well-studied examples of potential food risks created during food storage or preparation (Concon, 1988). The N-nitrosamines, along with the related nitrosamines and nitrosoguanidines, are of particular interest because so many members of this class are animal carcinogens, some of considerable potency, and because some can form both in foods and in the alimentary tract when the appropriate precursors and conditions are present (Tannenbaum, 1988). The N-nitrosamine precursors along with nitrite are either endogenous to or formed in many foods. Included are creatine, sarcosine, proline, pyrrolidine, and piperidine (meat and meat products), methyl guiaidine, diethylamine, and trimethylamine (fish), diethylamine and dipropylamine (cheese), and choline and lecithin (eggs and meat). There are also reports of N-nitrosamine formation from certain pesticides containing secondary amine of carbamate functions (Concon, 1988; Grasso, 1984). The nitrosamines most commonly reported in foods are dimethylnitrosamine (DMN), diethylnitrosamine (DEN), nitrosoproline (N Pro), and nitrosopyrrolidine (N Pyr). The levels of these tend to fall below 10 ppb, although levels up to several hundred micrograms per kilogram have been detected in certain smoked and nitrate or nitrite-treated foods. In vivo formation of N-nitroso compounds from precursors associated with foods apparently creates a greater exposure than that created by preformed products (Grasso, 1984). Dietary nitrosamine formation in food can be inhibited in several ways, most particularly by limiting the acidity. But, in many situations, this is not practically achievable. Ascorbic acid is an effective inhibitor because it reacts quickly with nitrite to form nitric oxide and dehydroascorbic acid. There is some evidence that ascorbic acid also reduces in vivo formation of nitrosamines, although not all attempts to achieve such an effect experimentally have been successful. It is not clear whether the use of this vitamin results in a significant reduction in the human exposures to nitrosamines. The principal reason for concern over nitrosamine formation in foods and beverages, and in other consumer and industrial products as well, is the marked carcinogenic activity exhibited by so many of these compounds in animals. Indeed, it is difficult to identify species of experimental animals that do not develop excess rates of tumors in response to exposures to nitrosamines and their chemical relatives. Typical targets following ingestion include the oral cavity, larynx, trachea, esophagus, liver, kidney, and skin. The multipotent properties of these substances are exemplified by DEN, which produces excess tumors at several sites in rats, mice, hamsters, guinea pigs, rabbits, dogs, monkeys, and several nonmammalian species. Although epidemiological evidence of nitrosamine involvement in human cancers is limited, it is difficult to imagine, given the nature of the animal evidence, that humans are somehow not among the susceptible species (Grasso, 1984; NRC, 1996). Recently it has been suggested that red meat enhances colonic formation of N-nitrosamines (Lewin et al., 2006).

228

FOOD

Acrylamide has been known for many years to be a cause of neurological disorders in excessively exposed production workers. The compound is also an animal carcinogen. In 2002 investigators in Sweden reported that the compound could be formed by reactions occurring when certain foods are heated to high temperatures. Starchy foods such as potato chips, french fries, cereals, and breads seem to be major sources; intakes from the sources seem to exceed guidelines that the EPA has established for drinking water. Levels in some heated foods have been reported to exceed 500 mg/kg. The compound seems to result from the reaction between glucose and the amino acid asparagine. Public health agencies have not published quantitative estimates of risks related to these surprising sources of acrylamide, but have made statements concerning the pressing need to understand both threats to health and control measures (www.ifst.org/acrylmd.htm). Many other compounds are produced during food processing, particularly when it entails high temperatures. The polycyclic aromatic hydrocarbons have been most widely investigated, with many others and under increasing scrutiny. 7.6.4

Dietary Supplements

In 1994, Congress passed the Dietary Supplement Health and Education Act (DSHEA). DSHEA was a compromise that tried to balance FDA’s need to regulate the dietary supplement industry with the public’s perception that access to these products should be given without undue or excessive restraint by the Agency. The definition of a dietary supplement is quite expansive and includes “a product intended to supplement the diet that bears or contains one or more dietary ingredients. . .a vitamin;. . .a mineral;. . .an herb or other botanical;. . .an amino acid;. . .a dietary substance for use by man to supplement the diet by increasing the total dietary intake; or. . .a concentrate, metabolite, constituent, extract, or combination [of any ingredient described above]” (Section 201(ff)(1) of the FDC Act, 21 U.S.C. x 321 (ff) (1)). There are additional criteria. The product must either be intended for ingestion in tablet, capsule, powder, softgel, gelcap, or liquid droplet form, or, if not intended for ingestion in such a form, not be “represented for use as a conventional food or as a sole item of a meal or the diet.” Pivotal to the regulation of dietary supplements is the provision that those containing new dietary ingredients (NDI) are subject to premarket FDA notification. An NDI is defined as “a dietary ingredient that was not marketed in the United States before October 15, 1994” presumably requiring the manufacturer or distributor to have written or other evidence that the ingredient in question is chemically identical to a dietary ingredient that was marketed in the United States before that date. A dietary supplement that contains an NDI is deemed to be adulterated unless, either (1) the supplement “contains only dietary ingredients which have been present in the food supply as an article used for food in a form in which the food has not been chemically altered,” or (2) there is a “history of use or other evidence of safety establishing the at the dietary ingredient when used under the conditions recommended or suggested in the labeling. . .will reasonably be expected to be safe.” The definition of an NDI brings up some difficult questions for both manufacturers of these products and for the FDA. The definition of “chemically altered” does not include the following physical modifications: minor loss of volatile components, dehydration, lyophilization, milling, tincture, or solution in water, slurry powder, or solid in suspension. This listing only explains what is not included in the definition, and leaves much room for debate over both the definition of a NDI, as well as the appropriate safety standard for these

FOOD SAFETY IN THE EUROPEAN UNION

229

ingredients. As written in the regulation, safety is defined as “reasonably expected to be safe under the conditions of use defined in the labeling.” The extent of scientific evidence to demonstrate safety, however, is undefined, although it is clear that FDA bears the burden of proof if it asserts that a dietary supplement is not safe under this standard. The value of understanding when an ingredient is “new,” and what the appropriate standard for safety should be, is more than a semantic argument. It is operational, in that it defines when new safety information must be generated and what safety information is sufficient to establish “reasonable expectation of safety” to protect the consumer. Current interpretation of the standard of safety for NDIs is revealed in the ruling on the litigation challenging the validity of the FDA’s February 2004 Final Rule that declared ephedra alkaloid dietary supplements (EDS) adulterated (unsafe) and not legally marketable in the United States. In the Final Rule, the FDA had concluded that when the minimal benefits of EDS were weighed against the substantial risks, EDS presented an unreasonable risk of illness or injury under the conditions of use recommended, and were therefore adulterated under DSHEA. The issues before the court were (1) whether FDA’s use of a risk–benefit analysis was appropriate under DSHEA; and (2) whether FDA had provided sufficient evidence to support the conclusion that EDS containing 10 mg or less per day of ephedrine alkaloids posed a significant or unreasonable risk of illness or injury. FDA had concluded that the words “significant” and “unreasonable” had two separate and independent meanings. FDA’s interpretation of a significant risk involved an evaluation of risk alone while unreasonable risk required a comparison of risks and benefits. FDA took the position that since EDS posed an unreasonable risk, it was not necessary to address the DSHEA’s significant risk standard. The court found that DSHEA contains no risk–benefit provision. A dietary supplement, as with any food, is presumed to be safe and is not required to establish benefit before sale. Therefore, FDA’s definition of “unreasonable” entailing a risk–benefit analysis was considered inappropriate. The court also found that although FDA could not determine a safe level, a negative inference is different from the affirmative proof of “significant or unreasonable” risk that is required to support a finding of adulteration. The court did not support FDA’s position in the Final Rule, thus providing current guidance on the regulation of dietary supplements as foods.

7.7 FOOD SAFETY IN THE EUROPEAN UNION 7.7.1

Legislation

Food safety is at the forefront of recent European Commission efforts to regulate chemicals. The general principles and requirements of food law are laid down in EU-Regulation 178/ 2002/EC, also referred to as the White Book on Food Safety. With this regulation, a comprehensive approach has been adopted for all food-related legislation in the EU and its member states. The aim of this regulation is to provide “the basis for the assurance of a high level of protection of human health and consumers’ interest in relation to food, taking into account in particular the diversity in the supply of food including traditional products, whilst ensuring the effective functioning of the internal market.” The guiding principle of the regulation is that food safety policy should be based on a comprehensive, integrated approach. This means that the origin of the food and the route it follows from farmer through the food industry and then to consumers has to be clear to all

230

FOOD

stakeholders, and that interventions to ensure food safety may occur at any of the stages “from farm to fork.” Where the primary responsibility for food safety is with the food and feed business, there must be adequate and effective controls organized by the competent authorities of the member states to complement and support this responsibility. Finally, consumers have their own responsibility for the proper storage, handling, and cooking of the food. The EU has based its food safety policy on the application of the three components of risk analysis: risk assessment, risk management, and risk communication. The precautionary principle is to be applied in risk management decisions when appropriate. Thus, a product or ingredient can be removed from the market when doubt about its safety arises, even when sufficient data to support a finding of excessive health risk are not available. When additional data are available to remove doubts about a product’s safety, the product may be marketed within the EU. Other factors relevant for the health protection of consumers can also be taken into account, such as environmental consequences, animal welfare, sustainable agriculture, and consumers’ expectations regarding product quality and information about the essential characteristics of products and of their production methods. The EU’s food safety policy appears to include a higher level of integration and comprehensiveness than can be found in U.S. policy. Risk assessments generally take the form of those conducted in the United States. There are no specific requirements for toxicological information; risk assessments will be performed with the information available. If risk cannot be fully assessed because of insufficient or inconclusive data, the precautionary principle may be applied. Interested parties may submit information to support better risk assessments and the substance under evaluation may be then authorized for marketing. The precautionary principle is a risk management tool, to be applied by decision makers, and is not to be confused with uncertainty measures that are used in risk assessment. These uncertainty factors are, for example, applied to account for inter- and intraspecies variability, when an ADI of a substance is derived. Another option, when an inadequate scientific database exists, is not to adopt an ADI but rather to apply the “ALARA” (as low as reasonably achievable) principle. This principle is often used as a basis for setting maximum levels of certain toxic contaminants, which cannot be entirely removed from food. Specific regulations exist for the regulation of novel foods, foods for particular nutritional uses, food additives, food supplements (only vitamins and minerals), and contaminants. Some aspects of these regulations are discussed below.

7.7.2

Novel Food Regulation (258/97/EC)

The novel food regulation is intended to manage the marketing of foods and food ingredients, which were not used to a significant degree for human consumption before 1997 within the EU community. Six categories have been distinguished: a. Foods and food ingredients containing or consisting of genetically modified organisms (GMO’s). b. Foods and food ingredients produced from, but not containing, genetically modified organisms (GMO). c. Foods and food ingredients with a new or intentionally modified primary molecular structure.

FOOD SAFETY IN THE EUROPEAN UNION

231

d. Foods and food ingredients consisting of, or isolated from, microorganisms, fungi or algae. e. Foods and food ingredients consisting of, or isolated from plants and food ingredients isolated from animals, except for foods and food ingredients obtained by traditional propagating or breeding practices and having a safe history of use. f. Foods and food ingredients to which has been applied a production process not currently used, where that process gives rise to significant changes in the composition or structure of the foods or food ingredients which affect their nutritional value, metabolism or level of undesirable substances. Those seeking to market a novel food are required to submit a dossier. A safety assessment, following the SAFEST concept (safety assessment of food by equivalence and similarity Targeting), is to be included in the dossier. This concept was developed by the ILSI Europe Task Force on Novel Foods (Jonas et al., 1996) and embraces both a toxicological assessment and a nutritional assessment. The SAFEST concept uses traditional foods, accepted as safe in use, as a basis for comparison. The traditional counterpart of a novel food or food ingredient is to be chosen so as to reflect not only its chemical composition, but also its intake, its role in the diet, and the effects of processing. The following SAFEST classes are distinguished: SAFEST Class 1: Substantially equivalent to a traditional counterpart, no further information required. SAFEST Class 2: Sufficiently similar to a traditional counterpart, focused safety evaluation needed, taking into account available literature on the safety of the component. Further testing may be necessary. SAFEST Class 3: Insufficiently similar to a traditional counterpart, safety assessment may have to be far more extensive, depending on the nature of the novel food or ingredient. Further testing likely to be necessary. Toxicological information may be required for novel foods or ingredients in SAFEST classes 2 and 3. For novel foods or ingredients in SAFEST class 2, the toxicological evaluation focuses on the identified differences between the novel food and the traditional counterpart, whereas the toxicological evaluation for novel foods or ingredients in SAFEST class 3 may require a more extensive test program. Studies that may be required are, for example, toxicokinetics, genotoxicity, allergenicity, potential for colonization, pathogenicity, 90-day subchronic feeding study in rodents or other toxicity studies, and confirmation of safety in humans (e.g., tolerance, examination of effects on intestinal microflora, effects on biomarkers). Several novel foods have already been authorized in the EU: sweet corn from genetically modified maize line Bt11, several products with added phytosterols and phytostanols, salatrim (acronym for short- and long-chain acyl triglyceride molecules), noni juice, and several others. 7.7.3

Food Additives

Directive 89/107/EEC is a comprehensive framework regulating the use of food additives as ingredients during the manufacture or preparation of food, and which become part of the

232

FOOD

finished product. In this respect, a food additive can be regarded as any substance that is not normally consumed as food itself, but is added intentionally and in this way results in becoming an ingredient. The only substances that may be used as food additives are those included in approved lists, and only under the conditions of use mentioned in the lists. These lists can be found in Directive 95/2/EC (general additives, such as emulsifiers, preservatives, thickeners), Directives 94/36/EC (colors), and Directives 94/35/EC and 2003/115/EC (artificial sweeteners). To introduce a new additive into the EU market, a dossier must be submitted, similar in content to a Food Additive Petition in the United States. It should be emphasized that substances proposed as food additives may only be authorized for use if a reasonable case of technological need can be demonstrated and if the food additive does not present a risk to the health of consumers at the proposed level of use. Toxicological studies that are required for authorization of a new food additive are not fixed, but depend, as in the United States, on the chemical nature of the additive and its proposed uses and levels of use in food. A guidance document is available containing a general framework from which can be determined which studies will be required to establish the safety of the additive, (Scientific Committee on Food, 2002). Principally, these guidelines follow the guidance published by the Joint FAO/WHO Expert Committee on Food Additives (JECFA). Core studies include information on metabolism/toxicokinetics, subchronic toxicity, genotoxicity, chronic toxicity, and carcinogenicity, and reproduction and developmental toxicity. In addition, other studies that may be helpful or necessary for certain substances may be required, for example studies on immunotoxicity, allergenicity, neurotoxicity, human volunteer studies, or other special studies. A dossier containing a review of the study results must be submitted, containing an overall evaluation of human risk. Determinations on whether safety criteria have been satisfied in particular cases are left to competent authorities in the EU and its Member States. 7.7.4

Food Supplements

Food supplements are concentrated sources of nutrients or other substances with a nutritional or physiological effect whose purpose is to supplement the normal diet, and which is marketed as, for example, pills, tablets, capsules, or liquids in measured doses. A list of permitted vitamin and mineral substances that may be sold as food supplements for nutritional purposes is available in Annex II of Directive 2002/46/EC. However, until December 31, 2009, the use of vitamins and minerals that are not listed in this Annex may still be allowed by member states in their territory, provided that the substance is used in one or more food supplements marketed in EU on the date of entry into force of the Directive, and that the EFSA has not given an unfavorable opinion in respect of the use of the substance. Furthermore, specific labeling requirements apply. To consider a substance for inclusion on the list, a dossier must be submitted for safety evaluation. The dossier should include biological and toxicological data, and information on possible sources that consist of, contain, or are derived from genetically modified organisms. Concerning the toxicological data, the guidance document states that in first instance, available data should be submitted. This means that, again, there is no specific guidance on minimal data requirements, and only the available information should be submitted for evaluation. Then, depending on safety

FOOD SAFETY IN THE EUROPEAN UNION

233

considerations in relation to the fate of the substance in the body, data requirements may be extended. 7.7.5

Contaminants

In Regulation 466/2001/EC, maximum levels are set for certain contaminants in foodstuffs. The maximum levels are intended to reduce the presence of these contaminants to the lowest levels reasonably achievable by means of good manufacturing or agricultural practices, in order to achieve a higher level of health protection, especially for sensitive population groups. Sampling and analysis methods for evaluating contaminant levels are proposed; member states are obliged to take appropriate monitoring measures regarding the presence of contaminants in foodstuffs. Four different categories of contaminants are at present listed in regulation: nitrates, aflatoxins, heavy metals (lead, cadmium, mercury) and 3-monochloropropane-1,2diol (3-MCPD). Amendments to the regulation include maximum levels for dioxins, ochratoxin A, patulin, and inorganic tin. .

.

.

.

.

.

Nitrate: In order to reduce the levels of nitrates in vegetables (mainly spinach and lettuce), the regulation stipulates that growing methods must be modified and codes of good practice applied. A transitional period enables member states to authorize marketing of territory spinach or lettuce with higher nitrate levels, as nitrate levels differ considerably depending on the climatic conditions. The Commission will review the maximum levels every 5 years. Aflatoxins: As aflatoxins are genotoxic carcinogenic substances, limits are set at the lowest possible level. Higher levels of aflatoxin are allowed for products such as groundnuts, nuts, dried fruit and maize, when they are not intended for direct human consumption or for use as an ingredient in foodstuffs. In such cases, they must bear a label showing their intended purposes and marked: “product that must be subjected to a sorting treatment or other physical treatments aimed at reducing the level of aflatoxin contamination.” Heavy Metals: principally, heavy metals maximum levels should be as low as reasonably achievable. 3-Monochloropropane-1,2diol (3-MCPD): Under certain conditions, 3-MCPD can be created during food processing, in particular ruing the manufacture of the savory food ingredient “hydrolyzed vegetable protein,” by the acid hydrolysis method. The substance is a carcinogenic agent therefore, maximum levels should be set as low as reasonably possible. A significant decrease in 3-MCPD productions has already been achieved by adjusting the production process. Ochratoxin A: A natural mycotoxin, which is produced by several fungi (species “penicillium” and “aspergillus”), and has carcinogenic, nephrotoxic, teratogenic, immunotoxic, and possibly neurotoxic properties, moreover, it has also been associated with nephropathy in humans. As it occurs naturally in many plant products from all over the world (cereals, coffee beans, cocoa, dried fruit, etc.), foodstuffs are often contaminated with ochratoxin A. Maximum levels are set for cereal-based preparations, baby foods, and special dietary preparations for infants. Patulin: A mycotoxin produced by several fungi, which may be found in fruit juices, particularly apple juice, and moldy foods such as bread.

234 .

FOOD

Dioxins: Tolerably weekly doses are set regarding the presence of dioxins and dioxintype PCBs in food. The values are designed to limit the presence of dioxins in feed and food by means of an integrated approach throughout the food chain.

The regulation is regularly amended on the basis of new knowledge and monitoring results. Recently, the European Commission has set up a new database containing research on acrylamide in foodstuffs. Furthermore, data on the presence of polycyclic aromatic hydrocarbons (PAH) has been collected by member states, and the Commission is developing legislation to set maximum levels in particular for benzo(a)pyrene in certain foods. Also, investigations on the possible presence and risks of organotins in food are ongoing. These substances are found in the aquatic environment due to their presence in paints as antifouling agents, and are known endocrine disruptors.

7.8 SUMMARY AND CONCLUSION Because foods and beverages are so complex and variable in composition, health risks associated with them can be understood fully only through the continued pursuit of longterm epidemiological investigations. There seems little doubt that the composition of the human diet strongly influences health status, in both positive and negative ways. Current evidence suggests that the major influences on long-term health status are those associated with total caloric intake and with the natural constituents of food, both nutritive and nonnutritive. A large impact from additives and contaminants seems unlikely, though the relative importance of these constituents, especially contaminants of both industrial and natural origin, varies considerably among the geographical regions of the earth. Recent evidence of large-scale pollution in rapidly developing countries suggests an increasing likelihood that food contamination could become a serious public health problem. As global trade in basic food commodities continues to increase, so will adverse public health impacts spread. Contamination by human pathogens––a not insignificant problem even in developed countries––is, on a global scale, almost certainly the most significant acute health problem associated with food. A review of what is known and unknown about the risks associated with the chemical constituents and contaminants of food, as has been attempted in this chapter, demonstrates that, on a chemical-specific basis, far more study has been devoted to substances intentionally added to food than to substances naturally present or contaminating food. This observation is confirmed by a recent NRC review of carcinogens and anticarcinogens in the diet (NRC, 1996). This state-of-affairs is perhaps largely explained by the fact that the laws under which foods are regulated, not only in the United States but around the world, require much closer examination of added substances, as we have earlier explained. The goal of understanding the effects on health of the diet as a whole, and of its myriad natural constituents and of its contaminants, is largely dictated by choices made in the research community and its funding agencies. If trends of the past decade are suggestive of the future, we can expect in the next decade or two a vastly increased understanding of the type of diet needed to maximize health benefits and to minimize the risks of chronic diseases. Of course, such an understanding will not, of itself, change individual behavior or that of the food production and distribution system; but without that understanding there is little hope for beneficial change.

REFERENCES

235

ACRONYMS ADI ARMS BHA CFSAN CNS DEN DMN DSHEA EDI EDS EFSA EPA EU FAO/WHO FDA FEMA FSIS GRAS IFBC JECFA NCI NDI NOAELS N Pro N Pyr NRC PBDE PCB PST RfD SAFEST SCOGS TSH T3 T4

Acceptable daily intake Adverse reaction monitoring system Butylated hydroxyanisole Center for Food Safety and Applied Nutrition Central nervous system Diethylnitrosamine Dimethylnitrosamine Dietary Supplement Health and Education Act Estimated Daily Intake Ephedra alkaloid dietary supplement European Food Safety Agency Environmental Protection Agency European Union Food and Agriculture Organization/World Health Organization Food and Drug Administration Flavor and Extract Manufacturers Association Food Safety and Inspection Service (USDA) Generally recognized as safe International Food Biotechnology Council Joint Expert Committee on Food Additives National Cancer Institute New dietary ingredient No observed adverse effect levels Nitrosoproline Nitrosopyrrolidine National Research Council Polybrominated diphenylethers Polychlorinated biphenyls Paralytic shellfish toxins Reference dose Safety assessment by equivalence and similarity targeting Select Committee on GRAS Substances Thyroid-stimulating hormone Triiodothyronine Thyroxine

REFERENCES Ames BN (1983a) Dietary carcinogens and anti-carcinogens. Science 221:1256–1264. Ames BN (1983b) Dietary carcinogens and anti-carcinogens, and changing patterns in cancer: some speculation. Environ. Res. 50(2):322–340. Ames BN, Gold LS (1989) Pesticides, risk and apple sauce. Science 244:757

236

FOOD

Bechtel DH, Wilcock KE (1990) Hepatitis B-independent hepatic cancer risk in aflatoxin exposed humans with estimation of a no-significant risk level. Paper presented at meeting of the Society Independent Epidemiologic Research. Snowbird, Utah. Brock WJ et al. (2003) Food safety and risk assessment. Int. J. Toxicol. 22:435–451. Committee on Food Protection (1973) Toxicants Occurring Naturally in Foods.Washington, DC: National Academy of Sciences. Concon J (1988) Food Toxicology. New York: Marcel Dekker. pp.511–603. Davis DL (1989) Natural anticarcinogens, carcinogens, and changing patterns in cancer: some speculations. Environ. Res. 50:322–340. Doull J (1981) Food safety and toxicology. In Roberts HR, editor. Food Safety. New York: John Wiley & Sons. Dourson ML, Stara JF (1983) Regulatory history and experimental support of uncertainty (safety) factors. Regul. Toxicol. Pharmacol. 3:224–238. Dourson ML, Felter SP, Robinson D (1996) Evolution of science-based uncertainty factors in noncancer risk assessment. Reg. Toxicol. Pharmacol. 24:108–120. EPA (1988) Superfund Exposure Assessment Manual. EPA-540/1-88/001. Washington, DC: Office of Emergency and Remedial Response. EPA (1997) Draft Health Assessment Document for 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) and Related Compounds. Chap. 8.Office of Research and Development.Washington, DC. FDA (1979a) Chemical compounds in food producing animals. Federal Register 44:17070– 17112. FDA (1979b) Polychlorinated biphenyls (PCBs): Reduction of tolerances, final rule. Federal Register 44:37336–37403. FDA (1981) Aspartame: Commissioner’s final decision. Federal Register 46:38283–38308. FDA (Bureau of Foods) (1982a) Toxicological Principles for the Safety Assessment of Direct Food Additives and Color Additives Used in Food. Washington, DC: U.S. Food and Drug Administration. Proposed revisions published by FDA (1993). FDA (1982b) Policy for regulating carcinogenic chemicals in food and color additives: Advance notice of proposed rulemaking. Federal Register 47:14464–14470. FDA (1983) Food Additives Permitted for Direct Addition to Food for Human Consumption.Section 172.804. Aspartame. Title 21. Washington, DC: Code of Federal Regulations. FDA (1986) Vinyl chloride polymers. Federal Register 51:4173–4185. FDA (1988) Recommendations for chemistry data for indirect food additives. Washington, DC: Center for Food Safety and Applied Nutrition. FDA (1990a) Color additives: Denial of petition for listing of FD&C Red No. 3 for use in cosmetics and externally applied drugs; withdrawal of petition for use in cosmetics intended for use in the area of the eye. Federal Register 55:3520–3543. FDA (1990b) Termination of provisional listings of FD&C Red No. 3 for use in cosmetics and externally applied drugs and of lakes of FD&C Red No. 3 for all uses. Federal Register 55:3516–3519. Fitzhugh OG, Nelson AA (1947) The comparative toxicities of fumaric, tartaric, oxalic and maleic acids. J. Am. Pharm. Assoc. 36:217–319. Food Chemical News (Sept. 8, 1997) EPA Still Wondering about Dioxins in Mississippi Clay. Washington, DC: CRC Press.p. 4. Food Chemical News (1970) Evaluating the Safety of Food Chemicals.Washington, DC:National Academy of Sciences/National Research Council. Food Protection Committee (1970) First National Conference for Food Protection. Washington, DC: U.S. Department of Health and Human Services.

REFERENCES

237

Frankos V, Rodricks JV (2001) Food Additives and Supplements in Regulatory Toxicology. 2nd ed., Chap. 5. S Gad. editor. Taylor and Francis: New York. Gartrell MJ, Craun JC, Podrebarac DS, Gunderson EL (1986a) Pesticides, selected elements, and other chemicals in adult total diet samples, October 1980–March 1982. J. Assoc. Off. Anal. Chem. 69:146–161. Gartrell MJ, Craun JC, Podrebarac DS, Gunderson EL (1986b) Pesticides, selected elements and other chemicals in infant and toddler total diet samples, October 1980–March 1982. J. Assoc. Off. Anal. Chem. 69:123–145. Goldman M, Doering J, Nelson RG (1977) Effect of dietary ingestion of oxalic acid on growth and reproduction in male and female Long-Evans rats. Res. Commun. Chem. Pathol. Pharmacol. 18:369–372. Goyer R (1986) Toxic effects of metals. In: Klaassen CD, Amdur MO, Doull J, editors. Casarett and Doull’s Toxicology, New York: Macmillan. Grasso P (1984) Carcinogens in food. In: Searle CE, editor. Chemical Carcinogens. 2nd ed. ACS Monograph 182.Washington, DC: American Chemical Society. Guest GB, Fitzpatrick SC (1990) Overview on bound residue issue––regulatory aspects. Drug Metab. Rev. 22 (6–8):595–599. Hall RL (1979) Food additives and ingredients. In:Clydesdale FM, editor. Food Science and Nutrition: Current Issues and Answers. Englewood Cliffs, NJ: Prentice-Hall. Halsted B (1978) Poisonous and Venomous Marine Animals of the World. Rev. ed. Princeton, NJ: Darwin Press. Hayes JR, Campbell TC (1986) Food additives and contaminants. In: Klaassen DC, Amdur MO, Doull J, editors. Casarett and Doull’s Toxicology. New York: Macmillan. Hill RN, Erdreich LS, Paynter OE, Roberts PA, Rosenthal SL, Wilkinson CF (1989) Thyroid follicular cell carcinogenesis. Fund. Appl. Toxicol. 12:629–697. International Food Biotechnology Council(1990) Biotechnologies and food: Assuring the safety of foods produced by genetic modification. Regul. Toxicol. Pharmacol. 12: part 2. IOM (1996) Dietary Reference Intakes, Vol. I. Washington, DC: National Academy Press. IOM (1998) Dietary Reference Intakes, Vol. II. Washington, DC: National Academy Press. IOM (1998) Ensuring Safe Food: From Production to Consumption. Washington, DC: National Academy Press. IOM (1997–2002) See the series of 6 volumes: Dietary Reference Intakes. Washington, DC: National Academy Press. Joint Codex Alimentarius Commission (1979) Guide to the Safe Use of Food Additives. Rome: Food and Agricultural Organization of the United Nations. Jonas DA, et al (1996) The safety assessment of novel foods. Food Chem. Toxicol. 34:931–940. Kokoski CJ, Flamm WG (1984) Establishment of acceptable limits of intake. InProceedings of the Second National Conference for Food Protection. Washington, DC: DHHS. Kotsonis FN, Burdock GA, Flamm WG (2001) Food Toxicology. In:Klaassen CD, editor. Casarett and Doull’s Toxicology.6th ed.McGraw Hill. New York. Lee C (1980) Fish poisoning with particular reference to ciguatera. J. Trop. Med. Hyg. 83:93–77. Lewin MH, et al(2006) Red meat enhances the colonic formation of the DNA adduct O6 carboxy-methyl guanine: implications for colorectal cancer risk. Cancer Res .66:1859– 1866. Maxim LD (1989) Problems associated with the use of conservative assumptions in exposure and risk analysis. In: Paustenbach DJ, editor. The Risk Assessment of Environmental Hazards. New York: John Wiley & Sons.

238

FOOD

Menzer RE, Nelson JO (1986) Water and soil pollutants. In: Klaassen CD, Amdur MO, Doull J, editors. Casarett and Doull’s Toxicology. New York: Macmillan. Merrill R (1979) Regulating carcinogens in food: A legislator’s guide to the food safety provisions of the Federal Food, Drug, and Cosmetic Act. Mich. Law Rev. 77:179–184. Merrill R (1996) Regulatory Toxicology. In: Klaassen CD, Amdur MO, Doull J, editors. Casarett and Doull’s Toxicology. 5th ed. New York: Macmillan. Munro I, Charbonneau SM (1981) Environmental contaminants. In: Robert HR, editor. Food Safety. New York: John Wiley & Sons. Munro IC (1990) Issues to be considered in the safety evaluation of fat substitutes. Food Chem. Toxicol. 28:751–753. National Academy of Sciences (NAS) (1987) Poultry Inspection: The Bases for a Risk Assessment Approach. Washington, DC: National Academy Press. NRC/NAS (1981) The Health Effects of Nitrate, Nitrite, and N-Nitroso Compounds.National Aca-demy Press. Washington, DC: NRC (1991) Improving America’s Diet and Health. Washington, DC: National Academy Press. NRC (1996) Understanding Risk. Washington, DC: National Academy Press. NRC. National Research Council (2002). Toxicological Effects of Methylmercury. Washington: National Academy Press. Oser B, Hall RL (1977) Criteria employed by the expert panel of FEMA for the GRAS evaluation of flavouring substances. Food Cosmet. Toxicol. 15:457–466. Pao EM, Fleming KH, Guenther PM, Mickle SJ (1982) Foods Commonly Eaten by Individuals: Amount Per Day and Per Eating Occasion. Washington, DC: U.S. Department of Agriculture. Park DL, Stoloff L (1989) Aflatoxin control––How a regulatory agency managed risk from an unavoidable natural toxicant in food and feeditor. Regul. Toxicol. Pharmacol. 9:109–130. Paynter OE, Burin GJ, Jaeger RB, Gregorio CA (1988) Goitrogens and thyroid follicular cell neoplasia: Evidence for a threshold process. Regul. Toxicol. Pharmacol. 8:102–119. Reddy BS, Cohen LA, editors. (1986) Diet, Nutrition and Cancer: A Critical Evaluation. Boca Raton, FL: CRC Press. Roberts HR (1981) Food safety in perspective. In: Roberts HR, editor. Food Safety. New York: John Wiley & Sons. Rodricks JV (1981) Regulation of carcinogens in food. In:Nicholson WS, editor. Management of Assessed Risks from Carcinogens. New York: New York Academy of Sciences. Rodricks JV, Pohland AD (1981) Food hazards of natural origin. In:Roberts HR, editor. Food Safety. New York: John Wiley & Sons. Rodricks JV, Park DL (1983) General aspects of food safety and an illustration using aflatoxin. Naguib K, Park DL, Pohland AE, Proceedings of the International Symposium on Mycotoxins, Cairo. pp.1–22. Rodricks JV (1988) Origins of risk assessment in food safety decision-making. J. Am. Coll. Toxicol. 7:539–542. Rodricks JV, Taylor M (1989) Comparison of risk management in U.S. regulatory agencies. J. Haz. Mater. 21:239–253. Rodricks JV, Jackson B (1992) Food constituents and contaminants. In: Lippman M, editor. Environmental Toxicants: Human Exposures and their Health Effects. New York: Van Nostrand Reinhold, pp.266–298. Rodricks JV (1994) Health and nutrition. In:Eblen RA, Eblen WR, editors. The Encyclopedia of the Environment. Boston and New York: The Rene Dubos Center for Human Environment, Houghton Mifflin Co. pp.321–324.

REFERENCES

239

Scientific Committee on Food (2002) White Paper on Food Safety, Commission of the European Communities, Brussels, Belgium. Select Committee on GRAS Substances (1981) Evaluation of health aspects of GRAS food ingredients: Lessons learned and questions unanswer. Feditor. Proc. 36:2519–2562. Sheikh-Omar AR, Schiefer HB (1980) Effect of feeding oxalic acid to pregnant rats. Pertanika 3:25– 31. Smith RL (1991) Does one man’s meat become another man’s poison?Trans. Med. Soc. Lond. Nov 11.6–17. Stagink LD (1979) Aspartame metabolism in human subjects. In:Guggenheim B, editor. Health and Sugar Substitutes. Basel, Switzerland: S. Karger. Stoloff L (1977) Aflatoxins––an overview. In: Rodricks JV, Hesseltine CW, Mehlman M, editors. Mycotoxins in Human and Animal Health. Park Forest South, IL: Pathotox Publishing. Stoloff L (1989) Aflatoxin is not a probable human carcinogen: The published evidence is sufficient. Regul. Toxicol. Pharmacol. 10:272–83. Tannen baum S (1988) N-nitroso compounds and carcinogenicity. Toxicol. Appl. Pharmacol. 31:1782–1791. Tennant DR, editor. (1997) Food Chemical Risk Analysis. Black’s Academic & Professional, London. Tollefson L (1988) Monitoring adverse reactions to food additives in the U.S. Food and Drug Administration. Regul. Toxicol. Pharmacol. 8:438–446. Tryphonas L, Truelove J, Iverson F (1990) Acute parental neurotoxicity of domoic acid in cynomolgus monkeys (M. fascicularis). Toxicol. Pathol. 18:297–303. Van Rensburg SJ, Cook-Mozafarri P, Schalwyk DJ, Van DerWatt JJ, Vincent TJ, Purchase IF (1985) Hepatocellular carcinoma and dietary aflatoxin in Mozambique and Transkei. Br. J. Cancer 57:713–726.

8 VOLATILE ORGANIC COMPOUNDS AND SICK BUILDING SYNDROME Lars Mølhave

8.1 INTRODUCTION 8.1.1

Volatile Organic Compounds

Volatile organic compounds (VOC) are frequent air pollutants in nonindustrial environments. A working group of World Health Organization (WHO) categorized the entire range of organic indoor pollutants into four groups, as indicated in Table 8.1 (WHO, 1989). No sharp limits exist between the categories, which were defined by boiling-point ranges. The VOC category was defined by a boiling-point range with a lower limit between 50 and 100 C and an upper limit between 240 and 260 C, where the higher values refer to polar compounds (WHO, 1989). 8.1.2

Health

Health has been defined by WHO (1961) as “a state of complete physical, mental, and social well-being and not merely the absence of disease or infirmity.” The toxic effects of VOCs may be divided into (1) those common to most VOCs; and (2) special effects caused by individual compounds. The special effects of individual compounds may be effects such as cancer or allergy that have been associated with specific volatile organic compounds, or effects caused by an unusually high potential of some compounds to cause more commonly seen effects. Reviews of such special effects are found in textbooks (e.g., Andrew and Snyder, 1980). A review of IAQ related health effects is found in Berglund et al. (1992). Special effects, such as genotoxic effects, or effects on the immune system, are severe for the few unlucky individuals affected by them. They are, from what we know, rare effects in relation to low-level VOC exposures in normal indoor environments and are not discussed further in this chapter. Such effects may, however, evade recognition if their prevalence is too

Environmental Toxicants, Third Edition Edited by Morton Lippmann Copyright
2009 John Wiley & Sons, Inc.

241

242

VOLATILE ORGANIC COMPOUNDS AND SICK BUILDING SYNDROME

TABLE 8.1

Classification of Indoor Organic Pollutants

Description Very volatile (gaseous) organic compounds Volatile organic compounds Semivolatile organic compounds Organic compounds associated with particulate matter or particulate organic matter

Abbreviation

Boiling-Point Range ( C)a

VVOC VOC SVOC POM

<0 to 50–100 50–100 to 240–260 240–260 to 380–400 >380

a Polar compounds appear at the higher end of the range. Source: WHO (1989).

small to allow a statistically significant association in a population of the size found in most normal nonindustrial buildings.

8.2 PREVALENCE OF EXPOSURES TO VOLATILE ORGANIC COMPOUNDS 8.2.1

Exposures to VOC in Nonindustrial Buildings

Building and furniture materials are known to emit VOCs. Ventilation transports outdoor pollutants to the indoor environment. The ventilation system itself may be a source of VOCs (Mølhave and Thorsen, 1991). Any human activity is a potential source of such volatile organic compounds. Maintenance, cleaning, and cooking create their own sources. Human metabolism and human activities, such as smoking, are other sources of gases and vapors. To these sources may be added copy machines, printing machines, glue, spray cans, and so on. From a number of early small-scale studies, it became evident that the concentration of many organic compounds in indoor air exceeds that in the outdoor air. An early review listed 307 VOCs identified in indoor air in different countries (Berglund et al., 1986). A WHO report (WHO, 1989) on VOCs indoors summarized the concentrations found in four major studies. These studies (Krause et al., 1987; De Bortoli et al., 1986; Lebret et al., 1986; Wallace, 1987) were used to construct one data set for each component, representative of an average home. The data set includes percentiles of the concentration distribution for individual compounds (WHO, 1989). These and a few other publications have been reviewed (Mølhave, 1986). Table 8.2 shows the total concentrations (TVOC) reported by Mølhave (1986) plus a few other buildings (Mølhave et al., 1982; Miksch et al., 1982). The measurement sites have been divided into dwellings, offices, and schoolrooms. The concentrations in older houses (range 0.02–1.7 mg/ m3) seem to be about 1/10 of that found in new houses, where the range of concentrations is from 0.5 to 19 mg/m3. Corresponding occupational total concentrations will typically be in the range of 0.1–1 times their respective occupational threshold limit values (TLV), which are about 40–400 mg/m3 (ACGIH, 2006). The question, therefore, is whether the nonindustrial concentrations are so low that the VOCs have no effect by themselves or do not interact with other factors to result in noticeable effects. In eight of the 12 measurements reported by Johansson (1982), human reactions were described. These eight measurements are shown in Table 8.3. The concentrations in houses with indoor climate problems are in the range from 0.09 to 13 mg/m3, whereas the concentration in houses where no problems were reported is from 0.02 to 1.7 mg/m3.

PREVALENCE OF EXPOSURES TO VOLATILE ORGANIC COMPOUNDS

243

TABLE 8.2 Total Concentrations of Volatile Organic Compounds (mg/m3) Reported from Nonindustrial Atmospheric Environments Uninhabited Environment

Inhabited

New

Old

New

Old

0.48–18.7a

0.24–0.52b

12.9c

0.02–1.7d 0.25e,f,g

Offices

–h

–h

–h

1.05i 0.09–1.51j

Schools

–o

0.01k,f 0.05o,f,g

0.86l,f,g

0.4–1.6m,n 0.13–0.18k,f 0.14o,f,g 0.29–0.50l,p,f,g 0.22–0.31f,g

Dwellings

a

Mølhave et al. (1979). Mølhave and Andersen (1980). c Frederiksson (1979). d Mølhave and Møller (1979). e Berglund et al. (1981). f Sum of selected compounds. g Johansson (1982). h No information. i Mølhave et al. (1982). j Mølhave et al. (1982). k Wang (1975). l Berglund et al. (1982). m Estimated value. n Miksch et al. (1982). o Johansson et al. (1978). p Johansson et al. (1979). b

The reported concentrations are improperly documented and may represent biased samplings. Further, the number of houses and the measurements are unsystematic and insufficient for a final conclusion. They do, however, indicate that the concentration of VOCs is generally higher in problem houses than in houses without problems and complaints seem to be present when the concentrations exceed 1.7 mg/m3. In summary, 50–300 volatile organic compounds are normally found in air samples from most nonindustrial environments. Each compound seldom exceeds 50 mg/m3, which is 100– 1000 times lower than relevant occupational threshold limit values (TLVs) (ACGIH, 2006). An upper extreme average total concentration of all VOCs in normally occupied homes seems to be 20 mg/m3. The total concentration of all VOCs is, however, normally well below 1 mg/m3, which is only 0.2% of the TLV for toluene. Toluene is one of the most frequently found compounds and one found in relatively high concentrations. 8.2.2

The Concept of Total Volatile Organic Compounds

At present there is no standardized way to summarize the combined effects of the many different compounds in the atmosphere. Addition of the masses of the various molecules has

244

VOLATILE ORGANIC COMPOUNDS AND SICK BUILDING SYNDROME

TABLE 8.3 Total Concentrations of Volatile Organic Compounds (mg/m3) Measured in Occupied Nonindustrial Indoor Environments Environment Dwellings

No Complaints or Information a

0.02–1.7

Offices

–

Schools

0.86g,d,e 0.22–0.31g,d,e 0.14j,d,e 0.13–0.18k,d 0.36 (n ¼ 47)

Weighted average

Complaints 1.05b 0.25c,d,e 12.9f 0.09–1.51b 0.4–1.6 0.29–0.50h,i,d,e 1.31 (n ¼ 24)

a

Mølhave and Møller (1979). Mølhave et al. (1982). c Berglund et al. (1981). d Sum of selected compounds. e Johansson (1982). f Frederiksson (1979). g Berglund et al. (1982). h Johansson et al. (1978). i Johansson et al. (1979). j Johansson et al. (1978). k Wang (1975). b

been suggested (Mølhave, 1986; Mølhave and Nielsen, 1992) in the form of a total volatile organic compound indicator (TVOC). This measure is easily obtained through chemical analysis. From a biological point of view, molar concentrations (number of molecules per cubic meter, in ppm of ppb) may be more relevant. Mathematical functions have also been suggested as indicators based on combinations of other variables such as type of radicals, vapor pressure, or polarity of the compounds. For practical and analytical reasons, identification and quantification of each of the hundreds of compounds in normal indoor air is impossible in most cases. The TVOC, therefore, is often measured by a flame ionization detector (FID) calibrated against toluene (or any other normally occurring organic compound). The use of FID and photo ionization detectors (PID) is discussed by Gammage (1986). Mølhave and Nielsen (1992) explain the necessary calibration procedures. The EU-ECA discourages the use of simple integrating instruments (EU-ECA, 1997). The simplifications used to develop the TVOC concept described here are based on some, but limited experimental evidence (Mølhave and Nielsen, 1992). However, it must be emphasized that the TVOC concept as a health risk indicator has not yet been thoroughly tested in practice and, therefore, is still a postulate. A standardized method for measuring TVOC has been proposed by an EU working group (EU-ECA, 1997). A Nordic consensus group (Andersson et al., 1997) discussed the use of this TVOC and concluded that the literature is inconclusive with respect to the use of TVOC for risk estimates and that insufficient data are available to establish threshold limit values/ guidelines based on TVOC.

HEALTH AND VOLATILE ORGANIC COMPOUNDS

245

Aworking group at the ASTM conference “Healthy Buildings 97” also discussed the use of TVOC and concluded that despite its obvious limitations and imperfections TVOC will be used in the future as an aid in limiting the concentrations of pollutants indoors. TVOC should, however, be used in the recommended standardized way and as a screening tool only, and should be reported together with a list of all identified VOC compounds. It should not be used as the only tool in making definitive conclusions about indoor air quality.

8.3 HEALTH AND VOLATILE ORGANIC COMPOUNDS 8.3.1

The Sick Building Syndrome

An international working group under WHO came to the conclusion in 1982 that many of the indoor climate problems dealt with in the literature seem to describe buildings with the same types of problems (WHO, 1982). These buildings are characterized by the same set of frequently appearing complaints and symptoms. The group suggested that the name “the Sick Building Syndrome” (SBS) be used for this set of symptoms. It appears as part of Table 8.4. In the literature, these and similar symptoms have been used to define other syndromes, which in many cases appear to be synonyms for the Sick Building Syndrome. These synonyms include the building disease, the building illness syndrome, building-related illness, or the tight office building syndrome. The WHO definition of the SBS was the first attempt to combine these syndromes into one general definition. Table 8.5 summarizes the definition of the SBS (WHO, 1982, 1984; Mølhave, 1986, 1990). The WHO group stated that more than 30% of all new buildings seem to be affected by these indoor climate problems which further, seem to have no evident cause (WHO, 1982). In these buildings, no excessive air pollution (e.g., formaldehyde), or defects in the technical installations, or in the construction were evident. The symptoms included in the syndrome may be observed in any group of persons, but “sick buildings” are characterized by a large fraction of the occupants having the TABLE 8.4 Five Categories of Symptoms Exemplified by Some Complaints Reported by Occupants Supposed to Suffer from the Sick Building Syndrome Sensory irritation in eyes, nose and throat Pain, sensation of dryness, smarting feeling, stinging, irritation, hoarseness, voice problems Neurological or general health symptoms Headache, sluggishness, mental fatigue, reduced memory, reduced capability to concentrate, dizziness, intoxication, nausea and vomiting, tiredness Skin irritation Pain, reddening, smarting or itching sensations, dry skin Nonspecific hypersensitivity reactions Running nose and eyes, asthma-like symptoms among nonasthmatics, sounds from the respiratory system Odor and taste symptoms Changed sensitivity of olfactory or gustatory sense, unpleasant olfactory or gustatory perceptions Source: WHO (1982).

246

VOLATILE ORGANIC COMPOUNDS AND SICK BUILDING SYNDROME

TABLE 8.5

Summary of the Sick Building Syndrome

1. The five categories of symptoms shown in Table 27.4 cover the major complaints in the building 2. Irritation of mucous membranes in eye, nose, and throat is among the most frequent symptoms 3. Other symptoms, e.g., from lower airways or from internal organs, should be infrequent 4. A large majority of occupants report symptoms 5. The symptoms appear especially frequent through occupancy of one building or in part of it 6. No evident causality can be identified in relation either to exposures or to occupant sensitivity Sources: WHO (1982, 1984); Mølhave (1986, 1990).

symptoms. The syndrome, therefore, seems to be a normal reaction of the normal population to an unfavorable indoor climate. The syndrome does not seem to be restricted to a minority reacting because of an unusually high sensitivity. The WHO group suggested the possibility that the SBS symptoms have a common causality and mechanism (WHO, 1982). No investigation of the content of this postulated SBS has yet been reported in which a well-defined spectrum of symptoms was used. In general, the descriptions of the symptoms in the literature are anecdotal and unsystematic. Therefore, the existence of SBS is still a postulate although individual symptoms have been related to indoor air quality. 8.3.2

Multiple Chemical Sensitivity

As stated by Ashford et al. (1994) most physicians generally acknowledge the existence of a small fraction of their patients with unexplainable symptoms and signs of unexplainable dysfunctions that may or may not be related to their environment. Because of the diversity of the health effects reported by such patients, most physicians abstain from classifying these as manifestations of a single disease. However, exposures to low levels of chemicals in industrial workplaces, in indoor environments, through consumer products, or pharmaceuticals have been suggested as causative agents for some of these health effects, and have given rise to a public health concern about what often is called multiple chemical sensitivity (MCS). The history of MCS has been summarized by Shorter (1997). The term MCS and a multitude of synonyms were first coined in the United States, but appears now also in Europe, (NAS, 1992; Ashford et al., 1994; Shorter, 1997). Descriptions of the observations and the nature of this controversial syndrome differ at the two sides of the Atlantic Ocean. No generally agreed definition exists and no firm scientific basis has yet been established for the syndrome. MCS could be called a phenomenon, rather than an illness (Levy, 1997). Most authors seem to refer to an intolerance or hyper-responsiveness to exposures or low levels of chemicals at home, during work or in the course of their life activities. Sensitization and subsequent responses are supposedly not associated with exposures to one specific chemical, but to the mere presence of chemicals as such. When exposed, the affected persons report a variety of symptoms in several organs or body systems in addition to allergy and asthma, and the “diagnosis” is often made by the patient, usually alone or with the help of clinical ecologists. As MCS is seen to affect few individual occupants with unusual high sensitivity, MCS must be different from SBS, which affects large fractions of building occupants.

PREVALENCE OF THE SICK BUILDING SYNDROME

247

No single widely accepted test of physiological function has been shown to correlate with MCS symptoms presented by patients. Medical diagnosis and treatments still need to be considered experimental.

8.4 PREVALENCE OF THE SICK BUILDING SYNDROME 8.4.1

Field Investigations

The SBS-symptoms referred to by different authors are nonspecific and undefined in most publications. In addition, exposures are poorly measured. Therefore, a scientifically performed comparison of the prevalence of symptoms found in different reports of field investigations is generally not possible. To review the available publications on investigations of SBS symptoms, these symptoms were instead classified according to the five categories of symptoms mentioned in Table 8.5 (Mølhave, 1991). Indoor climate reactions classified as SBS according to Tables 8.4 and 8.5 were found in 11 of 13 investigations identified (Mølhave, 1991). Most of these investigations dealt with comparison of a building with manifest indoor climate reactions among its occupants to a control building without such problems. Two publications dealt with the frequencies of symptoms among the normal populations of randomly selected buildings. These publications were used as references for comparison with the frequencies of symptoms among occupants supposed to suffer from SBS. Further reference was made to a major questionnaire investigation among Danes. Table 8.6 summarizes the 13 investigations and shows the range of peak frequencies of symptoms reported within each of the five categories of symptoms shown in Table 8.5. This table is explained in detail elsewhere (Mølhave, 1991). No definitive conclusion regarding the SBS and its symptoms is possible from the unsystematic investigations summarized in Table 8.6. Nevertheless, some tendencies are seen within the sparse material. The main conclusion is that if a SBS exists, it probably includes sensory irritation and headache. In six of the supposed sick buildings, information was given about both irritation and headache, TABLE 8.6 Ranges of Percentage of Complainers in Buildings with Occupants Apparently Suffering from the Sick Building Syndromea

Sensory irritation Neurological symptoms Skin irritation Unspecific reaction Odor and taste

Sick Buildings

Control Buildings

Random Buildings

Number of Significant Differencesb

Number of Investigations

35–90 31–100 5–38 4–41 (0)c

0–36 0–45 2–22 4–24 (25)c

8–56 20–56 2–25 2–21 –d

5 3 1 1 0

11 9 5 3 1

Source: Mølhave (1991). a

The frequencies are compared to frequencies in control buildings without complaints and in randomly selected buildings. The symptoms have been grouped according to the five categories of symptoms described in Tables 27.4 and 27.5. b Between sick buildings and control buildings. c Estimated values. d No information.

248

VOLATILE ORGANIC COMPOUNDS AND SICK BUILDING SYNDROME

and in all six cases, the symptom frequencies were highest in the supposed sick buildings. Further, these six investigations included five of the six cases where a significant difference was found between buildings with and without indoor climate problems. This supports the hypothesis that these two symptoms are part of an SBS. It is not possible from the available literature to evaluate if other symptoms are relevant for the SBS, although odor and taste symptoms appear in so few publications that their relevance for the SBS is unlikely. These two symptoms, however, may be mixed up with other complaints such as bad air quality or stuffiness. 8.4.2

Controlled Experiments

Occupational threshold limit values are normally based on evidence from exposures to concentrations much higher than those found in the indoor environment. Furthermore, such experiments often focus on health effects much more severe than the relatively harmless comfort-reducing mucous membrane irritation. Only a few relevant controlled experiments are, therefore, available for extrapolation to the low concentration range. The question is still open whether concentrations of VOCs in the nonindustrial indoor air are sufficiently below the threshold for comfort-reducing irritation and other symptoms included in the sick building syndrome. To test if low exposure levels of VOCs may cause discomfort, four controlled exposure experiments were established in climate chambers in Denmark (Mølhave et al., 1986, 1991; Kjœrgaard et al., 1989, 1991). A more recent experiment in the United States replicated the first of these Danish experiments (Otto et al., 1991). The exposure in four of these experiments consisted of a mixture of the same 22 compounds in the same relative concentrations. Only the total concentration of VOC was changed. These experiments have been summarized (Mølhave, 1991), and Table 8.7 brings together the conclusions of the five exposure experiments and the review of 13 field investigations mentioned previously. The table shows how many of the six data sets indicated positive or negative findings at different VOC levels in relation to five main types of effects.

TABLE 8.7 Summary of Effects on Human Health and Well-Being Caused by Exposure to VOCs in the Indoor Climatea Cofactors VOC Exposure (mg/m3) Effect

0–2.9

3–25

>25

Exposure Duration and Carryover

Subject Sensibility

Sensory irritation Olfaction Toxic irritation (tissue changes) Weak neurological Lung and lower airway Allergy, etc. Systemic and organ

1/1/4 2/0/4 1/1/4 1/1/4 0/0/6 0/0/6 0/0/6

5/0/1 5/0/1 5/0/1 5/0/1 0/0/6 0/0/6 0/0/6

2/0/4 2/0/4 2/0/4 2/0/4 1/0/5 0/0/6 0/0/6

1/2/3 2/1/3 1/1/4 1/0/5 0/0/6 0/0/6 0/0/6

1/1/4 1/1/4 2/0/4 0/1/5 1/0/5 0/0/6 0/0/6

Source: Mølhave (1991). a Six investigations (five exposure experiments and a review of 13 field investigations) are summarized, and the table shows the number of these studies in which either a positive effect, or tendencies to effect over the number showing negative effects, and the number in which the effect was not investigated (positive/negative/not investigated).

DOSE–RESPONSE RELATIONSHIPS FOR HEALTH EFFECTS

249

The five exposure experiments provide no consistent information about lung function, or allergic, or systemic effects at any exposure levels, and, except for olfaction, no effects at exposure levels below 3 mg/m3. At present the response of the olfactory sense seems to be the most sensitive indicator of VOC exposure. Investigations of sensory irritation, olfaction, skin irritation, and weak neurological effects caused by exposure levels of 3 mg/m3 and higher have indicated effects in all experiments dealing with such exposure levels. Further, in Table 8.7, the possibility of cofactor interaction is indicated for the two groups of factors: (1) effect of exposure duration, adaptation and carryover effects from one exposure episode to the next; and (2) personal factors, for example, relating to a subject’s sensitivity. No definitive conclusion can be made with respect to the influence of cofactors. In those few investigations dealing with such factors, both positive and negative indications were found. The information, however, indicates that future research with more sensitive experimental designs and analytical methods may show such effects.

8.5 DOSE–RESPONSE RELATIONSHIPS FOR HEALTH EFFECTS CAUSED BY LOW-LEVEL VOC EXPOSURE Little is known about the effects of low-level VOC exposures characteristic of nonindustrial environments. Evidence from experiments and field investigations indicate (Mølhave, 1990) that the most frequent effects of VOC exposure at low level fall into three classes: (1) perception of the environmental exposure caused by acute stimulations of sense; (2) perception or observation of acute or subacute inflammatory-like reactions, mostly in the exposed tissues; and finally (3) a number of effects that may be classified as a group of subacute environmental stress reactions caused by the perceptions (Mølhave, 1990). The primary processes are stimulation of sensory nerve endings and initiation of weak inflammatory tissue reactions. The secondary acute effects are perceptions of tissue reactions, reflexes initiated by the primary perception of exposures, or changed sensitivity of the senses from the tissue changes. Subacute effects may also occur. They are environmental stress reactions or more severe skin reaction. The three types of effects that are expected to follow from low-level exposure to VOC according to this model are summarized in Table 8.8. The intensity of each of the symptoms may be modified by additional factors, such as age, smoking habits, gender, and so on. Further, the number of symptoms observed and their intensity may have a feedback effect on people’s behavior, thereby causing them, for example, to modify their environment, or to focus on certain symptoms and thus suppress others. As a consequence, each subject may react differently to the mixed exposure and exhibit only a few of the spectrum of symptoms observed in the exposed population. The effects associated with VOC exposures in this model are nonspecific, and may be caused by other environmental exposures than chemicals. For example, physical exposures such as temperature or biological inert dust may cause a similar spectrum of symptoms. A discussion of causality between VOCs and the types of symptoms in Table 8.8 must consider not only the VOC exposure levels, but also the levels of other contributing exposures. As a consequence, under field conditions, three exposure ranges of VOCs are of interest. They are defined by the relative contribution of VOCs to the effects or symptoms observed.

250

VOLATILE ORGANIC COMPOUNDS AND SICK BUILDING SYNDROME

TABLE 8.8

Three Classes of Human Responses to VOCs in Normal Indoor Aira

Acutely perceived deterioration of the quality of the environment Primary effects Recognition of the environmental exposures (odors) Stinging, itching, or tingling feeling from tissues (irritation) Reduced air quality (need more ventilation) Secondary effects Related reflexes in eyes, upper or lower airways Changed mucosal secretion (changed tearfilm stability; changed cell counts in eye liquids) Difficulties in breathing Activities to change the environment Acute or subacute reactions in skin or mucous membranes similar to beginning inflammatory reactions Primary effects Dilatation of peripheral vessels Stinging, itching, or tingling feeling (irritation) Secondary effects Pain Changed skin temperature (face and body, subjective temperature) Subacute and weak stress-like reactions (“environmental stress”) Primary effects Discomfort, complaints (headache, drowsiness) Secondary effects Complications in other body functions and psychological effects such as mood changes and absenteeism Changes composition of eye and nose liquids Changed odor threshold Changed performance a Primary reactions are observed at acute low-level exposures. Secondary effects are observed after prolonged or more intense exposures. The table further shows in parentheses examples of effects, which have been found in controlled experiments with humans.

Below the lowest threshold (the no-effect level) no effect is expected to follow from exposure to VOCs despite any simultaneous exposure to other exposure factors. Above the upper threshold (the effect level), an effect of VOCs is always seen, even when all other factors are controlled and acceptable. Between the two thresholds, a correlation may or may not occur between exposure and effects of VOCs, depending on interactions with other exposure factors and the potency of the compounds in the mixture of VOCs. In this exposure range complaints do not necessarily disappear if one exposure factor (e.g., the VOCs) is identified and removed from the environment. Measurable group responses were found in the controlled exposure experiments summarized above. They follow a gradient from sensory effects: odor at 3 mg/m3 to indications of subacute stress reactions at 25 mg/m3. Regular toxic health effects, such as inflammatory tissue changes or neurotoxic effects such as intoxication in the normal healthy and unsensitized population, may be expected at higher exposures (>25 mg/m3). This exposure range, however, is outside the scope of this review. From the field investigations, it appears that although the symptoms observed are not systematically described, they are more frequent among those exposed than among the

GUIDELINES FOR VOLATILE ORGANIC COMPOUNDS

251

nonexposed. It was found that complaints seem to be present when the concentrations exceed 1.7 mg/m3. Below 1.7 mg/m3 complaints may occur if other types of simultaneous exposures are present (Mølhave, 1986). The concentrations reported from field investigations are improperly documented, and they may be biased. The exposure range of 0.19–0.66 mg/m3 was estimated in the Danish Town Hall Study for the lower, no-effect threshold (Zweers et al., 1990). This range corresponds to the range of lower limit of concentrations in buildings with complaints (Mølhave, 1986) and is at present the best first estimate of the lower exposure limit for no effects of VOC, but should, of this level of documentation, only be used for screening purposes. The laboratory experiments indicate that the main effects can be experimentally reproduced and acutely follow the exposure. No field investigations have been reported of tests of the effects of elimination or modification of VOC exposure. Postexposure measurements during the controlled experiments indicate, however, that the effects are reversible and disappear shortly after exposure (Mølhave et al., 1991). Exposures in most field investigations are multifactorial, as factors other than VOC exposure may exceed their no-effects levels and most of the effects reported in field investigations may have more than one cause. It is, therefore, not surprising that effect of VOC exposures in field investigations seem to occur at lower exposure levels than in controlled experiments, where most other factors are supposed to be below their noeffect levels. In the experiments, the exposure times were less than 3 h, which, from field experience, seem too short to cause severe subacute effects at low exposure levels. More research is needed to test if subacute effects may occur after prolonged exposures. In conclusion, there is no evidence to contradict the proposed causal link between low-level exposure to VOC and the effects shown in Table 8.8. On the contrary, evidence from both field investigations and controlled exposure experiments supports causality. The field investigations and controlled experiments as concluded by the Nordic consensus group (Andersson et al., 1997) are, however, still too few to allow a final conclusion.

8.6 GUIDELINES FOR VOLATILE ORGANIC COMPOUNDS IN NONINDUSTRIAL INDOOR ENVIRONMENTS-PRINCIPLES FOR ESTABLISHMENT OF GUIDELINES Two different concepts seem to be used by different authors in the evaluation of indoor air. The first refers to the quantitative evaluation of the risks of adverse irreversible health effects (e.g., asphyxiation by CO, or lung cancer associated with radon) or the risks of reversible or irreversible changes in the body’s physiological functions (e.g., nervous system effects). This is the traditional occupational or environmental evaluation of health risks, which is done according to standard toxicological principles including risk estimates and cost-benefit analysis. This is the background for air quality guidelines and ambient air standards for air pollutants and for TLVs and occupational exposure standards for light levels and sound levels, and is based on lists of high-risk compounds. The second concept refers to qualitative evaluations of the atmospheric environment and is, in many respects, a new concept for regulation that is often neglected in discussions of indoor air quality: Air deodorants, painting, wallpapers, or music can be liked by some persons and disliked by others, and generally accepted principles for regulation of the quality

252

VOLATILE ORGANIC COMPOUNDS AND SICK BUILDING SYNDROME

of the indoor environment with respect to odors, sounds, and colors may be impossible to establish, if indeed such regulation is wanted at all. Some general conclusions may, however, be drawn about the principles to apply in setting future regulations of VOCs in the indoor climate. Such principles can be extracted from existing regulations, for example, in building codes, for acoustics and lightning that contain additional qualitative concerns besides those used for the setting of TLV levels. For these guidelines the first basic principle implicit is that the building must support a specified range of human activities, habits, and preferences. Complaints and decreased performance will automatically follow if the occupants try to do activities outside this range, for example, reading in too dark a room with disturbing intermittent noise peaks or working with a video screen with many light reflections. This range of activity may be different for homes and offices, as the activity patterns in homes include recreation, rest, and sleep and other activities normally not found in offices. Further, the occupants of homes may be more sensitive than the working population, as they include the sick, young, and old fractions of the population. The second basic principle originates from the assumption that humans do not feel well if they do not have the optimal use of their senses to perceive their environment and to monitor the activities they are performing. The ideal indoor environment, therefore, seems to allow the occupants to use their senses to perceive their environment and to monitor the activities they are performing. The occupants should be able to use their senses to pick up wanted environmental signals undisturbed by exposures to unwanted information noise. This means that unwanted environmental information should be damped, such as the sound of a typing machine in an office or the neighbor’s radio in a home. On the contrary, our own conversation or the perception of sounds related to our own activities should be eased. In short, this second principle tells us to optimize the signal-to-noise ratio for our senses by allowing the wanted signals to propagate to the occupants and damping unwanted sensory signals. This principle, if true, explains why occupants have such different optimal environments. The signals that bear information to one person about his own activity and environment create sensory noise for the other person. Therefore, the signals relevant for one person differ from those relevant for his or her neighbor. To summarize the discussion on sensory perception of VOCs as air pollutants, the acceptable exposure range may be defined as follows. In normal rooms the air quality is acceptable if no unacceptable health risk exists, and if all sources for chemical stimulation can be identified by the occupants, and those sources bearing unwanted information can be removed, should the occupants desire to do so. 8.6.1

A Tentative Guideline for VOCs in Nonindustrial Environments

The observations summarized here are based on investigations that have major limitations. Presently, they do not form an acceptable basis for setting official recommendations or guidelines (Andersson et al., 1997), but discussions on the possible principles of such guidelines have been initiated (Mølhave, 1998). The observations do, however, indicate that VOCs may be important for indoor air quality, especially in the form of discomfort from odors, irritative symptoms in eyes, nose, and throat, and headache. The effects may also include effects related to productivity and performance. Such effects have not yet been positively identified.

GUIDELINES FOR VOLATILE ORGANIC COMPOUNDS

253

Most indoor environments in nonindustrial buildings are polluted with 10–100 VOCs as air pollutants, each generally in concentrations ranging from nanograms to few hundred micrograms per cubic meter. The list of compounds often identified indoors may include 1000 compounds. The toxicity of these compounds in the low exposure range is mostly unknown, and their possible interactions cannot be predicted with the existing knowledge. Lacking guidelines for indoor air quality, practitioners have had to use simple approximative tools for making evaluations of air quality. One of these is the TVOC, which is the sum of the individual concentrations of all VOC present. TVOC has been discussed by several international working groups (EU-ECA, 1997; Andersson et al., 1997) as described above. TVOC should only be used as a screening tool to identify potential unacceptable exposures to VOC, and should not be used as the only tool to make definitive conclusions. Tentative screening values for VOCs in nonindustrial environments have been offered (Mølhave, 1991). The tentative conclusion of the available epidemiological studies and the exposure experiments is that if the presence of specific potential irritants can be excluded (which they seldom can), irritation is unlikely to follow from exposure to VOCs below about 0.2 mg/m3. The ambient air levels are generally below that level, which for screening purposes may be used as a first estimate of a lower limit for possible effects of VOCs. At concentrations higher than about 3 mg/m3, complaints seemed to occur in most investigated buildings with occupants having symptoms. In controlled exposure experiments using on mixture of 22 VOCs, odors were significant at 3 mg/m3. At 5 mg/m3 in the same exposure experiments, objective effects were indicated in addition to subjective irritation. Exposures for 50 min to 8 mg/m3 led to significant irritation of mucous membranes in eyes, nose, and throat. Few indications are given in the literature that allows an estimate of the threshold for headache. Concentrations below 3 mg/m3 were found, in field investigation, to produce a significant difference in frequencies of headache between problem buildings and control buildings. On the contrary, significant headache was found in only one of the exposure experiments and then at 25 mg/m3. The reason for the lower threshold in field investigations may be either the interaction of other exposures, or the effect of longer exposure durations. Therefore, based on the present information, the threshold for TABLE 8.9 Tentative Dose–Response Relationship for Discomfort Resulting from Exposure to Solvent-like Volatile Organic Total Concentration (mg/m3) <0.20 0.20–3.00 3.0–25

>25

Irritation and Discomfort No irritation or discomfort Irritation and discomfort possible if other exposures interact Exposure effect and probable headache possible if other exposures interact Additional neurotoxic effects other than headache may occur

Exposure Range The comfort range The multifactorial exposure range The discomfort range

The toxic range

The table only refers to VOCs with a normal range of biological activity and should only be used for screening purposes. Source: Mølhave (1991).

254

VOLATILE ORGANIC COMPOUNDS AND SICK BUILDING SYNDROME

headache and other weak neurotoxic effects caused by exposure of less than a few hours duration are expected to be between 3 and 25 mg/m3. These conclusions refer to the most prevalent effects of VOCs on normal subjects. Risk groups may exist that will respond more strongly than the normal population. Furthermore, future investigations dealing with larger groups of persons may reveal special effects such as allergy or carcinogenicity associated with low-level exposures to VOCs. These special effects, however, have not been demonstrated for the type and concentration of VOCs found in indoor air. A tentative dose–response relationship for discomfort resulting from exposure to VOCs is shown in Table 8.9. The table, which should only be used for screening purposes, indicates a possible no-effect level at about 0.2 mg/m3. A multifactorial exposure range may exist from 0.2 to 3 mg/m3 in which odor irritation and discomfort may appear as consequences of VOC exposure depending on the types of compounds present and if other exposures contribute to the etiology. Above about 3 mg/m3, effects of VOC exposure are likely and exposures above 25 mg/m3 may be expected to cause toxic effects.

REFERENCES American Conference of Governmental Industrial Hygienists (ACGIH) (2006) TLVs: Threshold Limit Values and Biological Exposure Indices for 2006. Cincinnatti, USA:ACGIH. Andersson K, Bakke JV, Bj€orseth O, Bornhehag CG, Clausen G, Hongslo JK, Kjellman M, Kjœrgaard SS, Levy F, Mølhave L, Skerfving S, Sundell J (1997) TVOC and health in non-industrial indoor environments. Reports from a Nordic Scientific consensus meeting at La˚ngholmen in Stockholm, 1996. Indoor Air 7:78–91. Andrew LS, Snyder R (1980)Toxic effects of solvents and vapors. In:Casarett LJ, Doull J,editors. Casarett and Doull’s Toxicology. The Basic Science of Poisons. New York, USA: Macmillan. Ashford N, Hewinzow B, L€utjen K, Maruoli C, Mølhave L, M€onch B, Papadopoulos S, Rest K, Rosdahl D, Siskos P, Volonakis E (1994) Chemical sensitivity in selected European countries: an explorative study. Commission of the European Union, DG-V, agreement SOC 93 2027 48 05E01 (93CVVF1-613-0). Report from Ergonomia Ltd, Athens, Greece. Berglund B, Johansson I, Lindvall T (1981) Underlag f€or Ventilationsnormer. ETAPP II (Ventilation Requirements, in Swedish). Stockholm, Sweden: The National Institute of Environmental Medicine. Berglund B, Johansson I, Lindvall T (1982) A longitudinal study of air contaminants in a newly built preschool. Environ. Int. 8:111–116. Berglund B, Berglund U, Lindvall T (1986) Assessment of discomfort and irritation from the indoor air. In:Proceedings of IAQ-86. Managing Indoor Air for Health and Energy Conservation, American Society of Heating Refrigerating and Air Conditioning Engineers, Atlanta, USA, pp.138–149. Berglund B, Brunekreef B, Kn€oppel H, Lindvall T, Maroni M, Mølhave L (1992) Effects of indoor air pollution on human health. Indoor Air 22–25. De Bortoli A, Kn€oppel H, Pecchio E, Peil A (1986) Concentrations of selected organic pollutants in the indoor and outdoor air in northern Italy. Environ. Int. 12:343–356. (EU-ECA)Berglund B, Clausen G, De Ceaurriz J, Kettrup A, Lindvall T, Maroni M, Mølhave L, Pickering AC, Risse U, Rothweiler H, Seifert B, Younes M (1997) Total organic compounds (TVOC) in indoor air quality investigations. Report 19 of the European Collaborative Action “Indoor Air Quality & its Impact on Man”. EU-report EUR 17675 EN, Joint Research Center, Ispra, Italy.

REFERENCES

255

Frederiksson K (1979) Gifter i Bostadsluft (Indoor air pollution, in Swedish). H€alova˚rdskontakt 3:14–19. Gammage RB (1986) Volatile organic compounds. AIHA indoor environmental quality. reference manual (DE-AC05-840R-21400). Health and Safety Research Division, Oak Ridge National Laboratory, Oak ridge, TN, USA. Johannsson I (1978) Determination of organic compounds in indoor air with potential reference to air quality. Atmos. Environ. 12:1371–1377. Johansson I (1982) Kemiska Luftf€ororeningar Inomhus. En litteraturs Sammenst€allning (Chemical Indoor Air Pollution, a Swedish Review). Stockholm. Sweden: The National Institute of Environmental Medicine. Johansson I, Petterson S, Rehn T (1978) Gaschromatographic analysis of room air in recently built preschools (in Swedish). VVS J. 49:51–55. Johansson I, Petterson S, Rehn T (1979) Indoor air pollutants (in Swedish). VVS J. 50:6–7. Kjœrgaard SK, Mølhave L, Pedersen OF (1989) Human reactions to indoor air pollutants: n-decane. Environ. Int. 15:473–482. Kjœrgaard SK, Mølhave L, Pedersen OF (1991) Human reactions to a mixture of indoor air volatile organic compounds. Atmos. Environ. 24A (8):417–1426. Krause C, Mailahn W, Nagel R, Schulz C, Seifert B, Ulrich D (1987) Occurrence of volatile organic compounds in the air of 500 homes in the Federal Republic of Germany. In: Seifert B, Esdron H, Fischer M, F€uden H, Wegner J, editors. Indoor Air 87’. Berlin (FRG), Germany: Institute of Water, Soil and Air Hygiene. Lebret E, Van de Wiel HJ, Bos H, Noij D, Boleij JSM (1986) Volatile organic compounds in Dutch homes. Environ. Int. 12:323–332. Levy F (1997) Clinical features of multiple chemical sensitivity. Scand. J. Work. Environ. Health 23 (Suppl. 3):69–73. Miksch RR, Hollowell CD, Schmidt HE (1982) Trace Organic Chemical Contaminants in Office Spaces. Environ. Int. 8:129–137. Mølhave L (1986) Indoor air quality in relation to sensory irritation due to volatile organic compounds. ASHRAE Trans. 92 (1):306–316, Publication #2954. Mølhave L (1990) Volatile organic compounds, indoor air quality and health, Vol. 5. In: Walkinshaw D, editor. Indoor Air ’90: Proceedings of the 5th International Conference on Indoor Air Quality and Climate, Toronto. Ottawa, Ontario, Canada: Canada Mortgage and Housing Corporation. pp.15–33. Mølhave L (1991) Human response to volatile organic compounds as air pollutants in normal buildings. J. Exposure Anal. Environ. Epidemiol. 1(1):63–81. Mølhave L (1998) Principles for evaluation of health and comfort hazards caused by indoor air pollution. Indoor Air 8(Suppl. 4):17–25. Mølhave L, Andersen I (1980) Forureningskomponenter i indeluften i “Nulenergihuset” ved DTH (Air pollution in an experimental house, in Danish). Varme 45:121–125. Mølhave L, Møller J (1979) The atmospheric environment in modern Danish dwellings–measurements in 39 flats. In: Fanger PO, Valbjørn O, editors. Indoor Climate. Copenhagen, Denmark: Danish Building Research Institute. Mølhave L, Nielsen GD (1992) The TVOC indicator of human response to low level exposures to volatile organic compounds (VOC). Indoor Air 2:5–77. Mølhave L, Thorsen T (1991) A model for investigations of ventilation systems as sources for volatile organic compounds in indoor climate. Atmos. Environ. 25A(2):241–249. Mølhave L, Møller J, Andersen I. (1979) Luftens indhold af gasser, dampe og støv i nyere boliger (Indoor air pollution in home, in Danish). Ugesk. Lœger 141:956–961.

256

VOLATILE ORGANIC COMPOUNDS AND SICK BUILDING SYNDROME

Mølhave L, Andersen I, Lundqvist GR, Nielsen PA, Nielsen O (1982) Afgasning fra Byggematerialer—forekomst og Hygiejnisk Vurdering (Emission of Air Pollutants from Building Materials. In:Danish with English summary. SBI-report no. 137. Copenhagen, Denmark: The Danish Institute of Building Research. Mølhave L, Bach B, Pedersen OF (1986) Human reactions to low concentrations of volatile organic compounds. Environ. Int. 12:167–175. Mølhave L, Jensen JG, Larsen S (1991) Subjective reactions to volatile organic compounds as air pollutants. Atmos. Environ. 25A(7):1283–1293. (NAS) National Research Council (1992) Multiple Chemical Sensitivities. Addendum to Biological Markers in Immunotoxicology. Washington, DC, USA:National Research Council, National Academy Press. Otto DA, Mølhave L, Rose G, Hudnell HK, House D (1991) Neurobehavioral and sensory irritant effects of controlled exposure to a complex mixture of volatile organic compounds. Neurotoxicol. Teratol. 12:649–652. Shorter E (1997) Multiple chemical sensitivity: pseudo disease in historical perspective. Scand. J. Work Environ. Health 23(3):35–42. Wallace L (1987) The Total Assessment Methodology (TEAM) Study. Summary and Analyses, Vol. 1. Washington, DC, USA: U.S. Environmental Protection Agency. Wang TC (1975) A study of bioeffluents in a college classroom. ASHRAE Trans. 81:32–44. World Health Organization (WHO) (1961) Constitution of the World Health Organization: Basic Documents, 15th edn. Geneva, Switzerland: WHO. World Health Organization (WHO) (1982) Indoor air pollutants, exposure and health effects assessment. Euro Reports and Studies No. 78: Working Group Report. WHO Regional Office for Europe. Copenhagen, Denmark. World Health Organization (WHO) (1984) Indoor air quality research. Euro Reports and Studies No. 103. WHO Regional Office for Europe. Copenhagen, Denmark. World Health Organization (WHO) (1989) Indoor air quality: organic pollutants. Report on a WHO Meeting, Euro Reports and Studies No. 111. WHO Regional Office for Europe. Copenhagen, Denmark. Zweers T, Skov P, Valbjørn O, Mølhave L (1990) The effect of ventilation and air pollution on perceived indoor air quality in five town halls Energy Bldgs. 14:175–181.

9 FORMALDEHYDE AND OTHER ALDEHYDES George D. Leikauf

9.1 BACKGROUND Defined by a reactive, polarized carbonyl group, low-molecular-weight aldehydes are a family of organic compounds useful in a large number of industrial processes. The simplest aldehyde, formaldehyde (HCHO), is one of the top10 organic chemical feedstocks, and one of the top 20 five chemicals produced in the United States. Other widely used aldehydes include acetaldehyde (CH3CHO) and acrolein (CH2¼CHCHO), which differ from formaldehyde in carbon chain length and whether the chain is saturated or unsaturated. 9.1.1

Human Environmental Exposure

Human aldehyde exposures result from exogenous sources and endogenous formation (i.e., biogenesis through metabolism or oxidative stress) (Benedetti et al., 1980, 1984; Nilsson and Tottmar, 1987; Marnett, 1988; Anderson et al., 1997; Uchida et al., 1998; Lovell et al., 2001; Noiri et al., 2002; Shao et al., 2005). Exogenously formed or environmental aldehydes can be generated naturally through tropospheric reactions of terpenes and isoprene released from foliage with hydroxyl radicals. In addition, the major sources of aldehydes in ambient air are generated during incomplete combustion of alcohols, or are released from polymeric substances and solutions. Concern over the health effects of environmental aldehydes continues because of increasing usage of automotive fuels containing alcohols (ethanol and methanol) (Othmer, 1987). Sources other than motor vehicle exhaust include power plants, incinerators, paper mills, and refineries. This chapter reviews both the environmental sources and potential health effects, and is an update of the chapter that appeared in the second edition of this book. Other valuable reviews on aldehyde toxicity include reports by the National Research Council (NRC, 1981), Beauchamp et al. (1985), Feinman (1988), Marnett (1988),

Environmental Toxicants, Third Edition Edited by Morton Lippmann Copyright
2009 John Wiley & Sons, Inc.

257

258

FORMALDEHYDE AND OTHER ALDEHYDES

Council on Scientific Affairs (1989), World Health Organization (WHO, 1989, 1992, 1995), Heck et al. (1990), McLaughlin (1994), International Agency for Research on Cancer (IARC, 1982, 1985, 1995a, 1995b, 2004). 9.1.1.1 Indoor Air Estimated contributions from three indoor sources of formaldehyde are listed in Table 9.1. Considering the average time–activity pattern, poor indoor air quality in the home is the most common source of aldehyde exposure. Aldehyde-generating activities in the home include tobacco smoking, wood burning, and cooking. Other common sources include release from paint, structural materials, furnishings, clothing, cosmetics, and insulation (Pickrell et al., 1983; Feinman, 1988). We typically spend less than 1 h per day (18–42 min) outside (Fig. 9.1), so that outdoor exposures constitute only 3% of our average daily exposure (Chapin, 1974; Samet et al., 1987; Samet and Spengler, 1991). Excluding the other activities, such as time in transit (1.0–1.6 h per day) and occupational exposure for those working outside the home 5.2–6.7 h per day, the remaining and primary exposure for most individuals occurs at home. Indoor exposure therefore equals 55–65% for those working outside the home and 85% for those working inside the home (Songco and Fahey, 1987). In the past, urea formaldehyde foam insulation was a source leading to the highest home exposures, including exposures averaging 120 ppb (Gupta et al., 1982). Concentrations in older homes without foam insulation typically range from 30 to 90 ppb, whereas high-level exposures of up to 4200 ppb have been recorded in mobile homes (Georghiou et al., 1983; IARC, 1985, 1995a, 1995b, 2004). A recent study of home levels determined concentrations averaged about 30 ppb (Clarisse et al., 2003; Casset et al., 2006b). Formaldehyde is also released from polymeric resins such as urea formaldehyde (used in particleboard, plywood, paper and textile treatments, and surface coatings), melamine formaldehyde (used in laminates, surface coatings, wood adhesives, and molding compounds), or phenolic resins (used in plywood adhesives and insulation) (WHO, 1989; Gerberich and Seaman, 1994). Elevation of temperature or humidity can increase release rates of formaldehyde, and thereby lead to higher exposure concentrations. More recently, penetration of outdoor ozone

TABLE 9.1

Indoor Sources of Formaldehyde Exposure

Sources

Concentration

Cigarette smoke (40 ppm in 40-mL puff) Dose per pack for smoke Environmental tobacco smoke

0.38 mg per pack 0.25 ppm

Clothing made with synthetic fibers Men’s polyester cotton blend Women’s dress

2.7 mg/g per day 3.7 mg/g per day

Furnishings Particle boarda Plywood Paneling Draperies Carpet/Upholstery Fabric

0.4–8.1 mg/g per day 1.5–5.3 mg/g per day 0.9–21.0 mg/g per day 0.8–3.0 mg/g per day
0.1 ppm

a

Made with urea–formaldehyde resin.

Source: Pickrell et al. (1983); Feinman (1988).

BACKGROUND

259

FIGURE 9.1 Time spent per day at various locations for adults. Values for working outside home are for men and women and for working in the home are for women. Adapted from Chapin (1974); and Samet et al. (1987).

and reaction with indoor materials has been found to be an additional source of aldehydes (Morrison and Nazaroff, 2002; Destaillats et al., 2002, 2006). By far, the leading indoor source of formaldehyde and several other aldehydes today is mainstream cigarette smoke (the portion inhaled by the smoker), sidestream (the portion emitted from a burning cigarette), and environmental tobacco smoke (the aged combination of sidestream and exhaled mainstream smoke) (Table 9.2). Aldehydes are principally associated with the vapor phase of mainstream smoke, but have also been measured in the particulate phase (Ayer and Yeager, 1976; Godish, 1989; Nazaroff and Singer, 2004).

TABLE 9.2

Aldehydes in Cigarette Smoke Amount Released (mg/pack) Environmental

Aldehydea Formaldehyde Acetaldehyde Acrolein Propionaldehyde a

Mainstream

Sidestream

Tobacco Smokeb

3.4 12.5 1.5 1.3

14.5 84.7 25.2 18.8

1.3 3.2 0.6 0.9

Other aldehydes in cigarette smoke include isobutylaldehyde, methacrolein, butylaldehyde, isovaleraldehyde, crotonaldehyde, and 2-methylvaleraldehyde. b Environmental tobacco smoke ¼ 2-h integrated amount per pack. Source: R. J. Reynolds (1988).

260

FORMALDEHYDE AND OTHER ALDEHYDES

Of the various aldehydes present in mainstream smoke, acetaldehyde is the most abundant compound by weight, followed by formaldehyde, acrolein, and propionaldehyde (Fujioka and Shibamoto, 2006). Importantly, the aldehyde concentrations released in smoke from cigarettes between puffs at a lower temperature (600 C) are greater than those produced during smoking (900 C). Thus, sidestream aldehyde emissions are 5–15 times greater than mainstream levels. Because smokers inhale only about 45% of the smoke from each cigarette, sidestream smoke can add significantly to the indoor aldehyde burden. For example, the contribution of cigarette smoke alone to indoor aldehyde concentrations can be 100 ppb formaldehyde (Gammage and Gupta, 1984; Schaller et al., 1989) or acrolein (Badre et al., 1978; Jermini et al., 1976) in rooms where several individuals are smoking. 9.1.1.2 Occupational Exposures Occupational exposure to low-molecular-weight aldehydes is extensive, primarily because of the usefulness of their reactive carbonyl in chemical synthesis. Of the several aliphatic aldehydes available, formaldehyde has the greatest production and usage. Over 10 billion pounds in the United States, and over 45 billion pounds worldwide, of formaldehyde is produced annually, with about 90% being used as a chemical feedstock or an intermediate in the synthesis of a wide number of chemicals including urea-formaldehyde and phenol-formaldehyde resins, ethylene glycol, fertili-zers, dyes, disinfectants, germicides, hardening agents, and as a preservative in water-based paints, cosmetics, and hair shampoos. Urea- and phenolformaldehyde resins are used as adhesives in the manufacture of particleboard, fiberboard, and plywood, or are used in molding, paper treating and coating, surface coating, textile treating, and insulation foam. Over one million people in the United States are estimated to be occupationally exposed to formaldehyde alone. This includes persons working in medical and health services (approximately one third), funeral homes, textiles, furniture, paper, and agriculture industries (Consensus Workshop on Formaldehyde, 1984; IARC, 1995a). Of these, over 20,000 individuals have routine exposure to concentrations greater than 1000 ppb, with over 500,000 exposed to concentrations of 500–1000 ppb (Occupational Safety and Health Administration, 1985; Noisel et al., 2007). Along with inhalation, occupational exposures to low-molecular-weight-aldehydes can also be topical, as these compounds are used as aqueous solutions (e.g., formalin, which is typically 37% formaldehyde in water and methanol) or as a polymerized solid (e.g., paraformaldehyde). Absorption through the skin is limited (typically <10%) because of the high reactivity and volatility of aldehydes. Additional aldehydes with the greatest industrial production include acetaldehyde, acrolein, crotonaldehyde, chloroacetaldehyde, and furfural. The current recommended occupational threshold limit values (TLV) for these compounds are presented in Table 9.3. Of the compounds listed, acetaldehyde and acrolein have the greatest usage. Acetaldehyde is used primarily as a chemical intermediate, principally for the production of acetic acid, crotonaldehyde, pyridine and pyridine bases, glycols, and esters (IARC, 1985), and as a food additive, being listed as general recognized as safe (GRAS), or in the synthesis of flavors and fragrances. Acetaldehyde has been used commercially in the synthesis of aniline dyes, synthetic rubber, resins, disinfectants, drugs, explosives, lacquers, and photographic chemicals, and annual U.S. production exceeds 700 million pounds. Over 400 million pounds of acrolein are produced annually for use in the production of glycerine, synthetic methionine, glutaraldehyde, acrylic acid, and other chemicals. Acrolein also is applied directly as a biocide/slimicide/herbicide in the control

BACKGROUND

261

TABLE 9.3 Occupational Threshold Limit Values (TLVs) as Recommended by the American Conference of Governmental Industrial Hygienists (1998) Low Molecular Weight Aldehyde Acetaldehyde Acrolein Chloroacetaldehyde Crontonaldehyde Formaldehyde Furfural glutaraldehyde

Time-Weighted Averagea ppm (mg/m3)

Ceilingb ppm (mg/m3) 25.0 (45) 0.1 (0.23) 1.0 (3.2) 0.3 (0.86) 0.3 (0.37)

2.0 (7.9) 0.05 (0.2)

Carcinogenicityc

Skind

A3 A4

Skin

A3 A2 A3 A4

Skin Skin

a

Threshold limit value–time-weighted average is the mean concentration obtained normalized to an 8-h work day and a 40-h work week. At this concentration nearly all worker s may be repeatedly exposed, day after day, without adverse effect. b Threshold limit value–ceiling limit is a concentration that should not be exceeded during any part of a working exposure. When continuous monitoring is not feasible sampling period should not exceed 15 min. c Carcinogenicity are compounds that may cause or contribute to increased risk of cancer in workers and are separated into five categories. A1: Confirmed human carcinogen. A2: Suspected human carcinogen. This agent is carcinogenic in experimental animals at dose levels, by routes of administration, at sites, histological type, or by a mechanism considered relevant to worker exposure. Available epidemiologic studies are conflicting or insufficient to confirm the risk of cancer in exposed humans. A3: Animal carcinogen. This agent is carcinogenic in experimental animals at relatively high dose levels, by route of administration, at site, histological type, or by a mechanism not considered relevant to worker exposure. Available epidemiologic studies do not confirm an increased risk of cancer in exposed humans. Available evidence suggests that the agent is not likely to cause cancer in humans except under uncommon or unlikely routes or levels of exposure. A4: Not classifiable as a human carcinogen. Inadequate data exist on which to classify the agent as a carcinogen in either humans or animals. d Skin are agents with potential significant contribution of the overall exposure by the cutaneous route, including mucous membranes and the eyes, either by direct contact with vapors or, of probable greater significance, by direct skin contact with the substance.

of algae, weeds, and molluscs in re-circulating process water systems, drainage systems, the paper industry, or in oil wells and liquid petrochemical fuels. Acrolein has usage in histology, and to make modified food starch, synthetic glycerine, acrolein polymers, polyurethane, and polyester resins (ATSDR, 2005). 9.1.1.3 Ambient Air and Alternative Fuels Low-molecular-weight aldehydes are lowlevel contaminants of the urban environment. Concentrations in ambient air are typically much lower than those encountered in occupational or indoor settings. Along with the direct release of aldehyde from stationary sources (coal-fired power plants, home wood fires and incinerators), these compounds are either released directly from mobile sources (car, truck, or jet engines) or formed by secondary photochemical reactions from emitted hydrocarbons. Nonetheless, aldehydes are important components of atmospheric chemistry because, as products of oxidation of almost all hydrocarbons, they are precursors of free radicals, ozone, and peroxyacyl nitrates (Grosjean, 1991; Grosjean et al., 1996; Tuazon et al., 1997). As such, aldehydes are an excellent species to evaluate predictions of ozone formation in kinetic models of urban air quality (Carter, 1990; Grosjean et al., 1992; Harley and Cass, 1995; Liu et al., 2006).

262

FORMALDEHYDE AND OTHER ALDEHYDES

In urban areas, aldehyde levels, like those of other hydrocarbons, exhibit diurnal variations that precede ozone peaks. In the past, total aldehyde levels have occasionally reached maximums of 300 ppb, but more typically are less than 40 ppb. Approximately 20–50% of the total aldehyde is in the form of formaldehyde, with typical values ranging from 5 to 30 ppb (with maximal values as high 50–86 ppb in urban air masses contained by atmospheric inversions) (Dasgupta et al., 2005, Bruinen de Bruin et al. 2008). Historical episodic values recorded in Los Angeles for formaldehyde have reached 90–150 ppb; in contrast, ambient values in rustic environments can be
1.0 ppb. Because other aldehydes are not routinely measured in urban air; few data exist on the ambient concentrations of most aldehydes. Acetaldehyde levels are often 2–39 ppb, which is about 75% of formaldehyde levels (range 32–155%) (Grosjean et al., 1996). These levels are sufficient to make acetaldehyde the major aldehyde in the removal of hydroxyl radicals from the atmosphere. Other aldehydes that are important in this process include formaldehyde and nonanal. Estimates for acrolein are between 5% and 10% of the total aldehyde burden (or 5–30 ppb for peaks), and those for other aliphatic aldehydes between 35% and 40% of total aldehyde concentrations. Of these, higher molecular weight carbonyls (nine species: C8–C14) typically account for 10–15% of the total aldehydes (Grosjean et al., 1996). Recently atmospheric hemispheric background, biogenic- dominated regions, and urban environments atmospheric acrolein concentrations were measured to be 128, 204, and 667 ppb acrolein (0.056, 0.089, and 0.29 mg/m3), respectively, which are all above the EPA Reference Concentration of 0.02 mg/m3 (Grosjean and Grosjean, 2002; Destaillats et al., 2002; Seaman et al., 2006). Of the compounds found on the current list of 43 hazardous air pollutants (HAPs), formaldehyde is often above the cancer benchmark concentrations (with a risk level of one-in-one-million) and acrolein is often at or above the noncancer hazard index (Morello-Frosch et al., 2000; Leikauf, 2002; Tam and Neumann, 2004; Dasgupta et al., 2005; McCarthy et al., 2006; Seaman et al., 2006). Of these compound, acrolein is often associated with noncancer risks with one study estimating median additional number of adverse sC(L) outcomes across the United States was approximately 2.5 cases per 1,000 people (Woodruff et al. 2007). Between 55% and 75% of ambient aldehydes are from mobile sources, and proximity to such sources may produce greater localized exposures. For example, exposures near garages, tunnels, or in city street canyons may be much greater than average, since exhaust from passenger cars equipped with either spark-ignition or diesel engines can produce aldehydes (Grosjean and Grosjean, 2002). Estimated concentrations in emissions from warm gasolinefueled automobiles (running at 40-mph cruising speeds) are about 1000, 200, and 100 ppb for formaldehyde, acrolein, and acetaldehyde, respectively (Swarin and Lipari, 1983). In contrast, starts of cold engines yield levels in which formaldehyde can exceed 5000 ppb. Diesel engines also produce significant amounts of formaldehyde, acrolein, and acetaldehyde, with levels produced by warm engines equaling about 4000, 500, and 1000 ppb, respectively. Aldehydes are enriched in motor vehicle emissions when fuels containing oxygenated additives (especially alcohols) are combusted (Grosjean, 1990; Carter, 1990; Shah and Singh, 1998; Magnusson et al., 2002; Dasgupta et al., 2005). The formulation of these alternative fuels may vary and contain either methanol (from methane) or ethanol (from agricultural sources); these are likely to be mixtures of alcohol and gasoline (e.g., M-85, 85% methanol–15% gasoline mixture). The addition of methanol to gasoline may increase formaldehyde emissions, whereas ethanol is likely to increase acetaldehyde concentrations. Acetaldehyde concentrations, for example, for internal combustion engines using ethanol rather than gasoline can increase from <100 ppb to over 190,000 ppb

BACKGROUND

263

following cold starts and from <100 ppb to 8400 ppb acetaldehyde from warm engines (Marnett, 1988). Formaldehyde concentration can also increase from 480 to 3600 ppb. 9.1.2

Cellular Exposure and Dosimetry

9.1.2.1 Nasal Deposition and Penetration Regional deposition of an inhaled gas is controlled by water solubility, concentration, and physiological and pathological features of the respiratory tract. Measurements of formaldehyde deposition in rats (Dallas et al., 1985; Patterson, 1986) and in dogs (Egle, 1972) found essentially 100% total respiratory tract deposition. One interesting observation made in these studies was the secretion of low levels from control animals/individuals, presumably from biogenic formation. Aldehydes are highly soluble in water compared to other common air pollutants (Table 9.4) and the principal site of deposition of these compounds is the upper respiratory tract (Aharonson et al., 1974). When inhaled in low concentrations, almost all (>98%) of formaldehyde is deposited in the moist layers covering the nasal mucosa. This deposition occurs during the inspiration, with little (<2%) or no formaldehyde being exhaled (Egle, 1972; Heck et al., 1983; Dallas et al., 1985). In rats, Heck et al. (1983) reported that radiolabeled formaldehyde deposition was proportional to the concentration and length of exposure. Deposition rate decreased over a 6-h exposure period resulting, in part, from a decrease in ventilatory rate during exposure (Kimbell et al., 2001). TABLE 9.4 Agents

Physical and Chemical Properties of Aldehyde and Other Environmental

Compound

Structure

Mol. Wt.

Aqueous Sol. (g/L)

Organic Sol. (Most–Least)a

I. Aldehydes a. Saturated aliphaticaldehydes Formaldehyde HCHO Acetaldehyde CH3CHO Propionaldehyde CH3CH2CHO Glutaraldehyde OHC(CH2)3CHO

30.03 44.05 58.08 100.12

560.0 200.0 160.0 al, bz

Eth, ace, bz, al, chl Eth, al, bz Eth, al

b. Unsaturated aliphatic aldehydes Acrolein CH2¼CHCHO Crotonaldehyde CH3CH¼CHCHO

56.06 70.09

210.0 181.0

Eth, ace, al Bz, eth, ace, al

c. Aromatic or cyclic aldehydes Benzaldehyde C6H5CHO Cinnamaldehyde C6H5CH¼CHCHO

106.11 132.16

3.0 10.5

Eth, al, ace, bz Eth, al, chl

II. Other environmental Sulfur dioxide Carbon dioxide Carbon monoxide Oxygen Ozone

agents and commoninorganic molecules SO2 64.07 106.4 CO2 44.01 1.74 CO 28.01 0.03 O2 O3

32.00 48.00

0.04 0.02

a

Chl, ether meth, al – Eth, meth, eth ace, ace acid – Meth cl

Abbreviations. Eth: ethanol; ace: acetone; bz: benzene; al: alcohol; chl: chloroform; meth: methanol; eth ace: ethyl acetate; ace acid: acetic acid; meth cl: methylene chloride. Source: Comey and Hahn (1921); Seidell (1940, 1941); Windholz (1983).

264

FORMALDEHYDE AND OTHER ALDEHYDES

Aqueous solubility of aliphatic low-molecular-weight aldehydes decreases with increasing carbon chain length. For this reason, greater nasal penetration has been noted for proprionaldehyde (CH3CH2CHO) and for acetaldehyde (CH3CHO) than for formaldehyde (HCHO). Egle (1972), for example, reported approximately 60% retention (40% penetration) for proprionaldehyde and acetaldehyde as compared to >90% retention for formaldehyde in the canine upper respiratory tract during inspiration. Aliphatic aldehydes with carbon chain lengths greater than four carbon atoms are much less miscible in water, and penetration is therefore more likely to be even greater than that for proprionaldehyde. Values vary for other gases of low solubility (e.g., only about 20% of ozone is retained in the upper respiratory tract of dogs) and few measures are currently available for aldehydes other than formaldehyde (Overton et al., 2001). Additional physiological factors, such as altered clearance mechanisms (epithelial metabolism, mucociliary clearance, and possibly regional air, and blood flow) also affect aldehyde dosimetry. This has been suggested for formaldehyde, as the magnitude of the decrease in the rate of deposition was not solely related to the decrease in ventilation. Studies with other inhalation hazards (sulfur dioxide and ozone) have suggested that other factors influencing nasal penetration include vascular congestion, respiratory flow rate, concentration, duration of exposure, and airway caliber. Morris (1997) has proposed that aldehyde penetration, at least for the rat nasal passages, is strongly influenced by the metabolic capacity of the epithelium. Aldehydes are handled by a number of specific and nonspecific enzymes, including aldehyde dehydrogenases, aldehyde oxidases, or xanthine oxidase (Bohren et al., 1989). The predominant elimination pathway for aldehydes by dehydrogenation yields a carboxylic acid (e.g., formic acid from formaldehyde) through an irreversible step that requires glutathione (Lam et al., 1985; Bhatt et al., 1988) and a hydrogen carrier (e.g., nicotinamide adenine dinucleotide or its phosphate, NAD(P)). Formaldehyde deposits mainly (93%) within the respiratory epithelium with little in the olfactory epithelium of the rat nose (Kimbell et al., 1993; Cohen Hubal et al., 1997; Kimbell et al., 1997). Therefore, the dose rate at low concentrations (<5000 ppb at a breathing rate of 200 mL/min) produces a tissue dose rate approximately equaling the maximal metabolic rate (Vmax) of formaldehyde dehydrogenase (estimated capacity 40 nmol/min) reported by Casanova-Schmitz et al. (1984) for the rat nasal epithelium. Likewise, Morris (1997) proposed that 800,000 ppb acetaldehyde could produce a dose rate at the Vmax of aldehyde dehydrogenase of the rat respiratory epithelium. Accordingly, inhaled concentrations greater than 800,000 ppb would have greater penetration. This proposition is in general agreement with the nasal depositions of 76%, 48%, 41%, and 26% for inspired acetaldehyde concentrations of 1000, 10,000, 100,000, and 1,000,000 ppb (1 ppm), respectively. In addition, differences in airflow patterns within the nasal passages produce different doses to various types of epithelium within the nose. For example due to difference of the metabolic capacity and air flow over the olfactory epithelium, the acetaldehyde concentration necessary to overwhelm metabolism would be lower, 300,000 ppb, than of the respiratory epithelium, 800,000 ppb (Morris, 1997). Because formaldehyde deposits more in the anterior respiratory epithelium, this difference has less of an effect for this compound (Monticello et al., 1996). Morris (1998) also demonstrated that aldehydes can interact in coexposures. For example, simultaneous exposure to acrolein resulted in nonsteady-state acetaldehyde uptake behavior, with uptake efficiencies steadily decreasing with extension of exposure. Acrolein also produced a concentration-dependent reduction in net uptake during the exposure. For example at a flow rate of 200 mL/min, net upper respiratory tract acetaldehyde uptake

BACKGROUND

265

efficiency averaged 43%, 39%, 24%, and 24% in animals simultaneously exposed to 0, 2000, 10,000, or 20,000 ppb acrolein, respectively. The mechanisms of these responses are not known. However, these results demonstrate that caution is necessary in utilizing dosimetric data obtained during exposure to individual vapors to predict relationships that may exist under complex exposure scenarios to multiple vapors. Relative reactivity of the inhaled aldehyde also will influence retention in the upper respiratory tract (Aharonson et al., 1974). In the dog, nasal retention of acrolein (80%) was intermediate between those of formaldehyde (>95%) and proprionaldehyde (60%). The difference between acrolein and proprionaldehyde, both three-carbon molecules, can only be partially explained by acrolein’s higher solubility (acrolein ¼ 210 g/L: propionaldehyde 160 g/L, Table 9.4). The higher chemical reactivity of acrolein may also contribute significantly. Acrolein possesses both a reactive carbonyl group and an electrophilic a-carbon ttributable to the a,b-unsaturated carbon bond positioned next to the carbonyl group (þCHC¼CHOH). Presumably covalent binding with surface macromolecules, like mucin, is greater for acrolein than proprionaldehyde, and would prevent acrolein reentry to a greater extent than that of proprionaldehyde (Bogdanffy et al., 1987). Aldehydes that penetrate the nasal cavity or enter through oral breathing are almost completely deposited in the lower respiratory tract. Heck et al. (1983) examined deposition of [14 C]HCHO in nasal breathing rats and mice. In 6-h tests with formaldehyde concentrations up to 24,000 ppb, approximately 250 nmol [14 C]HCHO/g tissue was deposited in the trachea as compared to 2200 nmol/g tissue in the nasal mucosa (the amount of deposition was nonlinear in relation to dose, with saturation at doses exceeding 6000 ppb). Autoradiography of mouse and rat airways by Swenberg and associates (Chang et al., 1983; Swenberg et al., 1983) demonstrated the presence of radioactivity in the trachea and main bronchi following [14 C]HCHO exposure. Deposition of formaldehyde in the trachea and bronchi of monkeys was greater than that of rats, presumably because rodents breathe principally through their nose. Humans typically breathe between 30% and 60% through their mouth at rest and more during exercise. 9.1.2.2 DNA–Protein Cross-Links and Molecular Dosimetry Initial covalent interactions at the sites of exposure could serve as a sensitive indicator of the dose to the critical biological target site. If such interactions are irreversible and stable, these methods could enable a measure of regional molecular dosimetry. Aldehydes can form cross-links, bridging macromolecules through attack of the electrophilic carbonyl carbon with various reactive groups on proteins, RNA, or DNA (French and Edsall, 1945; Naylor et al., 1988). One reactive group contained in proteins is the amino group, which is also contained in each of these types of macromolecules. Thus, this reaction site can lead to mixed cross-links (e.g., DNA–protein cross-links) via a two-step process initiated by the formation of an unstable hydroxymethyl intermediate (through loss of water from a methylene Schiff base), followed by a nucleophilic attack on the carbon donated by formaldehyde by a second amino group contained on either DNA or a protein. Bifunctional aldehydes, that is, malondialdehyde, or glutaraldehyde, are more effective at forming cross-links. The resulting methylene bridge can either be repaired rapidly, or become temporally stable, and an integrated measurement would thus reflect these competitive processes of formation and degradation (Graftstrom et al., 1984). Although DNA–DNA cross-links are possible, attempts at detection in intact cells have met with limited success (Feldman, 1973; Chaw et al., 1980; Hemminki, 1981). On the contrary, DNA–protein cross-links have been more readily measured (Wilkins and Macleod, 1976; Magana-Schwenke et al., 1978; Magana-Schwenke and Ekert, 1978; Graftstrom et al., 1984; Casanova et al., 1994; Heck and Casanova, 1999). The relationship between

266

FORMALDEHYDE AND OTHER ALDEHYDES

formaldehyde dose and DNA–protein cross-links has, however, been found to be nonlinear over the range of 300–1000 ppb (Swenberg et al., 1983). This nonlinearity makes evaluation of cellular dose by measurements of DNA–protein alone difficult. Lastly, protein–protein cross-links have been detected, and are linear with dose; however, evidence suggests that this covalent binding is primarily between extracellular proteins and therefore does not reflect intracellular dose. This limitation may be important in the evaluation of dose in certain toxicological processes leading to cancer. In addition to the instability of cross-links formed by formaldehyde exposure (Graftstrom et al., 1984), a principal reason for the nonlinearity in the formation of DNA–protein crosslinks is likely to be simply the rapid enzymatic removal of aldehyde before cross-links are formed. Depletion of glutathione alone increases the amount of detectable DNA–protein cross-links and nearly restores the linearity of the dose–response relationship (Lam et al., 1985; Casanova and Heck, 1987), suggesting that this is an important defense mechanism at lower doses. Ultimately the carboxylic acid formed from each aldehyde is converted to CO2 and H2O, either through a multiple-step process involving tetrahydrofolate (initiated by formyl-THF-synthetase), or a reaction catalyzed by catalase. Conceivably, other methods also will prove useful in assessing regional aldehyde dosimetry. Such an indicator could be DNA-adduct levels, and at least two adducts, N6-hydroxymethyl-deoxyadenosine (HOMedA), and N2-hydroxymethyldeoxy guanosine (HOMedG), are possible. However, these hydroxymethyl adducts, like DNA–protein crosslinks, are not stable. Fennel (1990) has presented methods that stabilize these adducts so that they would be suitable for 32 P or electrophore postlabeling. Such a technique could reduce the need for the use of radiolabeled formaldehyde. More recently, the problems of DNA– protein cross-linking estimation of dosimetry have been somewhat circumvented by a combined approach, which includes cytolethality-regenerative cellular proliferation, which is also used to model formaldehyde dosimetry in rats (Conolly et al., 2003, 2004). 9.1.2.3 Populations of Concern At present, no specific human subpopulation has been clearly identified to have an increased risk for adverse effects from envrionmental aldehyde exposures. Previously, persons with asthma were identified as being of special concern when the National Ambient Air Quality Standard was established for sulfur dioxide. By analogy, formaldehyde or other aldehydes with bronchoconstrictive properties, may also represent an added risk to this population (see later for more details). An aspect of this concern includes persons who have developed an immunologic hypersensitivity, but this is a complex issue (see later for details). Aldehydes other than formaldehyde that can penetrate the upper respiratory tract, much like ozone, may produce similar adverse responses. Currently, children have been suggested as a population of concern for ozone and environmental tobacco smoke exposure, and thus aldehyde effects among this population deserve attention. Genetic differences in metabolism could be predisposing risk factors for adverse responses to aldehyde exposure. In general, aldehydes can be oxidized to carboxylic acids by aldehyde dehydrogenases, or reduced to alcohols by alcohol dehydrogenases and aldose (aldose-keto) reductases. Elimination of aldehydes is typically rapid and dependent on these enzymes and pathways that regulate levels of glutathione, tetrahydrofolate, and NAD(P). Alteration of the metabolism responsible for normal function of these defense mechanisms could therefore place individuals at risk. Currently, more that 30 aldehyde dehydrogenase (ALDHs) genes have been identified in humans, and aldehydes are substrates for alcohol dehydrogenases (ADHs) as well (Yoshida et al., 1998; Vasiliou et al., 2004).

BACKGROUND

267

The accumulation of circulating acetaldehyde is responsible for a dysphoric response among individuals with reduced acetaldehyde elimination following ethanol ingestion, which leads to sympathomimetic responses of facial flushing, tachycardia, and muscle weakness, and a rise in circulating catecholamines. Aldehyde dehydrogenase can also be inhibited by disulfiram (tetraethylthiuram disulfide) (DeMaster and Stevens, 1988; Helander et al., 1988) used clinically for the treatment of alcohol abuse, or its metabolites. The potential pathophysiological significance of depressed aldehyde dehydrogenase activity, and its relationship to adverse reaction to aldehydes in the lung, may be important in understanding individual susceptibility. Previously, studies of adverse effecs due to poor ethanol metabolism have identified specific polymorphisms in NADþ-dependent aldehyde dehydrogenase enzymes among Asians and Native Americans (Hsu et al., 1987; Card et al., 1989; Farres et al., 1989; Goedde and Agarwal, 1989; Kurys et al., 1989; Shibuya et al., 1989). Ingestion of ethanol among susceptible individuals includes normal, but rapid, formation of acetaldehyde via an allelic variant in alcohol dehydrogenase (an NAD-dependent cytosolic enzyme contained principally in the liver) that increases aldehyde formation. Ethanol also is converted to acetaldehyde by microsomal ethanol-oxidizing system (mainly by cytochrome P450, family 2, subfamily E, polypeptide 1: CYP2E1) or catalase (Crabb, 1990; Vasiliou et al., 2004). Normally, acetaldehyde is rapidly converted to acetic acid. However, because of an additional alteration in aldehyde dehydrogenase 2 (ALDH2), concentrations of this intermediate are elevated. This alteration is often due to a polymorphism in ALDH2 (G ! A transition in exon 12) that produces a Glu ! Lys substitution at position 487. This allelic variant has been found in high frequency (40–60%) among, and may be confined to, Asians (Yoshida et al., 1998). Based on blood acetaldehyde levels following alcohol consumption, individuals with this polymorphism have about a 20-fold increase in acetaldehyde levels (Mizoi et al., 1994). Besides having a role in systemic aldehyde metabolism, ALDH2 may have a tissue specific role dependent on the route of exposure. For example, gene-targeted mice lacking functional Aldh2 are more susceptible to adverse pulmonary responses during inhalation of 5 ppm acetaldehyde (Isse et al., 2005). Importantly, although individuals carrying ALDH2 variants are resistant to alcoholism, they are at increased risk for several cancers including esophageal, head and neck, oropharyngolaryngeal, and gastrointestinal cancer (Muto et al., 2002; Yokoyama and Omori, 2005). Persons with ALDH2 deficiency who drink alcohol excessively have increased acetaldehyde-derived DNA adducts (Matsuda et al., 2006). Brooks and Theruvathu (2005) also noted that the repair of such adducts is complex, involving multiple pathway, and inherited variation in the genes encoding the proteins involved in the repair acetaldehyde secondary adducts may contribute to susceptibility to alcoholic beverage-related carcinogenesis. In addition to altered ethanol-related aldehyde formation and elimination, other dehydrogenases include alcohol dehydrogenase 5 (ADH5, as known as chi-ADH, formaldehyde dehydrogenase, and S-nitrosoglutathione reductase) may have a role in individual susceptibility to the adverse health effects induced by aldehyde exposure. Despite being named alcohol dehydrogenase, ADH5 has a high affinity for formaldehyde but only a low affinity for ethanol. ADH5 catalyzes the oxidation of S-hydroxymethylglutathione, a spontaneous adduct between formaldehyde and glutathione, into S-formylglutathione in the presence of NAD (Koivusalo et al., 1989). ADH5 activity is increased by fatty acids (Engeland et al., 1993). Five nonsynonymous single nucleotide polymorphisms existing in exons 5, 6, 7, and 8 of ADH5, but the functional significance of the resulting amino acid substitutions has yet to be determined.

268

FORMALDEHYDE AND OTHER ALDEHYDES

Interestingly, ADH5 is also responsible for regulating the levels of S-nitrosoglutathione (GSNO), which is formed from and provides a source nitric oxide (Liu et al., 2004). Genetargeted mice lacking Adh5 have increased nitric oxide available, and thereby are protected from airway hyperreactivity following inhaled antigen challenge (Que et al., 2005). Other aldehydes have tissue specific metabolism, which can differ from that of formaldehyde. For example, acrolein also can be a substrate of liver aldehyde dehydrogenase (EC 1.2.1.5) and lung or liver microsomal epoxidase (EC 1.14.14.1) (Patel et al., 1980). The exact enzyme subtypes have yet to be determined, but probably include multiple ALDH members including ALDH2, ALDH3A1, and ALDH5A1 (Vasiliou et al., 2004) and possibly cytochrome P450 isozymes including CYP3A1. Clear evidence exists indicating that acrolein oxidation can be catalyzed by human ALDH3A1, but not ALDH1A1 (Bunting and Townsend, 1998; Townsend et al., 2001). This metabolism is tissue specific. For example, acrolein can be converted to acrylic acid by rat liver S9 supernatant, cytosol, or microsomes, but not by lung fractions (Patel et al., 1980). In the lung a major class 3 ALDH is ALDH3B1 (Hsu et al., 1994), and polymorphisms in this gene have been associated with difference in individual susceptibility to schizophrenia (Sun et al., 2005). Acrolein incubation with rat liver or lung microsomes in the presence NADPH yields glycidaldehyde and its hydration product glyceraldehyde, suggesting involvement of microsomal cytochrome P-450-dependent epoxidase (Patel et al., 1980). Glycidaldehyde is a substrate for lung and liver cytosolic glutathione S-transferases (EC 2.5.1.18) and can also be hydrated to glyceraldehyde (Patel et al., 1980). Glyceraldehyde can be metabolized via the glycolytic pathways. In the lung, the primary elimination pathway of many aldehydes is involves glutathione conjugation by glutathione S-transferases (GSTs). The cytosolic GSTs exist as monomers, and are catalytically active as homo- or heterodimers and are divided into seven classes: alpha, mu, omega, pi, sigma, theta, and zeta (Nebert and Vasiliou, 2004; Hayes et al., 2005; McIlwain et al., 2006). Of these, members of the mu (GSTM1), theta (GSTT1), and pi (GSTP1) families have garnered the most attention. Both GSTM1 and GSTT1 have null genotypes that yield no protein and these genotypes are common, being present in about 50% and 25% in Caucasian populations, respectively. In addition, the GSTT1 null variant is present in about 40% of Asian populations (Gilliland et al., 2002; Hayes et al., 2005). The GSTP1 gene has two common polymorphisms (with a frequency of about 5–10%) have been found to alter catalytic efficiency with isoleucine 105 valine (ile105val) and alanine 114 valine (ala114val) transitions (Ali-Osman et al., 1997; Watson et al., 1998). In terms of aldehyde conjugation, GSTM1 is more efficient using 4-hydroxyalkenals as substrates whereas GSTP1 is more efficient using acrolein or crotonaldehyde as substrates (Berhane et al., 1994), and the Val105 Val114 isoform of GSTP1 is less efficient in catalyzing acrolein (but not crotonaldehyde) than other variants (Pal et al., 2000). GSTP1 is a major form found in the lung (Watson et al., 1998) and the Val105 genotype is associated with an increase susceptibility to lung cancer from environmental tobacco smoke (Miller et al., 2003) especially when combined with GSTM1 null genotype (Wenzlaff et al., 2005). Polymorphisms in GSTP1 is also associated with younger age of onset of lung cancer (Miller et al., 2006) In addition to being associated with lung cancer, the Val105 GSTP1 genotype is associated with decreased lung function in smokers that have a serpin peptidase inhibitor, clade A (alpha-1 antiproteinase, antitrypsin), member 1-deficiency, suggesting GSTP1 has a modulatory role in chronic obstructive pulmonary diseases (Rodriguez et al., 2005). It should be noted that aldehydes, when conjugated to glutathione, have toxic potential, because the aldehyde group remains intact (Horvath et al., 1992).

SINGLE-EXPOSURE HEALTH EFFECTS

269

The catalytic activity and protein levels of these enzymes can be altered by exposures to aldehydes. For example, nasal metabolic activity following exposure of formaldehyde, acetaldehyde, and acrolein on male Wistar rats (5–6 per group) exposed 6 h/day, for 3 consecutive days, in a nose-only exposure chamber to acrolein at concentrations of 0, 250, 670, or 1400 ppb (Cassee et al., 1996b). Mixture (rather than single chemical exposure) effects were noted on regional nasal histopathological and biotransformation enzymes measured in homogenates of nasal tissue. Only in the only 1400 ppb acrolein exposure group was glutathione S-transferase activity depressed, while ADH5 and ALDH activities increased. Similar, in vitro acrolein can covalently modify and inactivate GST (Berhane and Mannervik, 1990) and carbonic anhydrase II (Tu et al., 1989). In contrast, acrolein, a substrate for aldo-keto reductase family members (including AKR1A1, AKR1B10, AKR1C3, and AKR7A1) (yielding allyl alcohol), can activate this enzyme in vitro, but this is not likely to occur in vivo because it is prevented by NADPþ (Lan et al., 2004; Fukumoto et al., 2005). Other aldehyde substrates for aldo-keto reductase included crotonaldehyde and cinnamaldehyde (Ellis and Hayes, 1995; Gardner et al., 2004).

9.2 SINGLE-EXPOSURE HEALTH EFFECTS 9.2.1

Formaldehyde

9.2.1.1 Symptomatic Response in Controlled Human Exposures Moderate concentrations of formaldehyde produce rapid responses including eye (50–2000 ppb) and upper respiratory tract (100–25,000 ppb) irritation that can become intolerable (Table 9.5). Eye TABLE 9.5

Symptomatic Responses in Humans to Formaldehyde

Formaldehyde Concentration (ppm)

Response

Eye irritation (blinking rate, lacrimation, conjunctivitis) 0.01 Detectable by some 0.30 Slight but tolerable response 0.50 Intermediate response 0.80 1.7–2.0

Severe response Marked eye blinking

References Schuck et al. (1966) Rader (1977) Bourne and Seferian (1959) Lang et al. (2008) Wayne et al. (1977) Bender et al. (1983)

Upper respiratory tract Irritation (nasal secretion or dryness, throat irritation) 0.03 Minimal or no effect Weber-Tschopp et al. (1977) 0.25–1.39 Moderate irritation Kerfoot and Mooney (1975) Schoenberg and Mitchell (1975) Anderson and Molhave (1983) Lang et al. (2008) 1.7–2.1 Significant throat irritation Weber-Tschopp et al. (1977) 3.1 Severe to intolerable Kane and Alarie (1977) irritation Odor threshold 0.05 Odor threshold Pettersson and Rehn (1977) 0.17 Detected by 50% exposed Pettersson and Rehn (1977) 1.5 Detected by all subjects Pettersson and Rehn (1977)

270

FORMALDEHYDE AND OTHER ALDEHYDES

irritation progresses to a greater extent with continuous exposure than with discontinuous exposure, whereas nose and throat irritation were significantly greater in discontinuous exposures (Weber-Tschopp et al., 1977). Tolerance to nose and throat irritation can develop in the same individual who continues to become increasingly sensitive to eye irritation during exposure (Anderson and Molhave, 1983). The characteristic odor of formaldehyde is detectable at concentrations below those that produce irritation (Bender, 2002; Arts et al., 2006). However, as with many compounds, the odor threshold (20–500 ppb) can vary widely between individuals, and acclimatization can occur during exposure (Berglund and Nordin, 1992). For these reasons, the detection of odor (or more importantly, the lack thereof) must not be used as an indicator of safety. Sequelae to massive acute exposure, an aversion response elicited merely by the odor, can develop (Shusterman et al., 1988). This can be viewed as a psycho-protective response resulting from behavioral conditioning in which the initial single acute exposure accompanied by observable pathophysiological effects serves as the unconditional stimulus. The odor (at subirritant concentrations) during a secondary, re-exposure after the initial toxic acute exposure, then serves as the conditioned stimulus. Following such a regimen, conditioned subjects can develop subjective (including those outlined above as well as a perception of poor air quality, discomfort, or a “desire to leave the room”) or objective (changes in heart rate and breathing pattern) responses. Although the enhanced response is psychogenic in origin, response generation is often involuntary, and may involve minimal personality components. In general, formaldehyde exposures of 50–150 ppb produce mild, transient, concentration-dependent responses in 10–50% of the persons exposed (Bernstein et al., 1984). These types of responses are common, and often are the primary or solitary finding associated with low-level environmental exposure (see below). In one study, 410 ppb formaldehyde exposure of 2 h, increased mucosal inflammation (leukocytes and albumin recovered in nasal lavage recovered 4 and 18 h after exposure) in healthy and asthmatic subjects (Pazdrak et al., 1993). Eye irritation develops frequently in the range of 1000–3000 ppb in most subjects, and upper respiratory irritation is frequent at concentrations of 1000–11,000 ppb (NRC, 1981; IARC, 1995a; Yang et al., 2001; Arts et al., 2006). At concentrations of 4000–5000 ppb, many subjects cannot tolerate prolonged exposure (IARC, 1982; WHO, 1989; IARC, 1995a). 9.2.1.2 Respiratory Mechanics in Laboratory Animals Inhalation of formaldehyde decreases minute volume in laboratory animals, primarily as a result of a decrease in respiratory breathing frequency (more than through a decrease in tidal volume, which may even increase) (Table 9.6). Concomitantly, pulmonary resistance increases with or without a slight decrease in pulmonary compliance. These responses are rapid in onset, can remain constant during exposure (response plateau), and are readily reversible with exposure cessation. The magnitude of the “response plateau” developed during exposure is dose dependent. The persistence of the response, however, varies somewhat with species. In studies with rats (F-344) and mice (Swiss-Webster but not B6C3F1), decreases in minute volume are not maintained during exposure, suggesting tolerance. In B6C3F1 mice, the induced decrease in minute volume remains more consistent during exposure. These differences in minute volume, as noted by Barrow et al. (1983), could alter the dose from exposure to 15,000 ppb (a dose found sufficient to induce nasal carcinoma in rats but not mice). In B6C3F1 mice, minute volume is decreased 75% as compared to 45% in F-344 rats. This would reduce the estimated dose to the mouse nasal epithelium from 0.156 to of 0.076 mg/min/cm2. Changes in breathing pattern, therefore, may be viewed as protective.

SINGLE-EXPOSURE HEALTH EFFECTS

TABLE 9.6

271

Acute Pulmonary Responses in Animals to Formaldehyde Concentration (ppm)

Species

Threshold

Decreased minute volume/respiratory rate Mice (Swiss-Webster) 0.50 (B6C3F1)a 0.50 Rats (F-344)a 0.95 Guinea pig 0.30 Increased airway resistance Guinea pig a

0.30

ED25%

ED25%

References

0.8 1.8 3.1 11.0

3.1 4.4 13.1 >49.0

Kane and Alarie (1977) Barrow et al. (1983) Chang et al. (1981) Amdur (1960)

2.0

11–49

Amdur (1960)

Tidal volume was noted to increase with exposure in animals during exposures to 6.4 ppm.

The formaldehyde dose producing decreases in respiratory rate in mice is comparable to the dose producing increases in symptomatic responses in humans (see above). This led Alarie and associates to propose that evaluation of sensory irritation in mice might serve as a useful test system to evaluate relative irritation potential of various compounds (Kane and Alarie, 1977; Arts et al., 2006). From these finding, the authors suggested that the threshold limit value should be set in the range of 30–300 ppb to prevent irritant responses. 9.2.1.3 Respiratory Mechanics in Humans after Single Exposure When inhaled alone in controlled exposure chambers, formaldehyde produces few changes in respiratory mechanics among healthy subjects either during or shortly after a single exposure at environmentally relevant levels (Table 9.7). Following exposures ranging from 300 to 7500 ppb, essentially no change has been noted in this group. For example, Day et al. (1984) exposed 18 volunteers to 1000 ppb formaldehyde, with nine subjects having complained of various nonrespiratory adverse effects from the urea formaldehyde foam insulation in their homes. No effects were noted in subjects who exposed to 1000 ppb formaldehyde for 90 min or 1100 ppb formaldehyde (produced from urea formaldehyde foam insulation) for 30 min. Exposure to 3000 ppb for 60 min did produce a small change in forced expiratory volumes and flows (Green et al., 1987), but this response was transient, and reversed when exposures were extended to 120–180 min (Sauder et al., 1986). Formaldehyde’s ability to alter pulmonary function has also been tested between two additional, possibly more susceptible, populations. The first population, persons with asthma, was selected based on the observation that another soluble irritant, like SO2, had lower dose thresholds to initiate bronchoconstriction in this group compared to control subjects (Sheppard et al., 1980). One study involved oral breathing (mouthpiece) for 10 min to 1000–3000 ppb (Sheppard et al., 1984), another involved chamber exposures for 40 min to 2000 ppb (Witek et al., 1987). In both studies, subjects failed to develop an increase in pulmonary resistance or a decrease in flow rate when tested at rest or with exercise. In the latter study, however, airway reactivity may have been altered (see below). The second clinical population tested were persons with previous occupational exposures. The major objective of these studies was to reproduce, in the laboratory, observed across-shift changes in pulmonary function. This effect is often evident and is discussed later under Effects of Multiple Exposure section. Another objective of these studies was to uncover an immunologically based response in subjects referred to the clinic for a direct

272

Length of Exposure (min) Route of Exposure

Respiratory Function of Humans Exposed to Formaldehyde

40

60

180

2

2.0

3.0

3.0

7.5

Oral (mouthpiece)

Oronasal (chamber, with exercise)

Oronasal (chamber, with exercise)

Oronasal (chamber, with or without exercise) Oronasal (chamber)

3.0 3.0

10 60

Oral (mouthpiece, with exercise) Oronasal (chamber)

Individual with asthma in clinical studies 2.0 40 Oronasal (chamber, with exercise)

180

0.5–3.0

Healthy individual (nonsmoking) in clinical studies 0.3–1.6 300 Oronasal (chamber)

Formaldehyde Concentration (ppm)

TABLE 9.7

No change in FVC, FEV1.0, MMEF (n ¼ 12) No change in sRaw No change in FVC, FEV1.0, FEF25–75% SGaw

No change in FVC, FEV1.0, FEF25–75% or Raw (n ¼ 16) No change in FVC, FEV1.0, FEF25–75% sGaw (n ¼ 9–10) No change in FVC, FEV1.0, MMEF (n ¼ 15) 2.5–3.8% change in FVC and FEV1.0 (n ¼ 22) 2–7% change in FVC, and FEV1.0 in 60 min; no change in FVC and FEV1.0 at 180 min No change in FEV1.0 or Raw

Findingsa (No. of Subjects)

Sheppard et al. (1984) Green et al. (1987)

Witek et al. (1987)

Rader (1977)

Sauder et al. (1986)

Green et al. (1987)

Schachter et al. (1986)

Kulle et al. (1987)

Anderson and Molhave (1983)

References

273

20

30 20

30

0.1–3.0

1.8 2.0

3.2

Oronasal (chamber)

Oronasal (chamber) Oronasal (chamber)

Oronasal (face mask)

Oronasal (chamber, with exercise)

Small (200 mL) change in FVC and FEV1.0 (n ¼ 38) No change in FVC, FEV1.0, MMEF (n ¼ 15) 12  2% decrease in FEV1.0 as compared to 8  3% in control; little or no change in MMEF and other measures 13% decrease in FEV1.0 (n ¼ 4) Decreases in Raw in 12 of 230 (5%) persons 12% decrease in FEV1.0 (n ¼ 8); three with decrease >25% in FEV1.0

Burge et al. (1985)

Burge et al. (1985) Nordman (1985)

Frigas et al. (1984)

Alexandersson and Hedenstierna (1988) Schachter et al. (1987)

a Abbreviations. FVC: forced vital capacity; FEV1.0: forced expiratory flow at 1.0 s; FEF25–75%: mean of forced expiratory flow at 25% and 75% vital capacity; (s)Raw: (specific) airway resistance; MMEF: midmaximal expiratory flow; (s)Gaw: (specific) airway conductance.

40

2.0

Individual with Occupational Exposure 0.3–0.6 8 h work Oronasal (at work)

274

FORMALDEHYDE AND OTHER ALDEHYDES

bronchial provocation. Such investigations have documented immediate and delayed responses, much like those found in antigen-induced immediate hypersensitivity reactions, but these responses appear to be rare (occurring in 12 of 230 persons tested, or 5%, in one study). Given the widespread exposure to formaldehyde and documented formaldehyde skin sensitization (see below), these results suggest that there are remarkably few cases of pulmonary hypersensitivity. This implies that sensitization by aldehyde inhalation is not a primary mechanism for a respiratory effect, but rather, observed effects are associated with a direct (nonimmuno-specific) irritant effect. This limited role for immunologic sensitization is particularly evident if hypersensitivity is narrowly defined as a response to extremely low doses of a compound, well below the dose necessary to induce irritation. Few data exist comparing possible formaldehyde effects on minute volumes in humans with this well-documented effect noted in laboratory animals. This is unfortunate, because this information would be useful if directly compared to the symptomatic responses noted within the same species. 9.2.1.4 Airway Reactivity Aldehydes can influence the underlying bronchial reactivity of the airways following an initial, transient bronchoconstriction (typically produced only at high (10,000 ppb) concentrations). Several other irritants (sulfur dioxide, ozone, and toluene diisocyanate) that induce an immediate bronchoconstriction also can induce bronchial hyperreactivity. Hyperreactivity is experimentally defined as heightened responsiveness to inhaled methacholine (a stable form of acetylcholine) or histamine, and is a diagnostic feature of asthma (Boushey et al., 1980; Barnes et al., 1989). Following a single initial exposure of healthy individuals, this condition is typically reversible and lasts for 12–48 h. In persons with asthma, however, this condition can persist for several years. Hyperreactivity frequently lacks an immunologic component, and is thus termed nonspecific. In these cases no identifiable causative antigen can be found, and a general heightened response is noted after a wide range of irritant stimuli. The relationship between each phase of the dual responses (e.g., an immediate and/or a delayed decrease in FEV1.0) after antigen presentation/challenge and hyperreactivity is currently unclear. Animal models have been informative in the past. Formaldehyde exposure of guinea pigs for 2 h produced a small change in pulmonary resistance, with an estimated half-maximal change in bronchial reactivity at 8000 ppb (Swiecichowski et al., 1993). When the duration of exposure was extended to 8.0 h, 1100 ppb formaldehyde produced a significant increase in reactivity, greater in magnitude as that produce when exposure was short but to a eightfold higher concentration. This study suggests that low-level exposures of several hours may have effects not detectable with shorter (
2 h) exposures. In another animal model, ICR strain mice were sensitized intraperitoneally with dust mite allergen (Der f) prior to exposure to 0.5% formaldehyde mist once a week for 4 weeks (Sadakane et al., 2002). Mice were then challenged by intratracheal instillation with Der f and airway inflammation examined. When combined with following Der f challenge, formaldehyde exposure enhanced the histopathology (including eosinophil infiltration and goblet cell formation) and lung levels of two cytokines associated with asthma [interleukin-5 (IL-5) and chemokine (C-C motif) ligand 5 also known as Regulated on activation, normal T cell expressed, and presumably secreted (RANTES)]. Bronchial reactivity has been examined in humans and no changes have been recorded in healthy persons after exposures up to 3000 ppb formaldehyde for 3 h (Sauder et al., 1986; Kulle et al., 1987; Green et al., 1987). In studies with persons with asthma, however, Witek et al. (1987) reported a decrease in the dose of methacholine necessary to produce a 20%

SINGLE-EXPOSURE HEALTH EFFECTS

275

decrease in forced expiratory volume in 1.0 s (FEV1) in 8 of 12 individuals following a 40-min exposure to 2000 ppb formaldehyde. The mean response for all 12 subjects was not statistically different from control; thus, further study is required before any conclusions on the effect of formaldehyde on airway reactivity can be drawn. Burge et al. (1985) also has suggested that a relationship exists between bronchial responses of subjects with previous occupational exposure to formaldehyde and their underlying bronchial reactivity. Lemiere et al. (1995) reported that three individuals developed decreased FEV1 after being exposed to formaldehyde resin dust, with one of these subjects also being responsive to formaldehyde gas. Kim et al. (2001) presented a case report of an individual with occupational asthma. Working in an environment with a mean formaldehyde level of 60 ppb (with occasional peaks of 120–130 ppb), this worker had across-shift decreases in FEV1, which was reversed by inhaltion of a bronchodilatory (beta-2 adrenergic agent). In addition, shortterm exposure (20 min) at concentrations of 500 ppb in the laboratory lead to an immediate and delay (lasting over 20 h) decreased. No circulating antibody or cutaneous reactivity specific to formaldehyde–human serum albumin conjugate was detected. In another case report, by Vandenplas et al. (2004), a single accidental high level formaldehyde exposure led to changes in FEV1 and a persistent (lasting over 1 year) increase in bronchial hyperactivity to histamine, and again this seemed not to involve a specific immunological mechanism. To begin to understand the possible mechanism, Kim et al. (2002) examined human mucosal microvascular endothelial cells following in vitro exposure to formaldehyde, and found an increase in the surface expression of intercellular adehesion molecule 1 (ICAM1) and vascular cellular adehesion molecule 1 (VCAM-1). In addition, the adhesiveness between endothelial cells and eosinophils was also increased by formaldehyde exposure. This implies that formaldehyde is acting as an irritant of the nasal mucosa that may lead to an increasing the expressions of adhesion molecules and interaction that increase eosinophil trafficking (an even critical to allergic rhinitis and possible asthma). In sensitized persons with asthma, Casset et al. (2006b) found that a short, low dose preexposure (30 min  30 ppb formaldehyde) lowered the threshold dose of dust mite need to induce allergen-mediated bronchoconstriction. In addition, the late-phase reaction (defined as a 15% decrease in FEV1) was also more common in persons exposed to both allergen and formaldehyde as compared to persons exposed to allergen and filtered air. Consistent with an increased asthmatic response, the level of eosinophil cationic protein in serum or induced sputum were greater following exposure to formaldehyde with antigen. In a recent review of the literature, these authors also noted that the risk for development of asthma is increased (approximately 1.4-fold) in homes that exceed levels of 50 ppb formaldehyde (Casset et al., 2006a). Similarily, in several other studies, children exposed to formaldehyde levels of >50 ppb were at increased risk of having asthma (see Other Responses to Multiple Exposure section). 9.2.1.5 Mucociliary Clearance Formaldehyde markedly inhibits mucociliary clearance in animals and humans (Table 9.8). This effect was noted as early as 1942 by Cralley (1942), who observed an inhibition and complete stasis of ciliary beating after formaldehyde exposure. As concerns arose about the health effects of cigarette smoke, several investigators (Guillerm et al., 1961; Wynder and Hoffman, 1963; Falk, 1963; Kensler and Battista, 1963; Carson et al., 1966; Dalhamn and Rosengren, 1971; Sisson and Tuma, 1994) confirmed that aldehydes (including formaldehyde and acetaldehyde) in smoke can alter ciliary function and extended this observation, finding that formaldehyde exposure as short as 12 s could reversibly inhibit ciliary activity.

276

FORMALDEHYDE AND OTHER ALDEHYDES

TABLE 9.8 Concentration

Formaldehyde Inhibition of Mucociliary Clearance in Laboratory Animals Duration of Exposure

Response

Species

20 ppm

10 min

10 ppm

–

Depression of ciliary Rabbit activity without recovery Cilia stasis in 20% Rabbit

52 ppm

–

Cilia stasis in 90%

Rabbit

32 ppm

11.5 min

Cilia stasis

Guinea pig

20 ppm

4h

Rat (Sprague– Dawley)

9.6 ppm

30 min

4.4 ppm 1.4 ppm 0.2 ppm 66 mg/cm2

30 min 30 min 30 min 60 min

33 mg/cm2

60 min

16 mg/cm2

60 min

Decrease early clearance of particles; no effect on delayed clearance (alveolar) Increase mucus flow rate at 2 min; muco- and cilia stasis at 2 and 3 min Mucostasis in 8 min Increase in clearance No effect Reduced ciliary activity within 10 min Reduced ciliary activity within 10–20 min Reduced ciliary activity within 30–40 min

References Cralley (1942) Dalhamn and Rosengren (1971) Dalhamn and Rosengren (1971) Oomichi and Kita (1974) Mannix et al. (1983)

Frog palate

Morgan et al. (1984)

Frog palate Frog palate Frog palate Rabbit and pig

Morgan et al. (1984) Morgan et al. (1984) Morgan et al. (1984) Hastie et al. (1990)

Rabbit and pig

Hastie et al. (1990)

Rabbit and pig

Hastie et al. (1990)

The onset and duration of the effects of formaldehyde on mucociliary clearances are dose dependent. Formaldehyde exposures initially diminish the movement of the surface mucus layer before decreasing ciliary beat frequency (Morgan et al., 1984), a result could be due to covalent reaction of formaldehyde with mucus macromolecules that alters the physical (rheologic) properties essential for effective energy coupling with the underlying cilia. Doses of formaldehyde sufficient to inhibit ciliary activity can also reduce the number of cilia extractable from exposed tissue preparations (Hastie et al., 1990). In cilia recovered from exposed epithelia, the specific activity of ATPase and dyneins and tubulin protein content are decreased. These responses were reversible in less than 2 h after exposure, suggesting that recovery is not dependent on de novo protein synthesis, and that ciliary loss and recovery are a dynamic process. In studies with a related aldehyde of concern due to it formation from alcohol and possible interaction with cigarette smoke, Wyatt et al. (1999) found that acetaldehyde diminished ciliary beat frequency through a protein kinase C dependent pathway. In isolated tissue preparations, responses to formaldehyde are dose dependent. In the past, a single exposure to a concentration of 1000–5000 ppb was thought to have no effect, or increases mucus transport rates, whereas concentrations between 5000 and 10000 ppb depress transport (Table 9.8). However, Schafer et al. (1999) reported a decrease in ciliary beat frequency in nasal epithelial cells isolated from persons exposed to 4100 ppb formaldehyde for 2 h. Similarly, Flo´-Neyret et al. (2001) performed a dose–response analysis, and found that 1250 ppb formaldehyde inhibited mucociliary

SINGLE-EXPOSURE HEALTH EFFECTS

277

transport rate within 30 min. In animals, depression of overall particle clearance has been noted at a concentration of 20,000 ppb (4-h exposure) in rats (Mannix et al., 1983; Adams et al., 1987). In humans, however, nasal particle clearance was depressed at concentrations as low as 240 ppb (4–5 h), with maximal effects observed at concentrations of 400 ppb (Anderson and Molhave, 1983). These effects on clearance are most prominent in the anterior portion of the nose. Clearance is dependent on mucus composition, mucus quantity, and ciliary function. Clearance can be compromised in humans at concentrations below those necessary to produce single effects in in vitro preparations (i.e., decreased ciliary beat frequency). This suggests that this compound acts through interference with more than a single cellular or extracellular process. For example, it is probable that ciliary proteins are sensitive to formaldehyde (Hastie et al., 1990), protein–protein methylene cross-links in mucus can alter the tertiary structure of this glycoprotein (Morgan et al., 1984), or protein kinase C activation diminishes ciliary function (Salathe et al., 1993; Wong et al., 1998; Wyatt et al., 1999). Acting together, these effects could diminish particle clearance at lower doses. 9.2.1.6 Effects of Aerosol–Formaldehyde Coexposures Amdur (1960), a founder of inhalation toxicology, reported that formaldehyde-induced increases in the respiratory resistance of guinea pigs were potentiated by the simultaneous administration of submicrometer sodium chloride aerosol (sodium chloride alone produced no effect). A response, induced by a formaldehyde concentration delivered as a formaldehyde–aerosol mixture when breathed through the nose, was greater than the response to the same formaldehyde dose (gas alone) when administered directly to the lung through a tracheal cannula. This interesting observation suggests that the increment added by the aerosol is not solely a result of the transfer of more formaldehyde to the lungs but may reflect additional factors. Similarly, Kilburn and McKenzie (1978) reported that coexposure of formaldehyde with carbon particles produces greater recruitment of leukocytes into the airway epithelium and epithelial cytotoxicity (cytoplasmic vacuolization and nuclear aberrations) in trachea and bronchi of Syrian golden hamsters. This effect peaked 24–48 h after exposure. Because many environmental exposures involve coexposure to ambient and occupational aerosols, for example, resin particulate matter (Lemiere et al., 1995), future investigations clarifying these issues would add greatly to our understanding of how complex mixtures act in concert to exert aldehyde toxicity. 9.2.2

Other Aldehydes

9.2.2.1 Symptomatic Responses and Respiratory Mechanics The decrease in respiratory rate, induced by a wide array of aldehydes, has been studied in mice by Steinhagen and Barrow (1984). In this experimental system, a,b-unsaturated aliphatic aldehydes, acrolein and crotonaldehyde, decreased respiratory rate at half-maximal concentration (ED50) of 1000 and 3500 ppb as compared to 3100 ppb for formaldehyde. In contrast, the half-maximal dose for acetaldehyde was much greater (>2.85 ppm). Acrolein is consistently more potent that formaldehyde in multiple assays. Saturated aliphatic aldehydes with two or more carbons (e.g., butyraldehyde or propionaldehyde) had half-maximal concentrations of 0.75–4.2 ppm, whereas cyclic aldehydes (e.g., 3-cylcohexane-1-carboxyldehyde or benzaldehyde) exert its effects in intermediate doses ranging from 60000 to 400,000 ppb. Thus, the relative potency was acrolein > crotonaldehyde  formaldehyde > benzaldehyde  acetaldehyde.

278

FORMALDEHYDE AND OTHER ALDEHYDES

This apparent relationship holds for most other toxic endpoints including measures of pulmonary function in humans (Pattle and Cullumbine, 1956), half-maximal lethal dose in mice, guinea pigs, and rabbits (Salem and Cullumbine, 1960), and nasal pathology in rats (Lam et al., 1985; Roemer et al., 1993; Cassee et al., 1996b). Acrolein, at doses less than 1000 ppb, can produce a number of pulmonary effects. Murphy et al. (1963) reported that increases in pulmonary resistance and decreases in respiratory rate in guinea pigs exposed to 400–1000 ppb acrolein. As with exposure to formaldehyde, these changes were rapid in onset, remained somewhat constant (“response plateau”) during exposure, and reversed within 60 min after exposure. Atropine inhibited the acrolein-induced change in resistance, suggesting that involvement of a vagally mediated cholinergic pathway. That this response is so readily reversible, with cessation of exposure, also suggests that continuous occupancy of an irritant receptor by acrolein during inhalation is necessary for the initiation of this reflex. Microelectrode recordings of trigeminal nerve fibers during inhalation of aldehydes (Kulle and Cooper, 1975) are consistent with involvement of this neural pathway in the decrease in respiratory rate (Kane and Alarie, 1977). The role of specific sensory afferent pathways was further examined by Lee et al. (1992), who reported that acrolein inhalation evoked an inhibitory effect on breathing with a prolongation of expiration and bradycardia. As determined by a number of interventions, acrolein activated both vagal C-fiber active afferents and rapidly adapting irritant receptors, and suggested that the elongation of expiration was due to stimulation of the former afferent pathway. Similar to acrolein, formaldehyde stimulates C-fiber nerves and can stimulate the release of substance P, which may induce a neurogenic inflammatory response. In rat airways, this is marked by an increase in vascular permeability that is mediated predominantly by stimulation of the tachykinin NK1 receptor (Ito et al., 1996). In addition, Fujimaki et al. (2004) found that mice exposures to 2000 ppb formaldedhye for 12 weeks increased substance P levels in plasma. High acrolein concentrations (>20,000 ppb) are an irritant component of smoke, and are thought to be a causative agent in pulmonary edema, pulmonary hypertension, and acute lung injury that result from such exposures (Hales et al., 1988, 1992; Barrow et al., 1992). This process is dependent on eicosanoid formation and inhibitors of specific pathways may be benefitial therapeutic approaches (Janssens et al., 1994; Hales et al., 1995). 9.2.2.2 Airway Reactivity In addition to the transient increase in baseline pulmonary resistance, acrolein exposure of 800 ppb (2 h) can produce bronchial reactivity in guinea pigs (Leikauf et al., 1989a). Hyperreactivity occurred as early as 1 h after exposure to 1300 ppb, became maximal at 2–6 h, and lasted for longer than 24 h. This response was accompanied by an increase in three bronchoactive eicosanoids (prostaglandin F2a thromboxane B2, and leukotriene C4) in bronchoalveolar lavage fluid. Inhibition of 5-lipoxygenase diminished the response (Leikauf et al., 1989b), indicating lower respiratory tract epithelial injury. An increase in leukocyte infiltration was also noted, occurring 6–24 h after exposure. These findings suggest that acrolein-induced hyperreactivity occurred by a pathway dependent on acute injury to the airway epithelium and mediator release. In addition, migration of leukocytes into the airway does not precede hyperreactivity, suggesting that injury to cells normally present in the lung during exposure produces the mediators responsible for hyperreactivity. Ben-Jebria et al. (1994, 1995 found that lumenal exposure of isolated ferret trachea to 300 ppb acrolein for 1 h decreased the contractile dose of cholinergeric agonists (carbachol or acetylcholine), and increased the maximal contraction (indicative of increased smooth muscle reactivity). Subsequent studies with

SINGLE-EXPOSURE HEALTH EFFECTS

279

human and rat tracheal smooth muscle exposed to acrolein (effective dose typically 0.1–0.2 mM) demonstrated that cholinergic enhancement of contraction was accompanied by increased membrane current (Hyvelin et al., 2001), and oscillations in intracellular calcium (Roux et al., 1998; Hyvelin et al., 2000). Turner et al. (1993a) examined airway responses to intravenous substance P following acrolein exposure. Guinea pigs were exposed twice to 1600 ppb acrolein (7.5 h/day on 2 consecutive days) and followed for up to 28 days. Again, pulmonary inflammation and epithelial damagewere prominent 1 day after acrolein exposure. Neutral endopeptidase (NEP) activity was decreased in the lungs, trachea, and liver 1 and 7 days after acrolein exposure. Twenty-eight days after exposure, NEP activity in the lungs and liver was not significantly different in vehicle- and acrolein-exposed guinea pigs, but was still reduced in tracheal tissue. Acrolein increased airway reactivity to substance P that lasted for up to 28 days following exposure. Thiorphan, a NEP inhibitor, potentiated this response. To further investigate the role of neuropeptides in acrolein-induced airway responses, a subsequent study with capsaicintreated guinea pigs exposed to acrolein was performed (Turner et al., 1993b). Capsaicin depletes neuropeptides from C-sensory fibers and resulted in 100% mortality (12 of 12 guinea pigs) within 24 h of two 7.5 h  1600 ppb exposures. This compared with only 14% mortality in guinea pigs exposed to acrolein alone. Pretreatment with capsaicin also exacerbated pulmonary inflammation and epithelial necrosis and denudation. Thus, acrolein activates airway C-fibers, which release neuropeptides and alters breathing. The resulting shallow breathing patterns may be protective by reducing deposition in the distal airways. Human studies have been performed with acetaldehyde. Myou et al. (1993) found that nine subjects with asthma had a dose-dependent decrease in FEV1 following 2-min inhalations of an aerosol of saline and acetaldehyde (0, 5, 10, 20, and 40 mg/mL) solution. In control subjects without asthma, acetaldehyde had no effect on lung function. Five of the subjects with asthma also had alcohol-induced bronchoconstriction, common among the Japanese population. As stated above, alcohol sensitivity is due to a dual inheritance of a rapid isoform of alcohol dehydrogenase with a slow isoform of aldehyde dehydrogenase-2. This allelic combination increases the formation of acetaldehyde and decreases subsequent clearance, and is carried by 20% of persons of Oriental lineage. It is further enriched among Japanese asthma patients, with over 50% of these individuals having alcohol intolerance (Gong et al., 1981; Watanabe, 1991). However, this factor did not explain the effect of inhaled acetaldehyde observed by Myou et al. (1993), because four of the nine asthma patients were not alcohol intolerant, yet they responded like those patients that were. Although the sample size of these populations is small, it is possible that persons with asthma may be at greater risk to the bronchoconstrictive effects of acetaldehyde. These results are like the immediate bronchoconstriction that has been noted after sulfur dioxide exposure (Sheppard et al., 1984). 9.2.2.3 Mucociliary Clearance and Defense Mechanisms Along with studies with formaldehyde, mucociliary function has been studied with other aldehydes. Unlike reported bronchoconstriction, Dalhamn and Rosengren (1971) found that the effects and dosimetry of formaldehyde were comparable to acrolein and crotonaldehyde in potency in the inhibition of cilia beating. Acetaldehyde was much less potent. Changes in ciliary beat can also be accompanied by alteration in mucus load. Borchers et al. (1998) found that acrolein exposure increases mucin (predominately MUC5ac) transcript and protein levels in rats. Furthermore, Borchers et al. (2008) found that CD8(þ) T cells are important mediators of macrophage accumulation in the lung and the progressive airspace enlargement in response to chronic acrolein exposures. The expression of several inflammatory cytokines (IP-10, IFN-gamma,

280

FORMALDEHYDE AND OTHER ALDEHYDES

IL-12, RANTES, and MCP-1), MMP2 and MMP9 gelatinase activity, and caspase3 immunoreactivity in pulmonary epithelial cells were attenuated in the Cd8-deficient mice compared to wild-type. Because acrolein has greater penetration of the upper respiratory tract than formaldehyde, effects of acrolein on other lung defense mechanisms has been examined. Exposure to 1000–2000 ppb acrolein significantly suppresses intrapulmonary killing in mice challenged with bacteria (Jakab, 1977; Astry and Jakab, 1983; Aranyi et al., 1986; Jakab, 1993), suggesting that host defense mechanisms of distal airway and alveolus are impaired. Macrophage activation and dysfunction induced by acrolein have also been observed in vitro (Leffingwell and Low, 1979; Grundfest et al., 1982; Sherwood et al., 1986; Jakab, 1993; Li et al., 1997). Thus, damage to the lower respiratory tract is much more likely after exposures to aldehydes other than formaldehyde, and some of these compounds (i.e., unsaturated aliphatic aldehydes) have equal or greater potency than formaldehyde. Even though acrolein can penetrate to the distal lung more than formaldehyde, the effects of acrolein (2500 ppb) can be enhanced by coexposure with carbon black particulate matter (10 mg/m3). Following exposure to each agent or coexposures of 4 h/day for 4 days, Jakab (1993) challenged Swiss strain mice with different infectious agents: Staphylococcus aureus to evaluate alveolar macrophage (AM) surveillance, Proteus mirabilis to evaluate AMs and polymorphonuclear leukocytes (PMNs) surveillance, Listeria monocytogenes to evaluate lymphokine-mediated cellular immunity, and influenza A virus for the cytotoxic T-cellmediated of cellular immunity. Only coexposures produced effects suppressed the intrapulmonary killing of S. aureus a day after exposure with a return to control levels by Day 7. In contrast, the coexposure enhanced the intrapulmonary killing of P. mirabilis possibly resulting from a significant increase in PMNs recovered in lung lavage fluid (also only noted following coexposure with infection). Combined exposure to carbon black and acrolein also impaired elimination of L. monocytogenes and influenza Avirus from the lungs. Exposure of alveolar macrophage to these concentrations directly suppressed Fc-receptor-mediated phagocytosis for up to 11 days (Jakab and Hemenway, 1993), which agrees with diminished S. aureus surveillance. However, the effect was short-lived in vivo. These studies suggest that the carbon black particle acts as a carrier for acrolein to enhance penetration to the distal airway and alveolar regions of the lung. The mechanisms of diminished innate immunity following acrolein exposure are not fully understood. Witz et al. (1985) noted the reactive aldehydes diminish superoxide anion production by neutrophils. In addition, a,b-unsaturated aldehydes have been found to alter NADH activity and macrophage and neutrophil membrane function, fluidity and sulfhydryl status (Witz et al., 1988). These studies were extended by Li et al. (1997) that demonstrated that acrolein increased cell death (necrosis) and programmed cell death (apoptosis) of macrophages. These effects were thought to be mediated by secondary intracellular formation of oxidants (Nardini et al., 2002; Misonou et al., 2006; Yousefipour et al., 2005) and activation of NFkB, and acrolein was also found to inhibit endotoxin-induced NFkB activation and decrease the basal level NFkB activity, which may be responsible for the inhibition of cytokine release and the induction of apoptosis in human alveolar macrophages (Li et al., 1999). The induction of apoptosis may be mediated by caspase-7 and -9 activation that induces cytochrome c release from the mitochondria (Tanel and Averill-Bates, 2005; Luo et al., 2005). In addition, acrolein and crotonaldehyde may diminish innate immunity by diminishing T-cell cytokine production, which can be reversed by the thiol compound, N-acetylcysteine (Lambert et al., 2005). Lastly, recent studies suggest that in vivo acrolein exposure of mice suppresses LPS-induced Th1 cytokine responses without affecting acute

EFFECTS OF MULTIPLE EXPOSURES

281

neutrophilia (Kasahara et al. 2008). This finding suggests that cytokine signaling can be disrupted by acrolein may represent a mechanism by which smoking contributes to chronic disease in chronic obstructive pulmonary disease and asthma. In contrast, acrolein diminishes apoptosis of neutrophils (Finkelstein et al., 2001). The inhibition of constitutive neutrophil apoptosis is mediated by common mechanisms, involving changes in cellular-reduced glutathione (GSH) status, resulting in reduced activation of initiator caspases as well as inactivation of caspase-3 by modification of its critical cysteine residue (Finkelstein et al., 2005). It is thought that diminished apoptosis in leukocytes at site of injury can lead to persistent inflammation.

9.3 EFFECTS OF MULTIPLE EXPOSURES 9.3.1

Formaldehyde

9.3.1.1 Carcinogenesis According to IARC (2004), formaldehyde is carcinogenic to humans (Group 1) on the basis of sufficient evidence in experimental animals and in human. This is a higher classification than the previous IARC evaluations of Group 2A––probably carcinogenic to humans, which were based on sufficient evidence in experimental animals, but insufficient evidence in humans (IARC, 1995a). In addition, in vitro and in vivo genotoxic studies support this classification. The major reason for the stricter classification was due to updated epidemiological studies that found that occupational formaldehyde exposure was associated with induction of nasopharyngeal cancer (Hildesheim et al., 2001; Hauptmann et al., 2004). Formaldehyde has genotoxic effects in a wide number of in vitro systems (Auerbach et al., 1977; Ma and Harris, 1988; WHO, 1989; Feron et al., 1991; IARC, 1995a). The level of activity in many experimental systems is often either weak or moderate compared to other known mutagens and can depend on what are sometimes unique experimental conditions. Regardless of these caveats, formaldehyde is clearly capable of reacting with cellular macromolecules (see above), and DNA damage induced by formaldehyde includes DNA– protein cross-links and single-strand breaks (Fornace, 1982; Fornace et al., 1982; Graftstrom et al., 1984; IARC, 1995a). DNA–protein cross-links have been found in vivo in the nasal passages of Fisher 344 rats and rhesus monkeys (Heck et al., 1989; Casanova et al., 1991). Graftstrom et al. (1984) also found that DNA single-strand breaks could accumulate in the presence of DNA excision repair inhibitors and that formaldehyde can inhibit repair. Studies examining genotoxic effects in bacteria, yeast, fungi, and Drosophila produce varied responses. For example, mutations in Drosophila are only noted in male larvae under culture conditions requiring adenosine, adenylic acid, or RNA medium supplementations. In mammalian systems the results are also subtle, complex, and varied. As noted by Boreiko and Ragan (1983), formaldehyde-induced sister-chromatid exchange in Chinese hamster ovary cells has been found by one investigator (Obe and Beck, 1979), but not by another (Brusick, 1983). In addition, formaldehyde is not mutagenic in either of these cells (Hsie et al., 1979). Another illustration of the complexity associated with the evaluation of formaldehyde’s genotoxic effects is that it can induce unscheduled DNA synthesis, which has been recorded in HeLa cells (Martin et al., 1978), but not in monkey kidney cells (Nocentini et al., 1980). In vivo dominant lethal assays in mice are negative or transiently positive (Epstein, 1972; Fontignie-Houbrechts, 1981). In any case, formaldehyde can facilitate malignant transformation of mouse embryo C3H10T1/2 C18 fibroblasts, acting

282

FORMALDEHYDE AND OTHER ALDEHYDES

through either an “initiating” or “promoting” mechanism (Ragan and Boreiko, 1981; Boreiko and Ragan, 1983; Frazelle et al., 1983), and produce DNA single-strand breaks and DNA–protein cross-links (Graftstrom et al., 1984). Mutational spectra induced by formaldehyde have been studied in human lymphoblasts in vitro, in Escherichia coli and in naked pSV2gpt plasmid DNA (Crosby et al., 1988). In lymphoblasts large visible deletions of some or all the X-linked hprt bands were detect by Southern blot analysis Liber et al., 1989). In E. coli, mutations in the xanthine guanine phosphoribosyl transferase (gpr) gene included large insertions (41%), large deletions (18%), and point mutations (41%). Many of the point mutations were GC transversions. Formaldehyde-induced genetic alterations in E. coli are concentration dependent, with higher doses yielding 92% point mutations, 62% of which were single AT transitions. Exposure of a SV2gpt plasmid DNA (with transformation into E. coli) resulted in frame shift mutations. Through a comparison of DNA repair proficient and deficient heterokaryons of Neurospora crassa, de Serres and Brockman (1999) examined formaldehyde induced specific-locus mutations at two closely linked loci in the adenine-3 (ad-3) region and inactivation of heterokaryotic conidia. As in many other systems, formaldehyde was a weak mutagen in the DNA repair proficient strain. However, in the DNA repair-deficient strain, formaldehyde caused about a 35-fold higher frequency of ad-3 mutations and pronounced inactivation of heterokaryotic conidia. In addition, formaldehyde induced a 5.4-fold higher frequency of ad3 mutations resulting from multilocus deletion mutation. As with X-ray-induced multilocus deletion mutations, the formaldehyde-induced ad-3 mutations could lead to deleterious heterozygous effects. Overall, the in vitro assays demonstrate the feasibility that formaldehyde could be mutagenic, but the results are complex and not always consistent. In contrast, studies with laboratory animals are clearer. Chronic inhalation of formaldehyde induces squamous cell carcinoma in the nasal cavity of rats (Swenberg et al., 1980; Albert et al., 1982; Kerns et al., 1983; Sellakumar et al., 1985; Feron et al., 1988; Woutersen et al., 1989). Exposure to 14,300 ppb formaldehyde (6 h/day  5 days/week  24 weeks) produced carcinomas in 103 of 240 (43%) Fischer 344 rats and in 2 of 240 (1%) C57BL/ 6  C3H/He F1 mice (Kerns et al., 1983; Morgan et al., 1986). (Please note the C57BL/6 mouse strain is a resistant to many lung carcinogens, and the F1 offspring of this strain is likely to be resistant). Similarly in a 28 month study (15,000 ppb formaldehyde  6 h/ day  5 days/week), Kamata et al. (1997) reported that male Fisher 344 rats developed nasal tumors that were macroscopically evident within 14 months, with 8 of 32 rats developing tumors (squamous cell papillomas and carcinomas) at 28 months. No nasal tumors were observed with lower concentrations (300 and 2000 ppb groups), but nasal epithelial cell hyperplasia, hyperkeratosis, squamous metaplasia, inflammatory cell infiltration, erosion, and edema was evident in all groups. To understand how formaldehyde exposure leads to cancer, Recio et al. (1992) examined 11 rat nasal squamous cell carcinomas and found point C or G mutations in regions II–Vof the tumor suppressor gene, p53, in five tumors. Proliferating cell nuclear antigen (PCNA) staining was similar in pattern and distribution to p53 immunoreactivity (Wolf et al., 1995). These mutations [particularly a CpG dinucleotide at rat codon 271 (codon 273 in humans)] were mutational hot spots that occur in many human cancers. Studies with different strains of rats suggest that a gene–environment interaction may be controlling the cancer response. For example, Sprague–Dawley rats also respond to 14,200 ppb formaldehyde, but the time to onset was delayed, and the total carcinoma incidence was less (38 squamous cell carcinoma (38%) and 10 polyps or papillomas in 100 rats) (Albert et al., 1982). Wistar rats yielded six tumors in 132 animals exposed to 20,000 ppb formaldehyde for 4–13 weeks and followed by observation to 126 weeks (Feron

EFFECTS OF MULTIPLE EXPOSURES

283

et al., 1988). Although the exposure period and concentration varied, these results suggest that strain differences exist in rats (Fisher 344  Sprague–Dawley > Wistar strain) and that rats are more sensitive than one mouse strain derived from a resistant cross. Future studies to identify the genetic loci controlling these responses would be very informative. The formaldehyde concentration–carcinoma response relationship is nonlinear in rats, and also varies among strains. In Fisher 344 rats, exposure to 5600 ppb formaldehyde resulted in two rats (of 240 exposed) with squamous carcinomas, and 2000 ppb produced no carcinomas (Kerns et al., 1983). Similarly, Woutersen et al. (1989) found squamous-cell carcinoma in 1 of 26 Wistar rats exposed to 10,000 ppb for 28 months. (However, 15 of 58 rats develop carcinomas when the nasal epithelium was injured before exposure.) Swenberg et al. (1983) ascribed this nonlinearity in response to be due to nonlinearity in the formaldehyde dosimetry to its macromolecular target site. This nonlinear dosimetry results in part from variability of the breathing pattern during exposure, which influences the inhaled dose. (As noted earlier, formaldehyde deposition in mice exposed to 14,000 ppb was equivalent to that in rats exposed to 5600 ppb. Consistent with this dosimetry estimate was the incidence of carcinomas: 2/240 in mice after 14,000 ppb and in 2/240 rats after 5600 ppb). In addition to altered breathing pattern, these investigators have ascribed the nonlinearity in the dose–response relationship to the nonlinearity of cytotoxicity and overloading of protective mechanisms (including metabolic biotransformation, mucociliary clearance, and DNA repair, all of which might be altered by multiple genetic variants) induced at higher formaldehyde concentrations. In Syrian golden hamsters, 10,000 ppb formaldehyde alone (5 h/day  5 days/week for life) did not produce nasal carcinomas (Dalbey, 1982). In the exposed group, however, the mortality was marked, with only 20 of 88 animals surviving 10 weeks of treatment. In these animals, nasal epithelial cell hyperplasia or metaplasia was observed. When hamsters were exposed to formaldehyde before exposure to another carcinogen, diethylnitrosamine, the incidence of respiratory carcinomas increased. Formaldehyde (30,000 ppb) exposures were for 48 h before subcutaneous injections with 0.5 mg diethylnitrosamine (each once a week for 10 weeks). About twice as many tracheal adenomas (per tumor-bearing hamster) were observed when compared to hamsters exposed to diethylnitrosamine alone. When formaldehyde was given after diethylnitrosamine, no increase in adenomas was observed. Formaldehyde clearly produces nasal carcinomas in rats, and although mice and hamsters may be less sensitive than rats, nasal tumors have been observed in mice, and nasal hyperand metaplasia has been observed in hamsters. The nasal epithelial disruption has been investigated in several species including mice, rats, hamsters, and monkeys (Chang et al., 1983; Rusch et al., 1983; Maronpot et al., 1986; Morgan et al., 1986; Al-Abbas et al., 1986; Zwart et al., 1988; Monticello et al., 1989; St. Clair et al., 1990; Monticello et al., 1991; Bhalla et al., 1991; Cassee et al., 1996b; Cohen Hubal et al., 1997). In each species tested, the anterior nasal epithelium portion is most affected, although some species differences exist in the regional distribution of epithelial lesions. Repeated exposures to nearly comparable concentrations (about 5000–6000 ppb) produced loss of cilia and various stages of hyperplasia and squamous metaplasia of pseudostratified columnar respiratory epithelium and leukocyte infiltration. In Fischer 344 rats, these changes were most severe in the maxilloturbinate, the lateral aspect of the nasoturbinate, and the wall of the lateral meatus (Morgan et al., 1986), whereas in the rhesus monkey changes were most severe in the middle turbinate. In both species, effects noted after a single exposure become more persistent, and the percentage of nasal surface area affected progressed, when exposures are extended. The Fisher 344 appears to be more sensitve to formaldehyde-induced nasal pathology than the Brown Norway strain of rat following a single exposure (Ohtsuka et al., 1997).

284

FORMALDEHYDE AND OTHER ALDEHYDES

Formaldehyde also increases nasal epithelial proliferation rate. In monkeys, effects were more persistent than those observed in rats. In rats, the proliferation rate returned to control values after 3–9 days at 1000–6000 ppb (Morgan et al., 1986; Cassee et al., 1996a, 1996b). In contrast, in monkeys the proliferation rates remained elevated longer and areas of effected epitheliumweremorewidespread.InaninvitrocorrelatedstudybyTyihaketal.(2001),10 mM formaldehyde caused extensive cell damage to human colon carcinoma (HT-29) and human endothelial (HUV-EC-C) cell cultures, 1.0 mM enhanced apoptosis and reduced mitosis, and 0.1 mMenhancedcell proliferationanddecreasedapoptoticinbothcelltypesbut moresointhe tumor cells. Changes in rat epithelium were mimicked by changes in cell cycle transcript expression (Hester et al., 2003). Increases were seen in cell proliferation in the nasopharynx, trachea, and carina of the lungs of monkeys, as compared to only in the anterior nasal cavity of rats (Monticello et al., 1989). Since monkeys, like humans, breathe through both the nose and the mouth, the involvement of the lower respiratory tract also may be possible in humans. Epidemiological studies have examined the cancer mortality among persons with occupational and residential exposure to formaldehyde. This topic remains controversial, and has been reviewed extensively elsewhere (Blair et al., 1990; Feron et al., 1991; Partanen, 1993; Tarone and McLaughlin, 1995; IARC, 1995a; Collins et al., 1997; Coggon et al., 2003; Heck and Casanova, 2004; Cole and Axten, 2004; IARC, 2004; Marsh and Youk, 2004; Collins and Lineker, 2004; Binetti et al., 2006). Based on animal toxicological data, most concern has been directed at carcinomas of the nasal cavity because formaldehyde is thought to be so reactive that the target tissue must directly be exposed to inhaled formaldehyde. Evidence supporting an association between formaldehyde and nasal cancer in humans was lacking for several years. While increased relative risk for nasal cancer had been associated with formaldehyde exposure in many case-control studies (Olsen and Asneas, 1986; Hayeset al., 1986; Roush et al., 1987; Vaughan et al., 1986a, 1986b; Luce et al., 1993; West et al., 1993; Armstrong et al., 2000; Vaughan et al., 2000; Hildesheim, 2001), there were concerns about consistency in this association because it was clearly not found in two others studies (Hernberg et al., 1983; Brinton et al., 1985). In addition, this association was not always been supported by comparable increases in the standard mortality ratio in several cohort studies for nasal cancer (Marsh, 1982; Acheson et al., 1984a, 1984b; Levine et al., 1984; Stroup et al., 1986; Blair et al., 1986; Hayes et al., 1990; Gardner et al., 1993; Marsh et al., 1994). However, a re-analysis formaldehyde exposed workers by Marsh et al. (2002), when combined with the analysis by Hauptmann et al. (2004), supported associations with cancers of the upper respiratory tract and nasal cavity with formaldehyde, importantly even after adjusting for 11 potential confounding substances (Hauptmann et al., 2004). In addition, this study was supported by several case-control that were consistent with this view (Hauptmann et al., 2005). Nonetheless, the controversy has not been resolved, and reanalyses of these data suggest that uncertainities still remain (Tarone and McLaughlin, 2005; Marsh et al., 2006). In contrast to nasal cancer, positive associations of formaldehyde exposure have been noted more frequently with nasopharyngeal cancer. This association has been noted in five of seven case-control (Olsen et al., 1984; Olsen and Asneas, 1986; Vaughan et al., 1986a, 1986b; Roush et al., 1987; West et al., 1993; Armstrong et al., 2000; Vaughan et al., 2000; Hildesheim et al., 2001) and two of five cohort (Blair et al., 1986; Hayes et al., 1990; Gardner et al., 1993; Marsh et al., 1994; Andjelkovich et al., 1995) epidemiological studies. Three meta-analyses (Blair et al., 1990; Partanen, 1993; Collins et al., 1997) of the later data suggest a small to moderate increase in relative risk (1.2 to 2.1). This relationship was further re-considered by Collins et al. (1997), who proposed that underreporting (when no nasopharyngeal cancer is observed in a number of studies of small sample size) influenced

EFFECTS OF MULTIPLE EXPOSURES

285

the outcome. They found that when studies of small sample size are added to the metaanalysis, the metarelative risk for cohort studies decreases to 1.0 (with 95% confidence interval of 0.5–1.8). It is also important to consider the small number of observed cases (less than 10 nasopharyngeal cancer in total) used in these analyses. Nonetheless, the re-analysis by Hauptmann et al. (2004) also supported associations between nasopharyngeal cancer and formaldehyde exposure (including adjustment for potential confounding substances), whereas Marsh et al. (2006) challenge these conclusions. None to slight increases in risks for other cancers have been noted and include respiratory (buccal cavity, oropharynx, pharynx, and lung) (Walrath and Fraumeri, 1983; Acheson et al., 1984b; Liebling et al., 1984; Sterling and Arundel, 1985; Stayner et al., 1986; Vaughan et al., 1986; Bertazzi et al., 1986; Blair et al., 1987) and nonrespiratory sites (brain, lymphatic and hematopoietic, prostate, and skin) (Harrington and Shannon, 1975; Walrath and Fraumeri, 1984; Blair et al., 1986; Hagmar et al., 1986; Stayner et al., 1988; Hauptmann et al., 2003; Pinkertonetal.,2004;HeckandCasanova,2004).However,most oftheincreasesincancerrisk at other sites lack correlation with duration or intensity of exposure. This has been explained in part by a strong self-selection effect because of irritant responses (healthy worker effect in occupational populations), difficulties in retrospective assessments of exposure, and potential confounding factors (e.g., wood dust exposure) (Acheson et al., 1967; Olsen et al., 1984; Sterling and Weinkam, 1988; Blair and Stewart, 1990; IARC, 1995a; Collins et al., 1997; Hauptmann et al., 2003; Pinkerton et al., 2004). For example, an increased risk of nonrespiratory cancer mortality has often been associated with formaldehyde exposure in embalmers, and these individuals have exposures to complex mixtures of materials. Some argue that formaldehyde alone is unlikely to be responsible for these systemic carcinomas, givenitsrapidmetabolismatsitesofentry.Nonetheless, thepossibleassociation withleukemia (Hauptmann et al., 2003) is supported by multiple reports of increased frequency of sister chromatid exchange in peripheral lymphocytes noted of formaldehyde exposed workers (Yager et al., 1986; Suruda et al., 1993; Ying et al., 1999; Shaham et al., 2002; Ye et al., 2005). Cytology of the nasal mucosa has also been examined in formaldehyde-exposed workers (Berke, 1987; Edling et al., 1988; Holmstro¨m, 1989a, 1989b; Boysen et al., 1990). Abnormalities observed include loss of cilia, goblet cell hyperplasia, squamous metaplasia, and mild dysplasia, but, as with cancer mortality, these changes do not exhibit dose–response relationships and are confounded by coexposure to particulate matter including wood dust. Increases in squamous metaplasia in individuals living in homes with urea–formaldehyde insulation have also been inconsistent with level of exposure (Broder et al., 1988; Broder et al., 1991). In contrast, Ballarin et al. (1992) found significant levels of epithelial abnormalities in 15 subjects exposed to urea–formaldehyde glue in a plywood factory. Exposed workers had higher frequencies of micronuclei and dysplasia of nasal epithelial cells and nasal inflammation (leukocyte infiltrates) when compared to age and sex matched control subjects. Formaldehyde exposure may lead to these effects through disruption of several cellular processes. DNA–protein cross-links can arrest DNA replication and lead to the induction of other genotoxic effects such as sister chromatid exchanges in proliferating cells (Merk and Speit, 1998). Incomplete repair of DNA–protein cross-links can lead to the formation of mutations (Barker et al., 2005), which can be detected as chromosomal aberrations and micronuclei, rather than gene mutations at specific loci. These errors, in turn, can lead to larger deletions and recombinations and thereby increase micronuclei frequency (Speit and Merk, 2002). As noted above, several studies suggest increased micronuclei frequency occur in the nasal or buccal mucosa cells of formaldehyde exposed workers (Ballarin et al., 1992; Burgaz et al., 2001; Ye et al., 2005). However, the effects are not consistent across studies

286

FORMALDEHYDE AND OTHER ALDEHYDES

(e.g., the threshold dose may vary greatly). Speit and Schmid (2006) have suggested that this may be due to the lack of standardization of micronucleus tests, the high assay variability, and the fact that effects can be focal, and that sampling may be incomplete. In summary, formaldehyde produces a complex array of genotoxic effects in in vitro systems and consistently has been found to induce squamous carcinoma in the nasal cavity of laboratory animals (at concentrations exceeding 5000 ppb) with correlated histopathological changes (at concentrations exceeding 500 ppb). Available epidemiological evidence is conflicting, but has recently been deemed to be sufficient to support the hypothesis that formaldehyde leads to an increased risk of cancer in exposed humans (IARC, 2004). Controversy exists as to whether formaldehyde is carcinogenic to humans (Group 1), but there is little controversy that formaldehyde is probably carcinogenic to humans (Group 2A). This is because the evidence of an increased risk of nasopharyngeal cancer in humans from epidemiological studies is suggestive but controversial, and evidence for nasal cancer is limited. Evidence for cancer at other sites (including lung) is equivocal, and hard to reconcile with exposure, although recent studies suggest that leukemia may be associated with exposure. Together these findings indicate that formaldehyde is at least a Group 2A suspected (or probable) human carcinogen (IARC, 1995a; ACGIH, 2007) and exposures should be restricted accordingly. 9.3.1.2 Other Responses to Multiple Exposures Although much attention has been place on the carcinogenic potential of formaldehyde, the irritant capacity of this compound is irrefutable. Repeated formaldehyde exposure can result in eye and upper respiratory tract irritation, declines in pulmonary function, and can initiate skin sensitization. Except for skin sensitization, these effects are often readily reversible with cessation of exposure and depend more on the exposure concentration (threshold dose) than on the exposure duration (cumulative dose). Such effects may be viewed as repeated immediate responses to acute exposure rather than persistent, irreversible dysfunction produced by cumulative degradation of defense mechanisms and initiation of compensatory processes. Eye, nose, and throat irritation are the most common complaint of individuals with occupational or residential formaldehyde exposures. In occupational studies, Alexandersson and Hedenstierna (1988) found that exposures to 300–500 ppb during a work shift led to symptoms. Likewise, Horvath et al. (1988) found that across-shift responses of sore throat and burning nose increased at concentrations
400 ppb. Control responses were 3–4%, and after exposure 8–15% responded. These responses followed a dose–response relationship with 22–36% and 33–55%, responding at <1000 and <3000 ppb, respectively. Although these exposures occurred in a population with repetitive exposure, the threshold dose or frequency of reported symptoms does not differ remarkably from that observed after a single exposure (Anderson and Molhave, 1983; Bender et al., 1983). Residential exposures also demonstrate similar results, with one difference: effects, particularly eye irritation, are noted with greater frequency at low concentrations. For example, whereas Horvath et al. (1988) found about 8–15% positive respondents after occupational exposure, Hanrahan et al. (1984) and Ritchie and Lehnen (1987) found 22–32% positive respondents for eye irritation after residential exposure to nearly equivalent concentrations (300 ppb). Ritchie and Lehnen (1987) also found nearly 90%, of persons exposed to 300 ppb reporting eye irritation. One reason for this difference may be that residential exposures are of longer duration (Fig. 9.1). As mentioned previously, Anderson and Molhave (1983) also found that reports of eye irritation increase, whereas nose and throat irritation complaints decrease, during extended exposure. Because these responses are subjective, this difference may also be explained by a greater willingness to tolerate symptoms at work than at

EFFECTS OF MULTIPLE EXPOSURES

287

home. The exact mechanism of eye irritation is not fully understood, but studies in rabbits suggest that topical formaldehyde can lead to release of secretoneurin, a 33-amino-acid chromogranis neuropeptide (Kralinger et al., 2003). Together these studies imply that irritant related responses will occur in humans at concentrations of 300 ppb. Pulmonary function after repeated formaldehyde exposure in occupational settings has also been examined. One type of study involves comparison of forced expiratory volumes before and after an 8-h work shift. Exposures in these studies ranged from 100 to 3000 ppb (typically averaging 500 ppb) and yielded slight declines that are associated with single exposures (Schoenberg and Mitchell, 1975: Gamble et al., 1976; Alexandersson et al., 1982; Kilburn et al., 1985; Horvath et al., 1988; Uba et al., 1989; Khamgaonkar and Fulare, 1991). Baseline values obtained Monday morning (after no exposures for 2 days) in each of these studies, however, typically was not different from control values, indicating that the crossshift declines are reversible and are unlikely to lead to a persistent pulmonary dysfunction. In one study by Alexandersson and Hedenstierna (1988), however, Monday-morning decrements in forced expiratory volume-1.0 s and forced vital capacity were noted, suggestive of a persistent change. However, these changes correlate neither with peak exposure nor with duration of employment and thus only offer limited support for the contention that persistent changes are a consequence of repetitive exposure. In each of the above cross-shift studies, small but measurable decreases in FEV1 were observed. In most cases, these changes resulted from exposure to threshold concentrations ’ 500 ppb. This threshold is in good agreement with the 300 ppb threshold for acute respiratory effects reported in guinea pigs by Amdur (1960). Individuals may vary in dose threshold, however. For example, Kim et al. (2001) presented a case report of an individual with occupational asthma. Working in an environment with a mean formaldehyde level of 60 ppb (with occasional peaks of 120–130 ppb), this worker had across-shift decreases in FEV1 that were reversed by inhaltion of a bronchodilatory (beta-2 adrenergic) agent. Thus, individual susceptibility may vary greatly, and occupational environments with irritant atmospheres are likely to be selective for healthy workers who tolerate exposures upon acclimiation. This type of complex effect may not always readily apparent in exposed populations. In numerous studies of healthy subjects, or persons with asthma (exposed in inhalation chambers) effects observed on respiratory function are difficult to demonstrate, even at concentrations of 3000 ppb. Possible explanations include the following:(a) occupational exposures involve a complex atmosphere that may or may not include particulate matter or other contaminants (Amdur, 1960; Kilburn and McKenzie, 1978; Jakab, 1993); (b) occupational exposure can be to transient peak levels that produce irritation, but are undetected by time weight averaging (Ryan et al., 2003); (c) the extended occupational exposures (8 h) produce greater effects than 10–120-min tests even at equivalent doses (concentration time) (Leikauf and Doupnik, 1989; Swiecichowski et al., 1993); or (d) repeated exposure lowers the threshold of responsiveness without producing chronic effects. Because symptomatic responses (e.g., eye irritation) occur at a greater frequency following residential exposures to 50–300 ppb formaldehyde, when compared to occupational exposures to equivalent concentrations, residential exposure to 50–300 ppb could also produce pulmonary effects. Inasmuch as the current occupational threshold limit value has been set at 300 ppb (ACGIH, 2007), a prudent environmental/indoor air level might be between 50 and 100 ppb, a value also recommended by Ritchie and Lehnen (1987) and more recently by Arts et al. (2008). This value may be difficult to achieve, however, unless indoor cigarette smoking is curtailed (see Indoor Air section). Epidemiological evidence supportive of a conservative environmental standard has been obtained by Krzyzanowski et al. (1990), who found an increased pulmonary morbidity

288

FORMALDEHYDE AND OTHER ALDEHYDES

(prevalence of asthma and bronchitis) among children living in homes with 60–120 ppb formaldehyde when compared to children living in homes with
40 ppb. Decrements in peak expiratory flow rates were also correlated with formaldehyde exposure. Similarly, a study of health outcomes for children that that seasonal differences in formaldehyde levels found that children exposed to formaldehyde levels of >50 ppb are at increased risk of having asthma (Rumchev et al., 2002). Confirmation of this effect has also been reported in an abstract by Czap et al. (1993). These investigators found that the prevalence of physiciandiagnosed asthma and asthma-related symptoms were about double for individuals living in home with formaldehyde levels averaging about 50 ppb as compared to controls that averaged about 5 ppb. Other investigators, including Norback et al. (1995) and Wieslander et al. (1997), also have reported formaldehyde exposure (mixed with other volatile organic compounds) is associated with an increase in prevalence of asthma, asthma-related symptoms, and blood eosinophil counts. In addition, sensitized persons with asthma develop bronchoconstriction to lower threshold dose of dust mite following short, low dose 30 ppb formaldehyde exposure (Casset et al., 2006b). In a review of the literature, Casse et al. (2006a) also noted that the risk for development of asthma is increased (approximately 1.4-fold) in homes that exceed levels of 50 ppb formaldehyde. Thus, these studies suggest that persistent respiratory effects can result from low level indoor formaldehyde exposures, and environmental exposures produce effects at concentration below those that produce observable effects in short-term clinical studies. Along with direct irritation, topical application of formalin (37% formaldehyde in methanol/water) can initiate allergic reactions including contact dermatitis or systemic responses (anaphylactic shock). Such reactions may result from formaldehyde usage in industrial processes including histological laboratories, dental procedures, or kidney dialysis (reviewed in Chapters 3–13 in Feinman, 1988, and by Bardana and Montanaro, 1991; Braun et al., 2003). The mechanisms of these immunologic reactions are still somewhat unclear, but it appears that formaldehyde reactions involve both immediate (antibody–antigen-mediated) and delayed (cell-mediated) hypersensitivity. In immediate hypersensitivity, formaldehyde acts as an incomplete antigen (hapten) through its covalent binding to constituent proteins in the skin or blood, forming new antigenic determinants (Horsfall, 1934). Antibodies to formaldehyde–hemolytic red blood cell membrane protein or formaldehyde–human serum albumin conjugates have been identified and include immunoglobin (Ig)E, IgG, and IgM subtypes (Sandler et al., 1979; Lynen et al., 1983; Maurice et al., 1986; Wilhelmsson and Holmstro¨m, 1987; Thrasher et al., 1987, 1988; Broughton and Thrasher, 1988; Patterson et al., 1989; Hilton et al., 1996; Braun et al., 2003). Concentrations of formaldehyde causing immediate or delayed skin hypersensitivity (30,000–55,000 ppb) in human volunteers are typically lower than those causing irritation (20 ppm) (Feinman, 1988). These changes in acquired immunity may also modulate innate immunity. For example, Go´rski et al. (1992) reported that neutrophils isolated from persons with formaldehyde contact sensitivity that had been exposed to 400 ppb formaldehyde for 2 h had higher responses (activation measured by chemiluminescence) than neutrophils isolated from control subjects. Similarly, Lyapina et al. (2004) found that the neutrophil respiratory burst activity was diminished in workers exposed to formaldehyde, especially among those with chronic mucosal inflammation. Modulation of cell-mediated immunity, on the other hand, is not as well documented (Pross et al., 1987; Thrasher et al., 1988; Patterson et al., 1989). It also remains unclear whether allergic reaction involving the lung can be initiated solely by inhalation exposure

EFFECTS OF MULTIPLE EXPOSURES

289

(Hendrick and Lane, 1975, 1977; Hendrick et al., 1982; Lee et al., 1984; Patterson et al., 1989; Bardana and Montanaro, 1991; Liteplo and Meek, 2003), but it does appear that, although rare, documented dual (immediate and delayed) pulmonary reactions can be evoked in persons with previous occupational exposure (Nordman, 1985; Burge et al., 1985; Kim et al., 2001). Although, formaldehyde is a weak sensitizing agent in allergic reactions, it may modulate immunity to other more common allergens. Enhancement of sensitization has been demonstrated in mice by Tarkowski and Go´rski (1995), who found higher IgE ovalbumin antibody titers following exposure to 1600 ppb formaldehyde for 10 days after presentation of antigen. Similarly, guinea pigs exposed to 130–250 ppb formaldehyde for 5 days before presentation of antigen developed higher IgG ovalbumin antibody levels (ELISA units) when compared to guinea pigs exposed to filtered air (Riedel et al., 1996). Likewise, repeated transnasal administration of formaldehyde also enhanced allergic bronchoconstriction and potentiated IgG production in guinea pigs (Kita et al., 2003). These changes in the Riedel et al. (1996) study were associated with greater epithelial pathology and eosinophilic infiltrates. Together, these investigations suggest that immunologically based bronchial hypersensitivity can develop following formaldehyde exposure alone, but that this response is rare (based on human studies of clear responses among a few individuals). More frequently, formaldehyde may enhance sensitization to other, more common respiratory antigens (based mainly on animal evidence). 9.3.2

Repeated Exposure to Other Aldehydes

9.3.2.1 Mutagenicity and Carcinogenicity of Other Aldehydes Like formaldehyde, several other aldehydes have been found to be genotoxic in microbial, insect, and mammalian systems (Graftstrom 1990; Feron et al., 1991; Graftstrom et al., 1994; WHO, 1995; Yang et al., 2002; Feng et al., 2006). Of the more common environmental aldehydes, acrolein, acetaldehyde, benzaldehyde, crotonaldehyde, furfural, and glutaraldehyde can be mutagenic in in vitro assay systems (Vegheli and Osztovics, 1978; Izard and Liberman, 1978; Obe and Beck, 1979; Hemminki et al., 1980; Marnett and Tuttle, 1980; Neudecker et al., 1981; Bird et al., 1982; Marnett et al., 1985; Cooper et al., 1987; Galloway et al., 1987; Reynolds et al., 1987; Hadi et al., 1989; He and Lambert, 1990; Graftstrom et al., 1994,Vaca et al., 1998; Feng et al., 2006). Acrolein, for example, can induce DNA single-strand breaks and DNA–protein crosslinks in human bronchial cells (Graftstrom et al., 1988) and in guanosine (Chung et al., 1984; Nath and Chung, 1994) and adenosine adducts (Kawai et al., 2003). The process of DNA– protein cross-links is likely to involve formation of acrolein–DNA adducts followed by a Schiff-base linkage to protein (Kurtz and Lloyd, 2003). Acrolein–DNA adducts have been detected in human tissues (Nath et al., 1996) and are increased in the oral mucosa of smokers as compared to nonsmokers (Nath et al., 1998). Similar to polycyclic aromatic hydrocarbons (PAHs)-DNA adducts, acrolein–DNA adducts induce predominantly G-to-T transversions in human cells. Feng et al. (2006) found that the acrolein–DNA binding pattern in tumor protein p53 (TP53) gene was similar to the p53 mutational pattern in human lung cancer. Acrolein preferentially binds at CpG sites, and this enhancement of binding is due to cytosine methylation at these sequences. In addition, acrolein reduces the DNA repair capacity for damage induced by benzo[a]pyrene diol epoxide. Based on a comparison of the concentrations and effects of acrolein to those of PAHs, these findings led the authors to suggest that acrolein may be a major etiological agent in cigarette smoke-related lung cancer, and that it

290

FORMALDEHYDE AND OTHER ALDEHYDES

contributes to lung carcinogenesis through two detrimental effects: DNA damage and inhibition of DNA repair. Acrolein also can be generated by endogenous lipid peroxidation (Esterbauer et al., 1982; Uchida et al., 1998; Noiri et al., 2002) and by myleoperoxidase degradation of amino acids (Anderson et al., 1997; Shao et al., 2005; Stevens and Maiers, 2007), the latter could link chronic inflammation with cancer as well as atherosclerosis. Acrolein also affects cell growth, membrane integrity, and differentiation in human bronchial epithelial cells. Responses may be mediated, in part, through acrolein’s action on cell thiol status (Graftstrom et al., 1988; Graftstrom 1990), cytoplasmic free Ca2þ (Trump et al., 1988), reduction of thioredoxin activity (Yang et al., 2004), or eicosanoid metabolism (Doupnik and Leikauf, 1990). However, even at concentrations well below those necessary to deplete glutathione or activate phospholipases, acrolein also activates metalloproteinases that cleave membrane ligands for the epidermal growth factor receptor and downstream mitogen-activated protein kinase signalling, and thereby lead to mucin (Deshmukh et al., 2005, 2008) and heme oxygenase 1 (Wu et al., 2006; Zhang and Forman, 2008) gene expression. The activation of the epidermal growth factor receptor-signaling pathway, which stimulates proliferative growth, is abnormally in many lung cancers. Thus, the augmentation of this pathway and the inhibition of tumor suppressor genes (like p53) could work in consort. When various aldehydes are compared in various geno- and cytotoxic assays, acrolein is often at least as potent as formaldehyde, and both compounds are markedly more potent than acetaldehyde. Interestingly, inhalation studies with acetaldehyde, but not acrolein, produce tumors in vivo. Like formaldehyde, acetaldehyde is a suspected human carcinogen because acetaldehyde can induce nasal adeoncarcinoma and squamous cell carcinoma in rats (Woutersen et al., 1986a; Woutersen and Feron, 1987) and laryngeal carcinoma in hamsters (Feron and Kruysse, 1977; Feron et al., 1982). The sites of tumors are somewhat different with acetaldehyde than with formaldehyde inasmuch as acetaldehyde (750,000 ppb) produced adenocarcinomas of the olfactory epithelium in rats (Woutersen et al., 1986a, 1986b). Histopathological correlates at lower acetaldehyde concentrations (400,000–1,000,000 ppb) include hyper- and metaplasia, principally in the nasal olfactory epithelium rather than the anterior respiratory epithelium (Appelman et al., 1982; Woutersen et al., 1984, 1986a, 1986b; Cassee et al., 1996b; Morris, 1997). In addition acetaldehyde shares an ability to enhance the tumorigenicity of benzo(a)pyrene in hamsters with formaldehyde (Feron, 1979). In contrast to the animal evidence for acetaldehyde, the animal inhalation studies with acrolein lack evidence for carcinogenesis (Feron et al., 1991; WHO, 1995). A study comparing formaldehyde with glutaraldehyde also may provide insights in to the comparative carcinogenic potential of aldehydes. Formaldehyde is cytotoxic and prduces rat nasal epithelial cell tumors after 12 months, whereas glutaraldehyde, while also cytotoxic, is not carcinogenic to nasal epithelium, even after 24 months. Both aldehydes induce similar acute and histopathology that is characterized by inflammation, hyperplasia, and squamous metaplasia. Hester et al. (2005) found differences among gene expression patterns in rat nasal tissue after instillation of glutaraldehyde and formaldehyde, and suggested that glutaraldehyde has a greater toxicity through diminishing DNA repair, and increasing mitochondrial damage and apoptosis. Histological findings noted following repeated acrolein exposure are similar and dissimilar to those developed after repeated formaldehyde inhalation. Like formaldehyde, but unlike acetaldehyde, the effects of 3000 ppb acrolein (6 h/day  5 days/week  3 weeks) in the nasal cavity are primarily limited to the respiratory epithelium and include degenerative change, squamous metaplasia, and neutrophilic inflammation (Feron et al., 1978; Leach

EFFECTS OF MULTIPLE EXPOSURES

291

et al., 1987; Monticello et al., 1991; Cassee et al., 1996a). In contrast to formaldehyde, acrolein can also produce effects in the mid- to distal rather than solely anterior portions of the rat nasal cavity (Leach et al., 1987) and can alter the epithelium of the lower respiratory tract (conductive airways and alveolar regions) (Lyon et al., 1970; Kutzman et al., 1985; Costa et al., 1986; Astry and Jakab, 1983; Borchers et al., 1998, 2008; Deshmukh et al., 2008). Another environmental and endogenously generated aldehyde, crotonaldehyde also is a mutagen and carcinogen (IARC, 1995b). In rats, crotonaldehyde induces liver tumors when administered in the drinking water (Chung et al., 1986), and crotonaldehyde–DNA adducts are readily detected in various tissues of untreated laboratory animals, as well as in human DNA (Chung et al., 1984; Stein et al., 2006; Zhang et al., 2006). Thus, crotonaldehyde–DNA adducts can apparently occur as a result of exogenous exposures and endogenous processes such as lipid peroxidation, and also have been proposed to contribute to age-related tumor incidence (Stein et al., 2006). Lastly, another aldehyde, malondialdehyde has also been found to be mutagenic in in vitro assays (Mukai and Goldstein, 1976; Shamberger et al., 1979; Yau, 1979; Bird et al., 1982; Basu and Marnett, 1983; Marnett et al., 1985) and carcinogenic to mice (Shamberger et al., 1974). Like acrolein, this naturally occurring compound, a 3-carbon dialdehyde (O¼CH-CH2CH¼O), is produced from auto-oxidation (peroxidation) of unsaturated fatty acids (Bernheim et al., 1948; Esterbauer et al., 1982), but unlike acrolein, malondialdehyde is also generated during synthesis of prostaglandins (Hamberg and Samuelsson, 1967; Marnett and Tuttle, 1980). Malondialdehyde levels in plasma (Gonenc et al., 2001) and malondialdehyde–DNA adducts in bronchial tissue (Munnia et al., 2006) increase in lung cancer patients. Other related compounds, b-ethoxyacrolein and b-methoxyacrolein, are 25–40 times more mutagenic than malondialdehyde. Inasmuch as these compounds are the result of common biogenic pathways, acrolein, crotonaldehyde, and malondialdehye and other possible aldehydes carry potential importance as mediators of spontaneous carcinogenesis, particularly with respect to background level age-related tumor incidence. 9.3.2.2 Other Response to Repeated Exposure The noncarcinogenic effects of repeated exposure to low-molecular-weight aldehydes other than formaldehyde are less studied. Extended exposure to certain aldehydes may lead to similar symptomatic (upper respiratory tract irritation) effects as formaldehyde (NRC, 1981; Feinman, 1988). These compounds may also produce respiratory effects (WHO, 1992, 1995). The chronic effects of acrolein and acetaldehyde are the most studied of the various aldehydes. Persistent decrements in pulmonary function in rats have been noted by Costa et al. (1986) after 4000 ppb acrolein (62 days  6 h/day  5 days/week). Airflow dysfunction was accompanied by focal peribronchial lesions and alterations of structural proteins (elastin). In tests with rats, guinea pigs, dogs, and monkeys, 700 ppb acrolein produced squamous metaplasia of the lungs in monkey (Lyon et al., 1970). This study, like those of Monticello et al. (1989) and Morris (1997), with formaldehyde suggests that obligatory nasal breathers (mouse, rats, and guinea pigs) are somewhat less responsive to chronic lower respiratory effects than oronasal breathers (dogs and monkeys). For example, Lyon et al. (1970) report histopathological effects of 220 ppb acrolein (24 h  90 days) in the lungs of dogs and monkey that were only apparent after 1000–1800 ppb exposure in rats and guinea pigs. Rats, hamsters, and rabbits were compared after exposures to 400,1400, and 4900 ppb acrolein (6 h/day  5 days/week  13 weeks) by Feron et al. (1978), who found rats to be more susceptible, responding at 400 ppb, whereas hamsters and rabbits were nonresponsive at this level.

292

FORMALDEHYDE AND OTHER ALDEHYDES

Low acrolein concentrations (200 ppb) may produce persistent changes in lung function observed at higher doses of formaldehyde (2000 ppb). Unsaturated aliphatic aldehydes (e.g., acrolein) are more chemically reactiveand irritating than formaldehyde. Animal studies clearly demonstrate that exposure to 300 ppb acrolein can initiate bronchial hyperreactivity, mediator release, and changes in pulmonary histology (Leikauf et al., 1989a, 1989b) and two exposures to 1600 ppb produced a persistent airway hyerreactivity that lasted up to 28 days in guinea pigs (Turner et al., 1993a). These observations parallel a human account of persistent respiratory effects following a single accidental exposure (Champeix et al., 1966). Acrolein and other compounds are less water soluble than formaldehyde and thus affect the lower, in addition to the upper, respiratory tract. Moreover, acrolein (formed endogenously) is increased in sputum or expired breath condensate from persons with chronic obstructive pulmonary disease or asthma (Corradi et al., 2004) in concentrations of over 100 nM, which are sufficient to lead to mucin formation in vitro (Deshmukh et al., 2005). High (250,000 ppb) acetaldehyde concentrations will diminish lung function about as much as >2000 ppb formaldehyde. Acetaldehyde produces nasal lesions in the rat at concentrations of 0.40 ppm (6 h/day  5 days/week  4 week) (Appelman et al., 1982), and in the hamster at concentration of 40 ppm (6 h/day  5 days/week  13 week) (Kruysse et al., 1975). The nasal area most damaged with acetaldehyde was the olfactory epithelium. In hamsters, tracheal epithelial metaplasia was noted following the 4.0 ppm and a lower 1.34 ppm exposure, suggesting that the nasal passages of the hamster are less sensitive than those of the rat. Chronic acetaldehyde exposure may produce persistent changes in lung function in that in Wistar rats exposed to 243,000 ppb acetaldehyde (8 h/day  5 days/ week  5 week) had altered functional residual capacity, residual volume, total lung capacity, and respiratory frequency (Saldiva et al., 1985). Less is currently known about the potential pulmonary effects of repeated exposures to other aldehydes. Longer-chain saturated aliphatic aldehydes (e.g., propionaldehyde) are less toxic, but cyclic aldehydes (e.g., benzaldehyde) have intermediate toxic potency when compared to formaldehyde. Inasmuch as a number of environmental exposures involve the cogeneration of these aldehydes with formaldehyde, need exists for more details on the extent of environmental exposure (through routine environmental sampling) and toxicological structure–activity relationships of these compounds. One aldehyde that may need further study is glutaraldehyde becaue occupational asthma has been reported following exposure (Chan-Yeung et al., 1993; Gannon et al., 1995; Ong et al., 2004). Lastly, a number of aldehydes, including acrolein, malondialdehyde, or hexanal, are naturally occurring or produced metabolically during oxidative metabolism, which is likely to be augmented by chronic inflammation when combined with exogenous exposure to another aldehyde. A better understanding of the molecular and cellular toxicology of these and other aldehydes is warranted, and may be useful to establish a relationship between human exposure and spontaneous carcinogenesis and perhaps even chronic obstructive pulmonary disease.

REFERENCES American Conference of Governmental Industrial Hygienist (ACGIH) (2007) TLVs and BEIs: Threshold Limit Values for Chemical Substances and Physical Agents Biological Exposure Indices. Cincinnati, OH:ACGIH. Acheson ED, Hadfield EH, Macbeth RG (1967) Carcinoma of the nasal cavity and accessory sinuses in woodworkers. Lancet 1:311–312.

REFERENCES

293

Acheson ED, Barnes HR, Gardner MJ, Osmond C, Pannett B, Taylor CP (1984a) Formaldehyde in the British chemical industry. Lancet 1:611–616. Acheson ED, Barnes HR, Gardner MJ, Pannett B, Taylor CP, Osmond C (1984b) Formaldehyde in process workers and lung cancer. Lancet 1:1066–1067. Adams DO, Hamilton TA, Lauer LD, Dean JH (1987) The effect of formaldehyde exposure upon the mononuclear phagocyte system of mice. Toxicol. Appl. Pharmacol. 88:165–174. Aharonson EF, Menkes H, Gatner G, Swift DL, Procter DF (1974) Effect of respiratory airflow on the removal of soluble vapors by the nose. J. Appl. Physiol. 27:654–657. Al-Abbas AH, Krajci D, Ibrahim KB (1986) The effect of formaldehyde vapour on the tracheal epithelium of albino rat. Acta Univ. Palacki Olomuc Fac. Med. 113:35–52. Albert RE, Sellakumar AR, Laskin S, Kuschner M, Nelson N, Snyder CA (1982) Gaseous formaldehyde and hydrogen chloride induction of nasal cancer in the rat. J. Natl. Cancer Inst. 68:597–603. Alexandersson R, Hedenstierna G (1988) Respiratory hazards associated with exposure to formaldehyde and solvents in acid-curing paints. Arch. Environ. Health 43:222–227. Alexandersson R, Kolmodin-Hedman B, Hedenstierna G (1982) Exposure to formaldehyde: Effects on pulmonary function. Arch. Environ. Health 37:279–284. Ali-Osman F, Akande O, Antoun G, Mao JX, Buolamwini J (1997) Molecular cloning, characterization, and expression in Escherichia coli of full-length cDNAs of three human glutathione S-transferase Pi gene variants. Evidence for differential catalytic activity of the encoded proteins. J. Biol. Chem. 272(15):10004–10012. Amdur MO (1960) The response of guinea pigs to inhalation of formaldehyde and formic acid alone and with a sodium chloride areosol. Int. J. Air Poll. 3:201–220. Anderson I, Molhave L (1983) Controlled human studies in formaldehyde. In: Gibson JE, editor. Formaldehyde Toxicity.Washington, DC: Hemisphere. Anderson MM, Hazen SL, Hsu FF, Heinecke JW (1997) Human neutrophils employ the myeloperoxidase-hydrogen peroxide-chloride system to convert hydroxy-amino acids into glycolaldehyde, 2-hydroxypropanal, and acrolein. A mechanism for the generation of highly reactive alphahydroxy and alpha, beta-unsaturated aldehydes by phagocytes at sites of inflammation. J. Clin. Invest. 99:424–432. Andjelkovich DA, Janzen DB, Brown MH, Richardson RB, Miller FJ (1995) Mortality of iron foundry workers. IV. Analysis of a subcohort exposed to formaldehyde. J. Occup. Med. 37:826–837. Appelman LM, Woutersen RA, Feron VJ (1982) Inhalation toxicity of acetaldehyde in rats. I. Acute and sub-acute studies. Toxicology 23:293–307. Aranyi C, O’Shea WJ, Graham JA, Miller FJ (1986) The effects of inhalation of organic chemical air contaminants on murine lung host defenses. Fund. Appl. Toxicol. 6:713–720. Armstrong RW, Imrey PB, Lye MS, Armstrong MJ, Yu MC, Sani S (2000) Nasopharyngeal carcinoma in Malaysian Chinese: Occupational exposures to particles, formaldehyde and heat. Int. J. Epidemiol. 29(6):991–998. Arts JH, Rennen MA, de Heer C (2006) Inhaled formaldehyde: Evaluation of sensory irritation in relation to carcinogenicity. Regul. Toxicol. Pharmacol. 44(2):144–160. Arts JH, Muijser H, Kuper CF, Woutersen RA (2008) Setting an indoor air exposure limit for formaldehyde: Factors of concern. Regul. Toxicol. Pharmacol. Nov; 52(2):189-94. Epub 2008 Aug 26. Astry CL, Jakab GJ (1983) The effects of acrolein exposure on pulmonary antibacterial defenses. Toxicol. Appl. Pharmacol. 67:49–54. ATSDR Agency for Toxic Substances and Disease Registry (ATSDR).(2005) Toxicological profile for acrolein (draft for public comment).Atlanta, GA:U.S. Department of Health and Human Services, Public Health Service. Auerbach C, Moutschen-Dahman M, Moutschen J (1977) Genetic and cytogenetic effects of formaldehyde and related compounds. Mutat. Res. 39:317–362.

294

FORMALDEHYDE AND OTHER ALDEHYDES

Ayer HE, Yeager DW (1976) Irritants in cigarette smoke plumes. Am. J. Public Health 72:1283–1285. Badre R, Guillerme R, Abram N, Bourdin M, Dumas C (1978) Pollution atmospherique par la fume’e de tabac. Ann. Pharm. Fr. 36:443–452. Ballarin C, Sarto F, Giacomelli L, Bartolucci Battista LG, Clonfero E (1992) Micronucleated cells in nasal mucosa of formaldehyde-exposed workers. Mutat. Res. 280:1–7. Bardana Jr EJ, Montanaro A (1991) Formaldehyde: An analysis of its respiratory, cutaneous, and immunologic effects. Ann. Allergy 66:441–452. Barker M, Weinfeld D, Murray D (2005) DNA–protein crosslinks: their induction, repair, and biological consequences. Mutat. Res. 589:111–135. Barnes PJ, Rodger IW, Thomson NC (1989) Asthma: Basic Mechanisms and Clinical Management. London:Academic Press. Barrow RE, Wang CZ, Cox RA, Evans MJ (1992) Cellular sequence of tracheal repair in sheep after smoke inhalation injury. Lung. 170(6):331–338. Barrow CS, Steinhagen WH, Chang JCF (1983) Formaldehyde sensory irritation. In: Gibson JE, editor. Formaldehyde Toxicity.Washington, DC: Hemisphere. Basu AK, Marnett LJ (1983) Unequivocal demonstration that malondialdehyde is a mutagen. Carcinogenesis 4:331–333. Beauchamp Jr RQ, Morgan KT, Kligerman AD, Anjelkovich DA, Heck HA (1985) A critical review of the literature on acrolein toxicity. CRC Crit. Rev. Toxicol. 14:309–380. Bender J (2002) The use of noncancer endpoints as a basis for establishing a reference concentration for formaldehyde. Regul. Toxicol. Pharmacol. 35(1):23–31. Bender JR, Mullin LS, Graepel GJ, Wilson WE (1983) Eye irritation response of humans to formaldehyde. Am. Ind. Hyg. Assoc. J. 44:463–465. Benedetti A, Comport MI,Esterbauer H (1980) Identification of 4-hydroxynonenal as a cytotoxic product originating from the peroxidation of liver microsomal lipids. Biochim. Biophys. Acta 620:281–296. Benedetti AM, Comport I, Fulceri R, Esterbauer H (1984) Cytotoxic aldehydes originating from the peroxidation of liver microsomal lipids. Biochim. Biophys. Acta 792:172–181. Ben-Jebria A, Marthan R, Rossetti M, Savineau JP, Ultman JS (1994) Human bronchial smooth muscle responsiveness after invitro exposure to acrolein. Am. J. Respir. Crit. Care Med. 149:(2 Pt 1):382–386. Ben-Jebria A, Crozet Y, Eskew ML, Rudeen BL, Ultman JS (1995) Acrolein-induced smooth muscle hyperresponsiveness and eicosanoid release in excised ferret trachea. Toxicol. Appl. Pharmacol. 135:35–44. Berglund B, Nordin S (1992) Detectability and preceived intensity for formaldehyde for smokers and non-smokers. Chem. Senses 17:291–306. Berhane K, Mannervik B (1990) Inactivation of the genotoxic aldehyde acrolein by human glutathione transferases of classes alpha, mu, and pi. Mol. Pharmacol. 37(2):251–254. Berhane K, Widersten M, Engstrom A, Kozarich JW, Mannervik B (1994) Detoxication of base propenals and other alpha, beta-unsaturated aldehyde products of radical reactions and lipid peroxidation by human glutathione transferases. Proc. Natl. Acad. Sci. U.S.A. 91(4):1480–1484. Berke JH (1987) Cytologic examination of the nasal mucosa in formaldehyde-exposed workers. J. Occup. Med. 29:681–684. Bernheim F, Bernheim MLC, Wilbur KC (1948) The reaction between thiobarbituric acid and the oxidation products of certain lipids. J. Biol. Chem. 174:257–264. Bernstein RS, Stayner LT, Elliott LJ, Kimbrough R, Falk H, Blade L (1984) Inhalation exposure to formaldehyde: An overview of its toxicology, epidemiology, monitoring, and control. Am. Ind. Hyg. Assoc. J. 45:778–785. Bertazzi BA, Pesatori AC, Radice L, Zocchetti C, Vai T (1986) Exposure to formaldehyde and cancer mortality in a cohort of workers producing resins. Scand. J. Work Environ. Health 12:461–468.

REFERENCES

295

Bhalla DK, Mahavni V, Nguyen T, McClure T (1991) Effects of acute exposure to formaldehyde on surface morphology of nasal epithelia in rats. J. Toxicol. Environ. Health. 33:171–188. Bhatt HS, Lober SB, Combes B (1988) Effect of glutathione depletion on aminopyrine and formaldehyde metabolism. Biochem. Pharmacol. 37:1581–1589. Binetti R, Costamagna FM, Marcello I (2006) Development of carcinogenicity classifications and evaluations: the case of formaldehyde. Ann. Inst. Super Sanita. 42(2):132–143. BirdRP,DraperHH,BasrurPK(1982)Effectofmalonaldehydeandacetaldehydeonculturedmammalian cells. Production of micronuclei and chromosomal aberrations. Mutat. Res. 101:237–246. Blair A, Stewart PA (1990) Correlation between different measures of occupational exposure to formaldehyde. Am. J. Epidemiol. 131:510–516. Blair A, Stewart PA, Hoover RN (1990) Mortality form lung cancer among workers employed in formaldehyde industries. Am. J. Ind. Med. 17:683–699. Blair A, Stewart P, O’Berg M, Gaffey W, Walrath J, Ward J, Bales R, Kaplan S, Cubit D (1986) Mortality among industrial workers exposed to formaldehyde exposure. J. Natl. Cancer Inst. 76:1071–1084. Blair A, Stewart P, O’Berg M, Gaffey W, Walrath J, Ward J, Bales R, Kaplan S, Cubit D (1987) Cancers of the nasopharynx and oropharynx and formaldehyde. J. Natl. Cancer Inst. 78:191–192. Bogdanffy MS, Morgan PH, Starr TB, Morgan KT (1987) Binding of formaldehyde to human and rat nasal mucus and bovine serum albumin. Toxicol. Lett. 38:145–154. Bohren KM, Bullock B, Wermuth B, Gabbay KH (1989) The aldo-keto reductase superfamily. J. Biol. Chem. 264:9547–9551. Borchers MT, Wert SE, Leikauf GD (1998) Acrolein-induced MUC5ac expression in rat airways. Am. J. Physiol. 274:L573–L581 Borchers MT, Wesselkamper SC, Harris NL, Deshmukh H, Beckman E, Vitucci M, Tichelaar JW, Leikauf GD (2007) CD8+ T cells contribute to macrophage accumulation and airspace enlargement following repeated irritant exposure. Exp. Mol. Pathol. 83(3):301–310. Boreiko CJ, Ragan DL (1983) Formaldehyde effects in the C3H/10T-1/2 cells transformation assay. In:Gibson JE, editor. Formaldehyde Toxicity.Washington:Hemisphere. Bourne HG, Seferian S (1959) Formaldehyde in wrinkle-proof apparel produces tears for milady. Ind. Med. Surg. 28:232–233. Boushey HA, Holtzman MJ, Sheller JR, Nadel JA (1980) Bronchial hyperreactivity. Am. Rev. Respir. Dis. 121:389–413. Boysen M, Zadig E, Digernes V, Abeler V, Reith A (1990) Nasal mucosa in workers exposed to formaldehyde: a pilot study. Br. J. Ind. Med. 47:116–121. Braun JJ, Zana H, Purohit A, Valfrey J, Scherer P, Haikel Y, de Blay F, Pauli G (2003) Anaphylactic reactions to formaldehyde in root canal sealant after endodontic treatment: four cases of anaphylactic shock and three of generalized urticaria. Allergy 58(11):1210–1215. Brinton LA, Blot WJ, Fraumeri JF (1985) Nasal cancer in the textile and clothing industries. Br. J. Ind. Med. 42:469–474. Broder I, Corey P, Brasher P, Lipa M, Cole P (1991) Formaldehyde exposure and health status in households. Environ. Health Perspect. 95:101–104. Broder I, Corey P, Cole P, Lipa M, Mintz S, Nethercott JR (1988) Comparison of health of occupants and characteristics of houses among homes and homes insulated with urea formaldehyde foam. II. Initial health and house variable and exposure-response relationships. Environ. Res. 45:156–178. Brooks PJ, Theruvathu JA (2005) DNA adducts from acetaldehyde: implications for alcohol-related carcinogenesis. Alcohol. 35(3):187–193. Broughton A, Thrasher JD (1988) Antibodies and altered cell mediated immunity in formaldehyde exposed humans. Comments Toxicol. 2:155–174.

296

FORMALDEHYDE AND OTHER ALDEHYDES

Bruinen de Bruin Y, Koistinen K, Kephalopoulos S, Geiss O, Tirendi S, Kotzias D (2008) Characterisation of urban inhalation exposures to benzene, formaldehyde and acetaldehyde in the European Union: comparison of measured and modelled exposure data. Environ. Sci. Pollut. Res. Inst.15 (5):417–430. Brusick DJ (1983) Genetic and transforming activity of formaldehyde. In: Gibson JE, editor. Formaldehyde Toxicity.Washington: Hemisphere. Bunting KD, Townsend AJ (1998) Dependence of aldehyde dehydrogenase-mediated oxaphosphorine resistance on soluble thiols. Importance of thiol interactions with the secondary metabolite acrolein. Biochem. Pharmacol. 56:31–39. Burgaz S, Cakmak G, Erdem O, Yilmaz M, Karakaya AE (2001) Micronuclei frequencies in exfoliated nasal mucosa cells from pathology and anatomy laboratory workers exposed to formaldehyde. Neoplasma 48(2):144–147. Burge PS, Harries MG, Lam WK, O’Brien IM, Patchett PA (1985) Occupational asthma due to formaldehyde. Thorax 40:255–260. Card SE, Tompkins SF, Brien JF (1989) Ontogeny of the activity of alcohol dehydrogenase and aldehyde dehydrogenases in the liver and placenta of the guinea pig. Biochem. Pharmacol. 38:2535–2541. Carson S, Goldhamer R, Carpenter R (1966) Mucus transport in the respiratory tract. Am. Rev. Respir. Dis. 93:86–92. Carter WPL (1990) Atmos. Environ. 24A:481–518. Casanova M, Heck Hd’A (1987) Further studies of the metabolic incorporation and covalent binding of inhaled [3 H]- and [14 C]-formaldehyde in F-344 rats: Effects of glutathione depletion. Toxicol. Appl. Pharmacol. 89:105–121. Casanova M, Morgan KT, Gross EA, Moss OR, Heck HA (1994) DNA-protein cross-links and cell replication at specific sites in the nose of F344 rats exposed subchronically to formaldehyde. Fundam. Appl. Toxicol. 23(4):525–536. Casanova M, Morgan KT, Steinhagen WH, Everitt JI, Popp JA, Heck Hd’A (1991) Covalent binding of inhaled formaldehyde to DNA in the respiratory tract of rhesus monkeys: Pharmacokinetics, rat-tomonkey interspecies scaling, and extrapolation to man. Fundam. Appl. Toxicol. 17:409–428. Casanova-Schmitz M, David RM, Heck Hd’A (1984) Oxidation of formaldehyde and acetaldehyde by NADþ-dependent dehydrogenases in rat mucosal homogenates. Toxicol. Appl. Pharmacol. 33:1137–1142. Cassee FR, Arts JHE, Groten JP, Feron VJ (1996a) Sensory irritation to mixtures of formaldehyde, acrolein, and acetaldehyde in rats. Arch. Toxicol. 70:329–337. Cassee FR, Groten JP, Feron VJ (1996b) Changes in the nasal epithelium of rats exposed by inhalation to mixtures of formaldehyde, acetaldehyde, and acrolein. Fundam. Appl. Toxicol. 29:208–218. Casset A, Purohit A, Marchand C, Le Calve S, Donnay C, Uring-Lambert B, Bahram S, Pauli G, de Blay F (2006a) The bronchial response to inhaled formaldehyde. Rev. Mal. Respir. 23:(1 Suppl.):3S25–3S34. Casset A, Marchand C, Purohit A, Le Calve S, Uring-Lambert B, Donnay C, Meyer P, de Blay F (2006b) Inhaled formaldehyde exposure: effect on bronchial response to mite allergen in sensitized asthma patients. Allergy 61:1344–1350. Champeix J, Courtial L, Perche E, Catilina P (1966) Broncho-pneumopathie aigue par vapeurs d’acroline. Arch. Mal. Prof. 27:794–796. Chang JCF, Steinhagen WH, Barrow CS (1981) Effect of single or repeated formaldehyde exposure on minute volume of B6C3F1 mice and F-344 rats. Toxicol. Appl. Pharmacol. 61:451–459. Chang JCF, Gross EA, Swenberg JA, Barrow CS (1983) Nasal cavity deposition, histopathology, and cell proliferation after single and repeated formaldehyde exposures in B6C3F1 mice and F-344 rats. Toxiol. Appl. Pharmacol. 68:161–176.

REFERENCES

297

Chan-Yeung M, McMurren T, Catonio-Begley F, Lam S (1993) Occupational asthma in a technologist exposed to glutaraldehyde. J. Allergy Clin. Immunol. 91(5):974–978. Chapin Jr FS (1974) Human Activity Patterns in the City. New York:Wiley-Interscience. Chaw UFM, Crane LE, Lange P, Shapiro R (1980) Isolation and identification of crosslinks from formaldehyde treated nuclei acids. Biochemistry 19:5525–5531. Chung FL, Tanaka T, Hecht SS (1986) Induction of liver tumors in F344 rats by crotonaldehyde. Cancer Res. 46(3):1285–1289. Chung FL, Young R, Hecht SS (1984) Formation of cyclic l, N2-propanodeoxyguanosine adducts in DNA upon reaction with acrolein or crotonaldehyde. Cancer Res. 44:990–995. Clarisse B, Laurent AM, Seta N, Le Moullec Y, El Hasnaoui A, Momas I (2003) Indoor aldehydes: Measurement of concentration levels and identification of their determinants in Paris dwellings. Environ. Res. 92:388–395. Coggon D, Harris EC, Poole J, Palmer KT (2003) Extended follow-up of a cohort of British chemical workers exposed to formaldehyde. J. Natl. Cancer Inst. 95(21):1608–1615. Cohen Hubal EA, Schlosser PM, Conolly RB, Kimbell JS (1997) Comparison of inhaled formaldehyde dosimetry predictions with DNA-protein cross-link in the rat nasal passages. Toxicol. Appl. Pharmacol. 143:47–55. Cole P, Axten C (2004) Formaldehyde and leukemia: an improbable causal relationship. Regul. Toxicol. Pharmacol. 40(2):107–112. Collins JJ (2004) Formaldehyde exposure and leukemia. Occup. Environ. Med. 61:875–876. Collins JJ, Lineker GA (2004) A review and meta-analysis of formaldehyde exposure and leukemia. Regul. Toxicol. Pharmacol. 40(2):81–91. Collins JJ, Acquivella JF, Esmen NA (1997) An updated meta-analysis of formaldehyde exposure and upper respiratory tract cancers. J. Occup. Med. 39:639–651. Comey AM, Hahn DA (1921) Dictionary of Chemical Solubilities: Inorganic.New York:Macmillan. p 158, 635, 900. Conolly RB, Kimbell JS, Janszen D, Schlosser PM, Kalisak D, Preston J, Miller FJ (2003) Biologically motivated computational modeling of formaldehyde carcinogenicity in the F344 rat. Toxicol. Sci. 75(2):432–447. Conolly RB, Kimbell JS, Janszen D, Schlosser PM, Kalisak D, Preston J, Miller FJ (2004) Human respiratory tract cancer risks of inhaled formaldehyde: dose-response predictions derived from biologically-motivated computational modeling of a combined rodent and human dataset. Toxicol. Sci. 82(1):279–296. Consensus Workshop on Formaldehyde. (1984) Deliberations. Environ. Health Perspect. 58:323–328. Cooper KO, Witz G, Witmer CM (1987) Mutagenicity and toxicity studies of several a,b-unsaturated aldehydes in Salmonella typhimurium mutagenicity assay. Environ. Mol. Mutagen. 9:289–295. Corradi M, Pignatti P, Manini P, Andreoli R, Goldoni M, Poppa M, Moscato G, Balbi B, Mutti A (2004) Comparison between exhaled and sputum oxidative stress biomarkers in chronic airway inflammation. Eur. Respir. J. 24(6):1011–1017. Costa DL, Kutzman RS, Lehmann JR, Drew RT (1986) Altered lung function and structure in the rat after sub chronic exposure to acrolein. Am. Rev. Respir. Dis. 133:286–291. Council on Scientific Affairs.(1989) Formaldehyde. JAMA 261:1183–1187. Crabb DW (1990) Biological markers for increased risk of alcoholism and for quantitation of alcohol consumption. J. Clin. Invest. 85:311–315. Cralley LV (1942) The effect of irritant gases upon the rate of ciliary activity. J. Ind. Hyg. Toxicol. 24:193–198. Crosby RM, Richardson KK, Kraft TR, Benforado KB, Liber HL, Skopek TR (1988) Molecular analysis of formaldehyde-induced mutations in human lymphoblasts and E.coli. Environ. Mol. Mutagen. 12:155–166.

298

FORMALDEHYDE AND OTHER ALDEHYDES

Czap C, Nicolai T, Wolf R, Von Mutius E (1993) Asthma, bronchial hyperreactivity and formaldehyde exposure in children living near a chip-board factory. Eur. Respir. J. 6:235s Dalbey WE (1982) Formaldehyde and tumors in hamster respiratory tract. Toxicology 24:9–15. Dalhamn T, Rosengren A (1971) Effect of different aldehydes on tracheal mucosa. Arch. Otolanngol. Head Neck Surg. 93:496–500. Dallas CE, Thetas JC, Harrist RB, Fairchild EJ (1985) Effect of subchronic formaldehyde inhalation on minute volume and nasal deposition in Sprague-Dawley rats. J. Toxicol. Environ. Health 16:553–564. Dasgupta PK, Li J, Zhang G, Luke WT, McClenny WA, Stutz J, Fried A (2005) Summertime ambient formaldehyde in five U.S. metropolitan areas: Nashville, Atlanta, Houston, Philadelphia, and Tampa. Environ. Sci. Technol. 39(13):4767–4783. Day JH, Lees REM, Clark RH, Pattee PL (1984) Respiratory response to formaldehyde and off-gas of urea formaldehyde foam insulation. Can. Med. Assoc. J. 131:1061–1065. de Serres FJ, Brockman HE (1999) Comparison of the spectra of genetic damage in formaldehydeinduced ad-3 mutations between DNA repair-proficient and -deficient heterokaryons of Neurospora crassa. Mutat. Res. 437(2):151–163. DeMaster EG, Stevens JM (1988) Acute effects of the aldehyde dehydrogenase inhibitors, disulfiram, pargyline and cyanamide, on circulating ketone body levels in the rat. Biochem. Pharmacol. 37:229–234. Deshmukh HS, Case LM, Wesselkamper SC, Borchers MT, Martin LD, Shertzer HG, Nadel JA, Leikauf GD (2005) Metalloproteinases mediate mucin 5AC expression by epidermal growth factor receptor activation. Am. J. Respir. Crit. Care Med. 171(4):305–314. Deshmukh HS, Shaver C, Case LM, Dietsch M, Wesselkamper SC, Hardie WD, Korfhagen TR, Corradi M, Nadel JA, Borchers MT, Leikauf GD (2008) Acrolein-activated matrix metalloproteinase 9 contributes to persistent mucin production. Am. J. Respir. Cell Mol. Biol. 38(4):446–454. Destaillats H, Lunden MM, Singer BC, Coleman BK, Hodgson AT, Weschler CJ, Nazaroff WW (2006) Indoor secondary pollutants from household product emissions in the presence of ozone: A benchscale chamber study. Environ. Sci. Technol. 40(14):4421–4428. Destaillats H, Spaulding RS, Charles MJ (2002) Ambient air measurement of acrolein and other carbonyls at the Oakland-San Francisco Bay Bridge toll plaza. Environ. Sci. Technol. 36(10):2227– 2235. Doupnik CA, Leikauf GD (1990) Acrolein stimulates eicosanoid release from bovine airway epithelial cells. Am. J. Physiol. 259:222–229 Edling C, Hellquist H, Odkvist L (1988) Occupational exposure to formaldehyde and histopathological changes in the nasal mucosa. Br. J. Ind. Med. 45:761–765. Egle Jr L (1972) Retention of inhaled formaldehyde, propionaldehyde, and acrolein in the dog. Arch. Environ. Health 25:119–124. Ellis EM, Hayes JD (1995) Substrate specificity of an aflatoxin-metabolizing aldehyde reductase. Biochem. J. 312: Pt 2:535–541. Engeland K, Hoog J-O, Holmquist B, Estonius M, Jornvall H, Vallee BL (1993) Mutation of arg-115 of human class III alcohol dehydrogenase: A binding site required for formaldehyde dehydrogenase activity and fatty acid activation. Proc. Nat. Acad. Sci. 90:2491–2494. Epstein SS (1972) Detection of chemical mutagens by dominant lethal assay in the mouse. Toxicol. Appl. Pharmacol. 23:288–298. Esterbauer H, Cheesman KH, Dianzani MU, Poli G, Slater TF (1982) Separation and characterization of aldehydic products of lipid per oxidation stimulated by ADP-Feþ2 in rat liver microsomes. Biochem. J. 208:129–140. Falk HL (1963) The response of mucus-secreting epithelium and mucus to irritants. Ann. N. Y. Acad. Sci. 106:583–608.

REFERENCES

299

Farres J, Guan K-L, Weiner H (1989) Primary structures of rat and bovine liver mitochondrial aldehyde dehydrogenases deduced from cDNA sequences. Eur. J. Biochem. 180:67–74. Feinman SE (1988) Formaldehyde Sensitivity and Toxicity. Boca Raton, FL: CRC Press. Feldman MY (1973) Reactions of nuclei acids and nucleoproteins with formaldehyde. Prog. Nucleic Acid Res. Mol. Biol. 13:1–19. Feng Z, Hu W, Hu Y, Tang MS (2006) Acrolein is a major cigarette-related lung cancer agent: Preferential binding at p53 mutational hotspots and inhibition of DNA repair. Proc. Natl. Acad. Sci. U.S.A. 103(42):15404–15409. Fennel TR (1990) Development of methods for measuring formaldehyde exposure. In Proceedings Seventh Annual Health Effects Institute Conference. Cambridge, MA: Health Effects Institute. Feron VJ (1979) Effects of exposure to acetaldehyde in Syrian hamsters simultaneously treated with benzo(a)pyrene or diethylnitrosamine. Prog. Exp. Tumor Res. 24:162–173. Feron VJ, Kruysse A (1977) Effects of exposure to acrolein vapor in hamsters simultaneously treated with benzo(a)pyrene or diethylnitrosamine. J. Toxicol. Environ. Health 3:379–394. Feron VJ, Kruysse A, Woutersen RA (1982) Respiratory tract tumours in hamsters exposed to acetaldehyde vapour alone or simultaneously to benzo(a)pyrene or diethylnitrosamine. Eur. J. Cancer Clin. Oncol. 18:13–22. Feron VJ, Kruysse A, Til HP, Immel HR (1978) Repeated exposure to acrolein vapor: Subacute studies in hamsters, rats, and rabbits. Toxicology 9:47–57. Feron VJ, Brruyntjes JP, Woutersen RA, Immel HR, Appelman LM (1988) Nasal tumours in rats after short-term exposure to cytotoxic concentrations of formaldehyde. Cancer Lett. 39:101–111. Feron VJ, Til HP, DeVrijer F, Wostersen RA, Cassee FR, Van Bladeren PJ (1991) Aldehydes: occurrence, carcinogenic potential, mechanism of action and risk assessment. Mutat. Res. 259:363–385. Finkelstein EI, Nardini M, van der Vliet A (2001) Inhibition of neutrophil apoptosis by acrolein: A mechanism of tobacco-related lung disease? Am. J. Physiol. Lung Cell. Mol. Physiol. 281:L732– L739 Finkelstein EI, Ruben J, Koot CW, Hristova M, van der Vliet A (2005) Regulation of constitutive neutrophil apoptosis by the alpha,beta-unsaturated aldehydes acrolein and 4-hydroxynonenal. Am. J. Physiol. Lung Cell Mol. Physiol. 289(6):L1019–L1028. Flo´-Neyret C, Lorenzi-Filho G, Macchione M, Garcia ML, Saldiva PH (2001) Effects of formaldehyde on the frog’s mucociliary epithelium as a surrogate to evaluate air pollution effects on the respiratory epithelium. Braz. J. Med. Biol. Res. 34:5:639–643. Fontignie-Houbrechts N (1981) Genetic effects of formaldehyde in the mouse. Mutat. Res. 88109– 119. Fornace Jr, AJ (1982) Detection of DNA single-strand breaks produced during the repair of damage by DNA-protein cross-linking agents. Cancer Res. 42:145–149. Fornace Jr, AJ, Lechner JF, Graftstrom RC, Harris CC (1982) DNA-repair in human bronchial epithelial cells. Carcinogenesis 3:1373–1377. Frazelle JH, Abernethy DJ, Boreiko CJ (1983) Weak promotion of C3H/l0T-1/2 cell transformation by repeated treatments with formaldehyde. Cancer Res. 43:3236–3239. French D, Edsall JT (1945) The reactions of formaldehyde with amino acids and proteins. Adv. Protein Chem. 2:277–335. Frigas E, Filley WV, Reed CE (1984) Bronchial challenge with formaldehyde gas: Lack of bronchoconstriction in 13 patients suspected of having formaldehyde-induced asthma. Mayo Clin. Proc. 59:295–299. Fujimaki H, Kurokawa Y, Kunugita N, Kikuchi M, Sato F, Arashidani K (2004) Differential immunogenic and neurogenic inflammatory responses in an allergic mouse model exposed to low levels of formaldehyde. Toxicology 197(1):1–13.

300

FORMALDEHYDE AND OTHER ALDEHYDES

Fujioka K, Shibamoto T (2006) Determination of toxic carbonyl compounds in cigarette smoke. Environ. Toxicol. 21(1):47–54. Fukumoto S, Yamauchi N, Moriguchi H, Hippo Y, Watanabe A, Shibahara J, Taniguchi H, Ishikawa S, Ito H, Yamamoto S, Iwanari H, Hironaka M, Ishikawa Y, Niki T, Sohara Y, Kodama T, Nishimura M, Fukayama M, Dosaka-Akita H, Aburatani H (2005) Overexpression of the aldo-keto reductase family protein AKR1B10 is highly correlated with smokers’ non-small cell lung carcinomas. Clin. Cancer Res. 11(5):1776–1785. Galloway SM, Armstrong MJ, Reubin C, Coleman S, Brown B, Cannon C, Bloom AD, Nakamura F, Ahmed M, Duk S, Rimpo J, Margolin BH, Resnick MA, Anderson B, Zeiger E (1987) Chromosome abberations and sister chromatid exchanges in Chinese hamster ovary cells: Evaluations of 108 chemicals. Environ. Mol. Mutagen. 10(Suppl):1–175. Gamble JF, McMichael AJ, Williams T, Battigelli M (1976) Respiratory function and symptoms: An environmental/epidemiologic study of rubber workers exposed to pheno-formaldehyde resin. Am. Ind. Hyg. Assoc. J. 37:499–512. Gammage RB, Gupta KC (1984) Formaldehyde. In: Walsh PJ, Dabney CS, Copenhaver ED, editors. Indoor Air Quality.Boca Raton, FL:CRC Press. Gannon PF, Bright P, Campbell M, O’Hickey SP, Burge PS (1995) Occupational asthma due to glutaraldehyde and formaldehyde in endoscopy and X-ray departments. Thorax 50(2):156– 159. Gardner MJ, Pannett B, Winter PD, Cruddas AM (1993) A cohort study of workers exposed to formaldehyde in the British chemical industry: an update. Br. J. Ind. Med. 50:872–834. Gardner R, Kazi S, Ellis EM (2004) Detoxication of the environmental pollutant acrolein by a rat liver aldo-keto reductase. Toxicol. Lett. 148(1–2):65–72. Georghiou PE, Snow D, Williams DT (1983) Formaldehyde monitoring in urea-formaldehyde houses in St John’s, Newfoundland, Canada: correlative field evaluation of a real-time infrared spectrophotometric method. Environ. Int. 9:279–287. Gerberich HR, Seaman GC (1994) In: Kroschwitz JI, Howe-Grant M, editors. Kirk-Othmer Encyclopedia of Chemical Technology.New York:J. Wiley and Sons. Gerin M, Siemiatycki J, Nadon L, Dewar R, Krewski D (1989) Cancer risk due to occupational exposure to formaldehyde: Results of a multi-site case-control study in Montreal. Int. J. Cancer 44:53–58. Gilliland FD, Gauderman WJ, Vora H, Rappaport E, Dubeau L (2002) Effects of glutathione-Stransferase M1, T1, and P1 on childhood lung function growth. Am. J. Respir. Crit. Care Med. 166 (5):710–716. Godish T (1989) Formaldehyde exposures from tobacco smoke: a review. Am. J. Public Health 79:1044–1045. Goedde HW, Agarwal DP (1989) Pharmacogenetics of aldehyde dehydrogenase (ALDH). Pharmacol. Ther. 45:345–371. Gonenc A, Ozkan Y, Torun M, Simsek B (2001) Plasma malondialdehyde (MDA) levels in breast and lung cancer patients. J. Clin. Pharm. Ther. 26(2):141–144. Gong Jr, H, Tashkin DP, Calvarsee BM (1981) Alcohol-induced bronchospasm in an asthmatic patient. Pharmacologic evaluation of the mechanism. Chest. 80:167–173. Go´rski P, Tarkowski M, Krakowiak A, Kiec´-Swierczyn´ska M (1992) Neutrophil chemiluminescence following exposure to formaldehyde in healthy subjects and in patients with contact dermatitis. Allergol. Immunopath. 20:20–23. Graftstrom RC (1990) In vitro studies of aldehyde effects related to human respiratory carcinogenesis. Mutat. Res. 238:175–184. Graftstrom RC, Fornace Jr, A, Harris CC (1984) Repair of DNA damage caused by formaldehyde in human cells. Cancer Res. 44:4323–4327.

REFERENCES

301

Graftstrom RC, Dypbukt JM, Willey JC, Sundqvist K, Edman C, Atzori L, Harris CC (1988) Pathobiological effects of acrolein in cultured human bronchial epithelial cells. Cancer Res. 48:1717–1721. Graftstrom RC, Dypbukt JM, Sundqvist K, Atzori L, Nielsen I, Curren RD, Harris CC (1994) Pathobiological effects of acetaldehyde in cultured human bronchial epithelial cells and fibroblasts. Cancinogenesis 15:985–990. Grandis JR, Tweardy DJ (1993) TGF-alpha and EGFR in head and neck cancer. J. Cell Biochem. Suppl. 17F:188–191. Green DJ, Sauder LR, Kulle TJ, Bascom R (1987) Acute responsed to 3.0 ppm formaldehyde in exercising healthy nonsmokers and asthmatics. Am. Rev. Respir Dis. 135:1261–1266. Grosjean D (1990) Atmospheric chemistry of toxic contaminants 2.saturated aliphatics acetaldehyde dioxane ethylene glycol ethers propylene oxide. J. Air Waste Manage. Assoc. 40:1522–1531. Grosjean D (1991) Ambient levels of formaldehyde, acetylaldehyde, and formic acid in Southern California: Results of a one-year base-line study. Environ. Sci. Technol. 25:710–715. Grosjean D, Grosjean E (2002) Airborne carbonyls from motor vehicle emissions in two highway tunnels. Res. Rep. Health Eff. Inst. 107:57–78. Grosjean D, Williams EL, Seinfeld JH (1992) Atmospheric oxidation of selected terpnes and related carbonyls gas-phase carbonyl products. Environ. Sci. Tech. 26:1526–1533. Grosjean E, Grosjean D, Fraser MP, Cass GR (1996) Air quality model evaluation data for organics. 2. C1-C 14 Carbonyls in Los Angeles Air. Environ. Sci. Technol. 30:2687–2703. Grundfest CC, Chang J, Newcombe D (1982) Acrolein: A potent modulator of lung macrophage arachidonic acid metabolism. Biochim. Biophys. Acta 713:149–159. Guillerm R, Badre R, Vignon B (1961) Effects inhibiteurs de la fume’e de tabac sur l’activie’ ciliaire de l’epithe’lium respiratoire et nature tea compsants responsables. Bull. Acad. Natl. Med. 145:416– 423. Gupta KC, Ulsamer AG, Preuso PW (1982) Formaldehyde in indoor air. Sources and toxicity. Environ. Int. 8:349–354. Hadi SM, Shahabuddin H, Rehman A (1989) Specificity of the interaction of furfural with DNA. Mutat. Res. 225:101–106. Hagmar L, Bellander T, Englander V, Ranstam J, Attewell R, Skerfring S (1986) Mortality and cancer among workers in a chemical factory. Scand. J. Work Environ. Health l2:545–551. Hales CA, Barkin PW, Jung W, Trautman E, Lamborghini D, Herrig N, Burke J (1988) Synthetic smoke with acrolein but not HCl produces pulmonary edema. J. Appl. Physiol. 64(3):1121–1133. Hales CA, Musto S, Hutchison WG, Mahoney E (1995) BW-755C diminishes smoke-induced pulmonary edema. J Appl Physiol. 78(1):64–69. Hales CA, Musto SW, Janssens S, Jung W, Quinn DA, Witten M (1992) Smoke aldehyde component influences pulmonary edema. J. Appl. Physiol. 72(2):555–561. Hamberg M, Samuelsson B (1967) Oxygenation of unsaturated fatty acids by the vesicular gland of sleep. J. Biol. Chem. 242:5341–5354. Hanrahan LP, Dally KA, Anderson HA, Kanarek MS, Ralkin J (1984) Formaldehyde vapor in mobile homes: A cross sectional survey of concentrations and irritant effects. Am. J. Public Health 74:1026–1027. Harley RA, Cass GR (1995) Modeling the atmospheric concentrations of individual volatile organic compounds. Atmos. Environ. 29:905–922. Harrington JM, Shannon HS, Mortality study of pathologists and medical laboratory technicians. (1975)BMJ 4:329–332. Hastie AT, Patrick H, Fish JE (1990) Inhibition and recovery of mammalian respiratory ciliary function after formaldehyde exposure. Toxicol. Appl. Pharmacol. 102:282–291.

302

FORMALDEHYDE AND OTHER ALDEHYDES

Hauptmann M, Lubin JH, Stewart PA, Hayes RB, Blair A (2003) Mortality from lymphohematopoietic malignancies among workers in formaldehyde industries. J. Natl. Cancer Inst. 95(21):1615–1623. Hauptmann M, Lubin JH, Stewart PA, Hayes RB, Blair A (2004) Mortality from solid cancers among workers in formaldehyde industries. Am. J. Epidemiol. 159:1117–1130. Hauptmann M, Lubin JH, Stewart PA, Hayes RB, Blair A (2005) The Authors Reply. Am. J. Epidemiol. 161(11):1090–1091. Hayes JD, Flanagan JU, Jowsey IR (2005) Glutathione transferases. Annu. Rev. Pharmacol. Toxicol. 45:51–88. Hayes RB, Raatgever JW, De Bruyn A, Gerin M (1986) Cancer of the nasal cavity and paranasal sinuses, and formaldehyde exposure. Int. J. Cancer 37:487–492. Hayes RB, Blair A, Stewart PA, Herrick RF, Mahar H (1990) Mortality of U. S. embalmers and funeral directors. Am. J. Ind. Med. 18:641–652. He S-M, Lambert B (1990) Acetaldehyde-induced mutation at the hprt locus in human lymphocytes in vitro. Environ. Mol. Mutagen. 16:57–63. Heck H, Casanova M (1999) Pharmacodynamics of formaldehyde: applications of a model for the arrest of DNA replication by DNA-protein cross-links. Toxicol. Appl. Pharmacol. 160(1):86–100. Heck H, Casanova M (2004) The implausibility of leukemia induction by formaldehyde: a critical review of the biological evidence on distant-site toxicity. Regul. Toxicol. Pharmacol. 40(2):92– 106. Heck H, d’A Casanova M, Starr TB (1990) Formaldehyde toxicity-new understanding. CRC Crit. Rev. Toxicol. 20:397–426. Heck H, d’A, Casanova M, Steinhagen WH, Everitt JL, Morgan KT, Popp JAS (1989) Formaldehyde toxicity: DNA-protein cross-linking studies in rats and nonhuman primates. In:Feron VJ, Bosland MC, editors. Nasal Carcinogenesis in Rodents: Relevance to Human Risk.Wageningen, The Netherlands:Pudoc. Heck H,d’A, Chin TY, Schmitz MC (1983) Distribution of 14C formaldehyde in rats after inhalation exposure. In:Gibson JE, editor. Formaldehyde Toxicity.Washington, DC:Hemisphere. Helander A, Carlsson S, Tottmar O (1988) Effects of disulfiram therapy on aldehyde dehydrogenase activity in human leukocytes and erythrocytes. Biochem. Pharmacol. 37:3360–3363. Hemminki K (1981) Reactions of formaldehyde with guanosine. Toxicol. Lett. 9:161–164. Hemminki K, Falck K, Vainio H (1980) Comparison of alkylation rates and mutagenicity of directly acting industrial and laboratory chemicals. Arch. Toxicol. 46:277–283. Hendrick DJ, Lane DJ (1975) Formal in asthma in hospital staff. BMJ 1:607–608. Hendrick DJ, Lane DJ (1977) Occupational formal in asthma. Br. J. Ind. Med. 34:11–18. Hendrick DJ, Rando RJ, Lane DJ, Morris MJ (1982) Formaldehyde asthma: Challenge exposure levels and fate after five years. J. Occup. Med. 24:893–897. Hernberg S, Westerholm P, Schults-Larsen K, Degerth R, Kuosma E, Englund A, Engzell U, Hansen HS, Mutanen P (1983) Nasal and sinonasal cancer. Scand. J. Work Environ. Health 9:315– 326. Hester SD, Barry WT, Zou F, Wolf DC (2005) Transcriptomic analysis of F344 rat nasal epithelium suggests that the lack of carcinogenic response to glutaraldehyde is due to its greater toxicity compared to formaldehyde. Toxicol. Pathol. 33(4):415–424. Hester SD, Benavides GB, Yoon L, Morgan KT, Zou F, Barry W, Wolf DC (2003) Formaldehydeinduced gene expression in F344 rat nasal respiratory epithelium. Toxicology 187(1):13–24. Hildesheim A, Dosemeci M, Chan CC, Chen CJ, Cheng YJ, Hsu MM, Chen IH, Mittl BF, Sun B, Levine PH, Chen JY, Brinton LA, Yang CS (2001) Occupational exposure to wood, formaldehyde, and solvents and risk of nasopharyngeal carcinoma. Cancer Epidemiol. Biomark. Prev. 10:1145– 1153.

REFERENCES

303

Hilton J, Dearman RJ, Basketter DA, Scholes EW, Kimber I (1996) Experimental assessment of the sensitizing properties of formaldehyde. Food Chem. Toxicol. 34(6):571–578. Holmstro¨m M, Wilhelmsson B, Hellquist H, Rose´n G. (1989a) Histological changes in the nasal mucosa in persons occupationally exposed to formaldehyde alone and in combination with wood dust. Acta Otolaryngol. 107(1-2):120–129. Holmstro¨m M, Wilhelmsson B, Hellquist H (1989b) Histological changes in the nasal mucosa in rats long-term exposure to formaldehyde and wood dust. Acta Otolaryngol. 108:274–283. Horsfall FL (1934) Formaldehyde hypersensitiveness: An experimental study. J. Immunol. 27:569– 581. Horvath JJ, Witmer CM, Witz G (1992) Nephrotoxicity of the 1, 1 acrolein-glutathione adduct in the rat. Toxicol. Appl. Pharmacol. 117(2):200–207. Horvath Jr, EP, Anderson Jr H, Pierce WE, Hanrahan L, Wendick JD (1988) Effects of formaldehyde on the mucous membranes and lungs: A study of an industrial population. JAMA 259(5):701– 707. Hsie AW, O’Neill J, San Sabastian JR, Couch OB, Brimen PA, Sun WNC, Fucoe JC, Forbes NL, Machanoff R, Riddle JC, Hsie MH (1979) Quantitative mammalian cell toxicology: Study of the cytotoxicity and mutagenicity of seventy individual environmental agents related to energy technologies and three subfractions of a crude synthetic oil in the CHO/HGPRT system. E.P. A. Report 600/9-78-027. pp.293–315. Hsu LC, Chang WC, Yoshida A (1994) Cloning of a cDNA encoding human ALDH7, a new member of the aldehyde dehydrogenase family. Gene 151(1–2):285–289. Hsu LC, Bendel RE, Yoshida A (1987) Direct detection of usual and atypical alleles on the human aldehyde dehydrogenase-2 (ALDH2) locus. Am. J. Hum. Genet. 41:966–1001. Hyvelin JM, Roux E, Prevost MC, Savineau JP, Marthan R (2000) Cellular mechanisms of acroleininduced alteration in calcium signaling in airway smooth muscle. Toxicol. Appl. Pharmacol. 164 (2):176–183. Hyvelin JM, Savineau JP, Marthan R (2001) Selected contribution: Effect of the aldehyde acrolein on acetylcholine-induced membrane current in airway smooth muscle cells. J. Appl. Physiol. 90 (2):750–754. International Agency for Research on Cancer (IARC). (1982) IARC Monographs on the Evaluation of the Carcinogeneic Risk of Chemicals to Humans, Vol. 29. Some Industrial Chemicals and Dyestuffs.Lyon.:International Agency for Research in Cancer,. International Agency for Research on Cancer (IARC). (1985) IARC Monographs on the Evaluation of the Carcinogeneic Risk of Chemicals to Humans. Vol. 36. Allyl Compounds, Aldehydes, Epoxides, and Peroxides.Lyon, France:International Agency for Research in Cancer. International Agency for Research on Cancer (IARC). (1995a) IARC Monographs on the Evaluation of the Carcinogeneic Risk of Chemicals to Humans. Vol. 62. Wood dust and Formaldehyde. Lyon: International Agency for Research in Cancer. International Agency for Research on Cancer (IARC). (1995b) IARC Monographs on the Evaluation of the Carcinogeneic Risk of Chemicals to Humans. Vol. 63. Dry cleaning, some chlorinated solvents and other industrial chemicals. Lyon:International Agency for Research in Cancer. pp373–391. International Agency for Research on Cancer (IARC).(2004) IARC Monographs on the Evaluation of the Carcinogeneic Risk of Chemicals to Humans. Vol. 88. Formaldehyde, 2-Butoxyethanol and 1-tert-Butoxy-2-propanol. Lyon.: International Agency for Research in Cancer. Isse T, Oyama T, Matsuno K, Ogawa M, Narai-Suzuki R, Yamaguchi T, Murakami T, Kinaga T, Uchiyama I, Kawamoto T (2005) Paired acute inhalation test reveals that acetaldehyde toxicity is higher in aldehyde dehydrogenase 2 knockout mice than in wild-type mice. J. Toxicol. Sci. 30 (4):329–333.

304

FORMALDEHYDE AND OTHER ALDEHYDES

Ito K, Sakamoto T, Hayashi Y, Morishita M, Shibata E, Sakai K, Takeuchi Y, Torii S (1996) Role of tachykinin and bradykinin receptors and mast cells in gaseous formaldehyde-induced airway microvascular leakage in rats. Eur. J. Pharmacol. 307(3):291–298. Izard C, Liberman C (1978) Acrolein. Mutat. Res. 47:115–138. Jakab GJ (1977) Adverse effect of a cigarette smoke component, acrolein, on pulmonary antibacterial defenses and on viral-bacterial interactions in the lung. Am. Rev. Respir. Dis. 115:33–38. Jakab GJ (1993) The toxicologic interactions resulting from inhalation of carbon black and acrolein on pulmonary antibacterial and antiviral defenses. Toxiclol. Appl. Pharmaacol. 121:167–175. Jakab GJ, Hemenway DR (1993) Inhalation coexposure to carbon black and acrolein suppresses alveolar macrophage phagocytosis and TNF-a release and modulates peritoneal macrophage phagocytes. Inhalation Toxicol. 5:265–279. Janssens SP, Musto SW, Hutchison WG, Spence C, Witten M, Jung W, Hales CA (1994) Cyclooxygenase and lipoxygenase inhibition by BW-755C reduces acrolein smoke-induced acute lung injury. J. Appl. Physiol. 77(2):888–895. Jermini C, Weber A, Grandjean E (1976) Quantitative determination of various gas-phase components of the side-stream smoke of cigarettes in the room air as a contribution to the problem of passivesmoking. Int. Arch. Occup. Environ. Health 36:169–181. Kamata E, Nakadate M, Uchida O, Ogawa Y, Suzuki S, Kaneko T, Saito M, Kurokawa Y (1997) Results of a 28-month chronic inhalation toxicity study of formaldehyde in male Fisher-344 rats. J. Toxicol. Sci. 22(3):239–254. Kane LE, Alarie Y (1977) Sensory irritation to formaldehyde and acrolein during single and repeated exposures in mice. Am. Ind. Hyg. Assoc. J. 38:509–522. Kasahara DI, Poynter ME, Othman Z, Hemenway D, van der Vliet A (2008) Acrolein inhalation suppresses lipopolysaccharide-induced inflammatory cytokine production but does not affect acute airways neutrophilia. J. Immunol. 181(1):736–745. Kawai Y, Furuhata A, Tokyokuni S, Arantani Y, Uchida K (2003) Formation of acrolein-derived 20 deoxyadenosine adduct in an iron-induced carcinogenesis model. J. Biol. Chem. 278(50):50346– 50354. Kensler CJ, Battista SP (1963) Components of cigarette smoke with ciliary-depressant activity. N. Engl. J. Med. 269:1161–1166. Kerfoot EJ, Mooney TF (1975) Formaldehyde and paraformaldehyde study in funeral homes. Am. Ind. Hyg. Assoc. J. 36:533–537. Kerns WD, Pavkov KL, Donofrio DJ, Gralla EJ, Swenberg JA (1983) Carcinogenicity of formaldehyde in rats mice after long-term inhalation exposure. Cancer Res. 43:4382–4392. Khamgaonkar MB, Fulare MB (1991) Pulmonary effects of formaldehyde exposure. An environmental-epidemiological study. Indian J. Chest Dis. Allied Sci. 33:9–13. Kilburn KH, McKenzie WN (1978) Leukocyte recruitment to airways by aldehyde-carbon combinations that mimic cigarette smoke. Lab. Invest. 38:134–142. Kilburn KH, Seidman BC, Warshaw R (1985) Neurobehavioral and respiratory symptoms of formaldehyde and xylene exposure in histology technicians. Arch. Environ. Health 40:229–233. Kim CW, Song JS, Ahn YS, Park SH, Park JW, Noh JH, Hong CS (2001) Occupational asthma due to formaldehyde. Yonsei Med. J. 42(4):440–445. Kim WJ, Terada N, Nomura T, Takahashi R, Lee SD, Park JH, Konno A (2002) Effect of formaldehyde on the expression of adhesion molecules in nasal microvascular endothelial cells: the role of formaldehyde in the pathogenesis of sick building syndrome. Clin. Exp. Allergy. 32 (2):287–295. Kimbell JS, Subramaniam RP, Gross EA, Schlosser PM, Morgan KT (2001) Dosimetry modeling of inhaled formaldehyde: comparisons of local flux predictions in the rat, monkey, and human nasal passages. Toxicol. Sci. 64(1):100–110.

REFERENCES

305

Kimbell JS, Gross EA, Joyner DR, Godo MN, Morgan KT (1993) Application of computational fluid dynamics to regional dosimetry to inhaled chemicals in the upper respiratory tract of the rat. Toxicol. Appl. Pharmacol. 121:253–263. Kimbell JS, Gross EA, Richardson RB, Conolly RB, Morgan KT (1997) Correlation of regional formaldehyde flux predictions with the distribution of formaldehyde-induced squamous metaplasia in F344 rat nasal passages. Mutat. Res. 380:143–154. Kita T, Fujimura M, Myou S, Ishiura Y, Abo M, Katayama N, Nishitsuji M, Yoshimi Y, Nomura S, Oribe Y, Nakao S (2003) Potentiation of allergic bronchoconstriction by repeated exposure to formaldehyde in guinea-pigs in vivo. Clin. Exp. Allergy 33(12):1747–1753. Koivusalo M, Baumann M, Uotila L (1989) Evidence for the identity of glutathione-dependent formaldehyde dehydrogenase and class III alcohol dehydrogenase. FEBS Lett. 257(1):105–109. Kralinger MT, Hinterholzl J, Troger J, Nguyen QA, Kremser B, Fischer-Colbrie R, Kieselbach GF (2003) Elevated levels of secretoneurin in the rabbit aqueous humor in response to formaldehyde irritation. Graefes Arch. Clin. Exp. Ophthalmol. 241(7):577–581. Kruysse A, Feron VJ, Til HP (1975) Repeated exposure to acetaldehyde vapour. Studies in Syrian golden hamsters. Arch. Environ. Health. 30:449–452. Krzyzanowski M, Quackenboss JJ, Lebowitz MD (1990) Chronic respiratory effects of indoor formaldehyde exposure. Environ. Res. 52:117–125. Kulle T, Cooper G (1975) Effects of formaldehyde and ozone on the trigeminal sensory system. Arch. Environ. Health 30:327–333. Kulle TJ, Sauder LR, Hebel JR, Green DJ, Chatham MD (1987) Formaldehyde dose-response in healthy nonsmokers. J. Air Pollut. Control Assoc. 37:919–924. Kurtz AJ, Lloyd RS (2003) 1,N2-deoxyguanosine adducts of acrolein, crotonaldehyde, and trans-4hydroxynonenal cross-link to peptides via Schiff base linkage. J. Biol. Chem. 278(8):5970–5976. Kurys G, Ambroziak W, Pietruszko R (1989) Human aldehyde dehydrogenase. J. Biol. Chem. 264:4715–4721. Kutzman RS, Popenoe EA, Schmaeler M, Drew RT (1985) Changes in rat lung structure and composition as a result of subchronic exposure to acrolein. Toxicology 34:139–151. Lam C-W, Casanova M, Heck Hd’A (1985) Depletion of nasal mucosal glutathione by acrolein and enhancement of formaldehyde-induced DNA-protein cross-linking by simultaneous exposure to acrolein. Arch. Toxicol. 58:67–71. Lambert C, McCue J, Portas M, Ouyang Y, Li J, Rosano TG, Lazis A, Freed BM (2005) Acrolein in cigarette smoke inhibits T-cell responses. J. Allergy Clin. Immunol. 116(4):916–922. Lan Q, Mumford JL, Shen M, Demarini DM, Bonner MR, He X, Yeager M, Welch R, Chanock S, Tian L, Chapman RS, Zheng T, Keohavong P, Caporaso N, Rothman N (2004) Oxidative damagerelated genes AKR1C3 and OGG1 modulate risks for lung cancer due to exposure to PAH-rich coal combustion emissions. Carcinogenesis 25(11):2177–2181. Lang I, Bruckner T, Triebig G (2008) Formaldehyde and chemosensory irritation in humans: a controlled human exposure study. Regul. Toxicol. Pharmacol. 50(1):23–36. Leach CL, Hatoum NS, Ratajczak HV, Gerhart JM (1987) The pathologic and immunologic effects of inhaled acrolein in rats. Toxicol. Lett. 39:189–198. Lee BP, Morton RF, Lee L-Y (1992) Acute effects of acrolein on breathing: role of vagal bronchopulmonary afferents. J. Appl. Physiol. 72:1050–1056. Lee H-K, Alarie Y, Karol MH (1984) Induction of formaldehyde sensitivity in guinea pigs. Toxicol. Appl. Pharmacol. 75:147–155. Leffingwell CM, Low RB (1979) Cigarette smoke components and alveolar macrophage protein synthesis. Arch. Environ. Health 34:97–102. Leikauf GD (2002) Hazardous air pollutants and asthma. Environ. Health Perspect. 110(Suppl. l4):505–526.

306

FORMALDEHYDE AND OTHER ALDEHYDES

Leikauf GD, Doupnik CA (1989) Formaldehyde induced bronchial hyperresponsiveness in guinea pig. Am. Rev. Respir. Dis. 139:A392 Leikauf GD, Leming LM, O’Donnell JR, Doupnik CA (1989a) Bronchial responsiveness and inflammation in guinea pigs exposed to acrolein. J. Appl. Physiol. 66:171–178. Leikauf GD, Doupnik CA, Leming LM, Wey HE (1989b) Sulfidopeptide leukotrienes mediate acrolein-induced bronchial hyperresponsiveness. J. Appl. Physiol. 66:1838–1845. Lemiere C, Desjardins A, Cloutier Y, Drolet D, Perrault G, Cartier A, Malo JL (1995) Occupational asthma due to formaldehyde resin dust with and without reaction to formaldehyde gas. Eur. Respir. J. 8(5):861–865. Levine JR, Anjelkovich D, Shaw LK (1984) The mortality of Ontario undertakers and a review of formaldehyde related mortality studies. J. Occup. Med. 26:740–746. Li L, Hamilton Jr, RF, Holian A (1999) Effect of acrolein on human alveolar macrophage NF-kappaB activity. Am. J. Physiol. 277(3 Pt 1):L550–L557. Li L, Hamilton Jr RF, Taylor DE, Holian A (1997) Acrolein-induced cell death in human alveolar macrophages. Toxicol. Appl. Pharmacol. 145:331–339. Liber HL, Benforado K, Crosby RM, Simpson D, Skopek TR (1989) Formaldehyde-induced and spontaneous alterations in human hprt DNA sequence and mRNA expression. Mutat. Res. 226:31–37. Liebling T, Rosenman KD, Pastides H, Griffith RG, Lemeshow S (1984) Cancer mortality among workers exposed to formaldehyde. Am. J. Ind. Med. 5:423–428. Liteplo RG, Meek ME (2003) Inhaled formaldehyde: exposure estimation, hazard characterization, and exposure-response analysis. J. Toxicol. Environ. Health B. Crit. Rev. 6(1):85–114. Liu L, Yan Y, Zeng M, Zhang J, Hanes MA, Ahearn G, McMahon TJ, Dickfeld T, Marshall HE, Que LG, Stamler JS (2004) Essential roles of S-nitrosothiols in vascular homeostasis and endotoxic shock. Cell. 116(4):617–628. Liu W, Zhang J, Kwon J, Weisel C, Turpin B, Zhang L, Korn L, Morandi M, Stock T, Colome S (2006) Concentrations and source characteristics of airborne carbonyl compounds measured outside urban residences. J. Air Waste Manag. Assoc. 56(8):1196–1204. Lovell MA, Xie C, Markesbery WR (2001) Acrolein is increased in Alzheimer’s disease brain and is toxic to primary hippocampal cultures. Neurobiol. Aging 22:187–194. Luce D, Gerin M, Leclerc A, Morcet JF, Brugere J, Goldberg M (1993) Sinonasal cancer and occupational exposure to formaldehyde and other substances. Int. J. Cancer 53:224–231. Luo J, Robinson JP, Shi R (2005) Acrolein-induced cell death in PC12 cells: role of mitochondriamediated oxidative stress. Neurochem. Int. 47(7):449–457. Lyapina M, Zhelezova G, Petrova E, Boev M (2004) Flow cytometric determination of neutrophil respiratory burst activity in workers exposed to formaldehyde. Int. Arch. Occup. Environ. Health 77(5):335–40. Lynen R, Rothe M, Gallasch E (1983) Characterization of formaldehyde-related antibodies in hemodialysis patients at different stages of immunization. Vox Sang. 44:81–89. Lyon JP, Jenkins LJ, Jones RA, Coon RA, Siegel J (1970) Repeated and continuous exposure of laboratory animals to acrolein. Toxicol. Appl. Pharmacol. 17:726–732. Ma TH, Harris MM (1988) Review of the genotoxicity of formaldehyde. Mutat. Res. 196:37–59. Magana-Schwenke N, Ekert B (1978) Biochemical analysis of damage induced in yeast by formaldehyde. II. Induction of crosslinks between DNA and protein. Mutat. Res. 51:11–19. Magana-Schwenke N, Ekert B, Moustacchi E (1978) Biochemical analysis of damage induced in yeast by formaldehyde. 1. Induction of single-strand breaks in DNA and their repair. Mutat. Res. 50:181–193. Magnusson R, Nilsson C, Andersson B (2002) Emissions of aldehydes and ketones from a two-stroke engine using ethanol and ethanol-blended gasoline as fuel. Environ. Sci. Technol. 36(8):1656– 1664.

REFERENCES

307

Mannix RC, Phalen RF, Walters RB, Kurosaki TT (1983) Effects of sulfur dioxide and formaldehyde on particle clearance in the rat. J. Toxicol. Environ. Health 12:429–440. Marnett LJ (1988) Health effects of aldehydes and alcohols in Mobil source emissions. In: Watson AY, Bates RR, Kennefy D, editors. Air Pollution, the Automobile, and Public Health. Washington, DC: National Academy Press. Marnett LJ, Tuttle M, A (1980) Comparison of the mutagenicities of malondialdehyde and side products formed during its chemical synthesis. Cancer Res. 40:276–282. Marnett LJ, Holly HK, Hollstein MC, Levin DE, Esterbauer H, Ames BN (1985) Naturally occurring carbonyl compounds are mutagens in Salmonella tester strain TA104. Mutat. Res. 148:25–34. Maronpot RR, Miller RA, Clarke WJ, Westerberg RB, Decker JR, Moss OR (1986) Toxicity of formaldehyde vapor in B6C3F, mice exposed for 13 weeks. Toxicology 1:253–266. Marsh GM (1982) Proportional mortality patterns among chemical plant workers exposed to formaldehyde. Br. J. Ind. Med. 39:313–322. Marsh GM, Youk AO (2004) Reevaluation of mortality risks from leukemia in the formaldehyde cohort study of the National Cancer Institute. Regul Toxicol Pharmacol. 40(2):113–124. Marsh GM, Youk AO, Buchanich JM, Cassidy LD, Lucas LJ, Esmen NA, Gathuru IM (2002) Pharyngeal cancer mortality among chemical plant workers exposed to formaldehyde. Toxicol. Ind. Health 18(6):257–268. Marsh GM, Youk AO, Morfeld P (2006 Sep 22) Mis-specified and non-robust mortality risk models for nasopharyngeal cancer in the National Cancer Institute formaldehyde worker cohort study. Regul Toxicol Pharmacol. 2006: [Epub ahead of print]. Marsh GM, Stone RA, Esmen NA, Henderson VL (1994) Mortality patterns among chemical plant workers exposed to formaldehyde and other substances. J. Natl. Cancer Inst. 86:384–386. Martin CN, McDermid AC, Garner RC (1978) Testing of known carcinogens and non carcinogens for their ability to induce unscheduled DNA synthesis in HeLa cells. Cancer Res. 38:2621– 2627. Matsuda T, Yabushita H, Kanaly RA, Shibutani S, Yokoyama A (2006) Increased DNA damage in ALDH2-deficient alcoholics. Chem. Res. Toxicol. 19(10):1374–1378. Maurice F, Rivory J-H, Larsson PH, Johansson SGO, Bouquest J (1986) Anaphylactic shock caused by formaldehyde in a patient undergoing long-term hemodialysis. J. Allergy Clin. Immunol. 77:594– 597. McCarthy MC, Hafner HR, Montzka SA (2006) Background concentrations of 18 air toxics for North America. J. Air Waste Manage. Assoc. 56(1):3–11. McIlwain CC, Townsend DM, Tew KD (2006) Glutathione S-transferase polymorphisms: cancer incidence and therapy. Oncogene 25(11):1639–1648. McLaughlin JK (1994) Formaldehyde and cancer: a critical review. Int. Arch. Occup. Environ. Health 66:295–301. Merk O, Speit G (1998) Significance of formaldehyde-induced DNA–protein crosslinks for mutagenesis. Environ. Mol. Mutagen. 32:260–268. Miller DP, Asomaning K, Liu G, Wain JC, Lynch TJ, Neuberg D, Su L, Christiani DC (2006) An association between glutathione S-transferase P1 gene polymorphism and younger age at onset of lung carcinoma. Cancer. 107(7):1570–1577. Miller DP, De Vivo I, Neuberg D, Wain JC, Lynch TJ, Su L, Christiani DC (2003) Association between self-reported environmental tobacco smoke exposure and lung cancer: modification by GSTP1 polymorphism. Int. J. Cancer. 104(6):758–763. Misonou Y, Asahi M, Yokoe S, Miyoshi E, Taniguchi N (2006) Acrolein produces nitric oxide through the elevation of intracellular calcium levels to induce apoptosis in human umbilical vein endothelial cells: implications for smoke angiopathy. Nitric Oxide 14(2):180–187.

308

FORMALDEHYDE AND OTHER ALDEHYDES

Mizoi Y, Yamamoto K, Ueno Y, Fukunaga T, Harada S (1994) Involvement of genetic polymorphism of alcohol and aldehyde dehydrogenases in individual variation of alcohol metabolism. Alcohol 29 (6):707–710. Monticello TM, Miller FJ, Morgan KT (1991) Regional increases in rat nasal epithelial cell proliferation following acute and subchronic inhalation of formaldehyde. Toxicol. Appl. Pharmacol. 134:515–527. Monticello TM, Swenberg JA, Gross EA, Leininger JR, Kimbell JS, Selikop S, Starr TB, Gibson JE, Morgan KT (1996) Correlation of regional and nonlinear formaldehyde-induced cancer with proliferating populations of cells. Cancer Res. 56:1012–1022. Monticello TM, Morgan KT, Everitt JI, Popp JA (1989) Effects of formaldehyde gas on the respiratory track of rhesus monkeys. Am. J. Pathol. 134:515–527. Morello-Frosch RA, Woodruff TJ, Axelrad DA, Caldwell JC (2000) Air toxics and health risks in California: The public health implications of outdoor concentrations. Risk Anal. 20(2):273– 291. Morgan KT, Patterson DL, Gross EA (1986) Responses of the nasal mucociliary apparatus of F-344 rats to formaldehyde gas. Toxicol. Appl. Pharmacol. 82:1–13. Morgan KT, Patterson DL, Gross EA (1984) Frog palate mucociliary apparatus: Structure, function. and response to formaldehyde gas. Fundam. Appl. Toxicol. 4:58–68. Morris JB (1997) Uptake of acetaldehyde vapor and aldehyde dehydrogenase levels in the upper respiratory tracts of the mouse, rat, hamster, and guinea pig. Fund. Appl. Toxicol. 35:91–100. Morris JB (1998) Effect of acrolein vapor on upper respiratory tract uptake of acetaldehyde. Inhal. Toxicol. 10:843–845. Morrison GC, Nazaroff WW (2002) Ozone interactions with carpet: secondary emissions of aldehydes. Environ. Sci Technol. 36(10):2185–2192. Mukai FH, Goldstein BD (1976) Mutagenicity of malondialdehyde, a decomposition product of peroxidized polyunsaturated fatty acids. Science 191:868–869. Munnia A, Bonassi S, Verna A, Quaglia R, Pelucco D, Ceppi M, Neri M, Buratti M, Taioli E, Garte S, Peluso M (2006) Bronchial malondialdehyde DNA adducts, tobacco smoking, and lung cancer. Free Radic. Biol. Med. 41(9):1499–1505. Murphy SD, Klingshirn DA, Ulrich CE (1963) Respiratory response of guinea pigs during acrolein inhalation and its modification by drugs. J. Pharmacol. Exp. Ther. 141:79–83. Muto M, Nakane M, Hitomi Y, Yoshida S, Sasaki S, Ohtsu A, Yoshida S, Ebihara S, Esumi H (2002) Association between aldehyde dehydrogenase gene polymorphisms and the phenomenon of field cancerization in patients with head and neck cancer. Carcinogenesis 23(10):1759– 1765. Myou S, Fujimura M, Nishi K, Ohka T, Matsuda T (1993) Aerosolized acetaldehyde induces histamine-mediated bronchoconstriction in asthmatics. Am. Rev. Respir. Dis. 148:940–943. Nardini M, Finkelstein EI, Reddy S, Valacchi G, Traber M, Cross CE, van der Vliet A (2002) Acroleininduced cytotoxicity in cultured human bronchial epithelial cells. Modulation by alpha-tocopherol and ascorbic acid. Toxicology 170(3):173–185. Nath RG, Chung FL (1994) Detection of exocyclic 1,N2-propanodeoxyguanosine adducts as common DNA lesions in rodents and humans. Proc. Natl. Acad. Sci. U.S.A. 91(16):7491–7495. Nath RG, Ocando JE, Chung FL (1996) Detection of 1, N2-propanodeoxyguanosine adducts as potential endogenous DNA lesions in rodent and human tissues. Cancer Res. 56(3):452–456. Nath RG, Ocando JE, Guttenplan JB, Chung FL (1998) 1,N2-propanodeoxyguanosine adducts: potential new biomarkers of smoking-induced DNA damage in human oral tissue. Cancer Res. 58 (4):581–584. National Research Council Formaldehyde and Other Aldehydes. Washington:National Academy Press.(1981)

REFERENCES

309

Naylor S, Mason RP, Sanders JKM, Williams DH, Moneti G (1988) Formaldehyde adducts of glutathione. Biochem. J. 249:573–579. Nazaroff WW, Singer BC (2004) Inhalation of hazardous air pollutants from environmental tobacco smoke in US residences. J. Expo. Anal. Environ. Epidemiol. 14(Suppl. 1):S71–S77. Nebert DW, Vasiliou V (2004) Analysis of the glutathione S-transferase (GST) gene family. Hum. Genomics 1(6):460–464. Neudecker T, Lutz D, Eder E, Hensehler D (1981) Crotonaldehyde is mutagenie in a modified Salmonella typhimurium mutagenieity testing system. Mutat. Res. 91:27–39. Nilsson GE, Tottmar O (1987) Effects of biogenic aldehydes and aldehyde dehydrogenase inhibitors on rat brain tryptophan hydroxylase activity in vitro. Brain Res. 409:374–379. Nocentini S, Moreno S, Coppey J (1980) Survival, DNA synthesis and ribosomal RNA transeription in monkey kidney eells treated by formaldehyde. Mutat. Res 70:231–240. Noiri E, Yamada S, Nakao A, Tsuchiya M, Masaki I, Fujino K, Nosaka K, Ozawa T, Fujita T, Uchida K (2002) Serum protein acrolein adducts: utility in detecting oxidant stress in hemodialysis patients and reversal using a vitamin E-bonded hemodialyzer. Free Radic. Biol. Med. 33:1651–1656. Noisel N, Bouchard M, Carrier G (2007) Evaluation of the health impact of lowering the formaldehyde occupational exposure limit for Quebec workers. Regul. Toxicol. Pharmacol. 48(2):118–127. Norback D, Bjornsson E, Jansen C, Widstrom J, Borman G (1995) Asthmatic symptoms and volatile organic compounds, formaldehyde, and carbon dioxide in dwellings. Occup. Environ. Med. 52:388–395. Nordman H (1985) Formaldehyde asthma––rare or overlooked? J. Allergy Clin. Immunol. 75:91–99. Obe G, Beck B (1979) Mutagenic activity of aldehydes. Drug Alcohol Depend. 4:91–94. Occupational Safety and Health Administration (1985) Occupational exposure to formaldehyde: Proposed rule and notice of hearing, 29 CFR Part 1920. Federal Register 50:50412. Ohtsuka R, Shuto Y, Fujie H, Takeda M, Harada T, Itagaki S (1997) Response of respiratory epithelium of BN and F344 rats to formaldehyde inhalation. Exp. Anim. 46(4):279–286. Olsen JH, Asneas S (1986) Formaldehyde and the risk of squamous cell carcinoma of the sinonasal cavities. Br. J. Ind. Med. 43:769–774. Olsen JH, Plough-Jensen S, Hink M, Faurbo K, Breum NO, Moller Jensen O (1984) Occupational formaldehyde exposure and inereased nasal cancer risk in man. Int. J. Cancer 34:639–644. Ong TH, Tan Kl, Lee HS, Eng P (2004) A case report of occupational asthma due to gluteraldehyde exposure. Ann. Acad. Med. Singapore 33(2):275–278. Oomichi S, Kita H (1974) Effect of air pollutants on ciliary activity of respiratory tract. Bull. Tokyo Med. Dent. Univ. 21:327–343. Othmer DF (1987) Methanol to gasoline? Chem. Eng. News 30:2–3. Overton JH, Kimbell JS, Miller FJ (2001) Dosimetry modeling of inhaled formaldehyde: the human respiratory tract. Toxicol. Sci. 64(1):122–134. Pal A, Hu X, Zimniak P, Singh SV (2000) Catalytic efficiencies of allelic variants of human glutathione S-transferase Pi in the glutathione conjugation of alpha, beta-unsaturated aldehydes. Cancer Lett. 154(1):39–43. Partanen T (1993) Formaldehyde exposure and respiratory cancer––a meta-analysis of the epidemiologic evidence. Scand. J. Work Environ. Health 11:409–415. Patel JM, Wood JC, Leibman KC (1980) The biotransformation of allyl alcohol and acrolein in rat liver and lung preparations. Drug Metab. Dispos. 8(5):305–308. Patterson R, Dykewicz MS, EvansIII R, Grammer LC, Greenberger PA, Harris KE, Lawrenee ID, Pruzansky JJ, Roberts M, Shaughnessy MA, Zeiss CR (1989) IgG antibody against formaldehyde human serum proteins: A eomparison with other IgG antibodies against inhalant proteins and reactive chemicals. J. Allergy Clin. Immunol. 9:359–366.

310

FORMALDEHYDE AND OTHER ALDEHYDES

Patterson DL, Gross EA, Bogdanffy MS, Morgan KT (1986) Retention of formaldehyde gas in the nasal passages of Fisher-344 rats. Toxicologist 6:55 Pattle RE, Cullumbine H (1956) Toxicity of some atmospheric pollutants. BMJ 2:913–916. Paustenbach D, Alarie Y, Kulle T, Schachter N, Smith R, Swenberg J, Witschi H, Horowitz SB (1997) A recommended occupational exposure limit for formaldehyde based on irritation. J. Toxicol. Environ. Health 50(3):217–263. Pazdrak K, Go´rski P, Krakowiak A, Ruta U (1993) Changes in nasal lavage fluid due to formaldehyde inhalation. Int. Arch. Occup. Environ. Health. 64:515–519. Pettersson S,RehnT (1977) Determination ofthe odorthresholdforformaldehyde.Hyg.Miljo. 10:53–59. Pickrell JA, Griffs LC, Mokler BV, Vanapilly GM, Hobbs CH (1983) Release of formaldehyde from various consumer products. Environ. Sci. Technol. 17:753–764. Pinkerton LE, Hein MJ, Stayner LT (2004) Mortality among a cohort of garment workers exposed to formaldehyde: an update. Occup. Environ. Med. 61(3):193–200. Pratt GC, Palmer K, Wu CY, Oliaei F, Hollerbach C, Fenske MJ (2000) An assessment of air toxics in Minnesota. Environ. Health Perspect. 108(9):815–825. Pross HF, Day JH, Clark RH, Lees REM (1987) Immunologic studies of subjects with asthma exposed to formaldehyde and urea-formaldehyde foam insulation (UFFI) off products. J. Allergy Clin. Immunol. 79:797–810. Que LG, Liu L, Yan Y, Whitehead GS, Gavett SH, Schwartz DA, Stamler JS (2005) Protection from experimental asthma by an endogenous bronchodilator. Science 308(5728):1618–1621. R. J. Reynolds Tobacco Company. (1988) Chemical composition of new cigarette smoke. In: Chemical and Biological Studies on New Cigarette Prototypes that Heat Instead of Burn Tobacco. Winston-Salem, NC:R. J. Reynolds Tobacco Company. Rader J Irritative Effects of Formaldehyde in Laboratory Halls. University of Wurzberg:J. D. Dissertation. Ragan DL, Boreiko CJ (1981) Initiation of C3H/l0T-1/2 cells transformation by formaldehyde. Cancer Lett. 13:325–331. Recio L, Sisk S, Phutts L, Bermudez E, Gross EA, Chen Z, Morgan K, Walker C (1992) p53 mutations in formaldehyde-induced nasal squamous cell carcinomas in rats. Cancer Res. 52:6113–6116. Reynolds SH, Stowers SJ, Patterson RM, Maronpot RR, Aaronson SA, Anderson MW (1987) Activated oncogenes in B6C3HF1 mouse liver tumors: implications for risk assessment. Science 237:1309–1316. Riedel F, Hasenauer E, Barth PJ, Koziorowski A, Reiger CRL (1996) Formaldehyde exposure enhances inhalative allergic sensitization in the guinea pig. Allergy 51:94–99. Ritchie M, Lehnen RG (1987) Formaldehyde-related health complaints of residents living in mobile and conventional homes. Am. J. Public Health 77:323–328. Rodriguez F, de la Roza C, Jardi R, Schaper M, Vidal R, Miravitlles M (2005) Glutathione S-transferase P1 and lung function in patients with alpha1-antitrypsin deficiency and COPD. Chest 127(5):1537–1543. Roemer E, Anton HJ, Kindt R (1993) Cell proliferation in the respiratory tract of the rat after acute inhalation of formaldehyde or acrolein. J. Appl. Toxicol. 13:103–107. Roush GC, Walrath J, Stayner LT, Kalplan SA, Flannery JT, Blair A (1987) Nasopharyngeal cancer, sinonasal cancer, and occupations related to formaldehyde: a case-control study. J. Natl. Cancer Inst. 79:1221–1224. Roux E, Hyvelin JM, Savineau JP, Marthan R (1998) Calcium signaling in airway smooth muscle cells is altered by in vitro exposure to the aldehyde acrolein. Am. J. Respir. Cell Mol. Biol. 19(3):437–444. Rumchev KB, Spickett JT, Bulsara MK, Phillips MR, Stick SM (2002) Domestic exposure to formaldehyde significantly increases the risk of asthma in young children. Eur. Respir. J. 20 (2):403–408.

REFERENCES

311

Rusch GM, Bolte HF, Rinehart WE A 26-week inhalation toxicity study with formaldehyde in the monkey, rat, and hamster In: Gibson JE, editor. Formaldehyde Toxicity.Washington: Hemisphere. (1983) Ryan TJ, Burroughs GE, Taylor K, Kovein RJ (2003) Video exposure assessments demonstrate excessive laboratory formaldehyde exposures. Appl. Occup. Environ. Hyg. 18(6):450–457. Sadakane K, Takano H, Ichinose T, Yanagisawa R, Shibamoto T (2002) Formaldehyde enhances mite allergen-induced eosinophilic inflammation in the murine airway. J. Environ. Pathol. Toxicol. Oncol. 21(3):267–276. Salathe M, Pratt MM, Wanner A (1993) Protein kinase C-dependent phosphorylation of a ciliary membrane protein and inhibition of ciliary beating. J. Cell Sci. 106 Pt 41211–1220. Saldiva PH, do Rio Caldeira MP, Massad E, Calheiros DF, Cardoso LM, Bohm GM, Saldiva CD (1985) Effects of formaldehyde and acetaldehyde inhalation on rat pulmonary mechanics. J. Appl. Toxicol. 5:288–292. Salem H, Cullumbine H (1960) Inhalation toxicities of some aldehydes. Toxicol. Appl. Pharmacol. 2:183–187. Samet JM, Marbury MC, Spengler JD (1987) Health effects and sources of indoor air pollution. Part 1. Am. Rev. Respir. Dis. 136:1486–1508. Samet JH, Spengler JD (1991) Indoor Air Pollution. Baltimore: J. Hopkins University Press. Sandler SG, Sharon R, Stroup M, Sabo B (1979) Formaldehyde-related antibodies in hemodialysis patients. Transfusion 19:682–687. Sauder LR, Chatham MD, Green DJ, Kulle TJ (1986) Acute pulmonary response to formaldehyde exposure in healthy nonsmokers. J. Occup. Med. 28:420–424. Schachter EN, Witek Jr TJ, Brody DJ, To sun T, Beck GJ, Leaderer BP (1987) A study of respiratory effects from exposure to 2.0 ppm formaldehyde in occupationally exposed workers. Environ. Res. 44:188–205. Schachter EN, WitekJrTJ, Tosun T, Leaderer B, Beck GJ (1986) A study of respiratory effects from exposure to 2 ppm formaldehyde in health subjects. Arch. Environ. Health 41:229–239. Schafer D, Brommer C, Riechelmann H, Mann JW (1999) In vivo and in vitro effect of ozone and formaldehyde on human nasal mucociliary transport system. Rhinology 37(2):56–60. Schaller KH, Triebig G, Beyer B (1989) Formaldehyde determination in tobacco smoke—studies under experimental and actual conditions. Zentrabl. Hyg. Umweltmed. 189:103–110. Schoenberg JB, Mitchell CA (1975) Airway disease cause by phenolic (phenol-formaldehyde) resin exposure. Arch. Environ. Health 30:574–577. Schuck EA, Stephens ER, Middleton JT (1966) Eye irritation responses at low concentrations of irritants. Arch. Environ. Health 13:570–575. Seaman VY, Charles MJ, Cahill TM (2006) A sensitive method for the quantification of acrolein and other volatile carbonyls in ambient air. Anal Chem. 78(7):2405–2412. Seidell A (1940) Solubilities of Inorganic and Metal Organic Compounds. New York:D. Van Nostrand. Seidell A (1941) Solubilities of Organic Compounds. New York:D. Van Nostrand. Sellakumar AR, Synder CA, Solomon JJ, Albert RE (1985) Carcinogenicity of formaldehyde and hydrogen chloride in rats. Toxicol. Appl. Pharmacol. 81:401–406. Shah JJ, Singh HB (1998) Distribution of volatile organic chemicals in outdoor and indoor air. Environ. Sci. Techol. 22:1381–1388. Shaham J, Gurvich R, Kaufman Z (2002) Sister chromatid exchange in pathology staff occupationally exposed to formaldehyde. Mutat. Res. 514(1–2):115–123. Shamberger RJ, Andreone TL, Willis CE (1974) Anti oxidants and cancer IV. Initiating activity of malonaldehyde as a carcinogen. J. Natl. Cancer Inst. 53:1171–1773. Shamberger RJ, Corlett CL, Beaman KD, Kasten BL (1979) Antioxidant reduce the mutagenic effect of malondialdehyde and 13-propiolactone. Part IX. Antioxidants and cancer. Mutat. Res. 66:349–355.

312

FORMALDEHYDE AND OTHER ALDEHYDES

Shao B, Fu X, McDonald TO, Green PS, Uchida K, O’Brien KD, Oram JF, Heinecke JW (2005) Acrolein impairs ATP binding cassette transporter A1-dependent cholesterol export from cells through site-specific modification of apolipoprotein A-I. J. Biol. Chem. 280:36386– 36396. Sheppard D, Eschenbacher WL, Epstein J (1984) Lack of bronchomotor response to up to 3 ppm formaldehyde in subjects with asthma. Environ. Res. 35:133–139. Sheppard D, Wong WS, Uehara CF, Nadel JA, Boushey HA (1980) Lower threshold and greater bronchomotor responsiveness of asthmatic subjects to sulfur dioxide. Am. Rev. Respir Dis. 122:873–878. Sherwood RL, Leach CL, Hatoum NS, Aranyi C (1986) Effects of acrolein on macrophage functions in rats. Toxicol. Lett. 32:41–49. Shibuya A, Yasunami M, Yoshida A (1989) Genotypes of alcohol dehydrogenase and aldehyde dehydrogenase loot in Japanese alcohol flushers and nonflushers. Hum. Genet. 82:14–16. Shusterman D, Salines J, Cone J (1988) Behavioral sensitization to irritants/odorants after acute overexposures. J. Occup. Med 30:565–567. Sisson JH, Tuma DJ (1994) Vapor phase exposure to acetaldehyde generated from ethanol inhibits bovine bronchial epithelial cell ciliary motility. Alcohol Clin. Exp. Res. 18(5):1252–1255. Songco G, Fahey PJ (1987) Indoor and outdoor air pollutants and health. Compr. Ther. 13(7):41–48. Speit G, Merk O (2002) Evaluation of mutagenic effects of formaldehyde in vitro: detection of crosslinks and mutations in mouse lymphoma cells. Mutagenesis 17(3):183–187. Speit G, Schmid O (2006) Local genotoxic effects of formaldehyde in humans measured by the micronucleus test with exfoliated epithelial cells. Mutat. Res. 613(1):1–9. St Clair MB, Gross EA, Morgan KT (1990) Pathology and cell proliferation induced by intra-nasal instillation of aldehydes in the rat: comparison of glutaraldehyde and formaldehyde. Toxicol. Pathol. 18:353–361. Stayner LT, Elliott L, Blade L, Keenlyside R, Halperin W (1988) A restrospective cohort mortality study of workers exposed to formaldehyde in the garment industry. Am. J. Ind. Med. 13:667–681. Stayner L, Smith AB, Reeve G, Blade L, Elliott L, Keenlyside R, Halperin W (1986) Proportionate mortality study of workers in the garment industry exposed to formaldehyde. Am. J. Ind. Med 7:229–240. Stein S, Lao Y, Yang IY, Hecht SS, Moriya M (2006) Genotoxicity of acetaldehyde- and crotonaldehydeinduced 1,N2-propanodeoxyguanosine DNA adducts in human cells. Mutat. Res. 608(1): 1–7. Steinhagen WH, Barrow CS (1984) Sensory irritation structure-activity study of inhaled aldehydes in B6C3F, and Swiss-Webster mice. Toxicol. Appl. Pharmacol. 72:495–503. Sterling TD, Weinkam JJ (1988) Reanalysis of lung cancer mortality in a National Cancer Institute study on mortality among industrial workers exposed to formaldehyde. J. Occup Med. 30:895–901. Sterling T, Arundel A (1985) Formaldehyde and lung cancer. Lancet 2:1366–1367. Stevens JF, Maier CS (2008) Acrolein: sources, metabolism, and biomolecular interactions relevant to human health and disease. Mol. Nutr. Food. Res. 52(1):7–25. Stroup NE, Blair A, Erikson GE (1986) Brain cancer and other causes of death in anatomists. J. Natl. Cancer. Inst. 77:1217–1224. Sun X, Jia Y, Zhang X, Xu Q, Shen Y, Li Y (2005) Multi-locus association study of schizophrenia susceptibility genes with a posterior probability method. Sci. China C. Life Sci. 48(3):263–269. Suruda A, Schulte P, Boeniger M, Hayes RB, Livingston GK, Steenland K, Stewart P, Herrick R, Douthit D, Fingerhut MA (1993) Cytogenetic effects of formaldehyde exposure in students of mortuary science. Cancer Epidemiol. Biomarkers Prev. 2(5):453–460. Swarin SJ, Lipari F (1983) Determination of formaldehyde and other aldehydes by high performance liquid chromatography with fluorescence detection. J. Liq. Chromatogr. 6:425–444.

REFERENCES

313

Swenberg JA, Gross EA, J Martin J, Popp JA (1983) Mechanisms of formaldehyde toxicity. In:Gibson JA, editor. Formaldehyde Toxicity.Washington:Hemisphere. Swenberg JA, Kerns WD, Mitchell RI, Gralla EJ, Pavkov KL (1980) Induction of squamous cell carcinomas of the rat nasal cavity by inhalation exposure to formaldehyde vapor. Cancer Res. 40:3398–3402. Swiecichowski A, Long KJ, Milller ML, Leikauf GD (1993) Formaldehyde-induced airway hyperreactivity in vivo an ex vivo in guinea pigs. Environ. Res. 61:185–199. Tam BN, Neumann CM (2004) A human health assessment of hazardous air pollutants in Portland. OR. J. Environ. Manage. 73(2):131–145. Tanel A, Averill-Bates DA (2005) The aldehyde acrolein induces apoptosis via activation of the mitochondrial pathway. Biochim. Biophys. Acta. 1743(3):255–267. Tarkowski M, Go´rski P (1995) Increased IgE antiovalbumin level in mice exposed to formaldehyde. Int. Arch. Allergy Immunol. 106:422–424. Tarone RE, McLaughlin JK (2005) Re: “Mortality from Solid Cancers among Workers in Formaldehyde Industries”. Am. J. Epidemiol. 161(11):1089–1090. Thrasher JD, Broughton A, Micevich P (1988) Antibodies and immune profiles of individuals occupationally exposed to formaldehyde: Six case reports. Am. J. Ind. Med 14:479–488. Thrasher JD, Wojdani A, Cheung G, Heuser G (1987) Evidence for formaldehyde antibodies and altered cellular immunity in subjects exposed to formaldehyde in mobile homes. Arch. Environ. Health 42:347–350. Townsend AJ, Leone-Kabler S, Haynes RL, Wu Y, Szweda L, Bunting KD (2001) Selective protection by stably transfected human ALDH3A1 (but not human ALDH1A1) against toxicity of aliphatic aldehydes in V79 cells. Chem. Biol. Interact. 130–132(1–3):261– 273. Trump BF, Smith MW, Miyashita M, Phelps PC, Jones RT, Harris CC (1988) The effects of aldehydes on cytosolic Ca2 of normal and transformed human bronchial cells. J. Cell. Biol. 107:72a. Tu C, Wynns GC, Silverman DN (1989) Chemical modification of carbonic anhydrase II with acrolein. J. Biol. Chem. 264(21):12389–12393. Tuazon EC, Aschmann SM, Arey J, Atkinson R (1997) Products of the gas-phase reactions of O-3 with a series of methyl-substituted ethenes. Environ. Sci. Technol. 31:3004–3009. Turner CR, Stow RB, Hubbs SJ, Gomes BC, Williams JC (1993a) Acrolein increases airway sensitivity to substance P and decreases NEP activity in guinea pigs. J. Appl. Physiol. 74:1830–1839. Turner CR, Stow RB, Hubbs SJ, Talerico SD, Christian EP, Williams JC (1993b) Protective role for neuropeptides in acute pulmonary response to acrolein in guinea pigs. J. Appl. Physiol. 75:2456– 2465. Tyihak E, Bocsi J, Timar F, Racz G, Szende B (2001) Formaldehyde promotes and inhibits the proliferation of cultured tumour and endothelial cells. Cell Prolif. 34:135–141. Uba G, Pachorek D, Bernstein J, Garabrant DH, Salines JR, Wright WE, Amar RB (1989) Prospective study of respiratory effects of formaldehyde among healthy and asthmatic medical students. Am. J. Ind. Med. 15:91–101. Uchida K, Kanematsu M, Morimitsu Y, Osawa T, Noguchi N, Niki E (1998) Acrolein is a product of lipid peroxidation reaction. Formation of free acrolein and its conjugate with lysine residues in oxidized low density lipoproteins. J. Biol. Chem. 273:16058–16066. Vaca CE, Nilsson JA, Fang JL, Graftstrom RC (1998) Formation of DNA adduct in human buccal epithelial cells exposed to acetaldehyde and methylglyoxal in vitro. Chemico-Biol. Interact. 108:197–208. Vandenplas O, Fievez P, Delwiche JP, Boulanger J, Thimpont J (2004) Persistent asthma following accidental exposure to formaldehyde. Allergy 59(1):115–116.

314

FORMALDEHYDE AND OTHER ALDEHYDES

Vasiliou V, Pappa A, Estay T (2004) Role of human aldehyde dehydrogenases in endobiotic and xenobiotic metabolism. Drug Metab. Rev. 36:279–299. Vaughan TL, Strader C, Davis S, Daling S (1986) Formaldehyde and cancers of the pharynx, sinus, and nasal cavity. 1. Occupational exposures. Int. J. Cancer 38:677–683. Vaughan TL, Strader C, Davis S, Daling JR (1986b) Formaldehyde and cancers of the pharynx, sinus and nasal cavity. II. Residential exposures. Int. J. Cancer 38:685–688. Vaughan TL, Stewart PA, Teschke K, Lynch CF, Swanson GM, Lyon JL, Berwick M (2000) Occupational exposure to formaldehyde and wood dust and nasopharyngeal carcinoma. Occup. Environ. Med. 57(6):376–384. Vegheli PV, Osztovics M (1978) The alcohol syndromes: The intrarecombigenic effects of acetaldehyde. Experientia 34:195–201. Walrath J, Fraumeri Jr, JF (1983) Mortality patterns among embalmers. Int. J. Cancer 31:407–411. Walrath J, Fraumeri Jr, JF (1984) Cancer and other causes of death among embalmers. Cancer Res. 44:4638–4641. Watanabe T (1991) Mechanism of ethanol-induced bronchoconstriction in Japanese asthmatic patients. Japan J. Allergy 40:1210–1217. Watson MA, Stewart RK, Smith GBJ, Massey TE, Bell DA (1998) Human glutathione S-transferase P1 polymorphisms: relationship to lung tissue enzyme activity and population frequency. Carcinogenesis 19:275–280. Wayne LG, Bryan RJ, Ziedman KJ (1977) Irritant Effects of Industrial Chemicals: Formaldehyde.US DHEW, PHS:NIOSH Publication. 77:117. Weber-Tschopp A, Fisher T, Grahdjean E (1977) Reizwirkugen tea Formaldehyds (HCHO) auf den Menschen. Int. Arch. Occup. Environ. Health 39:207–218. Wenzlaff AS, Cote ML, Bock CH, Land SJ, Schwartz AG (2005) GSTM1, GSTT1 and GSTP1 polymorphisms, environmental tobacco smoke exposure and risk of lung cancer among never smokers: a population-based study. Carcinogenesis 26(2):395–401. West S, Hildesheim A, Dosemeci M (1993) Non-viral risk factors for nasopharyngeal carcinoma in the Philippines: results from a case-control study. Int. J. Cancer 55(5):722–727. WHO. World Health Organization. International Programme on Chemical Safety. (1989) Environmental Health Criteria 89. Formaldehyde. Geneva:World Health Organization. WHO. World Health Organization. International Programme on Chemical Safety. (1992) Environmental Health Criteria 127. Acrolein. Geneva:World Health Organization. WHO. World Health Organization. International Programme on Chemical Safety. (1995) Environmental Health Criteria 167. Acetaldehyde.Geneva:World Health Organization. Wieslander G, Norbackvd D, Bjornsson E, Janson C, Borman G (1997) Asthma and the indoor environment: the significance of emission of formaldehyde and volatile organic compounds from newly painted indoor surfaces. Int. Arch. Occup. Environ. Health. 69:115–124. Wilhelmsson B, Holmstro¨m M (1987) Positive formaldehyde-rast after prolonged formaldehyde exposure by inhalation. Lancet 2:164. Wilkins RJ, Macleod HD (1976) Formaldehyde induced DNA-protein cross links in E. coli. Mutat. Res. 36:11–16. Windholz M (1983) Merck Index. Rahway, NJ:Merck and Co. WitekJrTJ, Schachter EN, Tosun T, Beck GJ, Leaderer BP (1987) An evaluation of respiratory effects following exposure to 2.0 ppm formaldehyde in asthmatics: Lung function, symptoms, and airway reactivity. Arch. Environ. Health 42:230–237. Witz G, Lawrie NJ, Amoruso MA, Goldstein BD (1985) Inhibition by reactive aldehydes of superoxide anion radical production in stimulated human neutrophils. Chem. Biol. Interact. 53 (1–2):13–23.

REFERENCES

315

Witz G, Lawrie NJ, Goldstein BD, Ryer-Powder J, Amoruso MA (1988) Effects of alpha, betaunsaturated aldehydes on macrophage and neutrophil membrane function, fluidity and sulfhydryl status. Basic Life Sci. 49:849–851. Woodruff TJ, Wells EM, Holt EW, Burgin DE, Axelrad DA (2007) Estimating risk from ambient concentrations of acrolein across the United States. Environ. Health Perspect. 115(3):410–415. Wolf DC, Gross EA, Lyght O, Bermudez E, Recio L, Morgan KT (1995) Immunohistochemical localization of p53, PCNA, and TGF-alpha proteins in formaldehyde-induced rat nasal squamous cell carcinomas. Toxicol. Appl. Pharmacol. 132(1):27–35. Wong LB, Park CL, Yeates DB (1998) Neuropeptide Y inhibits ciliary beat frequency in human ciliated cells via nPKC, independently of PKA. Am. J. Physiol. 275(2 Pt 1):C440–C448. Woutersen RA, Feron VJ (1987) Inhalation toxicity of acetaldehye in rats. IV. Progression and regression of nasal lesions after discontinuation of exposure. Toxicology 47:295–305. Woutersen RA, van Garderen-Hoetmer A, Bruitntjes JP, Zwart A, Feron VJ (1989) Nasal tumours in rats after severe injury to the nasal mucosa and prolonged exposure to 10 ppm formaldehyde. J. Appl. Toxicol. 9:39–46. Woutersen RA, Appelman LM, Feron VJ, Van der Heijden CA (1984) Inhalation toxicity of acetaldehyde in rats. II. Carcinogenicity study: interim results after 15 months. Toxicology. 31(2):123–133. Woutersen RA, Appelman LM, van Garderen-Hoetmer A, Feron VJ (1986a) Inhalation toxicity of acetaldehyde in rats. III. Carcinogenicity study. Toxicology 41:213–231. Woutersen RA, Appelman LM, Feron VJ, van Garderen-Hoetmer A (1986b) Inhalation toxicity of acetaldehyde in rats. II. Carcinogenicity study: interim results after 15 months. Toxicology 31:123–133. Wu CC, Hsieh CW, Lai PH, Lin JB, Liu YC, Wung BS (2006) Upregulation of endothelial heme oxygenase-1 expression through the activation of the JNK pathway by sublethal concentrations of acrolein. Toxicol. Appl. Pharmacol. 214(3):244–252. Wyatt TA, Heires AJ, Sanderson SD, Floreani AA (1999) Protein kinase C activation is required for cigarette smoke-enhanced C5a-mediated release of interleukin-8 in human bronchial epithelial cells. Am. J. Respir. Cell Mol. Biol. 21(2):283–288. Wynder EL, Hoffman D (1963) Ein experimenteller Beitrag zur Tabakrauchkanzerogenese. Dtsch. Med. Wochenschr. 88:623–628. Yager JW, Cohn KL, Spear RC, Fisher JM, Morse L (1986) Sister-chromatid exchanges in lymphocytes of anatomy students exposed to formaldehyde-embalming solution. Mutat. Res. 174(2):135–139. Yang IY, Chan G, Miller H, Huang Y, Torres MC, Johnson F, Moriya M (2002) Mutagenesis by acroleinderived propanodeoxyguanosine adducts in human cells. Biochemistry 41(46):13826–13832. Yang X, Wu X, Choi YE, Kern JC, Kehrer JP (2004) Effect of acrolein and glutathione depleting agents on thioredoxin. Toxicology 204(2–3):209–218. Yang X, Zhang YP, Chen D, Chen WG, Wang R (2001) Eye irritation caused by formaldehyde as an indoor air pollution––A controlled human exposure experiment. Biomed. Environ. Sci. (3):229–236. Yau TM (1979) Mutagenicity and cytotoxicity of malonaldehyde in mammalian cells. Mech. Aging Dev. 11:137–144. Ye X, Yan W, Xie H, Zhao M, Ying C (2005) Cytogenetic analysis of nasal mucosa cells and lymphocytes from high-level long-term formaldehyde exposed workers and low-level short-term exposed waiters. Mutat. Res. 588(1):22–27. Ying CJ, Ye XL, Xie H, Yan WS, Zhao MY, Xia T, Yin SY (1999) Lymphocyte subsets and sisterchromatid exchanges in the students exposed to formaldehyde vapor. Biomed. Environ. Sci. 12 (2):88–94. Yokoyama A, Omori T (2005) Genetic polymorphisms of alcohol and aldehyde dehydrogenases and risk for esophageal and head and neck cancers. Alcohol 35(3):175–185.

316

FORMALDEHYDE AND OTHER ALDEHYDES

Yoshida A, Rzhetsky A, Hsu LC, Chang C (1998) Human aldehyde dehydrogenase gene family. Eur. J. Biochem. 251:549–557. Yousefipour Z, Ranganna K, Newaz MA, Milton SG (2005) Mechanism of acrolein-induced vascular toxicity. J. Physiol. Pharmacol. 56(3):337–353. Zhang S, Villalta PW, Wang M, Hecht SS (2006) Analysis of crotonaldehyde- and acetaldehydederived 1,n(2)-propanodeoxyguanosine adducts in DNA from human tissues using liquid chromatography electrospray ionization tandem mass spectrometry. Chem. Res. Toxicol. (10):1386– 1392. Zhang H, Forman HJ (2008) Acrolein induces heme oxygenase-1 through PKC-delta and PI3K in human bronchial epithelial cells. Am. J. Respir. Cell Mol. Biol. 38(4):483–490. Zimmermann M, Zouhair A, Azria D, Ozsahin M (2006) The epidermal growth factor receptor (EGFR) in head and neck cancer: its role and treatment implications. Radiat. Oncol. 1:11 Zwart A, Woutersen RA, Wilmer JWGM, Spit BJ, Feron VJ (1988) Cytotoxic and adaptive effects in rat nasal epithelium after 3-day and 13-week exposure to low concentrations of formaldehyde vapour. Toxicology 51:87–99.

10 AMBIENT AIR PARTICULATE MATTER Morton Lippmann

A broad variety of processes produce suspended particulate matter (PM) in the ambient air in which we live and breathe, and there is an extensive body of epidemiological literature that demonstrates that there are statistically significant associations between the concentrations of airborne PM and the rates of mortality and morbidity in human populations. The PM concentrations have almost always been expressed in terms of mass, although recent studies suggest that number concentration may correlate better with some effects than does mass (Peters et al., 1997; Stolzel et al., 2003). Also, in studies that reported on associations between health effects and more than one mass concentration, the strength of the association generally improves as one goes from total suspended particulate matter (TSP) to thoracic particulate matter, a.k.a. PM less than 10 mm in aerodynamic diameter (PM10), to fine particulate matter, a.k.a. PM less than 2.5 mm in aerodynamic diameter (PM2.5). The influence of a sampling system inlet on the sample mass collected is illustrated in Fig. 10.1. The PM2.5 distinction, while nominally based on particle size, is a means of measuring the total gravimetric concentration of several specific chemically distinctive classes of fine particles that are emitted into or formed within the ambient air as very small particles. In the former category (emitted) are carbonaceous particles in wood smoke and diesel engine exhaust. In the latter category (formed) are carbonaceous particles formed during the photochemical reaction sequence that also leads to ozone formation, as well as acidic sulfur and nitrogen oxide compounds resulting from the oxidation of sulfur dioxide and nitrogen oxide vapors released during fuel combustion, and were chemically changed by their neutralization by ammonia. The coarse particle fraction, that is, those particles with aerodynamic diameters larger than 2.5 mm, are largely composed of soil and mineral ash that are mechanically dispersed into the air. Both the fine and coarse fractions are complex mixtures in a chemical sense. To the extent that they are in equilibrium in the ambient air, it is a dynamic equilibrium in which they enter the air at about the same rate as they are removed. In dry weather, the concentrations of coarse particles are balanced between dispersion into the air, mixing

Environmental Toxicants, Third Edition Edited by Morton Lippmann Copyright
2009 John Wiley & Sons, Inc.

317

318

AMBIENT AIR PARTICULATE MATTER

∆MASS/∆(log Da), (µg/m-3)

70 60

Fine mode particles

Coarse mode particles TSP Hi Vol

50

WRAC

40

PM10

30 20 10 PM2.5

0 5 10 20 50 100 0.1 0.2 0.5 1.0 2 Aerodynamic particle diameter (Da), (µm) Total suspended particulate matter (TSP)

PM10 PM2.5

PM(10-2.5)

FIGURE 10.1 Representative bimodal mass distribution as a function of aerodynamic particle diameter for Phoenix, AZ, Showing effect of size-selective sampling inlet on mass collected for (a) wide-ranging aerosol classifier (WRAC), (b) standard total suspended particulate (TSP) highvolume sampler, (c) Sampler following EPA’s (PM10) criteria for thoracic dust, and (d) sampler following EPA’s criteria for fine particulate matter (PM2.5). Source: U.S. EPA (1996a, 1996b).

with air masses, and gravitational fallout, while the concentrations of fine particles are determined by rates of formation, chemical transformation, and meteorological factors. PM concentrations of both fine and coarse PM are effectively depleted by rainout and washout associated with rain. The coarse particle fraction includes those less than 10 mm, which can penetrate into the thorax and cause some of the health effects associated with PM in ambient air (PM10–2.5). Further elaboration of these distinctions is provided in Table 10.1. In the absence of any detailed understanding of the specific chemical components responsible for the health effects associated with exposures to ambient air PM10, PM2.5, and PM10–2.5, and in the presence of a large and consistent bodies of epidemiological evidence associating ambient air PM size fractions with mortality and morbidity that cannot be explained by potential confounders such as other pollutants, aeroallergens, or ambient temperature or humidity, public health authorities have established ambient air standards based on mass concentrations within the fine and coarse thoracic size fractions. This chapter summarizes the nature and extent of the health effects associated with gravimetric ambient air PM concentrations and a limited number of specific components such as sulfate and nickel. Further discussions of the health effects of some of the specific constituents of the ambient air PM are discussed in the chapters on asbestos and other mineral fibers, diesel engine exhaust, lead, nitrogen oxides, and sulfur oxides.

10.1 SOURCES AND PATHWAYS FOR HUMAN EXPOSURE As indicated in Table 10.1, fine and coarse particles generally have distinct sources and formation mechanisms, although there may be some overlap. Primary fine particles are

319

Sources:

Solubility:

Composed of:

Formed by:

Metals: compounds of Pb, Cd, V, Ni, Cu, Zn, Mn, Fe, etc.

Organic compounds with very low saturation vapor pressure at ambient temperature

Probably less soluble than accumulation mode Combustion

Largely soluble, hygroscopic, and deliquescent Combustion of coal, oil, gasoline, diesel fuel, wood

Particle-bound water

Large variety of organic compounds

Metal compounds

Sulfate

Elemental carbon

Condensation

Break-up of large solids/droplets

Accumulation

Coagulation Reactions of gases in or on particles Evaporation of fog and cloud droplets in which gases have dissolved and reacted Sulfate, nitrate, ammonium, and hydrogen ions Elemental carbon

Condensation Coagulation

Combustion, high temperature processes, and atmospheric reactions Nucleation

Ultrafine

Fine Coarse

(continued)

Resuspension of industrial dust and soil tracked onto roads and streets

CaCO3, CaSO4, NaCl, sea salt Pollen, mold, fungal spores Plant and animal fragments Tire, brake pad, and road wear debris Largely insoluble and nonhygroscopic

Fly ash from uncontrolled combustion of coal, oil, and wood Nitrates/chlorides/sulfates from HNO3/HCl/SO2 reactions with coarse particles Oxides of crustal elements (Si, Al, Ti, Fe)

Suspended soil or street dust

Mechanical disruption (crushing, grinding, abrasion of surfaces) Evaporation of sprays Suspension of dusts Reactions of gases in or on particles

Comparison of Ambient Particles, Fine Particles (Ultrafine Plus Accumulation Mode) and Coarse Particles

Formation processes:

TABLE 10.1

320

Minutes–hours Grows into accumulation mode Diffuses to raindrops <1–10 s of km

High-temperature processes

Source: Adapted from Wilson and Suh (1997); CD, pp. 2–52.

Travel Distance:

Ultrafine

Atmospheric transformation of SO2 and some organic compounds

(Continued)

Atmospheric half-life: Removal Processes:

TABLE 10.1 Fine

100–1000 s of km

Days–weeks Forms cloud droplets and rains out

Atmospheric transformation products of NOx, SO2, and organic compounds, including biogenic organic species (e.g., terpenes) High-temperature processes, smelters, steel mills, etc.

Accumulation

Uncontrolled coal and oil combustion Ocean spray Biological sources Minutes–hours Dry deposition by fallout Scavenging by falling rain drops <1–10 s of km (small size tail, 100–1000 s in dust storms)

Construction and demolition

Suspension from disturbed soil (e.g., farming, mining, unpaved roads)

Coarse

SOURCES AND PATHWAYS FOR HUMAN EXPOSURE

321

formed from condensation of high temperature vapors during combustion. Secondary fine particles are usually formed from gases in three ways: (1) nucleation (i.e., gas molecules coming together to form a new particle); (2) condensation of gases onto existing particles; and (3) by reaction of absorbed gases in liquid droplets. Particles formed by nucleation also coagulate to form relatively larger aggregate particles or droplets with diameters between 0.1 and 1.0 mm, and such particles normally do not grow into the coarse mode. Particles form as a result of chemical reaction of gases in the atmosphere that lead to products that either have a low enough vapor pressure to form a particle, or react further to form a low vapor pressure substance. Some examples include: (1) the conversion of sulfur dioxide (SO2) to sulfuric acid droplets (H2SO4); (2) reactions of H2SO4 with ammonia (NH3) to form ammonium bisulfate (NH4HSO4) and ammonium sulfate (NH4)2SO4; (3) the conversion of nitrogen dioxide (NO2) to nitric acid vapor (HNO3), which reacts further with NH3 to form the semivolatile particulate ammonium nitrate (NH4NO3). Although directly emitted particles are found in the fine fraction (the most common being particles less than 1.0 mm in diameter from combustion sources), particles formed secondarily from gases generally dominate the fine fraction mass. By contrast, most of the coarse fraction particles are emitted directly as particles, and result from mechanical disruption such as crushing, grinding, evaporation of sprays, or suspensions of dust from construction and agricultural operations. Basically, most coarse particles are formed by breaking up bigger masses into smaller ones. Energy considerations normally limit coarse particle sizes to greater than 1.0 mm in diameter. Some combustiongenerated particles, such as fly ash, are also found in the coarse fraction. Fine and coarse mode particles generally have distinct chemical composition, solubility, and acidity. Fine mode PM is mainly composed of varying proportions of several major components: inorganic ions (Hþ, NH4þ, NO3 , and SO4 ¼ ); elemental carbon (EC); organic carbon (OC) compounds; trace elements; and water. By contrast, coarse fraction constituents are primarily crustal in origin, consisting of oxides of Si, Al, Fe, and K (note, however, that small amounts of Fe and K are also found among the fine mode particles that stem from different sources). Biological material such as bacteria, pollen, and spores may also be found in the coarse mode. As a result of the fundamentally different chemical compositions and sources of fine and coarse fraction particles, the chemical composition of the sum of these two fractions, PM10, is more heterogeneous than either mode alone. Figure 10.2 presents a synthesis of the available published data on the chemical composition of PM2.5 in U.S. cities by region. Each fraction also has regional patterns resulting from the differences in sources and atmospheric conditions. Figure 10.3 illustrates the differences in chemical composition of PM10–2.5, PM2.5, and PM0.1 in Los Angeles. In addition to the larger relative shares of crustal materials in the U.S. west, total concentrations of coarse fraction particles are generally higher in the arid areas of the southwestern United States. In general, fine and coarse particles exhibit different degrees of solubility and acidity. With the exception of carbon and some organic compounds, fine particle mass is largely soluble in water and hygroscopic (i.e., fine particles readily take up and retain water). Except under fog conditions, the fine particle mode also contains almost all of the strong acid. By contrast, coarse mineral particles are mostly insoluble, nonhygroscopic, and generally basic. Fine and coarse particles typically exhibit different behavior in the atmosphere. These differences affect several exposure considerations including the representativeness of central-site monitored values and the behavior of particles that were formed outdoors after they penetrate into homes and buildings where people spend most of their time.

322

AMBIENT AIR PARTICULATE MATTER

25 20

1st Quarter

15 10 5 0 25 20

2nd Quarter

15

Crustal TCM Ammonium nitrate Ammonium Sulfate

10 5 0 25 20

3rd Quarter

15 10 5 0 25 20

4th Quarter

15 10 5 0 Northeast

Industrial Southwest Southern Midwest California Southeast Northwest Upper Midwest

FIGURE 10.2 Seasonal average composition of urban PM2.5 by region, 2003. Data from EPA Speciation Network. Components (from top to bottom) are crustal material, total carbonaceous mass (TCM), ammonium nitrate, and ammonium sulfate Source: Schmidt et al. (2005).

Fine accumulation mode particles typically have longer atmospheric lifetimes (i.e., days to weeks) than coarse particles, and tend to be more uniformly dispersed across an urban area or large geographic region, especially in the eastern United States. Atmospheric transformations can take place locally, during atmospheric stagnation, or during transport over long distances. For example, the formation of sulfates from SO2 emitted by power plants with tall stacks can occur over distances exceeding 500 km and 12 h of transport time; therefore, the resulting particles are well mixed in the air shed. Once formed, the very low dry deposition velocities of fine particles contribute to their persistence and uniformity throughout an air mass. Larger particles generally deposit more rapidly than small particles; as a result, total coarse particle mass will be less uniform in concentration across a region than are fine particles. Because coarse particles may vary in size from about 1 mm to over 100 mm, it is important to note their wide range of atmospheric behavior characteristics. For example, the larger coarse particles (>10 mm) tend to rapidly fall out of the air and have atmospheric lifetimes of only minutes to hours depending on their size, wind velocity, and other factors. Their spatial impact is typically limited by a tendency to fall out in the proximate area downwind of their emission point. The atmospheric behavior of the smaller particles within

AMBIENT AIR PM CONCENTRATIONS

12.9%

323

PM10-2.5 (“course”)

0.7% 21.0%

5.8%

59.5%

PM2.5 (“fine”)

4.4% 30.1% 30.0%

13.0%

22.5%

13.0% 1.2% 5.7% 8.9%

PM0.1 (“ultrafine”)

71.2%

Ammonium nitrate Ammonium sulfate Crustal OC EC

FIGURE 10.3 Average PM10–2.5, PM2.5, and PM0.1 (ultrafine) chemical composition at an EPA “supersize” monitor in Los Angeles, CA, October 2001 to September 2002. Components shown in clockwise order (starting with ammonium nitrate) as listed in legend from top to bottom. Source: EPA (2006a).

the “coarse fraction” (PM10–2.5) is intermediate between that of the larger coarse particles and fine particles. Thus, some of the smaller coarse fraction particles may have lifetimes on the order of days and travel distances of up to 100 km or more. In some locations, source distribution and meteorology affect the relative homogeneity of fine and coarse particles, and in some cases, the greater measurement error in estimating coarse fraction mass precludes clear conclusions about relative homogeneity. Nevertheless, because fine particles remain suspended for longer times (typically on the order of days to weeks as opposed to days for coarse fraction particles) and travel much farther (i.e., hundreds to thousands of kilometers) than coarse fraction particles (i.e., tens to hundreds of kilometers), all else being equal, fine particles are theoretically likely to be more uniformly dispersed across urban and regional scales than coarse fraction particles. In contrast, coarse particles tend to be less evenly dispersed around urban areas and exhibit more localized elevated concentrations near sources, especially under windy conditions.

10.2 AMBIENT AIR PM CONCENTRATIONS This discussion focuses primarily on the concentrations of thoracic PM (PM10) and fine PM (PM2.5), on the basis that health considerations are our primary concern. This has the unfortunate effect of limiting our more detailed examination of temporal trends to the period beginning in the 1980s when such size-selective concentrations measurements were first

324

AMBIENT AIR PARTICULATE MATTER

FIGURE 10.4 Total suspended particulates in New York City (1930–1976). Date assembled by Eisenbud (1980).

made. There are, however, times when historic measurements of less current relevance can be useful for gaining a holistic appreciation of the effects of ambient PM on human health, and this section begins with a brief review of that experience of most relevance to today’s concerns about excess mortality and morbidity in relation to ambient PM. Up until the mid-1980s, available PM concentrations in the United States were generally measured gravimetrically as TSP. Because TSP includes, and can be dominated by, particles too large to penetrate into the thorax, its mass concentration is a poor index of inhalation hazard. Since the dispersion of large particles is limited, proximity of the sampler to local sources of dust has a major influence on measured TSP concentrations. The artifacts also vary with season and climate, and can be especially severe in the arid portions of the western United States. With these limitations in mind, it is still useful, in terms of historical perspective, to examine the major downward trend in TSP in large U.S. cities, such as New York City, between the 1930s when coal was used for domestic and commercial building heating, to the 1970s, when coal had been almost completely replaced by light oil and gas as fuels for such purposes. The approximately fivefold reduction in TSP that occurred by 1970 is illustrated in Fig. 10.4, as well as the further reduction into the late 1970s. Further reductions in both PM10 and PM2.5 have occurred since then, as illustrated in Fig. 10.5 for PM2.5 and its components in rural sites in western and eastern U.S. sites, and in Washington, DC. 10.2.1

Nongravimetric PM Monitoring

Some PM monitoring systems have determined PM concentrations by measuring the optical properties of the particles collected on a filter in terms of light reflectance (as for black smoke (BS) or “British smoke,” or for a similar index known as KM), or in terms of light transmission through the filter (coefficient of haze (CoH)). The use of such optical metrics can lead to underestimation of the mass concentration of light-colored ash particles, and/or to overestimation of sample masses containing diesel engine soot. This complicates the interpretation of epidemiological data based upon exposures determined using these optically based measurements. On the contrary, when a major change in pollution sources is taking place, there are generally parallel reductions in all of the pollutants affected by those sources. Figure 10.6

AMBIENT AIR PM CONCENTRATIONS

325

FIGURE 10.5 Annual average concentration of PM2.5 and components at background sites in the western United States, eastern United States, and in Washington, DC, 1993–2003.

shows that, as coal smoke came under control in London, England, between 1960 and 1980, there were essentially proportionate reductions in BS, SO2, and the sulfuric acid content of ambient air PM. Furthermore, even when there are relatively precise gravimetric measurements of sampled particles, there can be significant measurement artifacts. Positive artifacts can occur when the filters, or the particles collected on the filters, extract gas phase pollutants from the sampling stream by chemical reaction or sorption. Negative artifacts can occur when the sampled particles are volatilized and carried away from the filter by the sampling stream.

326

AMBIENT AIR PARTICULATE MATTER

FIGURE 10.6 Long-term trends in annual mean atmospheric conditions of BS and SO2 at seven stations in Greater London, and annual mean concentration of H2SO4 at St. Bartholomew’s Hospital in Central London.

Semivolatile constituents of the ambient air PM include ammonium nitrate (NH4NO3) and some organics formed by photochemical reactions. In southern California, especially on hot summer days, particle volatilization can account for substantial underestimations of PM10 concentrations, and even greater underestimations of PM2.5 concentrations, since the more volatile components are largely within the PM2.5 fraction. Despite the inherent limitations of: (1) the assumption of equivalent toxicity of all sampled particles; and (2) the sampling and analytical artifacts that limit the accuracy and precision of measured PM concentrations, there is a substantial body of epidemiological evidence for statistically significant associations between airborne PM concentrations and excess mortality and morbidity. Furthermore, the mortality and morbidity effects appear to be coherent, and not explicable on the basis of known potential confounding factors or coexisting gas phase pollutants (Bates,1992; Pope et al., 1995a; U.S. EPA, 2006a). The nature of this evidence will be discussed following a discussion of ambient air quality data and the relationships between ambient air PM concentrations and actual human exposures to PM.

10.3 EXTENT OF POPULATION EXPOSURES TO AMBIENT AIR PM The concentrations of constituents of PM in the ambient air are important determinants of human exposure to PM of outdoor origin, but other factors also greatly influence exposure. For exposures occurring indoors, where most people spend most of their time, these include: (1) limited penetrability to indoor spaces; (2) removal to indoor surfaces; and (3) chemical transformations. Each of these factors tends to reduce exposures to particles of outdoor origin for people spending time indoors.

EXTENT OF POPULATION EXPOSURES TO AMBIENT AIR PM

327

The penetrability of particles into the indoor environment varies with the air exchange between outdoors and indoors, which, in turn, varies with building size, type of construction, heating and cooling systems, and wind velocity. There will also be variations by season, with minimal air exchange in mid-winter and, for air-conditioned homes, in mid-summer as well. Particle size affects penetrability, especially when infiltration pathways are reduced in order to save energy. Under such conditions coarse particle penetration can be greatly reduced. Once PM penetrates indoors, particle size and chemical composition become major determinants of its fate. Coarse particles deposit by sedimentation relatively rapidly under the generally quiescent conditions indoors. Ultrafine particles diffuse to and deposit on the walls and other indoor surfaces. Acidic particles will be neutralized by ammonia released into the indoor air by people, pets, and household products. Thus, the indoor/outdoor concentration ratio can be close to unity for a component such as SO4 ¼ , which is: (1) present in the ambient air almost entirely in the accumulation mode (0.1–1 mm); (2) is chemically nonreactive; and (3) has no indoor sources in most circumstances. This is illustrated by the data from EPA’s Particle Total Exposure Assessment Methodology Study (PTEAM) in Riverside, CA, that are shown in Fig. 10.7, where the ratio of personal to outdoor SO4 ¼ was 0.78 þ 0.02. For a chemically reactive component, such as hydrogen ion (Hþ), the ratio was much lower. For many constituents of outdoor PM, there are significant indoor sources, and the ratio of personal exposure to outdoor concentration for many substances can be much greater than unity. Major PM sources indoors include smoking and cooking, as well as cleaning activities that release or resuspend PM such as dusting, sweeping, and conventional vacuum cleaning. Furthermore, there is a major source associated with the personal cloud created by each of us as we engage in routine activities whereby our motions resuspend settled dust from floors, furniture, and other surfaces. Additional data from the PTEAM study showed that elements enriched in settled dust can be enriched in the personal cloud by a factor of about 2. On

FIGURE 10.7 Personal versus outdoor SCV. Open circles represent children living in air-conditioned homes; the solid line is the 1_1 line. Source: U.S. EPA (1996a).

328

AMBIENT AIR PARTICULATE MATTER

average, personal activities were associated with 37% of overall personal PM exposure, as compared to 21% for identifiable indoor sources (cooking, environmental tobacco smoke, and other), and 42% for PM of outdoor origin. There are few such extensive datasets for locations other than Riverside, CA, but less detailed data from other studies elsewhere indicate that the PTEAM results are not atypical (U.S. EPA, 1996a). Important lessons from such studies are that gravimetric PM concentrations measured indoors may bear little relation to ambient mass PM concentrations, and that the compositions of indoor and outdoor PM can be very different. If the objective is to determine total exposure to PM of outdoor origin, then it may be best to use a conservative tracer of PM from outdoor sources, such as sulfate (by ion chromatography) or sulfur (by X-ray fluorescence analysis), along with data on the ratio of SO4 ¼ (or S) to ambient PM2.5 or PM10.

10.4 NATURE OF THE EVIDENCE FOR HUMAN HEALTH EFFECTS OF AMBIENT AIR PM For ambient air PM, there is abundant evidence, to be discussed, for significant associations between elevated concentrations and excess human mortality, morbidity, and lost-time, as well as reduced lung function. However, except under the most extreme exposure conditions, such as in Donora, PA, in 1948 (Schrenk et al., 1949), only a small percentage of the population has been observed to suffer these adverse effects. Furthermore, the affected populations are generally limited to those who are very young or elderly, and within these groups, those with preexisting disease or special sensitivity. In this context, the lack of confirming data and mechanistic understanding of the basis for the adverse effects from controlled human and animal inhalation studies should come as no great surprise. Short-term human exposure studies can seldom be ethically conducted on especially vulnerable subjects, and chronic long-term exposure studies on humans that may produce cumulative damage are both unethical and technically infeasible. Controlled inhalation studies on laboratory animals, which have enhanced sensitivity have recently been initiated, using concentrated ambient air PM (CAPs), and the results of some of these studies, to be discussed later in this chapter, have produced mortality and functional changes that are relevant to the epidemiological findings. There is only one component of ambient air PM that has produced functional responses that can readily be related to some of the epidemiologic results following controlled inhalation exposures in healthy humans and laboratory animals, and that component is Hþ. These studies are discussed in detail in Chapter 23 on the sulfur oxides. Other chapters that discuss the biological effects that can be produced by normal components of ambient air PM include Chapter 12 (Mineral Fibers), 16 (Diesel Exhaust), 20 (Lead), and 22 (Nitrogen Oxides). However, the ambient air PM concentrations for each of these constituents are almost certainly much too low to account for the effects that have been associated with ambient PM. It is possible that the effects associated with ambient air PM are initiated by nonspecific responses or reactions to deposited particles by lung epithelial cells that can be triggered by most, if not all, deposited particles. In that case the appropriate index of challenge may be associated with the number of depositing particles, or the number having a surface area or volume greater than some threshold level. More research is needed before such speculations can be confirmed or refuted.

EPIDEMIOLOGICAL EVIDENCE FOR HUMAN HEALTH EFFECTS OF AMBIENT AIR PM

329

10.5 EPIDEMIOLOGICAL EVIDENCE FOR HUMAN HEALTH EFFECTS OF AMBIENT AIR PM 10.5.1

Mortality Associated with Coal Smoke

Quantitative information on adverse health effects associated with particulate matter dates back to the London episode of 1873. A summation of bronchitis mortality during and following the December 9–11, 1873, fog episode was tabulated in the Ministry of Health (1954) report of the subsequent December 5–9, 1952, episode. As shown in Table 10.2, various nineteenth-century fog episodes produced excesses in bronchitis deaths that were comparable to that reported for the more famous 1952 episode. Also, it is important to note the higher baseline bronchitis mortality for London in the late nineteenth century, when the population was below three million (compared to about eight million in 1952), and at a time when that cigarette smoking could not have been a contributory cause. Two other pollution episodes that produced excess mortality occurred outside the United Kingdom before December 1952. These were reported by Firket (1936) on the December 1930 fog episode in the Meuse Valley in Belgium, and by Schrenk et al. (1949) on the October 1948 Donora, PA, fog episode. The December 1930 fog in the Meuse Valley was associated with 60 deaths from a population of 6000, but the pollutant concentrations were not measured. In Donora, a valley town of about 10,000 people at a bend in the Monongahela River south of Pittsburgh, there were steel mills, wire mills, zinc works, and a sulfuric acid plant along the riverbank for the entire length of the town. As reported by Schrenk et al. (1949), a persistent valley fog was associated with 20 excess deaths as well as acute morbidity among 43% of the population. About 10% were reported to have severe effects requiring medical attention. In a 10-year follow-up of the affected population, Ciocco and Thompson (1961) reported greater mortality rates and incidences of heart disease and chronic bronchitis among the residents who had reported acute illness in 1948 in comparison to residents who did not report such illness. As shown in Fig. 10.8, the daily death rate rose rapidly with the onset of the London fog on December 5, 1952, and peaked one day after the peak of pollution, as it was indexed by the measured pollutants, that is, black smoke (BS) and sulfur dioxide (SO2). There was also a rise in hospital emergency bed admissions, which peaked 2 days after the pollutant peaks. Both the deaths and hospital bed admissions remained elevated for several weeks after the fog lifted (see Table 10.2). Note also that hospital admissions exhibited declines on Sundays, a finding consistent with the known practices for hospital admissions. The Ministry of Health (1954) report attributed an excess of 4000 deaths from all causes to the exposures during the 1952 episode. Deaths peaked in the first full week, and were still above baseline levels 2 weeks after that. The specific cause with the greatest number of excess deaths over the 4 weeks was bronchitis (1156 excess deaths) and it had the greatest relative risk (RR ¼ 6.67). The next greatest increase, for heart disease (737 excess deaths), had an RR of only 1.82. The all cause relative risk was somewhat higher (1.96). Most of the excess deaths occurred in individuals over 55 years of age (2616 excess deaths over 4 weeks), but there was an excess in deaths for all age groups beyond 4 weeks of age. Overall, the excess mortality was concentrated among the elderly with preexisting disease. A recent reanalysis of the December 1952 episode in London that extended the time interval during which there was excess mortality out to 3 months suggested that the overall impact of this episode included 12,000 excess deaths (Bell et al., 2004).

330 86

65

451

375

357

294

228

133 25–31 Jan. 258 29 Jan.–4 Feb. 14 20–26 Dec. 35 25–31 Dec. 55 21–27 Nov. 14 1–6 Dec. 3

7–13 Dec. 424 1–7 Feb. 939 5–11 Feb. 324 27 Dec.–2 Jan. 583 1–7 Jan. 208 28 Nov.–4 Dec. 84 7–13 Dec. 621

14–20 Dec. 129 8–14 Feb. 453 12–18 Feb. 186 3–9 Jan. 333 8–14 Jan. 154 5–11 Dec. 33 14–20 Dec. 308

21–27 Dec. 102 15–21 167 19–25 31 10–16 437 15–21 2 12–18 20 21–27 92

Dec.

Dec.

Jan.

Jan.

Feb.

Feb.

28 Dec.–3 Jan.

Excess Bronchitis Deaths in Week of Fog and During Succeeding 3 Weeks

1018

151

309

1388

555

1817

788

Total 4 Week Excess in Bronchitis Deaths

Source: Adapted from Ministry of Health (1954). Report #95 on Public Health and Medical Subjects. Mortality and Morbidity During the London Fog of December 1952. London, H.M. Stationery Office.

26 Nov.–1 Dec. 1948 5–9 Dec. 1952

28–30 Dec. 1892

21–24 Dec. 1891

2–7 Feb. 1882

26–29 Jan. 1880

Av. Weekly Bronchitis Mortality in Previous 10 Years

Excess Bronchitis Deaths Associated with Historic London Fogs

Dates of Fog 9–11 Dec. 1873

TABLE 10.2

EPIDEMIOLOGICAL EVIDENCE FOR HUMAN HEALTH EFFECTS OF AMBIENT AIR PM

331

FIGURE 10.8 Metropolitan London total mortality and emergency bed admissions during the 1952 pollution episode in relation to black smoke, expressed as mg/m3, and sulfur dioxide, expressed as parts per million by volume.

It is of particular interest to current concerns that while recent daily mortality studies show much lower absolute risk levels from the much lower peaks in PM pollution, the elevated relative risks among the very young and oldest cohorts and the risk rankings among causes of death are quite similar today to those of December 1952. The Ministry of Health (1954) report also noted that there was a clear association between chronic air pollution and the incidence of bronchitis and other respiratory diseases. The death rate from bronchitis in England and Wales (where coal smoke pollution was very high) was much higher than in other northern European countries (with much lower levels of coal smoke pollution). The very high chronic coal smoke exposure in the United Kingdom, associated with a high prevalence of chronic bronchitis, appears to have created a large pool of individuals susceptible to “harvesting” by an acute pollution episode. The December 1962 London fog episode was the last to produce a clearly evident acute harvest of excess deaths, albeit a much smaller one than that of December 1952. Commins and Waller (1963) developed a technique to measure H2SO4 in urban air, and made daily measurements of H2SO4 at St. Bartholomew’s Hospital in Central London during the 1962 episode. As shown in Fig. 10.9, the airborne H2SO4 rose rapidly during the 1962 episode, with a greater relative increase than that for BS. The UK Clean Air Act of 1954 had led to the mandated use of smokeless fuels and, as shown in Fig. 10.6, annual mean smoke levels had declined by 1962, to about one half

AMBIENT AIR PARTICULATE MATTER

5000

Total deaths per day

Deaths 400

4000

300

3000 Acid

200

2000 SO2 1000

100

500 400 300 200

Acid (μg/m3)

500

Smoke and SO2 (μg/m3)

332

100

Smoke 0

0 1

2

3

4 5 6 7 8 December, 1962

FIGURE 10.9

0

9 10

London pollution episode, December 1962.

of the 1958 level. The annual average SO2 concentrations had not declined by 1962, but dropped off markedly thereafter, along with a further marked decline in BS levels. For the period between 1964 and 1972, the measured levels of H2SO4 followed a similar pattern of decline. During the later part of the coal smoke era in the United Kingdom, researchers began to study the associations between long-term daily records of mortality and morbidity and ambient air pollution. In the first major time-series analysis of daily London mortality for the winter of 1958–1959, Martin and Bradley (1960) and Lawther (1963) used the readily available BS and SO2 data. They estimated that both pollutants were associated with excess daily mortality when their concentrations exceeded about 750 mg/m3. However, additional analyses of this dataset led to different conclusions. For example, Ware et al. (1981) concluded that there was no demonstrable lower threshold for excess mortality down to the lowest range of observation (BS  150 mg/m3), as illustrated in Fig. 10.10. Although 150 mg/m3 is now near the upper end of observed concentrations rather than at the lower end, time-series analyses still indicate an increasing slope as concentrations decrease.

Daily mortality

+40

(5)

(10)

(6)

+20 (9)

(7)

0

(19) (18) ( )Indicates number of days during

(12)

-20

the winter with concentrations in the range surrounding the point

(6)

0

400

800

1200

BS (µg/m3), 24 h average

FIGURE 10.10 Martin and Bradley (1960) data for winter of 1958–1959 in London as summarized by Ware et al. (1981), showing average deviations of daily mortality from 15-day moving average by concentration of black smoke (BS).

EPIDEMIOLOGICAL EVIDENCE FOR HUMAN HEALTH EFFECTS OF AMBIENT AIR PM

333

TABLE 10.3 Standardized Annual Mortality Rate Regression Coefficients on Smokea for 64 UK County Boroughs Cancer of Trachea, Bronchus, and Lung

Chronic Bronchitis

1969–1973 1958–1964 1948–1954 1969–1973 1958–1964 1948–1954

0.07 0.53** 0.71*** 0.15 0.68*** 0.87***

0.02 0.32* 0.48*** 0.06 0.31 0.37*

45–64

1969–1973 1958–1964 1948–1954

0.02 0.64** 0.49*

0.02 0.33* 0.49**

65–74

1969–1973 1958–1964 1948–1954

0.07 0.25 0.61**

0.03 0.40* 0.31

Sex

Ages

Males

45–64

65–74

Females

Mortality in Year

Source: Adapted from Chinn et al. (1981). a

Based on index of black smoke pollution 20 years before death of Daly (Br. J. Prev. Soc. Med. 13: 14–27, 1959). p < 0.05. ** p < 0.01. *** p < 0.001. *

The marked reduction in UK smoke pollution levels during the 1960s by means of changes to smokeless fuels was shown to be associated with a marked reduction in annual mortality in County Boroughs by Chinn et al. (1981). As shown in Table 10.3, mortality rates in middle-aged and elderly men and women for the 1969–1973 period were no longer associated with an index of smoke pollution. By contrast, for both the 1948–1954 and 1958– 1964 periods, the index of smoke exposure correlated strongly with annual mortality rates for both chronic bronchitis and respiratory tract cancers. On the basis of such evidence of improved health status, UK colleagues considered air pollution to be a problem solved, essentially halted further investigations for the next several decades, and concluded that air quality standards were not needed. 10.5.2

Morbidity Associated with Coal Smoke

In terms of time-series analyses of morbidity, a study by Lawther et al. (1970) reported the daily symptom scores of a panel of patients with chronic bronchitis in relation to the daily concentrations of BS and SO2. There was a close correspondence between symptom scores and both pollutant indices. Chronic coal smoke exposure also affected baseline lung function. Holland and Reid (1965) analyzed spirometric data collected on British postal workers in 1965. By that time, pollution levels were well below their peaks, but the postal workers had been exposed out-ofdoors for many years when pollution levels were higher. The London postal workers had lower forced expiratory volumes in one second (FEV1) and peak expiratory flow rates (PEFR) than their country town counterparts. The deleterious effects of smoking were

334

AMBIENT AIR PARTICULATE MATTER

accounted for in these analyses. Within each smoking category, the differences between the London and country town means were attributed to pollution on the basis that pollution levels were, on average, twice as high in London as in the country towns. Mortality and Morbidity Associated with Other PM Indices With the phasing out of bituminous coal as a fuel for domestic heating, the use of the optical density of smoke samples as an index of the health risk associated with ambient particulate matter became increasingly problematic. It has also become clear that TSP, the standard index of PM pollution in the United States prior to 1987, was also far from ideal. Under high wind conditions, gravimetric TSP concentrations are dominated by PM too large to penetrate into the human thorax, even during oral inhalation. Some U.S. investigators chose the SO4 ¼ content of TSP samples as an alternate index of PM-associated health risk. Because of the nature of its sources, essentially all of the SO4 ¼ in the ambient air is on fine particles below 2.5 mm in aerodynamic diameter (PM2.5). As discussed in Chapter 23, SO4 ¼ is often a relatively large fraction of PM2.5, is nonvolatile, is stable on filters used for air sampling, can be easily extracted from the filters, and can be accurately analyzed with relatively simple and inexpensive procedures. Furthermore, it generally correlates with mortality and indices of morbidity as well as, or better than, other frequently measured PM indices, such as TSP, BS, CoH, and PM10. 10.5.3

Health Effects Associated with TSP

As indicated in Fig. 10.4, TSP levels in New York City had declined markedly between the 1950s and 1970s, and similar declines had taken place in most other urban areas. Thus, indices of chronic effects of exposure to PM made in the early 1970s for U.S. populations could have been due to earlier and higher exposures as well as to contemporary levels of exposure. This is relevant background for a discussion of results obtained during the first National Health and Nutrition Examination Survey (NHANES I), which was conducted between 1971 and 1975. In one paper, Chestnut et al. (1991) studied the relationship between pulmonary function and quarterly average levels of TSP for adults who resided in 49 of the locations where the NHANES I was conducted. Statistically significant relationships were observed between TSP levels and FVC and FEV1.0. Anthropometric measurements and socioeconomic characteristics of the subjects were included in the analysis, and the sample was restricted to “never” smokers. A 1 standard deviation increase (about 34 mg/m3) in TSP from the sample mean of 87 mg/m3 was associated with an average decrease in FVC of 2.25%. In another report of results from NHANES I, Schwartz (1993) examined reported rates of chronic respiratory illness by standardized questionnaire across 53 urban areas in the United States. Diagnosis of respiratory illness by an examining physician was also considered as an outcome. After controlling for age, race, sex, and cigarette smoking, a 10 mg/m3 increase in annual average TSP was associated with increased risk of chronic bronchitis (odds ratio (OR) ¼ 1.07, 95% confidence interval (CI) ¼ 1.02–11.2) and of a respiratory diagnosis by the examining physician (OR ¼ 1.06, 95% CI ¼ 1.02–1.11). When the analysis was restricted to never smokers, the associations remained, with a slight increase in the relative odds. In a study of the influence of chronic PM exposure on lung function of children and young adults (ages 6–24 years), Schwartz (1989) used TSP and NHANES II data for the period 1976–1980. TSP, NO2 and O3 were all significantly associated with reduced FVC, FEV1 and PEFR, but SO2 was not. For TSP and O3, there appeared to be thresholds for response at

EPIDEMIOLOGICAL EVIDENCE FOR HUMAN HEALTH EFFECTS OF AMBIENT AIR PM

335

40 ppb (daily average) O3 and 90 mg/m3 TSP. The relationships held whether or not children with respiratory conditions, or smokers were included. Demographic and geographic variables had little or no impact on the pollution relationships, which also held when only persons still residing in their state of birth were considered. 10.5.4

Time-Series Mortality and Morbidity Studies Based on PM10 and PM2.5

During the 1990s, there was a great increase in the number of peer-reviewed papers describing time-series studies of the associations between daily ambient air pollutant concentrations and daily rates of mortality and hospital admissions for respiratory diseases. In terms of morbidity, there was also been a rapid growth of the literature showing associations between airborne particle concentrations and exacerbation of asthma, increased symptom rates, decreased respiratory function and restricted activities. Much of the literature from the early 1990s was summarized by Pope et al. (1995a). They converted historically measured values for CoH and TSP to estimated levels of PM10, and remarked that very similar coefficients of response for the PM10-daily mortality associations were determined in all locations. In Thurston’s (1995) analysis of acute mortality studies in nine communities with measured PM10 concentrations, including 4 of the 10 studies cited by Pope et al. (1995a), the coefficients of response tended to be higher when the PM10 was expressed as a multiple-day average concentration, and lower when other air pollutants were included in multiple-regression analyses. In any case, the results in each city (except for the very small city of Kingston, TN) indicated a statistically significant association. The number of time-series studies of the associations between daily mortality and morbidity, and ambient air concentrations of PM10, PM2.5, and PM10–2.5, has grown substantially in recent years. There are still more studies involving PM10 than its fine and coarse components, but both size ranges appear to make similar contributions to the PM10 associations. Time-series studies of the associations between daily cardiovascular and respiratory mortality and PM10 were carried out in 29 European cities by Analitis et al. (2006). An increase in PM10 by 10 mg/m3 (lag 0 þ 1) was associated with increases of 0.76 (CI: 0.47– 1.05%) in cardiovascular deaths and 0.58% (CI: 0.21–0.95%) in respiratory deaths. In the first study to examine the roles of the chemical components of PM on daily mortality, Laden et al. (2001) performed a source apportionment on the PM2.5 in the SixCities Study, and reported that the mobile-source component accounted, per 10 in PM2.5, for a 3.4% (CI: 01.7–5.2%) increase in daily mortality, while the coal combustion source accounted for a 1.1% (CI: 0.3–2.0%) increase, while crustal materials were not associated with excess daily mortality. The CI are smaller for PM10, largely due to more years of measurement data than either of its components, and to the greater measurement error for PM10–2.5. It is noteworthy that, consistent with the coherence argument of Bates (1992), the excess risks for morbidity are higher than for mortality. Also, the relative risks (RRs) for respiratory mortality are greater than for total mortality, and the RRs for the less serious events are higher than those for mortality and hospital admissions. In its 1996 PM Criteria Document, EPA concluded that the associations between PM10 and daily mortality were not seriously confounded by weather variables or the presence of other criteria pollutants. Subsequent research, summarized in the 2004 PM criteria document (CD), led EPA to somewhat soften this position with respect to the weather variables.

336

AMBIENT AIR PARTICULATE MATTER

FIGURE 10.11 Marginal posterior distributions for effect of PM10 on total mortality at lag 1, with and without control for other pollutants, for the NMMAPS 90 cities. The numbers in the upper right legend are the posterior probabilities that the overall effects are greater than 0. Source: Dominici et al. (2003b).

Figure 10.11, from the 2004 PM CD, shows that the calculated relative acute mortality risks for PM10 are relatively insensitive to the concentrations of SO2, NO2, CO, and O3. 10.5.5

Time-Series Studies of Cardiac Function Based on PM2.5 and PM10–2.5

Recent studies have examined associations between cardiac function and PM concentrations on the basis that PM-associated mortality and morbidity can be explained, at least in part, by alterations in cardiac autonomic balance, as measured by heart rate variability (HRV). Associations between decreased HRV and increased PM2.5 in ambient air (Creason et al., 2001; Gold et al., 2000; Holguin et al., 2003; Park et al., 2005; Pope et al., 1999, 2004a, 2004b; Schwartz et al., 2005). Riediker et al. (2004) reported that HRV was significantly associated with PM2.5 concentration for young highway patrol troopers inside their patrol cars. While Gold et al. (2000) and Liao et al. (1999) found no such association for PM10–2.5 in communities with low concentrations of PM10–2.5, Lipsett et al. (2006) found that PM10–2.5 and PM10, but not PM2.5 were associated with reduced HRV in a California community where PM10–2.5 was higher than PM2.5. While there is mounting evidence that excess daily mortality, morbidity, and cardiac function are associated with short-term peaks in PM10 and PM2.5 pollution, the public health implications of this evidence are not yet fully clear. Key questions remain the following: .

.

.

Which specific components of the PM2.5 and PM10–2.5 are most influential in producing the responses? Do the effects of the PM depend on coexposure to irritant vapors, such as ozone, sulfur dioxide, or nitrogen oxides? What influences do multiple day pollution episode exposures have on daily responses and response lags?

EPIDEMIOLOGICAL EVIDENCE FOR HUMAN HEALTH EFFECTS OF AMBIENT AIR PM .

.

337

Does long-term chronic exposure predispose sensitive individuals to being “harvested” on peak pollution days? How much of the excess daily mortality is associated with life shortening measured in days or weeks versus months, years, or decades?

The first four questions above are complex, and difficult to answer at this time on the basis of current knowledge. The discussion that follows will examine them in greater detail. 10.5.6

Effects of Chronic Exposures to Ambient Air PM

The last question above is a critical one in terms of the public health impact of excess daily mortality. If, in fact, the bulk of the excess daily mortality were due to “harvesting” of terminally ill people who would have died within a few days, then the public health impact would be much less than if it led to prompt mortality among acutely ill persons who, if they did not die then, would have recovered and lived productive lives for years or decades longer. An indirect answer to this question is provided by the results of prospective cohort studies of annual mortality rates in relation to long-term pollutant exposures in the Harvard Six-Cities Study (Dockery et al., 1993; Laden et al., 2006), the American Cancer Society (ACS) cohort (Pope et al., 1995b, 2002, 2004a, 2004b), the AHSMOG study (Adventist Health Study on Smog-1998; Abbey et al., 1999; McDonnell et al., 2000), and the Veterans study (Lipfert et al., 2000b). The Six-Cities and the ACS cohort studies are considered to have the greater strengths and are discussed most fully in the text that follows. The limitations of these key studies and the AHSMOG and Veterans studies are addressed in the text that follows. Dockery et al. (1993) reported on a 14–16 year mortality follow-up of 8111 adults in six U.S. cities in relation to average ambient air concentrations of TSP, PM2.5, fine particle SO4 ¼ , O3, SO2, and NO2. Concentration data for most of these pollutant variables were available for 14–16 years. The mortality rates were adjusted for cigarette smoking, education, body mass index, and other influential factors not associated with pollution. The two pollutant variables that best correlated with total mortality (which was mostly attributable to cardiopulmonary mortality) were PM2.5 and SO4 ¼ . The overall mortality rate ratios were expressed in terms of the range of air pollutant concentrations in the six cities. The rate-ratios (and 95% confidence intervals) for both PM2.5 and SO4 ¼ were 1.26 (1.08–1.47) overall, and 1.37 (1.11–1.68) for cardiopulmonary. The mean life shortening was in the range of 2–3 years. Pope et al. (1995b) linked SO4 ¼ data from 151 U.S. metropolitan areas in 1980 with individual risk factor on 552,138 adults who resided in these areas when enrolled in a prospective study in 1982, as well as PM2.5 data for 295,223 adults in 50 communities. Deaths were ascertained through December 1989. The relationships of air pollution to allcause; lung cancer; and cardiopulmonary mortality were examined using multivariate analysis, which controlled for smoking, education, and other risk factors. PM air pollution was associated with cardiopulmonary and lung cancer mortality, but not with mortality due to other causes. Adjusted relative risk (RR) ratios (and 95% confidence intervals) of all-cause mortality for the most polluted areas compared with the least polluted equaled 1.15 (1.09–1.22) and 1.17 (1.09–1.26) when using SO4 ¼ and PM2.5 respectively. The mean life shortening in this study was between 1.5 and 2 years. The results (for SO4 ¼ ) were similar to those found in the previous cross-sectional studies of Ozkaynak and Thurston (1987) and Lave and Seskin (1970). Thus, the results of these earlier studies provided some confirmatory

338

AMBIENT AIR PARTICULATE MATTER

Fine particles divided into sulfate and nonsulfate particles 1.3

1.3

Risk ratio

S

Thoracic particles divided into fine and coarse particles

1.3

1.3 Risk ratio

S

1.2 1.1 1.0 P

1.2 1.1 1.0

T

P

L W T

30 50 70 90 Total particles, (µg/m3)

Risk ratio

S

0.9

1.2

L W

P

1.0

H

W TP

1.2 1.1

L

H

W

1.0 P T 0.9

5 9 13 17 Nonsulfate fine particles, (µg/m3)

L

W P

1.0

S

1.2 1.1

1.1

T

0.9

H

1.1 1.0

W

1.0 PT

S

30 50 70 90 Thoracic particles, (µg/m3) 1.3

H L

0.9 4 6 8 10 12 Fine sulfate particles, (µg/m3) 1.3

H

10 15 20 25 30 Fine Particles, (µg/m3) 1.3

W

0.9

H

L

1.1

0.9

H L

Risk ratio

Total particles

Risk ratio

S

1.2

Risk ratio

Divided into thoracic and nonthoracic particles

Risk ratio

S

1.2

6 10 14 Coarse particles, (µg/m3)

T

0.9 30 50 70 90 Nonthoracic particles, (µg/m3)

FIGURE 10.12 Adjusted relative risks for annual mortality are plotted against each of seven longterm average particle indexes in the Six-City study, from largest size range––total suspended particulate matter (lower left)––through sulfate and nonsulfate fine particle concentrations (upper right). Note that a relatively strong linear relationship is seen for fine particles, and for its sulfate and nonsulfate components. Topeka, which has a substantial coarse particle component of thoracic particle mass, stands apart from the linear relationship between relative risk and thoracic particle concentration. Source: Adapted from Fig. V-5 of PM Staff Paper; U.S. EPA (1996b).

support for the findings of Pope et al. (1995b), while the Pope et al. (1995b) results indicated that the concerns about the credibility of the earlier results, due to their inability to control for potentially confounding personal factors such as smoking and socioeconomic variables, could be eased. The Dockery et al. (1993) study had the added strength of data on multiple PM metrics. As shown in Fig. 10.12, the association became stronger as the PM metric shifted from TSP to PM10. Within the PM10, the association was much stronger forPM2.5 than for the coarse component. Within the PM2.5 fraction, both the SO4 ¼ and non-SO4 ¼ fractions correlated very strongly with annual mortality, suggesting a nonspecific response to PM2.5. The findings of Dockery et al. (1993) and Pope et al. (1995b), in carefully controlled prospective cohort studies, indicating that mean lifespan shortening is of the order of 2 years, implied that many individuals in the population had lives shortened by many years, and that there is excess mortality associated with fine particle exposure greater than that implied by the cumulative results of the time-series studies of daily mortality. An extensive reanalyses (Krewski et al., 2000) of the Six-Cities and ACS studies indicated that the published findings of the original investigators (Dockery et al., 1993; Pope et al., 1995b) were based on substantially valid datasets and statistical analyses. Krewski et al. (2000) investigated, via sensitivity analyses (in effect, new analyses), the Six-Cities and ACS studies’ datasets, including consideration of a much wider range of confounding variables.

EPIDEMIOLOGICAL EVIDENCE FOR HUMAN HEALTH EFFECTS OF AMBIENT AIR PM

339

Subsequent analyses of ACS data for more extended time periods (Pope et al., 2002) further substantiated the original findings, and also provided much clearer, stronger evidence for ambient PM exposure relationships with increased lung cancer risk previously indicated for the AHSMOG study by (Beeson et al., (1998). For the Six-Cities Study, Krewski et al. (2000) reported the excess relative risk of mortality from all causes associated with an increase in PM2.5 of 10 mg/m3 to be 14%, close to the 13% reported by the original investigators. For the ACS Study, they reported the relative risk of all-cause mortality associated with a 10 mg/m3 increase in PM2.5 to be 7.0% in the reanalysis, close to the original 6.6% value. The part II sensitivity analysis of Krewski et al. (2000) applied an array of different models and variables to determine whether the original results would remain robust to different analytic assumptions and model specifications. The reanalysis team first applied the standard Cox model used by the original investigators and included variables in the model for which data were available from both original studies, but had not been used in the published analyses (e.g., physical activity, lung function, marital status). The reanalysis team also designed models to include interactions between variables. None of these alternative models produced results that materially altered the original findings. Next, for both the Six-Cities and ACS studies, the reanalysis team investigated the possible effects of fine particles and sulfate on a range of potentially susceptible subgroups of the population. These analyses did not find differences in PM-mortality associations among subgroups based on various personal characteristics (e.g., including gender, marital status, smoking status, and exposure to occupational dusts and fumes). However, estimated effects of fine particles did vary with educational level: the association between an increase in fine particles and mortality tended to be higher for individuals without a high school education than for those with more education. The reanalysis team found little evidence that questionnaire variables had led to confounding in either study, thereby strengthening the conclusion that the observed association between fine particle air pollution and mortality was not the result of a critical covariate that had been neglected by the original investigators. For the ACS Study, the reanalysis team tested whether the relationship between ambient concentrations and mortality was linear. They found some indications of both linear and nonlinear relationships, depending upon the analytic technique used, suggesting that the shapes of the concentration–response relationships warrant additional research in the future. For the Six-Cities Study, however, when the general decline in fine particle levels over the monitoring period was included as a time-dependent variable, the association between fine particles and all-cause mortality was reduced (Excess RR ¼ 10.4% (CI: 1.5, 20)). Both the original ACS Study air quality data and the newly constructed dataset contained SO4 ¼ levels inflated by 50% due to SO2 collection by the filter. For the Six-Cities Study, the relative risks of mortality were essentially unchanged with adjusted or unadjusted sulfate. For the ACS Study, adjusting for sulfate artifact resulted in slightly higher RRs for all-cause mortality and from cardiopulmonary disease compared with unadjusted data, while the RR for mortality from lung cancer was lower after data adjustment. Thus, the Reanalysis Team found essentially the same results as the original Harvard Six-Cities and ACS studies, even after using independently developed pollution datasets and adjustment for sulfate artifact. The reanalysis team conducted most of its sensitivity analyses using only the ACS Study dataset that considered 151 cities. When a range of city-level (ecologic) variables (e.g., population change, measures of income, maximum temperature, number of hospital beds, water hardness) was included in the analyses, the results generally did not change. The only

340

AMBIENT AIR PARTICULATE MATTER

exception was that associations with PM2.5 and SO4 ¼ were reduced when city-level measures of population change or SO2 were included in the model. Pope et al. (2002) extended the analyses of Pope et al. (1995a, 1995b) and the reanalyses of Krewski et al. (2000) of the ACS CPS-II cohort to include an additional 9 years of followup data. Advantages over the previous analyses include: (a) doubling the follow-up time from 7 to 16 years and tripling the number of deaths; (b) expanding the ambient air pollution data substantially, including two additional years of PM2.5 data and adding data on gaseous copollutants; (c) improving statistical adjustments for occupational exposure; (d) incorporating data on dietary covariates believed to be important factors in mortality, including total fat consumption, and consumption of vegetables, citrus fruit, and high-fiber grains; and (e) using recent developments in nonparametric spatial smoothing and random effects statistical models as input to the Cox proportional hazards model. In an attempt to estimate the concentration during this period, the integrated average of PM2.5 concentrations during 1999–2000 was averaged with the earlier 1979–1983 period. For the 51 cities where paired data were available, the concentrations of PM2.5 were lower in 1999–2000 than in 1979–1983 for most cities. Mean PM2.5 levels for the two periods were highly correlated (r ¼ 0.78), and the rank order of the cities by relative pollution levels remained nearly the same. The Pope et al. (2002) paper confirmed that the general pattern of findings for the first 7 years of the study (Pope et al., 1995b; Krewski et al., 2000) could be reasonably extrapolated to the patterns that remain present with twice the length of time on study and three times the number of deaths. PM2.5 mortality RRs also tended to be higher for those with less education, which may be due to related socioeconomic factors or, more likely, to the generally greater interstate mobility of higher-educated persons. Based on the above patterns of results, the authors drew the following conclusions: (1) The apparent association between long-term exposure to PM2.5 pollution and mortality persists with longer follow-up as the participants in the cohort grow older and more of them die. (2) The estimated PM2.5 effect on cardiopulmonary and cancer mortality remained relatively stable even after adjustment for smoking status. The estimates were relatively robust against inclusion of many additional covariates: education, marital status, body mass index (BMI), alcohol consumption, occupational exposure, and dietary factors. Education was an effect modifier, with larger and more statistically significant PM2.5 effect estimates for persons with less education. Because this cohort has a much higher percentage of well-educated persons than the general public, the education effect modification seen suggests that the overall PM2.5 effect estimates are likely underestimated by this study cohort than are likely to be found for the general public. (3) PM2.5 was associated with elevated total, cardiopulmonary, and lung cancer mortality risks, but not with other-cause mortality. PM10 for 1987–1996 and PM15 for 1979–1983 were just significantly associated with cardiopulmonary mortality, but neither PM10–2.5 nor TSP were associated with total, or with any cause-specific mortality. All end points but lung cancer mortality were very significantly associated with SO4 ¼ , except for lung cancer with 1990 SO4 ¼ data. All end points, except lung cancer mortality, were significantly associated with SO2 using 1980 data. None of the other gaseous pollutants showed significant positive associations with any end

EPIDEMIOLOGICAL EVIDENCE FOR HUMAN HEALTH EFFECTS OF AMBIENT AIR PM

341

point. Thus, neither coarse thoracic particles nor TSP were significantly associated with mortality; nor were CO and NO2 on a long-term exposure basis. (4) The excess risk from PM2.5 exposure is much smaller than that estimated for cigarette smoking for current smokers in the same cohort (Pope et al., 1995b): RR ¼ 2.07 for total mortality, RR ¼ 2.28 for cardiopulmonary mortality, and RR ¼ 9.73 for lung cancer mortality. In the more polluted areas of the United States, the relative risk for substantial obesity (a known risk factor for cardiopulmonary mortality) is larger than that for PM2.5, but the relative risk from being moderately overweight is somewhat smaller. Jerrett et al. (2005) studied the annual mortality rates in the members of the ACS cohort living in Los Angeles through the year 2000, using interpolated PM2.5 concentration data from multiple monitors in order to get better exposure estimates for members of this subcohort. They reported a RR of 1.17 for an increase of 10 mg/m3 in PM2.5, which was a much larger RR than that for the ACS cohort as a whole. Laden et al. (2006) extended the annual mortality analysis of the Six-Cities cohort by 8 years (till 1998), and reported that total, cardiovascular, and lung cancer mortality rates were each positively associated with ambient PM2.5 concentrations. There was a substantial reduction in PM2.5 concentrations in the later years, and the reductions in concentrations were associated with reduced mortality risks. Lipfert et al. (2006) examined the influence of PM2.5 components on annual mortality in the Veterans cohort, and reported that traffic density was the most important predictor of survival, with significant contributions from some specific PM2.5 components, that is, nitrate, elemental carbon (EC), nickel (Ni), and vanadium (V).

10.5.7 Relationship of Six-Cities, ACS, AHSMOG, and Veterans Cohort Study Findings The number of subjects in these studies varies greatly: 8111 subjects in the Six-Cities Study; 295,223 subjects in the 50 PM2.5 cities, and 552,138 subjects in the 151 SO4 ¼ cities of the ACS Study; 6338 in the AHSMOG Study; and 26,000 in the VA Study for PM2.5. This may partially account for differences among their results. The Six-City and AHSMOG studies were designed specifically as prospective studies to evaluate long-term effects of air pollution and included concurrent air pollution measurements. McDonnell et al. (2000) used PM10 as its PM mass index and found some significant associations with total mortality and deaths with contributing respiratory causes, even after controlling for potentially confounding factors (including other pollutants). In further evaluation of results found for PM10 among males, McDonnell et al. (2000) reported larger associations with PM2.5 than PM10–2.5 for males in the AHSMOG cohort, though none of the 11 PM2.5 associations reached statistical significance. For the Veterans study, few statistically significant associations were found with PM indicators; in fact, some statistically significant negative associations were reported for some subset analyses. There is no clear consistency in relationships among PM effect sizes, gender, and smoking status across these studies. The AHSMOG study cohort is a primarily nonsmoker group, while the VA study cohort had a large proportion of smokers and former smokers in an allmale population. The ACS results show similar and significant associations with total mortality for both “never smokers” and “ever smokers,” although the ACS cohort may

342

AMBIENT AIR PARTICULATE MATTER

include a substantial number of long-term former smokers with much lower risk than current smokers. The Six-Cities Study cohort showed the strongest evidence of a higher PM2.5 effect in current smokers than in nonsmokers, with female former smokers having a higher risk than male former smokers. This suggested that smoking status might be viewed as an effect modifier for ambient PM, just as smoking may be a health effect modifier for ambient O3 (Cassino et al., 1999). In considering the results of these studies together, statistically significant associations are reported between PM2.5 and mortality in the ACS and Six-Cities analyses, inconsistent but generally positive associations with PM were reported in the AHSMOG analyses, and distinctly inconsistent results were reported in the VA study. Based on several factors, the larger study population in the ACS study, the larger air quality data set in the Six-Cities Study, the more generally representative study populations used in the Six-Cities and ACS studies, and the fact that these studies have undergone extensive reanalyses ––the greatest weight should be placed on the results of the ACS and Six-Cities cohort studies in assessing relationships between long-term PM exposure and mortality. The results of these studies, including the reanalyses results for the Six-Cities and ACS studies, and the results of the ACS study extension, provide substantial evidence for positive associations between long-term ambient PM (especially PM2.5) exposure and mortality. Furthermore, a peer-reviewed EPAsponsored expert elicitation study (http://www.epa.gov//ttn/ecas/ria.html) that was focused on the size of the coefficient for excess annual mortality reported a consensus judgment that the best estimate for the coefficient was closer to those of the Six-Cities Study (Laden et al., 2006; Jerrett et al., 2005) than that of Pope et al. (2002) because they both used better exposure assessments and because the Six-Cities Study used a more representative population. Studies on the effects of long-term exposure have also been performed in European populations. Hoek et al. (2002) assessed the effect of traffic pollutants on in a Dutch cohort study of on diet and cancer in 5000 randomly selected people (ages 55–69 years). They found that traffic markers and black smoke were significantly associated with annual mortality. Gehring et al. (2006) studied long-term exposure to ambient air pollution in Germany among 4800 women participating in a series of cross-sectional studies. They reported that cardiopulmonary mortality was significantly associated with living within 50 m of a major road and with the concentrations of PM10 and NO2. Rosenlund et al. (2006) studied longterm exposure (over 30 years) to ambient air pollution in Sweden among first time cases of myocardial infarction in 1397 cases during 1992–1994 and 1870 matched controls. The OR for fatal MI for the 5th to 95th percentile difference in average exposure was 1.39 (CI: 0.94–2.07). Similar ORs were seen for CO and NO2. Brauer et al. (2006) studied the association of average concentrations of air pollutants in the first 2 years of life in the Netherlands and Germany with otitis media, a common childhood infection. Increases of 3 mg/m3 PM2.5, 0.5 mg/m3 EC, and 10 mg/m3 NO2 were associated with ORs of 1.13 (CI: 1.00–1.27), 1.10 (CI: 1.00–1.22), and 1.14 (CI: 1.03–1.27) in the Netherlands, and 1.24 (CI: 0.84–1.83), 1.10 (CI: 0.86–1.41), and 1.14 (CI: 0.87–1.49) in Germany. 10.5.8

Other Health Effects and Their Coherence with Excess Mortality

If, in fact, more people are dying of cardiopulmonary causes on a given day because of exposures to elevated concentrations of PM, it would be reasonable to expect higher daily rates of emergency hospital admissions and visits to emergency rooms and clinics for similar causes. This expectation is consistent with the results summarized in the latest PM Criteria Document (U.S. EPA, 2006). These studies indicate that indices of PM, such as daily

EPIDEMIOLOGICAL EVIDENCE FOR HUMAN HEALTH EFFECTS OF AMBIENT AIR PM

343

concentrations of PM10, SO4 ¼ , and BS are generally significantly associated with excess daily emergency admissions to hospitals for either respiratory diseases or cardiac diseases, or both. These studies have not shown associations with non-cardiopulmonary causes, and the influence of PM has generally been found to remain in multiple regression analyses that included other criteria pollutants. However, for respiratory diseases, the influence of summertime O3 has generally been greater than that of PM. This is in contradistinction to excess daily mortality, where the influence of PM is generally much greater than that of O3. For hospitalizations for cardiac diseases, the most influential criteria pollutants appear to be PM and CO. Further discussion on pollutant interactions and joint effects is provided in Chapters 23 and 25. PM2.5 is a risk factor for subnormal respiratory function in children as illustrated in Fig. 10.13, which shows data collected in the Children’s Health Study (CHS) in 12 Southern California communities (Gauderman et al., 2000a, 2000b, 2004). There were significant associations between the percentage of children with forced expiratory volume in one second (FEV1) < 80% of predicted and the mass concentrations of PM2.5, PM10, and elemental carbon (EC), but not of ozone (Gauderman et al., 2004). Most of the recent epidemiological studies have not had the advantage of available PM2.5, PM10, and PM10–2.5 component data and were limited to mass concentrations within those size ranges. Summaries of such epidemiology are shown in Fig. 10.14 for mortality and Fig. 10.15 for morbidity. There is coherence in the data, as defined by Bates (1992), in terms of the RR ratings, with mortality risks increasing from total to cardiovascular to respiratory, and with emergency department visits being more frequent than hospital admissions. Another aspect of the influence of PM on short-term mortality is its role in sudden infant death syndrome (SIDS). Woodruff et al. (1997) examined the relationship between postneonatal infant mortality (28 days to 1 year) and PM10 in the United States. The study of approximately four million infants born between 1989 and 1991 in 86 metropolitan statistical areas (MSAs) in the United States combined data from the National Center for Health Statistics birth/infant death records with measurements of PM10. Infants were categorized as having high, medium, or low exposure. After adjustment for other covariates, the odds ratio (OR) and 95% confidence intervals (CI) for total postneonatal mortality for the high exposure versus the low exposure group was 1.10 (1.04, 1.16). In normal birth weight infants, high PM10 exposure was associated with respiratory causes (OR ¼ 1.40 (1.05, 1.85)) and SIDS (OR ¼ 1.26 (1.14, 1.99)). In a follow-up study of postneonatal deaths in relation to PM2.5 in California, Woodruff et al. (2006) reported that the OR for a 10 mg/m3 increase in PM2.5 was 1.07 (0.93–1.24) overall, 2.13 (1.12–4.05) for respiratory causes, and 0.82 (0.55–1.23) for SIDS. Studies in Mexico City and in the Czech Republic have also reported associations between PM and infant mortality. For Mexico City (average PM2.5 of 27 mg/m3), Loomis et al. (1999) reported 18.2% excess in infant mortality per 25 mg/m3 (CI: 6.4–30.7). Boback and Leon (1999) reported that long-term exposure to PM for the whole Czech Republic was associated with excess neonatal and postneonatal deaths. Ritz et al. (2000) studied the association between pollutant exposures and low birth weight (LBW) in 97,518 neonates in Southern California, after adjustment for maternal age, race, smoking during pregnancy, etc. A 50 mg/m3 increase in PM10 exposure averaged over the first month of pregnancy was associated with a 16% increase in preterm delivery, while a 50 mg/m3 increase in PM10 exposure, averaged over the 6 weeks prior to birth, was associated with a 20% increase. However, a study by Maisonnet et al. (2001) of the association between LBW and PM10 in eastern U.S. cities (Boston and Springfield, MA, Hartford, CT, Philadelphia and Pittsburgh, PA, and Washington, DC) was negative.

344

FIGURE 10.13 Relationship between RR for excess daily mortality associated with and peak daily levels of other criteria pollutants. Source: Adapted from Fig. V-3a of PM Staff Paper; U.S. EPA (1996b).

345

FIGURE 10.14 Relationships between relative risks per 50 mg/m3 PM10, PM2.5, and PM10–2.5 and various mortality categories. Source: Adapted from Fig. V-2 in PM Staff Paper; U.S. EPA (1996b).

346

FIGURE 10.15 Relationships between relative risks per 50 mg/m3 PM10, PM2.5, and PM10–2.5 and various nonmortality health effects. Source: Adapted from Fig. V-2 in PM Staff Paper; U.S. EPA (1996b).

EPIDEMIOLOGICAL EVIDENCE FOR HUMAN HEALTH EFFECTS OF AMBIENT AIR PM

347

PM exposure, albeit at much higher ambient levels has been associated with preterm delivery and LBW. Xu et al. (1995) followed all registered pregnant women who lived in four residential areas of Beijing, PRC. The analysis included 25,370 women who gave first live births in 1988. Very high concentrations of ambient SO2 (mean ¼ 102 mg/m3, maximum 630 mg/m3) and TSP (mean ¼ 375 mg/m3, maximum ¼ 1003 mg/m3) were observed. There was a significant dose-dependent association between gestational age and TSP concentrations. The estimated reduced duration of gestation was 0.042 week (7.1 h) for each 100-mg/m3 increase in TSP 7-day lagged moving average. The adjusted OR for preterm delivery was 1.21 (95% CI ¼ 1.01–1.46) was 1.10 (95% CI ¼ 1.01–1.20) for each 100 mg/m3 increase in TSP. In addition, the gestational age distribution of high-pollution days was more skewed toward the left tail (i.e., very preterm and preterm) compared with low-pollution days. In a follow-on study, Wang et al. (1997) examined the relationship between maternal exposure to air pollution during periods of pregnancy (entire and specific periods) and birth weight in a well-defined cohort between 1988 and 1991. All pregnant women living in four residential areas of Beijing were registered and followed from early pregnancy until delivery. The sample for analysis included 74,671 first-parity live births with gestational age 37–44 weeks. Multiple linear regression and logistic regression were used to estimate the effects of air pollution on birth weight and LBW (<2500 g), adjusting for gestational age, residence, year of birth, maternal age, and infant gender. There was a significant exposure–response relationship between maternal exposures to TSP during the third trimester of pregnancy and infant birth weight. The adjusted OR for LBW was 1.10 (95% CI, 1.05–1.14) for each 100 mg/m3 increase in TSP. The estimated reduction in birth weight was 6.9 g for each 100 mg/m3 increase in TSP, respectively. The birth weight distribution of the high-exposure group was more skewed toward the left tail (i.e., with higher proportion of births <2500 g) than that of the low-exposure group. In the absence of any generally accepted mechanistic basis to account for the epidemiological associations between ambient PM2.5 on the one hand, and mortality, morbidity, and functional effects on the other, the causal role of PM2.5 remains questionable. However, essentially all attempts to discredit the associations on the basis of the effects being due to other environmental variables that may covary with PM2.5 have been unsuccessful. As shown in Fig. 10.11, the RR for daily mortality in relation to PM10 is remarkably consistent across communities that vary considerably in their peak concentrations of other criteria air pollutants. The possible confounding influence of adjustments to models to account for weather variables has also been found to be minimal (Samet et al., 1997; Pope and Kalkstein, 1996). While mechanistic understanding of processes by which ambient air PM causes human health effects remains quite limited, the credibility of ambient PM2.5 as a cause of excess human mortality and morbidity has been enhanced by a series of studies inhalation studies in which animals and humans have been exposed by inhalation to CAPs. 10.5.9

Human Inhalation Studies

Several CAPs acute exposure studies in healthy human volunteers have been performed. Ghio et al. (2000) exposed 38 healthy volunteers exercising intermittently at moderate levels of exertion for 2 h to either filtered air or CAPs (23–311 mg/m3) from the air in Chapel Hill, NC. Analysis of cells and fluid obtained 18 h after exposure showed a mild increase in neutrophils in the bronchial and alveolar fractions of bronchoalveolar lavage (BAL) in subjects exposed to the highest quartile concentration (mean of 206.7 mg/m3).

348

AMBIENT AIR PARTICULATE MATTER

Lavage protein did not increase, and there were no other indicators of pulmonary injury. No respiratory symptoms or decrements in pulmonary function were found after exposure to CAPs. The 38 human volunteers reported on by Ghio et al. (2000) were also examined for changes in host defense and immune parameters in BAL and blood (Harder et al., 2001). There were no changes in the number of lymphocytes or macrophages, subcategories of lymphocytes (according to surface marker analysis by flow cytometry), cytokines IL-6 and IL-8, or macrophage phagocytosis in BAL. Similarly, there was no effect of CAPs exposure on lymphocyte subsets in blood. Thus, a mild inflammatory response to CAP was not accompanied by an effect on immune defenses as determined by lymphocyte or macrophage effects. The increase in neutrophils may represent an adaptive response of the lung to PM2.5, although the presence of activated neutrophils may release biochemical mediators that produce lung injury. Whether this mild inflammatory increase in neutrophils constitutes a biologically significant injury to the lung is an ongoing controversial issue. Other human CAPs inhalation studies have had limited power because of the small numbers of subjects studied. Petrovic et al. (1999) exposed four healthy volunteers (ages 18– 40 years) under resting conditions to filtered air and three CAPs concentrations (23–124 mg/ m3) for 2 h using a face mask. The exposure was followed by 30 min of exercise. No cellular signs of inflammation were observed in induced sputum samples collected at 2 or 24 h after exposure. There was a trend toward an increase in nasal lavage neutrophils although no statistical significance was presented. The only statistically significant change in pulmonary function was a 6.4% decrease in thoracic gas volume after exposure to PM2.5 at124 mg/m3 versus a 5.6% increase after air. A similar, small pilot study has been reported (Gong et al., 2000) in which no changes in pulmonary function or symptoms were observed in four subjects aged 19–41 years after a 2 h exposure to air or CAPs at 148–246 mg/m3 in Los Angeles, CA. In a follow-up study, Gong et al. (2004) exposed 12 mildly asthmatic and four healthy adults to filtered air (FA) and concentrated ambient coarse particles (CCP) supplied via a coarse particle concentrator in a Los Angeles suburb with high levels of motor vehicle pollution for 2 h with intermittent exercise. Mean CCP concentration was 157 mg/m3 (range: 56–218 mg/m3). On average, 80% of mass was coarse (2.5–10 mm aerodynamic diameter) and the rest <2.5 mm. Relative to FA, CCP exposure did not significantly alter respiratory symptoms, spirometry, arterial oxygen saturation, or airway inflammation according to exhaled NO, and total and differential cell counts of induced sputum. After CCP exposure, Holter electrocardiograms showed small (p < .05) increases in heart rate (HR) and decreases in heart-rate variability (HRV), which were larger in healthy than in asthmatic subjects. A review of human CAPs inhalation studies by Ghio and Huang (2004) summarizes some other recent studies. These include a follow-up paper by Huang et al. (2003) to the Ghio et al. (2000) paper that applied principal-components analysis of the CAPs aerosol. They linked specific water-soluble PM components to both the neutrophil influx and elevation in blood fibrinogen (Huang et al., 2003). A SO4 ¼ /Fe/Se factor, which may be attributed to photochemical air pollution, was associated with the neutrophil increase in the lavage, while a Cu/Zn/V factor, related to various combustion processes, was linked to increases in blood fibrinogen. In another study, healthy and asthmatic individuals (18–45 years of age) were exposed (with 2 h of exercise) to both CAPs (mean concentration ¼ 174 mg/m3 and filtered air (Gong et al., 2003). There were no changes in symptomatology, pulmonary function, and hematologic measurements attributable to

EPIDEMIOLOGICAL EVIDENCE FOR HUMAN HEALTH EFFECTS OF AMBIENT AIR PM

349

CAPs. CAPs decreased columnar epithelial cells in induced sputum in both healthy and asthmatic subjects. There were also small changes in mediators of blood coagulability, inflammation, and HRV. CAPs have also been used to compare the vascular responses between exposures to CAPs/ ozone versus filtered air in Toronto (Brook et al., 2002). Nonsmoking adults were exposed to CAPs and FA separated by 2 days. CAPs exposure was 150 mg/m3 while ozone exposure was 120 ppb. High-resolution vascular ultrasonography was used to measure alterations in brachial artery diameter (BAD), endothelial-dependent flow-mediated dilatation, and endothelial-independent nitroglycerine-mediated dilatation. Exposure to CAPs and ozone was associated with small but statistically significant BAD constriction compared to filtered air. There were no differences in flow-mediated dilatation or blood pressure responses between exposures. In a follow-up paper, Urch et al. (2004) examined the relationship between total and constituent PM2.5 mass concentrations and the acute vascular response. They found a significant negative association between both the organic and elemental carbon concentrations and the difference in the postexposure change in the BAD (DBAD) between and CAP þ O3 and FA exposure days. Devlin et al. (2003) studied individuals between 60 and 80 years of age who were exposed to both CAPs and filtered air for 2 h without exercise. There were significant decrements in HRVin both time and frequency domains immediately following exposure to CAPs. Some of these changes persisted for at least 24 h. These results contrast with those in the previous study of Ghio et al. (2000) in which young, healthy subjects exposed to CAPs with exercise had no changes in HRV relative to subjects inhaling FA. 10.5.10

Animal Inhalation Studies

Studies in normal dogs exposed to Boston CAPS by inhalation (Clarke et al., 2000) showed increases in pulmonary inflammation by bronchoalveolar lavage and in circulating blood neutrophils associated with specific ambient particle components. In these experiments, mean concentrations were 203 and 361 mg/m3. Saldiva et al. (2002) studied the effects on rat lung of CAPs from Boston. Rats with chronic bronchitis and normal rats were exposed by inhalation either to filtered air or CAPs induced a significant increase in bronchoalveolar lavage (BAL) neutrophils and in normal and bronchitic animals. A significant dosedependent association was found between CAPs components and BAL neutrophils. The authors concluded that: (a) short-term exposures to CAPs from Boston induce a significant inflammatory reaction in rat lungs; and (b) the reaction is influenced by particle composition. Gurgueira et al. (2002) exposed adult Sprague-Dawley rats to either CAPs aerosols (mass concentration, 300  60 mg/m3) or filtered air for periods of 1–5 h. Rats breathing CAPs aerosols for 5 h showed significant oxidative stress, determined by in situ chemiluminescence in the lung and heart, but not in the liver. Increases in chemiluminescence showed strong associations with the CAPs content of Fe, Mn, Cu, and Zn in the lung and with Fe, Al, Si, Ti in the heart. CAPs inhalation also led to tissue-specific increases in the activities of the antioxidant enzymes superoxide dismutase and catalase, suggesting that episodes of increased particulate air pollution not only have potential for oxidant injurious effects but many also trigger adaptive responses. Cheng et al. (2003) exposed male Sprague-Dawley rats with implanted radiotelemetry devices to CAPs for 6 h/day for 3 consecutive days, with rest for 4 days in each week during the experimental period of 5 weeks. These animals were exposed to concentrated particles

350

AMBIENT AIR PARTICULATE MATTER

during weeks 2, 3, and 4 and exposed to filtered air during weeks 1 and 5. The particle concentrations ranged between 108 and 338 mg/m3. CAPs exposure was associated with changes in heart rate and mean blood pressure. Immediately after particle exposure, the heart rate decreased and reached the lowest at the first and second hour of exposure for a decrease of 14.9 (p < 0.01) and 11.7 (p ¼ 0.01) beats per minute, respectively. The hourly mean blood pressure also decreased after the particle exposure, with a maximal decrease of 3.3 (p < 0.01) and 4.1 (p < 0.01) mmHg at the first and second hour of exposure. Nadziejko et al. (2003) exposed Fischer 344 rats at 18 months of age with implanted EKG transmitters for 4 h to New York City CAPs at 160 and 200 mg/m3 or filtered air to determine the effects of PM on the frequency of spontaneous arrhythmias in old rats. The EKG tracings demonstrated a significant increase in the frequency of supraventricular arrythmias after exposure to CAPS compared to the sham exposed animals. The effects of PM on myocardial ischemia have also been studied. Inhaled PM exacerbated ischemia in a model of coronary arterial occlusion in conscious dogs. Exposures to Boston CAPs significantly increased peak electrocardiographic ST-segment elevation during a 5-min coronary artery occlusion compared to sham exposures in two different protocols (Godleski et al., 2000; Wellenius et al., 2003). Zelikoff et al. (2003) reported effects on pulmonary or systemic immune defense mechanisms in Fischer 344 rats exposed to New York City CAPs at 0 or 90–600 mg/m3 for 3 h prior to intratracheal instillation of Streptococcus pneumoniae. The number of lavageable macrophages and neutrophils increased in both control and experimental groups, but were elevated faster and were twice as high in the CAPs-exposed group, as well as staying elevated longer. Lymphocytes and white blood cells were significantly increased 24 and 72 h postinfection in both groups. CAPs exposure significantly increased bacterial burdens at 24 h postinfection. Thereafter, CAPs-exposed animals exhibited significantly lower bacterial burdens. Zelikoff et al. (2003) also evaluated the effects of a single 5 h exposure to CAPs in rats following an intratracheal instillation of S. pneumoniae CAPs exposure significantly reduced percentages of lavageable neutrophils 24 h following CAPs exposure. Lavageable macrophages were significantly increased in the CAPs exposed animals. CAPs exposure reduced the levels of TNF, IL-1, and IL-6. The bacterial burden decreased in both exposed groups over time; however, CAPs exposed animals had a significantly greater burden after 24 h than did control rats. Lymphocyte and monocyte levels were unaffected by CAPs exposure. In a series of studies at NYU, Gordon et al. (2000) examined rodent cardiovascular system responses to CAPs derived from New York City air. Particles of 0.2–2.5 mm diameter were concentrated up to 10 times their levels in ambient air (130–900 mg/m3) to maximize possible differences in effects between normal and cardiopulmonary-compromised laboratory animals. EKG changes were not detected in normal Fischer 344 rats or hamsters exposed by inhalation to the New York City CAPs for 1–3 days. Similarly, no deaths or EKG changes were seen in monocrotaline (MCT) treated rats (a model of pulmonary hypertension) or cardiomyopathic hamsters exposed to PM. In contrast to the nonsignificant decrease in heart rate observed in dogs exposed to Boston CAPs (Godleski et al., 2000), statistically significant heart rate increases (5%) were observed by Gordon et al. (1998) in both normal and MCT rats exposed to PM, but not on all exposure days. Thus, extrapolation of the heart rate changes in these animal studies to human health effects is difficult, although the increase in heart rate in rats is similar to that observed in some human population studies.

EPIDEMIOLOGICAL EVIDENCE FOR HUMAN HEALTH EFFECTS OF AMBIENT AIR PM

351

Gordon et al. (1998) reported other cardiovascular effects in animals exposed to inhaled CAPs. Increases in peripheral blood platelets and neutrophils were observed in control and MCT rats at 3 h, but not 24 h, after exposure to 150–400 mg/m3 of CAPs. This neutrophil effect did not appear to be dose-related and did not occur on all exposure days, suggesting that day-to-day changes in particle composition may play an important role in the systemic effects of inhaled particles. The number of studies reported was small; and, it is therefore not possible to statistically determine if the day-to-day variability was truly due to differences in particle composition or even to determine the size of this effect. Nadziejko et al. (2003) exposed healthy rats to CAPs from New York City air at a concentration range of 95–341 mg/m3 for 6 h and sampled blood at 0, 12, and 24 h postexposure. They found no consistent differences in counts of platelets, blood cells, or in levels of proteins in the blood coagulation system that included fibrinogen, thrombin– antithrombin complex, tissue plasminogen activator, plasminogen activator inhibitor, and factor VII. Kleinman, as described by Lippmann et al. (2003), exposed ovalbumin (OVA)-sensitized mice in a specially equipped van that was located 50 m downwind of a Los Angeles freeway. Groups of mice were exposed to CAPs at 400 and 800 mg/m3 for 5 or 10 days. Control mice were sham exposed. All mice received an inhalation challenge of OVA 2 weeks after their last CAPs or sham exposure. Eosinophils and OVA-IgE were increased, relative to sham exposed, after the 5- and 10-day CAPs exposures at 400 mg/m3. While acute CAPs exposure studies provide a useful supplement to the acute effects studies in human populations, there were not, until recently, any prolonged CAPs exposure studies to complement the human cohort studies that indicate PM2.5 concentration related differences in annual mortality rates. This deficiency was especially important when we consider that the ACS cohort studies (Pope et al., 1995a, 1995b, 2002, 2004a, 2004b) and the Harvard Six-Cities Study (Dockery et al., 1993; Laden et al., 2006) imply that there is an average longevity reduction of 1 to 2 years between the U.S. cities at the 5th and 95th percentiles of PM2.5 concentration, and that the mortality impact from these studies is several times greater than that indicated by the daily time-series mortality studies. Also, the most recent analysis of the ACS cohort (Pope et al., 2004a, 2004b) assigns most of the mortality impact to deaths from cardiac disease. Lippmann et al. (2005a) addressed this deficiency with a 6-month CAPs inhalation study in normal mice and a mouse model of atherosclerotic disease. The primary objective of this study was to determine whether cumulative daily exposures would cause progressive changes in cardiac function in an animal model for a susceptible human population. In both groups of mice, there were sham (clean air) exposures following the same protocols used to expose animals to CAPs. Other coordinate objectives were to look for other PM2.5-related responses including (1) short-term changes in cardiac function associated with daily peak PM2.5 concentrations and/or specific air trajectories; (2) aortic plaque formation and/or plaque size at the end of exposures; (3) Gene activation at the end of the exposures; (4) Morphologic changes in the heart, lungs, and brains at the end of the exposures. To complement the evidence for acute cardiac function changes related to daily variations in PM2.5 exposure, collected particles were collected each day in an air sampler that operated

352

AMBIENT AIR PARTICULATE MATTER

in parallel with the particle concentrators that fed aerosol into the exposure chambers, in vitro assays on the particles that were sampled were performed. Protocols for the analysis of the huge volume of cardiac function data were developed specifically for this study, and were described by Chen and Hwang (2005) and Hwang et al. (2005). Groups of normal mice (C57) and knockout mice that develop artherosclerotic plaques (ApoE / and ApoE / LDLr / ) for 6 h/day, 5 day/week for 5 or 6 months during the spring/summer of 2003 to either filtered air or 10-fold CAPs in Tuxedo, NY (average PM2.5 concentration during exposure ¼ 110 mg/m3). Some of the mice had implanted electrocardiographic monitors. As described by Lippmann et al. (2005b): (1) this complex interdisciplinary study was technically feasible in terms of: daily exposures; collection of air quality monitoring data; the collection, analysis, and interpretation of continuous data on cardiac function; and the collection and analyses of tissues of the animals sacrificed at the end of the study; (2) the daily variations in CAPs were significantly associated, in ApoE / mice, with daily variations in cardiac function; (3) there were significant differences between CAPs and sham exposed ApoE / mice in terms of cardiac function after the end of the exposure period, as well as small differences in atherosclerotic plaque density, coronary artery disease, and cell density in the substantia nigra in the brain in the ApoE / mice; and (4) suggestive indications of gene expression changes for genes associated with the control of circadian rhythm in the ApoE / LDLr / double knockout (DK) mice. Lippmann et al. (2005c) also examined temporal variations in heart rate (HR) and HR variability (HRV) responses during the 24-h beginning with the start of the 6-h CAPs exposure in relation to the components of the CAPs as determined by source apportionment of the daily 6-h sampling filter. For HR, there were significant transient associations with resuspended soil (RS) during the CAPs exposures, and for secondary sulfate (SS) in the afternoon after exposure. For HRV, there were significant transient associations with residual oil (RO) combustion effluents in the afternoon, and with and for both SS and RS late at night. In a second 6-month CAPs inhalation exposure study in Tuxedo, NY, to ApoE / mice either on a high-fat diet or a normal chow, the average PM2.5 concentration during exposure was 85 mg/m3. In the mice on the high-fat diet, there was a significant exposure-related increase in atherosclerosis as well as altered vasomotor tone and induced vascular inflammation. Similar tendencies were seen in the mice on normal chow, but they were not statistically significant (Sun et al., 2005). As shown in Fig. 10.16, during the course of this second 6-month CAPs inhalation exposure study in Tuxedo, NY to ApoE / mice on either a high-fat diet, there were 14 days during the fifth and sixth months when there were unusually large deviations from baseline levels in HR and HRV. During those 14 days, the concentrations of Ni, Cr, and Fe were unusually high and the concentrations of PM2.5 and all other measured elements were unusually low. Back 72-h air trajectory analyses on those 14 days led to, or close to, a Ni smelter in western Ontario. Ni was the only element that was significantly associated with the Daily variations in HR and HRV (Lippmann et al., 2006). These various CAPs related effects on cardiac function and the development of histological evidence of increased risk of clinically significant disease at the end of the exposures in animal models of atherosclerosis provide biological plausibility for the premature mortality associated with PM2.5 exposure in human subjects and suggestive evidence for neurogenic disease as well. Collectively, these animal inhalation studies indicate that diseased and elderly rats respond to PM2.5 inhalation with greater responses than healthy young animals, and produce responses that appear relevant to excess mortality and morbidity in sensitive

353

FIGURE 10.16 Elemental concentrations and HR and HRV (mean  SE) for 14 days when winds were from the northwest and for the 89 days with winds from all other directions, and the differences in heart rates in ApoE / “ mice exposed to CAPs and filtered air. CAPs concentrations are shown in mg/m3, elemental concentrations in ng/m3, HR in beats/min, and HRV (as log SDNN) in milliseconds.

354

AMBIENT AIR PARTICULATE MATTER

human populations. Future extensions of these studies using specific components of the ambient PM2.5 could help to identify the most biologically active components of the ambient PM2.5 mixtures, as well as the host factors that put an individual at greatest risk.

10.6 DISCUSSION AND CURRENT KNOWLEDGE ON THE HEALTH EFFECTS OF PM The results of studies in recent years, summarized above, have made it possible to frame the remaining unresolved issues in a more coherent and focused manner. One key issue is the role of SO4 ¼ , and why it consistently correlates with mortality and morbidity as well as, or better than, other metrics of PM pollution. It is extremely unlikely that SO4 ¼ , per se, is a causal factor. If it is not, then it must be acting as a surrogate index for one or more other components in the PM mixture. Evidence for and against an active role for some PM constituents is presented in Table 10.4. One possibility is that the effects are really due to the PM2.5 mass, irrespective of particle composition, and that SO4 ¼ is a more stable measurement of airborne PM2.5 than is the reported PM2.5 itself. The ambient PM2.5 includes semivolatile compounds, such as nitrates

TABLE 10.4 Components of Ambient Air PM That May Account for Some or All of the Effects Associated with PM Exposures Components þ

Strong acid (H )

Ultrafine particles (D  0.2 mm)

Soluble transition metals

Peroxides

Evidence for Role in Effects

Doubts

Statistical associations with health effects in most recent studies for which ambient Hþ concentrations were measured Coherent responses for some health end points in human and animal inhalation and in vitro studies at environmentally relevant doses Much greater potency per unit mass in animal inhalation studies (Hþ, Teflon, and TiO2 aerosols) than for same materials in larger diameter fine particle aerosols Concept of “irritation signaling” in terms of number of particles per unit airway surface Recent animal study evidence of capability to induce lung inflammation

Similar PM-associated effects observed in locations with low ambient Hþ levels

Close association in ambient air with SO4 ¼ Strong oxidizing properties

Very limited data base on ambient concentrations

Only one positive study on response in humans

Absence of relevant data base on ambient concentrations Absence of relevant data on responses in humans Absence of relevant data on ambient concentrations Absence of relevant data on responses in humans or animals Very limited database on ambient concentrations

DISCUSSION AND CURRENT KNOWLEDGE ON THE HEALTH EFFECTS OF PM

355

(primarily ammonium nitrate) and organics formed by photochemical reactions in the atmosphere. There can be considerable volatilization of these species on sampling filters, resulting in negative mass artifacts whose magnitude varies with source strengths and ambient temperature. Some of the semivolatile organics may account for the associations reported between indices of traffic-related pollution and health effects. Another possibility is that SO4 ¼ is serving as a surrogate for Hþ, a more likely active agent on the basis of the results of controlled exposure studies in humans and animals. The support for this hypothesis is summarized in Table 10.4 and discussed further in Chapter 29. A third possibility is that the causal factor is the number concentration of irritating particles, which would be dominated by the particles in the ultrafine mode (diameters below 50 nm) (Oberd€ orster et al., 1995). Epidemiologic support for this hypothesis has been provided by Peters et al. (1997), who reported closer associations between peak expiratory flow rates and symptoms in adult asthmatics with particle number concentration than with fine particle mass concentration in Erfurt, Germany. A fourth possibility is that soluble transition metals in the ambient PM generate sufficient amounts of reactive oxygen species in the respiratory tract airways to cause inflammatory responses and chronic lung damage (Pritchard et al., 1996). A fifth possibility has been proposed by Friedlander and Yeh (1996), that is, that reactive chemical species, such as peroxides, are responsible for the health effects associated with fine particles, and that SO4 ¼ , being a product of chemical reactions involving hydrogen peroxide, is serving as a surrogate measure of the airborne peroxides. Support for this hypothesis has been provided by an in vivo animal exposure study by Morio et al. (2001). It is also possible that effects are related to a hybrid of Hþ and ultrafines, that is, acidcoated ultrafine particles. As discussed in Chapter 29, sulfuric acid coatings on ultrafine zinc oxide particles produce about the same responses as pure sulfuric acid for a given number of equivalent sized particles, yet the coated particles only had one tenth of the acid content per unit volume of air. Thus, the response may be related to the number of acidic particles that deposit on the lung surfaces rather than the amount of acid deposited. In other words, the total concentration of Hþ may be a better surrogate of the active agent than SO4 ¼ or PM2.5, but it still is a crude index for the number concentration of irritant particles. Amdur and Chen (1989) suggested that number concentration was important for sulfuric acid aerosol, and Hattis et al. (1987, 1990) gave the concept a name, that is, “irritation signaling.” Research of Chen et al. (1995) indicate that acid-coated particles much smaller than those discussed by Hattis et al. (1987, 1990) were capable of producing lung responses. If the number concentration of acid-coated particles is the most relevant index of the active agent in ambient PM, then new sampling techniques will be needed to characterize ambient air PM concentrations and personal exposures. Other components of the ambient ultrafine aerosol have not been well characterized either, and they may also be important health stressors. One class is the volatile trace metals (such as As, Cd, Cu, Ni, Pb, V, Zn), which condense as ultrafine particles in the effluent airstream of fossil fuel combustors (Amdur et al., 1986), and are inefficiently captured by air cleaners for fly ash collection. Another class is the ultrafine organics from atmospheric photochemical reaction sequences. Any remaining inconsistency between the epidemiological findings and the results of the controlled exposure studies may be explicable on the basis that the relatively rare individuals who respond in the epidemiological populations are an especially responsive subset of the overall population, and the low probability that such sensitive individuals would be included

356

AMBIENT AIR PARTICULATE MATTER

in the controlled exposure studies in the laboratory. An alternative hypothesis is that the controlled exposure atmospheres have not contained the highly toxic components or ultrafine particle sizes that may be present in ambient atmospheres. In summary, excess daily mortality and morbidity have been related to ambient PM at current levels in many communities in the United States and around the world using available pollutant concentration data. However, it is not at all clear whether any of the pollutant indices used are causally related to the health effects or, if none of them are, which is the best index or surrogate measure of the causal factor(s). This gap can best be addressed by analyses of pollutant associations with mortality and morbidity in locations where a number of different pollutant metrics are available simultaneously, using analytic methods not dependent on arbitrary model assumptions.

10.7 STANDARDS AND EXPOSURE GUIDELINES Unlike the United Kingdom, which has relied on source controls and eschewed numerical guidelines and standards for ambient air pollutants, most national and international agencies with responsibilities involving public health protection from the effects of air pollution on human health have felt that quantitative concentration limits were needed to guide their emission controls. However, selecting concentration limits for PM, which is a class of pollutants of varying particle size and chemical composition that is defined only by existing in the condensed phase has always been rather challenging. While more research is needed on causal factors for the excess mortality and morbidity associated with PM in ambient air, and on the characterization of susceptibility factors, most responsible public health authorities have felt that they could not wait for the completion and peer review of such research. In addition, the evidence for adverse health effects attributable to PM challenges the conventional paradigm used for setting ambient air standards and guidelines, that is, a threshold for adversity can be identified, and a margin of safety can be applied. Excess mortality is clearly an adverse effect, and the epidemiological evidence has long been consistent with a linear nonthreshold response for the population as a whole. The World Health Organization-Europe (WHO-EURO) has issued Air Quality Guidelines in 1987, 2000, and 2005, and the International WHO has adopted the 2000 and 2005 for the world as a whole as guidance to the member states that wish to establish national standards (WHO-EURO, 1987, 2000; WHO, 2005). In the 1987 WHO-EURO Guidelines relied on the U.K. experience with the health effects of soft coal combustion effluents and established Guideline values for combined exposures to SO2 and PM. The 24-h Guidelines were 125 mg/m3 for SO2, 125 mg/m3 for black smoke, as measured by light reflectance from a filter (but only in locations where PM from coal smoke is the dominant PM component), 120 mg/m3 as measured gravimetrically on a TSP filter, and 70 mg/m3 as measured gravimetrically on a PM10 filter These values were based on a safety factor of 2 for mortality and morbidity effects. The guidelines for the annual average concentration were 50 mg/m3 for both SO2 and black smoke, and no annual average limits were suggested for PM mass concentrations. In 2000, with the availability of health effect studies with PM measured as SO4 ¼ , PM10, and/or PM2.5 that were consistent, and with there being no threshold for mortality effects, the WHO-Euro Working Group determined that it could not recommend a PM Guideline. Instead, it prepared a tabular presentation of the estimated changes in daily average PM concentrations needed to produce specific percentage changes in: (1) daily mortality;

STANDARDS AND EXPOSURE GUIDELINES

357

TABLE 10.5 Air Quality Guideline and Interim Targets for Particulate Matter: Annual Mean Annual Mean Level

PM10 (mg/m3)

PM2.5 (mg/m3)

WHO interim target-1 (IT-1)

70

35

WHO interim

50

25

?? Target-2 (IT-2)

?? ??

?? ??

WHO interim target-3 (IT-3)

30

15

WHO Air quality guidelines (AQG)

20

10

Basis For the Selected Level These levels are estimated to be associated with about 15% higher long-term mortality than at AQG In addition to other health benefits, these levels ?? Lower risk of premature mortality by approximately 6% (2–11%) compared to WHO-IT1 In addition to other health benefits, these levels reduce mortality risk by another approximately 6% (2–11) compared to WHO-IT2 levels. These are the lowest levels at which total, cardiopulmonary, and lung cancer mortality have been shown to increase with more than 95% confidence in response to PM2.5 in the ACS study (Pope et al., 2002). The use of PM2.5 guideline is preferred.

(2) hospital admissions for respiratory conditions; (3) bronchodilator use among asthmatics; (4) symptom exacerbation among asthmatics; and (5) peak expiratory flow. The concentrations needed to produce these changes were expressed in PM10 for all five of the response categories. For mortality and hospital admissions, they were also expressed only in terms of PM2.5 and SO4 ¼ . Using this guidance, each national or local authority setting air quality standards can decide how much adversity is acceptable for its population. Making such a choice proved to be too much of a challenge to most national authorities. The 2005 WHO Guidelines, which were intended to be used worldwide, and not only in developed countries, were structured differently than those developed in 1987 and 2000 by WHO-EURO. Table 10.5 describes the Guideline and interim targets for the annual means for PM10 and PM2.5. Table 10.6 describes the corresponding values for 24-h average concentrations. The higher interim targets were offered for those countries where concentrations currently greatly exceed the guidelines in order to demonstrate that important public health benefits could be achieved with phased progress in lowering exposures. In the United States, the EPA Administrator promulgated National Ambient Air Quality Standards (NAAQS) for PM in 1971, 1987, and 1997. In 1971, PM was defined as the total suspended particle (TSP) mass collected by a high-volume sampler, which collected essentially all particles below an inlet cutoff of 25–45 mm that varied with wind speed and direction. The primary (health based) TSP limit was 260 mg/m3 as a 24-h average not to be exceeded more than once per year, and 75 mg/m3 as an annual geometric mean. In 1987, in recognition that the UK health effects data from the 1950s and 1960s, based on exposures to coal combustion effluents and optical reflectance measurements of airborne PM was not highly relevant to then current U.S. conditions, used newer data on PM dosimetry and

358

AMBIENT AIR PARTICULATE MATTER

TABLE 10.6

Air Quality Guideline and Interim Targets for Particulate Matter: 24-h Mean PM10 (mg/m3)

PM2.5 (mg/m3)

WHO interim target-1 (IT-1)

150

75

WHO interim target-2 (IT-2)a

100

50

75

37.5

59

25

24-h Mean Levela

WHO Interim target-3 (IT-3)b WHO Air quality guidelines (AQG)

Basis for the Selected Level Based on published risk coefficients from multicenter studies and meta-analyses (about 5% increase of short-term mortality over AQG) Based on published risk coefficients from multicenter studies and meta-analyses (about 2.5% increase of short-term mortality over AQG) (About 1.2% increase in short-term mortality over AQG) Based on relation between 24-h and annual PM levels

a

99th percentile (3 days/year). For management purposes, based on annual average guideline values; precise number to be determined on basis of local frequency distribution of daily means. b

health effects analyses to revise the PM NAAQS. The indicator was changed from TSP to PM10, limiting particle collection to those that could penetrate into the thorax. For the primary (health based) PM NAAQS, the level and form were changed to a 24-h gravimetric limit of 150 mg/m3 with no more than 1 expected exceedance per year, and an annual average of 50 mg/m3. In 1997 (Federal Register 1997, 62: 38762–38896), the PM NAAQS was changed in recognition of the inadequate public health protection provided by enforcement of the 1987 NAAQS for PM10 for the control of the fine particles that were recognized as being particularly hazardous. The PM10 of annual average of 50 mg/m3 was retained without change, and the 24-h PM10 of 150 mg/m3 was relaxed by applying it only to the 98th% value (eighth highest in each year) rather than to the fourth highest over 3 yrs. These PM10 standards were supplemented by the creation of new PM2.5 standards. The annual average PM2.5 chosen was 15 mg/m3, and the 24-h PM2.5 of 65 mg/m3 applied to the 98th % value. In the process of performing the review and reconsideration of the PM NAAQS in 2005, the EPA Staff Paper (U.S. EPA, 2006b) recommended, to the EPA Administrator, that the PM10 NAAQS be replaced by a 24-h PM10–2.5 NAAQS for urban areas that would provide comparable protection against thoracic coarse particles (50–70 mg/m3 with the 98% form), and that the PM2.5 NAAQS be reduced, with a 24-h level of 35–30 mg/m3 with the 98% form, and a annual concentration average within the range of 15–12 mg/m3. The Clean Air Scientific Advisory Committee (CASAC), an external scientific advisory committee that was mandated by the Clean Air Act Amendments of 1977, to review new NAAQS and recommend changes, endorsed the establishment of a new PM10–2.5 NAAQS to replace PM10 for protection against coarse particle effects, and endorsed the lowering of the 24-h PM2.5 NAAQS, but advised the EPA Administrator that an annual limit of 15 mg/m3 was not protective of the public health and that it should be within the range of 13–14 mg/m3 (CASAC, 2006). The Administrator ignored this advisory and promulgated new NAAQS that retained the PM10 24-h limit rather than adopting the proposed PM10–2.5 limit,

REFERENCES

359

established the 24-h PM2.5 NAAQS at 35 mg/m3, and also retained the annual PM2.5 limit at 15 mg/m3. Implementation of the newly revised 24-h PM2.5 NAAQS will advance the degree of public health protection against acute effects in those cities where fine particle concentration have broad day-to-day variations and peak levels between 65 and 35 mg/m3. These new NAAQS are clearly not sufficiently stringent to the protect public health against adverse chronic health effects, and are poorly targeted against the acute effects resulting from exposure to the coarse fraction of PM10. In the view of the EPA Administrator in 2006, there remained significant knowledge gaps on both exposures, and the nature and extent of the effects, to make the need for more restrictive NAAQS difficult for him to justify. It is essential that adequate research resources be applied to filling these gaps before the next round of NAAQS revisions during the coming years.

REFERENCES Abbey DE, Nishino N, McDonnell WF, Burchette RJ, Knutson SF, Beeson WL, Yang JX (1999) Long-term inhalable particles and other air pollutants related to mortality in nonsmokers. Am. J. Respir. Crit. Care Med. 159:373–382. Amdur MO, Chen LC (1989) Furnace-generated acid aerosols: speciation and pulmonary effects. Environ. Health Perspect. 79:147–150. Analitis A, Katsouyanni K, Dimakopoulou K, Samoli E, Nikoloupoulos AK, Petasakis Y, et al. (2006) Short-term effects of ambient particles on cardiovascular and respiratory mortality. Epidemiol. 17:230–233. Bates DV (1992) Health indices of the adverse effects of air pollution: the question of coherence. Environ. Res. 59:336–349. Beeson WL, Abbey DE, Knutson SF (1998) Long-term concentrations of ambient air pollutants and incident lung cancer in California adults: results from the AHSMOG study. Environ. Health Perspect. 106:813–823. Bell ML, Davis DL, Fletcher T (2004) A retrospective assessment of mortality from the London smog episode of 1952; the role of influenza and pollution. Environ. Health Perspect. 112:6–8. Boback M, Leon DA (1999) Pregnancy outcomes and outdoor air pollution: an ecological study in districts of the Czech Republic 1986–1988. Occup. Environ. Med. 56:539–543. Brauer M, Gehring U, Brunekreef B, de Jongste J, Gerritsen J, Rovers M , et al. (2006) Traffic-related air pollution and otitis media. Environ. Health Perspect. 114:1414–1418. Brook RD, Brook JR, Urch B, Vincent R, Rajagopalan S, Silverman F (2002) Inhalation of fine particle air pollution and ozone causes acute arterial vasoconstriction in healthy adults. Circulation 105:1534–1536. CASAC (2006) Letter from CASAC Chair, Dr. Rogene Henderson, to EPA Administrator Stephen L. Johnson, dated September 29, 2006. Available from EPA as EPA-CASAC-LTR-06-003. Cassino C, Ito K, Bader I, Ciotoli C, Thurston G, Reibman J (1999) Cigarette smoking and ozoneassociated emergency department use for asthma by adults in New York City. Am. J. Respir. Crit. Care Med. 159:1773–1779. Chen LC, Hwang J-S (2005) Effects of subchronic exposures to CAPs in mice: IV. Characterization of acute and chronic effects of ambient air fine particulate matter exposures on heart rate variability. Inhal. Toxicol. 17:209–216. Chen LC, Nadziejko C (2005) Effects of subchronic exposures to CAPs in mice: V. Exacerbation of aortic plaque development in hyperlipidemic mice after subchronic exposures to CAPs. Inhal. Toxicol. 17:217–224.

360

AMBIENT AIR PARTICULATE MATTER

Chen LC, Wu CY, Qu QS, Schlesinger RB (1995) Number concentration and mass concentration as determinants of biological response to inhaled irritant particles. Inhal. Toxicol. 7:577–588. Cheng T-J, Hwang J-S, Wang P-Y, Tsai C-F, Chen C-Y, Lin S-H, Chang CC (2003) Effects of concentrated ambient particles on heart rate and blood pressure in pulmonary hypertensive rats. Environ. Health Perspect. 111:147–150. Chestnut LG, Schwartz F, Savitz DA, Burchfiel CV (1991) Pulmonary function and ambient particulate matter: epidemiological evidence from NHANES I. Arch. Environ. Health 46:135–144. Chinn S, Florey C, Du V, Baldwin IG, Gorgol M (1981) The relation of mortality in England and Wales 1969-73 to measurements of air pollution. J. Epidemiol. Commun. Health 35:174–179. Clarke RW, Coull B, Reinisch U, Catalano P, Killingsworth CR, Koutrakis P, Kavouras I, Murthy GGK, Lawrence J, Lorett E, Wolfson JM, Verrier RL, Godleski JJ (2000) Inhaled concentrated ambient particles are associated with hematological and bronchoalveolar lavage changes in canines. Environ. Health Perspect. 108:1179–1187. Ciocco A, Thompson DJ (1961) A follow-up of Donora ten years after: methodology and findings. Am. J. Public Health 51:155–164. Commins BT, Waller RE (1963) Determination of particulate acid in town air. Analyst 88:364–367. Creason J, Neas L, Walsh D, Williams R, Sheldon L, Liao D , et al. (2001) Particulate matter and heart rate variability among elderly retirees: the Baltimore 1998 PM study. J. Expos. Anal. Environ. Epidemiol. 11:116–122. Devlin RB, Ghio AJ, Kehrl H, Sanders G, Cascio W (2003) Elderly humans exposed to concentrated air pollution particles have decreased heart rate variability. Eur. Respir. J. 21(40 Suppl.):76s–80s Dockery DW, Pope III, CA, Xu X, Spengler JD, Ware JH, Fay ME, Ferris Jr, BG, Speizer FE (1993) An association between air pollution and mortality in six U.S. cities. N. Engl. J. Med. 329:1753–1759. Eisenbud, M. Personal Communication Firket J (1936) Fog along the Meuse Valley. Trans. Faraday Soc. 32:1192–1197. Friedlander SK, Yeh EK (1996) The submicron atmospheric aerosol as a carrier of reactive chemical species: case of peroxides. In: Lee J, Phalen R, editors. Proceedings of 2nd Colloquium on Particulate Air Pollution and Human Health. Univ. of CA––Air Pollution Health Effects Laboratory (December 1996). pp. 4-122–4-135 Gauderman WJ, McConnell R, Gilliland F, London S, Thomas D, Avol E , et al. (2000a) Association between air pollution and lung function growth in southern California children. Am. J. Resp. Crit. Care Med. 162:1383–1390. Gauderman WJ, Gilliland F, Vora H, Avol E, Stram D, McConnell R , et al. (2000) Association between air pollution and lung function growth in southern California children: results from a second cohort. Am. J. Resp. Crit. Care Med. 166:76–84. Gauderman WJ, Avol E, Gilliland F, Vora H, Thomas D, Berhane K , et al. (2004) The effect of air pollution on lung development from 10 to 18 years of age. New Engl. J. Med. 351:1057–1067. Gehring U, Heinrich J, Kramer U, Grote V, Hochadel M, Sugiri D , et al. (2006) Long-term exposure to ambient air pollution and cardiopulmonary mortality in women. Epidemiology 17:545–551. Ghio AJ, Huang YCT (2004) Exposure to concentrated ambient particles (CAPs): a review. Inhal. Toxicol. 16:53–59. Ghio AJ, Kim C, Devlin RB (2000) Concentrated ambient air particles induce mild pulmonary inflammation in healthy human volunteers. Am. J. Respir. Crit. Care Med. 162:981–988. Godleski JJ, Verrier RL, Koutrakis P, Catalano P (2000) Mechanisms of Morbidity and Mortality from Exposure to Ambient Air Particulate. Research Report # 91. Cambridge, MA: Health Effects Institute. Gold D, Litonjua A, Schwartz J, Lovett E, Larson A, Nearing B, et al. (2000) Ambient pollution and heart rate variability. Circulation 101:1267–1273.

REFERENCES

361

Gong Jr, H, Sioutas C, Linn WS, Clark KW, Terrell SL, Terrell LL, Anderson KR, Kim S, Chang M-C (2000) Controlled Human Exposures to Concentrated Ambient Fine Particles in Metropolitan Los Angeles: Methodology and Preliminary Health-Effect Findings. In: Phalen RF, editor. Proceedings of the Third Colloquium on Particulate Air Pollution and Human Health (first special issue); June, 1999; Durham, NC. Inhal. Toxicol. 12(Suppl. 1):107–119. Gong Jr, H, Linn WS, Sioutas C, Terrell SL, Clark KW, Anderson KR, Terrell LL (2003) Controlled exposures of healthy and asthmatic volunteers to concentrated ambient fine particles in Los Angeles. Inhal. Toxicol. 15:305–325. Gong Jr, H, Linn WS, Terrell SL, Clark KW, Geller MD, Anderson KR, Cascio WE, Sioutas C (2004) Altered heart-rate variability in asthmatic and healthy volunteers exposed to concentrated ambient coarse particles. Inhal. Toxicol. 16:335–343. Gordon T, Nadziejko C, Chen LC, Schlesinger R (2000) Effects of Concentrated Ambient Particles in Rats and Hamsters: An Exploratory Study. Research Report No 93.Cambridge, MA:Health Effects Institute. Gordon T, Nadziejko C, Schlesinger RB, Chen LC (1998) Pulmonary and cardiovascular effects of acute exposure to concentrated ambient particulate matter in rats. Toxicol. Lett. 96–97:285– 288. Gurgueira SA, Lawrence J, Coull B, Krishna Murthy GG, Gonzalez-Flecha B (2002) Rapid increases in the steady-state concentration of reactive oxygen species in the lungs and heart after particulate air pollution inhalation. Environ. Health Perspect. 110:749–755. Harder SD, Soukup JM, Ghio AJ, Devlin RB, Becker S (2001) Inhalation of PM2.5 does not modulate host defense or immune parameters in blood or lung of normal human subjects. Environ. Health Perspect. Suppl. 109(4):599–604. Hattis D, Wasson JM, Page GS, Stern B, Franklin CA (1987) Acid particles and the tracheobronchial region of the respiratory system––An ‘irritation-signaling’ model for possible health effects. J. Air Pollut. Control Assoc. 37:1060–1066. Hattis DS, Abdollahzadeh S, Franklin CA (1990) Strategies for testing the ‘irritation-signaling’ model for chronic lung effects of fine acid particles. J. Air Waste Manage. Assoc. 40:322–330. Hoek G, Brunekreef B, Goldbohm S, Fischer P, Van den Brandt PA (2002) Association between mortality and indicators of traffic-related air pollution in the Netherlands: a cohort study. Lancet 360:1203–1209. Holguin F, Tellez-Rojo MM, Hernandez M, Cortez M, Chow JC, Watson JG, et al. (2003) Air pollution and heart rate variability among the elderly in Mexico City. Epidemiology 14:521–527. Holland WW, Reid DD (1965) The urban factor in chronic bronchitis. Lancet 1:445–448. Huang Y-C T, Ghio AJ, Stoneheurner J, McGee J, Carter D, Grambow SC, Devlin RB (2005) The role of soluble components in ambient air fine particles-induced changes in human lungs and blood. Inhal. Toxicol. 15:327–342. Hwang J-S, Nadziejko C, Chen LC (2005) Effects of subchronic exposures to CAPs in mice: III. Acute and chronic effects of CAPs on heart rate, heart rate variance, and body temperature. Inhal. Toxicol. 17:199–207. Jerrett M, Burnett RT, Ma R, Pope CA III, Krewski D, Newbold KB, Thurston G, Shi Y, Finkelstein N, Calle EE, Thun MJ (2005) Spatial analysis of air pollution and mortality in Los Angeles. Epidemiology 16:727–736. Krewski D, Burnett RT, Goldberg MS, Hoover K, Siemiatycki J, Jerrett M, Abrahamowicz M, White WH (2000) Reanalysis of the Harvard Six Cities Study and the American Cancer Society Study of Particulate Air Pollution and Mortality. Special Report.Cambridge, MA:Health Effects Institute. Laden F, Neas LM, Dockery DW, Schwartz J (2001) Associations of fine particulate matter from different sources with daily mortality in six U.S. cities. Environ. Health Perspect. 108:941–947.

362

AMBIENT AIR PARTICULATE MATTER

Laden F, Schwartz J, Speizer FE, Dockery DW (2006) Reduction in fine particulate air pollution and mortality: extended follow-up in the Harvard Six Cities Study. Am. J. Respir. Crit. Care Med. 173:667–672. Lave LB, Seskin EP (1970) Air pollution and human health. The quantitative effect, with an estimate of the dollar benefit of pollution abatement, is considered. Science 169:723–733. Lawther PJ (1963) Compliance with the Clean Air Act: medical aspects. J. Inst. Fuel 36:341–344. Lawther PJ, Waller RE, Henderson M (1970) Air pollution and exacerbations of bronchitis. Thorax 25:525–539. Liao D, Creason J, Shy C, Williams R, Watts R, Zweidlinger R (1999) Daily variation in particulate air pollution and poor cardiac autonomic control in the elderly. Environ. Health Perspect. 107:521–525. Lipfert FW, PerryJr, HM, Miller JP, Baty JD, Wyzga RE, Carmody SE (2000)The Washington University-EPRI veterans’ cohort mortality study: preliminary results. In: Grant LD, PM2000: Particulate Matter and Health. Inhal. Toxicol. 12(Suppl. 4):41-73. Lipfert FW, Baty JD, Miller JP, Wyzga RE (2006) PM2.5 constituents and related air quality variables as predictors of survival in a cohort of U.S. military veterans. Inhal. Toxicol. 18:645–657. Lippmann M, Frampton M, Schwartz J, Dockery D, Schlesinger RB, Koutrakis P, Froines J, et al. (2003) The U.S. Environmental Protection Agency particulate matter health effects research centers program: a midcourse report of status, progress, and plans. Environ. Health Perspect. 111:1074–1092. Lippmann M, Gordon T, Chen LC (2005a) Effects of subchronic exposures to CAPs in mice. I. Introduction, objectives and experimental plan. Inhal. Toxicol. 17:177–187. Lippmann M, Gordon T, Chen LC (2005b) Effects of subchronic exposures to concentrated ambient particles (CAPs) in mice: IX. Integral assessment and human health implications of subchronic exposures of mice to CAPs. Inhal. Toxicol. 17:255–261. Lippmann M, Hwang J-S, Maciejczyk P, Chen LC (2005c) PM source apportionment for short-term cardiac function changes in ApoE / mice. Environ. Health Perspect. 113:1575–1579. Lippmann M, Ito K, Hwang J-S, Maciejczyk P, Chen LC (2006) Cardiovascular effects of nickel in ambient air. Environ. Health Perspect. 114:1662–1669. Lipsett MJ, Tsai FC, Roger L, Woo M, Ostro BD (2006) Coarse particles and heart rate variability among older adults with coronary disease in the Coachella Valley, California. Environ. Health Perspect. 114:1215–1220. Loomis D, Castillejos M, Gold DR, McDonnell W, Borja-Aburto VH (1999) Air pollution and infant mortality in Mexico City. Epidemiology 10:118–123. Maisonnet M, Bush TJ, Correa A, Jaakola JJK (2001) Relation between ambient air pollution and low birth weight in the northeastern United States. Environ. Health Perspect. 109 (Suppl. 3):351–356. Martin AE, Bradley WH (1960) Mortality, fog and atmospheric pollution: An investigation during the winter of 1958–59. Mon. Bull. Minist. Health Public Health Lab. Serv. (BG) 19:56–73. McDonnell WF, Nishino-Ishikawa N, Petersen FF, Chen KH, Abbey DE (2000) Relationships of mortality and the fine and coarse fractions of long-term fractions of ambient PM10 concentrations in nonsmokers. J, Expos. Anal. Environ. Epidemiol. 10:427–436. Morio LA, Hooper KA, Brittingham J, Li TH, Gordon RE, Turpin BJ, Laskin DL (2001) Tissue injury following inhalation of fine particulate matter and hydrogen peroxide is associated with altered production of inflammatory mediators antioxidants by alveolar macrophages. Toxicol. Appl. Pharmacol. 177:188–199. Ministry of Health (1954) Mortality and Morbidity During the London Fog of December 1952. London:Her Majesty’s Stationary Office. Nadziejko C, Fang K, Nadziejko E, Narciso SP, Zhing M, Chen LC (2003) Immediate effects of particulate air pollutants on heart rate and respiratory rate in hypertensive rats. Cardiovasc. Toxicol. 2:245–252.

REFERENCES

363

Oberd€orster G, Gelein RM, Ferin J, Weiss B (1995) Association of particulate air pollution and acute mortality: involvement of ultrafine particles. Inhal. Toxicol. 7:111–124. Ozkaynak H, Thurston GD (1987) Associations between 1980 U.S. mortality rates and alternative measures of airborne particle concentration. Risk Anal. 7:449–461. Park SK, O’Neill MS, Vokonas PS, Sparrow D, Schwartz J (2005) Effects of air pollution on heart rate variability: the VA normative aging study. Environ. Health Perspect. 113:304–309. Peters A, Wichmann E, Tuch T, Heinrich J, Heyder J (1997) Respiratory effects are associated with the number of ultrafine particles. Am. J. Respir. Crit. Care Med. 155:1376–1383. Petrovic S, Urch B, Brook J, Datema J, Purdham J, Liu L, Lukic Z, Zimmerman B, Tofler G, Downar E, Corey P, Tarlo S, Broder I, Dales R, Silverman F (1999) Cardiorespiratory effects of concentrated ambient PM2.5: a pilot study using controlled human exposures. In: Phalen RF, Proceedings of the Third Colloquium on Particulate Air Pollution and Human Health (first special issue); June, 1999; Durham, NC. Inhal. Toxicol. 12(Suppl. 1):173–188. Pope CA III, Kalkstein LS (1996) Synoptic weather modelling and estimates of the exposure–response relationship between daily mortality and particulate air pollution. Environ. Health Perspect. 104:414–420. Pope III, CA, Verrier RL, Lovett EG, Larson AC, Raizenne ME, Kanner RE , et al. (1999) Heart rate variability associated with particulate air pollution. Am. Heart J. 138:890–899. Pope III, CA, Schwartz J, Ransom MR (1992) Daily mortality and PM10 pollution in Utah Valley. Arch. Environ. Health 47:211–217. Pope III, CA, Dockery DW, Schwartz J (1995a) Review of epidemiological evidence of health effects of particulate air pollution. Inhal. Toxicol. 7:1–18. Pope III, CA, Thun MJ, Namboodiri M, Dockery DW, Evans JS, Speizer FE, Heath CW Jr (1995b) Particulate air pollution is a predictor of mortality in a prospective study of U.S. adults. Am. J. Resp. Crit. Care Med. 151:669–674. PopeIII, CA, Burnett RT, Thun MJ, Calle EE, Krewski D, Ito K, Thurston GD (2002) Lung cancer, cardiopulmonary mortality and long-term exposure to fine particulate air pollution. J. Am. Med. Assoc. 287:1132–1147. Pope III, CA, Hansen ML, Long RW, Nielsem KR, Eatough NL, Wilson WE , et al. (2004a) Ambient particulate air pollution, heart rate variability, and blood markers of inflammation in a panel of elderly subjects. Environ. Health Perspect. 112:339–345. Pope III, CA, Burnett RT, Thurston GD, Thun MJ, Calle EE, Krewski D, Godleski JJ (2004b) Cardiovascular mortality and long-term exposure to particulate matter. Circulation 109:71–77. Pritchard RJ, Ghio AJ, Lehmann JR, Winsett DW, Tepper JS, Park P, Gilmour MI, Dreher KL, Costa DL (1996) Oxidant generation and lung injury after particulate air pollution exposure increase with the concentrations of associated metals. Inhal. Toxicol. 8:457–477. Riediker M, Cascio WE, Griggs TR, Herbst MC, Bromberg P, Neas L , et al. (2004) Particulate matter exposure in cars is associated with cardiovascular effects in healthy young men. Am. J. Crit. Care Med. 169:934–940. Ritz B, Yu F, Chapa G, Fruin S (2000) Effect of air pollution on preterm birth among children born in southern California between 1989 and 1993. Epidemiology 11:502–511. Rosenlund M, Berglind N, Pershagen G, Hallqvist J, Jonson T, Bellander T (2006) Long-term exposure to urban air pollution and myocardial infarction. Epidemiology 17:383–390. Saldiva PHN, Clarke RW, Coull BA, Stearns RC, Lawrence J, Krishna-Murthy GG, Diaz E, Koutrakis P, Suh H, Tsuda A, Godleski JJ (2002) Lung inflammation induced by concentrated ambient air particles is related to particle composition. Am. J. Respir. Crit. Care Med. 165:1610–1617. Samet JM, Zeger SL, Kelsall JE, Xu J, Kalkstein LS (1997) Air Pollution, Weather, and Mortality in Philadelphia 1973–1988. Report on Phase 1. B of the Particle Epidemiology Project.Cambridge, MA:Health Effects Institute.

364

AMBIENT AIR PARTICULATE MATTER

Schmidt M, Mintz D, Rao V, McCluney L, Frank N (2005) U.S. EPA Memorandum to file. Subject: Analyses of 2001–2003 PM Data for the PM NAAQS, June 30, 2005. Available at www.epa.gov./ oar/oaqps/pm25/docs.html. Schrenk HH, Heimann H, Clayton GD, Gafater WM (1949) Air Pollution in Donora, Pennsylvania, Public Health Bull. No. 306. Washington, DC:U.S. Government Printing Office. Schwartz J (1989) Lung function and chronic exposure to air pollution: A cross-sectional analysis of NHANES II. Environ. Res. 50:309–321. Schwartz J (1993) Air pollution and daily mortality in Birmingham. Alabama. Am. J. Epidemiol. 137:1136–1147. Schwartz J, Litonjua A, Suh H, Verrier M, Zanobetti A, Syring M , et al. (2005) Traffic-related pollution and heart rate variability in a panel of elderly subjects. Thorax 60:455–461. Stolzel M, Peters A, Wichmann HE (2003) Daily mortality and fine and ultrafine particles in Erfurt, Germany. Revised Analyses of Time-Series Studies of Air Pollution and Health. Special Report. Boston, MA: Health Effects Institute. pp. 231–240. Sun Q, Wang A, Jin X, Natanzon A, Duquane D, Brook RD, Aguinqaldo J-GSFayad ZA, Fuster V, Lippmann M, Chen L-C, Rajagopalan S (2005) Long-term air Pollution exposure and acceleration of atherosclerosis and vascular inflammation in an animal model. J. Am. Med. Assoc. 294:3003– 3010. Thurston GD (1995) Personal communication of table prepared for draft EPA Criteria Document on Particulate Matter. Urch B, Brook JR, Wasserstein D, Brook RD, Rajagopalan S, Corey P, Silverman F (2004) Relative contributions of PM2.5 chemical constituents to acute arterial vasoconstriction in humans. Inhal. Toxicol.16:345–352. U.S. EPA (1993) Air Quality Criteria for Oxides of Nitrogen. EPA/600/8-91/049F. Washington, DC: U.S. Environmental Protection Agency. U.S. EPA (1996a) Air Quality Criteria for Particulate Matter. EPA/600/P-95/001F. Washington, DC: U.S. Environmental Protection Agency. U.S. EPA (1996b) Review of the National Ambient Air Quality Standards for Particulate Matter: OAQPS Staff Paper. EPA-452/R-96-013.Research Triangle Park, NC:U.S. Environmental Protection Agency. Wang X, Ding H, Ryan L, Xu X (1997) Association between air pollution and low birth weight: A community-based study. Environ. Health Perspect. 105:514–520. Ware JH, Thibodeau LA, Speizer FE, Colome S, Ferris BG Jr (1981) Assessment of the health effects of atmospheric sulfur oxides and particulate matter: Evidence from observational studies. Environ. Health Perspect. 41:255–276. Wellenius GA, Coull BA, Godleski JJ, Koutrakis P, Okabe K, Savage ST, Lawrence JE, Krishna Murthy GG, Verrier RL (2003) Inhalation of concentrated ambient air particles exacerbates myocardial ischemia in conscious dogs. Environ. Health Perspect. 111:402–408. WHO-EURO. (1987) Air Quality Guidelines for Europe. World Health Organization-Regional Office for Europe. WHO Publications, European Series No. 23, Copenhagen, Denmark. WHO-EURO. (2000) Air Quality Guidelines for Europe, Second Edition. World Health OrganizationRegional Office for Europe. WHO Publications, European Series No. 91, Copenhagen, Denmark. WHO (2005) WHO Air Quality Guidelines Global Update 2005. Report of a Working Group Meeting, Bonn, Germany, 18–20 October 2005. WHO Regional Office for Europe, Copenhagen, Denmark. Wilson WE, Suh HH (1997) Fine particles and coarse particles: Concentration relationships relevant to epidemiologic studies. J. Air Waste Manage. Assoc. 47:1238–1249.

REFERENCES

365

Woodruff TJ, Grillo J, Schoendorf KC (1997) The relationship between selected causes of postneonatal infant mortality and particulate air pollution in the United States. Environ. Health Perspect. 105:608–612. Woodruff TJ, Parker JD, Schoendorf KC (2006) Fine particulate matter (PM2.5) air pollution and selected causes of post-neonatal infant mortality in California. Environ. Health Perspect. 114:786– 790. Xu X, Ding H, Wang X (1995) Acute effects of total suspended particles and sulfur dioxide on preterm delivery: A community-based cohort study. Arch. Environ. Health 50:407–415. Zelikoff JT, Chen LC, Cohen MD, Fang K, Gordon T, Li Y, Nadziejko C, Schlesinger RB (2003) Effects of inhaled ambient particulate matter on pulmonary antimicrobial immune defense. Inhal. Toxicol. 15:131–150.

11 ARSENIC Toby G. Rossman

11.1 INTRODUCTION The word “arsenic” is derived from the Greek and Latin words for yellow orpiment, a sulfide of arsenic. Arsenic (As) has long had a reputation as a poison used for murders and suicides. An analysis of the hair of the British King George III (1738–1820) revealed an unusually high level of As, and it has been suggested that As poisoning contributed to his “madness” (Cox et al., 2005). Most people are familiar with the play “Arsenic and Old Lace,” but it was likely that the strychnine and cyanide rather than the arsenic in the elderberry wine caused the immediate death of the gentlemen callers. During the epidemic of bubonic plague in the seventeenth and eighteenth centuries, As compounds came into use as rat poisons. In the nineteenth century, As was used as a tonic, an aphrodisiac, and as an aid to digestion. It was also found in paints and dyes for clothes, cosmetics, paper, and wallpaper, most notably by the designer William Morris (Meharg, 2003). The fungus Scopulariopsis breviculis, present in damp wallpaper, was able to convert wallpaper dyes to “Gosio gas” (trimethylarsine) (Fig. 11.1), which was suspected of poisoning many people. However, unlike the highly toxic arsine, trimethylarsine is not very toxic, casting doubt on this “urban legend” (Cullen and Bentley, 2005). As has been used in medicine for centuries, and today more than 50 Chinese, Indian, and Korean folk medicines contain As. Paul Ehrlich designed Salvarsan, [(HO)C6H3(NH2) As]2
2HCl
2H2O, as the first effective cure for syphilis, and it was used until its replacement by antibiotics after World War II. Fowler’s solution (containing potassium arsenite) was used to treat skin conditions (e.g., psoriasis, eczema), leukemia, stomatitis, and as a health tonic. As was also a common contaminant of antimony-based medicines (Cox et al., 2005). When skin cancers resulted from treatment with Fowler’s solution, As was almost completely eliminated from human medicine until the recent discovery that As trioxide can be used to treat acute promyelocytic leukemia and other cancers (Miller et al., 2002; Hu et al., 2005).

Environmental Toxicants, Third Edition Edited by Morton Lippmann Copyright
2009 John Wiley & Sons, Inc.

367

368

ARSENIC

FIGURE 11.1

Some arsenic compounds found in water and food and their metabolites.

Besides their use in chemotherapy, As compounds still can be found in alternative and homeopathic medicines in Western countries.

11.2 PHYSICAL AND CHEMICAL PROPERTIES OF ENVIRONMENTAL AS AND ITS COMPOUNDS Arsenic is a ubiquitous element that is found in the atmosphere, in the aquatic environment, in soils and sediments, and in organisms. Pure As is a gray-colored semimetallic (metalloid) solid belonging to the group V elements in the periodic table, with an atomic number of 33 and an atomic weight of 74.9. The most common oxidation numbers of arsenic are 5, 3, and 3. Elemental As is very brittle and tarnishes in air. When heated, it rapidly oxidizes to As2O3 that has a garlic-like odor. In the environment, it is mostly found in compounds with other elements, as the sulfides and sulfosalts such as arsenopyrite (FeAsS), orpiment (As2S3), realgar (AsS), lollingite (FeAs2), and tennantite (Cu12As4S13). The most common mineral is arsenopyrite, which can be heated to sublime As, leaving ferrous sulfide. The two sulfides previously in common use in medicines are the yellow orpiment and the red realgar. As oxide compounds include As trioxide (As2O3), a transparent crystal or white powder that is slightly soluble in water. As trioxide resembles table sugar, a characteristic that made it a popular choice as a poison. As pentoxide (As2O5) is a white amorphous solid that is very soluble in water, forming As acid. Arsine (AsH3), a gas used in the microelectronics industry, is more toxic than other As compounds (reviewed in Carter et al., 2003). Since it is not a common environmental contaminant, its toxicity will not be discussed here, nor will that of gallium or indium arsenide, also used in industry. Levels of arsenic in the Earth’s crust are often quoted as 3 ppm, but can vary greatly depending on the types of rocks being considered (Table 11.1). As tends to be present in higher concentrations in sulfide-bearing mineral deposits and is often found in areas of gold and copper mines. It is also concentrated in hydrous iron oxides. Weathering of rocks

PHYSICAL AND CHEMICAL PROPERTIES OF ENVIRONMENTAL AS AND ITS COMPOUNDS

TABLE 11.1

369

Concentration of As in Some Rocks, Soils, and Dusts

Material Continental crust Igneous rocks Metamorphic rocks Shale (mid-Atlantic ridge) Sandstone Phosphorite Iron formations and iron-rich sediment Coals Coal fly ash from coal-fired power plants Uncontaminated mixed soils Peaty and bog soils Acid sulfide soils Soils near sulfide deposits New York State orchard soil Above, previously treated with lead arsenate Long Island, NY, potato soil treated with lead arsenate Mining-contaminated sediment Mine tailing-contaminated soils Zimbabwe gold/As mine dumps Northern Peru soil near copper mine Toronto, Canada soil near secondary lead smelter House dust (uncontaminated sites) Street dust (uncontaminated sites)

Concentration Range (mg/kg) 3.4 0.3–15.8 50.1–143 48–361 0.6–120 0.4–188 1–2900 0.3–35,000 800–2000 0.1–55 2–36 1.5–45 2–8000 1.8–3.0 1.6–141 27.8–51 (range of means) 80–1104 396–2000 9530 143–3052 17.9–3000 (range of means) 7.95–23.4 1.24–29.5

From sources summarized in Fergusson (1984), Wedepohl (1991), Smedley and Kinneburgh (2005), and World Health Organization (2001).

converts As sulfides to As trioxide and to the water-soluble oxyanion species arsenate and arsenite, the most common forms of arsenic in groundwater. Under moderately reducing conditions, arsenite is the predominant species. In oxygenated water, arsenate is the predominant species. Most of the As in the atmosphere is in the form of particulate matter. Less than 10% is present in the vapor phase or on particles smaller than 0.2 mm, and it is accumulated mainly ´ lvarez et al., 2004). However, because many microorganin fine particles (50.6 mm) (A isms can produce volatile methylated arsenicals such as MMAV, DMAV, and TMA (Fig. 11.1), all these, as well as arsine (AsH3), have been detected in samples collected close to their source (reviewed in Cullen and Reimer, 1989). The ratio of methylarsenic to inorganic As compounds is 5l0% in airborne particulate matter. Measured concentrations of As in air samples from both contaminated and noncontaminated sites are shown in Table 11.2. As in oceanic waters shows much less variability compared to fresh waters (Table 11.3) that show considerable variation with the geological composition of the drainage area and the extent of anthropogenic input. AsIII exists mostly as the neutral As(OH)3 (pKa ¼ 9.2) in neutral water. It is more mobile than AsV because the negatively charged oxyanions (H3AsO4; pKa ¼ 2.22, 6.98, 11.53) absorb more strongly on most mineral surfaces (NRC, 2000). Little is known about the adsorptive behavior of organic arsenicals.

370

ARSENIC

TABLE 11.2

Concentrations of As in Outdoor Air

Location

Particle Size, Species

Antarctic Ocean N. Pacific Ocean N. Atlantic Ocean Baltic Sea N. Chesapeake Bay, USA Urban Thessaloniki, Greece Urban Yokohama, Japan Los Angeles, USA

Wuhan City, China Caletones, Chile 510 km from 520 km from 510 km from 520 km from

Cu Cu Cu Cu

smelter smelter smelter smelter

Mean Concentration(ng/m3)

Inorganic Inorganic Inorganic 1.1 510 mm 0.45 mm 0.45 mm, inorganic 0.45 mm, organic 52.5 mm, AsIII 52.5 mm, AsV 42.5 mm, AsIII 42.5 mm, AsV 52.5 mm 42.5–510 mm

0.05 0.1 0.1 0.3–3.7 0.66 4.1 2.5 0.01 7.4 5.2 1.8 2.2 25 13

0.4 mm 0.4 mm 0.8 mm 0.8 mm

1483 14 29 5

Range 0.01–0.2 0.01–0.95 0.01–0.45 0.11–1.96 1.0–5.1 0.001–0.64 51.2–44 50.9–18.7 50.9–4.8 50.8–6.6

Data from sources summarized in World Health Organization (2001).

TABLE 11.3

Concentration of As in Some Waters

Location Seawaters Tamar estuary, UK Rhone estuary, France Huang He River estuary, China Finfeather Lake, Texas (near plant for As-based defoliants) Maurice River, NJ Upstream of pesticide plant 0.6 km downstream 4.4 km downstream 14–17 km downstream Bowrun Lake, BC, Canada Lakes, Northwest Territories, Canada (gold mining activities) Groundwater, Hungary Groundwater, SW Finland Groundwater, New Jersey Groundwater, Lagunera region, Mexico Groundwater, Chile Groundwater, West Bengal, India

Mean Concentration (mg/L)

Range

0.04–1.8 (range of means)

3.6 7900

2.7–8.8 1.3–3.8 2.8–4.3 6000–8600

3.3 2222 266 86.1 0.26

1.05–4.4 1320–4160 118–578 27.1–267 50.2–0.42 700–5500

68

1–174 17–980 1160 (max.) 8–624 470–770 3700 (max.)

1 (median)

193–737 (range of means)

Groundwater, Bangladesh Data from sources summarized in World Health Organization (2001).

510–41000

ENVIRONMENTAL EXPOSURES TO THE GENERAL POPULATION: SOURCES

371

Methods for determining total As and As species are summarized in documents by the National Research Council (NRC, 2000) and International Agency for Research in Cancer (IARC, 2004).

11.3 ENVIRONMENTAL EXPOSURES TO THE GENERAL POPULATION: SOURCES AND STANDARDS While there is still some concern about acute As toxicity, particularly as a result of occupational exposure to arsines, the major human health concern today is that of chronic low-level exposure to (mainly) inorganic As compounds. Humans are exposed to As compounds in air, water, and food coming from both natural and man-made sources. Recently, the world has become aware of the enormous disaster in the Bengal Delta Plain (Bengal region of India and neighboring Bangladesh), where millions have been exposed to high levels of As in water from tube wells. This contamination was the unfortunate result of a program designed to prevent waterborne diseases such as cholera and typhoid. In the 1970s, the World Health Organization (WHO) instituted a system of shallow tube wells to supply safe drinking water. Although water from these wells is free from dangerous bacteria, it contains high concentrations of naturally occurring As from dissolved minerals and ores. As accumulation in this region probably resulted from alluvial sediments deposited by the Ganges and other smaller rivers during the Holocene age (Mukherjee et al., 2001). As appears to be released by anaerobic metal-reducing bacteria after FeIII reduction (Islam et al., 2004). In 42 districts of southern Bangladesh and in 9 districts of neighboring West Bengal, India, very large numbers of people are exposed to drinking water with high levels of As (Guha Mazumder et al., 1998) (Table 11.4). Levels can be 1000 mg/L or higher. Other regions of the world with naturally elevated As levels in drinking water include Taiwan, Vietnam, China, Mongolia, Japan, Argentina, Chile, Bolivia, Mexico, Germany (Bavaria), Hungary, Romania, Spain, Greece, Ghana, and Canada (National Research Council, 2000). In certain areas of the U.S. West, Midwest, Southwest and Northeast, well water may contain As levels greater than 50 mg/L. Drinking water may also be contaminated with As from mining, agricultural runoff, or improperly disposed chemicals. Countries with documented drinking water contaminated by anthropogenic sources include India, Mexico, TABLE 11.4 Regions of the World with the Greatest Human Exposure to Elevated As in Drinking Water Location Bangladesh India (West Bengal) China Argentina (Chaco-Pampean Plain) Vietnam (Red River alluvial tract) Chile (Antofagasta) Mexico Hungary, Romania (Danube Basin) Taiwan

Potentially Exposed Population

As Concentration (mg/L)

3  107 46  106 42  106 2  106 4106 5  105 4  105 4  105 1.2  105

50.5–2500 510–3700 550–1860 51–7550 1–3050 100–1000 8–624 1–176 10–1820

Based on data from Mandal and Suzuki (2002), IARC (2004), and Tapio and Grosche (2006).

372

ARSENIC

Chile, Brazil, Nicaragua, and Thailand (IARC, 2004). In 2001, The U.S. EPA reduced the permissible drinking water limit for As from 50 to 10 mg/L (to take effect in 2006). The European standard and WHO guidelines are also 10 mg/L (World Health Organization, 2001), but in the developing world, many countries are struggling to reduce levels to 50 mg/L (Smith and Hira Smith, 2004). Natural sources of As in air include volcanic activity, wind erosion, sea spray, forest fires, and low-temperature volatilization (mainly biological formation of volatile arsenicals). Smelting operations and fossil fuel combustion contribute to anthropogenic sources of As (Cullen and Reimer, 1989; Smedley and Keinneburgh, 2005). The ratio of natural to anthropogenic inputs was estimated as 60:40 (Chilvers and Peterson, 1987). Natural sources of As were largely comprised of low-temperature volatilization by soil organisms with most of the remainder due to volcanic activity. Anthropogenic emissions were dominated by metal production, especially copper smelting. Other sources of air pollution by As include emissions from pesticide manufacturing facilities, cotton gins, glass manufacturing operations, tobacco smoke, and burning of fossil fuels that contain As. Occupational exposure to As occurs in those working with insecticides (calcium arsenate is used on cotton crops and sodium arsenate is used in ant killers and in animal dips), herbicides (as weed killers for railroad and telephone posts and in lawns to control weeds, and as Agent Blue in Vietnam), As-containing desiccants (to facilitate mechanical cotton harvesting), and As-containing algaecides. As is also used in glass manufacturing (lowmelting glasses) and in nonferrous alloys (to harden lead shot and to improve the toughness and corrosion resistance of copper) (Leonard, 1991; Landrigan, 1992; Garcia-Vargas and Cebrian, 1996; Bencko, 1987; Squibb and Fowler, 1983; Pinto and Nelson, 1976). A newer source of potential human exposure to As occurs as a result of the use of crystals made of gallium and indium arsenide in superconductors, semiconductors, light emitting diodes, integrated circuits, laser technology, and some transistors (Flora, 1996; Carter and Bellamy, 1988). In addition, the very toxic gas arsine (AsH3) is used to make gallium arsenide. These uses can lead to unintentional exposures among the general public from industrial effluents and/or from disposal of the As-containing products. OSHA has established an 8 h average limit of 10 mg/m3 for occupational exposure to inorganic As. NIOSH recommends a 15 min ceiling limit of 2 mg/m3. A permissible exposure limit (PEL) of 0.5 mg/m3 for organic arsenic (8 h time-weighted average) has also been established by OSHA. The amount of As in coal shows wide variation, and coal fly ash can contain high concentrations of As (Table 11.1). In Guizhou province, China, where As-rich coal is burned, people are exposed not only by inhalation of coal fly ash but also by eating hot peppers that have been dried over coal-fired stoves (Sun, 2004). The As concentration in kitchen air in this region was 160–760 mg/m3, compared to normal rural air levels of 0.0005–0.003 mg/m3 (World Health Organization, 2001). The As concentration in the hot peppers was 52.2– 1090 mg/kg. A major use of As compounds today is in chromated copper arsenate (CCA)-treated wood, although its use will probably decrease as a result of concerns about exposure. Workers handling CCA-treated wood (e.g., in making decks) may be exposed to As in the absence of adequate safety precautions, and inhalation of the combustion fumes from such wood caused severe As poisoning in Wisconsin (Aposhian, 1989). Direct contact with CCAtreated wood in wood play structures is the major contributor to arsenic on children’s hands (Kwon et al., 2004). Elevated levels of As have been found in the soil immediately adjacent to CCA-treated utility poles (153–410 mg/kg) or play structures (12.4–47.5 mg/kg) or under decks made with CCA-treated woods (76 mg/kg) (Zagury and Pouschat, 2005).

ENVIRONMENTAL EXPOSURES TO THE GENERAL POPULATION: SOURCES

373

Extensive use of As-containing pesticides and other agricultural products in the past led to high As concentrations in some soils. As compounds can accumulate in the soil because they are not biodegradable (Tamaki and Frankenberger, 1992). In the late 1980s, the U.S. EPA banned the use of many inorganic As pesticides. It is estimated that MMAV and DMAV herbicide application added 41000 metric tons of As into the environment during the 1990s (Bednar et al., 2002). Once in the soil, microorganisms can demethylate the compounds. Microorganisms can also act as As reducers and oxidizers (Rhine et al., 2005). Other sources of soil contamination include smelters, mines, and coal fly ash. Diet represents a major source of background exposure to As, which is estimated to be 16–19 mg/day (Schoof et al., 1999). Seafood is a major source of As within the diet. Arsenic concentrations in U.S. saltwater finfish can be as high as 6 mg/g, but the forms of As in seafood are primarily the less toxic arsenocholine and arsenobetaine (Fig. 11.1), which are excreted in the urine without metabolism (Vahter et al., 1983). Shrimp contains some DMA, and arsenosugars in seaweed can be metabolized to 12 (mostly unidentified) metabolites, with DMA being the most abundant (67% in urine) (Ma and Le, 1998; Francesconi et al., 2002). Because of the high concentrations of As in seafood, the U.S. government has defined the upper permissible limit of contamination to be 2.6 ppm total As (Wesbey and Kunis, 1981). A market basket survey of 40 foods expected to supply at least 90% of U.S. daily As intake found that total As in seafood ranged from 0.160 mg/g in freshwater fish to 2.36 mg/g in saltwater fish, but the average inorganic As levels were only 51 to 2 ng/g (Schoof et al., 1999). Among the grains in the United States, raw rice contains by far the highest levels of As (0.24–0.30 mg/g total As, of which 10–67% can be inorganic) compared to flour (0.030 mg/g) and corn meal (0.039 mg/g) (Schoof et al., 1999; Williams et al., 2005). Rice grown in As-contaminated water has increased levels of As, up to a value of 0.95 mg/g in Bangladesh (Williams et al., 2005), and additional As can be consumed when the rice is also cooked in As-contaminated water. In the Unites States, among fruits, the highest As concentrations were in grape juice (0.058 mg/g and watermelon (0.040 mg/g) and the highest concentrations in vegetables were in tomatoes, cucumbers, and onions (0.099, 0.096, and 0.096 mg/g respectively) (Schoof et al., 1999). A field test of vegetables grown in four As-contaminated sites in the UK measured the transfer coefficients for As uptake (concentration of As in crop/ concentration of As in soil). Values ranged from 0.0007 to 0.032. Highest values were for radishes 4 calabrese (broccoli) 4 lettuce. Beetroot, cauliflower, and potato showed low uptake. Uptake was increased by high sand content and low clay content of soil, and soil contaminated by air deposition of As (Warren et al., 2003). Roxarsone (3-nitro-4-hydroxyphenylarsonic acid) has been used in agriculture to fatten swine and poultry, and as an anticoccidial agent. Roxarsone, a nontoxic form of As, is eliminated from the diet for a few days before slaughter, and the amount of As in chicken in the United States in 1994–2000 was estimated (based on liver samples) at 0.33–0.43 mg/g (Lasky et al., 2004) and measured (4 samples in 1999) as 0.086 mg/g, almost all of which was organic (Schoof et al., 1999). The FDA allows up to 2 mg/g As in chicken. The excretion of Roxarsone is very rapid and the product is unchanged except for 18% conversion to 3-amino-4hydroxyphenylarsonic acid (Moody and Williams, 1964). Chicken litter is used as fertilizer, and microbial action converts the organic compounds to arsenate (Garbarino et al., 2003). This is now causing problems due to runoff into the Chesapeake Bay and possible leaching into groundwaters (Nachman et al., 2005). Elevated levels of As in soil (due to either natural or man-made contamination) may lead to exposure from ingested soil. This is of particular concern for small children who swallow bits of soil while playing. The bioavailability of As in soil or other materials (rate and extent

374

ARSENIC

of uptake to a target organ) is a function of its bioaccessability (dissolution in bodily fluids such as saliva, gastric fluid, and intestinal fluid) and its absorption through the intestine. Using a standard reference soil containing 626 mg As/g soil, bioaccessability was estimated at 65.9% and bioavailability at 34.3–41.2% (Ellickson et al., 2001).

11.4 PATHWAYS AND KINETICS FOR IN VIVO UPTAKE, DISTRIBUTION, AND ELIMINATION The major routes of As uptake are via ingestion of food and water and via inhalation of polluted air by the lung and, to a much lesser extent, via dermal absorption (Aposhian and Aposhian, 1989). The form and oxidation state of ingested As will depend on its source. As from deep wells will be predominantly arsenite (AsIII), whereas surface water will be predominantly arsenate (AsV). Nothing is known about the routes of uptake into intestinal epithelium and subsequent transport into blood. In other cells, aquaglyceroporins 9 and to some extent 7 (which usually transport water and glycerol) are effective arsenite transporters (Liu et al., 2004b), although the major pathway for arsenite uptake in yeast, and probably mammalian cells, is catalyzed by hexose permeases (Liu et al., 2004a). Approximately 95% of soluble trivalent As compounds are absorbed from the GI tract (Leonard, 1991). In the case of the airborne pollutant As2O3, intestinal absorption following mucociliary clearance of particles deposited in the conductive airways to the esophagus depends on a number of factors such as particle size and gastric juice pH (Fowler et al., 1979). In addition, As may be dermally absorbed as demonstrated by the occurrence of systemic toxicity following accidental topical exposure to arsenic acid and arsenic trichloride (Garb and Hine, 1977). After absorption through the lungs or the GI tract, As is transported in the blood to other parts of the body and distributed to the kidney, liver, spleen, skin, hair and nails in that order (Fowler, 1977). In humans who are not exposed to high levels of As, the highest accumulated tissue concentrations of As are found in hair and nails (Liebscher and Smith, 1968). Approximately 70% of As is excreted, mainly in urine but also via bile in the feces. Other routes of elimination, although less important, are via the skin, hair, nails, and sweat. The half-life of inorganic As in humans is about 10 h (Leonard, 1991). In liver, arsenite is methylated by enzymatic transfer of the methyl group from S-adenosylmethionine, to the less toxic methylarsonate (MMAV) and dimethylarsenate (DMAV) (reviewed in (Aposhian et al., 2004; Styblo et al., 2002). Reduction of arsenate to arsenite is necessary before methylation can occur, and this reaction requires glutathione (GSH) (Miller et al., 2002; Vahter et al., 1983). Arsenite has been shown to be methylated by AsIII methyltransferase (AS3MT) using S-adenosylmethionine (SAM), thioredoxin, and a thiol. Human metabolism of arsenic is as follows: Oxidative methylation: arsenite þ SAM ! MMAV Reduction: MMAV þ thiol ! MMAIII Oxidative methylation: MMAIII þ SAM ! DMAV Reduction: DMAV þ thiol ! DMAIII Humans excrete a mixture of inorganic, monomethylated, and dimethylated forms of As. Most humans excrete at least 50% dimethylated and 25% monomethylated arsenic, the remainder being inorganic (Buchet et al., 1981), but the proportions can vary among different

As ESSENTIALITY

375

individuals. In fibroblasts, and probably in other nonmethylating cells, protection against arsenite appears to be accomplished by specific mechanisms for arsenite efflux (Winski and Carter, 1995) catalyzed by members of the MRP (multidrug resistance-associated protein) group of the ABC super family of transport ATPases (Haimeur et al., 2004). In human liver, a major route of As detoxification is MRP2-catalyzed extrusion of the As-triglutathione complex As(GS)3 into bile (Kala et al., 2000). Arsenite is more toxic than arsenate, probably due, at least in part, to different rates of cellular uptake (Bertolero et al., 1987; Dopp et al., 2004). Arsenate (AsV) is similar in structure to inorganic phosphate, and is known to inhibit metabolic reactions in mitochondrial oxidative phosphorylation by substituting for inorganic phosphate with subsequent formation of an unstable arsenate ester that decomposes spontaneously. This “arsenolysis reaction” has the effect of uncoupling ATP synthesis in oxidative phosphorylation and also in glycolysis in the reaction catalyzed by glyceraldehyde 3-phosphate dehydrogenase (Vahter et al., 1983). Exposure of rodents to arsenate results in hepatic mitochondrial damage (Aposhian, 1989). Arsenite does not compete with phosphate, but instead tends to bind to dithiol groups (Fowler, 1977). The chemical warfare agent Lewisite (chlorvinyldichloroarsine) that was used during World War I prompted research into arsenite–thiol interactions. Some enzymes reacted with Lewisite by forming a ring structure involving As and two thiol groups. A search for a dithiol antidote that could displace arsenite from these enzymes led to the discovery of British Anti-Lewisite and the natural dithiol compound lipoic acid, a cofactor of the pyruvate dehydrogenase (PDH) multienzyme complex (Bencko, 1987; Vahter et al., 1983). Like Lewisite, arsenite exerts its toxic effects by reacting with vicinal thiols in the cell. These thiols can exist on adjacent carbon atoms, as in lipoic acid, or on proteins where two cysteine residues act as vicinal thiols by coming into close proximity through the folding of the protein molecule. Arsenite also binds specifically to hormone receptors that have vicinal thiol groups, and interferes with their activities (Lopez et al., 1990; Hoffman and Lane, 1992; Kaltreider et al., 2001). Compounds that contain vicinal thiol groups, such as meso2,3-dimercaptosuccinic acid (DMSA) and 2,3-dimercapto-1-propanesulfonic acid (DMPS), are used in chelation therapy for treating As intoxication (Aposhian et al., 1995). In contrast to the pentavalent As metabolites, which are less toxic than arsenite, the trivalent methylated As metabolites MMAIII and DMAIII are generally more toxic (Styblo et al., 2000; Petrick et al., 2000). Methylated trivalent metabolites are highly reactive and are more potent inhibitors of glutathione reductase (Styblo et al., 1997) and thioredoxin reductase (Lin et al., 1999) than are arsenite, MMAV or DMAV. Inhibition of these enzymes may alter cellular redox status and lead to changes in cell signaling, lipid peroxidation and DNA damage.

11.5 As ESSENTIALITY There is some evidence that As may be an essential element for humans, based on experiments on As-deprived animals (chick, goat, hamster, pig, rat) (reviewed in Uthus, 1992; Uthus and Nielson, 1993; Nielsen, 1996). For example, goats fed a diet deficient in As showed a substantial increase in the mortality rate as compared to the control group (Anke et al., 1980). After weaning, 60% of the As-deficient kids died within 140 days after birth. In addition, As deficiency resulted in decreased fertility, an increased spontaneous abortion rate, and a decreased hematocrit. The As-deprived goats showed evidence of muscle atrophy,

376

ARSENIC

reduction in oxidative enzymes, and abnormal mitochondria in muscle (Schmidt et al., 1984). Unfortunately, many reports of As essentiality exist only in abstracts or in meeting reports, and not in peer-reviewed journals. Exceptions are studies by Uthus (1990, 1992, 2003 that suggest that arsenic has a role in methionine/methyl metabolism. As-deprived rats had decreased plasma taurine, hepatic polyamines, and S-adenosylmethionine decarboxylase (needed for polyamine synthesis). As deprivation, as well as As excess, caused DNA hypomethylation in rat liver and in Caco-2 cells. As deprivation or excess also increased the formation of aberrant colon crypts in rats treated with dimethyl hydrazine (Uthus and Davis, 2005), suggesting a cocarcinogenic effect. So far, no biochemical mechanism linking As and methionine/methyl metabolism has been discovered, but the fact that many laboratories have reported effects of arsenite on DNA methylation makes this an important area of study.

11.6 HEALTH EFFECTS AND EXPOSURE–RESPONSE RELATIONSHIPS Acute symptoms of poisoning with inorganic As compounds are usually seen in humans who have ingested contaminated food or drink. The fatal human dose for ingested As trioxide is 70–180 mg or about 600 mg/kg/day (Leonard, 1991; Landrigan, 1992; Garcia-Vargas and Cebrian, 1996; ATSDR, 2000). Acute symptoms are characterized by profound gastrointestinal inflammation, sometimes with hemorrhage, and can include constriction of the throat followed by dysphagia, gastric pain, vomiting, diarrhea, dehydration, leg cramps, irregular pulse, shock, stupor, paralysis and coma (Leonard, 1991; Landrigan, 1992; GarciaVargas and Cebrian, 1996). The cause of death is massive fluid loss from the GI tract, resulting in severe dehydration, reduced blood volume and circulatory collapse. In survivors, electrocardiograph changes are characterized by T-wave inversion and persistent prolongation of the Q–T interval (Aposhian, 1989). Exfoliative dermatitis, peripheral neuritis, cardiac abnormalities, and reversible anemia and leucopenia may develop. As compounds can also cause contact dermatitis and hepatotoxic effects. Around 1900 in Manchester, England, Ascontaminated beer caused by contaminated sugar resulted in 6000 poisonings and approximately 71 deaths. Most of the patients presented with anorexia, brown pigmentation, peripheral neuritis characterized by muscular weakness, pain and parasthesis of the extremities, hepatic lesions, localized edema and fatty degeneration of the heart (Aposhian, 1989). Transverse white lines across the nails (Mees’ lines) often appear weeks after an episode of acute poisoning. Chronic exposure to As, particularly via drinking water, is now a major worldwide concern. Skin lesions are the earliest nonmalignant effect of chronic exposure to As (Yoshida et al., 2004; Guha Mazumder, 2003). The most characteristic effects of As ingestion are areas of hyperpigmentation interspersed with smaller areas of hypopigmentation (raindrop appearance) on the trunk and neck and (somewhat later) hyperkeratosis of palms and soles characterized by small corn-like elevations and diffuse keratosis (Tseng et al., 1968; Centeno et al., 2002). Pigment changes are often seen on the face, neck and back. In Taiwan, exposures of 4100 mg/L As in drinking water induced skin lesions with an average latency period of 23 years (range 10–42) (Haque et al., 2003). Hyperpigmentation is the most sensitive biomarker for assessing effects of As (Yoshida et al., 2004; Guha Mazumder, 2003), but does not occur in every exposed individual, suggesting possible genetic differences in susceptibility. In Bangladesh, 50% of individuals from two As-contaminated villages had some skin manifestations (Kadono et al., 2002), but these included very mild lesions that might not have been counted in other studies.

HEALTH EFFECTS AND EXPOSURE–RESPONSE RELATIONSHIPS

377

As exposure results in a different prevalence and spectrum of diseases in different areas of the world. This makes it difficult to establish strong associations between exposure and disease. Different Expert Committees have disagreed on the strength of evidence for the associations between As exposure and some diseases (National Research Council, 2000; World Health Organization, 2001; IARC, 2004). For example, long-term, low level ingestion of As-contaminated water leads to blackfoot disease in Taiwan, but not in other parts of the world. Blackfoot disease is a progressive loss of circulation in hands and feet, ultimately resulting in necrosis and gangrene. In several countries, As causes less severe forms of peripheral vascular disease, such as Raynaud’s disease and thickening of small and mediumsized arteries throughout the body in children who were autopsied in Chile (Garcia-Vargas and Cebrian, 1996). In some populations, chronic As exposure affects the cardiovascular system, causing cardiac arrhythmias and hypertension. Gastrointestinal symptoms, less severe than in acute poisoning, are seen in some exposed populations. Several studies have reported swollen and tender livers and sometimes evidence of liver damage in those chronically exposed. In an examination of the clinical features of 248 patients from the West Bengal area of India who exhibited signs of chronic As dermatosis, enlargement of the liver was seen in 76.6% of those who drank the contaminated water (Guha Mazumder, 2003). Polyneuritis and motor paralysis, primarily of the fingers and toes, may occur as the only symptoms in As exposed individuals (Bencko, 1987). Hyperkeratosis and hyper pigmentation are not commonly seen in workers exposed to As primarily by inhalation. Early symptoms include weakness, loss of appetite, nausea, and diarrhea. Later, there is often conjunctivitis, inflammation of the mucous membranes of the nose, larynx and respiratory passages, mild tracheobronchitis, skin lesions and sometimes, at high levels of exposure, perforation of the nasal septum. Chronic exposure may lead to peripheral neuropathy (both of sensory and motor neurons) and sometimes anemia, and leucopenia. An increased incidence of Raynaud’s disease at 0.05–0.5 mg/m3 airborne As has been reported (Garcia-Vargas and Cebrian, 1996). Epidemiological evidence indicates that As is a human skin, lung, and bladder carcinogen (IARC, 2004). Squamous cell carcinoma (SCC), basal cell carcinoma (BCC) and Bowen’s disease (squamous cell carcinoma in situ), but not melanoma, are associated with As in drinking water (Guo et al., 2001). In Taiwan, As levels 4 640 mg/L were associated with increased risk of BCC in men and SCC in both men and women (Guo et al., 2001), and the risk of developing skin cancers as well as internal cancers was increased in individuals who had hyperkeratosis or hyper pigmentation (Yeh, 1973; Tsuda et al., 1995). Skin cancers also increase the risk of later development of internal cancers in Taiwan (Tsuda et al., 1995; Chiou, 1995). Factors other than As in drinking water may contribute to the high incidence of bladder cancers in Taiwan, since the incidence is high even without As exposure (Yang et al., 2002). Men in India, Bangladesh, Japan, and Taiwan had a higher prevalence of cancers and/ or benign lesions compared to women at all levels of exposure (Tseng et al., 1968; Kadono et al., 2002; Tsuda et al., 1995; Chen et al., 2003a; Watanabe et al., 2001). It was suggested that men drink more water per day than women. However, urinary As levels in females were higher than those in their male partners drinking from the same wells, suggesting that women excrete As more efficiently than men (Watanabe et al., 2001). Men generally have more sun exposure than women, who spend more time indoors and wear more clothing. Skin cancer in Taiwan was associated with both As exposure and sun exposure, and skin lesions (thought to be precursors of skin cancers) in Bangladesh were associated with As exposure, sun exposure, smoking, and exposure to fertilizers (Chen et al., 2003a, 2006). The sites of tumors differ between males and females in that region (Yeh et al., 1968). Men work outdoors

378

ARSENIC

as farmers, fishermen and in salt flats, usually wearing only shorts, whereas women spend more time indoors and wear more clothing out of doors (Lung Chi Chen, personal communication). Hence men receive more sun exposure on chest, abdomen, and back where more tumors appear, in contrast to women who have few tumors on the chest or abdomen (Yeh et al., 1968). These results support the hypothesis that As in drinking water is synergistic with sunlight in causing skin cancer. Smoking increased the risk of As exposure for mortality from lung cancer in As-exposed Swedish copper smelter workers (Alain et al., 1993), in miners exposed to radon (ionizing radiation) and As-containing dusts (Xuan et al., 1993), and in Japanese and Chilean populations exposed to As in drinking water (Tsuda et al., 1995; Ferreccio et al., 2000). In a Taiwan population exposed to As in drinking water, there was no increased risk of lung cancer in nonsmokers, but a relative risk of 2.45 in smokers (Chiou, 1995). Until recently, there were no animal models for As carcinogenesis (National Research Council, 2000; IARC, 2004; Milner, 1969). Recently, two nontraditional approaches, transplacental carcinogenesis and cocarcinogenesis, have been developed for studying the role of As in carcinogenesis. Sodium arsenite in the drinking water (42.5 and 85 ppm) of pregnant C3H mice was a transplacental carcinogen, inducing hepatocellular carcinoma and benign adrenal tumors in male offspring and ovarian and lung tumors in female offspring (but no skin tumors) (Waalkes et al., 2003). In the cocarcinogenesis protocol, hairless Skh1 mice were given sodium arsenite in drinking water and irradiated three times per week with solar spectrum ultraviolet light (UV). Arsenite increased the skin tumor (squamous cell carcinoma) yield in a dose/dependent manner (the lowest dose was 1.25 ppm) compared to mice exposed to UV alone (Rossman et al., 2001; Rossman et al., 2004; Burns et al., 2004). No tumors appeared in any organs in Skh1 mice given arsenite alone, even after a lifetime of exposure. Tumors in mice given UV plus arsenite appeared earlier and were much larger and more highly invasive than those in mice receiving UV alone (Rossman et al., 2001). The carcinogenicity of DMAV in animals has been reviewed by Kenyon and Hughes (2001). Humans are exposed to DMAV (also called cacodylic acid) from its use in herbicides, as well as via metabolism of inorganic arsenic, but it is not known whether DMAV is a human carcinogen. In the rat, high concentrations of DMAV (10,000 mg/L and higher in drinking water) act as a promoter of carcinogenesis in a number of organs (but not in skin), with the strongest effect on the bladder. At even higher doses (50,000 mg/L), DMAV was a complete bladder carcinogen, causing transitional cell carcinoma at 50 ppm in water after 2 years (Wei et al., 1999; Cohen et al., 2002). No tumors were seen in any other organs. The high doses of DMAV used in rats cause necrosis and sustained cell proliferation in the bladder epithelium (Cohen et al., 2002). The urine of these rats contains DMAIII at levels that cause cytotoxicity in vitro, suggesting that DMAIII may mediate the necrosis (Cohen et al., 2001). MMAV was not tumorigenic in either rat or mouse (Arnold et al., 2003). The genetic toxicology of As compounds has been reviewed (Basu et al., 2001; Rossman, 2003). As compounds do not form DNA adducts and are not significantly mutagenic at doses giving high survival (Rossman, 2003; Li and Rossman, 1991). At high (toxic) concentrations, trivalent As compounds can induce chromosome aberrations (Kligerman et al., 2003) and micronucleus (MN) formation. At lower concentrations, aneuploidy (change in chromosome number from the normal diploid or haploid number other than an exact multiple), gene amplification, and DNA methylation changes (which affect gene expression) are seen (Sciandrello et al., 2002). Recently, a delayed mutagenic response (after 15 generations) to very low doses of arsenite (but not to MMAIII) was demonstrated in human osteosarcoma (HOS) cells (Mure et al., 2003), suggesting that this “genotoxicity” may be a secondary

HEALTH EFFECTS AND EXPOSURE–RESPONSE RELATIONSHIPS

379

result of epigenetic effects such as DNA methylation changes or aneuploidy. Arsenite induces transformation into malignancy of HOS and other cell types as well (Huang et al., 1999; Achanzar et al., 2002; Zhao et al., 1997; Takahashi et al., 2002; Mure et al., 2003; Chie et al., 2004). In some cases, transformation was associated with altered DNA methylation (Zhao et al., 1997; Takahashi et al., 2002; Mass and Wang, 1997; Zhong and Mass, 2001). The generation of reactive oxygen species (ROS) and nitric oxide (NO) has been shown to play an important role in As-induced cell injury (reviewed in Rossman, 2003), and may be involved in the etiology of cancer, atherosclerosis, and perhaps diabetes (Lee and Ho, 1995; Barchowsky et al., 1999a; Lynn et al., 2000; Klaunig and Kamendulis, 2004; Simeonova and Luster, 2004; Kaneto et al., 2005). Oxidative DNA damage (8-oxo-dG) levels were increased in the bladders of rats receiving 200 ppm DMAV for 2 weeks (Wei et al., 1999). Oxidative damage to DNA has been shown to be mutagenic (Klaunig and Kamendulis, 2004) and to be affecting DNA methylation (Cerda and Weitzman, 1997), both important for cancer causation and possibly atherosclerosis. It is important to note that low concentrations of arsenite can affect cell signaling (e.g., oxidant-dependent activation of nuclear factor kB (NFkB)) in the absence of DNA damage (Barchowsky et al., 1999). The increased oxidants appear to result from activation of membrane-bound NAD(P)H oxidase. Antioxidants protect cultured cells against arsenite toxicity and genotoxicity (Hei et al., 1998; Nordenson and Beckman, 1991; Lee and Ho, 1994; Bau et al., 2002; Rossman and Uddin, 2004). A number of mechanisms may be responsible for the ability of arsenite to enhance carcinogenesis. Although not a mutagen, arsenite can enhance the mutagenicity of other agents, including UV light (Lee et al., 1985, 1986; Li and Rossman, 1989a; 1989b; Yang et al., 1992; Wiencke and Yager, 1992; Jha et al., 1992; Wu et al., 2005). This may occur by interference with DNA repair (Li and Rossman, 1989a, 1989b; Hu et al., 1998; Vogt and Rossman, 2001; Hartwig et al., 2003). Interference with tumor suppressor P53 activity might be responsible for the ability of arsenite to override the growth arrest normally caused by UV (Vogt and Rossman, 2001; Hartwig et al., 2002; Mudipalli et al., 2005). Low concentrations of arsenite also stimulate cell proliferation in vitro (Germolec et al., 1997; Komissarova et al., 2005; Germolec et al., 1996; Barchowsky et al., 1999; Trouba et al., 2000) and in vivo (Burns et al., 2004; Germolec et al., 1998; Simeonova et al., 2000). Angiogenesis is stimulated by doses as low as 0.033 mM arsenite in the chick choreoallantoic membrane assay. Low concentrations of arsenite can block apoptosis (programmed cell death) induced by a second agent (Pi et al., 2005; Wu et al., 2005). This might allow survival of cells with DNA damage, thus facilitating tumorigenesis (Gerl and Vaux, 2005). Exposure to As has been associated with increased risk of vascular disease. Carotid atherosclerosis was increased in individuals exposed to As in drinking water (Wang et al., 2002). Blackfoot disease, an extreme form of peripheral vascular disease characterized by severe systemic atherosclerosis followed by gangrene in the lower extremities, is seen in Asexposed individuals in Taiwan, who are also at increased risk for ischemic heart disease and stroke (Tseng et al., 2005) as well as hypertension and diabetes mellitus (Chen et al., 1995; Tseng et al., 2000). Increased lipid peroxidation was found in the serum of individuals in China exposed to As in drinking water (Pi et al., 2002) and in rats exposed to arsenite (Ramos et al., 1995). Oxidation of low-density lipoprotein resulting from overproduction of ROS is thought to be an early event in atherosclerosis (Stocker and Keaney, 2004). Arsenite-induced oxidative signaling results in expression of inflammatory cytokines such as IL-8 that can mediate atherogenesis (Simeonova and Luster, 2004). It is not clear whether As compounds act alone in increasing atherosclerosis by increasing ROS. Arsenite is atherogenic in

380

ARSENIC

ApoE / mice (which are prone to develop atherosclerosis), but not in wild-type mice (Simeonova and Luster, 2004). This suggests that As may be a modifier of atherogenesis.

11.7 BIOMARKERS OF EXPOSURE, SUSCEPTIBILITY, AND EFFECT Because As is rapidly cleared from the blood, blood As levels are not useful as biomarkers of exposure. As in urine is considered the most reliable marker of recent (up to one week) exposure to As (ATSDR, 2000). An important limitation to the use of total urinary arsenic as a biomarker of exposure is the presence of organic As compounds in urine after ingestion of certain seafoods (see above). Excretion of 4200 mg/day As indicates a toxic intake unless seafood was consumed up to 48 h before sampling (Buchet et al., 1994). Thus, total urinary As is not as useful a biomarker of exposure as is speciated urinary As. Usually, inorganic As and its metabolites (MMA and DMA, both trivalent and pentavalent) are used. Background levels of these compounds are usually below 10 mg/L (NRC, 2000). This amount increases with exposure to As in drinking water (as well as to airborne As), but there is no simple relationship, since the amounts of urinary As will also be affected by the amount of water consumed, amounts used in cooking, and amounts of other fluids consumed. Arsenic tends to accumulate in hair and nails, and measurement of As levels in these tissues have been useful indicators of recent (up to 6–12 months, depending on hair length) exposures. However, analysis of hair As can be misleading because it is easily contaminated by air pollution. This can be minimized by collecting samples close to the scalp or from unexposed areas and by careful washing before analysis. Extensive washing of nails to remove surface As is also necessary. In Bangladesh, hair from As-exposed individuals showed a mean value of 3.39 mg/kg As (range of 0.280–28), compared to 0.08–0.25 in the unexposed. The level of As in hair in unexposed individuals is usually 51 mg/kg (Rahman et al., 2001). However, neither toenails nor hair As levels can be reliably related to As dose because of problems with external contamination (Harkins and Susten, 2003; Hinwood et al., 2003). Concerning biomarkers of effect, As-exposed humans show increased micronuclei (MN) and sometimes chromosome aberrations in lymphocytes, exfoliated bladder epithelial cells, and buccal epithelial cells (reviewed in Basu et al., 2002). These effects may represent toxicity and/or possible genotoxicity as a result of oxidant stress. Increased oxidant stress has been seen in As-exposed individuals in a number of different assays. Skin samples from individuals with As-related skin lesions had a higher frequency of the oxidative DNA lesion 8-oxo-dG (detected by immunohistochemistry) compared to skin samples from patients with non-As-related lesions (Matsui et al., 1999). Serum lipid peroxides and decreased levels of nonprotein sulfhydryls (indicating oxidant stress) were found in a Chinese population chronically exposed to As in drinking water (Pi et al., 2002). In Taiwan, As exposure was associated with reactive oxidants in plasma and inversely associated with plasma antioxidant capacity (Wu et al., 2001). Genetic factors appear to play a role in sensitivity to As. In Taiwan, subjects with blackfoot disease have a significantly higher risk of developing cancers compared to others drinking the same water developing blackfoot disease (Chiou et al., 1995). This is true even after controlling for cumulative As exposure, and points to interindividual variations in susceptibility to As toxicity and carcinogenicity. A number of laboratory studies reveal the existence of heterogeneity in the response of human cells to As compounds. Lymphoblasts from normal humans showed a wide range of sensitivities to the toxic effects of arsenite

MITIGATING EFFECTS AND CONTROLLING EXPOSURES

381

(Li et al., 2004). Large interindividual variations are also seen for arsenite-induced chromosomal aberrations and aneuploidy in peripheral blood lymphocytes from human donors (Wiencke and Yager, 1992; Vega et al., 1995). Interindividual variations in the rate of As methylation were observed in primary human hepatocytes (Drobna et al., 2004). Regarding biomarkers of susceptibility, many molecular epidemiological studies have found differences in As methylation capacity between population groups and individuals, based on urinary metabolites (Hopenhayn-Rich et al., 1998; Chiou et al., 1995; Concha et al., 2002; Meza et al., 2005; Loffredo et al., 2003; Marnell et al., 2003). A low rate of As methylation is related to a low rate of excretion (Vahter and Concha, 2001). In general, the fraction of urinary DMA decreases and that of MMA increases with total urinary As (reviewed in Vahter, 2002), most likely because high MMAIII levels inhibit further methylation (Styblo et al., 2002). Case-control studies of As-exposed populations show that cases with arsenicosis (including cancers) tend to have higher urinary inorganic As and MMA levels and lower DMA levels compared to controls (Valenzuela et al., 2005; Yu et al., 2000; Chen et al., 2003b). A specific polymorphism in the hGSTO1 gene, whose product can reduce MMAV to MMAIII, was associated with reduced As methylation in a Mexican population (Marnell et al., 2003). A polymorphism in AS3MT (AsIII methyltransferase) was associated with altered methylation, but only in children (Meza et al., 2005). Surprisingly, the cytotoxicity of arsenite to cultured cells did not correlate with their abilities to methylate arsenite (Styblo et al., 2002), suggesting that other mechanisms such as transport, antioxidant defenses, or resistance to apoptosis might be more important in protecting cells than methylation. Other genetic polymorphisms now under investigation include those in genes involved in pathways for DNA repair, protection against oxidant stress, GSH metabolism, methylation, and transport (Ahsan et al., 2003; Chen et al., 2005; Ghosh et al., 2006).

11.8 MITIGATING EFFECTS AND CONTROLLING EXPOSURES In the United States, As-associated excess cancer mortality has not generally been seen (Cantor, 1997; Schoen et al., 2004). This may be due, in part, to the fact that few Americans are exposed to 4100 ppb As in drinking water. In Taiwan, As levels 4640 ppb were associated with increased risk of BCC in men and SCC in both men and women (Guo et al., 2001). Also, most Americans drink beverages other than local drinking water, while individuals in the less-developed world are more dependant upon local drinking water. In addition, there are other important factors, such as poor nutritional status, exposure to sunlight, and chronic liver disease, which enhance the toxicity of As (Everall and Dowd, 1978; Hsueh et al., 1995), but are less prevalent in the United States. Improving the nutritional status of As-exposed individuals is expected to mitigate the effects of exposure. There is extensive evidence demonstrating that As-induced disease is increased in individuals who are malnourished or undernourished, possibly because As methylation is compromised (Hsueh et al., 1995; Mitra et al., 2004; Steinmaus et al., 2005). Individuals with low body weight were at increased risk for As-induced hyperkeratosis in India (Guha Mazumder et al., 1998). Women who were 20% below standard body weight showed a 2.1-fold higher age-adjusted risk for keratosis compared to women with normal body weight. For underweight men, the relative risk was 1.5-fold (Guha Mazumder, 2003). Undernourishment may result in increased uptake of arsenite into liver. In rats, the expression in liver of AQP9, which transports arsenite into cells, is elevated 20-fold by starvation (Carbrey et al., 2003). Studies in yeast show that glucose and other hexoses

382

ARSENIC

compete with arsenite for transport by the hexose transporters (Liu et al., 2004). If the same is true in humans, then poor nutrition could increase arsenite uptake into the liver and perhaps other cells, and overwhelm intrinsic detoxification mechanisms such as methylation. Because arsenic metabolism involves methylation, dietary methyl group donors such as folate might be important in As detoxification. Rabbits that have been maintained on a diet low in methionine, choline, or protein show reduced rates of As excretion, indicating reduced rates of methylation (Vahter and Marafante, 1987). Folate deficient mice are more susceptible to arsenite-induced micronuclei formation (McDorman et al., 2002). It is reasonable to propose that methyl group donors, such as folate, might be important in As detoxification in humans. However, in cells that do not methylate As, folate might also be important for other reasons. Folic acid deficiency promotes and folate supplementation lowers the risk of cancers in rodents and humans (reviewed in Mason and Levesque, 1996). Although many mechanisms have been proposed to explain the promotion of carcinogenesis by folate deficiency, most attention has concentrated on alterations of DNA metabolism. Decreased synthesis of thymidylate due to a lack of folate leads to extensive incorporation of uracil into DNA and repair-induced strand breaks (Blount et al., 1997). Folate deficiency also impairs DNA repair (Choi and Mason, 2000) and alters the pattern of DNA methylation (Pogribny et al., 1997). In Taiwan, where skin cancer prevalence was as high as 6.1%, chronic carriers of hepatitis B antigen with liver dysfunction were at increased risk, perhaps because As metabolism to the methylated species was deficient (Hsueh et al., 1995). Other factors increasing the risk of skin cancers in Taiwan include working in salt fields, cigarette smoking, and alcohol consumption (Hsueh et al., 1995). It has been suggested that in some parts of the world with high As in the drinking water, low selenium (Se) levels in soil may exacerbate As toxicity and carcinogenicity (Spallholz et al., 2004). Since 1938, there is evidence that As and Se might be mutually antagonistic (reviewed in Zeng et al., 2005). Se and As have adjacent positions on the periodic table and have similar chemical properties as metalloids. Due to low levels of Se in soil, the crops and forage in some areas of the world can be Se-deficient. Se is a trace element needed for antioxidant defense or redox regulation because of its essential role in a number of proteins including glutathione peroxidase and thioredoxin reductase. Dietary Se compounds protect human cells from arsenite-induced delayed mutagenesis (Rossman and Uddin, 2004) and the synthetic organoselenium compound pXSC as well as a-tocopherol (Vitamin E) protect mice from arsenite cocarcinogenesis (Uddin et al., 2005). In As-exposed individuals in Taiwan, the percentage of inorganic As in urine was reduced while the percentage of DMA was increased with the concentration of urinary Se and serum a-tocopherol, suggesting that higher Se levels promote the methylation of As (Hsueh et al., 2003). Arsenite and selenite enhance the biliary excretion of metabolites of each other, possibly through formation of a diglutathione compound [(GS)2AsSe] (Gailer et al., 2000). It may take many years before the drinking water in some parts of the world will have safe levels of As. Meanwhile, a combination of Se and a-tocopherol (and/or other antioxidants) may prove useful and safe in ameliorating the effects in As-exposed individuals. Because trivalent As compounds can form chelates with vicinal thiol groups, compounds that contain those groups, such as DMSA and DMPS, can be used in chelation therapy for treating As intoxication. (Aposhian et al., 1995). These compounds are less toxic and more lipid soluble than an older compound, British Anti-Lewisite (BAL), developed during World War II as an antidote to the chemical warfare agent Lewisite, a trivalent As compound (Aposhian and Aposhian, 2001).

REFERENCES

383

Encouraging people to switch to wells that are not contaminated is the least expensive solution for reducing exposure to As in drinking water. It is estimated that 90% of people in Bangladesh live within 200 meters of a safe well (McLellan, 2002). However, there are problems preventing some people from switching wells, since in some areas the wells are privately owned. Transition to surface water is not an option, as it would result in increased risk in water-related infectious disease (Lokuge et al., 2004). Harvesting of rainwater could be an alternative option in areas of high rainfall, but even in the Bengal valley there is little rainfall during part of the year. In some areas, drilling deep wells (4150 m) to reach uncontaminated aquifers may present a solution. This has been tried in Inner Mongolia, but the deep wells were also contaminated with As (Smedley and Kinneburgh, 2005). Ideally, As should be removed from drinking water, but this must be accomplished by methods that are affordable and sustainable by the local population. Among the methods now available are ion exchange, adsorption, and coagulation/filtration. It must be pointed out that since arsenate is negatively charged at typical drinking water pH, whereas arsenite is neutral, many methods that rely on a negative charge would remove only arsenate and not arsenite. One ion exchange method uses polystyrene-based resins containing positively charged sites to remove arsenate (Frazer, 2005). Adsorption usually involves metal oxides such as those of iron or aluminum. Granular ferric oxide can adsorb both arsenate and arsenite, but its usefulness is limited to pH 7 or below (Fraser, 2005). Iron-treated activated carbon or gel beads or iron oxide-coated sand has been tried as adsorption agents. In one study, the sand appeared to work best (Yuan et al., 2002). A new approach is to coat nanoparticles of iron oxide onto an inert silicate, which reduces the cost and increases the binding capacity (Fraser, 2005). In developing countries, adsorption is carried out using locally-available materials such as brick chips. The coagulation/filtration process uses ferric chloride plus an oxidizing agent such as sodium hypochlorite to form insoluble ferric hydroxide, to which As compounds bind. Regardless of the method used, disposal of the removed As remains problematic. Research into methods for As removal is very active, and up-to-date information and debate about methods for removal of As from drinking water can be found in articles in Chappell et al. (2002) and in the following websites: http://bicn.com/acic/resources/arsenic-on-the-www/safewater.htm http://www.unu.edu/env/Arsenic/Proceedings.htm http://phys4.harvard.edu/%7Ewilson/arsenic/remediation/arsenic_project_remediation_ technology.html

REFERENCES Achanzar WE, Brambila EM, Diwan BA, Webber MM, Waalkes MP (2002) Inorganic arseniteinduced malignant transformation of human prostate epithelial cells. J. Natl. Cancer Inst. 94:1888–1891. Ahsan H, Chen Y, Kibriya MG, Islam MN, Slavkovich VN, Graziano JH, Santella RM (2003) Susceptibility to arsenic-induced hyperkeratosis and oxidative stress genes myeloperoxidase and catalase. Cancer Lett. 201:57–65. Alain G, Tousignant J, Rozenfarb E (1993) Chronic arsenic toxicity. Int. J. Dermatol. 32:899–901. ´ lvarez FF, Rodrıguez MT, Fernandez Espinosa AJ, Gutierrez Daban A (2004) Physical speciation of A arsenic, mercury, lead, cadmium and nickel in inhalable atmospheric particles. Anal. Chim. Acta 524 (1–2):33–40.

384

ARSENIC

Anke M, Groppel B, Grun M, Hennig A, Meissner D (1980) The influence of arsenic deficiency on growth reproductiveness, life expectancy and health of goats. Spurenelement Symposium, pp. 25–32. Aposhian HV (1989) Biochemical toxicology of arsenic. Rev. Biochem. Toxicol. 10:265–299. Aposhian HV, Aposhian MM (1989) New developments in arsenic toxicity. J. Am. Coll. Toxicol. 8:1273–1281. Aposhian HV, Aposhian MM (2001) Arsenic mobilization by DMPS. In: Chappell WR, Abernathy CO, Calderon RL, editors. Arsenic Exposure and Health Effects, Vol. IV.Oxford, UK: Elsevier Science Ltd., pp. 397–406. Aposhian HV, Maiorion RM, Gonzales-Ramirez D, Zuniga-Charles M, Xu Z (1995) Mobilization of heavy metals by newer, therapeutically useful chelating agents. Toxicology 97:23–38. Aposhian HV, Zakharyan RA, Avram MD, Sampayo-Reyes A, Wollenberg ML (2004) A review of the enzymology of arsenic metabolism and a new potential role of hydrogen peroxide in the detoxication of the trivalent arsenic species. Toxicol. Appl. Pharmacol. 198:327–335. Arnold LL, Eldan M, van Gemert M, Capen CC, Cohen SM (2003) Chronic studies evaluating the carcinogenicity of monomethylarsonic acid in rats and mice. Toxicology 190:197–219. ATSDR (2000) Toxicological Profile for Arsenic.Atlanta:Agency for Toxic Substance and Disease Registry, U.S. Department of Health and Human Services. Barchowsky A, Klei LR, Dudek EJ, Swartz HM, James PE (1999a) Stimulation of reactive oxygen, but not reactive nitrogen species, in vascular endothelial cells exposed to low levels of arsenate. Free Radic. Biol. Med. 27:1405–1412. Barchowsky A, Roussel RR, Klei LR, James PE, Ganju N, Smith KR, Dudek EJ (1999b) Low levels of arsenic trioxide stimulate proliferative signals in primary vascular cells without activating stress effector pathways. Toxicol. Appl. Pharmacol. 159:65–75. Basu A, Mahata J, Gupta S, Giri AK (2001) Genetic toxicology of a paradoxical human carcinogen, arsenic: a review. Mutat. Res. 488:171–194. Basu A, Mahata J, Roy AK, Sarlar JN, Poddar G, Nandi A, Sarkar PK, Dutta PK, Banerjee A, Das M, Ray K, Roychaudhury S, Natarajan AT, Nilsson R, Giri AK (2002) Enhanced frequency of micronuclei in individuals exposed to arsenic through drinking water in West Bengal, India. Mutat. Res. 516:29–40. Bau D-T, Wang T-S, Chung C-H, Wang ASS, Jan K-Y (2002) Oxidative DNA adducts and DNA– protein cross-links are the major DNA lesions induced by arsenite. Environ. Health Perspect. 110:753–756. Bednar AJ, Garbarino JR, Ranville JF, Wildeman TR (2002) Presence of organoarsenicals used in cotton production in agricultural water and soil of the southern United States. J. Agric. Food Chem. 50:7340–7344. Bencko V (1987)Arsenic. In:Fishbein L, Furst A, Mehlman MA, editors. Advances in Modern Environmental Toxicology, Vol. XI: Genotoxic and Carcinogenic Metals: Environmental and Occupational Occurrence and Exposure.Princeton, NJ:Princeton Scientific Publishing Co. Inc., pp. 1–30. Bertolero F, Pozzi G, Sabbioni E, Saffiotti U (1987) Cellular uptake and metabolic reduction of pentavalent to trivalent arsenic as determinants of cytotoxicity and morphological transformation. Carcinogenesis 8:803–808. Blount BC, Mack MM, Wehr CM, MacGregor JT, Hiatt RA, Want G, Wickramsinghe SN, Everson RB, Ames BN (1997) Folate deficiency causes uracil misincorporation into human DNA and chromosome breakage: implications for cancer and neuronal damage. Proc. Natl. Acad. Sci. USA 94:3290–3295.

REFERENCES

385

Buchet JP, Lauwerys R, Roels H (1981) Comparison of the urinary excretion of arsenic metabolites after a single dose of sodium arsenite, monomethylarsonate or dimethylarsinate in man. Int. Arch. Occup. Environ. Health 48:71–79. Buchet JP, Pauwels J, Lauwerys R (1994) Assessment of exposure to inorganic arsenic following ingestion of marine organisms by volunteers. Environ. Res. 66:44–51. Burns FJ, Uddin AN, Wu F, Nadas A, Rossman TG (2004) Arsenic-induced enhancement of UVR carcinogenesis in mouse skin: a dose-response. Environ. Health Perspect. 112:599–603. Carbrey JM, Gorelick-Feldman DA, Kozono D, Praetorius J, Nielsen S, Agre P (2003) Aquaglyceroporin AQP9: solute permeation and metabolic control of expression in liver. Proc. Natl. Acad. Sci. USA 100:2945–2950. Cantor KP (1997) Drinking water and cancer. Cancer Causes Control 8:292–308. Carter D, Bellamy WTY (1988) Toxicology of the group III–V intermetallic semiconductor, gallium arsenide. In:Clarkson TW, Friberg L, Nordberg GF, Sager PR,editors. Biological Monitoring of Toxic Metals. New York:Plenum Press, pp. 455–468. Carter DE, Aposhian HV, Gandolfi AJ (2003) The metabolism of inorganic arsenic oxides, gallium arsenide, arsine: a toxicochemical review. Toxicol. Appl. Pharmacol. 193:309–334. Centeno JA, Mullick FG, Martinez L, Page NP, Gibb H, Longfellow D, Thompson C, Ladich ER (2002) Pathology related to chronic arsenic exposure. Environ. Health Perspect. 110:883–886. Cerda S, Weitzman SA (1997) Influence of oxygen radical injury on DNA methylation. Mutat. Res. 386:141–152. Chappell WR, Abernathy CO, Calderon RL, Thomas DJ, editors. (2002) Arsenic Exposure and Health Effects V. Amsterdam:Elsevier B.V. Chen C-J, Hsueh Y-M, Lai M-S, Shyu M-P, Chen S-Y, Wu M-M, Kuo T-L, Tai T-Y (1995) Increased prevalence of hypertension and long-term arsenic exposure. Hypertension 25:53–60. Chen YC, Guo YL, Su H, Hsueh YM, Smith TJ, Ryan LM, Lee MS, Chao SC, Lee JY, Christiani DC (2003a) Arsenic methylation and skin cancer risk in southwestern Taiwan. J. Occup. Environ. Med. 45:241–248. Chen YC, Su HJ, Guo YL, Hsueh YM, Smith TJ, Ryan LM, Lee MS, Christiani DC (2003b) Arsenic methylation and bladder cancer risk in Taiwan. Cancer Causes Control 14:303–310. Chen CJ, Hsu LI, Wang CH, Shih WL, Hsu YH, Tseng MP, Lin YC, Chou WL, Chen CY, Lee CY, Wang LH, Cheng YC, Chen CL, Chen SY, Wang YH, Hsueh YM, Chiou HY, Wu MM (2005) Biomarkers of exposure, effect, and susceptibility of arsenic-induced health hazards in Taiwan. Toxicol. Appl. Pharmacol. 206:198–206. Chen Y, Graziano JH, Parvez F, Hussain I, Momotaj H, van Geen A, Howe GR, Ahsan H (2006) Modification of risk of arsenic-induced skin lesions by sunlight exposure, smoking, and occupational exposure in Bangladesh. Epidemiology 17:454–467. Chie CW, Chiang MC, Ho IC, Lee TC (2004) Association of chromosomal alterations with arseniteinduced tumorigenicity of human HaCaT keratinocytes in nude mice. Environ. Health Perspect. 112:1704–1710. Chilvers DC, Peterson PJ (1987)Global cycling of arsenic. In:Hutchinson TC, Meema KM, editors. Lead, Mercury, Cadmium and Arsenic in the Environment, Scientific Committee on Problems of the Environment (SCOPE) 31. New York:John Wiley & Sons. Chiou HY, Hsueh YM, Liaw KF, Horng SF, Chiang MH, Pu YS, Lin JSN, Huang CH, Chen CJ (1995) Incidence of internal cancers and ingested inorganic arsenic: a seven-year follow-up study in Taiwan. Cancer Res. 55:1296–1300. Choi SW, Mason JB (2000) Folate and carcinogenesis: an integrated scheme. J. Nutr. 130:129–132. Cohen SM, Yamamoto S, Cano M, Arnold LL (2001) Urothelial cytotoxicity and regeneration induced by dimethylarsinic acid in rats. Toxicol. Sci. 59:68–74.

386

ARSENIC

Cohen SM, Arnold LL, Uzvolgyi E, Cano M, St. John M, Yamamoto S, Lu X, Le XC (2002) Possible role of dimethylarsinous acid in dimethylarsinic acid-induced urothelial toxicity and generation in the rat. Chem. Res. Toxicol. 15:1150–1157. Concha G, Vogler G, Nermell B, Vahter M (2002) Intra-individual variation in the metabolism of inorganic arsenic. Int. Arch. Occup. Environ. Health 75:576–580. Cox TM, Jack N, Lofthouse S, Watling J, Haines J, Warren MJ (2005) King George III and porphyria: an elemental hypothesis and investigation. Lancet 366:332–335. Cullen WR, Bentley R (2005) The toxicity of trimethylarsine: an urban myth. J. Environ. Monit. 7: 11–15. Cullen WR, Reimer KJ (1989) Arsenic speciation in the environment. Chem. Rev. 89:713–764. Dopp E, Hartmann LM, Florea AM, von Recklinghausen U, Pieper R, Shokouhi B, Rettenmeier AW, Hirner AV, Obe G (2004) Uptake of inorganic and organic derivatives of arsenic associated with induced cytotoxic and genotoxic effects in Chinese hamster ovary (CHO) cells. Toxicol. Appl. Pharmacol. 201:156–165. Drobna Z, Waters SB, Walton FS, LeCluyse EL, Thomas DJ, Styblo M (2004) Interindividual variation in the metabolism of arsenic in cultured primary human hepatocytes. Toxicol. Appl. Pharmacol. 201:166–177. Ellickson KM, Meeker RJ, Gallo A, Buckley BT, Lioy PJ (2001) Oral bioavailability of lead and arsenic from a NIST standard reference soil material. Arch. Environ. Contam. Toxicol. 40:128–135. Everall JD, Dowd PM (1978) Influence of environmental factors excluding ultra violet radiation on the incidence of skin cancer. Bull. Cancer 65:241–247. Fergusson JE (1984) The elemental composition of street dust from large and small urban areas related to city type, source and particle size. Sci. Total Environ. 34:101–116. Ferreccio C, Gonzalez C, Milosavjlevic V, Marshall G, Sancha AM, Smith AH (2000) Lung cancer and arsenic concentrations in drinking water in Chile. Epidemiology 11:673–679. Flora SJS (1996) Alterations in some hepatic biochemical variables following repeated gallium arsenide administration in rats. Int. Hepatol. Commun. 5:97–103. Fowler BA (1977) Toxicology of environmental arsenic. In: Goyer GA, Mehlman MA, editors. Toxicology of Trace Elements. New York: John Wiley & Sons, pp. 79–122. Fowler B, Ishinishi N, Tsuchiya K, Vahter M (1979) Arsenic. In: Friberg L, Nordberg GF, Vouk VB, editors. Handbook on The Toxicology of Metals. New York: Elsevier/North-Holland Biomedical Press, pp. 1721–1723. Francesconi KA, Tanggaard R, McKenzie CJ, Goessler W (2002) Arsenic metabolites in human urine after ingestion of an arsenosugar. Clin. Chem. 48:92–101. Frazer L (2005) Metal attraction: an ironclad solution to arsenic contamination? Environ. Health Perspect. 113:A399–A401. Gailer J, George GN, Pickering IJ, Prince RC, Ringwald SC, Pemberton JE, Glass RS, Younis HS, DeYoung DW, Aposhian HV (2000) A metabolic link between arsenite and selenite: the seleno-bis (S-glutathionyl) arsenium ion. J. Am. Chem. Soc. 122:4637–4639. Garb LG, Hine CH (1977) Arsenical neuropathy: residual effects following acute exposure. J. Occup. Med. 19:567–568. Garbarino JR, Bednar AJ, Rutherford DW, Beyer RS, Wershaw RL (2003) Environmental fate or roxarsone in poultry litter. I. Degradation of roxarsone during composting. Environ. Sci. Technol. 37:1509–1514. Garcia-Vargas GG, Cebrian ME (1996)Health effects of arsenic. In:Wang LW,editor. Toxicology of Metals.Boca Raton, FL:CRC Press, pp. 423–438. Gerl R, Vaux DL (2005) Apoptosis in the development and treatment of cancer. Carcinogenesis 26:263–270.

REFERENCES

387

Germolec DR, Yoshida T, Gaido K, Wilmer JL, Simeonova PP, Kayama F, Burleson F, Dong W, Lange RW, Luster MI (1996) Arsenic induces overexpression of growth factors in human keratinocytes. Toxicol. Appl. Pharmacol. 141:308–318. Germolec DR, Spalding J, Boorman GA, Wilmer JL, Yoshida T, Simeonova PP, Bruccoleri A, Kayama F, Gaido K, Tennant R, Burleson F, Dong W, Lang RW, Luster MI (1997) Arsenic can mediate skin neoplasia by chronic stimulation of keratinocyte-derived growth factors. Mutat. Res. 386:209– 218. Germolec D, Spalding J, Yu H-S, Chen GS, Simeonova PO, Humble MC, Bruccoleri A, Boorman GA, Foley JF, Yoshida T, Luster MI (1998) Arsenic enhancement of skin neoplasia by chronic stimulation of growth factors. Am. J. Pathol. 153:1775–1785. Ghosh P, Basu A, Mahata J, Basu S, Sengupta M, Das JK, Mukherjee A, Sarkar AK, Mondal L, Ray K, Giri AK (2006) Cytogenetic damage and genetic variants in the individuals susceptible to arsenicinduced cancer through drinking water. Int. J. Cancer 118:2470–2478. Guha Mazumder DN (2003) Chronic arsenic toxicity: clinical features, epidemiology, and treatment: experience in West Bengal. J. Environ. Sci. Part A 38:141–163. Guha Mazumder DN, Haque R, Ghosh N, De BK, Santra A, Chakroborty D, Smith AH (1998) Arsenic levels in drinking water and the prevalence of skin lesions in West Bengal, India. Int. J. Epidemiol. 27:871–877. Guo HR, Yu HS, Hu H, Monson RR (2001) Arsenic in drinking water and skin cancers: cell-type specificity. Cancer Causes Control 12:909–916. Haimeur A, Conseil G, Deeley RG, Cole SP (2004) The MRP-related and BCRP/ABCG2 multidrug resistance proteins: biology, substrate specificity and regulation. Curr. Drug Metab. 5:21–53. Haque R, Guha Mazumder DN, Samanta S, Ghosh N, Kalman D, Smith MM, Mitra S, Santra A, Lahiri S, Das S, De BK, Smith AH (2003) Arsenic in drinking water and skin lesions: dose–response data from West Bengal, India. Epidemiology 14:174–182. Harkins DK, Susten AS (2003) Hair analysis: exploring the state of the science. Environ. Health Perspect. 111:576–578. Hartwig A, Asmuss M, Ehleben I, Herzer U, Kostelac D, Pelzer A, Schgwerdtle T, Burkle A (2002) Interference by toxic metal ions with DNA repair processes and cell cycle control: molecular mechanisms. Environ. Health Perspect. 110:797–799. Hartwig A, Pelzer A, Asmuss M, Burkle A (2003) Very low concentrations of arsenite suppress poly (ADP-ribosyl)ation in mammalian cells. Int. J. Cancer 104:1–6. Hei TK, Liu SX, Waldren C (1998) Mutagenicity of arsenic in mammalian cells: role of reactive oxygen species. Proc. Natl. Acad. Sci. USA 95:8103–8107. Hinwood AL, Sim MR, Jolley D, de Klerk N, Bastone EB, Gerostamoulos J, Drummer OH (2003) Hair and toenail arsenic concentrations of residents living in areas with high environmental arsenic concentrations. Environ. Health Perspect. 111:187–193. Hoffman RD, Lane MD (1992) Iodophenylarsine oxide and arsenical affinity chromatography: new probes for dithiol proteins. Application to tubulins and to components of the insulin receptor– glucose transporter signal transduction pathway. J. Biol. Chem. 267:14005–14011. Hopenhayn-Rich C, Biggs ML, Smith AH (1998) Lung and kidney cancer mortality associated with arsenic in drinking water in Cordoba, Argentina. Int. Epidemiol. Assoc. 27:561–569. Hsueh Y-M, Cheng G-S, Wu M-M, Yu H-S, Kuo T-L, Chen C-J (1995) Multiple risk factors associated with arsenic-induced skin cancer: effects of chronic liver disease and malnutritional status. Br. J. Cancer 71:109–114. Hsueh YM, Ko YF, Huang YK, Chen HW, Chiou HY, Huang YL, Yang MH, Chen CJ (2003) Determinants of inorganic arsenic methylation capability among residents of the Lanyang Basin, Taiwan; arsenic and selenium exposure and alcohol consumption. Toxicol. Lett. 137 (1–2):49–63.

388

ARSENIC

Hu Y, Su L, Snow ET (1998) Arsenic toxicity is enzyme specific and its affects on ligation are not caused by the direct inhibition of DNA repair enzymes. Mutat. Res. 408:203–218. Hu J, Fang J, Dong Y, Chen SJ, Chen Z (2005) Arsenic in cancer therapy. Anti-Cancer Drugs 16:119–127. Huang C, Ma W-Y, Li J, Goranson A, Dong Z (1999) Requirement of Erk, but not JNK, for arseniteinduced cell transformation. J. Biol. Chem. 274:14595–14601. IARC Monograph on the Evaluation of Carcinogenic Risk to Humans.(2004) Some Drinking Water Disinfectants and Contaminants, Including Arsenic. Vol. 84, Lyon, France. Islam FS, Gault AG, Boothman C, Polya DA, Charnock JM, Chatterjee D, Lloyd JR (2004) Role of metal-reducing bacteria in As release from Bengal delta sediments. Nature 430:68–71. Jha AN, Noditi M, Nilsson R, Natarajan AT (1992) Genotoxic effects of sodium arsenite on human cells. Mutat. Res. 284:215–221. Kadono T, Inaoka T, Murayama N, Ushijima K, Nagano M, Nakamura S, Watanabe C, Tamaki K, Ohtsuka R (2002) Skin manifestations of arsenicosis in two villages in Bangladesh. Int. J. Dermatol. 41:841–846. Kala SV, Neely MW, Kala G, Prater CI, Atwood DW, Rice JS, Lieberman MW (2000) The MRP2/ cMOAT transporter and arsenic-glutathione complex formation are required for biliary excretion of arsenic. J. Biol. Chem. 275:33404–33408. Kaltreider RC, Davis AM, Lariviere JP, Hamilton JW (2001) Arsenic alters the function of the glucocorticoid receptor as a transcription factor. Environ. Health Perspect. 109:245–251. Kaneto H, Matsuoka TA, Nakatani Y, Kawamori D, Miyatsuka T, Matsuhisa M, Yamasaki Y (2005) Oxidative stress, ER stress, and the JNK pathway in type 2 diabetes. J. Mol. Med. 83:429–439. Kenyon EM, Hughes MF (2001) A concise review of the toxicity and carcinogenicity of dimethylarsinic acid. Toxicology 160:227–236. Klaunig JE, Kamendulis LM (2004) The role of oxidative stress in carcinogenesis. Annu. Rev. Pharmacol. Toxicol. 44:239–267. Kligerman AD, Doerr CL, Tennant AH, Harrington-Brock K, Allen JW, Winkfield E, Poorman-Allen P, Kundu B, Funasaka K, Roop BC, Mass MJ, DeMarini DM (2003) Methylated trivalent arsenicals as candidate ultimate genotoxic forms of arsenic: induction of chromosomal mutations but not gene mutations. Environ. Mol. Mutagen. 42:192–205. Komissarova ET, Saha SK, Rossman TG (2005) Dead or dying: the importance of time in cytotoxicity assays using arsenite as an example. Toxicol. Appl. Pharmacol. 202:99–107. Kwon E, Zhang H, Wang Z, Jhangri GS, Lu X, Fok N, Gabos S, Li XF, Le XC (2004) Arsenic on the hands of children after playing in playgrounds. Environ. Health Perspect. 112:1375–1380. Landrigan PJ (1992) Arsenic. In: Rom WN, editor. Environmental and Occupational Medicine, 2nd edn. Boston, MA: Little Brown and Co., pp. 773–779. Lasky T, Sun W, Kadry A, Hoffman MK (2004) Mean total arsenic concentrations in chicken 1989– 2000 and estimated exposures for consumers of chicken. Environ. Health Perspect. 112:18–21. Lee TC, Ho IC (1994) Expression of heme oxygenase in arsenic-resistant human lung adenocarcinoma cells. Cancer Res. 54:1660–1664. Lee T-C, Ho I-C (1995) Modulation of cellular antioxidant defense activities by sodium arsenite in human fibroblasts. Arch. Toxicol. 69:498–504. Lee T-C, Huang RY, Jan KY (1985) Sodium arsenite enhances the cytotoxicity, clastogenicity and 6thioguanine-resistant mutagenicity of ultraviolet light in Chinese hamster ovary cells. Mutat. Res. 148:83–89. Lee TC, Wang-Wu S, Huang RY, Lee KC, Jan KY (1986) Differential effects of the pre- and posttreatment of sodium arsenite on the genotoxicity of methylmethanesulfonate in Chinese hamster ovary cells. Cancer Res. 46:1854–1857.

REFERENCES

389

Leonard A (1991) Arsenic. In:Merian E,editor. Metals and Their Compounds in the Environment: Occurrence, Analysis and Biological Relevance. New York/Verlagsgesellschaft:VCH Publishers, Inc., pp.751–774. Li J-H, Rossman TG (1989a) Mechanism of comutagenesis of sodium arsenite with N-methyl-Nnitrosourea. Biol. Trace Elem. Res. 21:373–381. Li J-H, Rossman TG (1989b) Inhibition of DNA ligase activity by arsenite: a possible mechanism of its comutagenesis. Mol. Toxicol. 2:1–9. Li J-H, Rossman TG (1991) Comutagenesis of sodium arsenite with ultraviolet radiation in Chinese hamster V79 cells. Biol. Met. 4:197–200. Li P, Uddin AN, Liu Z, Mukhopadhyay RM, Komissarova EV, Rosen BP, Rossman TG (2004) Variability in sensitivity to arsenite does not correlate with arsenic accumulation rate in normal human lymphoblasts. Mol. Cell. Biochem. 255:79–85. Liebscher K, Smith H (1968) Essential and nonessential trace elements: a method of determining whether an element is essential or nonessential in human tissue. Arch. Environ. Health 17:881– 890. Lin S, Cullen WR, Thomas DJ (1999) Methylarsenicals and arsenothiols are potent inhibitors of mouse liver thioredoxin reductase. Chem. Res. Toxicol. 12:924–930. Liu Z, Boles E, Rosen BP (2004a) Arsenic trioxide uptake by hexose permeases in Saccharomyces cerevisiae. J. Biol. Chem. 279:17312–17318. Liu Z, Carbrey JM, Agre P, Rosen BP (2004b) Arsenic trioxide uptake by human and rat aquaglyceroporins. Biochem. Biophys. Res. Commun. 316:1178–1185. Loffredo CA, Aposhian HV, Cebrian ME, Yamauchi H, Silbergeld EK (2003) Variability in human metabolism of arsenic. Environ. Res. 92:85–91. Lokuge KM, Smith W, Caldwell B, Dear K, Milton AH (2004) The effect of arsenic mitigation interventions on disease burden in Bangladesh. Environ. Health Perspect. 112:1172–1177. Lopez S, Miyashita Y, Simons SS (1990) Structurally based, selective interaction of arsenite with steroid receptors. J. Biol. Chem. 265:16039–16042. Lynn S, Gurr J-R, Lai H-T, Yan K-Y (2000) NADH oxidase activation is involved in arsenite-induced oxidative DNA damage in human vascular smooth muscle cells. Circ. Res. 86:514–519. Ma M, Le XC (1998) Effect arsenosugar ingestion on urinary arsenic speciation. Clin. Chem. 44:539– 550. Mandal BK, Suzuki KT (2002) Arsenic round the world: a review. Talanta 58:201–235. Marnell LL, Garcia-Vargas GG, Chowdhury UK, Zakharyan RA, Walsh B, Avram MD, Kopplin MJ, Cebrian ME, Silbergeld EK, Aposhian HV (2003) Polymorphisms in the human monomethylarsonic acid (MMA V) reductase/hGSTO1 gene and changes in urinary arsenic profiles. Chem. Res. Toxicol. 16:1507–1513. Mason JB, Levesque T (1996) Folate: effects on carcinogenesis and the potential for cancer prevention. Oncology 10:1727–1743. Mass MJ, Wang L (1997) Arsenic alters cytosine methylation patterns of the promoter of the tumor suppressor gene p53 in human lung cells: a model for a mechanism of carcinogenesis. Mutat. Res. 386:263–277. Matsui M, Nishigori C, Toyokuni S, Takada J, Akaboshi M, Ishikawa M, Imamura S, Miyachi Y (1999) The role of oxidative DNA damage in human arsenic carcinogenesis: detection of 8hydroxy-200 -deoxyguanosine in arsenic-related Bowen’s disease. J. Invest. Dermatol. 113:26–31. McDorman EW, Collins BW, Allen JW (2002) Dietary folate deficiency enhances induction of micronuclei by arsenic in mice. Environ. Mol. Mutagen. 40:71–77. McLellan F (2002) Arsenic contamination affects millions in Bangladesh. Lancet 359 (9312):1127 Meharg A (2003) The arsenic green. Nature 423:688

390

ARSENIC

Meza MM, Yu L, Rodriguez YY, Guild M, Thompson D, Gandolfi AJ, Klimecki WT (2005) Developmentally restricted genetic determinants of human arsenic metabolism: association between urinary methylated arsenic and CYT19 polymorphisms in children. Environ. Health Perspect. 113:775–781. Miller WH Jr, Schipper HM, Lee JS, Singer J, Waxman S (2002) Mechanisms of action of arsenic trioxide. Cancer Res. 62:3893–3903. Milner JE (1969) The effects of ingested arsenic exposure on methyl cholanthrene induced skin tumors in mice. Arch. Environ. Health 18:7–11. Mitra SR, Guha Mazumder DN, Basu A, Block G, Haque R, Samanta S, Ghosh N, Smith MM, von Ehrenstein OS, Smith AH (2004) Nutritional factors and susceptibility to arsenic-caused skin lesions in West Bengal, India. Environ. Health Perspect. 112:1104–1109. Moody JP, Williams RT (1964) The metabolism of 4-hydroxy-3-nitrophenylarsonic acid in hens. Food Cosmet. Toxicol. 15:707–715. Mudipalli A, Owen RD, Preston RJ (2005) The effect of arsenicals on ultraviolet-radiation-induced growth arrest and related signaling events in human keratinocytes. Int. J. Oncol. 27:769–778. Mukherjee PK, Pal T, Sengupta S, Shome S (2001) Arsenic rich phases in aquifer sediments from southern West Bengal. J. Geol. Soc. India 58:173–176. Mure K, Uddin AN, Lopez LC, Styblo M, Rossman TG (2003) Arsenite induces delayed mutagenesis and transformation in human osteosarcoma cells at extremely low concentrations. Environ. Mol. Mutagen. 41:322–331. Nachman KE, Graham JP, Price LB, Silbergeld EK (2005) Arsenic: a roadblock to potential animal waste management solutions. Environ. Health Perspect. 113:1123–1124. National Research Council (NRC) (2000) Arsenic in Drinking Water.Washington, DC:National Academy Press. Nielsen FH (1996) How should dietary guidance be given for mineral elements with beneficial actions or suspected of being essential? J. Nutr. 126(9 Suppl.):2377S–2385S. Nordenson I, Beckman L (1991) Is the genotoxic effect of arsenic mediated by oxygen free radicals? Hum. Hered. 41:71–73. Petrick JS, Ayala-Fierro F, Cullen WR, Carter DE, Aposhian HV (2000) Monomethylarsonous acid (MMAIII) is more toxic than arsenite in Chang human hepatocytes. Toxicol. Appl. Pharmacol. 163:203–207. Pi J, Yamauchi H, Kumagai Y, Sun G, Yoshida T, Aikawa H, Hopenhayn-Rich C, Shimojo N (2002) Evidence for induction of oxidative stress caused by chronic exposure of Chinese residents to arsenic contained in drinking water. Environ. Health Perspect. 110:331–336. Pi J, He Y, Bortner C, Huang J, Liu J, Zhou T, Qu W, North SL, Kasprzak KS, Diwan BA, Chignell CF, Waalkes MP (2005) Low level, long-term inorganic arsenite exposure causes generalized resistance to apoptosis in cultured human keratinocytes: potential role in skin co-carcinogenesis. Int. J. Cancer 116:20–26. Pinto SS, Nelson KW (1976) Arsenic toxicology and industrial exposure. Annu. Rev. Pharmacol. Toxicol. 16:95–100. Pogribny IP, Miller BJ, James SJ (1997) Alterations in hepatic p53 gene methylation patterns during tumor progression in folate methyl deficiency in the rat. Cancer Lett. 115:31–38. Rahman MM, Chowdhury UK, Mukherjee SC, Mondal BK, Paul K, Lodh D, Biswas BK, Chanda CR, Basu GK, Saha KC, Roy S, Das R, Palit SK, Quamruzzaman Q, Chakraborti D (2001) Chronic arsenic toxicity in Bangladesh and West Bengal, India—a review and commentary. J. Toxicol. Clin. Toxicol. 39:683–700. Ramos O, Carrizales L, Yanez L, Mejia J, Batres L, Ortiz D, Diaz-Barriga F (1995) Arsenic increased lipid peroxidation in rat tissues by a mechanism independent of glutathione levels. Environ. Health Perspect. 103:85–88.

REFERENCES

391

Rhine ED, Garcia-Dominguez E, Phelps CD, Young LY (2005) Environmental microbes can speciate and cycle arsenic. Environ. Sci. Technol. 39:9569–9573. Rossman TG (2003) Mechanism of arsenic carcinogenesis: an integrated approach. Mutat. Res. 533:37–66. Rossman TG, Uddin AN (2004) Selenium prevents spontaneous and arsenite-induced mutagenesis. Int. Congr. Ser. 1275:173–179. Rossman TG, Uddin AN, Burns FJ, Bosland M (2001) Arsenic is a cocarcinogen with solar ultraviolet radiation for mouse skin: an animal model for arsenic carcinogenesis. Toxicol. Appl. Pharmacol. 176:64–71. Rossman TG, Uddin AN, Burns FJ (2004) Evidence that arsenite acts as a cocarcinogen in skin cancer. Toxicol. Appl. Pharmacol. 198:394–404. Schmidt A, Anke M, Groppel B, Kronemann H (1984) Effects of As-deficiency on skeletal muscle, myocardium and liver. A histochemical and ultrastructural study. Exp. Pathol. 25:195–197. Schoen A, Beck B, Sharma R, Dube E (2004) Arsenic toxicity at low doses: epidemiological and mode of action considerations. Toxicol. Appl. Pharmacol. 198:253–267. Schoof RA, Yost LJ, Eickhoff J, Crecelius EA, Cragin DW, Meacher DM, Menzel DB (1999) A market basked survey of inorganic arsenic in food. Food Chem. Toxicol. 37:839–846. Sciandrello G, Barbaro R, Caradonna F, Barbata G (2002) Early induction of genetic instability and apoptosis by arsenic in cultured Chinese hamster cells. Mutagenesis 17:99–103. Simeonova PP, Luster MI (2004) Arsenic and atherosclerosis. Toxicol. Appl. Pharmacol. 198:444–449. Simeonova PP, Wang S, Toriuma W, Kommineni V, Matheson J, Unimye N, Kayama F, Harki D, Ding M, Vallyathan V, Luster MI (2000) Arsenic mediates cell proliferation and gene expression in the bladder epithelium: association with activating protein-1 transactivation. Cancer Res. 60:3445–3453. Smedley P, Kinneburgh DG (2005)Arsenic in groundwater and the environment. In:Selinus O,editor. Essentials of Medical Geology.Burlingham, MA:Elsevier Academic Press, pp.263–299. Smith AH, Hira Smith MM (2004) Arsenic drinking water regulations in developing countries with extensive exposure. Toxicology 198:39–44. Spallholz JE, Boylan LM, Rhaman MM (2004) Environmental hypothesis: is poor dietary selenium intake an underlying factor for arsenicosis and cancer in Bangladesh and West Bengal, India?Sci. Total Environ. 323:21–32. Squibb KS, Fowler BA (1983)The toxicity of arsenic and its compounds. In: Fowler BA, editor. Biological and Environmental Effects of Arsenic. Amsterdam: Elsevier, p. 233. Steinmaus C, Carrigan K, Kalman D, Atallah R, Yuan Y, Smith AH (2005) Dietary intake and arsenic methylation in a U.S. population. Environ. Health Perspect. 113:1153–1159. Stocker R, Keaney JF Jr (2004) Role of oxidative modifications in atherosclerosis. Physiol. Rev. 84:1381–1478. Styblo M, Serves SV, Cullen WR, Thomas DJ (1997) Comparative inhibition of yeast glutathione reductase by arsenicals and arsenothiols. Chem. Res. Toxicol. 10:27–33. Styblo M, Del Razo LM, Vega L, Germolec DR, LeCluyse EL, Hamilton GA, Reed W, Wang C, Cullen WR, Thomas DJ (2000) Comparative toxicity of trivalent and pentavalent inorganic and methylated arsenicals in rat and human cells. Arch. Toxicol. 74:289–299. Styblo M, Drobna Z, Jaspers I, Lin S, Thomas DJ (2002) The role of biomethylation in toxicity and carcinogenicity of arsenic: a research update. Environ. Health Perspect. 110:767–771. Sun G (2004) As contamination and arsenicosis in China. Toxicol. Appl. Pharmacol. 198:268– 271. Tapio S, Grosche B (2006) Arsenic in the aetiology of cancer. Mutat. Res., 612:215–246.

392

ARSENIC

Takahashi M, Barrett JC, Tsutsui T (2002) Transformation by inorganic arsenic compounds of normal Syrian hamster embryo cells into a neoplastic state in which they become anchorage-independent and cause tumors in newborn hamsters. Int. J. Cancer 99:629–634. Tamaki S, Frankenberger WT Jr (1992) Environmental biochemistry of arsenic. Rev. Environ. Contam. Toxicol. 124:79–110. Trouba KJ, Wauson EM, Vorce RL (2000) Sodium arsenite-induced dysregulation of proteins involved in proliferative signaling. Toxicol. Appl. Pharmacol. 164:161–170. Tseng WP, Chu HM, How SW, Fong JM, Lin CS, Yeh S (1968) Prevalence of skin cancer in an endemic area of chronic arsenicism in Taiwan. J. Natl. Cancer Inst. 40:453–463. Tseng C-H, Tai T-Y, Chong C-K, Tseng C-P, Lai M-S, Lin BJ, Chiou H-Y, Hsueh Y-M, Hsu K-H, Chen C-J (2000) Long-term arsenic exposure and incidence of non-insulin-dependent diabetes mellitus: a cohort study in arseniasis-hyperendemic villages in Taiwan. Environ. Health Perspect. 108:847– 851. Tseng C-H, Huang Y-K, Huang Y-L, Chung C-J, Yang M-H, Chen C-J, Hsueh Y-M (2005) Arsenic exposure, urinary arsenic speciation, and peripheral vascular disease in blackfoot diseasehyperendemic villages in Taiwan. Toxicol. Appl. Pharmacol. 206:299–308. Tsuda T, Babazono A, Yamamoto E, Kurumatani N, Mino Y, Ogara T, Kishi Y, Aoyama H (1995) Ingested arsenic and internal cancer: a historical cohort study followed for 33 years. Am. J. Epidemiol. 141:198–209. Uddin AN, Burns FJ, Rossman TG (2005) Vitamin E and organoselenium prevent the cocarcinogenic activity of arsenite with solar UV in mouse skin. Carcinogenesis 26:2179–2186. Uthus EO (1990) Effects of arsenic deprivation in hamsters. Magnes. Trace Elem. 9:227–232. Uthus EO (1992) Evidence for arsenic essentiality. Environ. Geochem. Health 14:55–58. Uthus EO (2003) Arsenic essentiality: a role affecting methionine metabolism. J. Trace Elem. Exp. Med. 16:345–355. Uthus EO, Davis C (2005) Dietary arsenic affects dimethylhydrazine-induced aberrant crypt formation and hepatic global DNA methylation and DNA methyltranserase activity in rats. Biol. Trace Elem. Res. 103:133–145. Uthus EO, Nielsen FH (1993) Determination of the possible requirement and reference dose levels for arsenic in humans. Scand. J. Work Environ. Health 19:137–138. Vahter M (2002) Mechanisms of arsenic biotransformation. Toxicology 181:211–217. Vahter M, Concha G (2001) Role of metabolism in arsenic toxicity. Pharmacol. Toxicol. 89:1–5. Vahter M, Marafante E (1987) Effects of low dietary intake of methionine, choline or proteins on the biotransformation of arsenite in the rabbit. Toxicol. Lett. 37:41–46. Vahter M, Marafante E, Dencker L (1983) Metabolism of arsenobetaine in mice, rats and rabbits. Sci. Total Environ. 30:197–211. Valenzuela OL, Borja-Aburto VH, Garcia-Vargas GG, Cruz-Gonzalez MB, Garcia-Montalvo EA, Calderon-Aranda ES, Del Razo LM (2005) Urinary trivalent methylated arsenic species in a population chronically exposed to inorganic arsenic. Environ. Health Perspect. 113:250– 254. Vega L, Gonseblatt ME, Ostrosky-Wegman P (1995) Aneugenic effect of sodium arsenite on human lymphocytes in vitro: an individual susceptibility effect detected. Mutat. Res. 334:365–373. Vogt B, Rossman TG (2001) Effects of arsenite on p53, p21 and cyclin D expression in normal human fibroblasts—a possible mechanism for arsenite’s comutagenicity. Mutat. Res. 478:159–168. Waalkes MP, Ward JM, Liu J, Diwan BA (2003) Transplacental carcinogenicity of inorganic arsenic in the drinking water: induction of hepatic, ovarian, pulmonary, and adrenal tumors in mice. Toxicol. Appl. Pharmacol. 186:7–17.

REFERENCES

393

Wang C-H, Jeng J-S, Yip P-K, Chen C-L, Hsu L-I, Hsueh Y-M, Chiou H-Y, Wu M-M, Chen C-J (2002) Biological gradient between long-term arsenic exposure and carotid atherosclerosis. Circulation 105:1804–1809. Warren GP, Alloway BJ, Lepp NW, Singh B, Bochereau FJM, Penny C (2003) Field trials to assess the uptake of arsenic by vegetables from contaminated soils and soil remediation with iron oxides. Sci. Total Environ. 311:19–33. Watanabe C, Inaoka T, Kadono T, Nagano M, Nakamura S, Ushijima K, Murayama N, Miyazaki K, Ohtsuka R (2001) Males in rural Bangladeshi communities are more susceptible to chronic arsenic poisoning than females: analyses based on urinary As. Environ. Health Perspect. 109:1265–1270. Wedepohl KH (1991) The composition of the upper Earth’s crust and the natural cycles of selected metals. Metals in natural raw materials. Natural resources. In:Merian E,editor. Metals and Their Compounds in the Environment: Occurrence, Analysis and Biological Relevance.New York/ Verlagsgesellschaft:VCH Publishers, Inc., pp.3–17. Wei M, Wanibuchi H, Yamamoto S, Li W, Fukushima S (1999) Urinary bladder carcinogenicity of dimethylarsinic acid in male F344 rats. Carcinogenesis 20:1873–1876. Wesbey G, Kunis A (1981) Arsenical neuropathy. IMJ Ill. Med. J. 160(5):396–398. Wiencke JK, Yager JW (1992) Specificity of arsenite in potentiating cytogenetic damage induced by the DNA crosslinking agent diepoxybutane. Environ. Mol. Mutagen. 19:195–200. Williams PN, Price AH, Hossain SA, Feldmann J, Meharg AA (2005) Variation in arsenic speciation and concentration in paddy rice related to dietary exposure. Environ. Health Perspect. 39:5531– 5540. Winski SL, Carter DE (1995) Interactions of rat red blood cell sulfhydryls with arsenate and arsenite. J. Toxicol. Environ. Health 46:379–397. World Health Organization (2001) Environmental Health Criteria 224. Arsenic and Arsenic Compounds. Geneva. Wu MM, Chiou HY, Wang TW, Hsueh YM, Wang IH, Chen CJ, Lee TC (2001) Association of blood arsenic levels with increased reactive oxidants and decreased antioxidant capacity in a human population of northeastern Taiwan. Environ. Health Perspect. 109:1011–1017. Wu F, Burns FJ, Zhang R, Uddin AN, Rossman TG (2005) Arsenite-induced alterations of DNA photodamage repair and apoptosis after solar-simulation UVR in mouse keratinocytes in vitro. Environ. Health Perspect. 113:983–986. Xuan XZ, Lubin JH, Li JY, Yang LF, Luo AS, Lan Y, Wang JZ, Blot WJ (1993) A cohort study in southern China of tin miners exposed to radon and radon decay products. Health Phys. 64:120–131. Yang JL, Chen MF, Wu CW, Lee TC (1992) Posttreatment with sodium arsenite alters the mutation spectrum induced by ultraviolet light irradiation in Chinese hamster ovary cells. Environ. Mol. Mutagen. 20:156–164. Yang MH, Chen KK, Yen CC, Wang WS, Chang YH, Huang WJ, Fan FS, Chiou TJ, Liu JH, Chen PM (2002) Unusually high incidence of upper urinary tract urothelial carcinoma in Taiwan. Urology 59:681–687. Yeh S (1973) Skin cancer in chronic arsenicism. Hum. Pathol. 4:469–485. Yeh S, How SW, Lin CS (1968) Arsenical cancer of skin. Histologic study with special reference to Bowen’s disease. Cancer 21:312–339. Yoshida T, Yamauchi H, Sun GF (2004) Chronic health effects in people exposed to arsenic via the drinking water: dose–response relationships in review. Toxicol. Appl. Pharmacol. 198:243–252. Yu RC, Hsu KH, Chen CJ, Froines JR (2000) Arsenic methylation capacity and skin cancer. Cancer Epidemiol. Biomarkers Prev. 9:1259–1262. Yuan T, Hu JY, Ong SL, Luo QF, Ng WJ (2002) Arsenic removal from household drinking water by adsorption. J. Environ. Sci. Health A 37(9):1721–1736.

394

ARSENIC

Zagury GJ, Pouschat P (2005) Arsenic on children’s hands after playing in playgrounds. Environ. Health Perspect. 113:A508–A509. Zeng H, Uthus EO, Combs GF Jr (2005) Mechanistic aspects of the interaction between selenium and arsenic. J. Inorg. Biochem. 99:1269–1274. Zhao CQ, Young MR, Diwan BA, Coogan TP, Waalkes MP (1997) Association of arsenic-induced malignant transformation with DNA hypomethylation and aberrant gene expression. Proc. Natl. Acad. Sci. USA 94:10907–10912. Zhong CX, Mass MJ (2001) Both hypomethylation and hypermethylation of DNA associated with arsenite exposure in cultures of human cells identified by methylation-sensitive arbitrarily-primed PCR. Toxicol. Lett. 122:223–234.

12 ASBESTOS AND OTHER MINERAL AND VITREOUS FIBERS Morton Lippmann

12.1 IMPORTANT SPECIAL PROPERTIES OF FIBERS 12.1.1

Asbestos

Asbestos is a geological term used to describe silicate minerals that display special qualities and properties (Langer and Nolan, 1986; Langer et al., 1990). The mineralogy and chemistry of asbestos fibers are summarized in Table 12.1. It is the fibrous habit or “asbestiform” nature of these minerals, as well as their occurrence as millimeter-long fibers that separate them from other silicate minerals and gave them commercial importance. Asbestos fibers are good thermal and acoustic insulators, those low in iron are good electrical insulators, and different varieties show good stability in alkaline and in acid environments. In particular, their high tensile strength and flexibility make them useful as re-enforcing agents in building products. Over 200 commercial products have used asbestos (Zoltai, 1979). The minerals tremolite, actinolite, and anthophyllite have the same name for both their asbestos and their common rock-forming variety. Thus, there has been some confusion in regard to the presence of the asbestos in some products and environments. The asbestos minerals may occur in greater than trace amounts as both asbestiform fibers and mineral cleavage fragments in other commercial minerals. When the asbestiform fiber content exceeds 1% it becomes an asbestos-containing material (ACM) under the current EPA definition. The asbestiform nature is due to crystallographic properties. The sheet silicate structure of the serpentine mineral known as chrysotile has a dimensional mismatch between its tetrahedral coordinate Si2O5 layer and the octahedral coordinated MgOH (Brucite) layer, resulting in a structural deformation that causes the sheets to form cylindrical or “tubular” fibrils (Whittaker and Zussman, 1956). Weak interatomic bonds hold together many hundreds of these individual fibrils to form a fiber (Yada, 1967). The other asbestos minerals are in a class known as amphiboles, and have double chain silicate structures. Crystal

Environmental Toxicants, Third Edition Edited by Morton Lippmann Copyright
2009 John Wiley & Sons, Inc.

395

396

ASBESTOS AND OTHER MINERAL AND VITREOUS FIBERS

TABLE 12.1

Commercial Asbestos Fiber Types that may be Found in Buildingsa

Commercial Name Buildingsb

Mineral Occurrence Name

Chrysotile Crocidolite Anthophyllite Amosite Actinolite Tremolite

Chrysotile Riebeckite Anthophyllite Grunerite Actinolite Tremolite

Mineral Group

Chemical Formula

In

Serpentine Amphibole Amphibole Amphibole Amphibole Amphibole

(Mg, Fe)6(OH)8Si4O10 Na2(Fe3þ)2(Fe2þ)3(OH)2Si8O22(Mg) (Mg,Fe)2(OH)2Si8O22 Fe2(OH)2Si8O22(Mg,Mn) Ca2Fe5(OH)2Si8O22(Mg) Ca2Mg5(OH)2Si8O22(Fe)

xxx x x xx x x

a

Chrysotile always occurs with the asbestos habit and is therefore always asbestiform. Actinolite asbestos is found as a contaminant of amosite from South Africa. It is not known to be exploited anywhere in the world. Tremolite asbestos is exploited commercially in Korea. Anthophyllite asbestos is no longer commercially worked anywhere in the world. Note that the amphibole minerals anthophyllite, actinolite, and tremolite do not have a separate mineral name for their asbestos varieties as do riebeckite (crocidolite) and grunerite (amosite). b Occurrence in buildings, frequency of observation: xxx, very commonly found if product is asbestos containing; xx, commonly found; x, uncommonly found.

structure and defects facilitate the release of the ultimate amphibole asbestos fibers (Chisholm, 1983; Veblen, 1980). The formation, geological origin, and crystal structure of chrysotile inhibit cation substitution, and the extent of iron and magnesium depletion from weathering, are the main variations. By contrast, the amphiboles have a crystal structure more favorable to cation substitution. Most commercial amphibole deposits show relatively little variation in their chemistry, and can be reliably identified as individual fibers by analytical transmission electron microscopy (TEM). Walton (1982), Chisholm (1983), Langer and Nolan (1986), and Langer et al. (1990) have reviewed the chemistry, structure, and properties of asbestos. Each of the asbestos fiber types has a unique size range in terms of airborne and tissue evaluations (Pooley and Clark, 1980; Burdett, 1985). TEM-based size distributions of airborne asbestos were reviewed by Berman and Chatfield (1989), who concluded that some 9% (range: 1–50%) of chrysotile, 4% (range: 1–18%) of crocidolite, and 25% (range: 8–43%) of amosite meet the industrial hygiene definition of asbestos (i.e., fibers >5 mm long, >0.25 mm wide, and aspect (length/width) ratio >3:1. A functionally similar definition by the World Health Organization (WHO) for hazardous fibers is 5 mm long, <3 mm wide, and aspect (length/width) ratio >3:1. Fibers <0.25 mm wide would not be seen by optical microscopy, and fibers >3 mm wide would not penetrate to the lungs. Measurements in United Kingdom textile and friction product plants (Rood and Scott, 1989) found that only about 4% of chrysotile fibers fell into this category. However, a study by Dement and Wallingford (1990), of the same industries, reported some 20% of fibers that would be counted by the industrial hygiene definition. Evaluations by TEM reveal that: (1) crocidolite forms very fine fibers (0.04–0.15 mm in diameter); (2) amosite fibers are thicker (0.06– 0.35 mm); (3) chrysotile’s fibrils are smallest (0.02–0.05 mm). Specific properties that affect the biological activity of asbestos fibers include fiber type, length, diameter, and their durability within the lungs and at other sites in the body. Nonasbestiform asbestos minerals and synthetic vitreous fibers (SVF) can break up into cleavage fragments that include particles that are relatively long and thin, and meet the criteria for fiber counting protocols. Davis et al. (1991) studied the carcinogenicity of six

IMPORTANT SPECIAL PROPERTIES OF FIBERS

397

different tremolite fibrous asbestos minerals; three being asbestiform and three others being cleavage fragments. The carcinogenicity was high for all three asbestiform materials, very low for two of the cleavage fragment dusts, and intermediate for the third, which contained more fragments in the size range that would be counted as fibers. Similar findings were reported for silicon carbide (SiC) fiber whiskers and cleavage fragments that met the WHO size limits for fibers by Rodelsperger and Bruckel (2006). They found that the carcinogenicity of the whiskers was much greater than that of the cleavage fragments. 12.1.2

Synthetic Vitreous Fibers

Manmade mineral fibers (MMMF), manmade vitreous fibers (MMVF), and SVF are generic names for fibers that are generally made by spraying or extruding molten glass, rock, or furnace slag. The production technology and its historical development were summarized by Konzen (1984). Both fiber lengths and diameters are polydisperse, especially for sprayed fibers, and subsequent fabrication of fiber mats for commercial products such as insulation, filters and fiber-reinforced composites results in the breakage of fibers into shorter length segments. Konzen (1984) reported that the average fiber diameters decreased from 10 to 12 mm in the 1940s to 7.5 mm in later products, and that the percentage less than 1 mm decreased from 2.8 to 1.7% over the same period. Smaller diameter fibers (i.e., microfibers) can be fabricated, but require much more energy to produce, and are much more expensive. The superior properties of microfibers for insulation and tensile strength lead to their use in special applications where the greater costs can be justified, such as space shuttle cabins, earplugs, and so on. The chemical compositions of SVF vary with the source materials, and their mechanical properties and durability, both in the products and in the body after inhalation and deposition, can vary greatly with their composition. Rock, and slag wools are terms used for vitreous products made from the precursor materials that are melted, drawn, centrifugally spun and steam-, or air-jet blown. These processes produce discontinuous fibers that are relatively short in length. Feldspar and kaolinite, now principally confined to use in ceramic fiber manufacture, were once used to make mineral wool. Indiana limestone has been used for the production of rock wool. Slags, byproducts from many sources including: iron and steel making; and base-metal and copper smelting, have been used in slag wool production. The bulk and trace metal chemistry of these products vary greatly. The principal refractory ceramic fiber (RCF) in commerce is of alumina–silica composition. These fibers have great stability. The principal focus on SVFs in this section is on human exposures to widely used commercial fibrous glass, and the health risks associated with them. The principal basis for the assessment is: (1) the epidemiological data on occupational exposures and their effects; (2) the biological responses to fiber suspensions inhaled by laboratory animals, or injected into their lungs, or pleural or peritoneal spaces; and (3) the aerodynamics, deposition, and clearance of airborne fibers within the respiratory tract. An underlying hypothesis is that the biological effects of SVFs are essentially the same as those produced by asbestos fibers, varying in potency rather than in nature. This hypothesis is based on the morphological and toxicological similarities between SVFs and asbestos fibers. The concern arises from the well-documented evidence that asbestos fibers can cause lung fibrosis (asbestosis), bronchial cancer, and mesothelioma in humans, and that both asbestos and SVFs can cause these diseases in animals. Glass fibers have also been associated with dermatitis and eye irritation in industrial workers, but these nonrespiratory health effects will

398

ASBESTOS AND OTHER MINERAL AND VITREOUS FIBERS

not be reviewed further in this chapter, since they are not likely to be relevant to the much lower levels of nonoccupational exposures. 12.1.3

Properties of Vitreous Fibers

Different cationic metals in silica melts may coordinate oxygen differently, due to their cation charge (Z) and ionic radius (r). The immiscibility regions in glass melts are compositionally dependent, a function of Z/r2 (the determining factor in influencing cation field strength). Because of their different chemistries, SVFs behave differently in biological hosts. The aluminum–silicon ceramic fibers are stable (similar Z/r2 values), and are therefore durable in vivo. Slag wools, rich in trace metals, are usually neither stable nor durable in biological hosts, and high-soda glasses are most unstable. The issue of fiber durability is of extreme biological importance. Because SVFs are generally not stable in vivo, they appear to carry less risk of producing disease (Lippmann, 1990). Most commercially available vitreous fibers that are used in insulation, fire retardant, and acoustical applications have diameters that range from about 4 to 6 mm. These values are consistent with aerodynamic diameters in the range of 12–18 mm (Timbrell, 1972). Thus, most commercial SVF fibers are too large for penetration into the thorax, providing one important basis for the conclusion that commercial SVFs are less biologically hazardous than asbestos fibers (Lippmann, 1990). Insulating glasses may be coated with binders, for example, phenol–formaldehyde resins, or with mineral oil lubricants in a range of concentrations. The biological significance of these coating materials for long term, chronic, diseases is unknown. 12.1.4

Other Inorganic Fibers

Wollastonite, a naturally occurring calcium silicate (CaSiO3), is available in a fibrous form. However it is not biopersistent, and is not considered to be hazardous (Maxim and McConnell, 2005). Erionite, a naturally occurring zeolite mineral also exists in a fibrous form. It has been implicated as a potent cause of mesothelioma in humans and animals (Wagner et al., 1985). Durable manmade fibers can also be produced from pure chemicals. For example, DuPont developed Fybex (potassium octatitinate) and Kevlar, an aramid, as fibrous asbestos substitutes. They found that Fybex produced mesotheliomas in hamsters (Lee et al., 1981), and did not pursue commercial application. By contrast, Kevlar had biological effects similar to those associated with nuisance dusts (Lee et al., 1983). These “chemical” fibers will not be discussed in any detail here except insofar as toxicity data obtained using them sheds light on the physical properties of fibers affecting in vivo toxicity. 12.1.5

Major Commercial Uses of Inorganic Fibers

There are many ACMs found in buildings, including thermal system insulation, structural fireproofing, acoustical and decorative finishes, sheet products, floor and ceiling tiles, asbestos-containing felts, and so on. (Sawyer, 1989). A number of other construction products have, in the past, contained asbestos fibers as well, for example, spackling, patching, and plastering compounds used in dry-wall construction and interior repair. In addition to natural mineral fiber, many of these formulations contained SVF in combination with asbestos and, over recent decades, SVFs have replaced asbestos in many applications.

EXPOSURES TO FIBERS

399

The major asbestos-containing items to be found in buildings were outlined by Spengler et al. (1989).

12.2 EXPOSURES TO FIBERS 12.2.1

Exposure Indices

There has never been a fully satisfactory method for measuring airborne fiber exposures relevant to health effects, and much confusion has arisen because the various methods used do not produce indices that can readily be interconverted. Three different types of concentration indices have been used. Initially, the most widely used index was the number of particles per unit volume of air, expressed in millions of particles per cubic foot (MPPCF) and determined from dust samples collected in liquid impingers and analyzed by the now obsolete U.S. Public Health Service standard optical microscopic dust-counting technique. It used a 10 objective lens to count the number of particles that settled to the bottom of a dust-counting cell that was initially filled with a liquid suspension of particles from the impinger flask. Since there was no discrimination between fibrous and nonfibrous particles, and since fibers are a very variable fraction of the total dust in most cases, dust counting for occupational exposure evaluations was replaced by a technique that only counts fibers. At the time the fiber-counting technique was first adopted (in the United Kingdom), it was already clear that long fibers were of most concern. This, combined with the practical limitation that fibers shorter than 5 mm could not re reliably identified by light microscopy, led to the adoption of a counting procedure that uses a 45 phase-contrast objective lens to count the fibers collected on a membrane filter, provided that they have a length >5 mm and an aspect ratio >3 (ACGIH-AIHA Aerosol Hazards Evaluation Committee, 1975). The phasecontrast optical method (PCOM) is specified in the Occupational Safety and Health Administration (OSHA) occupational health standard for asbestos. Table 12.2 summarizes

TABLE 12.2 Group

Recommended Air Concentration Limits and Standards for Asbestos Year

ACGIH ACGIH ACGIH OSHA OSHA NIOSH ACGIH

1946 1968a 1970,a 1974c 1972 1976 1976 1978,a 1980c

ACGIH OSHA OSHA OSHA

1997a 1976 1986 1992

a

Limit 6

3

5  10 particles/ft 12 fibersb/mL or 2  106 particles/ft3 5 fibers/mL 5 fibers/mL 2 fibers/mL 0.1 fiber/mL 0.2 fiber/mL for crocidolite 0.5 fiber/mL for amosite 2.0 fiber/mL for chrysotile and other forms 0.1 fiber/mL for all forms 2.0 fiber/mL 0.2 fiber/mL 0.1 fiber/mL

Notice of intent. All fiber limits based on phase-contrast optical determination at 400–450 magnification. c Adopted as threshold limit value (TLV). b

400

ASBESTOS AND OTHER MINERAL AND VITREOUS FIBERS

recommended occupational exposure limits and standards used in the United States. Detailed guidance on the use of the PCOM method has been provided by the World Health Organization (WHO, 1997). The third type of concentration index is based on the mass concentration of asbestos or on the mass concentration passing a precollector meeting the British Medical Research Council (BMRC) or American Conference of Governmental Industrial Hygienists (ACGIH) sampler acceptance criteria for “respirable” dust. Some of the recent animal inhalation studies report the chamber concentrations in terms of the “respirable” mass based on samples collected using samplers that meet the BMRC criteria. Environmental exposures have been measured either in terms of fiber count or fiber mass. Fiber counts have been made using both PCOM and electron microscopy. The reported concentrations have differed according to the size distributions of the fibers, the resolving power of the microscope, and whether there was any discrimination in the analyses according to fiber type. The fiber mass index was developed by Selikoff et al. (1972). The fibers and fiber bundles in the sample are mechanically reduced to individual fibers and fibrils, which are then identified and measured by TEM. Mass concentrations in nanograms per cubic meter are calculated from the numbers of fibers and fibrils, and their dimensions. The use of these various exposure indices has sometimes led to the development of site- or industry-specific exposure–response relationships for one or more of the asbestos-related diseases, but it has not been possible to develop any generic relationships demonstrating their more general adequacy as indices of exposure or health risk. However, as discussed later in this chapter, the landmark report of the Health Effects Institute-Asbestos Research (HEI-AR, 1991), established that: (1) short asbestos fibers (i.e., those shorter than 5 mm in length) pose little, if any, health risk; and (2) the standard analyses, that count all asbestos fibers or the mass that they represent are seldom adequate to define the concentrations of the longer fibers that do pose cancer risks. Support for the need for such a fiber-length cut-off was provided by an Expert Panel for the Agency for Toxic Substances and Disease Registry (ATSDR) (ERG, 2003a). Their report was titled: “Report on the Expert Panel on Health Effects of Asbestos and Synthetic Vitreous Fibers: The Influence of Fiber Length,” and stated: “Many of the short fibers that reach the gas-exchange region of the lung are cleared by alveolar macrophages, and the rate of clearance by phagocytosis has been found to vary with fiber length. There is a strong weight of evidence that asbestos and SVFs shorter than 5 mm are unlikely to cause cancer in humans.” Similarly, in a Peer Consultation report prepared for the US EPA, it was stated that there was agreement among the panelists convened that “the available data suggest that the risk for fibers less than 5 mm in length is very low and could be zero” (ERG, 2003b). Independent analyses of published data from chronic rat inhalation studies having fiber length, diameter, and compositional data and biological outcomes have provided key evidence that the health risks are due to long fibers, especially those longer than 10 mm (Lippmann, 1988; Berman et al., 1995; Berman and Crump, 2001). 12.2.2

Exposure Levels

Esmen and Erdal (1990) reviewed published data on human occupational and nonoccupational exposure to fibers. They concluded that for the traditionally defined asbestos fibers, that is, fibers >5 mm long, large amounts of the available data suffer from the diversity of sample collection and analysis methods. Simple generalizations suggest that occupational exposures are generally several orders of magnitude higher than environmental exposures; and currently extant data and the current routine measurement practices present significant

EXPOSURES TO FIBERS

401

difficulties in the consistent interpretation of the data with respect to health effects. The human exposure data to many nonasbestos minerals that exist in fibrous habit are very scanty, and in view of the biological activity of some of these fibers, this lack of relevant data limits interpretability. With respect to asbestos exposures in buildings, the Literature Review Committee of the Health Effects Institute-Asbestos Research (HEI-AR, 1991) grouped building occupants into three main categories with respect to ACM: C1: Bystanders or nonoccupationally exposed building occupants, for example, office workers, visitors, students, and teachers. C2: Housekeeping or custodial employees who may disturb materials in the course of routine cleaning and service functions. C3: Maintenance or skilled workers that may disturb ACM in the course of making repairs, installing new equipment, or during minor renovation activity.Two other categories not often dealt with in the context of building occupancy were identified as: C4: Abatement workers or others involved in the removal or renovation of structures with ACM. C5: Firefighters and other emergency personnel who may be present during or after the fabric of the building has been extensively damaged by fire, wind, water, or earthquake. Building employees and contractor employees may disturb ACM during the course of their normal work assignments, especially during maintenance and custodial activities. They may or may not know that they are disturbing ACM, and, if they do, they may or may not have the equipment and motivation to take the appropriate precautions to minimize their exposure to airborne fibers. Exposures of such workers can be high, and warrant special concern. By far the most numerous are the C1 building occupants. Persons in categories C2–C5 fall under OSHA regulations for personal monitoring if their exposures exceed the permissible exposure limit (PEL) of 0.1 fiber/mL (f/mL) >5 mm in length as an 8-h time-weighted average, or the OSHA excursion limit of 1 f/mL in a half-hour, respectively, both as determined by PCOM. Persons in category C1 are not covered by any federal exposure limits. There are relatively few published data on the concentrations of airborne fibers in public buildings. The HEI-AR sponsored Literature Review Panel compiled the available data, both published and such unpublished data as it could assemble (HEI-AR, 1991). In their report, the panel concluded that: A large number of buildings in the U.S. and other countries have been examined for airborne asbestos fibers within the previous 20 years, and yielded many thousands of air measurements (most unpublished). However, few building environments have been characterized in sufficient detail or sampled with sufficient analytical sensitivity to adequately describe the exposures of C1 occupants. Specific details are especially lacking for episodic and point-source releases of fibers into the air of buildings from maintenance and engineering activities, from repair and renovation operations, and from normal custodial functions. Such data as are now available on the airborne concentrations of asbestos fibers of the dimensions most relevant to human health (i.e., fibers >5 mm long) generally show average concentrations on the order of 0.00001 f/mL for outdoor rural air (except near asbestoscontaining rock outcroppings) and average concentrations up to about 10-fold higher in the outdoor air of urban environments. However, outdoor urban average concentrations above

402

ASBESTOS AND OTHER MINERAL AND VITREOUS FIBERS

FIGURE 12.1 Summary of building average airborne asbestos fiber concentrations (in fibers per milliliter, with lengths >5 mm) in public and commercial buildings. (From data as reported by HEI-AR, 1991.)

0.0001 f/mL have been reported in certain circumstances as a result of local sources; for example, downwind from, or close to, frequent vehicle braking or activities involving the demolition or spray application of asbestos products. Data on ambient indoor levels of asbestos from direct TEM measurements were averaged for each of a number of individual buildings. The following data were based on 1377 air samples obtained in 197 different buildings not involved in litigation. The overall means of the studies on these buildings ranged from 0.00004 to 0.00063 f/mL, with upper 90th percentiles ranging from 0.00002 to 0.0008 f/mL. Grouped by building category, the mean concentrations were 0.00051, 0.00019, and 0.00021 f/mL in schools, residences, and public and commercial buildings, respectively, with upper 90th percentiles of 0.0016, 0.0005, and 0.0004, respectively (see Fig. 12.1).

12.3 FIBER DEPOSITION IN THE RESPIRATORY TRACT There are five mechanisms that are important with respect to the deposition of fibers in respiratory tract airways. These are impaction, sedimentation, interception, electrostatic precipitation, and diffusion (see Fig. 1.8). Impaction and sedimentation probabilities are governed by the aerodynamic diameter of the fibers, which, for long mineral fibers, are close to three times their physical diameters (St€ ober et al., 1970; Timbrell, 1972). Most impaction occurs downstream of air jets in the larger airways, where the flow velocities are high and the momentum of a fiber propels it out of the bending flow streamlines and onto relatively small portions of the epithelial surfaces (Balashazy et al., 2005; Su and Cheng, 2006). Sedimentation, on the contrary, is favored by low flow velocity, long residence times, and small airway size. Electrostatic precipitation occurs primarily by image forces, in which charged particles induce opposite changes on airway surfaces. It is dependent on the ratio of electrical charge

FIBER DEPOSITION IN THE RESPIRATORY TRACT

403

to aerodynamic drag. Little is known about the charge levels on SVFs in the workplace. Jones et al. (1983) have shown that asbestos-fiber processing operations do generate fibrous aerosols with relatively high charge levels, and that these charge levels are sufficient to cause an enhancement of fiber deposition in the lungs. Such an enhancement of fiber deposition for chrysotile asbestos was seen in rats that were exposed by inhalation (Davis, 1976). Interception increases with fiber length. The greater the length, the more likely it is that the position of a fiber end will cause it to touch a surface that the center of mass would have missed. Diffusional displacement results from collisions between air molecules and airborne fibers. For compact particles, diffusion becomes an important deposition mechanism for diameters smaller than about 0.5 mm. Fibers of similar diameter would be more massive and therefore be displaced less by a single molecular impact. Long fibers may have nearly simultaneous impacts from several gas molecules, and their random trajectories may tend to damp the net displacement. On the contrary, a single collision near a fiber end may rotate the fiber sufficiently to alter its interception probability. The role of diffusion in fiber deposition is poorly understood. Gentry et al. (1983) measured the diffusion coefficients of chrysotile and crocidolite asbestos fibers and found good agreement with theoretical predictions for chrysotile (0.4 mm mean diameter) but poor agreement with the more rod-like crocidolite (0.3 mm mean diameter). The conductive airway region of the human lung consists of a series of bifurcating airways. The trachea is the only airway segment with a length-to-diameter ratio much greater than three. Single symmetrical fibers suspended in a laminar flow stream tend to become aligned with the flow axis as they move through a lung airway. On the contrary, fiber agglomerates or nonfibrous particles would have more random orientations that would depend on their distributions of masses and drag forces. A fiber whose flow orientation differs from axial alignment would have an enhanced probability of deposition by interception. A fiber’s alignment is radically altered as it enters a daughter airway, and this loss of alignment with the flow at the entry contributes to its deposition by interception at or near the carinal edge. To the extent that a fiber is entrained in the secondary flow streams that form at bifurcations, its deposition probability by interception is further enhanced. Sussman et al. (1991a) performed an experimental study of fiber deposition within the larger tracheobronchial airways of the human lung using replicate hollow airway casts. For crocidolite fibers with diameters primarily in the 0.5–0.8 mm range, interception increased total deposition, with the effect increasing with fiber length, especially for fibers >10 mm in length. The effect was more pronounced at 60 L/min than at 15 L/min. This is consistent with greater axial alignment of the fibers during laminar flow within the airway. Morgan and Holmes (1984) and Morgan et al. (1980) exposed rats for several hours by inhalation (nose only) to glass fibers 1.5 mm in diameter and 5, 10, 30, or 60 mm long. For fibers longer than 10 mm, essentially all were deposited, mostly in the head. These results, together with the results of their earlier studies on asbestos fibers, indicate that penetrability of airborne fibers into the rat lung drops sharply with aerodynamic diameter above 2 mm. The results reported by Morgan and Holmes provided experimental verification that increasing fiber length increases lung deposition within the tracheobronchial airways. Sussman et al. (1991b) found that the deposition patterns of fibers in the larger lung airways are similar to those for particles of more compact shapes. In other words, the added deposition due to interception increased the deposition efficiencies without changing the pattern of deposition.

404

ASBESTOS AND OTHER MINERAL AND VITREOUS FIBERS

Most of the studies on particle deposition patterns and efficiencies in hollow bronchial airway casts of the larynx and the larger conductive airways of the human bronchial tree have been focused on deposition during constant flow inspirations. For studies of deposition during cyclic inspiratory flows, Gurman et al. (1984a, 1984b) used a variable-orifice mechanical larynx model (Gurman et al., 1980) at the inlet in place of the fixedorifice laryngeal models used in the prior constant flow tests. In one series of tests, two replicate casts were connected in tandem. The corresponding terminal endings were connected with rubber tubing. Deposition in the downstream cast was analyzed to determine the deposition pattern and efficiencies during expiratory flow (Schlesinger et al., 1983). Concern about sites of enhanced surface deposition density is stimulated by the observation that the larger bronchial airway bifurcations, which are favored sites for deposition, are also the sites most frequently reported as primary sites for bronchial cancer (Schlesinger and Lippmann, 1978). Deposition patterns within the nonciliated airways distal to the terminal bronchioles may also be quite nonuniform. Brody et al. (1981) studied the deposition of chrysotile asbestos in lung peripheral airways of rats exposed for 1 h to 4.3 mg/m3 of respirable chrysotile. The animals were killed in groups of 3 at 0, 5, and 24 h and at 4 and 8 days after the end of the exposure. The pattern of retention on the epithelial surfaces was examined by scanning electron microscopy of lung sections cut to reveal terminal bronchiolar surfaces and adjacent airspaces. The rat does not have recognizable respiratory bronchioles, and the airways distal to the terminal bronchioles are the alveolar ducts. In rats killed immediately after exposure, asbestos fibers were rarely seen in alveolar spaces or on alveolar duct surfaces except at alveolar duct bifurcations. There were relatively high concentrations on bifurcations nearest the terminal bronchioles and lesser concentrations on more distal duct bifurcations. In rats killed at 5 h, the patterns were similar, but the concentrations were reduced. Subsequent studies have shown that crocidolite asbestos (Roggli et al., 1987), Kevlar aramid synthetic fibers (Lee et al., 1983), and particles of more compact shape (Brody and Roe, 1983) deposit in similar patterns, and that the deposition patterns seen in the rat also occur in mice, hamsters, and guinea pigs (Warheit and Hartsky, 1990). The sudden enlargement in air path cross section at the junction of the terminal bronchiole and alveolar duct may play a role in the relatively high deposition efficiency at the first alveolar duct bifurcation. Little has previously been known about the flow profiles in this region of the lung. However, Briant (1988) has shown that a net axial core flow in a distal direction and a corresponding net annular flow in a proximal direction take place during steady-state cyclic flow in tracheobronchial airways and that this could account for such concentrated deposition on the bifurcations of distal lung airways.

12.4 FIBER RETENTION, TRANSLOCATION, DISINTEGRATION, AND DISSOLUTION The fate of fibers deposited on surfaces within the lungs depends on both the sites of deposition and the characteristics of the fibers. Within the first day, most fibers deposited on the tracheobronchial airways are carried proximally on the surface of the mucus to the larynx, to be swallowed and passed into the gastrointestinal tract. The residence time for fibers on the surface of the tracheobronchial region is too short for any significant change in the size or composition of the fibers to take place.

FIBER RETENTION, TRANSLOCATION, DISINTEGRATION, AND DISSOLUTION

405

Fibers deposited in the nonciliated airspaces beyond the terminal bronchioles are more slowly cleared from their deposition sites by a variety of mechanisms and pathways. These can be classified into two broad categories, that is, translocation and disintegration. 12.4.1

Translocation

Translocation refers to the movement of the intact fiber: (1) along the epithelial surface to dust foci at the respiratory bronchioles; (2) onto the ciliated epithelium at the terminal bronchioles; or (3) into and through the epithelium, with subsequent migration to interstitial storage sites within the lung, along lymphatic drainage pathways, and for very thin short fibers, access via capillary blood to distant sites, as suggested by Monchaux et al. (1982). Boutin et al. (1996) suggested that thin fibers longer than 5 or 10 mm migrate toward the parietal pleura via the lymphatic pathway, where they accumulate preferentially in anthracotic “black spots” of the parietal pleura. In a study by Dodson et al. (1990) comparing the fiber content of tissues from chronically exposed shipyard workers, they reported that while 10% of amphibole fibers in pleural plaque samples were longer than 5 mm and 8% were longer than 10 mm, the corresponding figures for chrysotile fibers were 3.1 and 0%. In lymph nodes, the corresponding figures for >10 and >5 mm lengths were 6.0 and 2.5% for amphiboles and 0 and 0% for chrysotile. In lung tissue, they were 41.0 and 20.0% for amphiboles and 14.0 and 4.0% for chrysotile. Boutin et al. (1996) noted that the black spots that concentrate longer fibers were in close contact with early pleural plaques. These studies indicate that fiber translocation is dependent on both fiber diameter and fiber length, and that length is an important determinant of biological responses. Translocation may also occur after ingestion of the fibers by alveolar macrophages if the fibers are short enough to be fully ingested by the macrophages. Holt (1982) proposed that fibers phagocytosed by alveolar macrophages are carried by them toward the lung periphery by passing through alveolar walls and that some of these cells aggregate in alveoli near larger bronchioles and then penetrate the bronchiolar wall. Once in the bronchiolar lumen, they can be cleared by mucociliary transport. Lentz et al. (2003) reviewed the literature on the dimensions of fibers that may translocate to the parietal pleura, and concluded that the critical dimensions were: diameter <0.4 mm and length <10 mm. They attributed the pleural plaques that developed in refractory ceramic manufacturing workers to these translocated fibers. 12.4.2

Disintegration

Disintegration refers to a number of processes, including: the subdivision of the fibers into shorter segments; partial dissolution of components of the matrix, creating a more porous fiber of relatively unchanged external size; or surface etching of the fibers, creating a change in the external dimensions of the fibers and/or complete dissolution. For SVF, fiber breakup is virtually all by length. The breakdown into smaller-diameter fibrils that is characteristic of asbestos fibers is seldom seen. For SVFs, the relative importance of breakage into length segments, partial dissolution, and surface etching to the clearance of fibers depends upon the size and composition of the fiber. In the inhalation study of Brody et al. (1981) with chrysotile, their examination of tissues by TEM revealed that fibers deposited on the bifurcations of the alveolar ducts were taken up, at least partially, by type I epithelial cells during the 1 h inhalation exposure. In the 5 h period after exposure, significant amounts were cleared from the surfaces, and there was further

406

ASBESTOS AND OTHER MINERAL AND VITREOUS FIBERS

uptake by both type I cells and alveolar macrophages. Within 24 h after the exposure, there was an influx of macrophages to the alveolar duct bifurcations. The observations provide a basis for fiber penetration of the surface epithelium that does not hypothesize movement within macrophages. Roggli and Brody (1984) exposed rats for 1 h to a chrysotile aerosol, and showed that fiber clearance was associated with sequential dimensional changes in the retained fibers, with a tendency for long, thin fibers to be retained within the interstitium of the lung parenchyma. Roggli et al. (1987) subsequently performed essentially the same study with a crocidolite aerosol. For the crocidolite, there was a progressive increase in mean fiber length with increasing time postexposure, but the change was less pronounced than that for chrysotile. In addition, there was no change in fiber diameter with time for the crocidolite. By contrast, the longitudinal splitting of the chrysotile into fibrils had caused a marked reduction of diameter with time. Accumulation of fibers in distal lung airways may, by itself, slows the clearance of fibers and other particles from the lung. Ferin and Leach (1976) exposed rats by inhalation to 10, 5, or 1 mg/m3 of UICC amosite or Canadian chrysotile for periods ranging from 1 h to 22 days. Exposures at 10 mg/m3 for 1–3 h or for >11 days at 1 mg/m3 suppressed the pulmonary clearance of TiO2 particles. Bellmann et al. (2001) demonstrated that during chronic exposure to a refractory ceramic aerosol containing both fibers and shot (more compact particles), that the clearance of the fibers was retarded by the presence of the shot. 12.4.3

Overload Associated with High Lung Burden

Based upon 6-week inhalation studies of UICC amosite in rats, Bolton et al. (1983) reported strong evidence for an overload of clearance at high lung burdens (exceeding about 1500 mg/ rat), in which a breakdown occurs of the intermediate-rate clearance mechanisms (time constants of the order of 12 days). Their hypothesis is consistent with the results of other inhalation studies in rats with asbestos (Wagner and Skidmore, 1965), quartz (Ferin, 1972), and diesel soot (Chan et al., 1984). Vincent et al. (1985) modified the above hypothesis on the basis of additional 1 year-long rat inhalation studies. They found lung burden to scale to exposure concentration in a way that seemed to contradict the overload hypothesis stated earlier. However, the general pattern exhibited by the results for asbestos is so similar to that for rats inhaling diesel fumes that they suggest that such accumulations are not specific to fibrous dust. They offered a modified hypothesis that, whereas overload of clearance can take place at high lung burdens after exposure has ceased, it is cancelled by the sustained stimulus to clearance mechanisms provided by the continuous challenge of chronic exposure. The linearity of the increase in lung burden is explained in terms of a kinetic model involving sequestration of some inhaled material to parts of the lung where it is difficult to clear. The particular sequestration model favored by Vincent et al. (1985) is one in which the longer a particle remains in the lung without being cleared, the more likely it will be sequestrated (and therefore less likely be cleared). Morrow (1988) developed a general hypothesis that dust overloading, which is typified by a progressive reduction of particle clearance from the deep lung, reflects a breakdown in alveolar macrophage (AM)-mediated dust removal as a result of the loss of AM mobility. The inability of the dust-laden AMs to translocate to the mucociliary escalator is correlated to an average composite particle volume per alveolar macrophage in the lung. When the volume of relatively nontoxic particles exceeds approximately 60 mm3/AM, the overload effect appears to be initiated. When the distributed particulate volume exceeds 600 mm3 per cell, the AM-

FIBER RETENTION, TRANSLOCATION, DISINTEGRATION, AND DISSOLUTION

407

mediated particle clearance virtually ceases, and agglomerated particle-laden macrophages remain in the alveolar region. For cytotoxic particles, these effects occur at lower loadings. Oberd€ orster et al. (1990) performed additional lung instillation and inhalation studies to further explore the Morrow hypothesis and the respective roles of both AMs and polymorphonuclear leukocytes (PMNs), whose influx is indicative of a cellular inflammatory response. On the basis of their studies, they concluded the following: 1. The delivered dose rate of particles to the lung is a determinant for the acute inflammatory PMN response: the same dose delivered over days by inhalation as opposed to sudden instillation leads to a very low response, conceivably reflecting the low release rate of phagocytosis-related inflammatory mediators (e.g., chemotactic factors) from AMs. 2. The process of phagocytosis of “nuisance” particles by AMs rather than the interstitial access of the particles appears to initiate the influx of PMNs into the alveolar space. 3. The surface area of the retained particles correlates best with inflammatory parameters rather than the phagocytized particle numbers, mass or volume. Specifically: (a) The surface area of the fraction of particles phagocytized by AMs correlates best with the influx of PMNs; (b) By contrast, increase in alveolar epithelial permeability, another sign of inflammation, correlates with the retained surface area of the particles in the total lung, rather than with the surface area retained in the alveolar space; and (c) The two inflammatory parameters “alveolar PMN influx” and “alveolar epithelial permeability” are therefore separate events triggered by different mechanisms. 4. Interstitialization of particles appears to be important for inducing interstitial inflammatory responses including the induction of fibrotic reactions. 5. If the interstitialized particle fraction exceeds the particle fraction remaining in the alveolar space, the influx of PMNs into the alveolar lumen decreases, conceivably reflecting a reversal of chemotactic gradients from alveolar space towards the interstitial space.” Jones et al. (1988) extended the inhalation studies to lower concentrations; rats inhaled International UICC amosite asbestos at 0.1 mg/m3 (equivalent to 20 fibers/mL) for 7 h a day, 5 days a week, for up to 18 months. The lung burdens were compared with the previous results for concentrations of 1 and 10 mg/m3. Taken together, these results showed lung burdens rising pro rata with exposure concentration and exposure time. This accumulation of lung burden fit a kinetic model that takes account of the sequestration of material at locations in the lung from where it cannot be cleared. Tran et al. (1997) showed that the overloading of the lung by fibers less than 15 mm long and particles follow the same kinetics, and are similarly affected by overloading. For long fibers (>25 mm), the disappearance was independent of length and lung burden, implying that the clearance of such fibers occurs by dissolution and fragmentation into shorter lengths. 12.4.4

Role of Fiber Dissolution

A differential lung clearance between fibers of chrysotile and the more rod-like amphibole asbestos fibers was shown for rats that underwent chronic inhalation exposures (Wagner et al., 1974). The lung fiber burdens of the amphiboles rose continuously throughout 2 years of exposure, and declined slowly in the rats removed from exposure after six months. By

408

ASBESTOS AND OTHER MINERAL AND VITREOUS FIBERS

contrast, the lung burdens in rats exposed to both Quebec and Zimbabwe chrysotile rose much more slowly during exposure, and seemed to decline after 12 months, even with further exposure. The biopersistence of chrysotile fibers from other locations has been studied by Bernstein and colleagues following inhalation exposures to aerosols with large number concentrations of fibers >20 mm in length. For chrysotile from the Cana Brava mine in central Brazil, the clearance half-time of the fibers >20, those 5–20, and those <5 mm in length were 1.3, 2.4, and 23 days, respectively (Bernstein et al., 2004). For chrysotile from the Coalinga mine in California (Calidria RG144), the clearance half-time of the fibers >20, those 5–20, and those <5 mm in length were 7 h, 7 days, and 64 days, respectively, while for tremolite asbestos, there was no clearance from the rat lungs over the 1 year period of observation (Bernstein et al., 2005a). The tremolite exposures produced lung inflammation, granulomas, and lung fibrosis, while the chrysotile, despite involving a much higher long fiber concentration, did not produce any measurable response. For chrysotile from the Eastern Townships of Quebec, the clearance half-time for fibers >20 mm in length was 11.4 d, which was similar to that for glass and stone wools previously studied (Bernstein et al., 2005b) Similar differential retention has been found in humans. Churg (1994) reported on analyses of lung tissue for 94 chrysotile asbestos miners and millers from the Thetford region of Quebec, Canada. The retained chrysotile and exposure atmosphere contained a very small percentage of tremolite, yet the lungs contained more tremolite than chrysotile, and the tremolite content increased rapidly with the duration of exposure. While most of the inhaled chrysotile was rapidly cleared from the lungs, a small fraction seemed to be retained indefinitely. After exposure ended, there was little or no clearance of either chrysotile or tremolite from the lungs. Albin et al. (1994) studied retention patterns in lung tissues from 69 Swedish asbestoscement workers and 96 controls. They reported that chrysotile has a relatively rapid turnover in human lungs, whereas amphiboles (tremolite and crocidolite) have a slower turnover. They also noted that: (1) chrysotile retention may be dependent on dose rate; (2) chrysotile and crocidolite retention may be increased by smoking; and (3) that chrysotile and tremolite retention may be increased by the presence of lung fibrosis. The most direct evidence for the effect of altered dust clearance rates on the retention of inhaled fibers in humans comes from studies of the fiber content of the lungs of asbestos workers in various countries. Timbrell (1982) developed a model for fiber deposition and clearance in human lungs based on his analysis of the bivariate diameter and length distributions found in air and lung samples collected at an anthophyllite mine at Paakkila in Finland. The length and diameter distributions of the airborne dust at this particular mine were exceptionally broad, and historic exposures were very high. For workers with the highest exposure and most severe lung fibrosis (Ashcroft et al., 1988), the fiber distributions in some tissue segments approached those of the airborne fibers. Adjacent tissue, analyzed for extent of fibrosis, showed severe fibrotic lesions. He concluded that long-term retention was essentially equal to deposition in such segments. Fig. 12.2 shows a series of retention curves for different degrees of lung fibrosis. These curves were determined by comparing the anthophyllite fiber size distributions in other tissue samples from the same lung with the distribution in the sample for which all fibers deposited were retained. Lung fibrosis is associated with increased fiber retention, and fiber retention is clearly associated with fiber length and diameter. The critical fiber length for mechanical clearance from the lungs is 17 mm. More precise descriptions of the effect of fiber loading in the lung on fibrosis need to be based on the use of the most appropriate index of fiber loading.

FIBER RETENTION, TRANSLOCATION, DISINTEGRATION, AND DISSOLUTION

409

FIGURE 12.2 Effect of lung fibrosis on fiber retention in human lungs as a function of fiber length. The scores for fibrosis are A: minimal, B: slight, C: moderate, D: marked, and E: severe. (Source: Lippmann and Timbrell, 1990.)

Morgan et al. (1982) and Morgan and Holmes (1984) studied the retention of 1.5 mm diameter glass fibers administered to rat lungs by intratracheal instillation. Retention at 1 year for 5 mm long fibers was 10%, whereas for 10 mm long fibers it was 20%. For the fibers that were 30 or 60 mm long, there was no measurable clearance during the first 9 months. Further retention measurements were not made for these long fibers because of evidence that they were disintegrating and dissolving. This macrophage-mediated mechanical clearance is less effective for 10 mm long fibers than for 5 mm fibers, and is ineffective for fibers 30 mm in length and longer. The results were based on the sizes of fibers recovered from rats’ lungs at various times following inhalation exposures. For the glass fibers, there was much less dissolution of the 5 and 10 mm fibers than of the 30 and 60 mm fibers. The dissolution of the long 1.5 mm diameter fibers was very nonuniform. Some were little changed in dimension, whereas others were reduced in diameter to 0.2 mm. On the contrary, for rockwool fibers >20 mm in length, there was no observable change in fiber dimensions after 6 months. Morgan and Holmes (1984) attributed the dependence of dissolution on fiber length to the differences and intra- and extracellular pH. The shorter fibers within macrophages are exposed to a pH of 7.2, whereas those outside were exposed to the extracellular pH of 7.4. Bernstein et al. (1984) and Hammad (1984) also found evidence of substantial in vivo dissolution of glass fibers. LeBouffant et al. (1984) used X-ray analysis on individual fibers recovered from lung tissue to show the exchange of cations between the fibers and the tissues. For example, the fibers can lose calcium and gain potassium. Insight on the solubility of fibers in vivo has also been obtained from in vitro solubility tests. Griffis et al. (1981) found that glass fibers suspended either in buffered saline or serum simulant at 37 C for 60 days exhibited some solubility and that the sodium content of the residual fiber was reduced. Forster (1984) used Gamble’s solution for tests on samples of

410

ASBESTOS AND OTHER MINERAL AND VITREOUS FIBERS

18 different SVFs at temperatures of 20 and 37 C and for exposure times ranging from 1 h to 180 days using static tests, tests with once-daily shaking, tests with continuous shaking, and tests with single fibers in an open bath. There was some solubility for all fibers. Klingholz and Steinkopf (1982, 1984) studied dissolution of mineral wool, glass wool, rock wool, and basalt wool at 37 C in water and in a Gamble’s solution modified by omission of the organic constituents. Most of the tests used a continuous-flow system in which the pH was 7.5–8. There was relatively little dissolution in distilled water in comparison to that produced by the modified Gamble’s solution. The surfaces developed a gel layer that, for the smaller diameters, extended throughout the fiber cross section. Thus, the fibers can become both smaller in outline and more plastic to deformation. Scholze and Conradt (1987) performed a comparative in vitro study of the chemical durability of SVFs in a simulated extracellular fluid under flow conditions. Seven vitreous, three refractory, and three natural fibers were involved. Samples of the leachate were analyzed, and the silicon concentrations were used to roughly classify the fibers according to their chemical durability in terms of glass network dissolution. A durability ranking of fiber materials was expressed in terms of a characteristic time required for the complete dissolution of single fibers of given diameter. SVFs exhibited relatively poor durability (with network dissolution velocities ranging from 3.5 to 0.2 nm per day for a glass wool and an E glass fiber, respectively), whereas natural fibers were very persistent against the attack of the biological fluid (e.g., less than 0.01 nm per day for crocidolite). Johnson et al. (1984) exposed rats to SVF aerosols at 10 mg/m3 for 7 h/day, 5 days/week for 1 year as compared to the single exposure of several hours duration used by Morgan and Holmes. The percentage of glass fibers with diameters less than 0.3 mm that were recovered from the lungs was consistently less than that in the original fiber suspension, and the reduction was more marked in the animals that were sacrificed following a period of recovery from the exposures than from those sacrificed at the end of the exposure. The degree of fiber etching increased with residence times in the lungs. Glass wool with and without resin was also etched, but to a lesser extent, and the etching of the rock wool fibers was considerably less. Bellmann et al. (1986) instilled reference suspensions of UICC crocidolite and chrysotile A, as well as suspensions of glass fibers in rat lungs and examined the residual fibers after 1 day and 1, 6, 12, and 15 months. Crocidolite fibers longer than 5 mm did not decrease in number for over 1 year. The number of chrysotile fibers >5 mm doubled, probably as a result of longitudinal splitting, while the number of glass fibers >5 mm was reduced by dissolution, with a half-time of 55 days. All fibers <5 mm in length were cleared with half-times of 100– 150 days. When the crocidolite fibers were pretreated in acid, there was no change in retention. On the contrary, acid-treated chrysotile and glass fibers had much more rapid clearance, with half-times of 2 and 14 days, respectively. In a 2-year follow-up study, Bellmann et al. (1987) reported the persistence of some SVFs, crocidolite and chrysotile in the rat lung after intratracheal instillation. Experiments were based on the assumption that thin, long and durable fibers are of special importance for the carcinogenic potency. Parameters measured included: number of fibers; diameter and length distribution of fibers retained in lung ash; and leaching of various elements from fibers longer than 5 mm. For a special type of glass microfiber and for ceramic wool, which both had low alkaline earth content, the half-times of lung clearance were similar to that for crocidolite. Another type of glass microfiber, with a very low halftime, had a high alkaline earth content and a median diameter of about 0.1 mm. The glass and rock wools studied, which were thicker than the other fibers, had intermediate half-times.

FIBER RETENTION, TRANSLOCATION, DISINTEGRATION, AND DISSOLUTION

411

Collier et al. (1994) studied the behavior of two experimental continuous filament glass fibers of 2 mm diameter and 50 mm length following intraperitoneal injections of 5 mg in rats. They had in vitro dissolution rates of 150 and 600 ng/cm2/h. In the lung, the diameters of the long fibers (>20 mm) declined at a rate consistent with their exposure to a neutral pH environment. The diameters of shorter fibers declined much more slowly, consistent with exposure to the more acidic environment found in the phagolysosomes of alveolar macrophages. In the peritoneal cavity, all fibers, regardless of length, dissolved at the same rate as short fibers in the lung. The effect of dose on the distribution of fibers in the peritoneal cavity was investigated using experimental glass fibers and a powder made from ground fibers. At doses up to 1.5 mg, both fibers and powder were taken up by the peritoneal organs in proportion to their surface area, and uptake was complete 1–2 days after injection. At higher doses, the majority of the material in excess of 1.5 mg formed clumps of fibers (nodules) that were either free in the peritoneal cavity or loosely bound to peritoneal organs. These nodules displayed classic foreign body reactions, with an associated granulomatous inflammatory response. Collier et al. (1997) reported on the clearance of two stone wool fibers administered to rats by intratracheal instillation; one a conventional product (MMVF21) and the other an experimental, more soluble fiber (HTN). Unlike glass wool, stone wool is more soluble at the acid pH in macrophages than in the more neutral lung tissue. They found that MMVF21 had relatively slow clearance, with somewhat faster clearance for short fibers. The clearance of HTN was much faster. Eastes and Hadley (1995) administered suspensions of fibers to rats by intratracheal instillation, and the numbers, lengths, and diameters of fibers recovered from the lungs were measured by PCOM at intervals up to 1 year. Five different glass fibers had dissolution rates ranging from 2 to 600 ng/cm2/h measured in vitro in simulated lung fluid at pH 7.4. For fibers longer than 20 mm, the peak diameter decreased steadily with time after instillation, at the same rate measured for each fiber in vitro, until it approached zero. Measurements of the total number of fibers remaining in the rats’ lungs at times up to 1 year after instillation suggest that few of the administered fibers were being cleared by macrophage-mediated transport via the conducting airways. They concluded that glass fibers longer than 20 mm are removed from the lung by dissolution at the same rate measured in vitro. In a study of dissolution of inhaled fibers by Eastes and Hadley (1995), rats were exposed for 5 days to four types of airborne, respirable-sized SVF and to crocidolite fibers. The SVFs included two glass wools, and one each of rock and slag wool. After exposure, animals were sacrificed at intervals up to 18 months, and the numbers, lengths, and diameters of a representative sample of fibers in their lungs were measured. Long fibers (>20 mm) were eliminated from the rats’ lungs at a rate predicted from the dissolution rate measured in vitro. The long SVFs were nearly completely eliminated in several months, whereas the long crocidolite asbestos fibers remained in significant numbers at the end of the study. The number, length, and diameter distributions of fibers remaining in the rats’ lungs agreed well with a computer simulation of fiber clearance that assumed that the long fibers dissolved at the rate measured for each fiber in vitro, and that the short fibers of every type were removed at the same rate as short crocidolite asbestos. Thus, long SVFs were cleared by complete dissolution at the rate measured in vitro, and short fibers did not dissolve and were cleared by macrophage-mediated physical removal. In an inhalation study, using nine fiber types, Bernstein et al. (1996) exposed rats to an aerosol (mean diameter of 1 mm) at a concentration of 30 mg/m3, 6 h/day for 5 days with postexposure sacrifices at 1 h, 1 day, 5 days, 4 weeks, 13 weeks, and 26 weeks. At 1 h

412

ASBESTOS AND OTHER MINERAL AND VITREOUS FIBERS

following the last exposure, the nine types of fibers were found to have lung burdens ranging from 7.4 to 33  106 fibers/lung with geometric mean diameters (GMD) of 0.40–0.54 mm, reflecting the different bi-variate distributions in the exposure aerosols. The fibers cleared from the lungs following exposure with weighted half-lives ranging from 11 to 54 days. The clearance was found to closely reflect the clearance of fibers in the 5–20 mm length range. An important difference in removal was seen between the long fiber (L > 20 mm) and shorter fiber (L between 5 and 20 mm and L < 5 mm) fractions, depending upon composition. For all glass wools and the stone wools, the longer fibers were removed faster than the shorter fibers. It was found that the time for complete fiber dissolution based on the acellular in vitro dissolution rate at pH 7.4 was highly correlated (r ¼ 0.97, p < 0.01) with the clearance halftimes of fibers >20 mm in length. No such correlations were found with any of the length fractions using the acellular in vitro dissolution rate at pH 4.5. Examination of the fiber length distribution and particles in the lung from 1 h through 5 days of exposure indicated that, especially for those fibers that form leached layers, fiber breakage may have occurred during this early period. These results demonstrate that, for fibers with high acellular solubility at pH 7.4, the clearance of long fibers is very rapid. Eastes and Hadley (1996) fitted much of the data cited above into a mathematical model of fiber carcinogenicity and fibrosis. Their model predicts the incidence of tumors and fibrosis in rats exposed to various types of rapidly dissolving fibers in an inhalation study or in an ip injection experiment. It takes into account the fiber diameter and the dissolution rate of fibers longer than 20 mm in the lung, and predicts the measured tumor and fibrosis incidence to within approximately the precision of the measurements. The underlying concept for the model is that a rapidly dissolving long fiber has the same response in an animal bioassay as a much smaller dose of a durable fiber. Long, durable fibers have special significance, since there is no effective mechanism by which these fibers may be removed. In particular, the hypothesis is that the effective dose of a dissolving long fiber scales as the residence time of that fiber in the extracellular fluid. The residence time of a fiber is estimated directly from the average fiber diameter, its density, and the fiber dissolution rate as measured in simulated lung fluid at neutral pH. The incidence of fibrosis in chronic inhalation tests, the observed lung tumor rates, and the incidence of mesothelioma in the ip model, were all well predicted by the model, The model allows one to predict, for an inhalation or ip experiment, what residence time and dissolution rate are required for an acceptably small tumorigenic or fibrotic response to a given fiber dose. For an inhalation test in rats at the maximum tolerated dose (MTD), the model suggests that less than 10% incidence of fibrosis would be obtained at the maximum tolerated dose of 1 mm diameter fibers if the dissolution rate were greater than 80 ng/cm2/h. The dissolution rate that would give no detectable lung tumors in such an inhalation test in rats is much smaller. Thus, a fiber with a dissolution rate of 100 ng/cm2/h has an insignificant chance of producing either fibrosis or tumors by inhalation in rats, even at the maximum tolerated dose. This model provides manufacturers of SVFs with design criteria for fibrous products that minimize, if not eliminate, their potential for producing adverse health effects. Support for the use of biopersistence data for the prediction of fibrosis and tumor responses in rats from both ip injection studies and chronic inhalation studies for fibers >5 and >20 mm in length was provided by Bernstein et al. (2001a, 2001b). For the inhalation studies they used collagen deposition at the broncho-alveolar junction as a predictor of interstitial fibrosis on the basis that it has been shown to be associated with tumor response in previous studies.

FIBER-RELATED DISEASES/PROCESSES

413

12.5 PROPERTIES OF FIBERS RELEVANT TO DISEASE Fiber dimensions, chemical composition, and surface properties are important factors in biological reactivity of mineral fibers. As discussed previously, fiber length influences the deposition, clearance, and translocation of fibers in the lungs. Fiber length also determines clearance from the pleural and peritoneal spaces—fibers longer than 8 mm are trapped at the mesothelial lining because the opening of lymphatic channels draining these spaces are 8– 12 mm in diameter (Moalli et al., 1987). This provides an anatomic basis for the Stanton hypothesis that long fibers, regardless of their chemical composition, are more effective in producing mesotheliomas than shorter fibers, after direct intrapleural or intraperitoneal injection into rodents (Stanton et al., 1977). Cations within the crystal lattice may affect the toxicity of asbestos fibers. Mg2þ ions on the surface of chrysotile asbestos are important in cytotoxicity and carcinogenicity; acidleached fibers are less active than native fibers (Monchaux et al., 1981). The Fe2þ and Fe3þ content of amphibole fibers may be important because these cations can catalyze the Fenton or Haber–Weiss reactions, generating highly toxic and potentially mutagenic ROS (Weitzman and Graceffa, 1984; Zalma et al., 1987). Asbestos fibers generate reactive oxygen and nitrogen species (ROS/RNS), causing oxidation and/or nitrosylation of proteins and DNA. The ionic state of iron within asbestos fibers influences the oxidant-inducing potential and its influence on macromolecules, signal conduction pathways, inflammation and proliferation (Shukla et al., 2003).

12.6 FIBER-RELATED DISEASES/PROCESSES Macrophages are the initial target cells of inhaled particles that are deposited in the terminal airways and alveolar spaces. Phagocytosis of mineral fibers by macrophages leads to generation of ROS, and release of lysosomal enzymes, arachidonic acid metabolites, neutral proteases, chemotactic factors, and growth factors (Adamson, 1997). The interactions between mediators released from macrophages and other inflammatory cells and the target cell populations can initiate a sequence of events culminating in: fibrosis of the lungs and pleura; bronchogenic carcinoma; and malignant mesothelioma. However, for shorter exposures (2 weeks) to chrysotile, the early fibrotic lesions in rats are gradually resolved over the course of the following year (Pinkerton et al., 1997). Diffuse, bilateral interstitial fibrosis of the lungs characterizes asbestosis, a disease that usually develops in humans after prolonged exposure to high doses of asbestos fibers. Progressive scarring of the alveolar walls due to increased proliferation of fibroblasts and deposition of collagen produces radiographic evidence of disease, and interferes with gas exchange, leading to disability and premature death. The sequence of events that lead to the development of asbestosis includes the following: 1. Asbestos fibers are phagocytized by alveolar and/or interstitial macrophages. 2. Release of ROS from alveolar macrophages causes acute injury to the alveolar epithelial lining. The importance of hydrogen peroxide in the pathogenesis of asbestosis was demonstrated by the protection against asbestos-induced pulmonary fibrosis provided by catalast conjugated to polyethylene glycol (Mossman et al., 1990a, 1990b).

414

ASBESTOS AND OTHER MINERAL AND VITREOUS FIBERS

3. Phagocytosis of asbestos fibers by alveolar or interstitial macrophages also triggers increased synthesis and release of growth factors for fibroblasts. Growth factors are released from macrophages exposed to asbestos fibers in vitro or in vivo: a homologue of platelet-derived growth factor (PDGF) (Bauman et al., 1990) and transforming growth factor-b (TGF-b) (Kane and McDonald, 1993). These growth factors cause chemotaxis of inflammatory cells and fibroblasts, stimulation of collagen synthesis, and inhibition of collagen degradation. Another reaction to asbestos exposure is the development of acellular fibrous scars, called pleural plaques, on the parietal pleural lining and diaphragm. Asbestos exposure may also lead to pleural effusions or diffuse fibrosis of the visceral pleura, but these reactions cause little disability. The most important reaction of the pleural and peritoneal linings to asbestos fibers is development of diffuse malignant mesothelioma, a rare neoplasm that is most closely associated with occupational or environmental exposure to amphibole forms of asbestos after a long latent period (up to 30–60 years). There is no increased incidence in cigarette smokers or in workers with asbestosis. This malignant neoplasm has a variable histology, ranging from epithelial to fibroblastic or mixed patterns. Malignant mesothelioma has usually spread diffusely when first diagnosed and responds poorly to radiation or chemotherapy (Craighead, 1987). The following sequence of events is hypothesized to lead to the development of diffuse malignant mesothelioma. 1. Phagocytosis of fibers that reach the pleura or peritoneal lining by macrophages. 2. Release of ROS, causes acute injury to the mesothelial cell monolayer lining the pleural or peritoneal spaces. This injury can be prevented by coating the administered fibers with the iron chelator, deferoxamine, with exogenous superoxide dismutase, or with catalase (Goodglick and Kane, 1990; Kane and McDonald, 1993). 3. Acute injury to the mesothelial lining, which is repaired by proliferation of adjacent, uninjured mesothelial cells. Growth factors released from macrophages, following phagocytosis of asbestos fibers, may modulate mesothelial cell regeneration (Kane and McDonald, 1993). 4. Direct interaction of asbestos fibers with the regenerating mesothelial cell population, which may cause chromosomal aberrations and aneuploidy. Additional DNA damage may be produced by reactive oxygen species, especially the hydroxyl radical produced by the iron-catalyzed Haber–Weiss reaction (Barrett et al., 1989; Floyd, 1990). 5. Repeated episodes of mesothelial cell injury and regeneration may lead to the emergence of a subpopulation of autonomously proliferating cells. 6. Neoplastic mesothelial cells may produce growth factors that promote growth of an invasive tumor. It is well established that workers exposed to asbestos fibers have an increased risk of developing bronchogenic carcinoma, and that workers who also smoke cigarettes have a greater risk. Cancers can arise from the epithelial lining of the large airways or terminal bronchioles. Bronchogenic carcinomas have a variety of histologic appearances: adenocarcinoma, squamous cell carcinoma (presumably arising in areas of squamous metaplasia of the respiratory epithelium), large cell carcinoma, and small cell (oat cell) carcinoma. These are the same histologic types of cancer associated with cigarette smoking in the absence of asbestos exposure (Mossman and Craighead, 1987).

REVIEW OF BIOLOGICAL EFFECTS OF SIZE-CLASSIFIED FIBERS

415

Lung cancer and mesothelioma are known to occur in people without radiographic evidence of lung fibrosis. deKlerk et al. (1997) demonstrated that the level of radiographic fibrosis conferred additional risk for lung cancer beyond that associated with level of exposure, but that asbestosis was not a prerequisite for asbestos-associated cancer. Bronchogenic carcinomas that develop in cigarette smokers show multiple alterations in proto-oncogenes and tumor-suppressor genes. It is unknown whether similar molecular changes are present in those malignant tumors that result from cigarette smoking in combination with asbestos exposure. Most of the experimental evidence suggests that asbestos fibers act as a cocarcinogen or tumor promoter in the respiratory lining, in conjunction with multiple components of cigarette smoke that may act as initiators. Some of these effects of asbestos fibers on the tracheobronchial epithelium may be mediated by ROS, since they are decreased by addition of various scavenging enzymes (e.g., superoxide dismutase, catalase) to these in vitro model systems.

12.7 REVIEW OF BIOLOGICAL EFFECTS OF SIZE-CLASSIFIED FIBERS IN ANIMALS AND HUMANS The pathological effects produced by fibers depend upon both the characteristics of the fibers and their persistence at sensitive sites. A number of carefully designed studies have been performed in which the size distributions of fiber suspensions have been well characterized as well as their persistence and/or effects. 12.7.1

In vivo Exposures by Instillation into Animal Lung Airways

King et al. (1946) instilled 100 mg of Rhodesian chrysotile into rabbit lungs at monthly intervals. One group received fibers microtomed to a length of 15 mm, and another group received fibers cut to 2.5 mm in length. At this huge dosage level, both groups showed foreign body reactions in the lungs. The long fiber produced a nodular reticulinosis, whereas the short fiber produced a diffuse interstitial reticulinosis. Wright and Kuschner (1977) used short and long asbestos and SVF in intratracheal instillation studies in guinea pigs. With suspensions containing an appreciable number of fibers longer than 10 mm, all of the materials produced lung fibrosis, although the yields varied with the materials used. However, with equal masses of short fibers of equivalent fiber diameters, none produced any fibrosis. The yields were lower for the long glass fibers than for the long asbestos, and this was attributed to their lesser durability within the lungs. 12.7.2

In vivo Exposures in Animals via Intraperitoneal Injection

For fibers injected ip (Davis, 1976; Pott et al., 1976; Wagner et al., 1976), or for fibers placed in a pledget against the lung pleura, a similar kind of fiber size and composition dependence was observed (Stanton and Wrench, 1972). The yield of mesotheliomas varied with fiber diameter and length, and with dose, with very little response when long, thin fibers were not included. Asbestos fibers were more effective than glass in these studies also. At a dose of 2 mg of chrysotile, crocidolite, or glass fiber, Pott et al. (1976) found only slight degrees of fibrosis, but tumor yields of from 16 to 38% in rats. When the chrysotile was milled to the extent that 99.8% of the fibers were shorter than 5 mm, the dose required to produce a comparable tumor yield (32%) was 50 times greater.

416

ASBESTOS AND OTHER MINERAL AND VITREOUS FIBERS

Various hypotheses have been proposed to account for the pathological effects produced by asbestos. One was the contamination of the surface by trace metal and/or organic carcinogens. However, the studies of Stanton and Wrench (1972) found that surface contaminants played no role in mesothelioma yield, and concluded that the carcinogenicity of asbestos and fibrous glass was primarily related to the structural shape of these fibrous materials rather than their surface properties. Miller et al. (1999a) reviewed the collective outcomes of 9 rat intraperitoneal injection studies involving fibers of amosite, silicon carbide, four vitreous products (100/475, MMVF10, MMVF21, MMVF22), and three refractory ceramic products (RCF1, RCF2, RCF4) on mesothelioma. They reported a link between the numbers of injected fibers >20 mm in length, and to the biopersistence in the rat lung of fibers >5 mm in length. 12.7.3

In vivo Exposures of Animals by Inhalation

The relative potencies of the various mineral forms of asbestos after inhalation exposure are still not firmly established. Crocidolite is generally considered the most hazardous because of its association with significant numbers of human mesotheliomas. The relative toxicity of various asbestos minerals have been compared in a variety of experimental inhalation studies on small animals, but the results are in apparent conflict. Wagner (1963) reported more asbestosis with amosite than with chrysotile in guinea pigs, rats, and monkeys. However, in studies involving inhalation exposures of rats for 1 day to 2 years, Wagner et al. (1974), found that amosite and crocidolite were the least fibrogenic of five types of UICC asbestos, the others being Canadian chrysotile, Rhodesian (Zimbabwe) chrysotile, and anthophyllite (see Table 12.3). Holt et al. (1965) found no differences in the fibrogenic potential of chrysotile, crocidolite, amosite, and anthophyllite. Davis et al. (1978) used UICC chrysotile A, amosite, and crocidolite in 12-month rat exposures at respirable mass concentrations comparable to those used by Wagner et al. (1974) and found a similar pattern; that is, chrysotile was the most fibrogenic, and amosite and crocidolite the least. Heitt (1978) exposed guinea pigs by inhalation for 9 and 18 days and also found that chrysotile was more fibrogenic than amosite. Davis et al. (1986b) subsequently repeated the protocol with amosite with fiber length both shorter and longer than UICC amosite. The short amosite produced virtually no fibrosis, whereas the long amosite was more fibrogenic than chrysotile. The most fibrogenic asbestos was tremolite (see Table 12.3). Berman et al. (1995) analyzed the lung tumor and mesothelioma responses from 13 of the chronic inhalation studies in rats performed by Davis and colleagues at the Institute of Environmental Medicine in Edinburgh in relation to new measurements of the fiber distributions on archived chamber atmosphere sampling filters. The measure most highly correlated with tumor incidence was the concentration of fibers >20 mm in length. Miller et al. (1999b) reviewed the collective outcomes of 18 rat inhalation studies involving fibers of amosite, silicon carbide, four vitreous products (100/475, MMVF10, MMVF21, MMVF22), and three refractory ceramic products (RCF1, RCF2, RCF4). The primary influences on biological responses was the number of fibers <1.0 mm in diameter and >20 mm in length, and the dissolution rate of the fibers. Another important observation was that in vivo and in vitro cell responses did not significantly predict the risk of cancer following inhalation. McConnell et al. (1999) exposed hamsters for 78 weeks to amosite and two different fibrous glasses (MMVF10a and MMVF33) at 250–300 f/mL for fibers >5 mm in length, and to amosite at 125 and 25 f/mL as well. MMVF10a produced only mild inflammation, while

417

0.25

0.32 0.37

10

10 10

Amostiteshort Amositelong

Davis et al. (1986)

0.38 0.38 0.38 0.42 0.42

NR NR

10 10

10 10 5 10 2

NRcj NR NR

dm (mm)

10 10 10

Conc (mg/m3)

Tremolite

Amosite Crocidolite Crocidolite Chrysotile Chrysotile

UICC

Amosite Anthophvillite Crocidolite Chrysotile Canadian Rhodesian

UICC

Fiber Type

Resp.

30

1.7

28

16 12 12 30 30

NR NR

NR NR NR

>5 mm

10

0.1

7

2.7 4 4 16 16

NR NR

NR NR NR

>10 mm

Length (%)

Fiber Parameters

40

42

39

43 40 43 40 42

137 144

146 145 141

No. Rats

3 (7.5)

1 (2.4)

2 (5.1)

0 0 0 0 1 (2.4)

4 (2.9) 0

1 (0.7) 2 (1.4) 4 (2.8)

(4.7) (2.5) (4.7) (18) (14)

3 (7.5)

0

2 (5.1)

2 1 2 7 6

20 (15) 19 (13)

19 (13) 22 (15) 26 (18)

Adenoma

3 (7.5)

0

8 (21)

0 0 0 6 (15) 1 (2.4)

11 (8.0) 19 (13)

5 (3.4) 8 (5.5) 7 (5.0)

Adeno-CA

4 (10)

0

8 (21)

0 0 0 2 (5.0) 1 (2.4)

6 (4.4) 11 (7.6)

6 (4.1) 8 (5.5) 9 (6.4)

Squam-CA

Number (%) of Rats with Tumors Mesothelioma

Summary of Results of Mineral Fiber Inhalation Studies in Rats

Davis et al. (1985)

Davis et al. (1978)

Wagner et al. (1974)

Reference

TABLE 12.3

6.0 5.8

4.3 6.2 4.8

11.0

0.15

14.5

2.6 1.4 0.8 9.2 3.5

%bi

(continued)

Interstitial Scoreah

Fibrosis

418

Chrysotileshort Chrysotilelong

0.17 0.18

10

12

5

53 44

2

0.7

12 7

>10 mm

Length (%) >5 mm

0.30 0.22

dm (mm)

10

10 10

Conc (mg/m3)

Resp.

Fiber Parameters

40

40

28 28

No. Rats

From Lippmann (1988). a Relative scale where 1, nil; 2, minimal; 4, slight; 6, moderate; 8, severe at 24 months. b Percentage of tissue with fibrosis at 27–29 months. c Not reported.

Davis (1987)

UICC

Wagner et al. (1985)

Crocidolite Erionite

Fiber Type

(Continued)

Reference

TABLE 12.3

3 (7.5)

1 (2.5)

0 27 (96)

Mesothelioma

8 (20)

1 (2.5)

0 0

Adenoma

6 (15)

6 (15)

0 0

Adeno-CA

Number (%) of Rats with Tumors

5 (13)

0

1 (3.6) 0

Squam-CA

Interstitial Scorea

Fibrosis

12.6

2.4

%b

REVIEW OF BIOLOGICAL EFFECTS OF SIZE-CLASSIFIED FIBERS

419

MMVF33 produced more severe inflammation and mild interstitial and pleural fibrosis, as well as one mesothelioma. Amosite produced severe pulmonary fibrosis and many mesotheliomas (3.6%, 25.9%, and 19.5% at the low, medium and high doses). The effects were most closely related to the retained fibers >20 mm in length, and inversely paralleled the in vitro dissolution rates. Hesterberg and Hart (2000) reviewed the results of chronic inhalation studies in rats of 8 SFVs (RCF1, thin E-glass, 475-glass, 901-glass, CT-glass, slag wool, stone wool, rock wool), as well as amosite and crocidolite. The aerosols contained 100 f/mL >20 mm in length. Rats were exposed for 6 h/day, 5 days/week for 2 years and hamsters for 1.5 years. They also had biopersistence data for 5-day inhalations with retention measurements extending over 1 year. The more biopersistant fibers were fibrogenic (rock wool), or fibrogenic and carcinogenic (amosite, crocidolite, RCF1, 475-glass, and thin E-glass). Cullen et al. (2000) compared the pathogenicity of amosite to that of two special purpose glass microfibers with low dissolution rates (104E and 100/475) in a study in which rats were exposed for 7 h/d, 5 days/week for 12 months to 1000 f/mL longer than 5 mm. In terms of mesothelioma and lung cancers produced after the exposures and twelve months without further exposure, 104E and amosite fibers were considerably more potent than 100/475 fibers. They attributed the lower pathogenicity of 100/475 to the greater leaching of its component elements while in the lungs. 12.7.4

Fibers Retained in the Lungs of Exposed Humans

It is generally believed that amphibole fibers account for much of the mesothelioma incidence among exposed workers, even when they are predominantly exposed to chrysotile, since amphibole fibers are more biopersistent. Pooley (1976) examined postmortem lung tissue from 20 workers with asbestosis in the Canadian chrysotile mining industry and found that amphibole and other fibers were present in 16 cases. In seven of these, they were more numerous than chrysotile. In a later study of lung asbestos in chrysotile workers with mesothelioma, Churg et al. (1984) reported that the concentration ratio between cases and controls was 9.3 for tremolite but only 2.8 for chrysotile. In a Norwegian plant using 91.7% chrysotile, 3.1% amosite, 4.1% crocidolite and 1.1% anthophyllite, Gylseth et al. (1983) reported that the percentage of chrysotile in lung tissue ranged between 0 and 9%, while the corresponding numbers for the amphiboles were 76–99%. Case et al. (2000) examined the relationships between asbestos fiber type and length in the lungs of chrysotile textile workers as well as miners and millers in Quebec. Despite the lower cancer risk for the Quebec workers, the chrysotile, tremolite, total amphibole, and total count of fibers longer than 18 mm were all highest in the Quebec workers. They concluded that the textile workers experience should not be used to assess the cancer risks in other cohorts. In their review of the assessment of mineral fibers from human lung tissue, Davis et al. (1986a) attributed the high amphibole/chrysotile ratios to the dissolution of chrysotile within lung tissue, and the generally poor correlation between dust counts and mesothelioma as likely to be due to the differences among the various asbestos types in the fraction that reaches the pleural surface. Amosite fibers need to be longer to produce pulmonary fibrosis and pulmonary tumors in experimental animals than to produce mesotheliomas after injection (Davis et al., 1986b). Davis (1987) notes that both chrysotile and amphibole asbestos fibers inhaled by rats are plentiful in the most peripheral alveoli bordering on the pleura, but penetration of the external elastic lamina of the lung appears to be a rare event. On the contrary, erionite, a natural zeolite fiber, causes a very high incidence of mesotheliomas

420

ASBESTOS AND OTHER MINERAL AND VITREOUS FIBERS

in humans exposed to low environmental concentrations (Baris et al., 1987) and 100% incidence in rats exposed by inhalation (Wagner et al., 1985). Davis (1987) reported that the erionite used by Wagner et al. (1985) had a general appearance and fiber size distribution very close to that of UICC crocidolite, which produces a much smaller mesothelioma yield in rats exposed by inhalation. He attributed the difference to the enhanced ability of erionite to cross the pleural membrane.

12.8 CRITICAL FIBER PARAMETERS AFFECTING DISEASE PATHOGENESIS 12.8.1

Critical Fiber Dimensions for Asbestosis

Asbestosis has been caused by exposure to high concentrations of respirable fibers of all of the commercially exploited kinds of asbestos. Within the respirable fraction, the fibers often differ in diameter and length distributions and in retention times. The influence of these variables on fibrosis in the human lung was systematically explored and described by Timbrell et al. (1987) through analyses of both retained fibers and fibrosis in lung samples from exposed workers. He analyzed the fiber distributions in 0.5 g lung samples from several hundred workers by a technique known as magnetic alignment and light scattering (MALS) that he had previously developed (Timbrell, 1982). Optical microscopy was applied to measure the degree of fibrosis in paraffin sections prepared from adjacent samples of the same specimens. Wide intra- and intersubject variations were observed in fiber concentration and fibrosis score. The quantitative relationships between fibrosis and amphibole fibers in mineworkers’ postmortem lung specimens were determined for the main sources of amosite (Transvaal, South Africa), anthophyllite (Paakkila, Finland), and crocidolite (NW Cape and Transvaal, South Africa and Wittenoom, Australia). As illustrated in Fig. 12.3, the fibrosis-producing ability of the fibers was independent of amphibole type when normalized by the total surface area of long resident fibers per unit weight of lung tissue, presumably because the surface area determined the magnitude of the fiber–tissue interface. The wide range of the concentrations of retained fiber required to produce the same degree of fibrosis in the groups of mineworkers, when fiber quantity is expressed as number or total mass stemmed from the large differences between the distributions in diameter and length of the airborne fibers. Although the main focus of the Timbrell et al. (1987) study was on amphibole asbestos, they also reported results for three Wittenoom workers whose dominant exposure was to chrysotile asbestos. For these workers, chrysotile produced a similar degree of fibrosis to Wittenoom crocidolite for equal fiber mass concentrations in the lungs. Long residence in the tissue had almost completely dispersed the chrysotile fibers into fibrils, to give them a ratio of total surface area to mass resembling that of the particularly fine Wittenoom crocidolite fibers. The result indicates that the fibrogenicity of the retained chrysotile per unit of surface area within the lungs was similar to that of the amphiboles. Timbrell et al. (1987) also reported that amphibole mineworkers with a given fiber mass concentration in their lungs showed much higher degrees of fibrosis than gold miners with roughly the same mass concentration of retained quartz grains. The amphibole and quartz produced about the same fibrogenicity per unit of surface area, but the smaller diameters and higher area/mass ratios of the amphibole fibers endowed them with the greater surface area and thereby the superior fibrosis-producing capability.

CRITICAL FIBER PARAMETERS AFFECTING DISEASE PATHOGENESIS

421

FIGURE 12.3 Relationships between lung fibrosis scale and relative concentrations of fibers per unit weight of dry lung tissues. The lines connect data points from the same subject. The relative fiber surface area normalizes the data better than either the relative fiber number concentration of the fiber mass concentration. (Source: Lippmann, 1988.)

Knowledge of the interrelationships between retained fibers and fibrosis is critical in understanding the pathogenesis of the disease but is inadequate, by itself, in evaluating exposures to airborne fibers. This was recognized by Timbrell (1983, 1984), who developed a mathematical model relating fiber deposition and retention to analyses of lung samples. Specifically, he used samples from a woman at Paakkila who worked at a job that gave her exposure to an amphibole (anthophyllite) at high concentrations of fibers with a range of diameters and lengths sufficiently wide to encompass the size limits of respirable fibers. Her lungs contained 1.3 mg fiber/g of dry tissue, and she had asbestosis. One lung sample contained a fiber distribution matching the expected deposition. Timbrell speculated that severe fibrosis in the tissue in this sample had blocked the macrophage-mediated clearance. Another sample from the same lung yielded a retention pattern more closely matching those found in other Paakkila workers, with small fiber burdens and virtually no short fibers. He assumed that the latter represents long-term retention in the normal lung.

422

ASBESTOS AND OTHER MINERAL AND VITREOUS FIBERS

From the differences in retention, Timbrell developed a model for the retention of fibers as a function of length and diameter. Fiber retention rises rapidly with fiber lengths between 2 and 5 mm and peaks at 10 mm. Fiber retention also rises rapidly with fiber diameters between 0.15 and 0.3 mm, peaks at 0.5 mm, and drops rapidly between 0.8 and 2 mm. The utility of the model was demonstrated by applying it to predict the lung retention of Cape crocidolite and Transvaal amosite workers on the basis of the measured length and diameter distributions of airborne fibers. The predicted lung distributions did, in fact, closely match those measured in lung samples from a Cape worker (Timbrell, 1984) and, as shown in Fig. 12.4, from a Transvaal worker (Timbrell, 1983). Thus, fibrosis is most closely related to the surface area of fibers with diameters between 0.15 and 2 mm and lengths greater than 2 mm. The work of King et al. (1946) showing that chrysotile with length of 2.5 mm produced interstitial fibrosis in rabbits following multiple intratracheal instillations is consistent with the retention shown in Fig. 12.3 and a critical fiber length of 2 mm. Churg et al. (2000) examined putative biological mechanisms involved in fibrogenesis and conclude that: (1) fiber length, biopersistance, and dose it remains uncertain whether alveolar macrophages are central to fibrosis or whether fibers penetrating tissue are the real effector agents; (2) short fibers, readily degraded fibers, and small numbers of any fibers are nonfibrogenic; and (3) the ability of macrophages to clear fibers is probably crucial to preventing fibrosis.

12.8.2

Critical Fiber Dimensions for Mesothelioma

A National Research Council study (NRC, 1984) summarized mortality data for mesothelioma and lung cancer in asbestos-exposed occupational cohorts. In 20 studies in which there was an excess in respiratory cancer and/or mesothelioma, the percentage of the excess that was mesothelioma varied from 0 to 100%, with a mean (SD) of 38  29%. A study with 0% was that of Meurman et al. (1974, 1979), who reported 44 observed lung cancers (versus 22 expected) in a population of 1045 workers exposed to anthophyllite in Finland. Anthophyllite is an amphibole with larger fiber diameters than other forms of asbestos. By contrast, in several occupational cohorts the mesotheliomas accounted for more that 70% of the total. These included: (1) the study of Newhouse et al. (1982) of 7474 British workers exposed to mixed asbestos, among whom there were eight mesotheliomas and only three more than the 140 expected lung cancers; (2) the study of Rossiter and Coles (1980) of 6076 British shipyard workers exposed to mixed asbestos, among whom there were 31 mesotheliomas and 13 fewer lung cancers than the expected number of 101; (3) the study of Jones et al. (1980) of 578 British female workers exposed to crocidolite, among whom there were 17 mesotheliomas and six lung cancers more than the six expected; and (4) the study of Newhouse et al. (1982) of 3708 British female workers exposed to mixed asbestos, among whom there were two mesotheliomas and five fewer lung cancers that the 11 expected. Timbrell (1983), Timbrell et al. (1987), and Harington (1981) have noted that animal inoculation experiments have been interpreted as suggesting a fairly high value of diameter, for example, 1.5 mm (Stanton et al., 1977), 1 mm (Pott et al., 1976; WHO, 1986) and 0.25 mm (Wagner and Pooley, 1986), below which a fibrous material, so long as it is durable in lung fluids, can produce mesothelioma. In their view, these diameter limits are too high for human fiber-induced mesothelioma. If fibers with diameters >0.5 mm produced mesothelioma, then Paakkila, where the dust clouds contained on the order of 50 fibers/mL (PCOM) and a high proportion of fibers in the 0.5–3 mm diameter range, should have produced many

CRITICAL FIBER PARAMETERS AFFECTING DISEASE PATHOGENESIS

423

FIGURE 12.4 Distributions of fiber lengths and diameters of amosite asbestos in the lungs of a Transvaal worker. The predicted distribution at the left is based on the lengths and diameters of the airborne fibers and on the lung retention as a function of length and diameter. This corresponds closely to the distribution in the right panel, which was measured in samples from the worker’s lung. (Source: Lippmann, 1988.)

mesotheliomas, as well as excesses in fibrosis and lung cancer. As noted earlier, an average of 38% of the excess lung cancer plus mesothelioma in working populations exposed to asbestos was expressed as mesothelima. Despite the very high exposures of the Paakkila population, few mesotheliomas were observed. Timbrell (1983)Timbrell’s (1983) examination of the size distributions and mesothelioma incidence at Paakkila and other asbestos mines worldwide led him to conclude that a good correlation was obtained if the threshold diameter was reduced to 0.1 mm. The mesotheliomas that Paakkila fiber has produced in animals were, most likely, caused by the use of

424

ASBESTOS AND OTHER MINERAL AND VITREOUS FIBERS

excessive doses, 10,000 times that observed in man. Paakkila asbestos contains only 1% of fibers with diameters below 0.1 mm, but with such a large dose this represents an enormous absolute number. Harington (1981) noted that the data for the northwest Cape in South Africa, where numerous mesotheliomas have been reported, and for the northeastern Transvaal, where mesotheliomas are rare, are consistent with a low fiber-diameter limit. In the northwest Cape, about 60% of the fibers have diameters <0.1 mm, whereas for the Transvaal, only about 1% have diameters <0.1 mm, comparable to Paakkila. Timbrell (1983) also noted that the length distributions at Paakkila and the northwest Cape point to a need to reduce the 10 mm length threshold in Stanton’s criteria. Paakkila had a high percentage of fibers longer than 10 mm, whereas the northwest Cape had virtually none. And yet, the northwest Cape has been the major source of mesothelioma. Attributing potential carcinogenicity to shorter fibers by lowering the length threshold brings the estimated levels of significant fibers into closer line with the observed mesothelioma rates. In reviewing the literature on mesothelioma induction in rats exposed by inhalation to fibrous aerosols, as summarized in Table 12.4, I concluded that, for mesothelioma, the relatively low tumor yields seemed to be highly dependent upon fiber type. Combining the data from various studies by fiber type, the percentage of mesotheliomas was 0.6% for Zimbabwe (Rhodesian) chrysotile, 2.5% for the various amphiboles as a group, and 4.7% for Quebec (Canadian) chrysotile. This difference, together with the fact that Zimbabwe chrysotile had 2–3 orders of magnitude less tremolite than Quebec chrysotile, provides support for the hypothesis that the mesotheliomas that have occurred among chrysotile miners and millers could be largely due to their exposures to tremolite fibers. The chrysotile fibers are insufficiently biopersistent because of dissolution during translocation from the sites of deposition to sites where more durable fibers can influence the transformation or progression to mesothelioma. Combining the findings of Timbrell with the results of rat inhalation experiments reported by Davis et al. (1986b) for studies with length-classified fibers leads to the conclusion that the critical fibers for mesothelioma induction have lengths between 5 and 10 mm. Davis et al. reported that intraperitoneal injections of short amosite (1.7% > 5 mm) produced only one mesothelioma among 24 rats (after 837 days), whereas UICC amosite (11% > 5 mm, 2.5% > 10 mm) produced 30 mesotheliomas among 32 rats, and long amosite (30% > 5 mm, 10% > 10 mm) produced 20 mesotheliomas among 21 rats. Thus, fibers shorter than 5 mm appear to be ineffective, and an appreciable fraction longer than 10 mm appears to be unnecessary. 12.8.3

Critical Fiber Dimension for Lung Cancer

Excess incidence of lung cancer has been reported for workers exposed to amphiboles (amosite, anthophyllite and crocidolite), to chrysotile, and to mixtures of these fibers (NRC, 1984), but these studies have been uninformative with respect to the fiber parameters affecting the incidence. The series of rat inhalation studies performed by Davis et al. (1978), which have also produced lung cancers, have provided the most relevant evidence on the importance of fiber length on carcinogenicity in the lung. The Wagner et al. (1974) study found that the yield of squamous cell carcinoma and adenocarcinoma was greatest with Rhodesian chrysotile, with decreasing yields for Canadian chrysotile, crocidolite, anthopyllite, and amosite, respectively. As shown in Table 12.3, Davis et al. (1978) reported two squamous cell carcinomas, six adenocarcinomas and seven adenomas in 40 rats exposed to 10 mg/m3 of respirable chrysotile. In 42 rats exposed to

CRITICAL FIBER PARAMETERS AFFECTING DISEASE PATHOGENESIS

TABLE 12.4

425

Mesotheliomas Produced by Asbestos in Chronic Rat Inhalation Studies

Type of Asbestos Zimbabwe chrysotile Wagner et al. (1974) Davis et al. (1978) Davis et al. (1980)

Source Tumors/Animals 10 mg/m3 10 mg/m3 2 mg/m3 10 mg/m3 (1 day/week)

UICC UICC UICC UICC

Overall Quebec chrysotile Wagner et al. (1974) Davis et al. (1988) Hesterberg et al. (1993)

0/44 0/40 1/42 0/43 1/169 (0.6%)

10 mg/m3 10 mg/m3 10 mg/m3 10 mg/m3

UICC Short Long NIEHS

Overall Davis–Wagner Subset

4/44 1/40 3/40 1/69 9/193 (4.7%) 8/124 (6.5%)

Amphiboles Wagner et al. (1974) Crocidolite Amosite Anthophyllite

10 mg/m3 10 mg/m3 10 mg/m3

UICC UICC UICC

2/44 0/46 2/46

Davis et al. (1978) Crocidolite Crocidolite Amosite

5 mg/m3 10 mg/m3 10 mg/m3

UICC UICC UICC

1/43 0/40 0/43

Davis et al. (1980) Amosite

50 mg/m3 (1 day/week)

UICC

0/44

Wagner et al. (1985) Crocidolite

10 mg/m3

UICC

1/24

Davis et al. (1985) Tremolite

10 mg/m3

Korea

2/39

10 mg/m3 10 mg/m3

Short Long

1/42 3/40

10 mg/m3

UICC

1/69

Davis et al. (1986) Amosite

McConnell (1994) Crocidolite Overall Davis–Wagner subset

13/520 (2.5%) 12/451 (2.7%)

426

ASBESTOS AND OTHER MINERAL AND VITREOUS FIBERS

2 mg/m3 of chrysotile, there were six adenomas, one adenocarcinoma and one squamous cell carcinoma. There were also adenomas in the groups exposed to amosite at 10 mg/m3 (two) and to crocidolite (one at 10 mg/m3, two at 5 mg/m3). Davis et al. (1978) attempted to examine the influence of fiber number concentration in relation to mass concentration in their inhalation studies. Their five exposure groups included three at the same respirable mass concentration of 10 mg/m3, one each with chrysotile, crocidolite, and amosite. Of these, the amosite produced the lowest number concentration of fibers >5 mm in length. This fiber count was then matched with crocidolite (5 mg/m3 respirable mass) and chrysotile (2 mg/m3 respirable mass). In attempting to explain the greater fibrogenic and carcinogenic responses in the chrysotile-exposed animals than the crocidolite- or amosite exposed groups, they suggested it might have resulted, at least in part, from the greater number of >20 mm long fibers in the chrysotile aerosol. The ratio of >20 to >5 mm long fibers in the chrysotile was 0.185 compared to 0.040 for crocidolite and 0.011 for amosite. The diameter distributions of all three types of asbestos were similar, with a median diameter of 0.4 mm. The importance of fiber length to toxicity was further demonstrated by Davis et al. (1986b) on the basis of inhalation studies with amosite aerosols that were both shorter and longer than the UICC amosite studied earlier with the same protocols. Both aerosols had median diameters between 0.3 and 0.4 mm. The short-fiber amosite (1.7% > 5 mm in length) produced no malignant cancers in 42 rats, whereas the long-fiber amosite (30% > 5 mm, 10% > 10 mm) produced three adenocarcinomas, four squamous carcinomas and one undifferentiated carcinoma in 40 rats. In terms of adenomas, the frequencies were 3/40, 2/43, 0/42, and 1/81 for the long, UICC, short, and control groups, respectively. Davis et al. (1985) also studied tremolite asbestos using the same protocols. Its length distribution was similar to those of the chrysotile in the 1978 study and the long amosite in the 1986 study (i.e., 28% > 5 mm, 7% > 10 mm), but its median diameter was lower, that is, 0.25 mm. It produced two adenomas, eight adenocarcinomas, and eight squamous carcinomas in 39 rats. Davis (1987) reported on a study comparing the carcinogenic effects of “long” and “short” chrysotile at 10 mg/m3. Unfortunately, the discrimination between “long” and “short” fibers was less successful than that achieved for amosite. PCOM fiber counts for the fibers >10 mm in length for the “long” and “short” chrysotile were 1930 and 330 f/mL, whereas for the amosite they were 1110 and 12 f/mL, respectively. Despite the much more rapid clearance of the chrysotile from the lungs, the tumor yields were higher. For the “long” fiber, there were 22 tumors for the chrysotile versus 13 for the amosite. For the “short” fiber, there were seven versus none. Davis (1987) concluded that fibers <5 mm in length may be innocuous, since the tumors produced by the “short” chrysotile are explicable by the presence of 330 f/mL longer than 10 mm. In another study, Wagner et al. (1985) exposed rats by inhalation to 10 mg/m3 of respirable dust composed of either UICC crocidolite (52.7% > 5 mm, 11.6% > 10 mm, median diameter 0.30 mm) or Oregon erionite (44% > 5 mm, 7.4% > 10 mm, median diameter of 0.22 mm). The UICC crocidolite produced one squamous carcinoma in 28 rats (but no mesotheliomas), whereas the erionite produced no carcinomas in 28 rats but did produce 27 mesotheliomas. In summary, Table 12.3 shows that 10 mg/m3 of short amosite (0.1% > 10 mm), UICC amosite (2.5% > 10 mm), UICC crocidolite (3% > 10 mm), and Oregon erionite (7.4% > 10 mm) failed to produce malignant lung cancers, whereas 10 mg/m3 of UICC chrysotile, long amosite and tremolite (all with 10% >10 mm) all produced malignant lung tumors. Although there was no clear-cut influence of fiber diameter on tumor yield, the results suggest that carcinogenesis incidence increases with both fiber length and diameter. Since Timbrell (1983) has shown that fiber retention in the lungs peaks between 0.3 and

CRITICAL FIBER PARAMETERS AFFECTING DISEASE PATHOGENESIS

427

0.8 mm diameter, it is likely that the thinner fibers, which are more readily translocated to the pleura and peritoneium, play relatively little role in lung carcinogenesis. Therefore, it appears that the risk of lung cancer is associated with long fibers, especially those with diameters between 0.3 and 0.8 mm, and that substantial numbers of fibers >10 mm in length are needed. In my own review of the literature on the chronic rat inhalation studies with amosite, brucite, chrysotile, crocidolite, erionite, and tremolite (Lippmann, 1994), I found that, for lung cancer, the percentage of lung tumors (y) could be described by a relation of the form y ¼ a þ bf þ cf2, where f is the number concentration of fibers, and a, b, and c are fitted constants. The correlation coefficients for the fitted curves were 0.76 for >5 mm f/mL, 0.84 for >10 mm f/mL, and 0.85 for >20 mm f/mL, and seemed to be independent of fiber type. This supports the hypothesis that the critical length for lung cancer induction is in the 10–20 mm range. In terms of the critical sites within the lungs for lung cancer induction, it has been shown that brief inhalation exposures to chrysotile fiber produces highly concentrated fiber deposits on bifurcations of alveolar ducts, and that many of these fibers are phagocytosed by the underlying type II epithelial cells within a few hours. Churg (1994) has shown that both chrysotile and amphibole fibers retained in the lungs of former miners and millers do not clear much with the years since last exposure. Thus, lung tumors may be caused by that small fraction of the inhaled long fibers that are retained in the interstitium below small airway bifurcations, where clearance processes are ineffective. One reason that short fibers may be less damaging could be the fact that they can be fully ingested by macrophages (Beck et al., 1971), and can therefore be more rapidly cleared from the lung. The fibrogenic response to long fibers could result from the release of tissuedigesting enzymes from alveolar macrophages whose membranes are pierced by the fibers they are attempting to engulf (Allison, 1977). The fibers may also cause direct physical injury to the alveolar membrane. A positive association of asbestosis with lung tumors was demonstrated by Wagner et al. (1974). The induction of fibrosis impairs clearance of deposited fibers, increasing the persistence of fibers in the lung. The preceding implies that short fibers will have a low order of toxicity within the lung, comparable to that of nonfibrous silicate minerals. Within this concept, the critical fiber length would most likely be on the order of the diameter of an alveolar macrophage, that is, about 10–15 mm. This line of reasoning leads to the same conclusion reached on the basis of the incidence of lung cancer in rats exposed to fibrous aerosols, that is, that the hazard is related to the number of fibers longer than 10 mm deposited and retained in the lungs. The Timbrell (1983) model predicts alveolar retention of deposited fibers approaching 100% for 10 mm long fibers in the 0.3–0.8 mm diameter range. Airborne fibers longer than 100 mm may be much less hazardous than those in the 10–100 mm range because they do not penetrate deeply into the airways, as interception increases with fiber length. 12.8.4

Summary of Critical Fiber Parameters

The various hazards associated with the inhalation of mineral fibers, that is, asbestosis, mesothelioma, and lung cancer, are all associated with fibers with lengths that exceed critical values. However, it now appears that the critical length is different for each disease, that is, 2 mm for asbestosis, 5 mm for mesothelioma, and 10 mm for lung cancer. There are also different critical values of fiber diameter for the different diseases. For asbestosis and lung cancer, which are related to fibers retained in the lungs, only fibers with diameters >0.15 mm need to be considered. On the contrary, for mesothelioma, which is initiated by fibers that

428

ASBESTOS AND OTHER MINERAL AND VITREOUS FIBERS

migrate from the lungs to the pleura and peritoneum, the hazard has been related to fibers with smaller diameters. A study, by Dufresne et al. (1996) of the fibers in lungs of Quebec miners and millers with and without asbestosis is supportive of the critical influence of long fibers on fibrosis and cancer incidence. They found that mean concentrations were higher in cases than in the controls for chrysotile fibers 5–10 mm long in patients with asbestosis with or without lung cancer; for tremolite fibers 5–10 mm long in all patients; for crocidolite, talc, or anthophyllite fibers 5–10 mm long in patients with mesothelioma; for chrysotile and tremolite fibers 10 mm long in patients with asbestosis; and crocidolite, talc, or anthophyllite fibers 10 mm long in patients with mesothelioma. Cumulative smoking index (pack-years) was higher in the group with asbestosis and lung cancer but was not statistically different from the two other disease groups. Although all durable fibers of sufficient length can produce fibrosis and cancer, as documented in various animal studies, it appears that factors other than fiber size can influence the extent of the response. For example, inhaled erionite appears to be much more potent for mesothelioma in both humans and animals because of its greater ability to penetrate the pleural surface. On the contrary, the animal and human data appear to differ on the ability of inhaled chrysotile to induce mesothelioma. Animal data indicate that chrysotile produces as much or more mesothelioma than the amphiboles, whereas human data more often implicate amphiboles, even when the predominant exposures are to chrysotile. Examination of the fiber content of the lungs of asbestos workers and animals exposed by inhalation shows that chrysotile is cleared much more rapidly than the amphiboles. It breaks down within the lungs both by disaggregation into fibrils and by dissolution. The differences between the responses in animals and humans may be in relative persistence, that is, time of persistence of the long fibers in the lung relative to the time interval between exposure and the expression of the disease. In other words, the long fibers may be retained in the lung for a longer fraction of the lifespan in the rat. Although all durable fibers in the right size range can cause the asbestos-related diseases, they may have different potencies and need different concentration limits. The remainder of this discussion addresses the indices of exposure but not the concentration limits for the fibers that fall within the indices. The concentration limits warrant separate and further discussion. 12.8.5 Implication of Critical Fiber Parameters to Health Relevant Indices of Exposure Although the current occupational exposure index, based on PCOM for fibers with an aspect ratio >3 and a length >5 mm, was a reasonable choice, when it was made, for occupational exposures involving specific known fiber types, it is now apparent that it cannot provide a scientifically adequate index for any of the differing hazards resulting from exposures to chrysotile, the various amphibole fibers, other mineral fibers, and the various SVFs. Its most important inadequacies for occupational exposure evaluations in mining, milling, and manufacturing industry include: (1) thin fibers of health relevance, that is, those with widths <0.25 mm, cannot be seen by PCOM: and (2) the PCOM protocol identifies the presence, but not the composition or the distribution of lengths and widths for the >5 mm long fibers that are visualized. It has further limitations for occupational exposures in demolition, building renovation, asbestos remediation projects, emergency response to steam pipe explosions, building collapses, and so on, where most of the dust collected on the sampling filter is

EXPOSURE–RESPONSE RELATIONSHIPS FOR ASBESTOS-RELATED LUNG CANCER

429

nonfibrous. The background dust makes the counting of fibers difficult, if not impossible, and the fibers that are seen have a variety of compositions and toxicities. Some of these limitations can be overcome by using analytical transmission electron microscopy and X-ray Diffraction Analysis for fiber counting and analysis. These measurement methods enable visualization of fiber widths down to 0.1 mm, and composition analysis of each individual fiber in the field of view. Unfortunately, such analyses are considerably more expensive than PCOM for routine analyses, and the standard laboratory protocols have not utilized the available capabilities to generate fiber size distribution data that recognize the rapidly increasing hazard with fiber length as length increases above 5 mm. There is an additional limitation of the TEM methodology endorsed by EPA when it is applied to most exposure assessments for evaluating carcinogenic risks to the general public, since its detection limit for asbestos fibers >5 mm in length is higher than the concentration considered acceptable by EPA. This problem was discussed by HEI-AR (1991) and is further summarized in the next section. The TEM methodology has traditionally been used in EPAendorsed protocols to enumerate the count of all fibers greater than 0.5 mm in length, which ends up in insufficient filter area scanning to get a statistically valid sample of the healthrelevant fibers longer than 5, 10, or 20 mm. EPA has been considering the adoption of new dimensional criteria for hazardous fibers based on a proposal by Berman and Crump (2001), and sponsored a Peer Consultation Workshop (ERG, 2003b) on the topic. The Berman and Crump index, based in large measure on modeling of the results of the chronic inhalation studies in rats for which detailed fiber length and width data were available (Berman et al., 1995). Zero risk is assigned to fibers <5 mm in length, some small risk is assigned to fibers between 5 and 10 mm in length, and the bulk of the risk is assigned to longer fibers. The Workshop Panel agreed that there is considerably greater risk for lung cancer for fibers longer than 10 mm, but was uncertain as to an exact cut size for length and the magnitudes of the relative potencies in each size range. They also agreed that the available data suggest that the risk for fibers<5 mm in length is verylow and could be zero (ERG, 2003b). The Report of an Expert Panel on the Health Effects of Asbestos and Synthetic Vitreous Fibers to the Agency for Toxic Substances and Disease Registry came to similar conclusions, agreeing that there is a strong weight of evidence that asbestos and SVFs shorter than 5 mm are unlikely to cause cancer in humans (ERG, 2003a). As of this writing, the EPA has not endorsed new dimensional criteria for hazardous airborne fibers. Therefore, for the foreseeable future, it would be desirable for risk assessors to use analytical protocols that permit the measurement of the lengths and diameters of all fibers longer than 5 mm, so that the results can be tabulated according to the conventional criteria as well as others that will be adopted in future years. In the discussion of the human experience and risk assessments that follows, the exposure concentrations are based upon the conventional 5 mm fiber-length criterion. 12.9 EXPOSURE–RESPONSE RELATIONSHIPS FOR ASBESTOS-RELATED LUNG CANCER AND MESOTHELIOMA: HUMAN EXPERIENCE 12.9.1

Epidemiologic Models for Cancer Risks Associated with Asbestos

As noted by the HEI-AR Literature Review Committee (HEI-AR, 1991), the majority of published dose-specific estimates of the cancer risk caused by asbestos exposure have been based on models that assume the following:

430

ASBESTOS AND OTHER MINERAL AND VITREOUS FIBERS

1. The increase in relative risk for lung cancer is proportional to cumulative asbestos exposure, and the effects of asbestos and cigarette smoking multiply each other. 2. The increase in mesothelioma incidence caused by each brief period of exposure is proportional to the amount of that incremental exposure (exposure level  duration) and to a power of time since it occurred, independent of age or smoking. The power of time is approximately 2 or 3. Similar risk assessment models were used in various government-sponsored reports, including those of the Environmental Protection Agency, the National Research Council, and the Consumer Product Safety Commission in the United States and comparable bodies in Britain and Canada. The validity of these models remains to be confirmed. Although dosespecific risk estimates differ substantially between cohorts, the results of analyses by different groups and authors for each cohort are similar. For lung cancer, the model implies that the effects of cigarette smoking and asbestos on lung cancer risk are multiplicative. This would mean that an exposure that doubles the rate among smokers (from a lifetime risk of about 0.1–0.2) will also double the rate among nonsmokers (from about 0.005–0.01). However, this may not be true. Lung cancer is so rare among nonsmokers, even after quite heavy asbestos exposure, that their risk cannot be estimated precisely. In fitting models, a major uncertainty is the choice of the lung cancer rate in unexposed workers. Local or national rates will not be appropriate if the workers’ smoking habits are (or were, in the past) atypical. An examination of the cancer experience of 7279 Quebec chrysotile miners and millers in the 1891–1920 birth cohort by Liddell and Armstrong (2002) concluded that the risks of chrysotile exposure and smoking were not multiplicative, but rather independent, casting doubt on the validity of the previously adopted predictive models.

12.9.2 Review of Epidemiological Data on Human Asbestos Exposure-Cancer Response The risk estimates (KL) derived from different studies vary by two orders of magnitude. This in part reflects statistical variation, but Fig. 12.5, in which 95% confidence limits are shown, indicates that there are highly significant differences between the different estimates. (Fig. 12.5 is based on the risk estimates derived by the EPA, which are similar to those calculated in other reviews.) Possible reasons for this extreme heterogeneity include errors in the model, differences in hazard associated with different fiber types and dimensions, inaccuracies in exposure estimates, and the use of inappropriate lung cancer rates in calculating standardized mortality ratios (SMRs). The EPA review attempted a formal analysis of this variation between different studies (Nicholson, 1986). It presented explicit confidence limits for each estimate of KL, taking account of statistical variation and assuming twofold (and in some cases greater) uncertainty in exposure estimates. For some cohorts, adjustments were made for suspected biases, particularly the use of inappropriate lung cancer rates. Excluding the significantly lower risks per unit exposure observed for chrysotile mining and milling, this analysis gave a geometric mean for KL of 0.01. The only study that gave a significantly higher risk estimate than this central estimate was the Ontario asbestos cement products factor of Finkelstein (1983), but it had quite questionable exposure estimates and an inconsistent exposure–response relationship.

EXPOSURE–RESPONSE RELATIONSHIPS FOR ASBESTOS-RELATED LUNG CANCER

431

FIGURE 12.5 Mean and 95% confidence limits of KL, the fractional increase in lung cancer in fiber/ mL-year of exposure in 14 asbestos-exposed cohorts. (Source: Nicholson, 1986.)

One of the studies that produced a higher than average risk coefficient for chrysotile was that of Dement et al. (1983) (see Fig. 12.5). A follow-up study (Dement et al., 1994) confirmed the relatively high rate of lung cancer in this population, and attributed it to the high proportions of long fibers. A model based on the available data on this cohort was presented by Stayner et al. (1997). They used a model designed to evaluate evidence of a threshold response. Lifetime risks of lung cancer and asbestosis were estimated with an actuarial approach that accounted for competing causes of death. They found a highly significant exposure–response relation for both lung cancer and asbestosis. The exposure– response relation for lung cancer seemed to be linear on a multiplicative scale, which is consistent with previous analyses of lung cancer and exposure to asbestos. By contrast, the exposure–response relation for asbestosis in this analysis seemed to be nonlinear on a multiplicative scale. There was no significant evidence for a threshold in models of either the lung cancer or asbestosis. The excess lifetime risk for white men exposed for 45 years at the current OSHA standard of 0.1 fiber/mL was predicted to be about 5/1000 for lung cancer, and 2/1000 for asbestosis. One unsatisfactory aspect of the published literature on mesothelioma is the lack of adequately analyzed mortality data. The death rate rises sharply with time since exposure, yet only a few data sets have been analyzed by time since first exposure, and only four of these reports also provided estimates of average exposure level. Furthermore, there are serious weaknesses in all four studies, particularly for assessing the effects of chrysotile. The exposure data and results for the cement factory workers’ study reported by Finkelstein (1983) are of doubtful reliability for quantitative risk assessment, and there are no contemporary exposure data for the insulation workers (Selikoff et al., 1979) or the amosite textile workers (Seidman et al., 1979). Moreover, none of these four cohorts was exposed only to chrysotile.

432

ASBESTOS AND OTHER MINERAL AND VITREOUS FIBERS

In the absence of any satisfactory basis for direct estimation of the dose-specific mesothelioma risk caused by any specific type of asbestos, particularly chrysotile, HEIAR (1991) evaluated and modified previous predictions for mesothelioma by comparing observed and predicted ratios of mesothelioma to excess lung cancer in different cohorts. The predictive model for mesothelioma used by HEI-AR was proposed to explain the observation that mesothelioma incidence is independent of age and approximately proportional to the third power of time since first exposure. The model has been formally fitted in the only cohort for whom individual exposure data were available (Peto et al., 1985). This limited analysis, based on only 10 cases, suggested that brief exposure causes less mesothelioma risk than that predicted. The eventual lung cancer risk is assumed to be independent of age at exposure, but the predicted mesothelioma risk is much greater when exposure begins at an early age. These models therefore predict that the mesothelioma risk exceeds the lung cancer risk, even among smokers, for childhood exposure, whereas exposure in middle age causes a relatively trivial mesothelioma risk. Among nonsmokers, the lung cancer risk is much less than the mesothelioma risk irrespective of age at exposure. A study of vermiculate miners exposed to tremolite fibers in Libby, Montana, but not to other forms of asbestos, has provided an opportunity to examine the role of amphibole contamination of commercially important asbestos-containing products (McDonald et al., 2004). The miners had significantly elevated risks for lung cancer, nonmalignant respiratory disease, and mesothelioma. They concluded that amphibole fibers, and tremolite in particular, are likely to be disproportionately responsible for cancer mortality in persons exposed to commercial products containing asbestos. 12.9.3

Mesotheliomas Associated with Asbestos Exposure

12.9.3.1 Peritoneal Mesothelioma The clearest difference between the effects of different fiber types is in the proportion of mesotheliomas that are present in the peritoneum. Almost all cases among chrysotile workers (usually with some exposure to crocidolite and/or tremolite) or among crocidilite miners are pleural, whereas workers with some amosite exposure have suffered similar and sometimes higher risks of peritoneal than pleural mesothelioma (Levin et al., 1998). The only exception appears to be female gas mask workers exposed mainly to crocidolite, for whom several mesotheliomas were peritoneal. The possibility that some amosite exposure occurred in these workers was, however, not discussed. The inference that most peritoneal mesotheliomas are caused by amosite exposure is generally accepted (HEI-AR, 1991). 12.9.3.2 Pleural Mesothelioma Direct comparison of workers employed for similar duration to different forms of asbestos (e.g., in mining or gas mask manufacture) indicates a much higher mesothelioma risk for amphiboles than for chrysotile. Chrysotile friction products workers in Britain suffered no detectable increase in lung cancer, and 11 of the 13 mesotheliomas in this cohort occurred in the subgroup of workers with known exposure to crocidolite (Berry and Newhouse, 1983; Newhouse and Sullivan, 1989). Chrysotile textile workers in Britain suffered a high risk of mesothelioma in contrast to those in South Carolina, although there was a substantial lung cancer risk in both cohorts. The only marked difference between these two textile plants was the use of some crocidolite (less than 5% of the fiber processed) in the British plant.

EXPOSURE–RESPONSE RELATIONSHIPS FOR ASBESTOS-RELATED LUNG CANCER

433

12.9.3.3 Mesothelioma and the “Amphibole Hypothesis” As discussed previously, there have been marked differences between cohorts in the ratio of excess lung cancer to mesothelioma. Peritoneal mesotheliomas can usually be attributed to amosite exposure, but even when only pleural tumors are considered, the cancer ratio varies remarkably. English shipyard workers with mixed exposure, including a substantial amount of crocidolite, suffered a high mesothelioma risk but no excess of lung cancer (Rossiter and Coles, 1980), whereas among workers at a South Carolina chrysotile textile plant there was a marked excess of lung cancer and a low incidence of pleural mesothelioma (McDonald et al., 1984; Dement et al., 1982, 1994). These data have been almost universally accepted as indicating that amphiboles, particularly crocidolite, cause a disproportionate mesothelioma risk. Based on a pooling various cohorts, Doll and Peto (1985) and Nicholson (1986) concluded that among men the ratio of excess lung cancer to pleural mesothelioma is about three times greater for chrysotile than crocidolite, varying from at least four for chrysotile to between one and two for crocidolite, with substantially lower ratios for women. However, such pooling of generally inconsistent data has dubious validity. In particular, it conceals the most extreme inconsistencies, most notably the marked excess of mesothelioma in the absence of any detectable excess of lung cancer observed among shipyard workers by Rossiter and Coles (1980) and in the subgroup of friction product workers with crocidolite exposure studied by Berry and Newhouse (1983). Other investigators believe that virtually all mesotheliomas are due to amphibole exposure and that inhaled chrysotile fibers pose a negligible mesothelioma risk. The only strong evidence against the inference that mesothelioma is seldom, if ever caused by chrysotile fibers alone is the observation of substantial numbers of cases among Quebec chrysotile miners and millers. It has, however, been suggested that these mesotheliomas are related to the presence of fibrous tremolite in this material (Mossman et al., 1990a, 1990b). Tremolite constituted less than 1% of the fiber extracted but more than half of the long (>5 mm) fibers found in the lung tissue of the workers, apparently because chrysotile is cleared much more rapidly (Sebastien et al., 1989). Similarly, high levels of crocidolite were found in lung tissue from British textile workers who were exposed mainly to chrysotile but suffered a high incidence of mesothelioma (Wagner et al., 1982). The early evidence that chrysotile fibers rarely cause pleural mesothelioma was consistent, but not conclusive. There have been only two cohorts of heavily exposed asbestos workers who worked only with chrysotile (in both cases exposed to chrysotile from Quebec that is often contaminated with tremolite). There was an initial absence of pleural mesothelioma in the South Carolina plant in spite of the substantial risk of lung cancer (59 observed, 29.6 expected; Dement et al., 1982; McDonald et al., 1984). The Quebec chrysotile miners and millers suffered 230 lung cancers compared with 184.0 expected and 10 mesotheliomas, and lung burden studies show no marked differences between these cohorts in the type, size or amount of either chrysotile or tremolite fibers (Sebastien et al., 1989). In follow-up study of the South Carolina cohort by Dement et al. (1994) of 15 years of additional experience, there were two deaths attributable to mesothelioma, and the number of lung cancers had risen to 126, and was expressed as an increase in relative risk of 2.3% for each year of cumulative chrysotile exposure. Begin et al. (1992) brought the experience in Quebec up to 1990, and concluded that the incidence of pleural mesothelioma in chrysotile miners and millers, while less than that for crocodolite workers, was well above the North American male rate.

434

ASBESTOS AND OTHER MINERAL AND VITREOUS FIBERS

In a further examination of the Quebec asbestos cohort of 11,000 chrysotile workers by McDonald and McDonald (1997), of whom 80% had already died, they addressed the amphibole hypothesis by analyzing the deaths among the 4000 miners employed by the largest company in the Thetford region. The number of cancer deaths, by type, were mesothelioma (21), lung (262), larynx (15), stomach (99), and colon and rectum (76). Risks, in relation to case referants, were analyzed by logistic regression separately for those working in the five chrysotile mines located centrally and for the 10 mines located more peripherally, on the basis that tremolite concentrations were four times higher in the central region. Odds ratios were significantly and substantially elevated for workers at the centrally located chrysotile mines for mesothelioma and lung cancer, but not for gastric, intestinal or laryngeal cancers; while for the workers at the more peripherally located chrysotile mines, there was little or no elevation in odds ratio for any of the cancer groups. Dust exposures of the two groups were similar, and lung tissue analyses showed that the concentration of tremolite fibers was much higher in the lungs from the central area in comparison to those from workers at the peripheral mines (McDonald et al., 1997). In an earlier study of Thetford mine workers by Gibbs (1979), pleural calcifications were also much more common among miners who had worked in the centrally located mines than those who worked in the peripherally located mines, suggesting that tremolite accounted for much of both pleural calcification and cancers of the lung and pleura. Furthermore, the studies cited earlier by Dodson et al. (1990) and Boutin et al. (1996) suggest that some of this difference in potency was due to the greater retention of longer amphibole fibers in the lung and pleural lymph nodes than is the case for chrysotile. Additionally, the experimental animal inhalation studies, summarized in Table 12.4 and discussed earlier, support the critical role of amphibole fibers in the causation of mesothelioma. Thus, despite the publication of contrary views (Frank et al., 1997; Smith and Wright, 1996; Stayner et al., 1996; Dodson et al., 2003), the hypothesis that mesothelioma is largely, if not exclusively caused by amphibole fibers remains consistent with the bulk of recently published evidence in humans and animals. No other epidemiological studies have addressed the mesothelioma-inducing potency of chrysotile directly. The substantial incidence of pleural mesothelioma in various cohorts exposed to both chrysotile contaminated by tremolite and by mixtures of chrysotile and other amphiboles is consistent with the hypothesis that these tumors were caused by amphibole exposure. However, it constitutes weak evidence that chrysotile per se does not cause mesothelioma. A prudent conclusion is that a very substantial proportion of the mesotheliomas in such cohorts are caused by amphibole exposure. 12.9.4

Workers Exposed To SVFs

12.9.4.1 Lung Cancer The epidemiological literature on SVFs has focused almost exclusively on lung cancer, and there have been no reports of mesothelioma in occupational groups without coexposures to asbestos. With respect to lung cancer, many of the studies of SVF production workers have reported excess lung cancer among some cohorts, primarily those heavily exposed to slag wool and rock wool in earlier years, when exposure levels were largely uncontrolled (Enterline et al., 1983, 1987; Shannon et al., 1984, 1987; Saracci et al., 1984; Simonato et al., 1986). In the large multinational study in Europe involving approximately 22,000 SVF production workers at 13 plants in seven countries, the overall lung cancer SMR was

EXPOSURE–RESPONSE RELATIONSHIPS FOR ASBESTOS-RELATED LUNG CANCER

435

125, and the SMR for the subcohort with more than 30 years was 170. Adjustment for regional variations in mortality substantially reduced the excess lung cancer incidence for those workers exposed only to glass wool but not for those exposed to rock wool/slag wool. Within this group, most of the excess occurred in the group with 20–29 years since first exposure (SMR of 270). The SMR was 244 for those with 30 or more years, who worked before dust controls were installed. On the basis of measurements of airborne fiber concentrations made in the period 1977– 1980 in the same European plants, Cherrie et al. (1986) reported that the average combined occupational group concentrations in the rock and glass wool plants were generally low (<0.1 fibers/mL). In the glass continuous-filament factories the airborne fiber concentrations were very low (<0.01 fibers/mL). The average plant median for fiber length ranged from 10 to 20 mm, and the corresponding median diameters ranged from 0.7 to 2 mm. In general the glass wool fibers were thinner than the rock wool fibers. Higher levels (between 0.1 and 1.0 fibers/ mL) were found in some insulation wool production, secondary production, and user industries. The highest levels (>1.0 fibers/mL) occurred in very fine glass-fiber production and in other specialty insulation wool usage. In a follow-up study of the sub-cohort of rock and slag wool workers in Scandanavia and Germany, for whom smoking and other occupational exposure data were available, Kjærheim et al. (2002) reported that, for a smoking-adjusted model with a 15 years lag, the lung cancer odds ratio for the second, third, and fourth cumulative exposure quartiles were 1.3, 1.0, and 0.7, suggesting that the fiber exposures were not causal. For the 17 U.S. plants producing and using ordinary fibrous glass insulation products that were studied by Enterline et al. (1983, 1987), Esmen (1984) reported that 35% of the airborne fibers were >5 mm in length and that only 3.9% were less than 1.0 mm in diameter. The average exposure concentrations, as determined by phase-contrast optical microscopy, were between 0.01 and 0.05 fibers/mL in 13 plants handling ordinary glass fibers. For a slag wool plant the average was 0.07 fibers/mL, while for a rock wool plant and a glass microfiber plant, the averages were 0.25 fibers/mL. Enterline et al. (1987) reported the 1946–1982 mortality experience of 16,661 SVF workers employed 6 months or more during 1940–1963 at one or more of 17 U.S. manufacturing plants. Using local death rates to estimate expected deaths, there was a statistically significant excess in all malignant neoplasms and in lung cancer 20 or more years after first employment. For respiratory cancer the excess was greatest for mineral wool workers and workers ever exposed in the production of small diameter fibers. These two groups of workers are believed to have had mean exposures to respirable fibers of around 0.3 fibers/mL. For glass wool workers and glass filament workers, SMRs for respiratory cancer were much lower. For these workers, exposures were estimated to be about 1/10 the level for mineral wool and small diameter fiber workers. There were few positive relationships between respiratory cancer SMRs and duration of exposure, time since first exposure, or measures of fiber exposure. A smoking survey showed SVF workers to have cigarettesmoking habits similar to all U.S. white males. In a case-referent study, which controlled for smoking, there was a statistically significant relationship between fiber exposure and respiratory cancer for mineral wool workers but not for fibrous glass workers. Marsh et al. (2001a) introduced and summarized the results of the 1986–1992 update on this large U.S. cohort. It involved a new historical exposure reconstruction for glass fibers, arsenic, asbestos, asphalt, epoxy, formaldehyde, PAHs, phenolics, silica, styrene, and urea and a nested case–control study of the 631 respiratory cancer cases, including their smoking histories (Stone et al., 2001; Marsh et al., 2001b). The only outcome with a statistically

436

ASBESTOS AND OTHER MINERAL AND VITREOUS FIBERS

significant 6% (p ¼ 0.05) excess risk was respiratory cancer. However, the duration of fiber and other exposures, the cumulative exposures, and the time since first exposure were not associated with the cancer risk. The smoking habit data (Buchanich et al., 2001) indicated more smoking in the exposure cohort than in the referent population, suggesting that at least some of the cancer excess was due to smoking. In their examination of mesothelioma risk, Marsh et al. (2001c) reported one case in the exposed cohort, while the expected for the referent group was 2.19. For female workers, there was no respiratory cancer excess for the period 1946–1992 (Stone et al., 2004). Shannon et al. (1987) conducted a historical prospective mortality study at an insulating wool plant in Ontario, Canada. It covered 2557 men who had worked for at least 90 days and were employed between 1955 and 1977, with follow-up to the end of 1984. There were 157 deaths in the 97% of men traced. Mortality was compared by the person-years method with that of the Ontario population. Overall mortality was below that expected (SMR ¼ 84). Cancer deaths were slightly raised, owing entirely to an excess in lung cancer. The 21 deaths from this cause give a significantly high SMR of 176. All but two of these cases occurred among “plant-only” employees. However, the interpretation of these data remains difficult because the SMRs by length of exposure and time since first worked were not consistent with a causal relationship. Another very large population studied by Engholm et al. (1987) for the incidence of respiratory cancer in 135,000 male Swedish construction workers in relation to exposure to SVFs. The men were all examined at regular health check-ups in 1971–1974, and the cohort was followed for mortality through 1983 and for cancer incidence through 1982 by linkage to various national registries. A case–control study within the cohort was carried out on 518 cases diagnosed as having respiratory cancer. The subjects were classified into categories based on self-reported exposure and on estimates of average intensity of exposure in occupations concerned. Smoking habits and density of population were included as potential confounders. Overall, there was an excess of mortality from industrial accidents and an excess incidence of mesothelioma, but in other respects the mortality and the cancer incidence in this population compared favorably with those of the general Swedish population. For lung cancer the overall incidence was below that expected, but there was a risk related to high asbestos exposure. The risk fell close to unity for SVFs when both exposures were fitted simultaneously. The human experience, based on long-term follow-up on SVF workers in the USA, Canada, and Europe, is encouraging. Many of these workers were exposed to very high concentrations of fibers in the 1940s and 1050s. As summarized by Doll (1987), the evidence of excess lung cancer among these workers appears confined to those subcohorts exposed to slag or rock wool, or to those exposed to small diameter biopersistent glass fibers as well as to conventional fibrous glass. When Doll excluded short-term and office workers and compared the numbers of deaths with those that would have been expected had the workers experienced national mortality rates (or provincial rates in the Canadian series), he found that the mortality from lung cancer (SMR ¼ 121) was raised but that the mortality from other cancers (SMR ¼ 101), other respiratory disease (SMR ¼ 103), and all other causes of death (SMR ¼ 100) was close to that expected. Division of the workers by type of product and time since first exposure showed that the mortality from lung cancer was highest in the rock or slag wool sector of the industry (SMR ¼ 128), intermediate in the glass wool sector (SMR ¼ 110), and lowest in the continuous-filament glass sector (SMR ¼ 93) and that within the first two groups mortality rose with time since first exposure to a maximum after 30 or more years (rock or slag wool,

EXPOSURE–RESPONSE RELATIONSHIPS FOR ASBESTOS-RELATED LUNG CANCER

437

SMR ¼ 141; glass wool, SMR ¼ 119). Within the U.S. glass wool industry, the mortality from lung cancer was higher in those men who had ever been exposed to small diameter fibers (SMR ¼ 124) than in others (SMR ¼ 108). No relationship was observed with duration of employment or with cumulative fiber dose. In a case–control study, however, a weak relationship with cumulative fiber dose was observed in the rock and slag wool sectors of the industry after differences in smoking habits had been taken into account. No evidence was obtained of a risk of mesothelioma or any other type of cancer. Doll concluded that an occupational hazard of lung cancer has been demonstrated in the rock and slag wool section of the industry and possibly in the glass wool section. Uncertainty about fiber counts in the early years of the industry and about the extent to which other carcinogens were present in the atmosphere of the plants precludes an estimate of the quantitative effects of exposure to current fiber levels, except that it is unlikely to be measurable. Wong and Musselman (1994) described results from an epidemiologic study of SVF workers in the U.S. for workers at nine plants that made or used slag wool. These included four plants previously studied and five additional plants. This was a nested case–control study based on 55 lung cancers. They analyzed lung cancer risk in relation to cumulative fiber exposure (concentration and duration) and smoking history, and controlled for other coexposures such as asbestos contamination. No increased lung cancer risk with exposure to slag wool fibers was found. This paper provided guidelines to estimate the magnitude of potential confounding effects of coexposures such as smoking. 12.9.4.2 Respiratory Morbidity Quantitative data on exposure–response relationships in occupational groups exposed to SVFs for effects other than lung cancer are sparse. The best-documented effects are for workers exposed to refractory ceramic fibers. LeMasters et al. (1998) reported on an industry-wide RCF cohort that was characterized as either production or nonproduction activities and duration of production employment. Both male and female production workers had significantly more respiratory symptoms. For male production workers, there was a significant decline, over 10 years, in FVC for both smokers and nonsmokers, while only the male smokers had a significantly greater decline in FEV1. Female nonsmokers had a greater decline in FVC than their male counterparts. In an earlier report on this population, LeMasters and Lockey (1994) reported a correlation between pleural changes and RCF exposures. In an analysis of the mortality experience of this cohort through, 2000, LeMasters et al. (2003) reported no excess for all deaths, all cancers, mesothelialomas, or respiratory diseases, but there was an excess for urinary organ cancer. Lockey et al. (2005) reported that the RCF exposures for this cohort were significantly associated with pleural plaques detected radiographic chest exams, but there was no significant increase in interstitial changes. By contrast, studies of a comparable industry-wide cohort in Europe by Trethowan et al. (1995) found no excess in illness or chest X-ray abnormalities related to fiber exposures. Cowie et al. (2001) found that pleural changes in this RCF cohort were related to age and exposure to asbestos, consistent with time since first RCF exposure. Among men, FEV1 and FVC were inversely related to fiber exposure, but only in smokers, and chronic bronchitis showed some association with recent fiber exposure. For symptoms, lung function changes related to exposures to other SVFs, the epidemiologic evidence had been largely negative (Utidjian and deTreville, 1970; Hook et al., 1970; Hughes et al., 1993; Malmberg et al., 1984). In terms of X-ray abnormalities, the only positive findings reported by Hughes et al. (1993) were for a population cohort exposed to very thin glass fibers.

438

ASBESTOS AND OTHER MINERAL AND VITREOUS FIBERS

12.10 RISK ASSESSMENT ISSUES 12.10.1

Measures of Asbestos Exposure Used in Past Epidemiologic Studies

The calculation of dose-specific risk depends as much on measurement of exposure as on estimation of excess risk, yet too little attention has been paid to the quality of these vital data. Substantial excess risks have only been observed for cohorts for which individual exposures can be estimated based on reasonably extensive historical dust measurements. In each case, however, there was little or no measurement of exposure in some of the dustiest areas, and the conversion of particle to fiber counts was based on inconsistent measurements at relatively low levels. In one other study, of a United Kingdom friction products factory, extensive and probably reliable individual exposure estimates were calculated. The study does not provide a very useful dose-specific lung cancer risk estimate, however, as exposures were so low that the risk estimate (which was virtually zero) has very wide confidence limits. 12.10.2

Dose–Response for Mesothelioma

The form of the mesothelioma dose response for asbestos is not known. In a review of occupational experience by Hodgson and Darnton (2000), they concluded that for comparable high-level occupational fiber exposures to chrysotile, amosite, and crocidolite the mesothelioma risks were 1:100:500, respectively. Rodelsperger et al. (1999), in a mesothelioma case–control study, examined fiber burdens in the lungs of 66 cases and controls. They did not find a significant odds ratio for chrysotile, but reported a significant exposure– response relationship for amphibole fibers longer than 5 mm. For eight environmental exposures that involved exposures judged to be high (but without extensive exposure measurements), Bourdes et al. (2000) concluded that the relative mesothelioma risk of household exposures was 8.1 and that for neighborhood risk was 7.0. The prediction that risks at high levels of exposure can be linearly extrapolated to very low concentrations cannot be tested epidemiologically. Exposure levels have never been recorded accurately, and the predicted risks at low levels are far too low to be observable. The opposite belief, that the mesothelioma risk is anomalously high following very low exposure, is not supported by observation. In particular, the mesothelioma risk following short exposure to chrysotile may, if anything, be less than that predicted. Peto et al. (1985) studied approximately 18,000 men with no previous asbestos exposure employed in 1933 or later in a chrysotile textile plant. The incidence of mesothelioma was high among men with 20 or more years’ exposure, but only two cases were observed among more than 16,000 men with under 10 years’ exposure, and one of these seems certain not to have been caused by his employment, as the man was employed for only 4 months and died 4 years later. The current model for mesothelioma may thus overestimate the risk for brief (under 10 years) exposure, at least for chrysotile. 12.10.3

Dose–Response for Lung Cancer

The observation that excess lung cancer risk is roughly proportional to cumulative dose at high concentrations does not constitute very strong evidence of a linear relationship with fiber level, particularly at very low levels. This prediction is even more difficult to test directly for lung cancer than for mesothelioma, as lung cancer is so common in the general population, affecting more than one smoker in 10 and about one nonsmoker in 200, that even quite large increases in risk are difficult to estimate reliably. Prolonged low exposure to

RISK ASSESSMENT ISSUES

439

chrysotile in friction products, asbestos cement, and chrysotile mining has produced no detectable excess of lung cancer (Paustenbach et al., 2004). Even in chrysotile textile production, the sector in which the highest dose-specific risks for chrysotile have been observed, over 10 years’ exposure at low average levels (about 5 f/mL) produced little increase in risk in the United Kingdom study reported by Peto et al. (1985), although workers employed for less than 10 years in the South Carolina plant studied by Dement et al. (1983), who were more heavily exposed, suffered an increased risk (SMR ¼ 1.9). Moreover, there is evidence that the relative risk for lung cancer in chrysotile-exposed workers eventually falls after exposure to chrysotile has ceased (Walker, 1984; Peto et al., 1985). The risk is very likely to differ between chrysotile and the amphiboles, with Hodgson and Darnton (2000) putting the risk differential between 1:10 and 1:50. In the absence of evidence that the model used for lung cancer underestimates the long-term risk for brief or low exposure, and in view of the previously cited reassuring observations, the resulting predictions may, if anything, be too high for environmental exposure. Camus et al. (1998) examined the risk of developing lung cancer among nonoccupationally exposed women living in the vicinity of the Quebec chrysotile mines and mills. While the relative risk predicted by EPA’s model was 2.1, the measured risk was 1.0. 12.10.4

Dose–Response for Amphiboles

No extensive measurements of historical exposure levels are available for the cohorts exposed predominantly to crocidolite or amosite. Estimated levels have been published for the crocidolite miners of Western Australia (Armstrong et al., 1988) and varied from 20 to 100 f/mL. Most were employed for less than a year, however, and more than half had estimated cumulative exposures under 10 f/mL years, although only 5% exceeded 100 f/mL years. In a subsequent publication deKlerk et al. (1989) reported a case–control analysis indicating a significantly elevated lung cancer risk only in the minority of workers (about 3%) exposed for over 5 years, among whom the relative risk was 2.2, based on 11 deaths. No other study provides any useful exposure data for pure crocidolite, however, and this study alone is inadequate as a basis for a firm conclusion. The situation for amosite is also unsatisfactory. The only study of amosite workers for which dose estimates have been provided (Seidman et al., 1979; see Nicholson, 1986 for updated lung cancer data) is a cohort of men manufacturing amosite insulation in Paterson, New Jersey at the beginning of World War II. The dose estimates were based on very limited measurements taken more than 25 years later in two different factories using similar materials and equipment. There was a marked increase in lung cancer SMR, even in men employed for less than 2 months (SMR ¼ 264, based on 15 deaths), and possible estimates of KL vary from 0.01 (using the lung cancer rate in short-term workers as the baseline) to 0.04 (by regression on the SMR, based on local rates) (Nicholson, 1986). The SMR for men exposed for over 2 years was 650, and there were 14 mesotheliomas (seven pleural, seven peritoneal). There are three major difficulties in interpreting this study: the lack of any direct exposure data, the anomalous pattern of SMR in relation to duration of exposure, and the uncertainties related to extrapolation from brief very high exposure to prolonged low exposure. Both amosite and crocidolite have caused high risks of mesothelioma after brief exposure, which has not been observed for chrysotile. Brief amosite exposure can also cause a high lung cancer rate. Moreover, there is consistent evidence that the ratio of mesothelioma to excess lung cancer is higher for amosite, and higher still for crocidolite, than for chrysotile alone.

440

ASBESTOS AND OTHER MINERAL AND VITREOUS FIBERS

One interesting inconsistency relates to the groups of workers exposed to some crocidolite who suffered a substantial risk of mesothelioma but no detectable excess of lung cancer, in contrast to the more heavily exposed crocidolite miners, who appear to have suffered a larger excess of lung cancer than of mesothelioma. Perhaps the most plausible interpretation of these (and many other) differences is that different sizes have different effects, either in their ability to reach the bronchus or to reach the lung and penetrate the pleura, or in their biological activity in different tissues. Unfortunately, however, our understanding of these processes is at present too limited to justify more specific conclusions. 12.10.5

Risks Associated with Nonoccupational Exposures

Mesothelioma among people not occupationally exposed to asbestos has been reported among people living near asbestos mining and processing areas, including members of households containing asbestos workers as well as those without. Presumably, exposures to fibers were higher in homes with workers bringing home dust on their work clothing and shoes, but quantitative exposure data are lacking. Wagner et al. (1960) reported that one-third of the mesothelioma cases reported in his South African population were not occupationally exposed to amphibole asbestos. Also, three studies from Europe and one from the U.S. reported excess neighborhood cases around factories processing South African amphiboles (Newhouse and Thompson, 1965; Hain et al., 1974; Magnani et al., 1995; Hammond et al., 1979). In the U.S., there was very heavy and visible community exposure to chrysotile asbestos in Manville, NJ (Borow and Livornese, 1973). Berry (1997) reported on the environmental, nonoccupational component of mesothelioma incidence among persons living in Manville. Prior to removal of occupational cases, residents of Manville had an average annual (1979–1990) mesothelioma rate of 636 male cases and 96 female cases per million population, about 25 times higher than average state rates. Cases were removed from the analysis when their “usual employment” was reported as being at the asbestos plant, as evidenced through union lists or occupational information from either the Cancer Registry or mortality records. Standardized incidence ratios (SIRs) were computed for residents of Manville and Somerset County (less the Manville population) by sex. New Jersey mesothelioma rates less than Somerset County, 1979–1990, were used to generate the expected number of cases. The SIRs for Manville males and females were respectively 10.1 [95% confidence interval (CI): 5.8–16.4] and 22.4 (95% CI: 9.7–44.2). Male and female Somerset County mesothelioma incidence rates were 1.9 (95% CI: 1.4–2.5) and 2.0 (95% CI: 1.0–3.6). Some of these excesses were due to household exposures, but clearly the generally community exposures caused some of the excess. Populations not occupationally exposed to mineral fiber may have very high incidences of mesothelioma. The most extreme case is the study of Baris et al. (1987) of people living in four villages in Central Cappadocia in Turkey. Three villages (Karain, Sarihidir, and Tuzkoy) were exposed to erionite, a fibrous zeolite, and a fourth (Karlik) lacked this exposure and served as a control. There were 141 deaths during the study period in the four villages, including 33 mesotheliomas, 17 lung cancers, 1 cancer of the larynx, 8 cancers of other sites, and 13 cancers not specified. Thus, there were 72 cancers out of 141 deaths, with at least 33 of them due to mesothelioma. The age- and sex-specific mortality rates per 1000 person-years from mesothelioma and respiratory cancer for the four villages were 20.2, 13.5, 5.2, and 0 for males from Karain, Sarihidir, Tuzkoy, and Karlik, respectively. The

RISK ASSESSMENT ISSUES

441

corresponding rates for females were 10.9, 3.9, 4.9, and 0. Sebastien et al. (1984) examined ferruginous bodies in the sputum of residents of Karain, Tuzkoy, and Karlik. They found that the content of ferruginous bodies increased with age in Karain and Tuzkoy, but only one of 19 specimens from Karlik had any. Mesotheliomas among nonoccupationally exposed people living near crocidolite-mining and -milling regions in South Africa and Western Australia have been known to occur for some time (Wagner and Pooley, 1986; Reid et al., 1990). For a population living near the Wittenoom crocidolite mine in Western Australia, Hansen et al. (1997) were able to show a significant exposure–response relationship based on proximity and duration of exposure. Mesothelioma cases among residents of Cyprus, who had no occupational exposures to fibers, were attributed to environmental tremolite fibers (McConnochie et al., 1987). In California, the incidence of mesothelioma was associated with distance of homes from natural outcrops containing asbestos (Pan et al., 2005) For modeling purposes, however, risk extrapolations for community residents have had to rely on the quite considerable cancer risks associated with past occupational exposures. Within the range of observation, the models are consistent with the conservative assumption of a linear, nonthreshold response. Thus, one can predict risks at the much lower exposure levels observed in schools and in commercial and public buildings. These predictions are unlikely to underestimate the risks and are more likely to overestimate them. Based on mean concentration data from Fig. 12.1 and the lung cancer risk model of Doll and Peto (1985), the increase in cancer risk associated with 20 years of exposure to daily 8-hr exposures in commercial buildings, public buildings, and schools at average concentrations of fibers >5 mm in length in such buildings of 0.0002 f/mL correspond to a lifetime risk of about 2  10 6. However, it should be noted that concentrations in buildings are seldom much higher than concentrations in the air outside the buildings, and therefore much of this small risk is related to the entry of outdoor fibers into the building with the ventilation air. In contrast to risk estimations based on human experience in occupational populations and logical extrapolations to background concentration levels, as in the model of Doll and Peto (1985) described above, Larson (2003) applied the EPA Proposed Guidelines for Carcinogen Risk Assessment (EPA, 1996) to a set of lung cancer mortality data to obtain a “safe” fiber concentration based on a default linear extrapolation to one excess death per one million people, as specified for carcinogenic hazardous air pollutants by the Clean Air Act. He found that the “safe” concentration was 1/1000 of ambient air background concentrations of asbestos fibers. Because the calculated “safe” level cannot be achieved, Larsen suggested that that his risk assessment techniques be used only for airborne carcinogens that have only anthropogenic sources. Perhaps because he is an EPA employee, he did not question the reliability of the EPA Proposed Guidelines, with their numerous conservative defaults (Lippmann, 2003). Our current inability to: (1) reliably measure the concentrations of health-relevant fibers at concentrations near background levels; and (2) reliably quantitate the risks, if any, of exposures at such levels, has often led to confusion, alarm, and misguided acts of risk avoidance. For example, removal of in-place asbestos insulation in schools and public buildings has often increased rather than decreased exposures to asbestos fibers for workers doing the remediation and for building occupants after the remediation. Another example was the inappropriate focus, following the collapse of the World Trade Center buildings, on asbestos fibers in air and residual dust as indices of health risk for rescue workers and volunteers, workers removing debris, and neighborhood residents, office workers, and

442

ASBESTOS AND OTHER MINERAL AND VITREOUS FIBERS

service workers. The measured airborne fiber levels neither warranted that level of concern nor could be related to the health effects that were documented. 12.10.6

Synthetic Vitreous Fibers

Although there has been a significant advance in our knowledge about the deposition and elimination of SVFs and other fibers in recent years, as well as some new knowledge about exposure–response in controlled animal inhalation studies, some further concern about lung cancer among heavily exposed workers in industry, and some new insight into the critical fiber dimensions affecting disease pathogenesis, there are also many important questions which remain to be addressed. In some cases, the behavior and risks of airborne SVFs can be inferred from those of either compact particles or asbestos fibers. On the contrary, the validity of such inferences depends on some critical assumptions about the aerodynamic properties of the various fibers and about the responses of lung and mesothelial cells to such fibers. The differences may be critical, and more in vivo studies with SVFs should be performed in order to further clarify these issues. In the interim, we already know a great deal about the nature and extent of fiber toxicity and the factors that modify its expression. This knowledge provides a good basis for a fairly definitive risk assessment for SVFs. SVFs differ from asbestos fibers in several critical ways and tend to produce less lung deposition and more rapid elimination of those fibers that do deposit in the lungs. One difference is in diameter distribution. Except for glass microfiber, SVFs tend to have relatively small mass fractions in diameters small enough to penetrate through the upper respiratory tract. Asbestos, on the contrary, usually contains more “respirable” fiber. Furthermore, once deposited, the asbestos fibers may split into a larger number of long thin fibers within the lungs. SVFs rarely split but are more likely to break into shorter length segments. There are also differences in solubility among the fibers that affect their toxic potential, among both the asbestos types and the SVFs. Conventional glass fibers appear to dissolve much more rapidly than other SVFs and asbestos. Dissolution of glass fibers takes place both by surface attack and by leaching within the structure. The diameters are reduced and the structure is weakened, favoring break up into shorter segments. Since the smallest diameter fibers have the greatest surface-to-volume ratio, they dissolve most rapidly. Thus, the relatively small fraction of the airborne glass fibers having diameters small enough to penetrate into the lungs are the most rapidly dissolved within the lungs. The more durable and less soluble SVFs, that is, slag and rock wool, some specialty glasses, and ceramic fibers, require a higher degree of concern because of their longer retention within the lungs. In vitro studies and studies of dissolution in simulated lung fluids can be very useful in preliminary evaluations of the toxic potential of the various SVFs. On the contrary, the dissolution of SVFs in vivo depends on many additional factors that cannot readily be simulated in model systems. For example, the differences in solubility in vivo of long and short fibers noted by Morgan and Holmes (1984) were attributed to small difference in intracellular and extracellular pH. The mechanical stress on fibers in vivo may also contribute to their disintegration, and cannot readily be simulated in model systems. Thus, hazard evaluations of specialty product SVFs made for limited and specific applications should include detailed in vivo studies in which animals are exposed to appropriate sizes and concentrations of the fibers of interest.

KEY FACTORS AFFECTING FIBER DOSIMETRY

443

In the case of conventional fibrous glasses, we have sufficient information to conclude that the occupational health risks associated with the inhalation of fibers dispersed during their manufacture, installation, use, maintenance, and disposal are not measurable (Doll, 1987), and hence of an extremely low order. The health risk from casual and infrequent indoor air exposure of building occupants to relatively low concentrations of fibrous glass is therefore essentially nil. These judgments are based on a series of interacting factors, each of which individually leads to a far lower order of risk for conventional glass fibers than asbestos. Specifically, 1. Conventional glass fibers are less readily aerosolized than asbestos during comparable operations, as demonstrated by the much lower fiber counts measured at various industrial operations (Cherrie et al., 1986; Esmen, 1984). 2. A much smaller fraction of conventional glass fibers than asbestos fibers have small enough aerodynamic diameters to penetrate into lung airways (i.e., fibers with diameters below 3 mm) (Konzen, 1984). 3. The glass fibers that can penetrate into the lungs are much less durable within the lung than asbestos. They tend to break up into shorter segments, so that fewer fibers longer than the critical length limits are retained at critical sites. They also tend to dissolve, further reducing their retention (Bernstein et al., 1984). 4. The inherent toxicity of conventional glass fibers is much lower than that of asbestos fibers of similar dimension, as shown by studies in which fiber suspensions are applied directly to target tissues by intratracheal instillation (Wright and Kuschner, 1977) or application of a fiber mat to the lung pleura (Stanton and Wrench, 1972). In consideration of these factors, the risk for lung fibrosis is virtually nil unless there is continuous exposure at concentrations high enough to maintain a high level of lung burden for this relatively rapidly cleared type of particulate. The risk of lung cancer is also virtually nil unless there is continuous exposure to long fibers at high concentrations because of the relatively rapid breakup of long fibers into short fiber segments within the lungs. Finally, the risk of mesothelioma from inhaled conventional glass fibers is virtually nil under almost any circumstance. There are hardly any glass fibers thin enough to cause mesothelioma in the aerosols, and the very few that may be present would dissolve rapidly within the lungs.

12.11 KEY FACTORS AFFECTING FIBER DOSIMETRY AND TOXICITY: RECAPITULATION AND SYNTHESIS 12.11.1

Critical Fiber Properties Affecting Toxicity

Review of the in vitro studies clearly indicates that fiber length, diameter, and composition are critical determinants of biopersistence, cytotoxicity and cell transformation. A review of the in vivo animal studies, both by inhalation and injection, shows that fiber dimensions and composition are important factors affecting pathological measures such as fibrosis and cancer yields. Review of human exposure–response shows that the proportions of the different diseases caused by asbestos, that is, asbestosis, lung cancer, and mesothelioma, vary greatly among occupational cohorts and that the mesothelioma/lung cancer ratio tends to increase with decreasing fiber diameter for the durable amphibole forms of asbestos.

444

12.11.2

ASBESTOS AND OTHER MINERAL AND VITREOUS FIBERS

Influence of Fiber Diameter

Fiber diameter affects airborne fiber penetration into and along the lung airways and thereby the initial deposition patterns. The aerodynamic diameters of mineral fibers are about three times their physical diameters (Timbrell, 1972; St€ ober et al., 1970). Thus, fibers with diameters larger than 3 mm will not penetrate in the lungs (Lippmann, 1990). Fibers with diameters 0.1 mm are less well retained in the lungs than larger fibers (Lippmann and Timbrell, 1990). Their large surface-to-volume ratio favors dissolution (Lippmann, 1990). Those sufficiently durable not to dissolve can readily penetrate the epithelial surface and be translocated to the lung interstitium and pleural surfaces. The fibers that remain in the lungs can cause fibrosis and lung cancer, and those durable fibers that are translocated to pleural surfaces can cause mesothelioma. Thus, for asbestosis and lung cancer, the upper fiber diameter limit is on the order of 3 mm. For mesothelioma, the upper fiber diameter limit is likely to be much less for two reasons. First, the thinner fibers penetrate to the gas-exchange region to a greater extent. Second, fibers thinner than 0.5 mm are translocated from the deposition sites to postnodal lymphatic channels more than the thicker fibers and thus reach any organ of the body (Oberd€ orster et al., 1988). 12.11.3

Influence of Fiber Length

Fiber length can also affect fiber penetration into and along the airways. As the length increases beyond 10 mm, the interception mechanism begins to significantly enhance deposition (Sussman et al., 1991a, 1991b). Thus, longer fibers have proportionately more airway deposition and less deposition in the gas-exchange region. Lung retention also increases markedly with increasing fiber length above 10 mm for biopersistent fibers, both on theoretical grounds and on the basis of analysis of residual lung dust in humans (Pooley and Wagner, 1988; Churg and Wiggs, 1987; Timbrell et al., 1987) and animals (Morgan, 1979). Furthermore, fibers shorter than about 6 mm in length can readily penetrate through tracheobronchial lymph nodes and be translocated to more distant organs (Oberd€ orster et al., 1988). Exact specification of the critical lengths for the different diseases remains difficult, since the experimental studies generally have had, of practical necessity, to use imperfectly classified fiber suspensions. Also, the experimental studies have used very large concentrations, and apportioning attribution of the cytotoxicity and pathology produced to the effects of fiber size versus dust overload phenomena is difficult. In other words, the results described in the in vivo section of this review would be consistent either with short fibers having a much smaller effect than long fibers or with their contributing to the growth of fibrotic lesions caused by the relatively few long fibers in the tail of the fiber length distribution. In any case, the fibers shorter than 5 mm have very much less toxicity; whereas cytotoxicity and disease increase with fiber length for fibers longer than 5 mm. 12.11.4

Influence of Fiber Composition

Comparative retention and toxicity studies with various kinds of asbestos and other fibrous minerals, ceramics and glasses indicate that properties other than fiber dimensions affect fiber retention and toxicity. Among these are solubility; specific surface area; surface

KEY FACTORS AFFECTING FIBER DOSIMETRY

445

electrical charges that may contribute to redox reactions generating active oxygen species; and so on. Thus dimensional characteristics alone, although important, are insufficient indicators of fiber toxicity. It is now time to revise the Stanton hypothesis, which acknowledges the critical importance of fiber length and diameter in biological responses, and recognize the importance of the other physical-chemical properties that impart biological potential to fibers. A major research need is a systemic exploration of the surface properties and factors affecting solubility of fibers in lung fluids and cells, so that due considerations can be given to fiber composition in hazard assessment. 12.11.5

Risk Assessment for Inorganic Fibers

For asbestos, there is general agreement that occupational exposures to all fibrous forms have caused asbestosis and contributed to excesses of lung cancers. For mesothelioma, it is accepted that inhalation of amphibole and erionite fibers in workers and the general population has been causal. It is also generally agreed that occupational exposure to chrysotile asbestos has been associated with cases of mesothelioma, but many believe that these cases were more likely due to the contamination of most commercial chrysotile with amphibole fibers, and if chrysotile fibers do cause mesothelioma, they are considerably less potent in that regard than amphibole fibers. More definitive conclusions will require studies having better descriptions of the fiber sizes and compositions. For SVFs, the International Agency for Research on Cancer (IARC, 2002) has, on the basis of their own review of the data on carcinogenic risk, concluded that: (1) for humans, there is inadequate evidence for the carcinogenicity of glass wool, continuous glass filament, rock (stone) wool/slag wool, and refractory ceramic fibers; (2) for experimental animals, there is inadequate evidence for the carcinogenicity of continuous glass filament, and for certain newly developed, less biopersistent fibers (X-607 and HT wools and A, C, F, and G), limited evidence for insulation glass wool, rock (stone) wool, slag wool, and more biopersistent fibers such as fiber H, and sufficient evidence for the carcinogenicity of special purpose glass fibers, including E-glass and “475” glass fibers, as well as for refractory ceramic fibers. For all inorganic fibers, as for other airborne toxicants, the dose makes the poison. However, for these fibrous toxicants, the physical form and properties can be as important, or more important, than the chemical form. The aerodynamic diameters of the fibers, and therefore their deposition patterns and efficiencies within the lung airways, is determined largely by fiber width. For fibers >10 mm in length, interception enhances airway deposition, but even more importantly, these longer fibers elicit cellular responses that shorter fibers do not, and they also are subject to different clearance pathways and rates. Another physical property, their solubility within lung fluids, then becomes a major determinant of their toxicity. While these special determinants of risk are being increasingly recognized, they are not yet reflected in Standards or Guidelines for exposure assessment, an essential tool in risk assessment. Thus, there is an urgent need for new and improved occupational and ambient air quality limits for specific fiber compositions that recognize fiber length and diameter as critical risk factors. Dissolution rate in vivo is the other main dimension in the risk equation. Fortunately, the SVF manufacturing industry in the U.S. and the European Community has recognized the critical importance of fiber biopersistence to risk, and has revised many product formulations so that they have higher dissolution constants.

446

ASBESTOS AND OTHER MINERAL AND VITREOUS FIBERS

ACKNOWLEDGMENTS This research was performed as part of a Center Program supported by NIEHS (Grant ES, 00260). It includes extensive review material from earlier review papers, specifically Lippmann (1988, 1990, 1994 and HEI-AR, 1991) Health Effects Institute-Asbestos Research (1991).

12.13 ACRONYMS ACGIH: ACM: AM: BMRC: HEI-AR: ip MALS: MMMF: MMVF: MPPCF: MTD OSHA: PCOM: PEL: PMN: RCF ROS SIR SMR: SVF TEM: TLV: UICC:

American Conference of Governmental Industrial Hygienists Asbestos-containing material Alveolar macrophage British Medical Research Council Health Effects Institute-Asbestos Research intraperitoneal, a dose delivery technique for fibers Magnetic alignment and light scattering Manmade mineral fiber, an alternate name for SVF Manmade vitreous fiber, an alternate name for SVF Millions of particles per cubic foot Maximum tolerated dose Occupational Safety and Health Administration Phase-contrast optical method Permissible exposure limit Polymorphonuclear leukocytes Refractory ceramic fiber Reactive oxygen species Standardized incidence ratio Standardized mortality ratio Synthetic vitreous fiber Transmission electron microscopy Threshold limit value International Union Against Cancer (English translation of name of organization in French

REFERENCES ACGIH-AIHA Aerosol Hazards Evaluation Committee (1975) Recommended procedures for sampling and counting asbestos fibers. Am. Ind. Hyg. Assoc. J. 36:83–90. Adamson IYR (1997) Early mesothelial cell proliferation after asbestos exposure: in vivo and in vitro studies. Environ. Health Perspect. 105(Suppl. 5):1205–1208. Albin M, Pooley FD, Str€omberg U, Atteweil U, Mitha R, Johansson L, Welinder H (1994) Retention patterns of asbestos fibers in lung tissue among asbestos cement workers. Occup. Environ. Med. 51:205–211. Allison AC (1977) Mechanisms of macrophage damage in relation to the pathogenesis of some lung diseases. In:Brain JD, Proctor DF, Reid LM, editors. Respiratory Defense Mechanisms Part II. New York: Marcel Dekker, pp. 1075–1102.

REFERENCES

447

Armstrong BK, deKlerk NH, Musk AW, Hobbs MST (1988) Mortality in miners millers of crocidolite in Western Australia. Br. J. Ind. Med. 45:5–13. Ashcroft T, Simpson JM, Timbrell V (1988) Simple method of estimating severity of pulmonary fibrosis on a numerical scale. J. Clin. Pathol. 41:467–470. Balashazy I, Moustafa M, Hofmann W, Szoke R, El-Hussein A, Ahmed A-R (2005) Simulation of fiber deposition in bronchial airways. Inhal. Toxicol. 17:717–727. Baris I, Simonato L, Artvinli M, Pooley F, Saracci S, Skidmore J, Wagner C (1987) Epidemiological and environmental evidence of the health effects of exposure to erionite fibers: a four-year study in the Cappadocian region of Turkey. Int. J. Cancer 39:10–17. Barrett JC, Lamb PW, Wisemen RW (1989) Multiple mechanisms for the carcinogenic effects of asbestos and other mineral fibers. Environ. Health Perspect. 81:81–89. Bauman MD, Jetten AM, Bonner JC, Kumar RK, Bennett RA, Brody AR (1990) Secretion of a platelet-derived growth factor homologue by rat alveolar macrophages exposed to particulates in vitro. Eur. J. Cell Biol. 51:327–334. Beck EG, Bruch J, Friedricks KH, Hilscher W, Pott F (1971) Fibrous silicates in animal experiments and cell-culture in morphological cell and tissue reactions according to different physical chemical influences. In: Walton WH, editor. Inhaled Particles III, Old Woking. U.K.: Unwin Bros., pp. 477–486. Begin R, Gauthier JJ, Desmeules M, Ostiguy G (1992) Work-related mesothelioma in Quebec, 1967– 1990. Am. J. Ind. Med. 22:531–542. Bellmann B, Konig H, Muhle H, Pott H (1986) Chemical durability of asbestos and of man-made mineral fibres in vivo. J. Aerosol Sci. 17:341–345. Bellmann B, Muhle H, Pott H, Konig H, Kloppel H, Spurny K (1987) Persistence of man-made mineral fibers (MMMF) and asbestos in rat lungs. Ann. Occup. Hyg. 31(4B):693–709. Bellmann B, Muhle H, Creutzenberg O, Ernst H, Brown RC, Sebastien P (2001) Effects of nonfibrous particles on ceramic fiber (RCF1) toxicity in rats. Inhal. Toxicol. 13:877–901. Berman W, Chatfield EJ (1989) Interim superfund method for the determination of asbestos in ambient air. Part II: Technical background document. Draft Report EPA Contract No. 68-61-7290, Task 117, USEPA, Region 9, San Francisco, CA 94105. Berman DW, Crump KS (2001) Technical support document for a proposal to assess asbestos-related risk. Final draft. Prepared for the U.S. Environmental Protection Agency and the U.S. Department of Transportation, September 4, 2001. Berman DW, Crump KS, Chatfield EJ, Davis JMG, Jones AD (1995) The sizes, shapes, and mineralogy of asbestos structures that induce lung tumors or mesothelioma in AF-HAN rats following inhalation. Risk Anal. 15:181–195. Bernstein DM, Drew RT, Schidlovsky G, Kuschner H (1984) Pathogenicity of MMMF and the contrasts to natural fibers. In: Biological Effects of Man-Made Mineral Fibers. Vol. 2. Copenhagen: World Health Organization, pp. 169–195. Bernstein DM, Morscheidt C, Grimm H-G, Thevenaz P, Teichert U (1996) Evaluation of soluble fibers using the inhalation biopersistence model, a nine fiber comparison. Inhal. Toxicol. 18:345–385. Bernstein DM, Riego Sintes JM, Ersboell BK, Kunert J (2001a) Biopersistence of synthetic mineral fibers as a predictor of chronic inhalation toxicity in rats. Inhal. Toxicol. 13:823– 849. Bernstein DM, Riego Sintes JM, Ersboell BK, Kunert J (2001b) Biopersistence of synthetic mineral fibers as a predictor of chronic intraperitoneal injection tumor response in rats. Inhal. Toxicol. 13:851–875. Bernstein DM, Rogers R, Smith P (2004) The biopersistence of Brazilian chrysotile asbestos following inhalation. Inhal. Toxicol. 16:745–761.

448

ASBESTOS AND OTHER MINERAL AND VITREOUS FIBERS

Bernstein DM, Chevalier J, Smith P (2005a) Comparison of calidria chrysotile asbestos to pure tremolite: final results of the inhalation biopersistence and histopathology examination following short-term exposure. Inhal. Toxicol. 17:427–449. Bernstein DM, Rogers R, Smith P (2005b) The biopersistence of Canadian chrysotile asbestos following inhalation: final results through 1 year after cessation of exposure. Inhal. Toxicol. 17:1–14. Berry M (1997) Mesothelioma incidence community asbestos exposure. Environ. Res. 75:34–40. Berry J, Newhouse M (1983) Mortality of workers manufacturing friction materials using asbestos. Br. J. Ind. Med. 40:1–7. Bolton RE, Vincent JH, Jones AD, Addison J, Beckett ST (1983) An overload hypothesis for pulmonary clearance of UICC amosite fibers inhaled by rats. Br. J. Ind. Med. 40:264–272. Borow C, Livornese LL (1973) Mesothelioma following exposure to asbestos: a review of 72 cases. Chest 201:444–449. Bourdes V, Boffetta P, Pisani P (2000) Environmental exposure to asbestos and risk of pleural mesothelioma: review and meta-analysis. Eur. J. Epidemiol. 16:411–417. Boutin C, Dumortier P, Rey F, Viallat JR, DeVuyst P (1996) Black spots concentrate oncogenic asbestos fibers in the parietal pleura: thorascopic and mineralogic study. Am. J. Respir. Crit. Care Med. 153:444–449. Briant JK (1988) Distinction of reversible and irreversible convection by high-frequency ventilation of a respiratory airway cast, Ph. D. Thesis, New York University. Brody AR, Roe MW (1983) Deposition pattern of inorganic particles at the alveolar level in the lungs of rats and mice. Am. Rev. Respir. Dis. 128:724–729. Brody AR, Hill LH, Adkins B Jr, O’Connor RW (1981) Chrysotile asbestos inhalation in rats: deposition pattern and reaction of alveolar epithelium and pulmonary macrophages. Am. Rev. Respir. Dis. 123:670–679. Buchanich JM, Marsh GM, Youk AO (2001) Historical cohort study of US man-made vitreous fiber production workers: V. Tobacco smoking habits. J. Occup. Environ. Med. 43:793–802. Burdett G (1985) Use of a membrane filter, direct-transfer technique for monitoring environmental asbestos releases. In:A Workshop on Asbestos Fiber Measurements in Building Atmospheres. Mississauga, Ontario: Ontario Research Foundation. Camus M, Siemiatycki J, Meek B (1998) Nonoccupational exposure to chrysotile asbestos and the risk of lung cancer. N. Engl. J. Med. 338:1565–1571. Case BW, Dufresne A, McDonald AD, McDonald JC, Sebastien P (2000) Asbestos fiber type and length in lungs of chrysotile textile and production workers: fibers longer than 18 mm. Inhal. Toxicol. 12 (Suppl. 3):411–418. Chan TL, Lee PS, Hering WE (1984) Pulmonary retention of inhaled diesel particles after prolonged exposures to diesel exhaust. Fundam. Appl. Toxicol. 4:624–631. Cherrie J, Dodgson J, Groat S, MacLaren W (1986) Environmental surveys in the European man-made mineral fiber production industry. Scand. J. Work Environ. Health 12(Suppl. 1):18–25. Chisholm JE (1983) Transmission electron microscopy of asbestos. In: Chissick SS, Derricott R, editors. Asbestos. Chichester: John Wiley & Sons, pp. 86–167. Churg A (1994) Deposition and clearance of chrysotile asbestos. Ann. Occup. Hyg. 38:625–633. Churg A, Wiggs B (1987) Accumulation of long asbestos fibers in the peripheral upper lobe in cases of malignant mesothelioma. Am. J. Ind. Med. 11:563–569. Churg A, Wiggs B, Depaoli L, Kampe B, Stevens B (1984) Lung asbestos content in chrysotile workers with mesothelioma. Am. Rev. Respir. Dis. 130:1042–1045. Churg A, Wright J, Gilks B, Dai J (2000) Pathogenesis of fibrosis produced by asbestos and man-made mineral fibers: what makes a fiber fibrogenic? Inhal. Toxicol. 12 (Suppl. 3):15–26.

REFERENCES

449

Collier CG, Morris KJ, Launder KA, Humphreys JA, Morgan A, Eastes W, Townsend S (1994) The behavior of glass fibers in the rat following intraperitoneal injection. Regul. Toxicol. Pharmacol. 20:S89–S103. Collier CG, Kamstrup O, Morris KJ, Applin RA, Vatter IA (1997) Lung clearance of experimental man-made mineral fibres, preliminary data on the effect of fiber length. Ann. Occup. Hyg. 41 (Suppl. 1):320–326. Cowie HA, Wild P, Beck J, Auburtin G, Piekarski C, Massin N, Hurley JF, Miller BG, Groat S, Soutar CA (2001). An epidemiological study of the respiratory health of workers in the European refractory ceramic fibre industry. Occup. Environ. Med. 58:800–810. Craighead JE (1987) Current pathogenetic concepts of diffuse malignant mesothelioma. Hum. Pathol. 18:544–557. Cullen RT, Searl A, Buchanan D, Davis JMG, Miller BG, Jones AD (2000) Pathogenicity of a specialpurpose glass microfiber (E glass) relative to another glass microfiber and amosite asbestos. Inhal. Toxicol. 12:959–977. Davis JMG (1976) Pathological aspects of the injection of fiber into the pleural and peritoneal cavities of rats and mice. In: Occupational Exposure to Fibrous Glass, HEW Publ. No. (NIOSH) 76–151. Washington, DC: DHEW, pp. 141–199. Davis JMG, Beckett ST, Bolton RE, Collings P, Middleton AP (1978) Mass and number of fibers in the pathogenesis of asbestos-related lung disease in rats. Br. J. Cancer 37:673–688. Davis JMG, Beckett ST, Bolton R, Donaldson K (1980) A comparison of the pathological effects in rats of the UICC reference samples of amosite and chrysotile with those of amosite and chrysotile collected from the factory environment. In: Wagner JC, editor. Biological Effects of Mineral Fibres. IARC Scientific Publications, pp. 285–292. Davis JMG (1987) Experimental data relating to the importance of fibre type, size, deposition, dissolution and migration. In: Proceedings of 1987 Mineral Fiber Symposium, Lyons, IARC. Davis JMG, Addison J, Bolton RE, Donaldson K, Jones AD, Miller BG (1985) Inhalation studies on the effects of tremolite and brucite dust in rats Carcinogenesis 6:667–674. Davis JMG, Addison J, Bolton RE, Donaldson K, Jones AD, Smith T (1986) The pathogenicity of long versus short fiber samples of amosite asbestos administered to rats by inhalation and intraperitoneal injection. Br. J. Exp. Pathol. 67:415–430. Davis JMG, Gylseth B, Morgan A (1986) Assessment of mineral fibers from human lung tissue Thorax 41:167–175. Davis JMG, Jones AD (1988) Comparisons of the pathogenicity of long and short fibres of chrysotile asbestos in rats. Br. J. Exp. Pathol. 69:717–737. Davis JMG, Addison J, McIntosh C, Miller BG, Niven K (1991) Variations in the carcinogenicity of tremolite dust samples of differing morphology. Ann. NY Acad. Sci. 643:473–490. deKlerk NH, Armstrong BK, Musk AW, Hobbs MST (1989) Cancer mortality in relation to measures of occupational exposure to crocidolite at Wittenoom Gorge in Western Australia. Br. J. Ind. Med. 46:529–536. deKlerk NH, Musk AW, Glancy JJ, Pang PS, Lund HG, Olsen N, Hobbs MST (1997) Crocidolite, radiographic asbestosis and subsequent lung cancer. Ann. Occup. Hyg. 41 (Suppl. 1):134– 136. Dement JM, Wallingford KM (1990) Comparison of phase contrast and electron microscopic methods for evaluation of occupational asbestos exposures. Appl. Occup. Environ. Hyg. 5:242–247. Dement JM, Harris RL, Symans MJ, Shy C (1982) Estimates of dose–response for respiratory cancer among chrysotile asbestosis workers. Ann. Occup. Hyg. 28:869–887. Dement JM, Harris RL Jr, Symans MR, Shy CM (1983) Exposures and mortality among chrysotile asbestos workers. Part II. Mortality. Am. J. Ind. Med. 4:421–433.

450

ASBESTOS AND OTHER MINERAL AND VITREOUS FIBERS

Dement JM, Brown DP, Okun A (1994) Follow-up study of chrysotile asbestos textile workers: cohort mortality case-control analyses. Am. J. Ind. Med. 26:431–447. Dodson RF, Williams MG Jr, Corn CJ, Brollo A, Bianchi A (1990) Asbestos content of lung tissue, lymph nodes, and pleural plaques from former shipyard workers. Am. Rev. Respir. Dis. 142:843–847. Dodson RF, Atkinson MAL, Levin JL (2003) Asbestos fiber length as related to potential pathogenicity: a critical review. Am. J. Indust. Med. 44:291–297. Doll R (1987) Symposium on MMMF, Copenhagen, October 1986: overview and conclusions. Ann. Occup. Hyg. 31 (4B):805–819. Doll R, Peto J (1985) Effects on Health of Exposure to Asbestos. A Report to the Health and Safety Commission. London: Her Majesty’s Stationery Office. Dufresne A, Begin R, Masse S, Dufresne CM, Loosereewanich P, Perrault G (1996) Retention of asbestos fibres in lungs of workers with asbestosis, asbestosis and lung cancer, and mesothelioma in Asbestos township. Occup. Environ. Med. 53:801–807. Eastes W, Hadley JG (1995) Dissolution of fibers inhaled by rats. Inhal. Toxicol. 7:179–196. Eastes W, Hadley JG (1996) A mathematical model of fiber carcinogenicity and fibrosis in inhalation and intraperitoneal experiments in rats. Inhal. Toxicol. 8:323–343. Engholm G, Englund A, Fletcher CC, Hallin N (1987) Respiratory cancer incidence in Swedish construction workers exposed to man-made mineral fibers and asbestos. Ann. Occup. Hyg. 31 (4B):663–675. Enterline PE, Marsh GM, Esmen NA (1983) Respiratory disease among workers exposed to manmade mineral fibers. Am. Rev. Respir. Dis. 128:1–7. Enterline PE, Marsh GM, Henderson V, Callahan C (1987) Mortality update of a cohort of U.S. manmade mineral fiber workers. Ann. Occup. Hyg. 31 (4B):625–656. EPA (April 1996) Proposed guidelines for carcinogen risk assessment. EPA/600/P-92/003C. U.S. Government Printing Office, Washington, DC. ERG (2003a) Report on the Peer Consultation Workshop to Discuss a Proposed Protocol to Assess Asbestos Related Risk. Eastern Research Group, Inc., Lexington, MA, May 30, 2003. ERG(2003b) Report on the expert panel on health effects of asbestos and synthetic vitreous fibers in the influence of fiber length. Eastern Research Group, Inc., Lexington, MA. March 17, 2003. Esmen NA (1984) Short-term survey of airborne fibres in U.S. manufacturing plants. In: Biological Effects of Man-Made Mineral Fibers, Vol. 1. Copenhagen: WHO-EURO, pp. 65–82. Esmen NA, Erdal S (1990) Human occupational and nonoccupational exposure to fibers. Environ. Health Perspect. 88:277–286. Ferin J (1972) Observations concerning alveolar dust clearance. Ann. NY Acad. Sci. 200:66–72. Ferin J, Leach LJ (1976) The effect of amosite and chrysotile asbestos on the clearance of TiO2 particles from the lung. Environ. Res. 12:250–254. Finkelstein MM (1983) Mortality among long-term employees of an Ontario asbestos-cement factory. Br. J. Ind. Med. 40:138–144. Floyd RA (1990) Role of oxygen free radicals in carcinogenesis and brain ischemia. FASEB J. 4:2587–2597. Forster H (1984) The behavior of mineral fibers in physiological solutions. In: Biological Effects of Man-Made Mineral Fibers, Vol. 2. Copenhagen: World Health Organization, pp. 27–59. Frank AL, Dodson RF, Williams MG (1997) Lack of tremolite in UICC reference chrysotile and the implications for carcinogenicity. Ann. Occup. Hyg. 41 (Suppl. 1):287–292. Gentry JW, Spurny KR, Schormann J, Opiela H (1983) Measurement of the diffusion coefficient of asbestos fibers. In: Marple VA, Liu BYH, editors. Aerosols in the Mining and Industrial Work Environments, Vol. 2. Ann Arbor: Ann Arbor Science Publishers, pp. 593–612.

REFERENCES

451

Gibbs GW (1979) Etiology of pleural calcification: a study of Quebec asbestos miners and millers. Arch. Environ. Health 34:76–82. Goodglick LA, Kane AB (1990) Cytotoxicity of long and short crocidolite asbestos fibers in vitro and in vivo. Cancer Res. 50:5153–5163. Griffis LC, Henderson TR, Pickrell JA (1981) A method for determining glass in rat lung after exposure to a glass fiber aerosol. Am. Ind. Hyg. Assoc. J. 42:566–569. Gurman JL, Schlesinger RB, Lippmann M (1980) A variable-opening mechanical larynx for use in aerosol deposition studies. Am. Ind. Hyg. Assoc. J. 41:678–680. Gurman JL, Lippmann M, Schlesinger RB (1984a) Particle deposition in replicate casts of the human upper tracheobronchial tree under constant and cyclic inspiratory flow. I. Experimental. Aerosol Sci. Technol. 3:245–252. Gurman JL, Lioy PJ, Lippmann M, Schlesinger RB (1984b) Particle deposition in replicate casts of the human upper tracheobronchial tree under constant and cyclic inspiratory flow. II. Empirical model. Aerosol Sci. Technol. 3:253–257. Gylseth B, Mowe B, Wannag A (1983) Fiber type and concentration in the lungs of workers in an asbestos cement factory. Br. J. Ind. Med. 40:375–379. Hain E, Dalquen P, Bohlig H ,et al. (1974) Retrospective study of 150 cases of mesothelioma in Hamburg area. Int. Arch. Argeits Med. 33:15–37. Hammad YY (1984) Deposition and elimination of MMMF. In: Biological Effects of Man-Made Mineral Fibers, Vol. 2. Copenhagen: World Health Organization, pp. 126–142. Hammond EC, Garfinkle L, Selikoff IT et al. (1979) Mortality experience of residents in the neighborhood of an asbestos factory. Ann. NY Acad. Sci. 330:417–422. Hansen J, deKlerk NH, Musk AW, Eccles JL, Hobbs MST (1997) Mesothelioma after environmental crocidolite exposure. Ann. Occup. Hyg. 41(Suppl. 1):189–193. Harington JS (1981) Fiber carcinogenesis: epidemiologic observations and the Stanton hypothesis. J. Natl. Cancer Inst. 67:977–989. HEI-AR(1991) Health effects institute-asbestos research, Report of Literature Review Panel, Cambridge, MA. Heitt DM (1978) Experimental asbestosis: an investigation of functional and pathological disturbances. Br. J. Ind. Med. 35:129–145. Hesterberg TW, Miller WC, McConnell EE, Chevalier J, Hadley JG, Bernstein DM, Thevenaz P, Anderson R (1993). Chronic inhalation toxicity of size-separated glass fibers in Fischer 344 rats. Fund. Appl. Toxicol. 20:464–476. Hesterberg TW, Hart GA (2000) Lung biopersistence and in vitro dissolution rate predict the pathogenic potential of synthetic vitreous fibers. Inhal. Toxicol. 12 (Suppl. 3):91–97. Hodgson JT, Darnton A (2000) The quantitative risks of mesothelioma and lung cancer in relation to asbestos exposure. Ann. Occup. Hyg. 44:565–601. Holt PR (1982) Translocation of asbestos dust through the bronchiolar wall. Environ. Res. 27:255–260. Holt PF, Mills J, Young KK (1965) Experimental asbestosis with four types of fibers. Ann. NY Acad. Sci. 132:87–97. Hook HL, Morrice G Jr, deTreville RT (1970) Fiberglass manufacturing and health: results of a comprehensive physiological study. Proceedings of the 35th Annual Meeting of the Industrial Health Foundation. October 13–14, 1970. Pittsburgh, PA. IHF Bulletin No. 44. pp. 103–111. Hughes JM, Jones RN, Glindmeyer HW, Hammad YY, Weill H (1993) Follow up study of workers exposed to man made mineral fibres. Br. J. Ind. Med. 50 (7):658–667. IARC(2002) IARC Monographs on the Evaluation of Carcinogenic Risks to Human: Volume 81— Man-Made Vitreous Fibres. World Health Organization. International Agency for Research on Cancer. Geneva: IARC Press.

452

ASBESTOS AND OTHER MINERAL AND VITREOUS FIBERS

Johnson NF, Griffiths DM, Hill RJ (1984) Size distributions following long-term inhalation of MMMF. In: Biological Effects of Man-Made Mineral Fibers, Vol. 2. Copenhagen: World Health Organization, pp. 102–125. Jones JSP, Pooley FD, Sawle GW, Madeley RJ, Smith PG, Berry G, Wignall BK, Aggarwal A (1980) The consequences of exposure to asbestos dust in a wartime gas-mask factory. In: Wagner JC, editor. Biological Effects of Mineral Fibres, IARC Scientific Pub. No. 30, Vol. 2. Lyons: International Agency for Research on Cancer, pp. 637–653. Jones AD, Johnston AM, Vincent JH (1983) Static electrification of airborne asbestos dust. In: Marple VA, Liu BYH, editors. Aerosols in the Mining and Industrial Work Environments, Vol. 2. Ann Arbor: Ann Arbor Science Publishers, pp. 613–622. Jones AD, McMillan CH, Johnston AM, McIntosh C, Cowie H, Bolton RE, Borzucki G, Vincent JH (1988) Pulmonary clearance of UICC amosite fibers inhaled by rats during chronic exposure at low concentration. Br. J. Ind. Med. 45:300–304. Kane AB, McDonald JL (1993) Mechanisms of mesothelial cell injury, proliferation and neoplasia induced by asbestos fibers. In: Warheit DB, editor. Fiber Toxicology: Contemporary Issues. Academic Press, pp. 323–347. King EJ, Clegg JW, Rae VM (1946) The effect of asbestos, and of asbestos and aluminum, on the lung of rabbits. Thorax 1:188–197. Kjærheim K, Boffetta P, Hansen J, Cherrie J, Chang-Claude J, Eilber U, Ferro G, Guldner K, Olsen J, Plato N, Proud L, Saracci R, Westerholm P, Anderson A (2002) Lung cancer among rock and slag wool production workers. Epidemiology 13:445–453. Klingholz R, Steinkopf B (1982) Das Verhalten von Kunstlichen Mineralfasern in einer physiologischen Modellflassigkeit und in Wasser. Staub-Reinhalt. Luft. 42:69–76. Klingholz R, Steinkopf B (1984) The reactions of MMMF in a physiological model fluid and in water. In: Biological Effects of Man-Made Mineral Fibers, Vol. 1. Copenhagen: WHO-EURO, pp. 60–86. Konzen JL (1984) Production trends in fibre sizes of MMMF insulation. In: Biological Effects of ManMade Mineral Fibers, Vol. 1. Copenhagen: World Health Organization, pp. 44–63. Langer AM, Nolan RP (1986) The properties of chrysotile asbestos as determinants of biological activity. Accom. Oncol. 1 (2):30–51. Langer AM, Nolan RP, Addison J (1990) Physico-chemical properties of asbestos as determinants of biological potential. In: Miller K, Liddle D, editors. CRC Monograph Asbestos and Disease. Boca Raton, FL: CRC Press. Larson RI (2003) An air quality data analysis system for interrelating effects, standards, and needed source reductions: Part 13. Applying the EPA Proposed Guidelines for Carcinogen Risk Assessment to a set of asbestos lung cancer mortality data. J. Air Waste Manage. Assoc. 53:1226–1339. LeBouffant L, Henin JP, Martin JC, Normand C, Tichoux G, Trolard F (1984) Distribution of inhaled MMMF in the rat lung—long term effects. In: Biological Effects of Man-Made Mineral Fibers, Vol. 2. Copenhagen: World Health Organization, pp143–168. Lee KP, Barras CE, Griffith FD, Waritz RS, Lapin CA (1981) Comparative pulmonary responses to inhaled inorganic fibers with asbestos and fiberglass. Environ. Res. 24:167–191. Lee KP, Kelly DP, Kennedy GL Jr (1983) Pulmonary response to inhaled Kevlar aramid synthetic fibers in rats. Toxicol. Appl. Pharmacol. 71:242–253. LeMasters G, Lockey J (1994) Radiographic changes among workers manufacturing refractory ceramic fibre and products. Ann. Occup. Hyg. 38 (Suppl. 1):745–751. LeMasters GK, Lockey JE, Levin LS, Mckay RT, Rice CH, Horvath EP, Papes DM, Lu JW, Feldman DJ (1998) An industry-wide pulmonary study of men and women manufacturing refractory ceramic fibers. Am. J. Epidemiol. 148:910–919. LeMasters GK, Lockey JE, Yiin JH, Hilbert TJ, Levin LS, Rice CH (2003). Mortality of workers occupationally exposed to refractory ceramic fibers. J. Occup. Environ. Med. 45:440–450.

REFERENCES

453

Lentz TJ, Rice CH, Succop PA, Lockey JE, Dement JM, LeMasters GK (2003) Pulmonary deposition modeling with airborne fiber exposure data: a study of workers manufacturing refractory ceramic fibers. App. Occup. Environ. Hyg. 18 (4):278–288. Levin JL, McLarty JW, Hurst GA, Smith AN, Frank AL (1998) Tyler asbestos workers: mortality experience in a cohort exposed to amosite. Occup. Environ. Med. 55:155–160. Liddell FDK, Armstrong BG (2002) The combination of effects on lung cancer of cigarette smoking and exposure in Quebec chrysotile miners and millers. Ann. Occup. Hyg. 46:5–13. Lippmann M (1988) Asbestos exposure indices. Environ. Res. 46:86–106. Lippmann M (1990) Man-made mineral fibers (MMMF): human exposures and health risk assessment. Toxicol. Ind. Health 6:225–246. Lippmann M (1994) Deposition and retention of fibres: effects on incidence of lung cancer and mesothelioma. Occup. Environ. Med. 51:793–798. Lippmann M (2003) Letter to the Editor. J. Air Waste Manage. Assoc. 53:1298–1300. Lippmann M, Timbrell V (1990) Particle loading in the human lung—human experience and implications for exposure limits. J. Aerosol Med. 3:S155–S168. Lockey JE, LeMasters GK, Levin L, Rice C, Yiin J, Reutman S, Papes D (2005) A longitudinal study of chest radiographic changes ofworkers inthe refractoryceramic fiber industry. Chest. 121:2044–2051. Magnani C, Terracini B, Ivaldi C, Botta M, Mancini A, Andrion A (1995) Pleural malignant mesothelioma and nonoccupational exposure to asbestos in Castale Monferrato. Italy. Occup. Environ. Med. 52:362–367. Malmberg P, Hedenstein HJ, Kolmodin-Hedran B, Krantz S (1984) Pulmonary function in workers of a mineral rock fibre plant. Biological effects of man-made mineral fibres. Proceedings of a WHO/ IARC Conference, Vol. 1, April 20–22, 1982. Copenhagen, Denmark. pp. 427–435. Marsh GM, Youk AO, Stone RA, Buchanich JM, Gula MJ, Smith TJ, Quinn MM (2001a) Historical cohort study of US man-made vitreous fiber production workers: I. 1992 Fiberglass cohort followup: initial findings. J. Occup. Environ. Med. 43:741–756. Marsh GM, Buchanich JM, Youk AO (2001b) Historical cohort study of US man-made vitreous fiber production workers: VI. Respiratory system cancer standardized mortality ratios adjusted for the confounding effect of cigarette smoking. J. Occup. Environ. Med. 43:803–808. Marsh GM, Gula MJ, Youk AO, Buchanich JM, Churg A, Colby TV (2001c) Historical cohort study of US man-made vitreous fiber production workers: II. Mortality from mesothelioma. J. Occup. Environ. Med. 43:757–766. Maxim LD, McConnell EE (2005) A review of the toxicology and epidemiology of wollastonite. Inhal. Toxicol. 17:451–466. McConnell EE, Kamstrup O, Musselman R, Hesterberg TW, Chevalier J, Miller WC, Thevenaz P (1994) Chronic inhalation study of size-separated rock and slag wool insulation fibers in Fischer 344/N rats. Inhal. Toxicol. 6:571–614. McConnell EE, Axten C, Hesterberg TW, Chevalier J, Miiller WC, Everitt J, Oberd€orster G, Chase GR, Thevenaz P, Kotin P (1999) Studies of the inhalation toxicology of two fiberglasses and amosite asbestos in the Syrian golden hamster. Part II. Results of chronic exposure. Inhal. Toxicol. 11:785–835. McConnochie K, Simonato L, Mavrides P, Christofides P, Pooley FD, Wagner JC (1987) Mesothelioma in Cyprus: the role of tremolite. Thorax. 42:342–347. McDonald AD, Fry JS, Woolley GJ, McDonald JC (1984) Dust exposure and mortality in an American chrysotile asbestos friction products plant. Br. J. Ind. Med. 41:151–157. McDonald AD, Case BW, Churg A, Dufresne A, Gibbs GW, Sebastien P, McDonald JC (1997) Mesothelioma in Quebec chrysotile miners and millers: epidemiology and aetiology. Ann. Occup. Hyg. 41:707–719.

454

ASBESTOS AND OTHER MINERAL AND VITREOUS FIBERS

McDonald JC, Harris J, Armstrong B (2004) Mortality in a cohort of vermiculite miners exposed to fibrous amphibole in Libby MT. Occup. Environ. Med. 61:363–366. McDonald JC, McDonald AD (1997) Chrysotile, tremolite and carcinogenicity. Ann. Occup. Hyg. 41:699–705. Meurman LO, Kiviluoto R, Hakama M (1974) Mortality and morbidity among the working population of anthophyllite asbestos miners in Finland. Br. J. Ind. Med. 31:105–112. Meurman LO, Kiviluoto R, Hakama M (1979) Combined effects of asbestos exposure and tobacco smoking on Finnish anthophyllite miners and millers. Ann. NY Acad. Sci. 330:491–495. Miller BG, Searl A, Davis JMG, Donaldson K, Cullen RT, Bolton RE, Buchanan D, Soutar CA (1999a) Influence of fibre length, dissolution and biopersistence on the production of mesothelioma in the rat peritoneal cavity. Ann. Occup. Hyg. 43:155–166. Miller BG, Jones AD, Searl A, Buchanan D, Cullen RT, Soutar CA, Davis JMG, Donaldson K (1999b) Influence of characteristics of inhaled fibres on development of tumours in the rat lung. Ann. Occup. Hyg. 43:167–179. Moalli PA, Macdonald J, Goodglick LA, Kane AB (1987) Acute injury and regeneration of the mesothelium in response to asbestos fibers. Am. J. Pathol. 128:425–445. Monchaux G, Bignon J, Jaurand MC, Lafuma JC, Sebastien P, Masse R, Hirsh A, Goni T (1981) Mesotheliomas in rats following inoculation with acid-leached chrysotile asbestos and other mineral fibers. Carcinogenesis 2:229–236. Monchaux G, Bignon J, Hirsch A, Sebastien P (1982) Translocation of mineral fibres through the respiratory system after injection into the pleural cavity of rats. Ann. Occup. Hyg. 26:309–318. Morgan A (1979) Fiber dimensions: their significance in the deposition and clearance of inhaled fibrous dusts. In: Lemen R, Dement LM, editors. Dusts and Disease, Park Forest South. IL: Pathotox Publishers, pp. 87–96. Morgan A, Holmes A (1984) The deposition of MMMF in the respiratory tract of the rat, their subsequent clearance, solubility in vivo and protein coating. In: Biological Effects of Man-Made Mineral Fibers, Vol. 2. Copenhagen: World Health Organization, pp1–17. Morgan A, Black A, Evans N, Holme A, Pritchard JN (1980) Deposition of sized glass fibers in the respiratory tract of the rat. Ann. Occup. Hyg. 23:353–366. Morgan A, Holmes A, Davison W (1982) Clearance of sized glass fibers from the rat lung and their solubility in vivo. Ann. Occup. Hyg. 25:317–331. Morrow PE (1988) Possible mechanisms to explain dust overloading of the lungs. Fundam. Appl. Toxicol. 10:369–384. Mossman BT, Craighead JE (1987) Mechanisms of asbestos associated bronchogenic carcinoma. In: Antman K, Aisner J, editors. Asbestos-Related Malignancy. Orlando: Grune & Stratton, pp. 137–150. Mossman BT, Marsh JP, Sesko A, Hill S, Shatos MA, Doherty J, Petruska J, Adler KB, Hemenway D, Mickey R, Vacek P, Kagan E (1990a) Inhibition of lung injury, inflammation, and interstitial pulmonary fibrosis by polyethylene glycol-conjugated catalase in a rapid inhalation model of asbestosis. Am. Rev. Respir. Dis. 41:1266–1271. Mossman GT, Bignon J, Corn M, Seaton A, Gee JBL (1990b) Asbestos: scientific developments and implications for public policy. Science 247:294–301. Newhouse ML, Thompson H (1965) Epidemiology of mesothelial tumors in the London area. Ann. NY Acad. Sci. 132:579–588. Newhouse ML, Sullivan KR (1989) A mortality study of workers manufacturing friction materials: 1941–86. Br. J. Ind. Med. 46:176–179. Newhouse ML, Berry G, Skidmore JW (1982) A mortality study of workers manufacturing friction materials with chrysotile asbestos. Ann. Occup. Hyg. 26:899–909.

REFERENCES

455

Nicholson WJ (1986) Airborne Asbestos Health Assessment Update, United States Environmental Protection Agency, Office of Health and Environmental Assessment, Washington, DC, EPA/600/ 8-84/003F. Washington, DC: EPA. NRC (1984) Asbestiform Fibers—Nonoccupational Health Risks. Washington, DC: National Academy Press. Oberd€orster G, Morrow PE, Spurny K (1988) Size dependent lymphatic short term clearance of amosite fibers in the lung. Ann. Occup. Hyg. 32 (S1):149–156. Oberd€orster G, Ferin J, Finkelstein J, Soderholm S, Gelein R (1990) Mechanistic studies on particleinduced acute and chronic lung injury. In: Masuda S, Takahashi K, editors. Aerosols-Science, Industry, Health and Environment, Vol. II. Oxford: Pergamon Press, pp. 1229–1233. Pan XL, Day HW, Wang W, Beckett LA, Schenker MB (2005) Residential Proximity to naturally occurring asbestos and mesothelioma risk in California. Am. J. Resp. Crit. Care Med. 172:1019– 1025. Paustenbach DJ, Finley BL, Lu ET, Brorby GP, Sheehan PJ (2004) Environmental and occupational health hazards associated with the presence of asbestos in brake linings and pads (1990 to present): a state-of-the-art review. J. Toxicol. Environ. Health, Part B 7:25–80. Peto J, Doll R, Hermon C, Binns W, Clayton R, Goffe T (1985) Relationship of mortality to measures of environmental asbestos pollution in an asbestos textile factory. Ann. Occup. Hyg. 29:305– 355. Pinkerton KE, Elliot AA, Frame SR, Warheit DB (1997) Reversibility of fibrotic lesions in rats inhaling size-separated chrysotile asbestos fibres for 2 weeks. Ann. Occup. Hyg. 41(Suppl. 1): 178–183. Pooley FD (1976) An examination of the fibrous mineral content of asbestos lung tissue from the Canadian chrysotile mining industry. Environ. Res. 12:281–298. Pooley FD, Clark NJ (1980) A comparison of fibre dimension in chrysotile, crocidolite, and amosite particles from samples of airborne dust and from post-mortem lung tissue specimens. In: Wagner JC, editor. Biological Effects of Mineral Fibres, Vol. 1. Lyons: International Agency for Research on Cancer, pp. 79–86. Pooley FD, Wagner JC (1988) The significance of the selective retention of mineral dusts. Ann. Occup. Hyg. 32(Suppl. 1):187–194. Pott F, Friedrichs KH, Huth F (1976) Results of animal experiments concerning the carcinogenic effect of fibrous dusts and their penetration with regard to the carcinogenesis in humans. Zentralbl. Bakteriol. Hyg. 162:467–505. Reid G, Keilkowski D, Steyn SD, Botha K (1990) Mortality of an asbestos-exposed birth cohort: a pilot study. S. Afr. Med. J. 78:584–586. Rodelsperger K, Woitowitz H-J, Bruckel B, Arhelger R, Pohlaben H, Jockel K-H (1999). Doseresponse relationship between amphibole fiber lung burden and mesothelioma. Cancer Detect. Prevent. 23:183–193. Rodelsperger K, Bruckel B (2006) The carcinogenicity of WHO fibers of silicon carbide: SiC whiskers compared to cleavage fragments of granular. SiC. Inhal. Toxicol. 18:623–631. Roggli VL, Brody AR (1984) Changes in numbers and dimensions of chrysotile asbestos fibers in lungs of rats following short-term exposure. Exp. Lung Res. 7:133–147. Roggli VL, George MH, Brody AR (1987) Clearance and dimensional changes of crocidolite asbestos fibers isolated from lungs of rats following short-term exposure. Environ. Res. 42:94–105. Rood AP, Scott RM (1989) Size distributions of chrysotile asbestos in a friction products factory as determined by transmission electron microscopy. Ann. Occup. Hyg. 33:583–590. Rossiter CF, Coles RM (1980) H.M. Dockyard, Devonport: 1947 mortality study. In: Wagner JC, editor. Biological Effects of Mineral Fibres, IARC Scientific Pub. No. 30. Lyons: International Agency for Research on Cancer, pp. 713–721.

456

ASBESTOS AND OTHER MINERAL AND VITREOUS FIBERS

Saracci R, Simonato L, Acheson ED, erson A, Bertazzi PA, Claude J, Charnay N, Esteve JJ, FrentzelBeyme RR, Gardne MJ, Jensen OM, Massing R, Olsen JH, Teppo L, Westerholm P, Zocchetti C (1984) Mortality and incidence of cancer of workers in the man-made vitreous fibres producing industry: an international investigation in 13 European plants. Br. J. Ind. Med. 41:425–436. Sawyer RN (1989) Asbestos material inventory, control concepts and risk communications. In: Symposium on Health Aspects of Exposure to Asbestos in Buildings. Cambridge, MA: Harvard University Kennedy School of Government, pp. 155–169. Schlesinger RB, Lippmann M (1978) Selective particle deposition and bronchogenic carcinoma. Environ. Res. 15:424–431. Schlesinger RB, Concato J, Lippmann M (1983) Particle deposition during exhalation: a study in replicate casts of the human upper tracheobronchial tree. In: Marple VA, Liu BYH, editors. Aerosols in the Mining and Industrial Work Environments, Vol. 1. Ann Arbor: Ann Arbor Science Publishers, pp. 165–176. Scholze H, Conradt R (1987) An in vitro study of the chemical durability of siliceous fibers. Ann. Occup. Hyg. 31 (4B):683–692. Sebastien P, Bignon J, Baris YL, Awad L, Petit G (1984) Ferruginous bodies in sputum as an indication of exposure to airborne mineral fibers in the mesothelioma villages of Cappadocia. Arch. Environ. Health 39:18–23. Sebastien P, McDonald JC, McDonald AD, Case B, Harley B (1989) Respiratory cancer in chrysotile textile and mining industries: exposure inferences from lung analysis. Br. J. Ind. Med. 46:180–187. Seidman H, Selikoff IJ, Hammond EC (1979) Short term asbestos work exposure and long term observation. Ann. NY Acad. Sci. 330:61–90. Selikoff IJ, Nicholson WJ, Langer AM (1972) Asbestos air pollution. Arch. Environ. Health 25:1–13. Selikoff IJ, Hammond EC, Seidman H (1979) Mortality experience of insulation workers in the United States and Canada: 1943–1976. Ann. NY Acad. Sci. 330:91–116. Shannon H, Hayes M, Julian J, Muir D (1984) Mortality experience of glass fibre workers. In: Biological Effects of Man-Made Mineral Fibers. Copenhagen: World Health Organization, pp. 347–349. Shannon HS, Jamieson E, Julian JA, Muir DCF, Walsh C (1987) Mortality experience of Ontario glass fiber workers—extended follow-up Ann. Occup. Hyg. 31 (4B):657–662. Shukla A, Gulumian M, Hei TK, Kamp D, Rahman Q, Mossman BT (2003) Multiple roles of oxidants in the pathogenesis of asbestos-induced diseases. Free Radical Biol. Med. 34:1117– 1129. Simonato L, Fletcher AC, Cherrie J, Andersen A, Bertazzi PA, Charney N, Claude J, Dodgson J, Esteve J, Frentzel-Beyme R, Gardner MJ, Jensen O, Olsen J, Saracci R, Teppo L, Westerholm P, Winkelmann R, Winter PD, Zocchetti C (1986) Updating lung cancer mortality among a cohort of man-made mineral fibre production workers in seven European countries. Cancer Lett. 30:189– 200. Smith AH, Wright CC (1996) Chrysotile asbestos is the main cause of pleural mesothelioma. Am. J. Ind. Med. 30:252–266. Spengler JD, Ozkaynak H, McCarthy JF, Lee H (1989) Summary of Symposium on Health Aspects of Exposure to Asbestos in Buildings. In: Symposium on Health Aspects of Exposure to Asbestos in Buildings. Cambridge, MA: Harvard University Kennedy School of Government, pp. 1–26. Stanton MF, Wrench C (1972) Mechanisms of mesothelioma induction with asbestos and fibrous glass. J. Natl. Cancer Inst. 48:797–821. Stanton MF, Layard M, Tegeris A, Miller E, May M, Kent E (1977) Carcinogenicity of fibrous glass: pleural response in the rat in relation to fiber dimension. J. Natl. Cancer Inst. 58:587–603. Stayner LT, Dankovic DA, Lemon RA (1996) Occupational exposure to chrysotile asbestos and cancer risk: a review of the amphibole hypothesis. Am. J. Public Health 86:179–186.

REFERENCES

457

Stayner L, Smith R, Bailer J, Gilbert S, Steenland K, Dement J, Brown D, Lemen R (1997) Exposure– response analysis of risk of respiratory disease associated with occupational exposure to chrysotile asbestos. Occup. Environ. Med. 54:646–652. St€ ober W, Flachsbart H, Hochrainer D (1970) Der aerodynamische Durchmesser von Latexaggregaten und Asbestfasern. Staub-Reinhalt. Luft 30:277–285. Stone RA, Youk AO, Marsh GM, Buchanich JM, McHenry MB, Smith TJ (2001) Historical cohort study of US man-made vitreous fiber production: IV. Quantitative exposure–response analysis of the nested case-control study of respiratory system cancer. J. Occup. Environ. Med. 43:779–792. Stone RA, Youk AO, Marsh GM, Buchanich JM, Smith TJ (2004) Historical cohort study of U.S. man-made vitreous fiber production workers IX: summary of 1992 mortality follow up and analysis of respiratory system cancer among female workers. J. Occup. Environ. Med. 46:55–67. Su W-C, Cheng YS (2006) Fiber deposition pattern in two human respiratory tract replicas. Inhal. Toxicol. 18:749–760. Sussman RG, Cohen BS, Lippmann M (1991a) Asbestos fiber deposition in a human tracheobronchial cast. I. Exp. Inhal. Toxicol. 3:145–160. Sussman RG, Cohen BS, Lippmann M (1991b) Asbestos fiber deposition in a human tracheobronchial cast. II. Empirical model. Inhal. Toxicol. 3:161–179. Timbrell V (1972) An aerosol spectrometer and its applications. In: Mercer TT, Morrow PE, St€ober W, editors. Assessment of Airborne Particles. Springfield, IL: Charles C. Thomas, pp. 290– 330. Timbrell V (1982) Deposition and retention of fibers in the human lung. Ann. Occup. Hyg. 26:347–369. Timbrell V (1983) Fibers and carcinogenesis. J. Occup. Health Soc. 3:3–12. Timbrell V (1984) Pulmonary deposition and retention of South African amphibole fibers: identification of asbestosis-related measure of fiber concentration. In:Proceedings of VIth International Pneumoconiosis Conference, Bochum, 1983, Vol. 2. Geneva: ILO, pp. 998–1008. Timbrell V, Ashcroft T, Goldstein B, Heyworth F, Meurman LO, Rendall REG, Reynolds JA, Shilkin KB, Whitaker D (1987) Relationships between retained amphibole fibers and fibrosis in human lung tissue specimens. In: Walton WH, editor. Inhaled Particles VI. Oxford: Pergamon Press. Tran CL, Jones AD, Cullen RT, Donaldson K (1997) Overloading of clearance of particles and fibres. Ann. Occup. Hyg. 41(Suppl. 1):237–243. Trethowan WN, Burge PS, Rossiter CE, Harrington JM, Calvert IA (1995) Study of the respiratory health of employees in seven European plants that manufacture ceramic fibres. Occup. Environ. Med. 52(2):97–104. Utidjian HMD, deTreville RTP (1970) Fibrous glass manufacturing and health: report of an epidemiological study. Parts I and II. Proceedings of the 35th Annual Meeting of the Industrial Health Foundation. October 13–14, 1970, Pittsburgh, PA, IHF Bulletin No. 44, pp. 98–102. Veblen DR (1980) Anthophyllite asbestos: microstructures, intergrown sheet silicates and mechanisms of fiber formation. Am. Mineral 65:1075–1086. Vincent JH, Johnston AM, Jones AD, Bolton RE, Addison J (1985) Kinetics of deposition and clearance of inhaled mineral dusts during chronic exposure. Br. J. Ind. Med. 42:707–715. Wagner JC (1963) Asbestosis in experimental animals. Br. J. Ind. Med. 20:1–12. Wagner JC, Skidmore JW (1965) Asbestos dust deposition and retention in rats. Ann. NY Acad. Sci. 132:77–86. Wagner JC, Pooley FD (1986) Mineral fibers and mesothelioma. Thorax 41:161–166. Wagner JC, Steggs A, Marchand P (1960) Diffuse pleural mesothelioma and asbestos exposure in northwestern Cape Province. Br. J. Ind. Med. 17:260–271.

458

ASBESTOS AND OTHER MINERAL AND VITREOUS FIBERS

Wagner JC, Berry G, Skidmore JW, Timbrell V (1974) The effects of the inhalation of asbestos in rats. Br. J. Cancer 29:252–269. Wagner JC, Berry G, Skidmore JW (1976) Studies of the carcinogenic effects of fiber glass of different diameters following intrapleural innoculation in experimental animals. In: Occupational Exposure to Fibrous Glass, HEW Publ. No. (NIOSH) 76-151, pp. 193–197. Wagner JC, Pooley FD, Barry G, Seal RME, Munday DE, Morgan J, Clark NJ (1982) A pathological and mineralogical study of asbestos-related deaths in the United Kingdom. Ann. Occup. Hyg. 26:417–422. Wagner JC, Skidmore JW, Hill RJ, Griffiths DW (1985) Erionite exposure and mesotheliomas in rats. Br. J. Cancer 51:727–730. Walker AM (1984) Declining relative risks for lung cancer after cessation of asbestos exposure. J. Occup. Med. 26:422–425. Walton WH (1982) The nature, hazards and assessment of occupational exposure to airborne asbestos dust: a review. Ann. Occup. Hyg. 25:117–247. Warheit DB, Hartsky MA (1990) Species comparisons of alveolar deposition patterns of inhaled particles. Exp. Lung Res. 16:83–99. Weitzman SA, Graceffa P (1984) Asbestos catalyzes hydroxyl and superoxide radical release from hydrogen peroxide. Arch. Biochem. Biophys. 288:373–376. Whittaker EJW, Zussman J (1956) The characterization of serpentine minerals by X-ray diffraction. Min. Mag. 32:107–115. WHO(1986) Asbestos and Other Natural Mineral Fibers. Environ. Health Criteria 53. Geneva: World Health Organization. WHO(1997) Determination of Airborne Fibre Number Concentrations. Geneva: World Health Organization. p.53 Wong O, Musselman RP (1994) An epidemiological and toxicological evaluation of the carcinogenicity of man-made vitreous fiber, with a consideration of coexposures. J. Environ. Pathol. Toxicol. Oncol. 13:169–180. Wright GW, Kuschner M (1977) The influence of varying lengths of glass and asbestos fibers on tissue response in guinea pigs. In: Walton WH, editor. Inhaled Particles IV, Part 2. New York: Pergamon Press, pp. 455–472. Yada K (1967) Study of chrysotile asbestos by a high resolution electron microscope. Acta Crystallogr. 23:704–707. Zalma R, Bonneau L, Guignard J, Pezerat H (1987) Formation of oxy radicals by oxygen reduction arising from the surface activity of asbestos. Can. J. Chem. 65:2338–2341. Zoltai T (1979) Asbestiform and acicular mineral fragments. Ann. NY Acad. Sci. 330:621–643.

13 BENZENE Bernard D. Goldstein and Gisela Witz

Understanding and preventing the threat of benzene (C6H6) to human health is one of the most important environmental issues facing national and international regulatory authorities. Benzene causes human leukemia. Among the known human cancer-causing agents, benzene is the organic chemical of highest volume and broadest distribution. Further, as an integral component of our petrochemical era and a constituent of crude oil, benzene cannot simply be banned from use. Understanding the mechanisms by which benzene leads to adverse health effects is of crucial importance. Uncertainties about health effects must be balanced against the potential for substantial economic and societal costs in regulating benzene, as well as the potential adverse effects of benzene substitutes. Current standards for benzene in the United States include a maximum permissible level in drinking water of 5 ppb by weight. As for any carcinogen, the U.S. Environmental Protection Agency (EPA) has set a drinking water goal of 0 ppb. The U.S. Occupational Safety and Health Administration (OSHA) has set a workplace standard for benzene of 1.0 ppm benzene by volume as a time-weighted average for an 8 h working day. The U.S. National Institute for Occupational Safety and Health (NIOSH) has recommended a workplace air standard of 0.2 ppm. The control of benzene in ambient air by EPA is currently based on emission standards set for selected industrial sources. Under the 1990 Clean Air Act Amendments, maximum available control technology for all significant atmospheric benzene point sources is required, followed by a risk-based approach that has yet to be clearly defined. Many states have developed their own drinking water or atmospheric standards for benzene. International standards vary greatly. Benzene is the smallest and most stable aromatic compound. It is a clear colorless liquid with a classic aromatic odor. It is minimally soluble in water (820 mg/L at 22 C) and has an octanol/water partition coefficient of 1.56–2.15 (Leo et al., 1971), a blood/air partition coefficient of 7.8 (Sato and Nakajima, 1979), and a vapor pressure of 0.125 atm at 25 C (Thibodeaux, 1981). It is thus a hydrophobic solvent that readily evaporates at room temperature and rapidly partitions into lipid. Benzene reacts with hydroxyl radicals and

Environmental Toxicants, Third Edition Edited by Morton Lippmann Copyright
2009 John Wiley & Sons, Inc.

459

460

BENZENE

participates in the photochemical process leading to the formation of ozone and other components of oxidant smog. Described below are a number of aspects of the toxicology and exposure pathways of benzene pertinent to understanding how the use of benzene in our modern society leads to the risk of adverse health effects. Pertinent review articles or documents include Goldstein (1977, 1989a, 1989b), Snyder (2002, 2004), Savitz and Andrews (1997), Smith (1996), Ross (2000, 2005), Lovern et al. (2001), Bird et al. (2005), Rana and Verma (2005), and Zhang et al. (2002). Both the U.S. Agency for Toxic Substances and Disease Registry (ATSDR) and EPA have draft revisions of their comprehensive benzene documents in the review stage, with publication expected in late 2006 or 2007.

13.1 BENZENE EXPOSURE Benzene is a ubiquitous agent. As a component of petroleum, it is widely distributed. Gasoline contains 1–2% benzene in the United States, and higher levels are reported elsewhere. Benzene is also an important starting agent for chemical synthesis. It is a valuable solvent, but its use in that regard has been decreasing, primarily because of health and safety concerns. Benzene exposure occurs in the workplace, in the general environment, and through the use of consumer products. Occupational exposures present the highest risks. Cigarette smoke contains relatively high levels of benzene. For nonsmokers, benzene sources in the home are usually the major component of exposure. Benzene is also present in some foods, but relatively little of the usual total daily body burden is likely to come from this source. The World Health Organization (WHO) estimates that total daily uptake from all sources ranges from 130 to 550 mg in nonsmokers (WHO, 1987). Much of what we know about the extent of individual benzene exposure in the general environment began with a series of pioneering studies by EPA’s Office of Research and Development. They developed miniaturized sampling and analytical techniques suitable for personal monitors, demographic sampling techniques to choose individuals representative of community exposure, and, most importantly, conceptual approaches allowing for integration of indoor and outdoor exposure data for individuals in conjunction with activity questionnaires (Wallace et al., 1985; Wallace, 1987). Perhaps the most startling information concerning benzene came from studies in northern New Jersey, near a major petrochemical refinery complex (Wallace et al., 1985; Wallace, 1987). Evaluation of 355 individuals failed to reveal any statistically significant impact of proximity to the refinery on individual benzene exposure. This does not mean there was no impact; outdoor monitors confirmed the human olfactory perception of higher outdoor levels of petrochemical vapors, including benzene, in proximity to the refinery complex. However, the most notable finding was the large variability in indoor benzene levels, a variability so great that it swamped the relatively small differences caused by geographical proximity to the refinery. For those with the higher individual levels of benzene exposure, that is, those at greatest risk, the indoor sources clearly predominated. Personal exposure is usually even higher than that predicted solely by indoor levels, reflecting activity patterns near benzene sources. In general, homes in northern communities have higher indoor benzene levels, reflecting the likelihood of both attached garages and restricted outdoor ventilation in the winter. The Total Exposure Assessment Methodology (TEAM) was also the basis for study of the risk of benzene exposure resulting from emissions from the Marine Oil Terminal in Valdez, Alaska. Despite exceptionally high emissions, personal monitoring

BENZENE EXPOSURE

461

coupled with extensive meteorological and tracer studies demonstrated that the Marine Oil Terminal contributed only minimally to the benzene cancer risk of the inhabitants living in Valdez, located 5 km away. Indoor sources predominated (Goldstein, 1994). The approaches pioneered by the TEAM study have continued to be adapted and improved, including use of breath analysis as a means of assessing body benzene burden (Yu and Weisel, 1996). Benzene may also be measured in blood, and benzene metabolites can be measured in urine (Ashley et al., 1994; Medeiros et al., 1997; Rappaport et al., 2005; Wallace et al., 2004). Many new techniques have been developed, such as protein adducts (Rappaport et al., 2005) and a variety of urinary and other markers (Qu et al., 2005). These techniques should be considered as state of the art when exploring individual human exposures. In particular, regulatory decisions made about a pollutant source should no longer depend solely on source-based mathematical modeling approaches to determine whether there is sufficient risk to warrant imposing control measures or closing the facility. In addition to background levels of benzene from industrial sources in the community, the general public may be exposed to benzene in a number of ways. Some of the major ones are as follows: 1. Cigarette smoking, including passive smoking: Benzene levels are elevated in areas with significant levels of cigarette smoke, and blood benzene levels have been shown to be higher in smokers than in nonsmokers (Wallace and Pellizari, 1986). One pack a day contributes about 600 mg benzene to the smoker (WHO, 1987). 2. Home use of solvents or gasoline: Many solvents contain benzene, in almost all cases at levels less than 0.1%. However, if allowed to evaporate freely, even at 0.1% (1000 ppm) solvents can be a measurable contributor to airborne benzene levels in the home. Gasoline, which contains 1–2% benzene, is often a major source of benzene at home, particularly for those who have gasoline-fueled machinery, such as automobiles or lawnmowers in attached garages. Other consumer products that are sources of benzene include household cleaning agents, art and other hobby supplies, and glues. 3. Leaky underground storage tanks: Water supplies have become contaminated with benzene as a result of leaks from underground gasoline storage tanks. Contamination of groundwater can lead to human exposure through three routes: ingestion, inhalation, and skin absorption. Inhalation and skin absorption can occur during such activities as showering with benzene-contaminated potable water (Weisel et al., 1996). Significant risk of benzene inhalation from groundwater contamination may occur even in situations in which a municipal water supply is unaffected. This can occur by off-gassing through basement walls and floors. In one instance on Long Island in New York, which has particularly porous sandy soils, a group of over 20 homes was sufficiently affected by a leaky underground storage tank from a nearby gasoline station that the gasoline levels in the basement reached potentially explosive concentrations and the families had to be evacuated. Eventually, the decision about when it would be safe to return the families to their homes was based on arguments concerning the leukemia risk of residual air benzene concentrations. As is common in these unfortunate situations, after a period of living in a motel the families preferred to move elsewhere rather than return to their homes. 4. Automotive sources: Automobile-related emissions remain a substantial source of community exposure to benzene. Evaporation of ambient gasoline from the fuel train within the car has been largely, but still incompletely, controlled. However, release of

462

BENZENE

benzene and other gasoline vapors during refueling remains a significant source of community benzene exposure, as well as exposure to the individual doing the refueling. Benzene is also emitted during the combustion of gasoline. The catalytic converter installed on automobile exhausts is an effective means of reducing benzene emissions. Inhalation of benzene while in an automobile can be a significant portion of total daily benzene burden (Dor et al., 1995; Lawyrk and Weisel, 1996).

13.2 UPTAKE Absorption of benzene occurs through inhalation, ingestion, and across the skin. Except for unusual circumstances, inhalation of benzene is the major route of absorption. Benzene is readily absorbed in the lung, directly entering the bloodstream, where it is distributed to the tissues. Benzene within the blood is in direct equilibrium with the benzene in expired air. Thus, measurement of end alveolar breath benzene concentration is a good indicator of body benzene concentration. Approximately 50% of benzene taken up into the body by any route is eventually exhaled, the extent being dependent on benzene dose and the rate of metabolism and respiratory mechanics (i.e., assisted ventilation is effective in removing benzene from the body for treating benzene-induced acute central nervous system toxicity). Ingested benzene is also assumed to be fully absorbed. The skin is a more effective barrier. Studies of the absorption rate of liquid benzene across the skin have demonstrated significant uptake both in vitro and in vivo, with time in contact with the skin being a major factor (Franz, 1984; Loden, 1986). The extent of worker transdermal exposure to benzene is being explored using the new technique of charcoal cloth pads (Van Wendel de Joode et al., 2005). There is no evidence of transdermal absorption of benzene vapor. Much more needs to be done to understand the rate of skin absorption of benzene in liquid mixtures, such as gasoline and commercial solvents. Furthermore, it is at least theoretically possible that blends of gasoline with oxygenated fuels such as methyl tert-butyl ether or ethanol will lead to a more rapid rate of benzene absorption across the skin.

13.3 METABOLISM AND DISPOSITION Benzene is relatively inert to chemical additions, eliminations, oxidations, and reductions because it lacks substituents that can be altered and/or that confer chemical reactivity to the aromatic ring. Benzene is a nonpolar organic compound that partitions into fatty tissues. Numerous studies (Andrews et al., 1977; Sammett et al., 1979; Bolcsak and Nerland, 1983) indicate that benzene requires metabolism to reactive intermediates in order to be toxic. This discussion briefly reviews the essentials of benzene metabolism and disposition. 13.3.1

In Vivo Metabolism

The major pathways of benzene metabolism are shown in Fig. 13.1. The earliest studies of benzene metabolism in vivo reported the formation of phenol (Schultzen and Naunyn, 1867), catechol, and hydroquinone (Nencki and Giacosa, 1880). Porteous and Williams (1953) found that phenol, catechol, p-benzoquinone, and hydroquinone are excreted as ethereal sulfates in the urine of rabbits dosed orally with benzene. Earlier, Jaffe (1909) and other

METABOLISM AND DISPOSITION

463

FIGURE 13.1 Structure of benzene and adducts investigated as potential biomarkers of benzene exposure. From Medeiros et al. (1997) and Taylor and Francis (1997), used with permission.

investigators (Drummond and Finar, 1938) reported the urinary excretion of muconic acid, a ring-opened six-carbon diene dicarboxylic acid, in rabbits. In l953, Parke and Williams, using [14 C]benzene, confirmed and extended the early studies on the metabolism of benzene to ring-hydroxylated metabolites and to trans,trans-muconic acid. In rabbits administered 0.3–0.5 mL/kg [14 C]benzene by gavage, the major metabolite formed was phenol. Catechol and hydroquinone were also detected. These, along with phenol, were eliminated in the urine, mainly as the ethereal sulfate or glucuronic acid conjugates. trans,trans-Muconic acid (muconic acid) was also detected in the urine. Labeled carbon dioxide, indicating benzene ring opening, and phenylmercapturic acid were also detected. Metabolic fate studies indicated that 43% of the administered benzene dose was expired unmetabolized, 1.5% was exhaled as CO2, 35% was recovered as urinary metabolites, and 5–10% was present in feces and body tissues. Urinary metabolites consisted of 23% phenol, 4.8% hydroquinone, 2.2% catechol, and 1–2% trans,trans-muconic acid. A second ring-opened metabolite, 6-hydroxy-trans,trans-2,4-hexadienoic acid (6-hydroxyhexadienoic acid, HHA), was recently identified by Kline et al. (1993) as a urinary metabolite of benzene in mice. In studies on benzene metabolism in the isolated perfused mouse liver, Hedli et al. (1997) found a significant difference between single-pass metabolism in the orthograde (normal) and that in the retrograde (reversed) direction. Although the amount of phenol plus its conjugates produced was the same regardless of the direction of perfusion, the amount of free phenol formed expressed as a percentage of total phenolic metabolites was twice as great following normal perfusion compared with reversed perfusion, indicating regional differences in the

464

BENZENE

location of cytochrome P450 and the conjugation enzymes. Phenol conjugates and small amounts of free and conjugated hydroquinone, but no free phenol, were detected after recirculation of products formed during single-pass orthograde perfusion. These results could, in part, explain why administration of phenol does not lead to bone marrow depression. These major pathways of benzene metabolism, originally determined in rabbits, were subsequently established in rats and mice (Sabourin et al., 1988a; for reviews see Snyder, 1987; Snyder et al., 1993). In general, in vivo metabolism of benzene results in the formation of ring-closed metabolites and the ring-opened metabolites muconic acid and 6-hydroxyhexadienoic acid. The latter two compounds can be formed by metabolism of trans,trans-muconaldehyde (Witz et al., 1990; Goon et al., 1992; Zhang et al., 1993), a reactive ring-opened microsomal metabolite of benzene (Latriano et al., 1986). The metabolism to phenol and subsequent formation of phenyl glucuronide or sulfate is a detoxication pathway, as is conjugation with glutathione and subsequent formation of the prephenylmercapturic acid [S-(1,2-dihydro-2-hydroxyphenyl)-N-acetyl cysteine]. Quinol thioethers were recently reported to be present in bone marrow of mice and rats administered 11.2 mmol/kg benzene twice daily for 2 days (Bratton et al., 1997). The quinol thioethers identified consisted of 2-(glutathione-S-yl)hydroquinone [2-(GSyl)HQ], 2-(cystein-S-ylglycinyl)hydroquinone [2-(Cys-Gly)HQ], 2-(cystein-S-yl)hydroquinone [2-(Cys)HQ], and 2-(N-acetyl-cysteine-S-yl)hydroquinone [2-(NAC)HQ]. The metabolite 2-(GSyl)HQ is most likely derived by the reaction of glutathione with p-benzoquinone, a product formed by the oxidation of hydroquinone. Metabolism via the mercapturic acid pathway was demonstrated to lead to the quinol thioethers derived from 2-(GSyl)HQ. Metabolism of phenol to hydroquinone and that of benzene to muconic acid are two pathways that are currently thought to lead to the formation of toxic benzene metabolites. Hydroquinone and p-benzoquinone, easily formed through the oxidation of hydroquinone, are reactive metabolites that have been suggested to play a role in benzene toxicity (Schwartz et al., 1985; Irons, 1985; Sawahata et al., 1985; see Ross, 2005 for a recent review). Muconaldehyde, a putative precursor of urinary muconic acid, is hematotoxic in mice (Witz et al., 1985) and may be responsible, in part, for benzene toxicity. p-Benzoquinone– glutathione adduct formation leading to quinol thioethers could potentially represent yet another pathway resulting in the formation of toxic benzene metabolites. Several quinol thioether metabolites of hydroquinone have been shown to inhibit erythropoiesis in rats (Bratton et al., 1997). Potential mechanisms for their bone marrow toxicity could involve acylation of critical cellular molecules, as well as oxidative damage by reactive oxygen species generated via redox cycling (Bratton et al., 1997; Rao, 1996; Ross, 2000; Smith, 1996). 13.3.2

Mechanism(s) of Metabolite Formation

The liver and, to a lesser extent, the bone marrow and more recently the lung are the organ systems examined for benzene metabolism. Studies by Sammett et al. (1979) originally demonstrated that partial hepatectomy inhibits benzene hematotoxicity. It also decreases benzene metabolism by 70%. Coadministration of toluene, a competitive inhibitor of benzene metabolism, also reduces benzene hematotoxicity and decreases the amount of benzene metabolites excreted in the urine and found in the bone marrow, blood, liver, spleen, and fat tissue (Andrews et al., 1977). Based mainly on these studies, benzene toxicity is thought to be mediated by reactive metabolites that are formed via pathways including

METABOLISM AND DISPOSITION

465

metabolism of benzene by the liver. This hypothesis does not preclude hepatic metabolism of benzene to intermediates that travel from the liver to the bone marrow, where they could be further metabolically activated to the ultimate myelotoxic species. Many in vitro studies utilizing cellular fractions or reconstituted purified enzyme systems have been carried out in order to elucidate the mechanisms of benzene metabolite formation. Benzene is initially metabolized by a cytochrome P450-dependent monooxygenase to phenol. The formation of phenol is believed to involve the intermediate formation of benzene oxide, followed by rearrangement to phenol (Jerina and Daly, 1974), or acid-catalyzed opening of the epoxide ring, followed by aromatization via loss of a proton. A direct insertion of oxygen for aromatic hydroxylation may also account for the formation of appreciable amounts of phenol (Hanzlik et al., 1984). Urinary S-phenylmercapturic acid and N7phenylguanine are adducts presumably derived from the reaction of benzene oxide with glutathione and DNA, respectively. Identification of these adducts (Parke and Williams, 1953; Norpoth et al., 1988) as well as of S-phenylcysteine (McDonald et al., 1994) in vivo after benzene exposure provides indirect evidence for the formation of benzene oxide as the initial benzene oxidation product. Using HPLC analysis, benzene oxide was tentatively identified as such by Lovern et al. (1997) in microsomal systems metabolizing benzene. Direct evidence for the formation of benzene oxide in vivo after benzene administration comes from studies by Lindstrom et al. (1997). These investigators initially spiked blood from F344 rats with benzene oxide and, using a GC–MS method for measuring the remaining benzene oxide, found that benzene oxide has an estimated half-life of 7.9 min. This half-life is considerably longer than that of less than 2 min reported by Jerina and Daly (1974) for benzene oxide in 1 M KCl at 30 C. An even longer half-life of 34 min was recently reported by Henderson et al. (2005) for benzene oxide dissolved in phosphate buffer in D2O containing deuterated dimethylsulfoxide (95:5, v:v) at 25 C at pD 7.0 (pD ¼ pH þ 0.4). This half-life did not significantly change in the presence of 2–15 mM GSH, but did decrease at higher GSH concentrations. The major product was phenol. Addition of glutathione transferase (GST) to a reaction mixture containing 2 mM GSH also did not appreciably affect the half-life of benzene oxide. The authors concluded that capture of benzene oxide (which is in equilibrium with benzene oxepin) by GSH is an inefficient process that may account for the low levels of S-phenylmercapturic acid present in the urine after benzene exposure. Little is known, however, about the stability of benzene oxide/oxepin in hydrophobic environments, such as lipophilic portions of cell membranes and hydrophobic pockets of proteins, in which it may be sequestered and protected from reacting with glutathione. The favorable entropy in such microenvironments may actually promote reaction of benzene oxide with thiols, such as cysteine residues of albumin and hemoglobin, reactions that form the basis of a specific biomarker assay for benzene exposure. Using the GC–MS method developed for benzene oxide in rat blood, Lindstrom et al. (1997) demonstrated that benzene oxide is indeed formed in vivo in F344 rats administered 400 mg/kg benzene. In subsequent studies, Lindstrom et al. (1998) reported similar half-lives for benzene oxide incubated with blood from humans (7.2 min) and mice (6.6 min). Although not formed in microsomal preparations of rat bone marrow, benzene oxide was also found to have an estimated half-life of 6 min in bone marrow homogenates of F344 rats (Lindstrom et al., 1999). Thus, in contrast to what was previously believed, benzene oxide is a relatively stable electrophile, with a half-life long enough to be distributed via the blood stream to the bone marrow and other target tissues. The studies by Lindstrom et al. described above not only definitely demonstrated that benzene oxide is formed in vivo, but also formed the basis for estimating second-order rate constants for the reaction of benzene oxide with cysteinyl residues of hemoglobin and

466

BENZENE

albumin, which these investigators subsequently used to assess systemic doses of benzene oxide arising from benzene exposure in previously published animal and human studies. Phenol can undergo further cytochrome P450-mediated oxidation to catechol and hydroquinone. Benzene oxide can also be metabolized by epoxide hydrolase to benzenetrans-dihydrodiol, a metabolite converted to catechol by the action of a dehydrogenase. The 1,2,4-benzenetriol, observed in some in vitro metabolism systems, is believed to be derived from the oxidation of dihydroxylated metabolites by cytochrome P450. The hydroxylated aromatic benzene metabolites can undergo further metabolic conversion to form sulfate or glucuronic acid conjugates. Benzene epoxide can also serve as substrate for glutathioneS-transferase, catalyzing the formation of the prephenylmercapturic acid, which is aromatized by dehydration under acidic conditions to S-phenylmercapturic acid (Sabourin et al., 1988b). Tunek et al. (1980) identified the glutathione conjugate of p-benzoquinone in an incubation of phenol and glutathione with microsomes. This product is thought to be formed mainly nonenzymatically (Lunte and Kissinger, 1983). The cytochrome P450 isozyme primarily responsible for the initial oxidation of benzene is cytochrome P4502E1 (Johansson and Ingelman-Sundberg, 1988; Koop et al., 1989; Schrenk et al., 1992). This P450 isozyme is induced by many chemicals, including ethanol, a chemical known to enhance benzene hematotoxicity in mice when administered in the drinking water (Baarson et al., 1982). In a study on benzene metabolism in relation to cytochrome P4502E1 activity, Seaton et al. (1994) reported that measured cytochrome P4502E1 activities varied 13-fold for microsomes prepared from human liver samples, and that the fraction of benzene metabolized in 16 min ranged from 10% to 59%. A model developed by the investigators predicted the dependence of benzene metabolism on the measured cytochrome P4502E activity in liver samples from humans, rats, and mice. The authors suggested that interindividual and interspecies variations in hepatic metabolism of benzene may be related to differences in liver cytochrome P4502E1 activity. Since benzene metabolism is required for toxicity, the findings suggest that interindividual differences in P4502E1 gene expression, including differences in P4502E1 induction, as a result of alcohol consumption, for example, could play a role in determining susceptibility to benzene toxicity. The relationship between cytochrome P450 expression and benzene metabolism and toxicity is further discussed below. A reactive ring-opened metabolite of benzene trans,trans-muconaldehyde, a reactive ring-opened metabolite of benzene, was identified by Latriano et al. (1986) as a microsomal metabolite of benzene, but it has not been identified in vivo. This metabolite, a six-carbon diene dialdehyde, is unique among individual benzene metabolites in its ability to cause bone marrow depression in mice (Witz et al., 1985). Muconaldehyde was originally identified as a product formed via hydroxyl radical-mediated ring opening in aqueous solutions of benzene irradiated with X-rays (Loeff and Stein, 1959). Studies by Latriano et al. (1985) showed that muconaldehyde is formed from benzene in the presence of a hydroxyl radical-generating Fenton system. In subsequent studies, both the trans,trans- and the cis,trans-isomers of muconaldehyde were identified in Fenton mixtures incubated with benzene (Zhang et al., 1995a). Identification of cis,trans-muconaldehyde, an isomer most likely derived by rearrangement of the less stable cis,cis-muconaldehyde, suggests that cis,cis-muconaldehyde is the initial product of benzene ring opening in a Fenton system. Benzene dihydrodiol, a newly identified product derived from benzene incubated in the Fenton system, was shown to form phenol, catechol and ring-opened a,b-unsaturated aldehydes of unknown structure. These results indicate that although benzene dihydrodiol can be ring opened to a,b-unsaturated aldehydic products, it is not the precursor of muconaldehyde formed from

METABOLISM AND DISPOSITION

467

benzene in a Fenton system. Studies by Zhang et al. (1995b) suggest that Fenton chemistry plays a role in the formation of ring-opened as well as ring-hydroxylated compounds derived from benzene upon incubation with liver microsomes. Using microsomes prepared from the liver of mice treated with acetone for induction of cytochrome P4502E1, Zhang et al. showed that ring opening and ring hydroxylation were enhanced by micromolar concentrations of iron and inhibited by addition of oxy radical scavengers. Cytochrome P4502E1, the major isozyme responsible for the initial oxidation of benzene to benzene oxide, is known to produce large amounts of hydrogen peroxide (Wu and Cederbaum, 1994; Kukielka and Cederbaum, 1995). The possibility exists that benzene ring opening is mediated by reactive oxygen species generated during P4502E1 metabolism and/or involves a series of steps that include enzymatic and nonenzymatic transformations. Oxidative and reductive metabolism of muconaldehyde leads to a variety of metabolites, some of which could be important in muconaldehyde toxicity, and consequently in benzene hematotoxicity (Witz et al., 1996). Of particular interest is 6-hydroxy-trans,trans2,4-hexadienal, a reduced muconaldehyde metabolite shown to be hematotoxic in mice (Zhang et al., 1995c) and mutagenic in V79 cells (Chang et al., 1994). 6-Hydroxy-trans, trans-2,4-hexadienal is less reactive than muconaldehyde and, if formed in the liver (Grotz et al., 1994), is more likely than muconaldehyde to survive transport to the bone marrow, the target tissue of benzene. The monoreduction of muconaldehyde is reversible (Zhang et al., 1995a), a finding that could relate to the hematotoxicity observed for muconaldehyde as well as for 6-hydroxy-trans,trans-2,4-hexadienal. The role of free radicals in benzene metabolism and benzene toxicity is relatively unexplored (Subrahmanyam et al., 1991). Using a reconstituted system containing rabbit liver P450 isozyme LM2 as well as microsomes, Johansson and Ingelman-Sundberg (1983) demonstrated that phenol formation from benzene is inhibited by presumed hydroxyl radical scavengers including mannitol and dimethyl sulfoxide (DMSO) and by catalase, horseradish peroxidase, and superoxide dismutase. The authors suggested that the cytochrome P450dependent metabolism of benzene to phenol is mediated by hydroxyl radicals generated from hydrogen peroxide, thought to be formed by the spontaneous dismutation of superoxide anion radicals released by cytochrome P450. In this mechanism of phenol formation, the initial reactive intermediate is a hydroxy cyclohexadienyl radical formed by the addition of a hydroxyl radical to the benzene ring. This reaction takes place readily (Walling and Johnson, 1975), followed by phenol formation via loss of a hydrogen atom. In a subsequent study on the role of free hydroxyl radicals in the cytochrome P450-catalyzed oxidation of benzene and cyclohexanol, Gorsky and Coon (1985) concluded that, in the presence of very low (micromolar) concentrations of benzene, the hydroxyl radical-mediated formation of phenol is the dominant pathway, whereas at higher concentrations (millimolar), the direct oxidation by P450 is quantitatively of much greater importance. One implication of these studies is that a shift in metabolic pathways (free radical-mediated compared with direct enzymatic conversion) may occur, depending on the exposure dose or concentration of benzene. More recent approaches to evaluating the potential role of oxidative stress in benzene toxicity have utilized toxicogenomics (Hirabayashi, 2005). The bone marrow, the target tissue of benzene, has been reported to contain small amounts of cytochrome P450 (Andrews et al., 1979) and to metabolize benzene to only a limited extent (Irons et al., 1980). Cytochrome P4502E1 was not detected in bone marrow of B6C3F1 mice (Genter and Recio, 1994), but it is not known whether other strains of mice or other species also lack this major benzene metabolizing P450 isozyme in the bone marrow. However, the hydroxylated metabolites of benzene are present in the bone marrow after

468

BENZENE

benzene administration, and the levels of catechol and hydroquinone persist in this tissue (Rickert et al., 1979; Irons et al., 1982), suggesting a role for these metabolites in benzene toxicity. Peroxidases are present in the bone marrow, and recent studies suggest that they may play a role in the metabolism of hydroxylated benzene metabolites to reactive toxic intermediates. Using horseradish peroxidase (HRP) in the presence of hydrogen peroxide as a model for bone marrow peroxidases, Sawahata and Neal (1982) demonstrated the formation of biphenols and of p-diphenoquinone from phenol. The formation of biphenols and covalent binding of 14 C to protein was also observed during the incubation of [14 C] phenol with a rat bone marrow homogenate in the presence of hydrogen peroxide. The peroxidative oxidation of phenol and other hydroxylated benzene metabolites to reactive intermediates and the ability of hydroxylated benzene metabolites to serve as good reducing cosubstrates for peroxidases in the oxidation of benzene metabolites to reactive intermediates are well documented (Subrahmanyam and O’Brien, 1985; Smart and Zannoni, 1985; Eastmond et al., 1987; Sadler et al., 1988). The reactive toxic intermediates generated by metabolism of hydroxylated benzene metabolites by myeloperoxidase, the major peroxidase in the bone marrow, are quinones and free radical metabolite intermediates. The quinones can be detoxified by a two-electron reduction by NAD(P)H:quinone acceptor oxidoreductase (NQO1), a process that regenerates the polyhydroxylated benzene metabolites. The balance of activation of hydroxylated benzene metabolites and detoxification by NQO1 has been suggested to be a determining factor in benzene bone marrow toxicity (Ross, 2000). Results from a study by Rothman et al. (1997) support the hypothesis that high cytochrome P4502E1 activity along with low NQO1 activity are susceptibility factors for benzeneinduced hematotoxicity. The major route of human exposure to benzene is via inhalation; consequently, the lung is the first site for absorption and potential metabolism of benzene. Little is known about lung metabolism of benzene in humans compared with experimental animals. A pharmacokinetic modeling study by Sherwood and Sinclair (1999) suggested that organs other than the liver may contribute significantly to benzene metabolism, and the lung certainly is an important organ to consider. Initial studies by Chaney and Carlson (1995) showed that rat pulmonary microsomes metabolize benzene, and later studies by Powley and Carlson (1999), on species comparisons of humans, mice, rabbits, and rats, demonstrated that the rat is most similar quantitatively and qualitatively to human pulmonary microsomal metabolism of benzene incubated at 24–1000 mM concentrations. In the same studies, the rat was also found to be similar to the human in the metabolism of low concentrations of benzene (24 and 200 mM) by hepatic microsomes, while benzene metabolism to phenol by mouse microsomes is most similar to human at higher benzene concentrations (700 and 1000 M). Interestingly, hepatic, but not pulmonary, microsomes exhibited saturation of benzene metabolism, and a greater proportion of phenol was converted to hydroquinone when the benzene concentration was increased in pulmonary microsomes, while the opposite was observed with hepatic microsomes. The authors concluded that overall the rat is most similar to the human in oxidative benzene metabolism at the lower environmentally relevant benzene levels. In subsequent studies using microsomal preparations from CYP2E1 knockout and wild-type mice, Powley and Carlson (2000) showed that, in contrast to the liver where CYP2E1 was the most important isozyme accounting for 96% of total hydroxylated metabolite formation, lung CYP2E1 was responsible for only 45% of total hydroxylated metabolite formation and that CYP2F2 also contributed to the formation of oxidized benzene metabolites. Additional studies suggest that CYP2E1 is less important in the lung than the liver and show that it has a lower affinity for benzene, but a higher rate of hydroxylated metabolite formation than

METABOLISM AND DISPOSITION

469

CYP2F2, which plays a predominant role in benzene metabolism in the mouse lung. Recent studies of benzene metabolism using human bronchiolar- and alveolar-derived cell lines (Sheets and Carlson, 2004) demonstrate that benzene is metabolized by CYP2E1 and CYP2F1 that are expressed in the lung. 13.3.3

Species and Strain Differences in the Metabolism of Benzene

Chronic toxicity studies by the National Toxicology Program (NTP) (Huff, 1983) have shown that B6C3F1 mice are more sensitive to the hematotoxic and carcinogenic effects of benzene than F344/N rats. The greater sensitivity of mice compared with rats to benzene toxicity may be related to a greater metabolism of benzene to toxic intermediates, differences in detoxification of toxic metabolites, or greater inherent susceptibility of target tissues to the action of toxic metabolites. Differential toxicity as related to metabolic differences between B6C3F1 mice and F344/N rats was investigated when Sabourin et al. (1988a) quantitated water-soluble metabolites for four metabolic pathways. In animals exposed to 50 ppm benzene for 6 h, phenylsulfate, a detoxification metabolite, was present in approximately equal concentrations in rats and mice. Hydroquinone glucuronide, hydroquinone, and muconic acid, which are thought to reflect pathways leading to potential toxic metabolites of benzene, were present in much greater concentration in the mouse than in the rat. These results suggest that greater metabolism to toxic intermediates may in part explain the higher susceptibility of B6C3F1 mice to benzene-induced toxicity. Significant differences in the metabolism of benzene to urinary muconic acid have also been demonstrated between DBA/2N and C57BL/6 mice (Witz et al., 1990), two strains previously reported to exhibit differences in benzene toxicity (Longacre et al., 1981). At hematotoxic benzene doses (220–880 mg/kg), benzene-sensitive DBA/2N mice excreted significantly more muconic acid than the less-benzene-sensitive C57BL/6 mice. No differences between the two strains were observed in urinary excretion of muconic acid after muconaldehyde administration or muconic acid administration. Assuming that urinary muconic acid is derived from muconaldehyde (Kirley et al., 1989), these findings suggest that strain sensitivity toward benzene may be related to differences in the metabolism of benzene to toxic intermediates, including toxic ring-opened compounds such as muconaldehyde. In contrast to the results at hematotoxic benzene doses, at lower benzene doses (0.5–2.5 mg/kg) C57BL/6 mice excreted significantly more muconic acid than DBA/2N mice. These findings may reflect a dose-dependent (and strain-specific) shift in the metabolic pathways of benzene analogous to that found by Gorsky and Coon (1985) for the in vitro metabolism of benzene. The role of benzene metabolism, in relation to toxicity, was recently investigated by Valentine et al. (1996) in male mice lacking cytochrome P4502E1 expression (Lee et al., 1996). The CYP2E1 knockout mice and their wild-type counterparts were F3 homozygous hybrids of SV/129  C57BL/6N. After a 6 h nose-only exposure to 200 ppm benzene along with a tracer dose of [14 C]benzene, total urinary metabolites in the knockout mice were decreased to 13% compared with total urinary metabolites excreted by wild-type control mice. The amount of phenylsulfate excreted in the knockout mice constituted a significantly larger percentage of urinary total radioactivity than that in the wild-type mice, indicating a substantial role for cytochrome P4502E1 in the oxidation of phenol formed from benzene in normal mice. Toxicity studies in the knockout mice exposed by whole body inhalation to 0 or 200 ppm benzene 6 h/day for 5 days showed no effects on bone marrow cellularity and no benzene-induced genotoxicity using micronuclei formation in bone marrow and blood as an

470

BENZENE

end point. In contrast, wild-type and B6C3F1 mice exhibited severe bone marrow cytotoxicity and genotoxic effects in bone marrow and blood. These studies indicate that cytochrome P4502E1 is the major benzene-metabolizing P450 isozyme in vivo and that benzene metabolism is required for toxicity. These conclusions are supported by other in vivo studies with male and female B6C3F1 mice, which exhibited sex-dependent differences in the rate of benzene metabolism that are correlated with known differences in genotoxicity (Kreyon et al., 1996). 13.3.4

Disposition of Benzene

In humans, the half-life of benzene is in the order of 1–2 days, and essentially all absorbed benzene is gone from the body within a week following exposure. Thus, long-term bioaccumulation of benzene or its metabolites is not of concern. Ideally, in order to extrapolate results from animal experiments to humans, investigators should know the effects of dose, exposure rate, route of exposure, and species on the metabolism and disposition of benzene. Sabourin et al. (1987) found virtually 100% absorption in F344/N and Sprague–Dawley rats and B6C3F1 mice that had received oral administration of 0.5–150 mg/kg benzene. This differs from inhalation, where the percentage of benzene absorbed and retained during a 6 h exposure decreases as the exposure concentration increases. For example, at 10 ppm benzene, mice and rats absorb and retain 33% and 50%, respectively, compared with 15% and 10% respectively, at 1000 ppm benzene. At oral doses below 15 mg/kg benzene, mice and rats excreted more than 90% of the administered dose as urinary metabolites. Above 15 mg/kg, increasing amounts of benzene were exhaled unmetabolized with increasing exposure concentration, suggesting saturation of metabolism of orally administered benzene in rats and mice. For inhalation exposures, saturation of metabolic routes occurred in mice, but not in rats, at higher exposure concentrations. The results indicate that a saturating dose, if given as a bolus by gavage, is not saturating when administered by inhalation over 6 h. The effect of exposure concentration, exposure rate, and route of administration on metabolism was studied by Sabourin et al. (1989) in F344/N rats and B6C3F1 mice. Animals were exposed orally to 1, 10, and 200 mg/kg benzene and by inhalation for 6 h to 5, 50, and 600 ppm benzene vapor. In addition, animals were exposed over different time intervals to the same total amount of benzene (C  T ¼ 300 ppm h). As the exposure concentration or oral dose increased, there was a shift in metabolism from putative toxification pathways to detoxification pathways. In mice, hydroquinone glucuronide and muconic acid (markers of toxification pathways) represented a greater percentage of the administered dose at low benzene doses than was evidenced at high doses. The percentage dose excreted as the detoxification products phenylglucuronide and prephenylmercapturic acid increased with increasing dose. Similar results were obtained in the rat, except that hydroquinone glucuronide was a minor benzene metabolite at all concentrations. No simple relationship between oral dosing and inhalation was observed in terms of metabolite dose to tissues. These studies indicate that extrapolation from high-exposure toxicity studies to low-level exposures or from oral to inhalation exposures may not result in a true estimate for the parameter being extrapolated to humans. The authors concluded that if hydroquinone glucuronide and muconic acid are markers of toxic benzene metabolites and are formed in significant amounts in humans, then linear extrapolation of health effects from high exposures to low-level exposures could underestimate the toxicity of benzene. These data as well as the previous studies of Witz et al. (1990) and recent work by Rappaport et al. (2005)

MECHANISMS OF TOXICITY

471

all suggest the possibility of a hyperbolic dose–response curve for benzene hematotoxicity and carcinogenesis

13.4 MECHANISMS OF TOXICITY Remaining as major uncertainties are the nature of the toxic metabolites responsible for benzene toxicity and the mechanisms of action of these toxic metabolites. Any hypothesis of benzene toxicity must account for the role of hepatic metabolism and the selective toxicity of benzene in the bone marrow. A longstanding puzzling aspect of benzene toxicology is its lack of hepatotoxicity. Adding to this puzzle is the finding by Heijne et al. (2005) that hepatic gene expression in Fisher rats exposed to benzene is similar to that observed following exposure to known hepatotoxins. As discussed above, benzene is metabolized in the liver, mainly to phenol and minor amounts of hydroquinone and catechol. In contrast to muconaldehyde, the hydroxylated benzene metabolites administered singly do not cause bone marrow toxicity in experimental animals. Phenol and hydroquinone, which accumulate in bone marrow (Greenlee et al., 1981), have been shown to cause significant decreases in bone marrow cellularity when coadministered to mice (Eastmond et al., 1987). Since the bone marrow is rich in peroxidative enzymes, including myeloperoxidase and potential oxidants such as hydrogen peroxide derived from leukocytes, it has been suggested (Eastmond et al., 1987; Smith et al., 1989) that a bone marrow-localized phenol-dependent stimulation of hydroquinone metabolism results in the formation of benzoquinone, the ultimate toxic benzene metabolite. Benzoquinone is a direct-acting alkylating agent. It readily reacts with sulfhydryls and has been shown to inhibit microtubule assembly by blocking the thiol-sensitive GTP binding site (Irons et al., 1981). Benzoquinone also forms DNA adducts (Jowa et al., 1990), causes DNA strand breakage (Pellack-Walker and Blumer, 1986), and is genotoxic in V79 cells (Glatt et al., 1989). Thus, binding of benzoquinone to critical cellular substituents may play a role in benzene myelotoxicity. Recent work by Gaskell et al. (2005) has explored the formation of DNA adducts from benzene and the benzene metabolites para-benzoquinone and hydroquinone. The bone marrow is a complex matrix harboring stem cells, progenitor cells of blood cells, and stromal cells, which provide growth factors necessary for the proliferation and differentiation of stem and progenitor cells (Tavassoli and Friedenstein, 1983). The stromal macrophage, a regulator of hematopoiesis (Bagby, 1987), has been proposed to be a specific target of benzene (Kalf et al., 1989). In DBA/2N and C57BL/6 mice, benzene caused a dose-dependent bone marrow depression and a significant increase in bone marrow prostaglandin E levels. Both effects were prevented by coadministration of indomethacin and other inhibitors of the cyclooxygenase component of prostaglandin H synthase (PHS). Benzene, or a reactive metabolite, is hypothesized to stimulate the release of arachidonic acid, which is further metabolized to the hydroperoxide PGG2. In this mechanism of benzene toxicity, the decreased bone marrow cellularity after benzene administration is attributed to the constitutive production of high levels of prostaglandins, known to be downregulators of hematopoiesis, coupled with the genotoxic damage from reactive metabolites such as benzoquinone. Using hydroquinone or phenol as electron donors, the endoperoxidase would metabolize PGG2 to PGH2, the immediate precursor molecule for prostaglandins, with the concomitant formation of benzoquinone. Phagocytic cells have the ability to produce a variety of toxic oxygen species including superoxide anion radical, hydrogen peroxide, and hydroxyl radical. Bone marrow

472

BENZENE

macrophages and granulocytes from Balb/c mice treated with 880 mg/kg benzene were found to produce elevated levels of hydrogen peroxide on stimulation with phorbol myristate acetate compared with the same types of cells from control mice (Laskin et al., 1989). In addition to toxic oxygen species, Laskin et al. (1995) demonstrated that bone marrow leukocytes from mice administered hematotoxic doses of benzene or the metabolites hydroquinone, p-benzoquinone, and 1,2,4-benzenetriol produced increased amounts of nitric oxide (NO) in response to the inflammatory mediators lipopolysaccharide or interferon gamma. The production of NO induced by the inflammatory mediators was further enhanced by granulocyte macrophage and macrophage colony-stimulating factor, that is, growth factors present in the bone marrow required for normal cell proliferation and differentiation. The authors suggest that elevated NO production in the bone marrow may be an important mediator of benzene-induced bone marrow suppression (Laskin et al., 2000). Further supporting a role for elevated bone marrow production of NO as an important mediator of bone marrow effects is the work of Chen et al. (2004) who identified nitrobenzene, nitrobiphenyl, and nitrophenol in bone marrow of mice after i.p. administration of 400 mg/kg benzene. Nitrated benzene metabolites were either not detected in liver, lung, and blood or were present at levels significantly less than that found in the bone marrow. The authors hypothesized that peroxynitrite and other NO-derived intermediates are formed in the bone marrow via reaction of NO with oxygen or superoxide anion generated by redox cycling of hydroxylated benzene metabolites. Using 3-nitrotyrosine as a biomarker for NO-induced damage to proteins, Chen et al. (2005) demonstrated that 3-nitrotyrosine contents in bone marrow proteins increased by 1.5–4.5-fold in B6C3F1 mice administered 50–200 mg/kg benzene compared with controls. At 400 mg/kg benzene, 3-nitrotyrosine content of bone marrow proteins was significantly lower than that observed at 200 mg/kg benzene, but still significantly higher than that of controls. The authors suggest that nitration of proteins by peroxynitrite and/or by bone marrow myeloperoxidase-dependent pathways in NO metabolism may account in part for the myelotoxicity and leukemogenic effects of benzene. Nitration of tyrosine residues of specific proteins could affect signal transduction pathways and could inactivate enzymes in the bone marrow. Whether nitrated benzene metabolites play a role in benzene toxicity is at present not known. Despite many interesting hypotheses and fruitful lines of investigation, it is not now possible to identify a specific metabolic pathway leading to a specific toxic intermediate producing a specific pathogenetic mechanism resulting in either bone marrow aplasia or leukemogenesis. In fact, it has become more apparent that the effect of benzene is likely to be exerted through the action of multiple metabolites on multiple end points through multiple biological pathways (Goldstein, 1989b; Eastmond et al., 1987). An overall hypothesis for benzene-induced leukemia was proposed by Smith (1996). The key elements of this hypothesis consist of generally accepted knowledge on benzene metabolism and disposition of metabolites, cellular targets, and effects leading to changes in structure, which result in protooncogene activation and the inactivation of tumor suppressor genes. A stem cell thus affected would proliferate and develop a leukemic clone, which then develops into the disease state. Among the molecular targets of toxic metabolites suggested by Smith are tubulin, DNA, and topoisomerase II (topo II). The inhibitory effect of p-benzoquinone on tubulin formation has been known for some time (Irons et al., 1981). In addition to DNA adduct formation, which could lead to mutations and cancer discussed above, Kolachana et al. (1993) showed that mice administered benzene or its phenolic metabolites have increased levels of 8-hydroxydeoxyguanosine (8-OHdG) in bone marrow cell DNA. A study by Lagorio et al. (1994) in 65 filling station attendants in Rome, Italy,

MECHANISMS OF TOXICITY

473

showed a dose–response relationship between personal exposure to benzene and urinary concentrations of 8-OHdG. Studies by Frantz et al. (1996) and Chen and Eastmond (1995a) indicate that topoisomerase II is inhibited directly by p-benzoquinone and muconaldehyde, and by phenol and the polyhydroxylated metabolites after their peroxidase activation. Work by these groups and others has continued to explore the role of topoisomerase in benzene hematotoxicity (Lindsey et al., 2005). Topo II is involved in breaking and resealing DNA strands during DNA replication and repair and inhibition of topo II could lead to chromosome breaks and aneuploidy or cell death. It is of interest to note that the use of topo II inhibitors in chemotherapy is associated with a high risk of developing acute myeloid leukemia (Francis et al., 1994). The inhibition of topo II by benzene metabolites could contribute to the clastogenic and carcinogenic effects of benzene. Synergistic clastogenic effects have been observed in bone marrow erythrocytes of mice coadministered phenol and hydroquinone (Barale et al., 1990). The increase in micronuclei induction in the bone marrow erythrocytes appears to originate mainly from breakage in the euchromatic region of mouse chromosomes (Chen and Eastmond, 1995b). Whysner et al. (2004) and Eastmond et al. (2005) have recently reviewed the evidence concerning benzene genotoxicity and have concluded that a role for topoisomerase is likely. Elevated levels of DNA–protein cross-links (DNAPC) have been demonstrated in bone marrow cells of mice administered benzene and in HL-60 cells exposed to muconaldehyde (Schoenfeld et al., 1996). Structure–activity relationship studies of muconaldehyde and its metabolites in HL-60 cells (Schoenfeld and Witz, 2000) indicate that 6-hydroxy-trans, trans-2,4-hexadienal, the initial reduction product, and 6-oxo-trans,trans-2,4-hexadienoic acid, the initial oxidation product of muconaldehyde, also have the ability to induce DNA–protein cross-links, albeit at 5–10 times higher concentrations than muconaldehyde. DNA–protein cross-links could be involved in the hematotoxicity of 6-hydroxy-trans, trans-2,4-hexadienal, as well as that of muconaldehyde and benzene. Hydroquinone was subsequently also shown to induce DNAPC, but to a considerably lesser extent than muconaldehyde (Amin and Witz, 2001). DNAPC formation by hydroquinone could potentially be mediated by 1,4-benzoquinone, its oxidation product, or by reactive oxygen species generated during hydroquinone metabolism. Incubation of HL-60 cells with equimolar mixtures of muconaldehyde and hydroquinone resulted in higher DNAPC levels relative to those expected if the effects were additive. The induction of DNA–protein cross-links by toxicants is often associated with DNA strand breaks (Cosma et al., 1988; Yamanaka et al., 1995), suggesting that DNA–protein cross-links could contribute to the observed clastogenic effects observed after benzene exposure. Studies by Amin and Witz (2001) indicate that both muconaldehyde and hydroquinone cause concentration- and time-dependent increases in DNA single- and double-strand breaks (DNASB) and alkalilabile sites in HL-60 cells as determined by a fluorometric assay based on DNA unwinding. Induction of DNASB was additive upon treatment with equimolar mixtures of muconaldehyde and hydroquinone. 30 OH DNASB levels determined by the TUNEL assay increased significantly in HL-60 cells in a concentration-dependent manner after treatment with muconaldehyde (5–25 mM, 1 h), while hydroquinone had no effect. Cotreatment with equimolar muconaldehyde/hydroquinone mixtures resulted in significant decreases in 30 OH DNASB compared to treatment with muconaldehyde without hydroquinone. If the 30 OH DNAPC caused by muconaldehyde are an indication that apoptosis has been induced, the results suggest that a muconaldehyde/hydroquinone interaction could potentially result in survival of cells with genotoxic damage. It would be of interest to pursue this interaction in bone marrow cells and study the effects of

474

BENZENE

mixtures of these and other benzene metabolites on signal transduction pathways involved in apoptosis. In vitro studies in bone marrow cells and HL-60 cells indicate induction of apoptosis by exposure to phenolic metabolites and to muconaldehyde, suggesting that programmed cell death could be involved in benzene bone marrow cytotoxicity (Moran et al., 1996; Hiraku and Kawanishi, 1996; Schoenfeld et al., 1997). Studies analyzing bone marrow cell populations in mice administered benzene or hydroquinone (Hazel et al., 1996) and in vitro studies in model cell systems with reactive benzene metabolites (Irons and Stillman, 1996; Hedli et al., 1996; Hazel and Kalf, 1996; Kalf et al., 1996) indicate changes in cell differentiation and/or effects on signal transduction pathways, which could play a role in the mechanism of benzene-induced leukemogenesis. Similarly, a possible role for benzene metabolites, and particularly muconaldehyde, in leukemogenesis through interference with gap junction intercellular communication has been suggested (Rivedal and Witz, 2005). One particularly exciting area of benzene research can be loosely grouped under the heading of biological markers (NAS, 1989). As more is learned about the metabolism of benzene and the pathogenesis of its effects, it becomes possible to develop markers of exposure or effect suitable for assay in body fluids of laboratory animals or of humans. These should be of value in elucidating the mechanism of benzene hematotoxicity, in determining the extent of human exposure to benzene, and in establishing the appropriate dose–response curve for the effects of benzene. Substantial efforts have been made in the past 10 years toward the development of biological markers or biomarkers of benzene exposure and their application to exposed populations. Biomarkers have the advantage over external monitoring of integrating exposure from all routes and sources. Their measurement quantifies internal dose and reflects metabolism, where applicable, and disposition. Potential biomarkers of benzene exposure are the parent compound benzene, ring-hydroxylated and ringopened metabolites, glutathione-derived metabolite adducts, and DNA- and protein-derived adducts. Analytical methods for measurement of urinary biomarkers based on benzene metabolites and protein adducts based on cysteinal adducts of benzoquinone and benzene oxide were developed in the 1990s and the sensitivity and specificity of these biomarker assays have been described in several reviews (Bechtold and Henderson, 1993; Ong et al., 1995; Medeiros et al., 1997). Improvement of these methods in recent years has led to the development of less cumbersome and more sensitive assays and their application has resulted in significant progress not only in measuring benzene exposure, but also in other important aspects of benzene toxicology including susceptibility factors for benzene toxicity and correlations with toxic effects. For a detailed description, the reader is referred to articles published in a Special Issue of Chemico-Biological Interactions (Vol. 153–154, May 2005) from the Proceedings of the International Symposium on Recent Advances in Benzene Toxicity, held in Munich, Germany, October 9–12, 2004. Among the highlights of recent biomarker studies is the finding of a significant correlation of 1,4-benzoquinone (1,4-BQ) adducts of albumin (Alb) and hemoglobin (Hb) in Chinese workers exposed to benzene (Yeowell-O’Connell et al., 2001). In this study, 1,4-Alb adducts and albumin adducts of benzene oxide (BO) were found to be highly correlated with each other and with urinary phenol (PH) and hydroquinone (HQ) when compared on an individual basis. Another study in Chinese workers (Qu et al., 2003) indicated that the urinary biomarkers S-phenylmercapturic acid (S-PMA) and trans,trans-muconic acid (t,t-MA) showed significant exposure–response trends even at low benzene levels, and that S-PMA, which detects benzene exposure of about 0.1 ppm, is superior to t,t-MA, which detects benzene levels of 1.0 ppm. The hydroxylated metabolites HQ, catechol (CAT), and PH were

MECHANISMS OF TOXICITY

475

only able to detect exposures above 5 ppm. Measurement of 1,4-BQ and BO albumin adducts in 134 workers exposed to benzene (0.07–46.6 ppm; median 3.55 ppm) and 51 unexposed controls in Tianjin, China, provided strong evidence that metabolism of benzene by CYP2E1 is saturated at occupational exposures of about 1 ppm (Rappaport et al., 2002). This is based on adduct levels that were not linear and less than proportional with benzene exposure. The 1,4-BQ/BO adduct ratio was found to decrease with age, coexposure to toluene, and alcohol consumption, findings interpreted to indicate that factors affecting CYP2E1 exert a greater effect on the production of 1,4-BQ than that of BO, most likely due to the second oxidation step required for the formation of HQ. In a subsequent study (Waidyanatha et al., 2004) on benzene-exposed workers (1.65–329 ppm) and 44 controls in Shanghai, China, a similar conclusion of saturation of benzene metabolism was reached based on urinary biomarkers. For these studies, a newly developed method based on solvent extraction of 0.5 mL acidified urine, trimethylsilyl derivatization, and GC–MS analysis was used to measure PH, CAT, HQ, 1,2,4-trihydroxybenzene (triOHBz), t,t-MA, and S-PMA. There was a greater than proportional production of PH, CAT, and S-PMA and a less than proportional production of HQ, t,t-MA, and triOHBz, findings that are consistent with a competitive inhibition of PH, BO, and HQ for the same CYP2E1 enzymes. HQ, t,t-MA, and triOHBz all require more than one oxidation step for their formation, and their production is expected to be decreased under conditions of competitive inhibition of CYP2E1. The fact that CATwas grouped with PH and S-PMA was interpreted to suggest that catechol is formed in humans from the dihydrodiol obtained by ring opening of benzene oxide, and not via a second oxidation of phenol. Additional studies on cysteinyl albumin adducts of 1,4-BQ and BO as biomarkers of human benzene metabolism in Chinese workers by Rappaport et al. (2005) support the above results, indicating that adduct formation is less than proportional to benzene exposure at above about 1 ppm. The authors suggest that the biologically effective dose of 1,4-BQ and BO should be proportionally greater in persons exposed to low rather than high levels of benzene. This could have profound implications with respect to risk assessment, as discussed later in this chapter. Since the toxicity of benzene requires metabolism, factors that affect enzymes involved in the activation of benzene and its metabolites to toxic intermediates as well as enzymes involved in the deactivation of toxic metabolites may modulate benzene toxicity by changing the levels of metabolites produced. In recent years, increasing use has been made of genetically altered animal models to probe not only the role of key enzymes in the hematotoxic and genotoxic effects of benzene (Long II et al., 2002; Bauer et al., 2003) but also regulatory mechanisms in signal transduction that could be dysregulated and involved in myelotoxicity (Boley et al., 2002; Nwosu et al., 2004). In human populations, studies have focused on polymorphisms in metabolic genes in relation to biomarkers of benzene exposure and correlation with increased susceptibility to benzene toxicity. Key enzymes studied in human populations include CYP2E1 and NAD(P)H quinone:oxidoreductase (NQO1), that is, the major enzyme involved in the initial activation of benzene to benzene oxide in the liver and a major enzyme involved in the deactivation of toxic quinone metabolites in the bone marrow, respectively, as well as myeoloperoxidase (MPO), which is involved in the metabolism of phenolic metabolites to toxic quinones in the bone marrow and glutathione transferase GSTT1, which catalyzes glutathione adduct formation with benzene oxide, thereby deactivating benzene oxide, the initial reactive oxidation product. A detailed account of these studies is beyond the scope of this review, and the reader is referred to the Special Issue of Chemico-Biological Interactions (Vol. 153–154, 2005) mentioned earlier. A highlight among these studies is the finding of increased susceptibility to benzene toxicity

476

BENZENE

in Chinese workers with a polymorphism in NQO1 consisting of a C to T point mutation at position 609 in exon 6. This mutation codes for a proline-to-serine change at position 187 in the protein, resulting in instability and rapid degradation. Individuals with two null alleles do not exhibit NQO1 activity. In a study by Rothman et al. (1997), NQO1 and CYP2E1 genotypes were determined and CYP2E1 activity was estimated from the fractional excretion of hydroxychlorzoxazone. A 7.6-fold increased risk of benzene poisoning was observed in rapid chlorzoxazone metabolizers who were homozygous for the NQO1 polymorphism compared with slow chlorzoxazone metabolizers who had two wild-type NQO1 alleles. The PstI/RsaI polymorphism was found not to influence the risk of benzene poisoning. In an NQO1 mouse animal model, benzene exposure was shown to decrease apoptosis and cause myelogenous hyperplasia in the bone marrow and significant increases in peripheral blood neutrophils, eosinophils, and basophils (Long II et al., 2002). These studies support a critical role of NQO1 in myelotoxicity and the authors suggest that NQO1 null mice may potentially be useful as an animal model for studying acute leukemias and chemotherapeutic agents against these diseases. For a detailed account of the NQO1 in relation to benzene myelotoxicity, the reader is referred to a review by Ross (2005). A direct correlation between a gene polymorphism and a biomarker was reported by Qu et al. (2005) in studies on urinary biomarkers and polymorphisms of several metabolic genes in Chinese workers. Subjects with GSTT1 null alleles excreted significantly less S-PMA than those with wild-type GSTT1. The authors concluded that GSTT1 plays a critical role in determining interindividual variation of S-PMA formation from benzene, and that it is important that GSTT1 genotype be known and taken into account when S-PMA is used as a marker to estimate personal exposure levels. A future goal is continued emphasis on studies on the correlation of biomarkers with markers of susceptibility and effect for elucidation of the processes involved in benzene-induced hematotoxicity and leukemogenesis. Many of the studies of human susceptibility described above depend upon modern advances in technology, grouped under the heading of ‘omics’ that are being applied to understanding the mechanism of benzene toxicity. This topic has recently been reviewed by Smith et al. (2005). Narrowing down the genes and proteins primarily involved in benzene toxicity is a promising avenue (see, for example, studies exploring the role of the oncogenes c-MYB (Wan et al., 2005) and p53 (Yoon et al., 2003; Hirabayashi, 2005). One of the more promising experimental designs to come out of advances in the microarray analysis of gene expression is that of studying pairs of individuals: one exposed and one unexposed control. This approach is based upon the strengths of classic epidemiological case–control analyses, in which a careful match of two individuals on all but one characteristic allows exploration of the impact of exposure. Forrest et al. (2005) used this technique in their microarray analysis of peripheral blood mononuclear cell gene expression from paired benzene-exposed and control workers. RNA from these cells obtained in the field from six exposed–control pairs was subjected to study using cDNA microarrays. Realtime polymerase chain reaction technique was used to follow up on selected genes of interest of which four genes were particularly prominent in differentiating between the benzeneexposed and control subjects. In further studies in benzene-exposed workers by this group, Shen et al. (2006) reported that polymorphisms involved in genes repairing DNA doublestrand breaks increased susceptibility to benzene hematotoxicity, and Lan et al. (2005) found that SNPs could be identified from cytokine, chemokine, and cell adhesion pathways involved in hematopoiesis that could influence benzene hematotoxicity. This is an exciting area that will impact both the risk assessment of benzene and the use of toxicogenomics in risk assessment.

MECHANISMS OF TOXICITY

13.4.1

477

Hematological Effects

13.4.1.1 Pancytopenic Effects Benzene was first identified as a hematological toxicant in the nineteenth century. Experience since that time has amply confirmed the ability of benzene to destroy bone marrow precursor cells that are responsible for the production of mature circulating blood cells in humans. Similar effects are noted in the many species of laboratory animals that have been experimentally exposed to benzene. Human blood cells are generally considered to be of three types: red blood cells, whose primary function is to deliver oxygen to the tissues; white blood cells, which are involved in the body’s defenses against infection; and platelets, which participate in normal blood coagulation. Exposure to benzene affects the formation of each of these cell types. Severe benzene toxicity produces aplastic anemia, which consists of a marked decrease in the cellularity of the bone marrow and highly significant decrements in circulating red blood cells (anemia), white blood cells (leukopenia), and platelets (thrombocytopenia). Aplastic anemia is a frequently fatal disorder with death usually occurring from infection or hemorrhage. The normal bone marrow has ample reserves. For example, under certain circumstances, six times as many red blood cells as normal can be made. Thus, relatively low levels of a toxicant may decrease bone marrow reserve without causing any clinically recognizable decrease in blood counts. As toxicant levels get higher, initially one can see a decrease in blood counts within the normal range, a seemingly selective fall in one of the three blood cell types, then a mild decrease in each, known as pancytopenia, followed by full-blown aplastic anemia. Benzene produces its pancytopenic and aplastic effects through damage to precursors within the marrow by its metabolites. The earliest bone marrow precursor is a pluripotential stem cell that can mature into precursors of red blood cells (erythoblastic cell line), platelets (megakaryocytic cell line), and granulocytic white blood cells (myelocytic cell line, resulting in polymorphonuclear leukocytes, basophils, and eosinophils). The pluripotential cell appears also to be a source of lymphocytic white blood cells. Circulating lymphocytes are relatively susceptible to benzene in laboratory animals (Wiedra et al., 1981; Snyder et al., 1978; Rozen and Snyder, 1985) and in humans (Goldstein, 1988; Rothman et al., 1996). As with other aplastic agents, an increase in red blood cell mean corpuscular volume is also noted. A particularly important recent study in Chinese workers suggests that careful observation of benzene exposure levels and blood counts will show a hematotoxic effect of less than 1 ppm benzene, a long-used benzene workplace standard. Lan et al. (2004) studied 250 workers exposed to benzene and 140 controls and found that platelet counts and white blood cell counts were significantly lower in the benzene-exposed group, even for exposure below 1 ppm. There was a declining trend in other blood counts with higher exposures. The data for red blood cell mean corpuscular volume, an early sign of benzene hematotoxicity, were not presented. Of note is that the authors also evaluated the level of circulating bone marrow progenitor cells in 9 benzene-exposed and 24 control workers. In keeping with the hypothesis that the decline in circulating blood counts reflected an effect of benzene on progenitor cells, they reported a statistically significant lower level of colony-forming units (CFU) for granulocyte macrophages (CFU-GM), for the granulocyte, erythroid, macrophage, megakaryocyte (CFU-GEMM), and erythroid burst-forming units (BFU-E). Similar observations have been made in the blood and bone marrow of laboratory animals exposed to benzene. However, these findings are in contrast with the

478

BENZENE

recent report of Quitt et al. (2004), who studied 24 petroleum refinery workers exposed to 0.28–0.41 ppm benzene and 17 age-matched working relatives. They reported an increase, not a decrease, in CFU-GM and BFU-E and noted an interaction with smoking. It is important that these two very different results be reconciled. Further replication of the Lan et al. (2004) finding of a decrease in blood counts at low levels of benzene exposure would also be of value. A variety of potential mechanisms have been suggested to account for presumed differences in individual susceptibility to the pancytopenic effects of benzene, including such factors as gender, increased bone marrow turnover, and various metabolic factors (see discussion below and Goldstein, 1988). A landmark study is the observation by Rothman et al. (1997) that those individuals with an elevation in both the activity of cytochrome P4502E1, as measured by rapid fractional excretion of chlorzoxazone, and gene mutations, reflecting a reduction in activity of NQO1, were associated with a 7.6-fold greater risk of benzene-induced pancytopenia. As discussed above, CYT P4502E1 is a major activator of benzene metabolism and NQO1 is a presumed detoxifying enzyme of benzene metabolites. Either variation by itself showed a two- to threefold increase in risk. Lan et al. (2004), in the study described above, evaluated four different single-nucleotide polymorphisms of potential significance to benzene toxicity. They found that genetic variations in NAD(P)H:quinone oxidoreductase and in myeloperoxidase influenced susceptibility to benzene hematotoxicity. Other work related to toxicogenomic exploration of susceptibility to benzene hematotoxicity was described earlier in this review. These toxicogenomic studies should be extended to the risk of benzene-induced neoplasia, although the presumption is that hematotoxicity is related to risk of cancer. 13.4.1.2 Neoplastic Effects Benzene is a known cause of acute myelogenous leukemia (AML), the adult form of acute leukemia that was almost uniformly fatal until recent advances in chemotherapy. Individual cases of AML and its variants (grouped under the heading acute nonlymphocytic leukemia—ANLL) in benzene-exposed individuals were first reported about seven decades ago, but it was not until the 1970s that the causal relationship was fully accepted. Benzene exposure leads to cytogenetic abnormalities in bone marrow cells and in circulating lymphocytes (Forni et al., 1971; Zhang et al., 2002), consistent with alteration of the genome and in keeping with a somatic mutation leading to cancer. Cytogenetic changes observed in humans exposed to benzene fall into three categories: micronuclei, sister chromatid exchanges, and chromosomal aberrations. Zhang et al. (2002) reviewed the published findings and concluded that there is, thus far, unfortunately, no evidence of a unique pattern of benzene-induced chromosomal aberrations in humans. The known increased likelihood of a demonstrable cytogenetic abnormality in those cases of leukemia associated with a high level of workplace benzene exposure has been extended recently to cryptic cytogenetic abnormalities determinable in AML patients only with advanced techniques (Cuneo et al., 2002). More recent studies in benzene-exposed Chinese workers have begun to apply molecular biological techniques to better understand the chromosomal effects of benzene. For example, Rothman et al. (1995) noted that benzene induced gene duplicating but not gene inactivating mutations at the glycophorin A locus in bone marrow cells. Of note is that hematologists have long recognized that anyone with an aplastic anemia from virtually any cause has an increased risk of developing AML. Thus, radiation and alkylating agents used in chemotherapy produce both aplastic anemia and AML.

MECHANISMS OF TOXICITY

479

Unequivocal evidence of the causal relationship between benzene and AML has come through studies in a number of countries. An epidemic of aplastic anemia followed by an epidemic of AML was observed in Turkish leather workers due to the introduction of a benzene-containing glue in their workshops (Aksoy, 1985; Aksoy et al., 1971). A cohort of workers in a Pliofilm factory in Ohio has become among the most thoroughly studied in the history of occupational medicine, particularly in relation to reconstructing their past exposure history (Infante et al., 1977; Rinsky et al., 1981, 1987; Ward et al., 1996). Ten cases of myelogenous leukemia were observed, with only two expected. Studies in Chinese workers have again demonstrated the leukomogenic potential of benzene (Yin et al., 1987a, 1996a; Hayes et al., 1997, 2000, 2001). Often individuals with aplastic anemia or some degree of pancytopenia are observed to go through a preleukemic stage of varying duration before developing frank AML. The manifestations observed, including morphological abnormalities in bone marrow precursors, have been classified together as the myelodysplastic syndrome (Layton and Mufti, 1987), a syndrome that is not unique to benzene hematotoxicity. Benzene also appears to share, with radiation and with chemotherapeutic alkylating agents, a predilection for a lag period between initial exposure and the development of frank myelogenous leukemia of 5–15 years (Goldstein and Kipen, 1991). This is relatively short for other types of cancers. Although the data for benzene are less clear cut, a case can be made that it would be distinctly unusual for a benzene exposure to result in AML in less than 2 years, and there does appear to be a lessening of risk perhaps 10–15 years following cessation of exposure. Benzene has also been highly associated with other neoplasms, both hematological and nonhematological (Young, 1989; Goldstein, 1977, 1990), although the evidence is not as definitive as it is for AML. An example is multiple myeloma. This is a bone marrow tumor of plasma cells that are antibody-producing cells derived from b-lymphocytes. A few individual cases of multiple myeloma have been reported in association with benzene exposure, including four cases in the Turkish group (Aksoy, 1985; Aksoy et al., 1984). Four deaths from multiple myeloma were initially reported in the Ohio Pliofilm cohort as compared to one expected, a statistically significant observation (Rinsky et al., 1987). A causal relationship is biomedically plausible in view of the fact that benzene unequivocally can cause a bone marrow tumor, and b-lymphocytes that are precursors of plasma cells develop cytogenetic abnormalities following benzene exposure. A related question is, if multiple myeloma is causally related to benzene exposure, why is the evidence less definitive than it is for AML? This important point is also pertinent when considering the relationship between benzene and other hematological and nonhematological neoplasms. There are three main considerations. The first is the relative risk per unit of benzene exposure for AML in comparison to multiple myeloma. It is conceivable, although speculative, that a given amount of benzene is more likely to produce the somatic mutation leading to AML than that it will produce the mutation responsible for multiple myeloma; or, similarly, benzene may be more likely to lead to promotion or progression of the mutation eventually resulting in AML. Collins et al. (2003) have also suggested that peak benzene exposures may be an important factor in causation of multiple myeloma and other lymphohematopoietic cancers. The second consideration, which is not speculative, is that multiple myeloma is less common a cancer than is AML. Thus, any multiplication of its original background risk caused by benzene exposure is less likely to be recognized given the usual size of cohorts (e.g., if the background incidence of death from tumor A is 10 in 1000 and of tumor B is 1 in 1000, then the doubling of risk by a carcinogen in a large cohort recording 1000 deaths would lead to 20 deaths as compared to 10 expected for tumor A, a

480

BENZENE

statistically significant finding; however, for tumor B there would be a statistically unrecognizable two deaths as compared to one expected). The third point concerns the relatively short latency period for AML as compared to other tumors (Goldstein, 1989b, 1990). A greater delay between onset of exposure and eventual tumor complicates the likelihood of observing a causal relationship. An example of the common problem inherent in detecting a relation between benzene and hematological tumors other than AML is given by the recent study of mortality in 5514 workers occupationally to benzene in England and Wales (Sorahan et al., 2005). The standard mortality ratio for all leukemias was 137, which was not statistically significant, and 183 for ANLL, which was statistically significant at p < 0.05. For any of the three reasons described above, it is not surprising that this large and sophisticated study, which barely is able to detect an increase in ANLL, would not have the power to detect a causal effect of benzene on multiple myeloma or other hematological tumors. Similarly, Huebner et al. (2004), in a study of mortality in two petrochemical refineries, found that, in one, there was a statistically significant increase in the broad category of lymphohematopoietic cancers (SMR 1.47) and for leukemia overall (SMR 1.69), but the SMR of 1.52 for ANLL was not statistically significant and the SMR for non-Hodgkin’s lymphoma (1.47) was “near significant.” A study reputed to be a reason to discard the hypothesis of a role of benzene in causing multiple myeloma, that of Bergsagel et al. (1999), has an intrinsic flaw that makes it incapable of addressing the question. Basically, the cohort had such low benzene exposure that it would not be expected to have an increase in AML (or ANLL), the signature of benzene effect. Looking at this group for a tumor other than AML is bound to be negative, as would asking whether cigarette smoking caused multiple myeloma in a cohort whose pack-years of smoking was too low to cause a detectable increase in lung cancer (Goldstein and Shalit, 2000). The evidence supporting benzene as a cause of non-Hodgkin’s lymphoma (NHL) is even greater than it is for multiple myeloma, in part because of the observation of lymphomas in mice exposed to benzene (Snyder et al., 1980). The findings in China of at least a borderline statistically significant increase in NHL (relative risk 3.0, 99% confidence interval (CI): 0.9–10.5) is particularly important in view of the low background incidence of lymphatic tumors in China. For those with 10 or more years of benzene exposure, the relative risk for NHL was 4.2 (95% CI: 1.1–15.9) (Hayes et al., 1997). A related study in the same cohort showed a statistically significant increase in risk for NHL (Yin et al., 1996b). Perhaps the most perverse positive evidence of the role of benzene in causing NHL, as well as the difficulty posed by the healthy worker effect to interpreting worker cohort studies, is that of Wong and Raabe (2000). They reported that the SMR for NHL was 0.90 (95% CI: 0.82–0.98) in a pooled multinational cohort of over 300,000 workers, which they stated demonstrated that benzene was not a cause of NHL. As with their study of multiple myeloma (Bergsagel et al., 1999), there is a fatal flaw in relating the findings to benzene carcinogenicity in that their cohort did not have an overall increase in AML, the signature event of benzene exposure. Further, they did not consider the healthy worker effect, which is particularly prominent as NHL is a not uncommon outcome in those with HIV infection or with immune-related disorders—and such individuals are far less likely to become members of a petroleum refinery workforce. In fact, in their subanalysis, they report that the SMR for NHL in U.S. refinery workers was 0.96 (95% CI: 0.86–1.07) and non-U.S. refinery workers was 1.12 (95% CI: 0.90–1.37), while for gasoline distribution workers it was 0.64 (95% CI: 0.50–0.82). The lack of overlap in confidence intervals between the refinery workers who can be exposed to pure streams of benzene and the distribution workers who

MECHANISMS OF TOXICITY

481

have far less opportunity for benzene exposure strongly indicates a statistically significant impact of working in petroleum refineries on increasing the incidence of NHL. This impact is obscured by comparison with the total population that includes a much higher incidence of those with a propensity to develop NHL than does the worker population. Acute lymphatic leukemia (ALL), which is the usual childhood form of leukemia, is also highly likely to be caused by benzene. In addition to the evidence cited above for other lymphatic neoplasms, there is some epidemiological evidence in workers. Further, the molecular evidence of a common bone marrow precursor for lymphatic and myelocytic cells is strongly supportive. Obviously, as children are rarely exposed to high levels of benzene in the workplace, epidemiological evidence will be difficult to obtain. Steffen et al. (2004) noted a statistically significant association between childhood leukemia and residence near a gasoline station or repair garage, but not between ALL and traffic density. Recently, Schnatter et al. (2005) have reviewed the literature on the relation of leukemia subtypes to benzene exposure. The possibility that benzene might be the cause of nonhematological neoplasms, such as lung or liver cancer, has been raised by animal studies in which it has been difficult to demonstrate hematological neoplasms. A variety of solid tumors have been observed in 2-year rodent studies performed by Maltoni et al. (1983) in Italy and by the U.S. National Toxicology Program (Huff, 1983). However, there is no convincing evidence of nonhematological tumors occurring in humans as a result of benzene exposure. In predicting whether these would occur, it is of conceptual importance to determine whether the proximal carcinogen is uniquely formed within the bone marrow or whether the carcinogenic metabolites are made primarily in the liver and then travel throughout the circulation, the bone marrow being particularly susceptible because of its special dynamics but as in radiation carcinogenesis, other organs also being at risk. 13.4.2

Nonhematological Effects

13.4.2.1 Central Nervous System Benzene has an odor threshold in the range of 4– 5 ppm. The acute central nervous system toxicity of benzene is similar to that of other alkyl benzenes and has much in common with the general anesthetic effects of lipophilic solvents. Acute central nervous system toxicity appears to be a direct effect of benzene, and not related to its metabolites. Symptoms following acute inhalation include drowsiness, lightheadedness, headache, delirium, vertigo, and narcosis leading to loss of consciousness. Benzene levels at which acute central nervous system effects become noticeable are at least above 100 ppm and are more likely to be in the few hundred or few thousand parts per million range. In view of the leukemogenicity of benzene, controlled human exposure studies of such effects are not appropriate. On structure–activity grounds, benzene might be expected to have acute central nervous system effects similar to toluene, albeit at a slightly lower dose for benzene. Chronic nervous system effects of benzene have not been unequivocally demonstrated. There is much debate about whether chronic nervous system effects may occur with alkyl benzenes, particularly in work groups such as commercial painters, and, if proven, such findings might conceivably be pertinent to chronic benzene exposure. However, the exposure of painters to benzene is now at a far lower level than in the past—the predominant solvent exposures being to toluene and other alkyl benzenes. 13.4.2.2 Reproductive and Developmental Effects Keller and Snyder (1986, 1988) have shown hematological effects in the developing fetus at relatively low maternal exposure

482

BENZENE

concentration. However, to date, animal studies have generally been negative, and there is no human evidence for reproductive or developmental toxicity from benzene despite substantial exposure of women of reproductive age. Studies evaluating the possible link of childhood leukemia to parental occupation have shown no clear pattern (Ali et al., 2004). 13.4.2.3 Effects on the Immune System There is no question that high levels of benzene produce effects on lymphocytes in laboratory animals exposed to benzene, and that these lymphocytopenic effects can result in deficits in immune function (Irons 1985). Immune function decrements can occur in the absence of lymphocytopenic effects (Rosenthal and Snyder, 1985, 1987). However, there is currently no evidence whatsoever to suggest that immune function is affected following exposure of humans to allowable workplace levels of benzene or to the much lower levels present in the general environment.

13.5 RISK ASSESSMENT Risk assessment for benzene remains controversial. As compared to many other carcinogens, there has long been comparatively good agreement about the risk-specific dose of benzene (Goldstein, 1985), yet the economic and societal stakes are so high as to make relatively small differences in the interpretation of great significance to industrial stakeholders. Zeise and McDonald (2000) reviewed some of the risk assessment issues considered by the state of California, and the use of nontumor data for the cancer risk assessment of benzene has been reviewed by Albertini et al. (2003). The four major components of risk assessment are hazard identification, dose–response estimation, exposure assessment, and risk characterization. Particularly controversial has been the exposure assessment for worker cohorts in which there has been epidemiological identification of an increase in leukemia risk. This will continue, particularly in relation to exposure of Chinese workers. In addition, the possibility of a hyperbolic shape of the dose–response curve will be debated, as will the implications of toxicogenomic data relevant to human susceptibility to benzene. 13.5.1

Hazard Identification

In terms of hazard identification, there is no question that benzene is a known cause of hematological cancers, particularly acute myelogenous leukemia, and of bone marrow damage leading to pancytopenia and aplastic anemia. Benzene unequivocally causes AML and its variants, grouped under the heading of ANLL. As discussed above, hazard identification of benzene as causal for certain other hematological cancers varies from likely to almost certain. High levels of benzene, like other hydrocarbon solvents, can produce acute central nervous system toxicity and are a legitimate workplace concern—particularly in enclosed workspaces. Other effects, such as chronic CNS toxicity and fetal abnormalities, are of theoretical concern, but there is no direct evidence to support these hazards in humans. 13.5.2

Dose–Response Estimation

13.5.2.1 Carcinogenicity of Benzene Dose–response estimation for benzene, as for other carcinogens, has generated controversy. Points of contention include the use of the standard EPA linearized multistage model, which in essence assumes that every

RISK ASSESSMENT

483

single molecule of benzene has a finite chance of producing the somatic mutation responsible for acute myelogenous leukemia. Industry at times has argued that aplastic anemia is a necessary precursor of AML, based on observations of individuals with aplastic anemia who evolved through a preleukemic phase to AML, on the argument that many of the reported cases of benzene-associated AML have occurred in individuals with relatively high levels of exposure, and on the knowledge that aplastic anemia from seemingly any cause seems to predispose toward AML. There is clearly a threshold for aplastic anemia; that is, there is a level of benzene exposure below which there is no apparent risk of aplastic anemia. Accordingly, if aplastic anemia is a necessary precursor of AML, then there must also be a threshold for benzene leukemogenesis, and this threshold must be well beyond any exposure level of environmental concern. However, the supporting evidence for such a threshold for AML is very weak. In terms of case reports, there have been numerous cases of benzene-associated AML in which there was no evidence of antecedent aplastic anemia, although some degree of pancytopenia cannot be ruled out because of lack of preceding blood counts. In addition, a mechanism that depended solely on some leukemogenic process occurring in response to aplastic anemia would have difficulty explaining the cytogenic abnormalities that occur at levels of benzene that do not produce a frank pancytopenia (Yardley-Jones et al., 1988; Smith, 1996). The slope of the dose–response curve for benzene leukemogenesis undoubtedly depends on the metabolism of benzene to proximal carcinogen(s). As described in detail above, the complex metabolic pathways leading to active species are in the process of being unraveled. Such information should be of great value in establishing the appropriate dose–response relationship for benzene leukemogenesis. Although it is difficult to overstate the importance of understanding benzene metabolism in relation to benzene leukemogenesis, it must also be emphasized that determining the kinetic relationship between benzene exposure and the formation of active species is not sufficient by itself to establish the dose–response relationship. One must also take into account the responsiveness of the target organ, which, through its own natural variation in sensitivity or in defense mechanisms, may play a role in determining the dose–response pattern. For example, Sabourin et al. (1990) and Witz et al. (1990) have shown that the percentage of benzene metabolized to the ring-opened form, as measured by muconic acid, is inversely proportional to the exposure level. This suggests that the dose pattern of concern might be chronic low-level exposure rather than the equivalent dose given as a short-term spike, and that there may be a hyperbolic dose–response curve (i.e., per unit dose there is greater potency at the lower end of the dose–response curve). Recent studies by Rappaport et al. (2005) lend credence to this possibility. However, Witz et al. (1985) have also shown that the bone marrow toxicity of trans,trans-muconaldehyde is greater with a single daily dose than it is with the same dose divided into three daily injections; that is, the short-term spike of the toxic metabolite is more harmful to the target organ. Integrating physiologically based pharmacokinetics for benzene metabolites with the dose pattern responsiveness of the bone marrow will present an intriguing challenge to those interested in modeling dose– response patterns. The dose–response estimation for benzene leukemogenesis using currently standard EPA models has been derived from the leukemia incidence in studies of occupationally exposed cohorts. The study that finally convinced regulators that benzene was a human leukemogen, that of Infante et al. (1977) on Pliofilm workers, was a major basis for the original EPA dose–response estimation of the leukemogenic potential of benzene. A highly controversial retrospective analysis of dose was required (see below). More recently epidemiological

484

BENZENE

studies of worker cohorts have generally paid more attention to the dose side of the dose– response analysis. Three groups of studies have been particularly notable for reporting an increase in leukemia risk at lower levels of benzene than previously reported or expected: studies in China by a group that includes the Chinese Academy of Preventive Medicine; the U.S. National Cancer Institute and academics at a variety of U.S. institutions, most notably the University of California at Berkeley School of Public Health and New York University; the second is from a group in France investigating workers at their electric power industry, and the third from a group evaluating workers at the Australian petrochemical industry. The latter two studies are relatively easy to discount in terms of their having any impact on the dose–response analysis for benzene leukemogenesis. However, the multiple ongoing studies in China will certainly have an impact on considerations of benzene cancer risk. The studies in China have been reviewed by Hayes et al. (2000, 2001). The rapid industrialization of China has taken a toll in worker health, and the adverse health impact of the inappropriate use of benzene is just one example. Not unexpectedly, an increase in hematological tumors in benzene-exposed Chinese workers has been documented (Yin et al., 1987a, 1996a; Hayes et al., 1997, 2000, 2001). However, in terms of the risk assessment of benzene, evaluation of the benzene exposure of these Chinese workers led to the observation of a higher relative risk per unit benzene than anticipated based upon the Pliofilm and other cohorts of benzene-exposed workers. Questions were immediately raised about whether there had been an underestimate of actual exposures. As described below (see Section 13.5.3), this has led to a more thorough evaluation of the benzene-exposed workforce in China with residual controversy. Among the surprising findings of leukemia seemingly associated with low-level benzene exposure is that of Guenel et al. (2002) funded by the power industry in France, which heavily relies on nuclear power. An increase in leukemia risk in a cohort of workers in this industry was tentatively ascribed by the authors to benzene exposure rather than to the more likely exposure to radiation, a known leukemogen. The authors estimated that mean benzene TWA exposure for exposed workers was 0.16 ppm, but this estimate was made despite an apparently complete lack of any actual benzene measurements. As pointed out by the authors, the observed risk may be related to other occupational factors. A particularly thorough but partially flawed nested case–control study of incidence of disease in refinery workers has been performed in Australia by Glass et al. (2003). While again demonstrating an increased risk for AML in those refinery workers most heavily exposed to benzene, the study is marred by a likely spurious finding due to surveillance bias of an increase in chronic lymphatic leukemia (CLL) in this workforce at particularly low levels of benzene (Goldstein, 2004). In contrast to AML, which almost always has a rapid onset and progression, CLL is a disease that develops relatively slowly. Not uncommonly, diagnosis of CLL results from an unexpected finding of lymphocytosis in a routine blood count in an otherwise healthy individual. In such cases, average life expectancy is more than 12 years. Obviously, the more frequently routine blood counts are performed, the more likely the diagnosis will be made. Glass et al. (2003) studied the entire refinery workforce, including those with negligible benzene exposure (e.g., office staff). Refinery workers exposed to benzene routinely have blood counts as part of standard workplace surveillance. Accordingly, those with CLL in the benzene-exposed part of the workforce would have been discovered relatively early in their long disease process, while those not subjected to hematological surveillance could have an undiagnosed case of early CLL that would not be detected. Because of the high likelihood of a surveillance bias, this study cannot be used as evidence that benzene causes CLL, nor can the CLL cases in the study be incorporated into

RISK ASSESSMENT

485

any extrapolation used for risk assessment purposes. The latter point is particularly important as the authors report an excess total leukemia incidence in association with cumulative lifetime benzene exposure levels as low as >2–4 ppm-years. However, in their conditional logistic regression analysis of the association of leukemia subtype with the lowest level of cumulative benzene exposure in this particular analysis, >4–8 ppm-years, the OR was only 0.52 (0.05–5.0) for ANLL. In contrast, at these low benzene levels the OR for CLL was 2.76 (0.42–18.1), again in keeping with surveillance bias accounting for the findings. 13.5.2.2 Noncarcinogenic End Points There seems to be no reasonable likelihood that aplastic anemia can be caused at the levels of exposure reported in the general environment. The threshold for aplastic anemia as a potentially crippling or fatal disease, even for benzenesensitive individuals, is likely to be significantly above the 10 ppm TWA (time-weighted average) benzene standard that was the U.S. workplace norm for many years. As discussed above, the recent report of observable differences in blood counts in Chinese workers exposed to less than 1 ppm benzene (Lan et al., 2004) will increase interest in estimating the risk for the noncancer effects of low levels of benzene. There are three issues that need to be addressed: the need for replication of these hematological findings in other closely observed worker groups, including resolving the discrepancy between the Lan et al. (2004) and Quitt et al. (2004) findings relating to precursor cells (see above); an understanding of the extent to which a small decrement of a blood count well within the normal range would be considered an adverse effect for risk assessment purposes; and the appropriate use of uncertainty (safety) factors for such an end point. The use of large safety factors resulting in a sub-ppm benzene standard to protect against noncancer hematological effects is open to question. Starting with animal data and then using multiple safety factors might be relevant for a new chemical, but is simply inappropriate given the literally millions of person-years of data available for benzene. An example of the misapplication of this approach occurred a few years ago in New Jersey, where the use of a 10 ppb indoor “acute action level” in a situation with gasoline-contaminated groundwater led to evacuation of homeowners and widespread community concern. This standard was based upon use of a 100-fold safety factor on a 1 ppm standard. The families were moved to a crossroads motel that likely contained higher indoor benzene levels than their homes. The level of 10 ppb is not uncommon in American basements. In view of the lack of positive data, no risk levels for fetal abnormalities or chronic nervous system effects from benzene are currently indicated. 13.5.3

Exposure Assessment

Exposure assessment for benzene presents some current areas of intense controversy, both in terms of retrospective estimates of exposure of benzene-exposed cohorts with elevated leukemia incidence and in terms of the appropriate approach to exposure estimation from known benzene sources in a community. Perhaps the most thorough retrospective evaluation of benzene exposure, or of any occupationally exposed cohort, has been that performed by Rinsky et al. (1987) on the employees of a Pliofilm plant whose benzene exposure resulted in a substantial risk of leukemia. The authors did a masterful job of reconstructing the location of each of the workers within the workplace and possible exposure levels. However, Rinsky et al. (1987) ended up with much lower exposure levels, particularly during the Second World War,

486

BENZENE

when most of the leukemia cases were at work, than did Crump and Allen (1984), who built their exposure estimate on the basis of the Rinsky et al. efforts. The major distinction between the two approaches is the much higher levels posited by Crump and Allen during World War II. The latter appears to be strongly supported by contemporary accounts of significant incidences of aplastic anemia and of central nervous system effects at Pliofilm factories as a result of wartime conditions, which would also expect to result in longer working hours and thus greater individual exposures. In addition, Kipen et al. (1988, 1989a) found that the historic blood counts of workers in this cohort had a statistically significant inverse correlation with the Crump and Allen exposure assessment but no relationship with the estimate of Rinsky et al. (see also Hornung et al., 1989; Kipen et al., 1989b; Paxton et al., 1994a, 1994b; Schnatter et al., 1996). Because almost all of the workers with AML had been in this cohort in the early 1940s, it appears that reliance on the Rinsky et al. exposure estimation to perform a risk assessment may lead to an overestimate of the risk. Because of its relevance to risk assessment, the reassessment of the exposure of this cohort continues (Williams and Paustenbach, 2003). Similar to the Pliofilm cohort, it can be anticipated that the extent to which the recent studies of benzene-exposed workers in China will impact on the cancer risk potency for benzene will depend in large part upon the scientific acceptability of the exposure assessments (e.g., Dosemeci et al., 1994; Vermeulen et al., 2004; Waidyanatha et al., 2004), which have been questioned (Wong, 2002). 13.5.4

Risk Characterization

The EPA has characterized the risk of benzene-induced leukemia as being 8  106 (mg/m3 benzene) (ppb) per 70-year lifetime. This can be translated into a risk of 8 in 1 million of dying from benzene-induced leukemia from breathing 1 mg/m3 benzene continually for 70 years. This risk is usually described as being the plausible upper boundary in that the conservative assumptions built into risk assessment are thought to make it unlikely that the risk is any higher, but it is conceivable that it is much lower. The level of benzene responsible for a one in 1 million lifetime risk, for drinking water, has been similarly calculated by EPA to be 0.66 ppb.

REFERENCES Aksoy M (1985) Benzene as a leukemogenic and carcinogenic agent. Am. J. Ind. Med. 8:9–20. Aksoy M, Dincol K, Akgun T, Erdem S, Dincol G (1971) Haematological effects of chronic benzene poisoning in 217 workers. Br. J. Ind. Med. 28:296–302. Aksoy M, Erdem S, Dincol G, Kutlar A, Bakioglu I, Hepyuksel T (1984) Clinical observations showing the role of some factors in the etiology of multiple myeloma. Acta Haematol. 71:116–120. Albertini R, Clewell H, Himmelstein MW, Morinello E, Olin S, Preston J, Scarano L, Smith MT, Swenberg J, Tice R, Travis C (2003) The use of non-tumor data in cancer risk assessment: reflections on butadiene, vinyl chloride, and benzene. Regul. Toxicol. Pharmacol. 37:105–132. Ali R, Yu CL, Wu MT, Ho CK, Pan BJ, Smith T, Christiani DC (2004) A case-control study of parental occupation, leukemia and brain tumors in an industrial city in Taiwan. J. Occup. Environ. Med. 46:985–992. Amin R, Witz G (2001) DNA–protein crosslink and DNA strand break formation in HL-60 cells treated with trans,trans-muconaldehyde, hydroquinone and their mixtures. Int. J. Toxicol. 20:69–80.

REFERENCES

487

Andrews LS, Lee EW, Witmer CM, Kocsis JJ, Snyder R (1977) Effects of toluene on the metabolism, disposition, and hemopoietic toxicity of 3 H-benzene. Biochem. Pharmacol. 26:293–300. Andrews LS, Lee EW, Witmer CM, Kocsis JJ (1979) 3 H-benzene metabolism in rabbit bone marrow. Life Sci. 25:567–572. Ashley DL, Bonin MA, Cardinali FL, McCraw JM, Wooten JV (1994) Blood concentrations of volatile organic compounds in a nonoccupationally exposed US population and in groups with suspected exposure. Clin. Chem. 40:1401–1404. Baarson KA, Snyder CA, Green JD, Sellakumar A, Goldstein BD, Albert RE (1982) The hematotoxic effects of inhaled benzene on peripheral blood, bone marrow, and spleen cells are increased by ingested ethanol. Toxicol. Appl. Pharmacol. 64:393–404. Bagby GC (1987) Production of multilineage growth factors by hematopoietic stromal cells: an intercellular regulatory network involving mononuclear phagocytes and interleukin-1. Blood Cells 13:147–159. Barale R, Marrazzini A, Betti C, Vangelisti V, Loprieno N, Barrai I (1990) Genotoxicity of two metabolites of benzene: phenol and hydroquinone show strong synergistic effects in vivo. Mutat. Res. 244:15–20. Bauer AK, Faiola B, Abernethy DJ, Marchan M, Pluta LJ, Wong VA, Roberts K, Jaiswal AK, Gonzalez FJ, Butterworth BE, Borghoff S, Parkinson H, Everitt J, Recio L (2003) Genetic susceptibility to benzene-induced toxicity: role of NADPH:quinone oxidoreductase-1. Cancer Res. 63:929–935. Bechtold WE, Henderson RF (1993) Biomarkers of human exposure to benzene. J. Toxicol. Environ. Health 40:377–386. Bergsagel DE, Wong O, Bergsagel PL, Alexanian R, Anderson K, Kyle RA, Raabe GK (1999) Benzene and multiple myeloma: appraisal of the scientific evidence. Blood 94:1174–1182. Bird MG, Greim H, Snyder R, Rice JM (2005) Recent advances in benzene toxicity. Chem. Biol. Interact. 153–154:1–5. Bolcsak LE, Nerland DE (1983) Inhibition of erythropoiesis by benzene and benzene metabolites. Toxicol. Appl. Pharmacol. 69:363–368. Boley S, Wong VA, French JE, Recio L (2002) p53 heterozygosity alters the mRNA expression of p53 target genes in the bone marrow in response to inhaled benzene. Toxicol. Sci. 66:209–215. Bratton SB, Lau SS, Monks TJ (1997) Identification of quinol thioethers in bone marrow of hydroquinone/phenol-treated rats and mice and their potential role in benzene-mediated hematotoxicity. Chem. Res. Toxicol. 10:859–865. Chang RL, Wang CQ, Kline SA, Conney AH, Goldstein BD, Witz G (1994) Mutagenicity of trans, trans-muconaldehyde and its metabolites in V79 cells. Environ. Mol. Mutagen. 24:112–115. Chaney AM, Carlson GP (1995) Comparison of rat hepatic and pulmonary microsomal metabolism of benzene and the lack of benzene-induced pneumotoxicity and hepatotoxicity. Toxicology 104:53–62. Chen H, Eastmond DA (1995a) Topoisomerase inhibition by phenolic metabolites: a potential mechanism for benzene’s clastogenic effects. Carcinogenesis 16:2301–2307. Chen H, Eastmond DA (1995b) Synergistic increase in chromosomal breakage within the euchromatin induced by an interaction of the benzene metabolites phenol and hydroquinone in mice. Carcinogenesis 16:1963–1969. Chen K-M, El-Bayoumy K, Cunningham J, Aliaga C, Li H, Melikian AA (2004) Detection of nitrated benzene metabolites in bone marrow of B6C3F1 mice treated with benzene. Chem. Res. Toxicol. 17:370–377. Chen K-M, El-Bayoumy K, Hosey J, Cunningham J, Aliaga C, Melikian AA (2005) Benzene increases protein-bound 3-nitrotyrosine in bone marrow of B6C3F1 mice. Chem. Biol. Interact. 156:81–91.

488

BENZENE

Collins JJ, Ireland B, Buckley CF, Shepperly D (2003) Lymphohematopoietic cancer mortality among workers with benzene exposure. Occup. Environ. Med. 60:676–679. Cosma GN, Jamasbi R, Marchok AC (1988) Growth inhibition and DNA damage induced by benzo[a]pyrene and formaldehyde in primary cultures of rat tracheal epithelial cells. Mutat. Res. 201:161–168. Crump K, Allen B (1984) Quantitative Estimates of Risk of Leukemia from Occupational Exposure to Benzene. OSHA Docket H-059B. Washington, DC: OSHA. Cuneo A, Bigoni R, Cavazzini F, Bardi A, Agostini P, Tammiso E, Ciccone N, Mancini M, Nanni M, De Cula R, Divona M, La Starza R, Crescenzi B, Testoni N, Rege Cambrin G, Mecuci C, Lo Coco F, Saglio G, Castoldi G (2002) Incidence and significance of cryptic chromosome aberrations detected by fluorescence in situ hybridization in acute myeloid leukemia with normal karyotype. Leukemia 16:1745–1751. Dor F, LeMoullec Y, Festy B (1995) Exposure of city residents to carbon monoxide and monocyclic aromatic hydrocarbons during commuting trips in the Paris metropolitan area. J. Air Waste Manage. Assoc. 45: 2103–110. Dosemeci M, Li G-L, Hayes RB, Yin S-N, Linet M, Chow W-H, Wang YZ, Jiang Z-L, Dai T-R, Zhang W-U, Chao X-J, Ye P-Z, Kou Q-R, Fam Y-H, Zhang X-C, Lin X-F, Meng J-F, Zho J-S, Wacholder S, Kneller R, Blot WJ (1994) Cohort study among workers exposed to benzene in China. II. Exposure assessment. Am. J. Ind. Med. 26:401–411. Drummond JC, Finar IL (1938) Muconic acid as a metabolic product of benzene. Biochemistry 32:79–84. Eastmond DA, Smith MT, Irons RD (1987) An interaction of benzene metabolites reproduces the myelotoxicity observed with benzene exposure. Toxicol. Appl. Pharmacol. 91:85–95. Eastmond DA, Mondrala ST, Hasegawa L (2005) Topoisomerase II inhibition by myeloperoxidaseactivated hydroquinone: a potential mechanism underlying the genotoxic and carcinogenic effects of benzene. Chem. Biol. Interact. 153–154:207–216. Forni AM, Cappellini A, Pacifico E, Vigliani EC (1971) Chromosome changes and their evolution in subjects with exposure to benzene. Arch. Environ. Health 23:385–391. Forrest MS, Lan Q, Hubbard AE, Zhang L, Vermeulen R, Zhao X, Li G, Wu YY, Shen M, Yin S, Chanock SJ, Rothman N, Smith MT (2005) Discovery of novel biomarkers by microarray analysis of peripheral blood mononuclear cell gene expression in benzene-exposed workers. Environ. Health Perspect. 113:801–807. Francis GE, Tejedor MC, Berney JJ, Chresta CM, Delgado C, Patel P (1994) Effects of DNA topoisomerase II inhibitors on human bone marrow progenitor cells. Leukemia 8(1):121–128. Frantz CE, Chen H, Eastmond DA (1996) Inhibition of human topoisomerase II in vitro by bioactive benzene metabolites. Environ. Health Perspect. 104:1319–1323. Franz TJ (1984) Percutaneous absorption of benzene. In: MacFarland HN, et al., editors. Advances in Modern Environmental Toxicology, Vol. VI. Applied Toxicology of Petroleum Hydrocarbons. Princeton, NJ: Princeton Scientific Publishers. Gaskell M, McLuckie KIE, Farmer PB (2005) Genotoxicity of the benzene metabolites para-benzoquinone and hydroquinone. Chem. Biol. Interact. 153–154:267–270. Genter MB, Recio L (1994) Absence of detectable P4502E1 in bone marrow of B6C3F1 mice: relevance to butadiene-induced bone marrow toxicity. Fundam. Appl. Toxicol. 22:469–473. Glass DC, Gray CN, Jolley DJ, Gibbons C, Sim MR, Fritschi L, Adams GG, Bisby JA, Manuell RW (2003) Leukemia risk associated with low level benzene exposure. Epidemiology 14:569–577. Glatt H, Padykdula R, Berchtold GA, Ludewig G, Platt KL, Klein J, Oesch F (1989) Multiple activation pathways of benzene leading to products with varying genotoxic characteristics. Environ. Health Perspect. 82:81–89. Goldstein BD (1977) Hematotoxicity in humans. J. Toxicol. Environ. Health (Suppl.) 2:69–105.

REFERENCES

489

Goldstein BD (1985) Risk assessment and risk management of benzene by the Environmental Protection Agency. In: Risk Quantitation and Regulatory Policy, Banbury Report, Vol. 19. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory, pp. 293–304. Goldstein BD (1988) Benzene toxicity. In: Weaver NK, editor. Occupational Medicine State of the Art Review: The Petroleum Industry. Philadelphia, PA: Hanley and Beltus, Inc., pp. 541–554. Goldstein BD (1989a) Clinical hematotoxicity of benzene. In: BT Advances in Modern Environmental Toxicology, Vol. XVI. Benzene: Occupational and Environmental Hazards—Scientific Update. Princeton, NJ: Princeton Scientific Publishing, pp. 55–65. Goldstein BD (1989b) Occam’s razor is dull. Environ. Health Perspect. 82:3–6. Goldstein BD (1990) Is exposure to benzene a cause of human multiple myeloma? Ann. N.Y. Acad. Sci. 609:225–334. Goldstein BD (1994) Letter to the editor: response to the Valdez Air Study Review Committee. Risk Anal. 14:891–893. Goldstein BD (2004) Letter to the editor re: Glass et al. Leukemia risk associated with low-level benzene exposure. Epidemiology 15:509–510. Goldstein BD, Kipen H (1991) Lessons on the second cancers resulting from cancer chemotherapy. In: Witmer CM, editor. Biological Reactive Intermediates, Vol. IV. New York: Plenum Press. Goldstein BD, Shalit SL (2000) The causal relation between benzene exposure and multiple myeloma (Letter). Blood 95:4. Goon D, Cheng X, Ruth J, Petersen DR, Ross D (1992) Metabolism of trans,trans-muconaldehyde by aldehyde and alcohol dehydrogenase: identification of a novel metabolite. Toxicol. Appl. Pharmacol. 114:147–155. Gorsky LD, Coon MJ (1985) Evaluation of the role of free hydroxyl radicals in the cytochrome P-450 catalyzed oxidation of benzene and cyclohexanol. Drug Metab. Dispos. 13:169–174. Greenlee WF, Gross EA, Irons RD (1981) Relationship between benzene toxicity and the disposition of 14 C-labeled benzene metabolites in rat. Chem. Biol. Interact. 33:285–299. Grotz VL, Ji S, Goldstein BD, Witz G (1994) Metabolism of benzene and trans,trans-muconaldehyde in the isolated perfused liver to trans,trans-muconic acid. Toxicol. Lett. 70:281–290. Guenel P, Imbernon E, Chevalier A, Crinquand-Calastreng A, Goldberg M (2002) Leukemia in relation to occupational exposures to benzene and other agents: a case-control study nested in a cohort of gas and electric utility workers. Am. J. Ind. Med. 42(2):87–97. Hanzlik RP, Hogberg K, Dugson CM (1984) Microsomal hydroxylation of specifically deuterated monosubstituted benzenes. Evidence for direct aromatic hydroxylation. Biochemistry 23:3048– 3055. Hayes RB, Yin S-N, Dosemeci M, Li G-L, Wacholder S, Travis LB, Li C-Y, Rothman N, Hoover RN, Linet MS (1997) Benzene and the dose-related incidence of hematologic neoplasms in China. J. Natl. Cancer Inst. 89:1065–1071. Hayes RB, Yin S, Rothman N, Dosemeci MM, Li G, Travis LT, Smith MT, Linet MS (2000) Benzene and lymphohematopoietic malignancies in China. J. Toxicol. Environ. Health 61:419–432. Hayes RB, Songnian Y, Dosemeci M, Linet M (2001) Benzene and lymphohematopoietic malignancies in humans. Am. J. Ind. Med. 40:117–126. Hazel BA, Kalf GF (1996) Induction of granulocytic differentiation in myeloblasts by hydroquinone, a metabolite of benzene, involves the leukotriene D4 receptor. Receptors Signal Transduct. 6:1–12. Hazel BA, O’Connor A, Niculescu R, Kalf GF (1996) Induction of granulocytic differentiation in a mouse model by benzene and hydroquinone. Environ. Health Perspect. 104:1257–1264. Hedli CC, Rao NR, Reuhl KR, Witmer CM, Snyder R (1996) Effects of benzene metabolite treatment on granulocytic differentiation and DNA adduct formation in HL-60 cells. Arch. Toxicol. 70:135–144.

490

BENZENE

Hedli CC, Hoffmann MJ, Ji S, Thomas PE, Snyder R (1997) Benzene metabolism in the isolated perfused mouse liver. Toxicol. Appl. Pharmacol. 146:60–68. Heijne WHM, Jonker D, Stierum RH, van Ommen B, Groten JP (2005) Toxicogenomic analysis of gene expression changes in rat liver after a 28-day oral benzene exposure. Mutat. Res./Fundam. Mol. Mech. Mutagen. 575:85–101. Henderson AP, Barnes L, Bleasdale C (2005) Reactions of benzene oxide with thiols including glutathione. Chem. Res. Toxicol. 18:265–270. Hirabayashi Y (2005) p53-dependent gene profiling for reactive oxygen species after benzene inhalation: special reference to genes associated with cell cycle regulation. Chem. Biol. Interact. 153–154:165–170. Hiraku Y, Kawanishi S (1996) Oxidative DNA damage and apoptosis induced by benzene metabolites. Cancer Res. 56:5172–5178. Hornung RW, Ward E, Morris JA, Rinsky RA (1989) Letters to the editor. Toxicol. Ind. Health 5:1153–1155. Huebner WW, Wojcik NC, Rosamilia K, Jorgensen G, Milano CA (2004) Mortality updates (1970–1997) of two refinery/petrochemical plant cohorts at Baton Rouge, Louisiana and Baytown. Texas J. Occup. Environ. Med. 46:1229–1245. Huff J (1983) National Toxicology Program. Technical Report on the Toxicology and Carcinogenesis Studies of Benzene (CAS No. 71-43-2) in F344/N Rats and B6C3F1 Mice (Gavage Studies). Research Triangle Park, NC: National Institutes of Environmental Health Sciences. Infante PF, Rinsky RA, Waggoner JK, Young RJ (1977) Leukemia in benzene workers. Lancet 2:76–78. Irons RD (1985) Quinones as toxic metabolites of benzene. J. Toxicol. Environ. Health 16:673–678. Irons RD, Stillman WS (1996) Impact of benzene metabolites on differentiation of bone marrow progenitor cells. Environ. Health Perspect. 104:1247–1250. Irons RD, Dent JG, Baker TS, Rickert DE (1980) Benzene is metabolized and covalently bound in bone marrow in situ. Chem. Biol. Interact. 30:241–245. Irons RD, Neptun DA, Pfeifer RW (1981) Inhibition of lymphocyte transformation and microtubule assembly by quinone metabolites of benzene: evidence for a common mechanism. J. Reticuloendothel. Soc. 30:359–371. Irons RD, Greenlee WF, Wierda D, Bus JS (1982) Relationship between benzene metabolism and toxicity: a proposed mechanism for the formation of reactive intermediates from polyphenol metabolites. Adv. Exp. Med. Biol. 136:229–243. Jaffe M (1909) Uber die Aufspaltung des Benzolring im Organismus. I. Mitteilung. Das Auftreten von Muconsaure im Harn nach Darreichung von Benzol. Hoppe-Seylers Z. Physiol. Chem. 62:58–67. Jerina DM, Daly JW (1974) Arene oxides: a new aspect of drug metabolism. Science 185:573–582. Johansson I, Ingelman-Sundberg M (1983) Hydroxyl radical-mediated, cytochrome P450-dependent metabolic activation of benzene in microsomes and reconstituted enzyme systems from rabbit liver. J. Biol. Chem. 258:7311–7316. Johansson I, Ingelman-Sundberg M (1988) Benzene metabolism by ethanol-, acetone- and benzeneinducible cytochrome P-450 (IIE1) on rat and rabbit liver microsomes. Cancer Res. 48:5387–5390. Jowa L, Witz G, Snyder R, Winkle W, Kalf GF (1990) Synthesis and characterization of deoxyguanosine–benzoquinone adducts. J. Appl. Toxicol. 10:47–54. Kalf GF, Schlosser MJ, Renz JF, Pirozzi ST (1989) Prevention of benzene-induced myelotoxicity by nonsteroidal anti-inflammatory drugs. Environ. Health Perspect. 82:57–64. Kalf GF, Renz JF, Niculescu R (1996) p-Benzoquinone, a reactive metabolite of benzene, prevents the processing of pre-interleukins-1a and -1b to active cytokines by inhibition of the

REFERENCES

491

processing enzymes, calpain, and interleukin-1b converting enzyme. Environ. Health Perspect. 104:1251–1256. Keller KA, Snyder CA (1986) Mice exposed in utero to low concentrations of benzene exhibit enduring changes in their colony forming hematopoietic cells. Toxicology 42:171–181. Keller KA, Snyder CA (1988) Mice exposed in utero to 20 ppm benzene exhibit altered numbers of recognizable hematopoietic cells up to seven weeks after exposure. Fundam. Appl. Toxicol. 10:224–232. Kipen HM, Cody RP, Crump KS, Allen B, Goldstein BD (1988) Hematologic effects of benzene: a thirty-five year longitudinal study of rubber workers. Toxicol. Ind. Health 4:411–430. Kipen HM, Cody RP, Goldstein BD (1989a) Use of longitudinal analysis of peripheral blood counts to validate historical reconstruction of benzene exposure. Environ. Health Perspect. 82:199–206. Kipen HM, Cody RP, Goldstein BD (1989b) Letters to the editor. Toxicol. Ind. Health 5:1156–1158. Kirley TA, Goldstein BD, Maniara WM, Witz G (1989) Metabolism of trans,trans-muconaldehyde, a microsomal hematotoxic metabolite of benzene by purified yeast aldehyde dehydrogenase and a mouse liver soluble fraction. Toxicol. Appl. Pharmacol. 100:360–367. Kline SA, Robertson JF, Grotz VL, Goldstein BD, Witz G (1993) Identification of 6-hydroxy-trans, trans-2,4-hexadienoic acid, a novel ring-opened urinary metabolite of benzene. Environ. Health Perspect. 101(4):310–312. Kolachana P, Subrahmanyam VV, Meyer KB, Zhang L, Smith MT (1993) Benzene and its phenolic metabolites produce oxidative DNA damage in HL-60 cells in vitro and in the bone marrow in vivo. Cancer Res. 53:1023–1026. Koop DR, Laethem CL, Schnier GG (1989) Identification of ethanol-inducible P450 isozyme 3a (P450IIE1) as a benzene and phenol hydroxylase. Toxicol. Appl. Pharmacol. 98:278–288. Kreyon EM, Kraichely RE, Hudson KT, Medinsky MA (1996) Differences in rates of benzene metabolism correlate with observed genotoxicity. Toxicol. Appl. Pharmacol. 136:49–56. Kukielka E, Cederbaum AI (1995) Increased oxidation of ethylene glycol to formaldehyde by microsomes after ethanol treatment: role of oxygen radicals and cytochrome P450. Toxicol. Lett. 78:9–15. Lagorio S, Tagesson C, Forastiere F, Iavarone I, Axelson O, Carere A (1994) Exposure to benzene and urinary concentrations of 8-hydroxydeoxy-guanosine, a biological marker of oxidative damage to DNA. Occup. Environ. Med. 51:739–743. Lan Q, Zhang L, Li G, Vermeulen R, Weinberg RS, Dosemeci M, Rappaport SM, Shen M, Alter BP, Wu Y, Kopp W, Waidyanatha S, Rabkin C, Guo W, Chanock S, Hayes RB, Linet M, Kim S, Yin S, Rothman N, Smith MT (2004) Hematotoxicity in workers exposed to low levels of benzene. Science 306:1774–1776. Lan Q, Zhang L, Shen M, Smith MT, Li G, Vermeulen R, Rappaport SM, Forrest MS, Hayes RB, Linet M, Dosemeci M, Alter BP, Weinberg RS, Yin S (2005) Polymorphisms in cytokine and cellular adhesion molecule genes and susceptibility to hematotoxicity among benzene-exposed workers. Cancer Res. 65:9574–9581. Laskin DL, MacEachern L, Snyder R (1989) Activation of bone marrow phagocytes following benzene treatment of mice. Environ. Health Perspect. 82:75–79. Laskin JD, Rao NR, Punjabi CJ, Laskin DL, Snyder R (1995) Distinct actions of benzene and its metabolites on nitric oxide production by bone marrow leukocytes. J. Leukocyte Biol. 57:422– 426. Laskin DL, Heck DE, Punjabi CJ, Laskin JD (2000) Nitric oxide as a mediator of benzene-induced hematosuppression and toxicity. J. Toxicol. Environ. Health A 61:413–417. Latriano L, Zaccaria A, Goldstein BD, Witz G (1985) Muconaldehyde formation from 14 C-benzene in a hydroxyl radical generating system. J. Free Radic. Med. Biol. 1:363–371.

492

BENZENE

Latriano L, Zaccaria A, Goldstein BD, Witz G (1986) Formation of muconaldehyde, an open-ring metabolite of benzene in mouse liver microsomes: a novel pathway for toxic metabolites. Proc. Natl. Acad. Sci. USA 83:8356–8360. Lawyrk N, Weisel C (1996) Concentrations of volatile organic compounds in the passenger compartments of automobiles. Environ. Sci. Technol. 30:810–816. Layton DM, Mufti GJ (1987) The myelodysplastic syndromes. Br. Med. J. 295:227–228. Lee SS, Buters JT, Pineau T, Fernandez-Salquero P, Gonzalez FJ (1996) Role of CYP2E1 in the hepatotoxicity of acetaminophen. J. Biol. Chem. 15:3012–3022. Lindsey RH, Bender RP, Osheroff N (2005) Stimulation of topoisomerase II-mediated DNA cleavage by benzene metabolites. Chem. Biol. Interact. 153–154:197–205. Lindstrom AB, Yeowell-O’Connel K, Waidyanatha SB, Golding T, Tornero-Velez R, Rappaport SM (1997) Measurement of benzene oxide in blood of rats following administration of benzene. Carcinogenesis 18:1637–1641. Lindstrom AB, Yeowell-O’Connel K, Waidyanatha S, McDonald TA, Golding BT, Rappaport SM (1998) Formation of hemoglobin and albumin adducts of benzene oxide in mouse, rat, and human blood. Chem. Res. Toxicol. 11:302–310. Lindstrom AB, Yeowell-O’Connel K, Waidyanatha S, McDonald TA, Rappaport SM (1999) Investigation of benzene oxide in bone marrow and other tissues of F344 rats following metabolism of benzene in vitro and in vivo. Chem. Biol. Interact. 122:41–58. Leo A, Hansch C, Elkins D (1971) Partition coefficients and their uses. Chem. Rev. 71:569. Loden M (1986) The in vitro permeability of human skin to benzene, ethylene glycol, formaldehyde, and n-hexane. Acta Pharmacol. Toxicol. 58:382–389. Loeff I, Stein G (1959) Aromatic ring opening in the presence of oxygen in irradiated solutions. Nature 184:901. Long DJ II Gaikwad A, Multani A, Pathak S, Montgomery CA, Gonzalez FJ, Jaiswal A (2002) Disruption of the NAD(P)H:quinone oxidoreductase 1 (NQO1) gene in mice causes myelogenous hyperplasia. Cancer Res. 62:3030–3036. Longacre LS, Kocsis JJ, Snyder R (1981) Influence of strain differences in mice on the metabolism and toxicity of benzene. Toxicol. Appl. Pharmacol. 60:398–409. Lovern MR, Turner MJ, Meyer M, Kedderis GL, Bechtold WE, Schlosser PM (1997) Identification of benzene oxide as a product of benzene metabolism by mouse, rat, and human liver microsomes. Carcinogenesis 18:1695–1700. Lovern MR, Cole CE, Schlosser PM (2001) A review of quantitative studies of benzene metabolism. Crit. Rev. Toxicol. 31:285–311. Lunte SM, Kissinger PT (1983) Detection and identification of sulfhydryl conjugates of p-benzoquinone in microsomal incubations of benzene and phenol. Chem. Biol. Interact. 47:195–212. Maltoni C, Conti B, Cotti G (1983) Benzene: a multipotential carcinogen. Results of long-term bioassays performed at the Bologna Institute of Oncology. Am. J. Ind. Med. 4:589–630. McDonald TA, Yeowell-O’Connel K, Rappaport SM (1994) Comparison of protein adducts of benzene oxide and benzoquinone in the blood and bone marrow of rats and mice exposed to [14 C=13 C6 ] benzene. Cancer Res. 54:4907–4914. Medeiros AM, Bird MG, Witz G (1997) Potential biomarkers of benzene exposure. J. Toxicol. Environ. Health 51:519–539. Moran JL, Siegel D, Sun X-M, Ross D (1996) Induction of apoptosis by benzene metabolites in HL-60 and CD34þ human bone marrow progenitor cells. Mol. Pharmacol. 50:610–615. National Academy of Sciences (NAS) (1989) Biomarkers in Reproductive Toxicology. Washington, DC: National Academy Press.

REFERENCES

493

Nencki M, Giacosa P (1880) Uber die oxydation der aromatischen wasserstoffe in Turkoyser. Z. Physiol. Chem. 4:325. Norpoth K, Stucker W, Krewet E, Muller G (1988) Biomonitoring of benzene exposure by trace analysis of phenylguanine. Int. Arch. Occup. Environ. Health 60:163–168. Nwosu VC, Kissling GE, Trempus CS, Honeycutt H, French JE (2004) Exposure of Tg.AC transgenic mice to benzene suppresses hematopoietic progenitor cells and alters gene expression in critical signaling pathways. Toxicol. Appl. Pharmacol. 196:37–46. Ong CN, Kok PW, Lee BL, Shi CY, Ong HY, Chia KS, Lee CS, Luo XW (1995) Evaluation of biomarkers for occupational exposure to benzene. Occup. Environ. Med. 52:528–533. Parke DV, Williams RT (1953) Studies in detoxification. 44. The metabolism of benzene containing 14 C-benzene. Biochem. J. 54:231–238. Paxton MB, Chinchilli VM, Brett SM, Rodricks JV (1994) Leukemia risk associated with benzene exposure in the Pliofilm cohort. I. Mortality update and exposure distribution. Risk Anal. 14:147–154. Paxton MB, Chinchilli VM, Brett SM, Rodricks JV (1994) Leukemia risk associated with benzene exposure in the Pliofilm cohort. II. Risk estimates. Risk Anal. 14:155–161. Pellack-Walker P, Blumer JL (1986) DNA damage in L5178YS cells following exposure to benzene metabolites. Mol. Pharmacol. 30:42–47. Porteous JW, Williams RT (1953) Studies in detoxication. 49. The metabolism of phenol, catechol, quinol and hydroxyquinol from the ethereal sulfate fraction of the urine of rabbits receiving benzene orally. Biochem. J. 44:56–61. Powley MW, Carlson GP (1999) Species comparison of hepatic and pulmonary metabolism of benzene. Toxicology 139:207–217. Powley MW and Carlson GP (2000) Cytochromes P450 involved with benzene metabolism in hepatic and pulmonary microsomes. J. Biochem. Mol. Toxicol. 14(6):303–309. Qu Q, Shore R, Li G, Jin X, Chen LC, Cohen B, Melikian AA, Eastmond D, Rappaport S, Li H, Rupa D, Waidyanatha S, Yin S, Yan H, Meng M, Winnik W, Kwok ES, Li Y, Mu R, Xu B, Zhang X, Li K (2003) Validation and Evaluation of Biomarkers in Workers Exposed to Benzene in China. HEI Research Report 115. Qu Q, Shore R, Li G, Su L, Yin X, Melikian AA, Roy N, Chen LC, Wirgin E, Cohen B, Yin S, Li Y, Mu R (2005) Biomarkers of benzene exposure: urinary metabolites in relation to individual genotype and personal exposure. Chem. Biol. Interact. 153–154:85–95. Quitt M, Cassel A, Yoffe A, Anatol AM, Froom P (2004) Autonomous growth of committed hematopoietic progenitors from peripheral blood of workers exposed to low levels of benzene. J. Occup. Environ. Med. 46:27–29. Rana SV, Verma Y (2005) Biochemical toxicity of benzene. J. Environ. Biol. 26:157–168. Rao GS (1996) Glutathionyl hydroquinone: a potent prooxidant and a possible toxic metabolite of benzene. Toxicology 106:49–54. Rappaport SM, Waidyanatha S, Qu Q, Shore R, Jin X, Cohen B, Chen LC, Melikian AA, Li G, Yin S, Yan H, Xu B, Mu R, Li Y, Zhang X, Li K (2002) Albumin adducts of benzene oxide and 1,4benzoquinone as measures of human benzene metabolism. Cancer Res. 62:1330–1337. Rappaport SM, Waidyanatha S, Yeowell-O’Connel K, Rothman N, Smith MT, Zhang L, Qu Q, Shore R, Li G, Yin S (2005) Protein adducts as biomarkers of human benzene metabolism. Chem. Biol. Interact. 153–154:103–109. Rickert DE, Baker TS, Bus JS, Barrow CS, Irons RD (1979) Benzene disposition in the rat after exposure by inhalation. Toxicol. Appl. Pharmacol. 49:417–423. Rinsky RA, Young RJ, Smith AB (1981) Leukemia in benzene workers. Am. J. Ind. Med. 2:217– 245.

494

BENZENE

Rinsky RA, Smith AB, Hornung R, Filloon TG, Young RF, Okun AD, Landrigan PJ (1987) Benzene and leukemia: an epidemiological risk assessment. N. Engl. J. Med. 316:1044–1050. Rivedal E, Witz G (2005) Benzene metabolites block gap junction intracellular communication. Role in hematotoxicity and leukemia? Chem. Biol. Interact. 153–154:257–260. Rosenthal GJ, Snyder CA (1985) Modulation of the immune response to Listeria monocytogenes by benzene inhalation. Toxicol. Appl. Pharmacol. 80:502–510. Rosenthal GJ, Snyder CA (1987) Inhaled benzene reduces aspects of cell-mediated tumor surveillance in mice. Toxicol. Appl. Pharmacol. 88:35–43. Ross D (2000) The role of metabolism and specific metabolites in benzene-induced toxicity: evidence and issues. J. Toxicol. Environ. Health A 61:357–372. Ross D (2005) Functions and distribution of NQO1 in human bone marrow: potential clues to benzene toxicity. Chem. Biol. Interact. 153–154:137–146. Rothman N, Smith MT, Hayes RB , et al. (1997) Benzene poisoning, a risk factor for hematological malignancy, is associated with the NQO1 609 C  T mutation and rapid fractional excretion of chlorzoxazone. Cancer Res. 57:2839–2842. Rothman N, Haas R, Hayes RB, Li G-L, Wiemels J, Campleman S, Quintana PJE, Xi LJ, Dosemeci M, Titenko-Holland N, Meyer KB, Lu W, Zhang LP, Bechtold W, Wang YZ, Kolachana P, Yin SN, Blot W, Smith MT (1995) Benzene induces gene-duplicating but not gene-inactivating mutations at the glycophorin A locus in exposed humans. Proc. Natl. Acad. Sci. USA 92:4069–4073. Rothman N, Li G-L, Dosemeci M, Bechtold W, Marti GE, Wang Y-Z, Linet M, Xi L-Q, Lu W, Smith MT, Titenko-Holland N, Zhang L-P, Blot W, Yin S-N, Hayes RB (1996) Hematotoxicity among Chinese workers heavily exposed to benzene. Am. J. Ind. Med. 29:236–246. Rozen MG and Snyder CA (1985) Protracted exposure of C57BL/6 mice 300 ppm benzene depresses B- and T-lymphocyte numbers and mitogen responses. Evidence for thymic and bone marrow proliferation in response to the exposures. Toxicology 37(1–2):13–26. Sabourin PJ, Chen BT, Lucier G, Birnbaum LS, Fisher E, Henderson RF (1987) Effect of dose on the absorption and excretion of [14 C] benzene administered orally or by inhalation in rats and mice. Toxicol. Appl. Pharmacol. 87:325–336. Sabourin PJ, Bechtold WE, Birnbaum LS, Lucier G, Henderson RF (1988a) Differences in the metabolism and disposition of inhaled [3 H] benzene by F344/N rats and B6C3F1 mice. Toxicol. Appl. Pharmacol. 94:128–140. Sabourin PJ, Bechtold WE, Henderson RF (1988b) A high pressure liquid chromatographic method for the separation and quantitation of water-soluble radiolabeled benzene metabolites. Anal. Biochem. 170:316–327. Sabourin PJ, Bechtold WE, Griffith WC, Birnbaum LS, Lucier G, Henderson RF (1989) Effect of exposure concentration, exposure rate, and route of administration on metabolism of benzene by F344 rats and B6C3F1 mice. Toxicol. Appl. Pharmacol. 99:421–444. Sabourin PJ, Sun JD, MacGregor JT, Wehr CM, Birnbaum CG, Lucier G, Henderson RF (1990) Effect of repeated benzene inhalation exposures on benzene metabolism, binding to hemoglobin, and induction of micronuclei. Toxicol. Appl. Pharmacol. 103:452–462. Sadler A, Subrahmanyam VV, Ross D (1988) Oxidation of catechol by horseradish peroxidase and human leukocyte peroxidase: reactions of o-benzoquinone and o-benzosemiquinone. Toxicol. Appl. Pharmacol. 93:62–71. Sammett D, Lee EW, Kocsis JJ, Sunder R (1979) Partial hepatectomy reduces both metabolism and toxicity of benzene. J. Toxicol. Environ. Health 5:785–792. Sato A, Nakajima T (1979) Partition coefficients of some aromatic hydrocarbons and ketones in water, blood and oil. Toxicol. Appl. Pharmacol. 48:49. Savitz D, Andrews KW (1997) Review of epidemiologic evidence on benzene and lymphatic and hematopoietic cancers. Am. J. Ind. Med. 31:287–295.

REFERENCES

495

Sawahata T, Neal RA (1982) Horseradish peroxidase-mediated oxidation of phenol. Biochem. Biophys. Res. Commun. 109:988–994. Sawahata T, Rickert DA, Greenlee WF (1985) Metabolism of benzene and its metabolites in bone marrow. In: Irons R, editor. Toxicology of the Blood and Bone Marrow.New York:Raven Press, pp. 141–148. Schnatter AR, Nicolich MJ, Bird MG (1996) Determination of leukemogenic benzene exposure concentrations: refined analyses for the Pliofilm cohort. Risk Anal. 16:833–840. Schnatter AR, Rosamilia K, Wojcik NC (2005) Review of the literature on benzene exposure and leukemia subtypes. Chem. Biol. Interact. 153–154:9–21. Schoenfeld H, Witz G (2000) Structure–activity relationships in the induction of DNA–protein crosslinks by hematotoxic ring-opened benzene metabolites and related compounds in HL60 cells. Toxicol. Lett. 116:79–88. Schoenfeld HA, Radbourne R, Goldstein BD, Witz G (1996) Mechanistic studies on benzene toxicity: formation of DNA–protein crosslinks induced by benzene in mice and muconaldehyde in HL-60 cells. Fundam. Appl. Toxicol. (Suppl.) 30:232. Schoenfeld HA, Mirochnitchenko O, Goldstein BD, Witz G (1997) Effects of trans,trans-muconaldehyde, a hematotoxic benzene metabolite, or DNA–protein crosslink levels and on apoptosis in mouse bone marrow cells. Fundam. Appl. Toxicol. (Suppl.) 31:163. Schrenk DM, Ingelman-Sundberg, Bock KW (1992) Influence of P-4502E1 induction on benzene metabolism in rat hepatocytes and on biliary metabolite excretion. Drug Metab. Dispos. 20: 137–141. Schultzen O, Naunyn B (1867) Uber das Verhalten des Kohlenwasserstoffes im Organisms. Arch. Anat. Physiol. 1:349. Schwartz CS, Snyder R, Kalf GF (1985) The inhibition of mitochondrial DNA replication in vitro by the metabolites of benzene hydroquinone and p-benzoquinone. Chem. Biol. Interact. 53:327– 350. Seaton MJ, Schlosser PM, Bond JA, Medinsky MA (1994) Benzene metabolism by human liver microsomes in relation to cytochrome P450 2E1 activity. Carcinogenesis 15:1799–1806. Sheets P, Carlson G (2004) Kinetic factors involved in the metabolism of benzene in mouse lung and liver. J. Toxicol. Environ. Health A 67:421–430. Shen M, Lan Q, Zhang L, Chanock S, Li G, Vermeulen R, Rappaport SM, Guo W, Hayes RB, Linet M, eager M, Welch R, Forrest MS, Rothman N, Smith MT (2006) Polymorphisms in genes involved in DNA double-strand break repair pathway and susceptibility to benzene-induced hematotoxicity. Carcinogenesis 27:2083–2089. Sherwood RJ, Sinclair GC (1999) New PBPK model applied to old occupational exposure to benzene. Am. Ind. Hyg. Assoc. J. 60:259–265. Smart RC, Zannoni VG (1985) DT-diaphorase and peroxidase influence the covalent binding of metabolites of phenol, the major metabolite of benzene. Mol. Pharmacol. 26:105–111. Smith MT (1996) The mechanism of benzene-induced leukemia: a hypothesis and speculations on the causes of leukemia. Environ. Health Perspect. 104:1219–1225. Smith MT, Yager JW, Steinmetz KL, Eastmond DA (1989) Peroxidase-dependent metabolism of benzene’s phenolic metabolites and its potential role in benzene toxicity and carcinogenicity. Environ. Health Perspect. 82:23–29. Smith MT, Vermeulen R, Li G, Zhang L, Lan Q, Hubbard AE, Forrest MS, McHale C, Zhao X, Gunn L, Shen M, Rappaport SM, Yin S, Channock S, Rothman N (2005) Use of ‘omic’ technologies to study humans exposed to benzene. Chem. Biol. Interact. 153–154:123–127. Snyder CA (1987) Benzene. In: Snyder R, editor. Ethel Browning’s Toxicity and Metabolism of Industrial Solvents, Vol. I. Hydrocarbons, 2nd edn.Amsterdam:Elsevier, pp. 3–27. Snyder R (2002) Benzene and leukaemia. Crit. Rev. Toxicol. 32:155–210.

496

BENZENE

Snyder R (2004) Xenobiotic metabolism and the mechanism(s) of benzene toxicity. Drug Metab. Rev. 36:531–547. Snyder CA, Goldstein BD, Sellakumar A (1978) Hematotoxicity of inhaled benzene to Sprague– Dawley rats and AKR mice at 300 ppm. J. Toxicol. Environ. Health 4:605–618. Snyder CA, Goldstein BD, Sellakumar AR, Bromberg I, Laskin S, Alberts RE (1980) The inhalation toxicology of benzene: incidence of hematopoietic neoplasms and hematotoxicity in ARK/J and C57BL/6J mice. Toxicol. Appl. Pharmacol. 54:323–331. Snyder R, Witz G, Goldstein BD (1993) The toxicology of benzene. Environ. Health Perspect. 100:293–306. Sorahan T, Kinlen LJ, Doll R (2005) Cancer risks in a historical UK cohort of benzene exposed workers. Occup. Environ. Med. 62:231–236. Steffen C, Auclerc MF, Auvrignon A, Baruchel A, Kebaili K, Lambilliotte A, Leverger G, Sommelet D, Vilmer E, Hemon D, Clavel J (2004) Acute childhood leukemia and environmental exposure to potential sources of benzene and other hydrocarbons: a case-control study. Occup. Environ. Med. 61:773–778. Subrahmanyam VV, O’Brien PJ (1985) Phenol oxidation product(s) formed by a peroxidase reaction that bind to DNA. Xenobiotica 15:873–885. Subrahmanyam VV, Ross D, Eastmond DA, Smith MT (1991) Potential role of free radicals in benzene-induced myelotoxicity and leukaemia. Free Radic. Biol. Med. 11:495–515. Tavassoli M, Friedenstein A (1983) Hemapoietic stromal microenvironment. Am. J. Hematol. 14:195–203. Thibodeaux LJ (1981) Estimating the air emissions of chemicals from hazardous waste landfills. J. Hazard. Mater. 4:235–244. Tunek A, Platt KL, Przybylski M, Oesch F (1980) Multistep metabolic activation of benzene. Effect of superoxide dismutase on covalent binding to microsomal macromolecules and identification of glutathione conjugates using high pressure liquid chromatography and field desorption mass spectrometry. Chem. Biol. Interact. 33:1. Valentine JL, Lee SS-T, Seaton MJ, Asgharian B, Farris G, Corton JC, Gonzalez FJ, Medinsky MA (1996) Reduction of benzene metabolism and toxicity in mice that lack CYP2E1 expression. Toxicol. Appl. Pharmacol. 141:205–213. Van Wendel de Joode B, Tielemans E, Vermeulen R, Wegh H, Kromhout H (2005) Dermal exposure assessment to benzene and toluene using charcoal cloth pads. J. Expo. Anal. Environ. Epidemiol. 15:47–50. Vermeulen R, Li G, an Q, Dosemeci M, Rappaport SM, Bohong X, Smith MT, Zhang L, Hayes RB, Linet M, Mu R, Wang L, Yu J, Yin S, Rothman N (2004) Detailed exposure assessment for a molecular epidemiology study of benzene in two shoe factories in China. Ann. Occup. Hyg. 48:105–116. Waidyanatha S, Rothman N, Li G, Smith MT, Yin S, Rappaport SM (2004) Rapid determination of six urinary benzene metabolites in occupationally exposed and unexposed subjects. Anal. Biochem. 327:184–199. Wallace LA (1987) The Total Exposure Assessment Methodology (TEAM) Study: Summary and Analysis, Vol. I. EPA-600-6-87-002a. Washington, DC: U.S. EPA, Office of Acid Deposition, Environmental Monitoring and Quality Assurance. Wallace LA, Pellizzari ED (1986) Personal air exposures and breath concentrations of benzene and other volatile hydrocarbons for smokers and nonsmokers. Toxicol. Lett. 35:113–116. Wallace LA, Sparacino C, Sheldon LS, Zelon H (1985) Personal exposures, indoor–outdoor relationships, and breath levels of toxic air pollutants measured for 355 persons in New Jersey. Atmos. Environ. 19:1651–1661.

REFERENCES

497

Wallace LA, Pellizzari ED, Harwell TD, Waidyanatha S, Rothman N, Li G, Smith MT, Yin S, Rappaport SM (2004) Rapid determination of six urinary benzene metabolites in occupationally exposed and unexposed subjects. Anal. Biochem. 327:184–199. Walling C, Johnson RA (1975) Fenton’s reagent. V. Hydroxylation and side chain cleavage of aromatics. J. Am. Chem. Soc. 97:363–367. Wan J, Badham HJ, Winn L (2005) The role of c-MYB in benzene-initiated toxicity. Chem. Biol. Interact. 153–154:171–178. Ward E, Hornung R, Morris J, Rinsky R, Wild D, Halperin W, Guthrie W (1996) Risk of low red or white blood cell count related to estimated benzene exposure in a rubberworker cohort (1940– 1975). Am. J. Ind. Med. 29:247–257. Weisel C, Yu R, Roy A, Georgopoulos P (1996) Biomarkers of environmental benzene exposure. Environ. Health Perspect. 104: (Suppl. 5)1141–1146. Whysner J, Reddy MV, Ross PM, Mohan M, Lax EA (2004) Genotoxicity of benzene and its metabolites. Mutat. Res. 566:99–130. Wiedra D, Irons RD, Greenlee WF (1981) Immunotoxicity in C57BL/6 mice exposed to benzene and Aroclor 1254. Toxicol. Appl. Pharmacol. 60:410–417. Williams PR, Paustenbach DJ (2003) Reconstruction of benzene exposure for the Pliofilm cohort (1936–1976) using Monte Carlo techniques. J. Toxicol. Environ. Health 66:677–781. Witz G, Rao GS, Goldstein BD (1985) Short term toxicity of trans,trans-muconaldehyde. Toxicol. Appl. Pharmacol. 80:511–516. Witz G, Maniara WM, Mylavarapu VJ, Goldstein BD (1990) Comparative metabolism of benzene and trans,trans-muconaldehyde to trans,trans-muconic acid in DBA/2N and C57BL/6 mice. Biochem. Pharmacol. 40:1275–1280. Witz G, Zhang Z, Goldstein BD (1996) Reactive ring-opened aldehyde metabolites in benzene hematotoxicity. Environ. Health Perspect. 104:1195–1199. Wong O (2002) Investigations of benzene exposure, benzene poisoning and malignancies in China. Regul. Toxicol. Pharmacol. 35:126–135. Wong O, Raabe GK (2000) Non-Hodgkin’s lymphoma and exposure to benzene in a multinational cohort of more than 308,000 petroleum workers, 1937 to 1996. J. Occup. Environ. Med. 42: 554–568. World Health Organization (WHO) (1987) Air Quality Guideline for Europe.Copenhagen:WHO Regional Publications. Wu D, Cederbaum AI (1994) Characterization of pyrazole and 4-methylpyrazole induction of cytochrome P4502E1 in rat kidney. J. Pharm. Exp. Ther. 270:407–413. Yamanaka K, Hayashi H, Kato K, Hasegawa A, Okada S (1995) Involvement of preferential formation of apurinic/apyrimidinic sites in dimethylarsenic-induced DNA strand breaks and DNA–protein crosslinks in cultured alveolar epithelial cells. Biochem. Biophys. Res. Commun. 207:244–249. Yardley-Jones A, Anderson D, Jenkinson PC, Lovell DP, Blowers SD, Davies MJ (1988) Genotoxic effects in peripheral blood and urine of workers exposed to low level benzene. Br. J. Ind. Med. 45:694–700. Yeowell O’Connell K, Rothman N, Waidyanatha S, Smith MT, Hayes RB, Li G, Bechtold WE, Dosemeci M, Zhang L, Yin S, Rappaport SM (2001) Protein adducts of 1,4-benzoquinone and benzene oxide among smokers and nonsmokers exposed to benzene in China. Cancer Epidemiol. Biomarkers Prev. 10:831–838. Yin SN, Li G-L, Tain F-D, Fu Z-I, Jin C, Chen Y-J, Luo S-J, Ye P-Z, Zhang J-Z, Wang G-C, Zhang X-C, Wu H-N, Zhong Q-C (1987a) Leukemia in benzene workers: a retrospective cohort study. Br. J. Ind. Med. 44:124–128.

498

BENZENE

Yin S-N, Hayes RB, Linet MS, Li G-L, Dosemeci M, Travis LB, Li C-Y, Zhang Z-N, Li D-G, Chow W-H, Wacholder S, Wang Y-Z, Jiang Z-L, Dai T-R, Zhang W-Y, Chao X-J, Ye P-Z, Kou Q-R, Zhang X-C, Lin X-F, Meng J-F, Ding C-Y, Zho J-S, Blot WJ (1996) A cohort study of cancer among benzene-exposed workers in China: overall results. Am. J. Ind. Med. 29:227–235. Yin SN, Hayes RB, Linet MS, Li GL, Dosemeci M, Travis LB, Zhang ZN, Li DG, Chow WH, Wacholder S, Blot WJ (1996b) An expanded cohort study of cancer among benzene-exposed workers in China. Benzene Study Group. Environ Health. Perspect. 104S6:1336–1341. Yoon B, Li G, Kitada K, Kawasaki Y, Igarashi K, Kodama Y, Inoue T, Kobayashi K, Kanno J, Kim D, Inoue T, Hirabayashi Y (2003) Mechanisms of benzene-induced hematoxicity and leukemogenicity: cDNA microarray analyses using moue bone marrow tissue. Environ. Health Perspect. 111:1411–1420. Young N (1989) Benzene and lymphoma. Am. J. Ind. Med. 15:495–498. Yu R, Weisel C (1996) Measurement of benzene in human breath associated with an environmental exposure. J. Expo. Anal. Environ. Epidemiol. 6:261–277. Zeise L, McDonald TA (2000) California perspective on the assessment of benzene toxicological risks. J. Toxicol. Environ. Health A 61:479–483. Zhang Z, Kline SA, Kirley TA, Goldstein BD, Witz G (1993) Pathways of trans,trans-muconaldehyde metabolism in mouse liver cytosol: reversibility of monoreductive metabolism and formation of end products. Arch. Toxicol. 67:461–467. Zhang Z, Xiang Q, Glatt H, Platt KI, Goldstein BD, Witz G (1995a) Studies on pathways of ring opening of benzene in a Fenton system. Free Radic. Biol. Med. 18:411–419. Zhang Z, Goldstein BD, Witz G (1995b) Iron-stimulated ring-opening of benzene in a mouse liver microsomal system. Mechanistic studies and formation of a new metabolite. Biochem. Pharmacol. 50:1607–1617. Zhang Z, Schafer F, Schoenfeld H, Cooper K, Snyder R, Goldstein BD, Witz G (1995c) The hematotoxic effects of 6-hydroxy-trans,trans-2,4-hexadienal, a reactive metabolite of trans, trans-muconaldehyde, in CD-1 mice. Toxicol. Appl. Pharmacol. 132:213–219. Zhang L, Eastmond DA, Smyth MT (2002) The nature of chromosomal aberrations detected in humans exposed to benzene. Crit. Rev. Toxicol. 32:1–42.

14 CARBON MONOXIDE Michael T. Kleinman

14.1 INTRODUCTION Carbon monoxide is emitted from virtually all sources of incomplete combustion, including internal combustion engines (e.g., automobiles, trucks, and gasoline-fueled small engines) (Duci et al., 2003; El-Fadel and El-Hougeiri, 2003; Mott et al., 2002; Ott et al., 1994; Renner, 1988; Utell et al., 1994); fires, both natural and man-made; improperly adjusted gas and oil appliances (e.g., space heaters (Anonymous, 1997; Setiani, 1994), water heaters (Howell et al., 1997), stoves (Guggisberg et al., 2003; Samet et al., 1987), and ovens (Anonymous, 1997; Angle, 1988)); and tobacco smoking (Calafat et al., 2004; Gourgoulianis et al., 2002; Murray et al., 2002; Viegi et al., 2004). Because of the large number and the ubiquity of CO sources with significant source strengths (e.g., tobacco smoke contains about 1% CO by volume, or 10,000 ppm CO), ambient CO concentrations show large temporal and spatial variations. The exposure of individuals to CO is, therefore, also quite variable, depending upon the types of activities in which a person is engaged and how long they are engaged in those activities (time–activity profiles), where the activity takes place (microenvironments, for example, indoors, at a shopping mall, outdoors, in a vehicle, at work or school, in a parking garage, or even in a skating rink) (Horner, 2000; Jovanovic et al., 1999; Levesque et al., 2000, 2005; Viala, 1994), and the proximity to CO sources (Campbell et al., 2005; Linn and Gong, 1999; Ott, 1990; Ott et al., 1992). Various methods have been used to document exposures, including the use of data from fixed-site ambient monitors, the use of microenvironmental exposure assessment models, personal exposure monitoring methods, and biological monitoring methods (Ott et al., 1992). Controls on motor vehicle exhaust and the use of catalytic converters on vehicles sold in the United States have been very effective in reducing ambient CO emissions and commuter exposures (Hinkle, 1980; Hutchinson and Pearson, 2004; Hysell et al., 1975; Mott et al., 2002). However, while the increased use of oxidizing catalysts resulted in decreased emissions of CO, they also increased emissions of noble metals and ultrafine particles, which may contribute to health effects in other ways

Environmental Toxicants, Third Edition Edited by Morton Lippmann Copyright
2009 John Wiley & Sons, Inc.

499

500

CARBON MONOXIDE

(Brubaker et al., 1975; Finklea et al., 1975). The fact remains that the reduction in automotive emissions brought about by the Clean Air Act has reduced traffic CO exposures and traffic-related CO concentrations well below those measured prior to the year 2000 (U.S. EPA, 1999).

14.2 CO EXPOSURE AND DOSIMETRY CO competes with oxygen for binding sites on the heme portion of the hemoglobin (Hb) molecules in red blood cells to form carboxyhemoglobin (COHb). The affinity of Hb for CO is about 240–250 times that for O2 (Roughton, 1970). The formation of COHb by the binding of CO to circulating Hb reduces the oxygen-carrying capacity of blood. In addition, binding of CO to one of the four hemoglobin binding sites increases the O2 affinity of the remaining binding sites, thus interfering with the release of O2 at the tissue level. When O2 content of blood (mL O2/mL blood) is plotted versus O2 partial pressure (mmHg) in blood, the increased O2 affinity is seen as the so-called leftward shift in the curve for blood partially loaded with CO (Okada et al., 1976; Zwart et al., 1984). CO-induced tissue hypoxia is therefore a joint effect of the reduction in O2-carrying capacity and the reduction of O2 release at the tissue level. The brain and heart, under normal conditions, utilize larger fractions of the arterially delivered O2 (about 75%) than do peripheral tissues and other organs (Ayres et al., 1970) and are therefore the most sensitive targets for hypoxic effects following CO exposures. The potential for adverse health effects is increased under conditions of stress, such as exercise, which increases O2 demands at the tissue level to sustain metabolism. The measure of biological dose that relates best to observed biological responses and deleterious health effects is the concentration of COHb expressed as a percentage of available, active Hb, thus representing the percent of potential saturation of Hb. COHb can be measured directly in blood or estimated from the CO content of expired breath (Berny et al., 2002; Attebring et al., 2001; Groman et al., 1998; Laranjeira et al., 2000; Rea et al., 1973; Vogt et al., 1977; Vreman et al., 1996; Wickramatillake, 1999). It is currently accepted that the most accurate and reliable method for measuring COHb concentration is by gas chromatographic analysis (Berny et al., 2002; Vreman et al., 1984, 1994). Spectrophotometric measures and instruments have been widely used in both clinical and occupational health settings (Boumba and Vougiouklakis, 2005). However, spectrophotometric instruments may have limited accuracy and precision at COHb concentrations below 5% (Allred et al., 1989; Chaitman et al., 1992; Johansson and Wollmer, 1989). If used for research studies, they should be calibrated properly and measurements should verified by a “gold standard,” such as gas chromatography (Johansson and Wollmer, 1989; Widdop, 2002; Wigfield et al., 1981). When direct measurements cannot be made, it is possible to estimate COHb from ambient air CO concentrations (Ott et al., 1988), indoor air CO concentrations, and personal CO monitoring data (Ott and Mage, 1978, 1979) using pharmacokinetic and other models (Benignus et al., 1994; Bruce and Bruce, 2003; Chung, 1988; Coburn et al., 1965; Hauck and Neuberger, 1984; Joumard et al., 1981; Peterson and Stewart, 1975; Tikuisis et al., 1987) that link the concentration of inhaled CO, breathing rate and volume, blood volume, metabolic production of endogenous CO, and rate of removal of CO. The Coburn– Forster–Kane (CFK) model (Coburn et al., 1965) has been widely used for this purpose. But perhaps more importantly, in the process of establishing ambient air quality guidelines, the CFK model has been the basis for associating ambient and workplace air CO concentrations

MECHANISMS OF CO TOXICITY

501

with concentrations of COHb that could be hazardous for sensitive exposed individuals. The CFK model has been experimentally verified for exposures at 25–5000 ppm, during rest and exercise (Peterson and Stewart, 1975; Tikuisis et al., 1987). Sensitivity analyses of CFK can be used to identify those model input parameters for which errors in the parameter values that are used will have the greatest errors in predicted COHb concentrations (McCartney, 1990). Validation of the CFK model under well-controlled conditions at environmental concentrations with larger numbers of subjects would be useful. However, Kleinman et al. (1998) used the CFK model to successfully predict blood COHb concentrations in 17 human volunteers exposed to 100 ppm CO, suggesting that the CFK model is valid under such circumstances.

14.3 MECHANISMS OF CO TOXICITY CO affects health indirectly by interfering with the transport of oxygen to tissues (especially the heart and other muscles and brain tissue) (McGrath, 2000). The resulting impairment of O2 delivery cause tissue hypoxia and interferes with cellular respiration. When CO is taken up by cells, it can complex with Fe2þ in hemoproteins such as myoglobin (McGrath, 2000), cytochrome oxidase, and cytochrome P450 (so named because the Fe2þ–CO complex absorbs light with a maximum absorption at 450 nm) (Williams, 1992), and thus interfere with electron transport processes and energy production at the cellular level. Thus, in addition to observed physiological effects and cardiovascular effects, CO can modify electron transport in nerve cells resulting in behavioral, neurological, and developmental toxicological consequences. The possible role of CO as an etiologic factor in development of atherosclerosis is suggested by effects of tobacco smoke exposure (Hart, 1993; Leone, 1995) and mobile source emissions (Utell et al., 1994), but long-term exposures to 200 ppm CO in a sensitive animal model failed to show an effect of CO (Penn, 1993). The role of CO as a causative factor in cardiac arrhythmias, sudden cardiac arrest, and myocardial infarctions is an area of active research activity. The hemodynamic responses to CO have been reviewed for both animal models and humans (Kanten et al., 1983; Penney, 1988). Chronic CO exposures, usually at COHb concentrations greater than 10%, produce several changes. These may be adaptive responses to induced hypoxia, such as increases in numbers of red blood cells (polycythemia), increased blood volume, and increased heart size (cardiomegaly). In addition, heart rate, stroke volume, and systolic blood pressure may be increased. Some of these effects have been seen in smokers. Other environmental factors, such as effects of other pollutants (both from conventional air pollution sources and from environmental tobacco smoke), interactions with drugs and medications, health and related factors (e.g., cardiovascular and respiratory diseases, anemia, or pregnancy), and exposures at high altitude are possible risk modifiers for the health effects of CO. Exposures to high concentrations of CO due, for example, to fires and emissions from faulty appliances result in over 2000 deaths per year in the United States and other countries (Abu-al Ragheb and Battah, 1999; Raub et al., 2000b; Sadovnikoff et al., 1992) and in illness sufficient to cause upward of 10,000 individuals to seek medical attention or to miss one or more days of work in the United States (Centers for Disease Control, 1982). The available data may substantially underestimate the total number of such cases, especially those related to unsuspected CO exposure in the home because some CO-related symptoms are similar to those of flu (headache and dizziness) and possible to those of certain seizure disorders (Heckerling, 1987; Heckerling et al., 1990a, 1990b; Kirkpatrick, 1987; Leikin et al., 1988).

502

CARBON MONOXIDE

Therefore, many cases may be misdiagnosed and missed as being related to CO exposure. As a consequence, patients may not receive the proper treatment and their cohabitants may go untreated if they did not independently seek medical help (Heckerling et al., 1990b; Kao and Nanagas, 2004). Many of the inadvertent incidents of morbidity and mortality are preventable through the use of CO detectors, which are now readily available at a moderate cost (Yoon et al., 1998). Blood tests for COHb concentrations or breath analyses for CO improve the accuracy of the diagnoses. While many improvements have been made with respect to exposure to ambient CO, this remains as an important issue. The remainder of this chapter will, therefore, deal with ambient environmental exposures and will focus on the recent findings of CO-related health effects.

14.4 POPULATIONS AT RISK OF HEALTH EFFECTS DUE TO CO EXPOSURE 14.4.1

People with Cardiovascular Diseases

Daily variations in CO were strongly associated with hospital admissions among persons with ischemic heart disease (IHD) conditions, even after controlling for potential effects of ozone, nitrogen dioxide, or particulate matter less than or equal to 10 mm in aerodynamic diameter (PM10). A 1 ppm increase in 8 h average CO was associated with a 3.60% increase in same-day IHD admissions (Mann et al., 2002). Ischemic heart disease, also categorized as coronary artery disease, is a leading cause of disability and death in industrialized nations and may be associated with chronic elevation of COHb (Mall et al., 1985). It is a clinical disorder of the heart resulting from an imbalance between oxygen demand of myocardial tissue and oxygen delivery via the bloodstream. The ability of the heart to adjust to increases in myocardial O2 demands resulting from increased activity, or to reductions in O2 delivery by arterial blood due, for example, to COHb or reduced partial pressure in O2 in inspired air, by increasing O2 extraction, is limited, because the extraction rate in myocardial tissue is already high. Normally, coronary circulation responds to increased O2 demands by increasing blood flow. In coronary artery disease, the coronary artery is occluded by lipid deposits, which can impede augmentation of local coronary blood flow in response to increased O2 demands. Under these conditions, the myocardium is forced to extract more O2 resulting in reduced coronary venous and tissue O2 tensions, which can produce myocardial ischemia. Severe myocardial ischemia can induce a myocardial infarction (heart attack) or can alter cardiac rhythms, that is, cause arrhythmias. The association of acute CO exposure to heart attacks has been described (Koskela et al., 2000; Marius-Nunez, 1990; Martys, 1994; Scharf et al., 1974; Tan et al., 1993). Individuals with obstructed peripheral arteries may experience intermittent claudication, which is severe pain, usually in their legs, during walking or other relatively mild activities. CO exposure, for example, from cigarette smoking, can exacerbate the imbalance between O2 demand by exercising peripheral muscular tissue and O2 delivery in individuals with diseased peripheral arteries (Wald et al., 1977). 14.4.2

People with Anemia and Other Blood Disorders

Individuals with reduced blood hemoglobin concentrations, or with abnormal hemoglobin, will have reduced O2-carrying capacity in blood. In addition, disease processes that result in increased destruction of red blood cells (hemolysis) and accelerated breakdown of

REGULATORY BACKGROUND

503

hemoproteins accelerate endogenous production of CO (Sannolo et al., 1992; Sears et al., 2001; Solanki et al., 1988), resulting in higher COHb concentrations than in normal individuals. For example, patients with hemolytic anemia have COHb concentrations two to three times those seen in normal individuals (Coburn et al., 1966). Endogenously produced CO, from the breakdown of hemoglobin by hemeoxygenase, was originally thought to be a superfluous by-product of heme catabolism. However, CO is now known to play a central role in blood pressure regulation, maintenance of organ-specific vascular tone, neurotransmission, stress response, platelet activation, and smooth muscle relaxation (Morse and Sethi, 2002; Wu and Wang, 2005). Thus, CO may be an important and beneficial mediator at normal physiological levels, but is toxic at elevated levels. 14.4.3

People with Chronic Lung Disease

Chronic lung diseases such as chronic bronchitis, emphysema, and chronic obstructive pulmonary disease (COPD) are characterized by impairment of the lung’s ability to transfer O2 to the bloodstream because diseased regions of the lung are poorly ventilated and blood circulating through these regions will therefore receive less O2 and accumulate carbon dioxide (so-called ventilation–perfusion mismatch) (Wagner et al., 1977; West, 1971, 1978). Exertional stress often produces a perception of difficulty in breathing, or breathlessness (dyspnea) in these individuals. Although exercise, and the metabolic acidosis associated with exercise in COPD patients, increases ventilatory drive, they have limited ventilatory capacity with which to respond (Sue et al., 1988). Reduction of blood O2 delivery capacity due to formation of COHb could exacerbate symptoms and further reduce exercise tolerance in these individuals. 14.4.4

Potential Risks for Pregnant Women, Fetuses, and Newborn Children

A CO-induced leftward shift in the O2Hb saturation curve may be significant for fetuses because the O2 tension in their arterial blood is low (20–30 mmHg) compared to adult values (100 mmHg) and because fetal Hb has a higher O2 affinity than does maternal Hb (Longo, 1976, 1977). Fetal blood has higher Hb concentrations than does maternal blood (Heilmann et al., 2005), which may compensate for the higher O2 affinity to some extent. In pregnant women, O2 consumption is increased 15–25%, and hemoglobin concentration may be simultaneously reduced, lowering the O2-carrying capacity of their blood (Sady and Carpenter, 1989). Epidemiological studies show that odds ratios for cardiac ventricular septal defects increased in a dose-responsive fashion with increasing CO exposure, a 1 ppm increase in mean exposure to CO during the first trimester of pregnancy is associated with a reduction of 23 g in birth weight, and first-trimester CO exposures were associated with 20% increased risk of intrauterine growth retardation (Gouveia et al., 2004; Newill, 1974; Ritz et al., 2002; Salam et al., 2005).

14.5 REGULATORY BACKGROUND The National Ambient Air Quality Standards (NAAQS) for CO were promulgated by the Environmental Protection Agency (EPA) in 1971 at levels of 9 ppm (10 mg/m3) for an 8 h average and 35 ppm (40 mg/m3) for a 1 h average, not to be exceeded more than once per year. (Primary and secondary standards were established at identical levels.) The 1970 CO criteria

504

CARBON MONOXIDE

document (NAPCA, 1970) cited as the standard’s scientific basis a study that indicated that subjects exposed to low levels of CO, resulting in COHb concentrations of 2–3% of saturation, exhibited neurobehavioral effects (Beard and Wertheim, 1967). A reexamination of the scientific evidence, as reported in a revised CO criteria document (U.S. EPA, 1979), concluded that it was unlikely that significant, and repeatable, neurobehavioral effects occurred at COHb concentrations below 5%. Medical evidence, accumulated during the intervening years, however, indicated that aggravation of angina pectoris, and other symptoms of myocardial ischemia, occurred in men with chronic cardiovascular disease, exposed to low levels of CO resulting in COHb concentrations of about 2.7% (Anderson et al., 1973; Aronow et al., 1972; Aronow and Isbell, 1973; Aronow, 1974, 1979, 1981; Goldsmith and Aronow, 1975). EPA proposed, in 1980, based in part on the above studies, to retain the 8 h 9 ppm primary standard level, to reduce the 1 h primary standard from 35 to 25 ppm, and to revoke the secondary CO standards (because no adverse welfare effects had been reported at near-ambient levels). An EPA investigation found flaws in some of the Aronow studies from which data were used as part of the basis for the proposed reduction in the 1 h standard (Budiansky, 1983); EPA later decided to keep the 1 h standard at 35 ppm. In l984, EPA published an addendum to the 1979 CO criteria document that reevaluated the CO health effects data previously reviewed and took into account research that had been published in the interim (U.S. EPA, 1984). The document reviewed four effects associated with low-level CO exposure: cardiovascular, neurobehavioral, fibrinolytic, and perinatal. Dose–response data provided by controlled human studies allowed the following conclusions to be drawn: (a) Cardiovascular Effects. Among those with chronic cardiovascular disease, a shortening of time to onset of angina was observed at COHb concentrations of 2.9–4.5%. A decrement in maximum aerobic capacity was observed in healthy adults at COHb concentrations at and above 5%. Patients with chronic lung disease demonstrated a decrease in walking distance when COHb concentrations were increased from 1.1– 5.4% to 9.6–14.9%. (b) Neurobehavioral Effects. Decrements in vigilance, visual perception, manual dexterity, and performance of complex sensorimotor tasks were observed at, and above, 5% COHb. (c) Effects on Fibrinolysis. Although evidence existed linking CO exposure to fibrinolytic mechanisms, controlled human studies did not demonstrate consistent effects of carbon monoxide exposure on coagulation parameters. (d) Perinatal Effects. While there were some epidemiological associations between CO exposure and perinatal effects, such as low birth weight, slowed postnatal development, and incidences of sudden infant death syndrome (SIDS), the available data were not sufficient to establish causal relationships. In September 1985, EPA issued a final notice that announced the retention of the existing 8 h 9 ppm and 1 h 35 ppm primary NAAQS for CO and the rescinding of the secondary NAAQS for CO. The EPA reviewed health-related data in 1991 and completed the most recent CO criteria document in 1999 (U.S. EPA, 1999). In that interval, several controlled human clinical exposures, population-based studies, and inhalation studies using laboratory animal models were added to the available database. These studies have provided important insights into the

HEALTH EFFECTS OF CO

505

possible mechanisms of toxic action of CO, in addition to those related to hypoxia, and illuminate effects not currently identified in human studies, or that might not be amenable to controlled human experimentation, such as perinatal and developmental effects. Following review of the 1991 and the 1999 CDs, the existing NAAQS for CO were retained and are the current standards.

14.6 HEALTH EFFECTS OF CO 14.6.1

Population-Based Studies

14.6.1.1 Acute Exposures and Their Effects Most of the population-based studies in the literature relating to the health effects of CO in humans have been concerned with exposures to combustion and pyrolysis products from sources such as tobacco, fires, motor vehicle exhaust, home appliances fueled with wood (Ellegard, 1996; Pierson et al., 1989), gas, or kerosene (Amitai et al., 1998; Cooper and Alberti, 1984), and small engines (Baldauf et al., 2006). The individuals in these studies are therefore exposed to variable, and usually unmeasured, concentrations of CO and also to high concentrations of other combustion products. Exposures to CO in occupational settings represent another substantial exposure classification, but such exposures are also often accompanied by exposures to other contaminants as well. The symptoms of CO poisoning are often nonspecific, or masked by an exacerbation of an underlying illness, such as congestive heart failure. The effects can range from mild, annoying symptoms that resolve after removal of the source to severe morbidity with profound central nervous system dysfunction and acute complications. Acute CO intoxication often results in neurologic and/or myocardial injury. Studies have reported that 2% to approximately 10% of patients display delayed neurological sequelae (Choi, 1983; Mathieu et al., 1985; Raub et al., 2000b; Thom and Keim, 1989). Estimates suggest that about one-third of nonfatal cases of CO poisoning go undetected and undiagnosed (Abelsohn et al., 2002; Heckerling, 1987; Heckerling et al., 1987). CO poisoning, even when treated with supplemental oxygen (Mathieu and Mathieu-Nolf, 2005), can cause permanent neurocognitive or affective deficits; thus, increased awareness and prevention of CO poisoning is imperative (Weaver, 1999). The mechanism involved in delayed neurological damage after CO exposure was studied in rats. There were significant increases in glutamate release and
OH generation during and immediately after CO hypoxia, and CO-exposed rats showed learning and memory deficits that were associated with cell loss in the cortex, globus pallidus, and cerebellum. Both neuronal necrosis and apoptosis were observed, indicating that both necrosis and apoptosis contribute to brain cell death after acute CO poisoning (Piantadosi et al., 1997). This lends some mechanistic support to findings that Parkinsonism, which is an outcome of lesions or losses of dopaminergic neurons, may be associated with exposures to CO (Bleecker, 1988; Choi and Cheon, 1999; Choi, 2002). Necrosis of muscle tissue (myonecrosis) has been reported as possible but fairly unusual sequelae to CO exposure (16–20 cases have been reported in the English-language literature) (Herman et al., 1988; Shapiro et al., 1989; Waisma et al., 1998; Wolff, 1994). Some of the cases involve firefighters, and it is not clear that CO alone is a causal factor. Cyanide, which is a frequent cocontaminant in fires, has been suggested as a contributor to myonecrosis (Shapiro et al., 1989). Marius-Nunez (1990) reported a case of an individual who suffered an acute myocardial infarction (shown by ECG

506

CARBON MONOXIDE

and serum enzyme findings) after an acute CO exposure. This case was of interest because the patient’s medical profile was negative for coronary heart disease risk factors, and because a coronary angiogram performed 1 week after admission failed to show coronary obstructive lesions. A similar case was reported (Ebisuno et al., 1986), and the circumstances of both cases suggest that contributing factors to the CO-induced reduction in oxygen supply to the myocardium might include induction of coronary artery spasm, inadequate myocardial perfusion, and a direct toxic effect on myocardial mitochondria. Leikin and Vogel (1986) reported that patients admitted to intensive care units with proven myocardial infarctions had higher COHb levels than a control group, but these differences could have been accounted for by smoking alone, and a relationship to ambient urban CO could not be established. Sokal and Kralkowska (1985) examined 39 cases of acute CO poisoning in Poland. The subjects were poisoned at home by emissions from household gas or coal stoves. The authors found that the duration of CO exposure and the degree of metabolic acidosis, indicated by lactate concentrations in the blood, were better predictors of the clinical severity of symptoms than was the COHb concentration in blood at the time of admission to the hospital. The importance of exposure duration has been suggested in earlier evaluations of CO toxicity and is consistent with the possible involvement of myoglobin in CO toxicology. The prognosis for patients who survive acute CO poisoning is uncertain, particularly in those who develop delayed sequelae after their initial recovery (Deschamps et al., 2003; Ersanli et al., 2004; Gupta et al., 2005; Hwang and Park, 1996; Kanazawa and Yoshikawa, 2004; Kelafant, 1996; Lam et al., 2004; Scheinkestel et al., 1999; Shahbaz Hassan et al., 2003; Webber, 2003). For example, Lee and Marsden (1994) followed 31 patients with CO poisoning sequelae for a year (Lee and Marsden, 1994). Eight had a progressive course and four of the eight died. Twenty-three had a delayed relapse after an initial recovery period of approximately 20 days. Nine of these developed a Parkinsonian state with behavioral and cognitive impairment, but 14 of the cases progressed further and were bed-bound; the deterioration to either condition occurred rapidly over a few days to a week and three died. The mean initial CO hemoglobin level was not different in the two groups. Brain computed tomography (CT) scans were obtained at the onset of sequelae in both groups. Ten patients had a normal CT scan, 13 had white matter low-density lesions, and 4 had globus pallidus low-density lesions. The mechanisms for these sequelae may involve ischemia/reperfusion injury (Mathieu et al., 1996; Wattel et al., 1996) or cerebral biochemical and metabolic changes (Pall, 2001; Thom et al., 2004). 14.6.2

Chronic Exposures

14.6.2.1 Cardiovascular Effects Kristensen (1989) examined the relationship between cardiovascular diseases and exposures in the work environment and concluded that CO exposure increases the acute risk of cardiovascular disease, but that there was no lasting atherosclerotic effect. Stern et al. (1988) performed a retrospective study of heart disease mortality in 5529 bridge and tunnel officers. The socioeconomic and smoking characteristics of the two groups were well matched and the populations were limited to individuals who were assigned their positions and did not transfer between groups. The bridge officers experienced significantly lower CO exposures than the tunnel officers. Significantly elevated risk of coronary artery disease was found in the tunnel officers relative to the bridge officers (61 deaths observed versus 45 deaths expected); however, the risk declined after cessation of exposure, dissipating substantially after 5 years. Although convincing evidence from animal studies is lacking, CO may elevate plasma cholesterol and does appear to enhance

HEALTH EFFECTS OF CO

507

atherosclerosis when serum cholesterol is greatly elevated by diet (Penney and Howley, 1991). 14.6.2.2 Effects on Lung Function In addition to cardiovascular effects, individuals exposed to relatively high concentrations of CO in both indoor and outdoor environments may also be at risk of lung function decreases. It should be noted, however, that in addition to CO, these individuals were also exposed to high concentrations of other products of combustion and pyrolysis as well, and it is difficult to separate the effects of CO from those of these other compounds, many of which are known to be respiratory system irritants. Firefighters exhibit losses of lung function associated with acute and chronic smoke and CO exposure. Bronchoalveolar lavage fluid from firefighters after smoke exposure shows evidence of inflammation (Bergstrom et al., 1997), and decrements in function (days in which fires were fought compared to routine work shifts without fires) lasted for up to 18 h in some individuals (Sheppard et al., 1986). However, Slaughter et al. (2004) did not find a significant association between CO exposure and pulmonary function deficits in firefighters after exposures during controlled burns. In a study of matched populations of tunnel and bridge officers, whose primary job was to collect tolls, tunnel officers consistently had greater concentrations of COHb, compared to a population of bridge officers with a similar demographic profile that performed essentially similar work. However, the differences were small. Lung function measures of forced vital capacity (FVC) and forced expiration volume in 1 s (FEV1.0) were slightly reduced in tunnel versus bridge officers (Evans et al., 1988). No changes in FVC or FEV1.0 were observed in loggers who complained of dyspnea and eye, nose, and throat irritation after felling trees and cutting logs using chain saws (Hagberg et al., 1985). Exposures to typical ambient concentrations of CO, both outdoors and indoors, have not been significantly associated with pulmonary diseases or lung function decrements (Lebowitz et al., 1987), although other components of ambient pollution do show some significant associations (ozone, particulate matter), as do the use of gas stoves and tobacco smoking. 14.6.2.3 Effects on Pregnancy Outcomes Alderman et al. (1987) performed a case– control study of the association between low birth weight infants and maternal CO exposures in approximately 1000 cases in Denver. CO exposures were assigned to residential locations using fixed-site outdoor monitor data. After controlling for race and education (a surrogate for smoking behavior), no relationship was detected between the assigned CO exposure during the last 3 months of pregnancy and lower birth weights. The investigators suggested that failure to directly account for unmeasured sources of CO exposure, such as smoking, emissions from gas appliances, and exposures to vehicular exhaust, was a limitation of the study design. They also noted that the use of personal monitors for CO would have permitted a more direct evaluation of the potential relationship (exposure evaluations could be made after cases were identified, the relationship of personal to fixed-site assignments could be established, and then applied to the retrospective fixed-site data, author’s note). More recent studies have borne out the association between CO exposure and low birth weight. A 1 ppm average exposure during the first trimester of pregnancy was associated with a 23 g decrease in infant birth weight (Gouveia et al., 2004). Similar results have been obtained in Los Angeles, CA, and Sydney, Australia (Mannes et al., 2005; Ritz and Yu, 1999; Salam et al., 2005). Fetotoxicity has been demonstrated in laboratory animal studies at elevated (125 ppm) levels (Singh and Scott, 1984). Moderate CO exposure can alter neuron development and

508

CARBON MONOXIDE

modify neurochemical signaling in rats (Fechter, 1987). Prenatal exposure can adversely alter responses of dopaminergic neurons (Cagiano et al., 1998) and alter development of serotenergic and adrenergic neurons (Fechter et al., 1986; Storm and Fechter, 1985) that can lead to behavioral changes later in life (Fechter and Annau, 1980). 14.6.2.4 Exposure and Relationship to COHb Concentrations A study of over 1500 nonsmoking people sampled as part of the second National Health and Nutrition Examination Survey (NHANES II) demonstrated that CO concentrations measured at fixed-site monitors accounted for only 3% of the variance in blood COHb concentrations, using Spearman rank order correlations between the fixed-site monitor readings and sampled blood COHb concentrations (Wallace and Ziegenfus, 1985). They found that the correlations were not significant (p < 0.05) for 24 of the 36 sampling stations in 20 U.S. cities surveyed. The failure of 8 h average concentrations to correlate strongly with measured COHb concentrations indicates that outdoor monitoring data do not adequately reflect personal CO exposure. Using data from personal monitors worn by a probability sample of over 1500 residents of Denver and Washington, DC, Akland et al. (1985) found that over 10% of Denver residents and 4% of Washington residents were exposed during the wintertime to CO concentrations in excess of 9 ppm for 8 h, or longer. 14.6.3

Controlled Human Studies

Several clinically based studies have been published that have provided a relatively coherent picture of the effects of CO on the cardiopulmonary system. Some of the key studies cited in the 1991 and 1999 CO criteria documents (U.S. EPA, 1991, 1999), as well as those published since then, are described below. 14.6.4

Cardiovascular Effects

Individuals with ischemic heart disease have limited ability to compensate for increased myocardial O2 demands during exercise; hence, exercise testing is often used as a means for evaluating the severity of an individual’s cardiovascular impairment. Four useful parameters of ischemia that are measurable during exercise testing are ST segment depression (at least 1 mVof horizontal or downsloping depression of the ST segment of an electrocardiographic tracing persisting for 70 ms in three successive complexes); exercise-induced angina (chest pain during exercise, which is increased with effort and then resolves with rest—some individuals may experience pain in the jaw, neck, or shoulder areas); impaired work capacity (maximum work levels expressed as a percentage of nomographically predicted, normal values (Bruce, 1971, 1974, 1994); and an inadequate blood pressure response to exercise (blood pressure that falls on exercise (test would be discontinued) or fails to rise more than 15 mmHg at a work level of at least 40% of the predicted norm). There are some individuals who exhibit one or more of these responses during exercise who do not have abnormal coronary arteries, as determined by measuring luminal narrowing using angiographic methods; however, these parameters, taken in combination, can identify 85–90% of people with coronary artery disease (Allison et al., 1996). However, exercise testing alone has limitations with respect to its ability to predict future cardiac events (Fubini et al., 1992). Since CO exposure impairs myocardial O2 delivery, CO exposure would be expected to worsen symptoms of ischemia in individuals with coronary artery disease. Therefore,

HEALTH EFFECTS OF CO

509

exercise tests of such individuals have been an important means of providing quantitative and dose-related estimates of the potential impact of CO on health. Sheps et al. (1987) exposed 30 subjects with ischemic heart disease, aged 38–75 years, to CO (100 ppm) or air, during a 3-day, randomized, double-blind protocol to achieve an average postexposure COHb concentration of 3.8% on the CO exposure day (COHb on the air exposure day averaged 1.5%). After exposure to either CO or air, subjects performed an exercise stress test. All exercise tests were performed with the subjects in a supine position using a cycle ergometer at the same time of day, with the subjects in a fasting state. The workload was set at 0 for the first min, increased to 200 kpm for the next 4 min, and then increased in 50–100 kpm increments at 4 min intervals until a maximal level was achieved. Exercise was continued until anginal pain required cessation of exercise, fatigue precluded further exercise, or blood pressure plateaued or decreased, despite the increase in workload. All of the subjects were nonsmokers and had documented evidence of ischemic heart disease, defined by exercise-induced ST segment depression (1 mV or more), exerciseinduced angina, or abnormal left ventricular ejection fraction response to exercise (failure to increase 5 units from rest). Not all of the subjects in the study, which included both men and women, reported exercise-induced angina and the CO exposure produced only small, and not significant, decreases in time to onset of angina (1.9%) and maximal exercise time (1.3%) compared to air exposures (Sheps et al., 1987). Times to significant ST decreases, double product (DP; heart rate  systolic blood pressure) at significant ST depression, and maximal DP were similar for both air and CO exposure conditions. Double product in the absence of arterial obstructions can be used to estimate myocardial O2 consumption during dynamic exercise (Sim and Neill, 1974). The change in ejection fraction (rest to maximal) was slightly lower for CO exposures (air ¼ 3.5%, CO ¼ 2%; p ¼ 0.049). The authors concluded that there were no clinically significant effects of low-level CO exposures at COHb concentrations of 3.8%. Adams et al. (1988) subsequently extended the above study to an average postexposure COHb concentration of 5.9%, during exercise, using an identical protocol and 30 subjects (22 men, 8 women; mean age 58 years). Not all of the subjects in this study experienced exercise-induced angina, and only 21 subjects reported angina on both exposure days. The time to onset of angina in these 21 subjects was slightly, but not significantly, decreased after CO exposure (10.3%) compared to air exposure. An actuarial analysis of the data, from all subjects reporting angina, indicated that subjects were likely to experience angina earlier during stress on the CO exposure day (p  0.05). The left ventricular ejection fractions at rest were the same after both air and CO exposures; however, the level of submaximal ejection fraction was significantly higher after air, when compared to the CO exposure (3.3%; p  0.05), and the change in ejection fraction, from rest to submaximal exercise, was significantly lower after CO exposure, compared to air exposure (air ¼ 1.6% and CO ¼ 1.2%; p  0.05). No statistically significant exposure-related differences were seen for either maximal ST segment depression, time to onset of significant ST segment depression, or maximal DP. The authors concluded that exposures to CO resulting in COHb concentrations of about 6% significantly impaired exercise performance in subjects with ischemic heart disease. Kleinman et al. (1989) exposed 24 nonsmoking male subjects with stable angina and positive exercise tests to 100 ppm CO or air to achieve an average COHb concentration of 2.9%, during exercise, on the CO exposure day. Subjects ranged in age from 51 to 66 years, with a mean age of 59 years. All but one of the subjects had additional confirmation of ischemic heart disease, such as previous myocardial infarction, coronary artery

510

CARBON MONOXIDE

bypass surgery, positive thallium isotope exercise test, or a positive angiogram or cardiac catheterization. Subjects were exposed to CO or to clean air in a randomized, double-blind protocol. Subjects performed an incremental exercise test on a cycle ergometer until the point at which they could detect the onset of their typical anginal pain, and then stopped exercising. Workload was set at 50 W initially and was increased in 25 W increments at 3 min intervals. Blood pressure was measured at the end of each 3 min of exercise, ECG tracings were taken at the end of each minute, and respiratory gas exchange was measured at 15 s intervals and averaged for each minute. Data were analyzed statistically using a twofactor analysis of variance and one-tailed tests of significance. The time to onset of angina was decreased after CO exposure (5.9%; p ¼ 0.046) relative to air exposure. The duration of angina was longer after CO exposure compared to air exposure (8.3%), but this change was not statistically significant. Oxygen uptake at the angina point was slightly reduced after CO exposure compared to air exposure (2.2%; p  0.04), but the increase in O2 uptake with increasing workload was similar on both exposure days. A subgroup of 11 subjects who, in addition to angina, exhibited arrhythmias or ST segment depressions during exercise showed a greater reduction in time to angina after CO exposure, compared to air exposure (10.6%; p  0.016), than did the overall group. The time to significant ST segment depression was significantly reduced for the eight subjects with this characteristic after CO exposure, compared to air exposure (19.1%; p  0.044). The number of subjects exhibiting exerciseinduced ST segment depression identified in this study was small; however, those subjects in whom angina preceded detection of ST segment changes would not have been identified in the protocol used because exercise was stopped at the point of onset of angina. A large multicenter CO exposure study was conducted in three different cities (Allred et al., 1989a, 1989b, 1991). Sixty-three men with documented coronary artery disease underwent exposure to air, 117 ppm CO, or 253 ppm CO, on three separate days in a randomized, double-blind protocol, followed by an incremental treadmill exercise test. Average COHb concentrations of 2.2% and 4.3%, during exercise, were achieved on the two CO exposure days (2.0% and 3.9%, respectively, at the end of exercise). All of the subjects were males, aged 41–75 years (mean age of 62 years), with stable exertional angina and a positive exercise stress test with ST segment changes indicative of ischemia. In addition, all of the subjects had objective evidence of coronary artery disease indicated by at least one of the following:(1) angiographic evidence of at least 70% obstruction in one or more coronary arteries; (2) previous myocardial infarction; and (3) a positive thallium stress test. On each of the exposure days, the subject performed a symptom-limited treadmill exercise test, was exposed to one of the three test atmospheres (clean air, 117 ppm CO, or 253 ppm CO), and then performed a second exercise test. The subjects exercised until the subjects (1) were too fatigued to continue; (2) experienced severe dyspnea; (3) experienced grade 3 angina (on a subjective scale where grade 1 indicated the first perception of angina and grade 4 represented the worst angina the subject had ever experienced); (4) exhibited ECG changes (ST depression 3 mV or important arrhythmias); (5) high systolic (240 mmHg) or diastolic (130 mmHg) blood pressure; (6) a 20 mmHg drop in systolic blood pressure; or (7) a request by the subject. The time to onset of angina and the time to significant ST depression were determined for each test, and the percent changes (preexposure versus postexposure) for the two CO exposure days were compared to the same subject’s response to the randomized clean air exposure. The time to onset of angina was significantly reduced by CO exposure, in a dose-dependent manner (4.2% at 2% COHb, p ¼ 0.054; 7.1% at 4% COHb, p ¼ 0.004). Linear regressions of time to angina versus COHb concentrations for

HEALTH EFFECTS OF CO

511

each subject indicated that time to angina decreased 1.9  0.8% for every 1% increase in COHb (p  0.01). The time to onset of 1 mV ST segment depression was also reduced by CO in a dose-dependent manner (5.1% at 2% COHb, p ¼ 0.02; 12.1% at 4% COHb, p  0.0001) compared to the clean air exposure. There was a decrease of approximately 3.9  0.6% in time to ST depression for every 1% increase in COHb (p  0.0001). There was a significant correlation between the percent change in the time to onset of angina and the time to onset of ST depression 1 mV ( p  0.0001). There is some evidence that acute hypoxia that can result in myocardial ischemia and reversible angina can also lead to arrhythmias (Dahms et al., 1993; Farber et al., 1990; Jacobs and Nabarro, 1970). Hinderliter et al. (1989) exposed 10 subjects, with ischemic heart disease and no ventricular ectopy at baseline, to air, 100 ppm CO, and 200 ppm CO; COHb concentrations averaged 4% and 6% on the two respective CO exposure days. The exposures were randomized and double blinded. Following exposure, each subject performed a symptom-limited supine exercise test; ambulatory electrocardiograms were obtained prior to exposure, during exposure, during exercise, and over a 5 h postexercise period. The ECGs were analyzed for the frequency and severity of arrhythmias. Eight of the 10 subjects demonstrated evidence of ischemia on one or more of the exposure days (angina, 1 mV ST segment depression, or abnormal ejection fraction response). There were no CO-related increases in the frequency of premature ventricular beats and no multiple arrhythmias occurred. The authors concluded that low-level CO exposure (4–6% COHb) was not arrhythmogenic in patients with coronary artery disease and no ventricular ectopy at baseline. However, researchers from the same team (Sheps et al., 1990) reported on a larger study population (41 subjects) with some evidence of ventricular ectopy, exposed to air, 100 ppm CO, and 200 ppm CO in a similar protocol to that described above. The frequency of single ventricular premature depolarizations (VPDs) per hour increased (p  0.03) from 127  28 (mean  SD) after the air exposure to 168  38 after exposure to achieve a COHb concentration of 6%. The frequency of multiple VPDs per hour increased approximately threefold during exercise at 6% COHb, compared to air exposure (p  0.02). No significant differences in these parameters occurred after exposures that achieved COHb concentrations of 4%, compared to air exposures. The subjects who exhibited single VPDs with increased frequency after CO exposure were significantly older than the subjects who had no increased arrhythmias. The subjects who exhibited increased frequencies of multiple VPDs were older, exercised for longer durations, and had higher peak workloads during exercise than those who did not have complex arrhythmias. Leaf and Kleinman (1996) and Kleinman et al. (1998) have also reported evidence of effects of CO exposure on cardiac rhythm after relatively low CO exposures (3% COHb) in a small group of volunteers with coronary artery disease that exhibited abnormal rhythms on one or more exercise test. In all of the above clinical studies of CO-related effects, subjects with coronary artery disease were maintained on individualized regimens of medications, some of which might interact with CO-induced responses, increasing the apparent variations in observed responses. Specifically, blockade of beta-adrenergic receptors (Melinyshyn et al., 1988) and alpha-adrenergic receptors (Villeneuve et al., 1986) was shown to modify hemodynamic responses to CO in animal studies. Examination of the potential influence of medications on observed responses to CO could provide additional insights on the possible mechanisms of action of CO in individuals with coronary artery disease.

512

14.6.5

CARBON MONOXIDE

Cardiopulmonary Effects (Lung Function and Exercise Tolerance)

14.6.5.1 Normal Individuals Reduction of O2 delivery could reduce the ability to perform work in healthy individuals. Studies of the cardiopulmonary effects of CO have demonstrated that maximal O2 uptake during exercise (VO2 max ) decreases linearly with increasing COHb concentrations ranging from 2.3% to 35% COHb, in normals (Horvath, 1981; Horvath et al., 1988b; Shephard, 1984). The linear relationship can be expressed as percent decrease in VO2 max ¼ 0:91 [% COHb] þ 2.2. Changes in VO2 max are significant because they represent changes in an individual’s maximal aerobic exercise (or work) capacity (Ekblom et al., 1975). Klausen et al. (1983) exposed 16 male smokers to CO (5.26% COHb) and compared the effects on maximal exercise performance to performance after 8 h without smoking and performance after smoking three cigarettes (4.51% COHb). Both exposures reduced VO2 max by about 7%, but exercise time was decreased more after cigarette smoking than after CO exposure, suggesting that other components of smoke may contribute to the observed effects (Ekblom and Huot, 1972; Klausen et al., 1983). In a controlled laboratory test, 23 subjects (11 male, 12 female) (Horvath et al., 1988a) were exposed to 0, 50, 100, and 150 ppm CO at four different simulated altitudes (55, 1524, 2134, and 3048 m); following each exposure, an incremental exercise test was performed. COHb concentrations ranged from 0.5  0.2% to 5.6  0.4% of saturation after sea level exposures. The study showed a significant effect of increased altitude on decreased work performance and VO2 max . CO exposure tended to slightly decrease these parameters at all altitudes; however, the statistical analyses did not demonstrate a CO  altitude interaction, suggesting that these factors acted independently, and perhaps additively, but not synergistically. The female subjects appeared to be more resistant to the hypoxic effects of altitude than the male subjects. The rate of CO uptake (i.e., formation of COHb) decreased with increasing altitude, in part due to the reduced driving pressure of CO at altitude. In this study, significant fractions of CO were moved to extravascular spaces during exercise, probably in temporary combination with myoglobin, when exercise levels exceeded 80% of VO2 max (i.e., COHb concentrations increased 5 min postexercise compared to concentrations measured at the point of maximum workload). While this might suggest a mechanism in which CO might act in part by directly affecting cardiac myoglobin, evidence for direct cardiotoxicity of CO is still lacking. Horvath and Bedi (1989) demonstrated that long-term, low-level (9 ppm for 8 h) exposures at 2134 m result in lower COHb concentrations than the same exposure at 55 m, again suggesting slower CO uptake during altitude exposure. However, endogenous CO production is increased in rats chronically maintained at high altitudes (1000–6000 m) (McGrath, 1989, 1992), suggesting that high-altitude residents have higher initial COHb concentrations and might therefore achieve 2% or greater COHb levels (the COHb level associated with the CO NAAQS) more quickly than sea level residents. It has been reported that unacclimated workers exposed to about 25 ppm CO at an altitude of 2.3 km above sea level exhibited significantly increased symptoms of headache, vertigo, fatigue, weakness, memory impairment, insomnia, and heart palpitations compared to local residents (Song, 1993). The subjects in these human clinical studies of exercise tolerance have been relatively young and all were in good health. There is not sufficient information available to determine if relationships between CO exposure, altitude, and COHb concentrations would be similar for individuals with coronary artery disease, chronic lung diseases, anemia, or in pregnant women. Kleinman et al. (1998) and Leaf and Kleinman (1996) have demonstrated that hypoxia due to high altitude and CO exposure may cause additive effects on exercise tolerance,

HEALTH EFFECTS OF CO

513

hemodynamic changes, and cardiologic parameters. The subjects in this study were older men with confirmed coronary artery disease. 14.6.5.2 Individuals with Chronic Obstructive Pulmonary Disease Individuals with COPD usually have limited exercise tolerance because they have low ventilatory capacity, which can result in desaturation of arterial blood and hypoxemia (a relative deficiency of O2 in the blood) and hypoxia (a relative deficiency of O2 in some tissue) during exercise. Exercise performance in such individuals can be improved by providing supplemental O2 (Knower et al., 2001; Kramer et al., 1999). Reduced O2-carrying capacity of blood due to formation of COHb could exacerbate this limitation; hence, individuals with COPD could represent a potentially sensitive group. Aronow et al. (1977) exposed 10 men with COPD, aged 53–67 years to 100 ppm CO for 1 h, achieving increases in COHb from baseline concentrations of 1.4% to postexposure concentrations of 4.1%. Mean exercise time was reduced by 33%. Calverley et al. (1981) exposed six smokers (who stopped smoking 12 h prior to testing) and nine nonsmokers to 200 ppm CO for 20–30 min (increasing COHb concentrations to between 8% and 12% COHb above baseline COHb) and measured the distance each subject walked in a 12 min period. All of the subjects had severe bronchitis and emphysema. Significant decreases in walking distance were seen in individuals with 12.3% COHb or greater (levels that are seen in smokers with COPD). Individuals with severe COPD, even without clinically apparent coronary artery disease, may exhibit exercise-related cardiac arrhythmias. The exercise-induced arrhythmias were associated with arrhythmias at rest, but were not related to the severity of pulmonary disease, O2Hb desaturation, or ECG evidence of chronic lung disease (Cheong et al., 1990). The Sheps et al. (1990, 1991) studies of exercise-related arrhythmias in CO-exposed subjects with coronary artery disease suggest that COPD subjects might be important to study, as well, if they have baseline ectopy. Overall, the information available on individuals with COPD is consistent with the hypothesis that they represent a population potentially at risk of CO-related health effects during submaximal exercise, as may occur during normal daily activities. The available data are, however, based on population group sizes that are too small and too diverse with respect to disease characteristics to draw firm conclusions. 14.6.5.3 Neurotoxicological and Behavioral Effects The neurologic effects of relatively high-level acute CO exposures have been well documented (Gilbert and Glaser, 1959; Lacey, 1981; Remick and Miles, 1977). Subtle neurotoxic effects associated with lower level CO exposures may be underreported or not associated with CO exposure because the symptoms, which resemble those of a flu-like viral illness, may be misdiagnosed (Ares et al., 2001; Balzan et al., 1996; Foster et al., 1999; Raub et al., 2000a). Population-based studies on the potential neurotoxicological and behavioral effects of chronic CO exposure at ambient concentrations have not been reported. However, several clinical studies of CO-related sensory effects that evaluated several different parameters, under controlled laboratory conditions, showed little or no effect at COHb levels up to 17%. Hudnell and Benignus (1989) demonstrated, in a double-blind study, that visual function in healthy, young adult males, as defined by measurements of contrast threshold, luminance threshold, and time of cone/rod break, was not affected by COHb concentrations maintained at 17% for over 2 h. von Restorff and Hebisch (1988) reported no changes in time to dark adaptation and sensitivity after adaptation, at COHb concentrations ranging from 9% to 17%. One earlier study had demonstrated CO-induced visual threshold effects, that is, a slowing of dark adaptation (McFarland, 1973). However, the number of subjects tested was

514

CARBON MONOXIDE

small and documentation of the study was scant. Recent studies of temporal resolution of the visual system associated with elevated COHb levels have been reported; however, those involved cigarette smoking studies. The normal capacity for increased blood flow velocity in the central retinal artery in darkness was markedly reduced in smokers, which might explain the reduced dark vision after recent smoking reported in some studies. But this could reflect the combined effects of an increased blood viscosity and the vasoconstrictive action of nicotine in addition to the reduced capacity of the blood to transport O2 due to COHb (Havelius and Hansen, 2005). In general, neurotoxicity at COHb levels near 5% has not been convincingly demonstrated in normal healthy adults (Benignus et al., 1987) even though several early studies had suggested possible effects on critical flicker fusion at COHb levels at or below 9% (Seppanen et al., 1977; Vonpost-Lingen, 1964; Weber et al., 1975). Benignus et al. (1987a) exposed 24 healthy nonsmoking males to 0 or 100 ppm CO for 4 h (mean COHb ¼ 8%). They measured the subject’s ability to perform fast and slow tracking tasks (maintaining the position of a moving point of light on a computer screen using a joystick) and monitoring tasks (judging the brightness of two red spots on a computer screen) once per hour during exposure. CO exposure increased tracking errors, but did not interfere in the monitoring task. An earlier study demonstrated significant decrements in both tracking and monitoring tasks at a COHb concentration of 4.6%, but not at 3.5% (Putz et al., 1979). A large number of studies have investigated the effects of CO on other behavioral parameters; however, effects in general are only seen at COHb concentrations above 5%, and there are inconsistencies among the study results. Other studies, published in 1984 and later, showed interactive effects of exercise and CO exposure (47% COHb) on cognitive tasks (Bunnell and Horvath, 1988), but no changes in visually evoked response potentials in young (23 years) and older (69 years) subjects were observed at 5.3% COHb (Harbin et al., 1988). 14.6.5.4 Fetal Developmental and Perinatal Effects There are both theoretical reasons and supporting experimental data that indicate that the fetus may be more susceptible to the effects of CO than the mother. Fetal Hb has greater affinities for CO and O2 than does maternal Hb. The partial pressure of O2 in fetal blood is about 20–30% of that in maternal blood, because of the greater O2 affinity of fetal Hb. In addition, COHb shifts the O2Hb dissociation curve to the left in maternal blood, reducing the transfer of O2 across the placenta from maternal to fetal circulation. As in adults, the nervous and cardiovascular systems of the fetus are the most sensitive to the effects of CO. For humans, information is available for women who smoked during pregnancy or were acutely exposed to CO; however, most of the available reports do not characterize the relevant CO exposure levels and cannot, in general, rule out toxic effects of cocontaminants. Acute CO exposure may play a role in fetal death (Caravati et al., 1988), and environmental exposures, as well as maternal smoking, have been linked to sudden infant death syndrome in some (Hoppenbrouwers et al., 1981; Hutter and Blair, 1996), but not all studies (Variend and Forrest, 1987). Prenatal exposure to CO affects cholinergic and catecholaminergic pathways in the medulla of the guinea pig fetus, particularly in cardiorespiratory centers, regions thought to be compromised in SIDS (Tolcos et al., 2000). Additional animal studies suggest that high-level maternal CO exposures can have other significant neurotoxicological consequences for the fetus including disruption of neuronal proliferation and possible disruption of markers of neurochemical transmission (Fechter, 1987). Neonatal mortality and low birth weights are more prevalent in children born in high-altitude regions (Moore, 1987; Unger et al., 1988; Yip, 1987), suggesting high-altitude hypoxia interuterine growth, and further suggesting

SUMMARY AND CONCLUSIONS

515

that low birth weights in children born to women who smoke during pregnancy could possibly be a result of CO-induced hypoxia. Immune system changes have also been noted in rats exposed to CO prenatally; however, the changes may be reversible (Giustino et al., 1993). 14.6.5.5 CO as a Risk Factor in Cardiovascular Disease Development Evidence from population-based studies indicates that workers exposed to CO in combination with other combustion products from automobile exhaust (Stern et al., 1988) and other workers as well (Kristensen, 1989) have increased risk of development of atherosclerotic heart disease. Also, individuals hospitalized for myocardial infarction frequently exhibit higher COHb concentrations than individuals hospitalized for other reasons (Leikin and Vogel, 1986). Central to the development of atheromatous plaques is the deposition and retention of fibrinogen and lipids within the arterial wall. It is known that cigarette smoke increases the permeability of the arterial wall to fibrinogen. Allen and colleagues (Allen et al., 1989; Allen and Browse, 1990) demonstrated in a canine model that both CO and nicotine in cigarette smoke can produce an atherogenic effect, but they act via different mechanisms. CO increases arterial wall permeability and nicotine reduces clearance of deposited fibrinogen. Activation and dysfunction of blood platelets is associated with production of chemokines that elicit the migration of smooth muscle and inflammatory cells into the vascular intima, which is a major factor in the process of atherogenesis (Munro and Cotran, 1988; Nomoto et al., 1988; Weber, 2005) and in cardiac-related sudden deaths due to the role of platelets in the initiation of thrombosis (Harker and Ritchie, 1980; Meade, 1992). Studies have reported biochemical evidence that cigarette smoking induced both platelet and vascular dysfunctions in apparently healthy individuals (Folts et al., 1990; Krupski, 1991). Production of platelet-derived growth factor (PDGF) by endothelial cells is upregulated in response to hypoxia and is a major growth factor for vascular smooth muscle cells and a powerful vasoconstrictor (Humar et al., 2002; Kourembanas et al., 1990). Platelet dysfunction may also be a contributory cause of thrombosis during pregnancy and may increase fetal mortality and morbidity among women who smoke (Davis et al., 1987). Abnormalities in platelet aggregation occur after CO exposure (Mansouri and Perry, 1982) and may be linked to guanylate cyclase activation (Brune and Ullrich, 1987). When 10 healthy nonsmokers were exposed passively to cigarette smoke (in hospital corridors), resulting in a small increase in COHb concentration from 0.9  0.3% to 1.3  0.6%, before and after passive exposure, respectively (Davis et al., 1989), they showed evidence of changes in platelet aggregation and endothelial cell damage. The changes in endothelial cell counts (preexposure to postexposure) were significantly correlated to changes in COHb concentrations from before to after exposure, but plasma nicotine levels were not. The contribution of CO relative to other components of tobacco smoke in causing platelet dysfunction is not established.

14.7 SUMMARY AND CONCLUSIONS The current CO ambient air standards are designed to protect susceptible individuals from exposures that would result in COHb concentrations of 2% and above. Occupational standards are designed to protect workers from concentrations of 5% COHb. Studies of individuals with coronary artery disease and residents of New York, NY, Denver, CO, Washington, DC, and Los Angeles, CA, suggest that susceptible individuals frequently

516

CARBON MONOXIDE

exceed 2% COHb in cities that frequently exceed NAAQS. Control of exposures is difficult because the sources of CO are widespread, the distribution of ambient CO is very nonuniform, and emissions from unregulated sources, especially indoors, probably contribute substantially to individual CO doses. The distributions of COHb concentrations in workers are also very nonuniform but may often reach the 5% level. The contribution of CO to the aggravation of symptoms of myocardial ischemia is reasonably well defined for a selected subset of people with existing coronary artery disease. The individuals comprising the populations tested in the various studies on which this conclusion was drawn were carefully selected to have sufficiently pronounced disease such that effects would be measurable, but they were also sufficiently healthy so that they could perform moderate levels of exercise with minimal risk. Thus, more impaired individuals, who might presumably be at equal or greater risk of detrimental CO-induced health effects, and relatively asymptomatic individuals, so-called silent ischemics, have not been well characterized. Incorporation of broader, possibly more representative, subject populations into the clinical studies of Sheps et al. (1990) and Adams et al. (1988) significantly increased the variance in subject responses and increased the difficulty of attributing statistical significance to observed findings. As shown in Figure. 14.1, there is a reasonable dose– response relationship over the range of 2–6% COHb for the decrease in time to onset of angina in data from five independent studies in which subjects with documented coronary artery disease were exposed to CO and then performed symptom-limited exercise tests. Convincing documentation for effects of CO on other potentially susceptible individuals at ambient exposure levels is becoming available. The most extensive body of evidence of CO effects on pregnant women, fetuses, and neonates comes from the literature on smoking and from acute, high-level accidental CO exposures. In most cases, actual CO exposures are poorly, if at all, documented, and the contribution of copollutants to the observed effects cannot be assessed. However, animal studies demonstrating developmental changes and associations between environmental CO and SIDS indicate that risks to pregnant women, fetuses, and neonates may be important.

FIGURE 14.1 Reduction in time to angina (TTA) following CO exposure in subjects with coronary artery disease. Linear regression shows that TTA is reduced in a dose-dose-dependent manner. Values shown are mean  SE.

REFERENCES

517

The importance of occult CO exposures leading to clinically significant symptoms and effects is becoming well appreciated. The large number of such incidents suggests the potential that there may be many undetected incidents that lead to subclinical manifestations but are ignored if they are not serious enough to prevent relatively normal daily activities. The home environment is very poorly characterized with respect to indoor pollutant levels, and given the apparently large potential for CO-related health effects, the home indoor environment should be the focus of significant new study initiatives. It would seem, from this study, that both occupational and ambient standards are placed near the limits at which significant effects are seen, albeit in sensitive individuals, thus affording a narrow, if any, margin of safety. The number of studies suggesting roles for CO in the development of cardiovascular disease and in infant mortality is increasing, but not yet conclusive. Additional studies under well-controlled conditions, with accurate estimates of CO exposure history, should be a priority.

ACKNOWLEDGMENTS This study was funded in part by the California Air Resources Board and the UCI Center for Occupational and Environmental Health.

REFERENCES Abelsohn A, et al. (2002) Identifying and managing adverse environmental health effects. 6. Carbon monoxide poisoning. Can. Med. Assoc. J. 166(13):1685–1690. Abu-al Ragheb SY, Battah AH (1999) Carbon monoxide fatalities in medicolegal autopsies. Med. Sci. Law 39(3):243–246. Adams KF, et al. (1988) Acute elevation of blood carboxyhemoglobin to 6% impairs exercise performance and aggravates symptoms in patients with ischemic heart disease. J. Am. Coll. Cardiol. 12(4):900–909. Akland GG, et al. (1985) Measuring human exposure to carbon monoxide in Washington, DC, and Denver, Colorado, during the Winter of 1982–1983. Environ. Sci. Technol. 19(10):911–918. Alderman BW, Baron AE, Savitz DA (1987) Maternal exposure to neighborhood carbon monoxide and risk of low infant birth weight. Public Health Rep. 102(4):410–414. Allen DR, Browse NL (1990) The effect of cigarette smoke, nicotine and carbon monoxide on arterial wall permeability and arterial wall uptake of 125I-fibrinogen. Adv. Exp. Med. Biol. 273:95–106. Allen DR, Browse NL, Rutt DL (1989) Effects of cigarette smoke, carbon monoxide and nicotine on the uptake of fibrinogen by the canine arterial wall. Atherosclerosis 77(1):83–88. Allison T, et al. (1996) Cardiovascular stress testing: a description of the various types of stress tests and indications for their use. Mayo Clin. Proc. 71(1):43–52. Allred EN, et al. (1989a) Short-term effects of carbon monoxide exposure on the exercise performance of subjects with coronary artery disease. N. Engl. J. Med. 321(21):1426–1432. Allred EN, et al. (1989b) Acute effects of carbon monoxide exposure on individuals with coronary artery disease. Res. Rep. Health Eff. Inst. (25):1–79. Allred EN, et al. (1991) Effects of carbon monoxide on myocardial ischemia. Environ. Health Perspect. 91:89–132. Amitai Y, et al. (1998) Neuropsychological impairment from acute low-level exposure to carbon monoxide. Arch. Neurol. 55(6):845–848.

518

CARBON MONOXIDE

Anderson EW, et al.(1973)Effectoflow-levelcarbon monoxide exposure ononsetandduration ofangina pectoris. A study in ten patients with ischemic heart disease. Ann. Intern. Med. 79(1):46–50. Angle CR (1988) Indoor air pollutants. Adv. Pediatr. 35:239–280. Anonymous (1997) Use of unvented residential heating appliances—United States, 1988–1994. MMWR Morb. Mortal. Wkly. Rep. 46(51):1221–1224. Ares B, et al. (2001) Headache secondary to carbon monoxide poisoning. Rev. Neurol. 32(4):339–341. Aronow WS (1974) Effects of carbon monoxide on coronary heart disease. Geriatrics 29(10):141– 146. Aronow WS (1979) Effect of carbon monoxide on cardiovascular disease. Prev. Med. 8(3):271–278. Aronow WS (1981) Aggravation of angina pectoris by two percent carboxyhemoglobin. Am. Heart J. 101(2):154–157. Aronow WS, Isbell MW (1973) Carbon monoxide effect on exercise-induced angina pectoris. Ann. Intern. Med. 79(3):392–395. Aronow WS, et al. (1972) Effect of freeway travel on angina pectoris. Ann. Intern. Med. 77(5):669– 676. Aronow WS, Ferlinz J, Glauser F (1977) Effect of carbon monoxide on exercise performance in chronic obstructive pulmonary disease. Am. J. Med. 63(6):904–908. Ayres SM, Giannelli S Jr, Mueller H (1970) Myocardial and systemic responses to carboxyhemoglobin. Ann. N.Y. Acad. Sci. 174(1):268–293. Baldauf R, et al. (2006) Air contaminant exposures during the operation of lawn and garden equipment. J. Expo. Sci. Environ. Epidemiol. 16:362–370. Balzan MV, Agius G, Debono AG (1996) Carbon monoxide poisoning: easy to treat but difficult to recognise. Postgrad. Med. J. 72(850):470–473. Beard RR, Wertheim GA (1967) Behavioral impairment associated with small doses of carbon monoxide. Am. J. Public Health 57(11):2012–2022. Benignus VA, et al. (1987a) Absence of symptoms with carboxyhemoglobin levels of 16–23%. Neurotoxicol. Teratol. 9(5):345–348. Benignus VA, et al. (1987b) Effect of low level carbon monoxide on compensatory tracking and event monitoring. Neurotoxicol. Teratol. 9(3):227–234. Benignus VA, et al. (1994) Prediction of carboxyhemoglobin formation due to transient exposure to carbon monoxide. J. Appl. Physiol. 76(4):1739–1745. Bergstrom CE, et al. (1997) Bronchoalveolar lavage findings in firefighters. Am. J. Ind. Med. 32 (4):332–336. Berny C, et al. (2002) Biomarkers of tobacco smoke exposure. Ann. Biol. Clin. (Paris). 60(3):263– 272. Bleecker ML (1988) Parkinsonism: a clinical marker of exposure to neurotoxins. Neurotoxicol. Teratol. 10(5):475–478. Boumba VA, Vougiouklakis T (2005) Evaluation of the methods used for carboxyhemoglobin analysis in postmortem blood. Int. J. Toxicol. 24(4):275–281. Brubaker PE, et al. (1975) Noble metals: a toxicological appraisal of potential new environmental contaminants. Environ. Health Perspect. 10:39–56. Bruce RA (1971) Exercise testing of patients with coronary heart disease—principles and normal standards for evaluation. Ann. Clin. Res. 3(6):323. Bruce RA (1974) Methods of exercise testing—step test, bicycle, treadmill, isometrics. Am. J. Cardiol. 33(6):715–720. Bruce RA (1994) Clinical exercise testing—a review of personal and community practice experience. Primary Care 21(3):405–414.

REFERENCES

519

Bruce EN, Bruce MC (2003) A multicompartment model of carboxyhemoglobin and carboxymyoglobin responses to inhalation of carbon monoxide. J. Appl. Physiol. 95(3):1235–1247. Brune B, Ullrich V (1987) Inhibition of platelet aggregation by carbon monoxide is mediated by activation of guanylate cyclase. Mol. Pharmacol. 32(4):497–504. Budiansky S (1983) Data falsification: innocents and culprits. Nature 303(5918):563. Bunnell DE, Horvath SM (1988) Interactive effects of physical work and carbon monoxide on cognitive task performance. Aviat. Space Environ. Med. 59(12):1133–1138. Cagiano R, et al. (1998) Effects of prenatal exposure to low concentrations of carbon monoxide on sexual behaviour and mesolimbic dopaminergic function in rat offspring. Br. J. Pharmacol. 125 (4):909–915. Calafat AM, et al. (2004) Determination of tar, nicotine, and carbon monoxide yields in the mainstream smoke of selected international cigarettes. Tob. Control 13(1):45–51. Calverley PM, Leggett RJ, Flenley DC (1981) Carbon monoxide and exercise tolerance in chronic bronchitis and emphysema. Br. Med. J. 283(6296):878–880. Campbell ME , et al. (2005) Should people be physically active outdoors on smog alert days? Can. J. Public Health 96(1):24–28. Caravati EM, et al. (1988) Fetal toxicity associated with maternal carbon monoxide poisoning. Ann. Emerg. Med. 17(7):714–717. Centers for Disease Control, (1982) Carbon monoxide poisoning- a preventable environmental health hazard Morb. Mort. Wkly. Rept. 31(39):529–531. Chaitman BR, et al. (1992) Carbon monoxide exposure of subjects with documented cardiac arrhythmias. Res. Rep. Health Eff. Inst. (52):1–26; discussion 27–37. Cheong TH, et al. (1990) Cardiac arrhythmias during exercise in severe chronic obstructive pulmonary disease. Chest 97(4):793–797. Choi IS (1983) Delayed neurologic sequelae in carbon monoxide intoxication. Arch. Neurol. 40 (7):433–435. Choi IS (2002) Parkinsonism after carbon monoxide poisoning. Eur. Neurol. 48(1):30–33. Choi IS, Cheon HY (1999) Delayed movement disorders after carbon monoxide poisoning. Eur. Neurol. 42(3):141–144. Chung SJ (1988) Formulas predicting carboxyhemoglobin resulting from carbon monoxide exposure. Vet. Hum. Toxicol. 30(6):528–532. Coburn RF, Forster RE, Kane PB (1965) Considerations of the physiological variables that determine the blood carboxyhemoglobin concentration in man. J. Clin. Invest. 44(11):1899–1910. Coburn RF, Williams WJ, Kahn SB (1966) Endogenous carbon monoxide production in patients with hemolytic anemia. J. Clin. Invest. 45(4):460–468. Cooper KR, Alberti RR (1984) Effect of kerosene heater emissions on indoor air quality and pulmonary function. Am. Rev. Respir. Dis. 129(4):629–631. Dahms TE, et al. (1993) Effects of carbon monoxide exposure in patients with documented cardiac arrhythmias. J. Am. Coll. Cardiol. 21(2):442–450. Davis RB, et al. (1987) Evaluation of platelet function in pregnancy. Comparative studies in nonsmokers and smokers. Thromb. Res. 46(2):175–186. Davis JW, et al. (1989) Passive smoking affects endothelium and platelets. Arch. Intern. Med. 149 (2):386–389. Deschamps D, et al. (2003) Memory one month after acute carbon monoxide intoxication: a prospective stud. Occup. Environ. Med. 60(3):212–216. Duci A, Chaloulakou A, Spyrellis N (2003) Exposure to carbon monoxide in the Athens urban area during commuting. Sci. Total Environ. 309(1–3)47–58.

520

CARBON MONOXIDE

Ebisuno S, et al. (1986) Myocardial infarction after acute carbon monoxide poisoning: case report. Angiology 37(8):621–624. Ekblom B, Huot R (1972) Response to submaximal and maximal exercise at different levels of carboxyhemoglobin. Acta Physiol. Scand. 86(4):474–482. Ekblom B, et al. (1975) Effect of changes in arterial oxygen content on circulation and physical performance. J. Appl. Physiol. 39(1):71–75. El-Fadel M, El-Hougeiri N (2003) Indoor air quality and occupational exposures at a bus terminal. Appl. Occup. Environ. Hyg. 18(7):513–522. Ellegard A (1996) Cooking fuel smoke and respiratory symptoms among women in low-income areas in Maputo. Environ. Health Perspect. 104(9):980–985. Ersanli D, et al. (2004) Visual loss as a late complication of carbon monoxide poisoning and its successful treatment with hyperbaric oxygen therapy. Swiss Med. Wkly. 134(43–44):650–655. Evans RG, et al. (1988) Cross-sectional and longitudinal changes in pulmonary function associated with automobile pollution among bridge and tunnel officers. Am. J. Ind. Med. 14(1):25–36. Farber JP, et al. (1990) Carbon monoxide and lethal arrhythmias. Res. Rep. Health Eff. Inst. (36):1–17; discussion 19–27. Fechter LD (1987) Neurotoxicity of prenatal carbon monoxide exposure. Res. Rep. Health Eff. Inst. (12):3–22. Fechter LD, Annau Z (1980) Prenatal carbon monoxide exposure alters behavioral development. Neurobehav. Toxicol. 2(1):7–11. Fechter LD, Mactutus CF, Storm JE (1986) Carbon monoxide and brain development. Neurotoxicology 7(2):463–473. Finklea JF, et al. (1975) Estimated changes in human exposure to suspended sulfate attributable to equipping light-duty motor vehicles with oxidation catalysts. Environ. Health Perspect. 1:29–34. Folts JD, et al. Effects of cigarette smoke and nicotine on platelets and experimental coronary artery thrombosisAdv. Exp. Med. Biol. 273:(1990)339–358. Foster M, et al. (1999) Recurrent acute life-threatening events and lactic acidosis caused by chronic carbon monoxide poisoning in an infant. Pediatrics 104(3) art. no.-e34. From Attebring M , et al. (2001) Are patients truthful about their smoking habits? A validation of selfreport about smoking cessation with biochemical markers of smoking activity amongst patients with ischaemic heart disease. J. Intern. Med. 249(2):145–151. Fubini A, et al. (1992) Value of exercise stress test, radionuclide angiography and coronary angiography in predicting new coronary events in asymptomatic patients after a 1st episode of myocardial infarction. Int. J. Cardiol. 34(3):319–325. Gilbert GJ, Glaser GH (1959) Neurologic manifestations of chronic carbon monoxide poisoning. New Engl. J. Med. 261(24):1217–1220. Giustino A, et al. (1993) Immunological changes produced in rats by prenatal exposure to carbon monoxide. Pharmacol. Toxicol. 73(5):274–278. Goldsmith JR, Aronow WS (1975) Carbon monoxide and coronary heart disease: a review. Environ. Res. 10(2):236–248. Gourgoulianis KI, et al. (2002) Indoor maternal smoking doubles adolescents’ exhaled carbon monoxide. Acta Paediatr. 91(6):712–713. Gouveia N, Bremner SA, Novaes HM (2004) Association between ambient air pollution and birth weight in Sao Paulo. Braz. J. Epidemiol. Community Health 58(1):11–17. Groman E , et al. (1998) Measurement of expired carbon monoxide among medical students to assess smoking behaviour. Soz. Praventivmed. 43(6):322–324. Guggisberg M , et al. (2003) Particulate matter and gaseous contaminants in indoor environments in an isolated northern community. Int. J. Circumpolar Health. 62(2):120–129.

REFERENCES

521

Gupta A, et al. (2005) Apparent suicidal carbon monoxide poisonings with concomitant prescription drug overdoses. J. Anal. Toxicol. 29(7):744–749. Hagberg M, et al. (1985) Irritative complaints, carboxyhemoglobin increase and minor ventilatory function changes due to exposure to chain-saw exhaust. Eur. J. Respir. Dis. 66 (4):240–247. Harbin TJ, et al. (1988) The effects of low-level carbon monoxide exposure upon evoked cortical potentials in young and elderly men. Neurotoxicol. Teratol. 10(2):93–100. Harker LA, Ritchie JL (1980) The role of platelets in acute vascular events. Circulation 62(6 Pt 2): V13–8. Hart BP (1993) Vascular consequences of smoking and benefits of smoking cessation. J. Vasc. Nurs. 11 (2):48–51. Hauck H, Neuberger M (1984) Carbon monoxide uptake and the resulting carboxyhemoglobin in man. Eur. J. Appl. Physiol. Occup. Physiol. 53(2):186–190. Havelius U, Hansen F (2005) Ocular vasodynamic changes in light and darkness in smokers. Invest. Ophthalmol. Vis. Sci. 46(5):1698–1705. Heckerling PS (1987) Occult carbon monoxide poisoning: a cause of winter headache. Am. J. Emerg. Med. 5(3):201–204. Heckerling PS , et al. (1987) Predictors of occult carbon monoxide poisoning in patients with headache and dizziness. Ann. Intern. Med. 107(2):174–176. Heckerling PS , et al. (1990a) Occult carbon monoxide poisoning in patients with neurologic illness. J. Toxicol. Clin. Toxicol. 28(1):29–44. Heckerling PS, et al. (1990b) Screening hospital admissions from the emergency department for occult carbon monoxide poisoning. Am. J. Emerg. Med. 8(4):301–304. Heilmann L, Rath W, Pollow K (2005) Fetal hemorheology in normal pregnancy and severe preeclampsia. Clin. Hemorheol. Microcirc. 32(3):183–190. Herman GD, Shapiro AB, Leikin J (1988) Myonecrosis in carbon monoxide poisoning. Vet. Hum. Toxicol. 30(1):28–30. Hinderliter AL , et al. (1989) Effects of low-level carbon monoxide exposure on resting and exerciseinduced ventricular arrhythmias in patients with coronary artery disease and no baseline ectopy. Arch. Environ. Health. 44(2):89–93. Hinkle LE Jr (1980) Automobile emissions in the perspective of human health: health benefits and social costs of pollution control. Bull. N.Y. Acad. Med. 56(9):948–979. Hoppenbrouwers T, et al. (1981) Seasonal relationship of sudden infant death syndrome and environmental pollutants. Am. J. Epidemiol. 113(6):623–635. Horner JM (2000) Anthropogenic emissions of carbon monoxide. Rev. Environ. Health 15(3):289– 298. Horvath SM (1981) Impact of air quality in exercise performance. Exerc. Sport. Sci. Rev. 9:265–296. Horvath SM, Bedi JF (1989) Alteration in carboxyhemoglobin concentrations during exposure to 9 ppm carbon monoxide for 8 hours at sea level and 2134 m altitude in a hypobaric chamber. J. Air Pollut. Control Assoc. 39(10):1323–1327. Horvath SM, et al. (1988a) Maximal aerobic capacity at several ambient concentrations of carbon monoxide at several altitudes. Res. Rep. Health Eff. Inst. (21):1–21. Horvath SM, et al. (1988b) Maximal aerobic capacity at several ambient concentrations of CO at several altitudes. J. Appl. Physiol. 65(6):2696–2708. Howell J, Keiffer MP, Berger LR (1997) Carbon monoxide hazards in rural Alaskan homes. Alaska Med. 39(1):8–11. Hudnell HK, Benignus VA (1989) Carbon monoxide exposure and human visual detection thresholds. Neurotoxicol. Teratol. 11(4):363–71.

522

CARBON MONOXIDE

Humar R, et al. (2002) Hypoxia enhances vascular cell proliferation and angiogenesis in vitro via rapamycin (mTOR)-dependent signaling. FASEB J. 16(8):771–780. Hutchinson EJ, Pearson PJ (2004) An evaluation of the environmental and health effects of vehicle exhaust catalysts in the UK. Environ. Health Perspect. 112(2):132–141. Hutter CD, Blair ME (1996) Carbon monoxide—does fetal exposure cause sudden infant death syndrome? Med. Hypotheses 46(1):1–4. Hwang JM, Park HW (1996) Carbon monoxide poisoning as an epigenetic factor for Leber’s hereditary optic neuropathy. Korean J. Ophthalmol. 10(2):122–123. Hysell DK, et al. (1975) Inhalation toxicology of automotive emissions as affected by an oxidation exhaust catalyst. Environ. Health Perspect. 10:57–62. Jacobs HS, Nabarro JD (1970) Arrhythmias during acute myocardial hypoxia. Lancet 1(7658):1224. Johansson MB, Wollmer P (1989) Measurement of carboxyhaemoglobin by spectrophotometry and gas chromatography. Clin. Physiol. 9(6):581–586. Joumard R, et al. (1981) Mathematical models of the uptake of carbon monoxide on hemoglobin at low carbon monoxide levels. Environ. Health. Perspect. 41:277–289. Jovanovic J, Jovanovic M, Dordevic D (1999) Professional exposure of drivers to carbon monoxide as a possible risk factor for the occurrence of traffic accidents in the road traffic. Vojnosanit. Pregl. 56 (6):587–592. Kanazawa H, Yoshikawa J (2004) Accelerated decline in lung function and impaired reversibility with salbutamol in asthmatic patients with chronic hepatitis C virus infection: a 6-year follow-up study. Am. J. Med. 116(11):749–752. Kanten WE, et al. (1983) Hemodynamic responses to acute carboxyhemoglobinemia in the rat. Am. J. Physiol. 244(3):H320–7. Kao LW, Nanagas KA (2004) Carbon monoxide poisoning. Emerg. Med. Clin. North Am. 22(4):985– 1018. Kelafant GA (1996) Encephalopathy and peripheral neuropathy following carbon monoxide poisoning from a propane-fueled vehicle. Am. J. Ind. Med. 30(6):765–768. Kirkpatrick JN (1987) Occult carbon monoxide poisoning. West. J. Med. 146(1):52–56. Klausen K, Andersen C, Nandrup S (1983) Acute effects of cigarette smoking and inhalation of carbon monoxide during maximal exercise. Eur. J. Appl. Physiol. Occup. Physiol. 51(3):371–379. Kleinman MT, et al. (1989) Effects of short-term exposure to carbon monoxide in subjects with coronary artery disease. Arch. Environ. Health 44(6):361–369. Kleinman MT, et al. (1998) Urban angina in the mountains: effects of carbon monoxide and mild hypoxemia on subjects with chronic stable anginaArch. Environ. Health 53(6):388–397. Knower MT, et al. (2001) Baseline oxygen saturation predicts exercise desaturation below prescription threshold in patients with chronic obstructive pulmonary diseaseArch. Intern. Med. 161 (5):732–736. Koskela RS, et al. (2000) Factors predictive of ischemic heart disease mortality in foundry workers exposed to carbon monoxide. Am. J. Epidemiol. 152(7):628–632. Kourembanas S, Hannan RL, Faller DV (1990) Oxygen tension regulates the expression of the platelet-derived growth factor-B chain gene in human endothelial cells. J. Clin. Invest. 86(2):670– 674. Kramer MR, et al. (1999) Quantitative 15 steps exercise oximetry as a marker of disease severity in patients with chronic obstructive pulmonary disease. Isr. Med. Assoc. J. 1(3):165–168. Kristensen TS (1989) Cardiovascular diseases and the work environment—a critical review of the epidemiologic literature on chemical factors. Scan. J. Work Environ. Health 15(4):245–264. Krupski WC (1991) The peripheral vascular consequences of smoking. Ann. Vasc. Surg. 5(3):291– 304.

REFERENCES

523

Lacey DJ (1981) Neurologic sequelae of acute carbon monoxide intoxication. Am. J. Dis. Child. 135 (2):145–147. Lam SP, et al. (2004) Delayed neuropsychiatric impairment after carbon monoxide poisoning from burning charcoal. Hong Kong Med. J. 10(6):428–431. Laranjeira R, Pillon S, Dunn J (2000) Environmental tobacco smoke exposure among non-smoking waiters: measurement of expired carbon monoxide levels. Sao Paulo Med. J. 118 (4):89–92. Leaf DA, Kleinman MT (1996) Urban ectopy in the mountains: carbon monoxide exposure at high altitude. Arch. Environ. Health 51(4):283–290. Lebowitz MD, Collins L, Holberg CJ (1987) Time series analyses of respiratory responses to indoor and outdoor environmental phenomena. Environ. Res. 43(2):332–341. Lee MS, Marsden CD (1994) Neurological sequelae following carbon monoxide poisoning clinical course and outcome according to the clinical types and brain computed tomography scan findings. Mov. Disord. 9(5):550–558. Leikin JB, Vogel S (1986) Carbon monoxide levels in cardiac patients in an urban emergency department. Am. J. Emerg. Med. 4(2):126–128. Leikin JB , et al. (1988) Carboxyhemoglobin levels in patients with flu-like symptoms. Ann. Emerg. Med. 17(4):383–384. Leone A (1995) Cigarette smoking and health of the heart. J. R. Soc. Health 115(6):354–355. Levesque B, et al. (2000) Air quality monitoring during indoor Monster Truck and car demolition shows. J. Expo. Anal. Environ. Epidemiol. 10(1):58–65. Levesque B, et al. (2005) Exposure to carbon monoxide during indoor karting. Int. J. Environ. Health Res. 15(1):41–44. Linn WS, Gong H Jr (1999) The 21st century environment and air quality influences on asthma. Curr. Opin. Pulm. Med. 5(1):21–26. Longo LD (1976) Carbon monoxide: effects on oxygenation of the fetus in utero. Science 194 (4264):523–525. Longo LD (1977) The biological effects of carbon monoxide on the pregnant woman, fetus, and newborn infant. Am. J. Obstet. Gynecol. 129(1):69–103. Mall T, et al. (1985) Influence of moderately elevated levels of carboxyhemoglobin on the course of acute ischemic heart disease. Respiration 48(3):237–244. Mann JK, et al. (2002) Air pollution and hospital admissions for ischemic heart disease in persons with congestive heart failure or arrhythmia. Environ. Health Perspect. 110(12):1247–1252. Mannes T, et al. (2005) Impact of ambient air pollution on birth weight in Sydney, Australia. Occup. Environ. Med. 62(8):524–530. Mansouri A, Perry CA (1982) Alteration of platelet aggregation by cigarette smoke and carbon monoxide. Thromb. Haemost. 48(3):286–288. Marius-Nunez AL (1990) Myocardial infarction with normal coronary arteries after acute exposure to carbon monoxide. Chest 97(2):491–494. Martys R (1994) Adverse cardiac effects of smoking. Wien. Med. Wochenschr. 144(22–23):556–560. Mathieu D, Mathieu-Nolf M (2005) Oxygen therapy for CO poisoning: rationale and recommendations. Przegl. Lek. 62(6):436–437. Mathieu D, et al. (1985) Acute carbon monoxide poisoning. Risk of late sequelae and treatment by hyperbaric oxygen. J. Toxicol. Clin. Toxicol. 23(4–6)315–24. Mathieu D, Mathieu-Nolf M, Wattel F (1996) Carbon monoxide poisoning: current aspects. Bull. Acad. Natl. Med. 180(5):965–971; discussion 72–73. McCartney ML (1990) Sensitivity analysis applied to Coburn–Forster–Kane models of carboxyhemoglobin formation. Am. Ind. Hyg. Assoc. J. 51(3):169–177.

524

CARBON MONOXIDE

McFarland RA (1973) Low level exposure to carbon monoxide and driving performance. Arch. Environ. Health 27(6):355–359. McGrath JJ (1989) Cardiovascular effects of chronic carbon monoxide and high-altitude exposure. Res. Rep. Health Eff. Inst. (27):1–23. McGrath JJ (1992) Effects of altitude on endogenous carboxyhemoglobin levels. J. Toxicol. Environ. Health 35(2):127–133. McGrath JJ (2000) Biological plausibility for carbon monoxide as a copollutant in PM epidemiologic studies. Inhal. Toxicol. 12(Suppl. 4):91–107. Meade TW (1992) Thrombosis and cardiovascular disease. Ann. Epidemiol. 2(4):353–364. Melinyshyn MJ , et al. (1988) Circulatory and metabolic responses to carbon monoxide hypoxia during beta-adrenergic blockade. Am. J. Physiol. 255(1 Pt 2):H77–84. Moore LG (1987) Altitude-aggravated illness: examples from pregnancy and prenatal life. Ann. Emerg. Med. 16(9):965–973. Morse D, Sethi J (2002) Carbon monoxide and human disease. Antioxid. Redox Signal 4(2):331–338. Mott JA, et al. (2002) National vehicle emissions policies and practices and declining US carbon monoxide-related mortality. JAMA 288(8):988–995. Munro JM, Cotran RS (1988) The pathogenesis of atherosclerosis: atherogenesis and inflammation. Lab. Invest. 58(3):249–261. Murray RP, et al. (2002) Relations of cotinine and carbon monoxide to self-reported smoking in a cohort of smokers and ex-smokers followed over 5 years. Nicotine Tob. Res. 4(3):287–294. NAPCA(1970) Air Quality Criteria for Carbon Monoxide. AP-62. National Air Pollution Control Administration. Newill VA , (1974) The susceptibility of the fetus and child to chemical pollutants. Nature and sources of pollutants. Pediatrics 53(5):785–788. Nomoto A, et al. (1988) Smooth muscle cell migration induced by inflammatory cell products and its inhibition by a potent calcium antagonist, nilvadipine. Atherosclerosis 72(2–3):213–219. Okada Y, et al. (1976) Effect of carbon monoxide on equilibrium between oxygen and hemoglobin. Am. J. Physiol. 230(2):471–475. Ott WR (1990) Total human exposure: basic concepts, EPA field studies, and future research needs. J. Air Waste Manage. Assoc. 40(7):966–975. Ott WR, Mage DT (1978) Interpreting urban carbon monoxide concentrations by means of a computerized blood COHb model. J. Air Pollut. Control Assoc. 28(9):911–916. Ott WR, Mage DT (1979) Evaluation of CO air quality criteria using a COHb model. Atmos. Environ. 13(11):1603. Ott W, et al. (1988) Validation of the simulation of human activity and pollutant exposure (shape) model using paired days from the Denver, CO, carbon monoxide field study. Atmos. Environ. 22 (10):2101–2113. Ott WR, Mage DT, Thomas J (1992) Comparison of microenvironmental CO concentrations in two cities for human exposure modeling. J. Expo. Anal. Environ. Epidemiol. 2(2):249–267. Ott W, Switzer P, Willits N (1994) Carbon monoxide exposures inside an automobile traveling on an urban arterial highway. Air Waste 44(8):1010–1018. Pall ML (2001) Common etiology of posttraumatic stress disorder, fibromyalgia, chronic fatigue syndrome and multiple chemical sensitivity via elevated nitric oxide/peroxynitrite. Med. Hypotheses 57(2):139–145. Penn A (1993) Determination of the atherogenic potential of inhaled carbon monoxide. Res. Rep. Health Eff. Inst. (57):1–20; discussion 21–30. Penney DG (1988) Hemodynamic response to carbon monoxide. Environ. Health Perspect. 77:121–30.

REFERENCES

525

Penney DG, Howley JW (1991) Is there a connection between carbon monoxide exposure and hypertension? Environ. Health Perspect. 95:191–198. Peterson JE, Stewart RD (1975) Predicting the carboxyhemoglobin levels resulting from carbon monoxide exposures. J. Appl. Physiol. 39(4):633–638. Piantadosi CA, et al. (1997) Apoptosis and delayed neuronal damage after carbon monoxide poisoning in the rat. Exp. Neurol. 147(1):103–114. Pierson WE, Koenig JQ, Bardana EJ Jr (1989) Potential adverse health effects of wood smoke. West. J. Med. 151(3):339–342. Putz VR, Johnson BL, Setzer JV (1979) A comparative study of the effects of carbon monoxide and methylene chloride on human performance. J. Environ. Pathol. Toxicol. 2(5):97–112. Raub JA, et al. (2000a) Carbon monoxide poisoning—a public health perspective. Toxicology 145 (1):1–14. Raub JA, et al. (2000b) Carbon monoxide poisoning—a public health perspective. Toxicology 145 (1):1–14. Rea JN, et al. (1973) Expired air carbon monoxide, smoking, and other variables. A community study. Br. J. Prev. Soc. Med. 27(2):114–120. Remick RA, Miles JE (1977) Carbon monoxide poisoning—neurologic and psychiatric sequelae. Can. Med. Assoc. J. 117(6):654. Renner MG (1988) Car sick. World Watch 1(6):36–43. Ritz B, Yu F (1999) The effect of ambient carbon monoxide on low birth weight among children born in southern California between 1989 and 1993. Environ. Health Perspect. 107(1):17–25. Ritz B, et al. (2002) Ambient air pollution and risk of birth defects in Southern California. Am. J. Epidemiol. 155(1):17–25. Roughton FJ (1970) The equilibrium of carbon monoxide with human hemoglobin in whole blood. Ann. N.Y. Acad. Sci. 174(1):177–188. Sadovnikoff N, Varon J, Sternbach GL (1992) Carbon monoxide poisoning. An occult epidemic. Postgrad. Med. 92(4): 86–88, 92–96. Sady SP, Carpenter MW (1989) Aerobic exercise during pregnancy. Special considerations. Sports Med. 7(6):357–375. Salam MT , et al. (2005) Birth outcomes and prenatal exposure to ozone, carbon monoxide, and particulate matter: results from the Children’s Health Study. Environ. Health Perspect. 113 (11):1638–1644. Samet JM, Marbury MC, Spengler JD (1987) Health effects and sources of indoor air pollution. Part I. Am. Rev. Respir. Dis. 136(6):1486–1508. Sannolo N, Farina V, Fiorillo A (1992) Abnormal endogenous carbon monoxide production in children with ineffective erythropoiesis. Ann. Clin. Biochem. 29(Part 4):397–9. Scharf SM, Thames MD, Sargent RK (1974) Transmural myocardial infarction after exposure to carbon monoxide in coronary-artery disease. Report of a case. N. Engl. J. Med. 291(2):85–86. Scheinkestel CD, et al. (1999) Hyperbaric or normobaric oxygen for acute carbon monoxide poisoning: a randomised controlled clinical trial. Med. J. Aust. 170(5):203–210. Sears DA, Udden MM, Thomas LJ (2001) Carboxyhemoglobin levels in patients with sickle-cell anemia: relationship to hemolytic and vasoocclusive severity. Am. J. Med. Sci. 322(6):345–348. Seppanen A, Hakkinen V, Tenkku M (1977) Effect of gradually increasing carboxyhaemoglobin saturation on visual perception and psychomotor performance of smoking and nonsmoking subjects. Ann. Clin. Res. 9(5):314–319. Setiani O (1994) Indoor air quality and ventilation strategies in the use of combustion space heating appliances in housing. Hiroshima J. Med. Sci. 43(4):163–167.

526

CARBON MONOXIDE

Shahbaz Hassan M, Ray J, Wilson F (2003) Carbon monoxide poisoning and sensorineural hearing loss. J. Laryngol. Otol. 117(2):134–137. Shapiro AB, et al. (1989) Carbon monoxide and myonecrosis: a prospective study. Vet. Hum. Toxicol. 31(2):136–137. Shephard RJ (1984) Athletic performance and urban air pollution. Can. Med. Assoc. J. 131 (2):105–109. Sheppard D, et al. (1986) Acute effects of routine firefighting on lung function Am. J. Ind. Med. 9 (4):333–340. Sheps DS, et al. (1987) Lack of effect of low levels of carboxyhemoglobin on cardiovascular function in patients with ischemic heart disease. Arch. Environ. Health 42(2):108–116. Sheps DS, et al. (1990) Production of arrhythmias by elevated carboxyhemoglobin in patients with coronary artery disease. Ann. Intern. Med. 113(5):343–351. Sheps DS, et al. (1991) Effects of 4 percent and 6 percent carboxyhemoglobin on arrhythmia production in patients with coronary artery disease. Res. Rep. Health Eff. Inst. (41):1–46; discussion 47–58. Sim DN, Neill WA (1974) Investigation of physiological basis for increased exercise threshold for angina pectoris after physical conditioning. J. Clin. Invest. 54(3):763–770. Singh J, Scott LH (1984) Threshold for carbon monoxide induced fetotoxicity. Teratology 30 (2):253–257. Slaughter JC, Koenig JQ, Reinhardt TE (2004) Association between lung function and exposure to smoke among firefighters at prescribed burns. J. Occup. Environ. Hyg. 1(1):45–49. Sokal JA, Kralkowska E (1985) The relationship between exposure duration, carboxyhemoglobin, blood glucose, pyruvate and lactate and the severity of intoxication in 39 cases of acute carbon monoxide poisoning in man. Arch. Toxicol. 57(3):196–199. Solanki DL, et al. (1988) Hemolysis in sickle cell disease as measured by endogenous carbon monoxide production. A preliminary report. Am. J. Clin. Pathol. 89(2):221–225. Song CP (1993) Health effects on workers exposed to low concentration carbon monoxide at high altitude. Zhonghua Yu Fang Yi Xue Za Zhi 27(2):81–84. Stern FB, et al. (1988) Heart disease mortality among bridge and tunnel officers exposed to carbon monoxide. Am. J. Epidemiol. 128(6):1276–1288. Storm JE, Fechter LD (1985) Prenatal carbon monoxide exposure differentially affects postnatal weight and monoamine concentration of rat brain regions. Toxicol. Appl. Pharmacol. 81(1):139– 146. Sue DY, et al. (1988) Metabolic acidosis during exercise in patients with chronic obstructive pulmonary disease. Use of the V-slope method for anaerobic threshold determination. Chest 94(5):931–938. Tan ES, van Veldhuisen DJ, Lie KI (1993) Carbon monoxide poisoning as a trigger for myocardial infarction. Ned. Tijdschr. Geneeskd. 137(44):2266–2268. Thom SR, Keim LW (1989) Carbon monoxide poisoning: a review epidemiology, pathophysiology, clinical findings, and treatment options including hyperbaric oxygen therapy. J. Toxicol. Clin. Toxicol. 27(3):141–156. Thom SR, et al. (2004) Delayed neuropathology after carbon monoxide poisoning is immune mediated. Proc. Natl. Acad. Sci. USA 101(37):13660–13665. Tikuisis P, et al. (1987) A critical analysis of the use of the CFK equation in predicting COHb formation. Am. Ind. Hyg. Assoc. J. 48(3):208–213. Tolcos M, et al. (2000) Chronic prenatal exposure to carbon monoxide results in a reduction in tyrosine hydroxylase immunoreactivity and an increase in choline acetyltransferase immunoreactivity in the fetal medulla: implications for sudden infant death syndrome. J. Neuropathol. Exp. Neurol. 59 (3):218–228.

REFERENCES

527

Unger C, et al. (1988) Altitude, low birth weight, and infant mortality in Colorado. JAMA 259 (23):3427–3432. U.S. EPA (1979) Air Quality Criteria for Carbon Monoxide. U.S. Environmental Protection Agency. U.S. EPA (1984) Revised Evaluation of Health Effects Associated with Carbon Monoxide Exposure: An Addendum to the 1979 EPA Air Quality Criteria Document for Carbon Monoxide. Final Report Office of Health and Environmental Assessment (EPA 600/8-83-033, U.S. Environmental Protection Agency). U.S. EPA (1991) Air Quality Criteria for Carbon Monoxide. U.S. EPA (1999) Air Quality Criteria for Carbon Monoxide. United States Environmental Protection Agency. Utell MJ, Warren J, Sawyer RF (1994) Public health risks from motor vehicle emissions. Annu. Rev. Public Health 15:157–178. Variend S, Forrest AR (1987) Carbon monoxide concentrations in infant deaths. Arch. Dis. Child. 62 (4):417–418. Viala A (1994) Indoor air pollution and health: study of various problems. Bull. Acad. Natl. Med. 178 (1):57–66; discussion 67–71. Viegi G, et al. (2004) Indoor air pollution and airway disease. Int. J. Tuberc. Lung. Dis. 8(12):1401– 1415. Villeneuve SM, et al. (1986) The role of alpha-adrenergic receptors in carbon monoxide hypoxia. Can. J. Physiol. Pharmacol. 64(11):1442–1446. Vogt TM, et al. (1977) Expired air carbon monoxide and serum thiocyanate as objective measures of cigarette exposure. Am. J. Public Health 67(6):545–549. Vonpost-Lingen ML (1964) The significance of exposure to small concentrations of carbon monoxide: results of an experimental study on healthy persons. Proc. R. Soc. Med. 57 (Suppl.):1021–1029. von Restorff W, Hebisch S (1988) Dark adaptation of the eye during carbon monoxide exposure in smokers and nonsmokers. Aviat. Space Environ. Med. 59(10):928–931. Vreman HJ, Kwong LK, Stevenson DK (1984) Carbon monoxide in blood: an improved microliter blood-sample collection system, with rapid analysis by gas chromatography. Clin. Chem. 30 (8):1382–1386. Vreman HJ, et al. (1994) Semiportable electrochemical instrument for determining carbon monoxide in breath. Clin. Chem. 40(10):1927–1933. Vreman HJ, et al. (1996) Evaluation of a fully automated end-tidal carbon monoxide instrument for breath analysis. Clin. Chem. 42(1):50–56. Wagner PD, et al. (1977) Ventilation–perfusion inequality in chronic obstructive pulmonary disease. J. Clin. Invest. 59(2):203–216. Waisma D, et al. (1998) Hyperbaric oxygen therapy in the pediatric patient: the experience of the Israel Naval Medical Institute. Pediatrics 102(5):E53. Wald N, Idle M, Smith PG (1977) Carboxyhaemoglobin levels in smokers of filter and plain cigarettes. Lancet 1(8003):110–112. Wallace LA, Ziegenfus RC (1985) Comparison of carboxyhemoglobin concentrations in adult nonsmokers with ambient carbon monoxide levels. J. Air Pollut. Control Assoc. 35(9):944–49. Wattel F, et al. (1996) Carbon monoxide poisoning. Presse Med. 25(31):1425–1429. Weaver LK (1999) Carbon monoxide poisoning. Crit. Care Clin. 15(2):297–317, viii. Webber AP (2003) Recurrent cardiac failure of environmental origin. J. R. Soc. Med. 96(9):458–459. Weber C (2005) Platelets and chemokines in atherosclerosis: partners in crime. Circ. Res. 96(6):612– 616.

528

CARBON MONOXIDE

Weber A, Jermini C, Grandjean E (1975) Effects of low carbon monoxide concentrations on flicker fusion frequency and on subjective feelings (author’s translation). Int. Arch. Occup. Environ. Health 36(2):87–103. West JB (1971) Causes of carbon dioxide retention in lung disease N. Engl. J. Med. 284 (22):1232–1236. West JB (1978) Regional differences in the lung. Chest 74(4):426–437. Wickramatillake HD (1999) Validation of the end-expired method for measuring carboxyhaemoglobin levels for the use in occupational and environmental exposure studies. Occup. Med. 49 (1):43–45. Widdop B (2002) Analysis of carbon monoxide. Ann. Clin. Biochem. 39(Pt 4):378–391. Wigfield DC, et al. (1981) Assessment of the methods available for the determination of carbon monoxide in blood. J. Anal. Toxicol. 5(3):122–125. Williams MT (1992) Cytochrome P450. Mechanisms of action and clinical implications. J. Fla. Med. Assoc. 79(6):405–408. Wolff E (1994) Carbon monoxide poisoning with severe myonecrosis and acute renal failure. Am. J. Emerg. Med. 12(3):347–349. Wu L, Wang R (2005) Carbon monoxide: endogenous production, physiological functions, and pharmacological applications. Pharmacol. Rev. 57(4):585–630. Yip R (1987) Altitude and birth weight. J. Pediatr. 111(6 Part 1):869–876. Yoon SS, Macdonald SC, Parrish RG (1998) Deaths from unintentional carbon monoxide poisoning and potential for prevention with carbon monoxide detectors. JAMA 279(9):685–687. Zwart A, et al. (1984) Human whole-blood oxygen affinity: effect of carbon monoxide. J. Appl. Physiol. 57(1):14–20.

15 CHROMIUM Mitchell D. Cohen

15.1 INTRODUCTION Chromium (Cr) is abundant in the Earth’s crust, with both the hexavalent (Cr(VI)) and more predominant trivalent (Cr(III)) forms readily found in nature. Chromite (FeCr2O4) is the most important Cr-containing ore, and is used for production of ferrochromium by direct reduction (Carson et al., 1986). Chemical treatment of chromite, followed by electrolysis, yields Cr metal. Commercially, Cr compounds are commonly used directly in leather/pelt tanning and for electroplating, and as additives in production of pigments, catalysts, corrosion inhibitors, and wood preservatives. Chromium metal is widely used in the steel industry, as a superalloy for jet engines, and for the formation of other alloys. Human exposure to Cr is primarily within the industrial setting or from contact with industrial effluents released into the general environment. Symptoms of acute toxicity include allergic contact dermatitis, skin ulcers, nasal membrane inflammation, and nasal ulceration, while chronic occupational exposure can result in nasal septum perforations, rhinitis, liver damage, pulmonary congestion, edema, and nephritis (Goyer, 1986). Increased incidences of lung and gastric cancers also occur among chronically exposed individuals, while elevations in other types of cancers are also evident (Costa, 1997). The toxicity and carcinogenicity of Cr are largely related to exposure to the metal in its hexavalent state.

15.2 ESSENTIALITY A possible essential role of Cr(III) was demonstrated in 1955 when weanling rats fed a torula yeast-based diet developed small progressive impairments in their glucose tolerance (Mertz and Schwarz, 1955). Subsequent studies with other experimental animals showed that small Cr deficiencies impaired their glucose tolerance, and that the rate of glucose removal was reduced to half its normal value (Schwarz and Mertz, 1959). In addition, severe Cr

Environmental Toxicants, Third Edition Edited by Morton Lippmann Copyright
2009 John Wiley & Sons, Inc.

529

530

CHROMIUM

deficiencies caused reductions in rodent growth, longevity, fertility, and sperm counts. Correspondingly, there was an increase in glycosuria, aortic plaques, and a rise in the fasting blood glucose and cholesterol levels. Evidence of the beneficial effects of Cr in human nutrition was obtained as a consequence of studies of patients who could no longer ingest food via the normal esophogeal–stomach– intestinal route due to disease or injury. In these patients, surgical implantation of a tube allows for parenteral delivery of fluids containing all essential nutrients. The beneficial effect of Cr was detected as a result of its omission from specially prepared total parental nutrition (TPN) regimens. Two case histories showed that the patients receiving chronic TPN developed considerable weight losses and hypoglycemia (Jeejeebhoy et al., 1977; Freund et al., 1979). These symptoms were reversed with Cr supplementation (as Baker’s yeast) of the parental fluids.

15.3 ENVIRONMENTAL EXPOSURES Occupational exposures to Cr occur during the various stages of its production. Because Cr can be used for many different purposes, there is the potential for exposure in a variety of industries. The most likely risk of occupational exposure is through inhalation of Cr-bearing aerosols. These mixtures are thought to have a wide spectrum of biological activities and are frequently contaminated by other metals (Stern et al., 1984), as well as other known carcinogens such as benzo(a)pyrene. Additionally, there are wide variations in the possible aerosol characteristics, such as the relative proportions of the major oxidation states of the Cr particles, as well as varying solubilities within these fractions (Hertel 1986). Chromium concentrations in soil can range from 0.1–250 ppm Cr and in certain areas, soil content may be as high as 400 ppm Cr (Langard and Norseth, 1979); overall, most soils have been shown to contain on average 50 ppm Cr (Hertel, 1986). However, industrial sources can contribute to significant elevations in the concentration of Cr found in soil. In cases of extensive Cr contamination, such as occurred in 42 Cr-contaminated sites in Hudson County, NJ, concentrations of Cr(VI) and Cr(III) up to 100 and 19,000 ppm, respectively, have been documented in the surrounding soils (ESE, 1989; Paustenbach et al., 1992; Sheehan et al., 1991). Other contaminated industrial sites that were deemed hazardous have included two sites in Odessa, Texas (total soil Cr levels ranging from 720–5000 ppm), and one site each in Woburn, MA (total soil Cr of 1000 ppm), Dixiana, SC (630 ppm Cr), and Vancouver, WA (550 ppm Cr) (U.S. EPA 1984, 1998a and b). Conversely, several agricultural regions throughout the world have been identified as being located upon Cr-deficient soils. This was demonstrated by the fact that both crop yield and quality were improved when Cr was added to the soil. However, it is not clear whether or not the beneficial aspects were due to an effect of the Cr upon the plants themselves, or were a result of interactions of the introduced Cr with other elements or biological agents already present in the soil. While the presence of Cr in phosphate fertilizers is an important source of Cr for crop growth, the downside to the introduction of Cr into the normally Cr-deficient soils also provides a major means for introducing Cr into the environment as a pollutant. Chromium in ambient air originates primarily from industrial sources (i.e., steel manufacturing and cement production) and the combustion of fossil fuels; the content in coal and crude oil varies from 1–100 mg Cr/L and 0.005–0.7 mg Cr/L, respectively (Pacyna, 1986). Airborne particulate matter from coal-fired power plants have been shown to contain Cr in the range of 2.3–31 ppm Cr; however, these levels are reduced to 0.19–6.6 ppm Cr by fly

ENVIRONMENTAL EXPOSURES

531

ash collection processes (Goyer, 1986). Typical atmospheric concentrations of Cr are, on an average, 0.2 to 1.0 ng Cr/m3 in remote continental regions, 1 to 10 ng Cr/m3 for rural and semirural areas, and 13 to 30 ng Cr/m3 in urban areas; other studies have put the latter range at a more expansive 10–100 ng Cr/m3 (see Nriagu and Nieboer, 1988; ATSDR, 2000; Federal Register, 2004), depending on the degree of industrialization. Overall, the distribution of Cr(III) to Cr(VI) is
2:1 in atmospheric Cr emissions; this arises as a result of the fact that most Cr(VI) that enters the air is reduced by the action of many common environmental constituents and other ambient pollutants, including aerosolized acids and dissolved sulfides (Sheehan et al., 1991; ATSDR, 2000). In general, removal of Cr from the atmosphere is the result of either precipitation events or dry deposition. In rural and urban areas, fallout rates for Cr average about 0.2–1.5 and 20– 60 mg Cr/m2/year, respectively (Nriagu et al., 1988); dry deposition rates in areas far away from the point source of emission average between 0.001–0.03 mg Cr/m2/year. The concentrations of Cr in water (U.S. EPA, 1984, 1998a and b) are variable and depend upon salinity. Average concentrations of Cr in American rivers and lakes range from 1–30 mg Cr/L; these values are considerably higher than those found in seawater (0.1–5 mg Cr/L). Drinking water has also been shown to contain higher Cr concentrations than that encountered in river water. For example, in a survey of 84 midwestern cities, the levels of Cr in tap water were found to range from 5–17 mg Cr/L. A controllable source of Cr waste in water is from chrome plating and metal finishing industries, as well as from textile and tanning plants. Industrial wastewater contains total Cr in the range of 0.005–525 mg Cr/L, with concentrations of Cr(VI) averaging from 0.004–335 mg Cr/L. As noted above, most of the Cr that can be encountered in both freshwater and seawater environs is the result of direct deposition of airborne Cr. Oddly, Cr, which is associated with soil, has been deemed not to pose a significant runoff hazard, nor does Cr present much of a threat to aquifers or groundwater supplies as it does not readily leach from soils (U.S. EPA, 1998a and b). This proposed low-risk scenario has been supported by the studies of groundwater Cr levels in the well-studied Hudson County sites; only
1% of the total Cr found in the polluted soils was found to be leachable under stringent extraction procedures (ESE, 1989). The most likely explanation for this was that the majority of the Cr in the soils occurred as water-insoluble Cr(III) (Rai et al., 1986, 1988; Sheehan et al., 1991). However, there may not be a complete absence of a threat from Cr pollution of utilizable water supplies as a result of soil contamination. It might be concluded that under conditions wherein levels of natural reductants might be low in the soil, or the rate of deposition of Cr(VI) onto the soil exceeds that of normal reduction processes, increased amounts of soluble Cr(VI) may penetrate further into the soils and, possibly, reach water-bearing strata. Conditions have been documented in which residents in the vicinity of a Cr-contaminated site were potentially exposed to Cr(VI) in their drinking water at levels up to 10 ppm. For example, in Hinkley, CA, an electric power company pumping station utilizing water coolant laced with potassium chromate routinely discharged the solutions into unlined ponds in the desert. Over time, the Cr permeated into the local aquifer, as well as into wells used for drinking water (Costa, 1997). 15.3.1

Exposure Scenarios

Significant non-occupational Cr exposure of animals and humans is provided through the intake of Cr-containing foodstuffs. The largest sources of dietary Cr are found in meat,

532

CHROMIUM

vegetables, and unrefined sugar, while fruit, fish, and vegetable oils contain fairly small quantities of the metal. Most food, with the exception of herbs and condiments, probably contain less than 100 ppb Cr; concentrations in meat range from 10–60 ppb Cr wet weight (Guthrie, 1982; Kiovistoinen, 1982). Higher Cr concentrations have been measured in some beverages, with typical values in spirits, beer, and wine being 175, 300, and 450 mg Cr/ml, respectively (Jenning and Howard, 1980). Estimated daily intake by adults is between 100–200 mg Cr/day, though large inter-individual variations of <850–2620 mg Cr/day have been identified in several studies (Guthrie, 1982). The highest bioavailability of Cr is from the glucose tolerance factor (GTF) that is predominantly found in Baker’s yeast, liver, and meat. There are several factors regulating whether certain foodstuffs, primarily vegetables, can be a major source for Cr. Apart from the fact that the vegetables/fruits must be grown on a Crcontaining site, the levels of Cr in the soil must be in the range for which plant growth is not retarded, and the Cr taken up from the soil must localize to those portions that are edible. Plants growing on soil with a low Cr content have been estimated to contain 0.02 ppm Cr wet weight (Hertel, 1986). Even in soils with higher Cr concentrations, plants usually contain low levels of Cr, although a higher Cr content is often found in the roots. This is most likely related to the fact that only chelated Cr compounds (and not soluble Cr molecules) are absorbed from the soil by plants (Kabata-Pendias and Pendias, 1984). Because most sites containing high levels of soil Cr are in urban areas, little commercial farming is expected, and the primary source for Cr-bearing vegetable/fruit consumption is via store-bought produce. However, risk of consuming Cr via ingestion of Cr-bearing soils in these environments still exists, primarily among children (Calabrese et al., 1989) or adults suffering from pica or geophagia. Among children, average soil/dirt consumption has been found to range from 10–90 mg/day; an average of 10 mg/day has been calculated for those above the age of 6 years (Paustenbach, 1987). Using these consumption rates for each age group, and the factor of amount of time available for possible soil ingestion, average daily uptakes/intakes of Cr (primarily as Cr(III)) have been calculated to be 0.07–0.2 mg Cr/kg/day for children, and
4–9 ng Cr/kg/day for adults, using the parameters of the most likely exposed individual (MLEI) or maximally exposed individual (MEI), respectively, during data analysis within each age group (Sheehan et al., 1991). Ingestion of Cr from groundwater, especially from that around Cr-contaminated sites, is not considered a major risk factor. As noted earlier, permeation of Cr from contaminated soils is very limited, and so polluting of major water/deep aquifer supplies, including wells, is not likely. While contamination of shallow aquifers can occur over time, most often this water is not considered potable due to contamination by other pollutants or even sewageassociated microbes. The only possible source for consumption of significantly polluted water occurs after ponding; this is especially significant if soluble Cr(VI) agents are present in the contaminated soils. In non-occupational settings, inhalation of suspended dust/soil particulates containing adsorbed Cr is one likely route of exposure deemed to present a significant hazard to health. The amount of exposure to any Cr-bearing soil particles in residential environs is expected to be low, with outside soil particles representing no more than 10% (on average) of the total composition of home-associated dusts (Sheehan et al., 1991). In general, following inhalation, the Cr-bearing soil/dust particles are expected to undergo redistribution, with 25% being exhaled, 50% landing in the upper airways (and subsequently swallowed), and the remainder deposited in the lungs (Cowherd et al., 1985). Even within that small fraction that can reach the lungs, mucociliary clearance leads to removal of a significant amount of the

ENVIRONMENTAL EXPOSURES

533

Cr-bearing material. Using this redistribution profile, an assumption of a level of 1 ng Cr/m3 air, and taking into account differences in breathing rates as a function of age, deposition of Cr (as Cr(III)) at the MEI level can be estimated to be
5 and 3.5 pg Cr/kg/day for an exposed child and adult, respectively. Correspondingly, levels of Cr(VI) alone in the airborne particulate matter would be much lower in this model (i.e., 3 pg/m3), and so daily deposition levels would be on the levels of 1015 g (fg) Cr(VI)/kg. Another variable needs to be considered when discussing daily deposition of Cr in the lungs; that is smoking. Tobacco grown in the United States has been shown to contain 0.24–6.30 ppm Cr (IARC, 1990); tobaccos in other nations have Cr levels that are highly variable as well (Chen, 2003; Grant et al., 2004; Gendreau and Vitaro, 2005). Although no estimates on the amounts of Cr inhaled daily by smokers has yet been reported, studies have shown that Cr is one of nine of the 44 chemical agents classified as “Group I carcinogens” by IARC reported to occur in mainstream cigarette smoke (Smith et al., 1997). The same equations used for determining the average daily dose of soil-associated Cr may not hold true with smokers due to induced changes in respiratory parameters/ functions. In addition, cigarette usage among individuals is highly variable, even within defined age groups. As such, Cr deposition in the lungs arising from cigarette smoking is difficult to estimate. The last major means for introduction of Cr into the body in nonoccupational situations is via dermal contact. Overall, Cr(III) is not dermally absorbed to any significant extent, since it binds readily to several constituents within the skin (Polak, 1983). Unless solubilized or suspended in solution, Cr in soil/dusts is very poorly absorbed through the skin. If soluble Cr (VI) is present in the contact sample, it can pass more readily through the epidermal barrier than its trivalent counterparts (Samitz and Katz, 1964). There are instances where contact surfaces may have elevated Cr levels (as observed on the cinderblocks used in many basement walls in Cr-contaminated sites in Hudson County, NJ), and the majority of the material is in the Cr(VI) form. This differs from a scenario involving skin contact with soil in that the cinderblocks are conducive to permeation by the solubilized Cr(VI) compounds while excluding Cr(III) agents (Sheehan et al., 1991). As a result of the differences in the potential risk posed between Cr(VI) and Cr(III) with respect to this route of exposure, along with the apparent selective concentration of Cr(VI) compounds on those surfaces most likely to pose a threat for human contact (i.e., basement walls), the MEI- and MLEI-associated values of average daily uptake via the dermal route, unlike those for intake via the diet or inhalation, are reported solely in terms of mg Cr(VI)/kg/day rather than Cr(III)/kg/day. Not all non-occupational dermal exposure to Cr is the result of contact with Cr-polluted waters, soils, or basement walls. Significant amounts of skin exposure can arise from daily contact with many household materials and clothing (reviewed in Paustenbach et al., 1992). In cleaning items such as bleaches and detergents, chromate has been included as both a stabilizing and a coloring agent; though this practice is not as common in the United States as it is in Europe, exposure to these Cr-containing cleaning agents has been associated with a condition known as “housewive’s eczema.” With clothing, particularly tanned leathers, sweat is the primary vehicle for both liberating Cr from the material and for providing a vehicle to concentrate Cr onto the skin during evaporation. Other less frequently reported sources for dermal exposure to Cr include military uniforms, match heads, magnetic tapes, and green felt used on gaming tables. The oddest source of introducing a significant concentration of Cr directly onto the skin may be via tatooing, though Cr is probably not as widely in use today as it was in earlier decades when use of “chromium green” was common in tattoo application.

534

15.3.2

CHROMIUM

Uptake and Distribution

As noted above, exposure to Cr can occur through one of the three major routes: via absorption through the skin, by direct ingestion, or by inhalation of Cr-containing particles. The absorption of Cr largely depends on the oxidation state of the metal and the physical characteristics of the compound itself. Hexavalent Cr compounds can penetrate the skin more readily than trivalent forms, and uptake is enhanced with increases in the pH of the Crcontaining substances (Nriagu and Nieboer, 1988). While under normal conditions, absorption of Cr through the skin is limited due to ongoing chemical reduction, these processes can be circumvented; absorption of Cr(VI) may be increased by the presence of broken skin, as occurs frequently with workers bearing Cr-induced dermal ulcerations. There have been documented cases in which extensive absorption of Cr(VI) occurred following a chromic acid burn, with the patient developing significant damage to tissues (i.e., kidney) at a distal site in the body (WHO, 1988). Soluble Cr(VI) is readily absorbed from the respiratory system, whereas Cr(III) is absorbed into a much lesser degree. However, when present as insoluble particles, Cr in either valence state can be phagocytized by epithelial cells. Under normal exposure conditions (i.e., atmospheric Cr), Cr absorption from the respiratory tract has been estimated to be 1 mg Cr/day (Hertel, 1986); however, occupationally exposed individuals may inhale several mg per day. Absorption from the lung depends on characteristics of the aerosol, including size, shape, hygroscopicity, and overall electric charge of the Cr-containing particles (Stern et al., 1984; Hertel, 1986). Other factors that may influence absorption of the particles include temperature, solubility in body fluids, and reactions with other airborne agents.In thegastrointestinal tract, only 1% of an ingested dose of Cr(III) is absorbed, whereas absorption of Cr(VI) is 3%–6% (Mertz, 1969; Offenbaucer et al., 1986). Recent studies suggest a wide variation in human absorption of Cr(VI) from drinking water, with some individuals absorbing >25% of an oral dose (Kuykendall et al., 1996). This variability may relate to the reduction of Cr(VI) to the trivalent form by components of saliva and gastric juice. The oxidation state of Cr is also the determining factor for its transportation via the bloodstream. Trivalent Cr is mainly transported via the serum, bound to the iron-binding transferrin and the b-globulin fraction of serum proteins, however, at high concentrations Cr(III) binds to serum albumin or a1- or a-globulin (Gray and Sterling 1950; Harris 1977). In contrast to Cr(III), Cr(VI) can readily cross the erythrocyte membrane and bind to the globulin portion of hemoglobin following oxidation of the heme group (Gray and Sterling, 1950; Saner, 1980; Nieboer and Jusy, 1986). Inside these cells, Cr(VI) is reduced to Cr(III) by glutathione and then becomes trapped intracellularly. Recent human studies where volunteers ingested Cr(VI) in drinking water showed that a substantial amount enters red blood cells indicated that not all Cr(VI) was reduced to Cr(III) in the GI tract (Kuykendall, 1996). Consequently, the degradation products of erythrocytes may explain, in part, the high concentration of Cr found in the spleen and the slow excretion of Cr from the body. The distribution of Cr from the bloodstream depends on its chemical state. Soluble chelated formsofCrarerapidlycleared,whereascolloidalorprotein-boundformsclearmoreslowly.The latter have a greater affinity for reticuloendothelial system components,such as the liver, spleen, and bone marrow (Hopkins, 1965; Langard, 1982). Accumulation of Cr also occurs in the kidney and testes, whereas retention is less in the heart, pancreas, lungs, and brain. The Cr retained in the liver and kidneys accounts for
45%–50% of total body Cr burdens (Saner, 1980). Excretion of Cr also depends on the oxidation state and occurs primarily via urine and, to a lesser degree, through the feces. Approximately 80% of a parental 51Cr dose is excreted in

TOXICOLOGICAL EFFECTS

535

the urine and 2%–20% in the feces. The biological halftime of 51Cr in humans has been estimated to be 50–60 days (Nieboer and Jusy, 1986). As urinary Cr is generally derived from the dialyzable fraction of serum, postglomerular reabsorption of Cr into renal tubuli results in an intrarenal circulation following exposure (Mutti et al., 1984). However, urinary excretion is generally <10 mg Cr/day in the absence of excessive exposure (Underwood, 1977). A minor route of excretion is through the skin and via sweat. Placental transfer of Cr has been indicated in animal studies; however, the transport across the placenta was time dependent (U.S. EPA, 1984, 1998a). For example, insignificant amounts of Cr were transferred if the metal was administered more than 10 days before birth, whereas larger amounts were transferred if the dose was given shortly before birth. This suggests that either inorganic Cr can cross the placenta or is converted to a form that can be readily transported (Danielsson et al., 1982). There were also considerable differences in distribution of Cr(III) and Cr(VI) in the fetal and embryonic tissues. On day 13 of gestation, the embryonic content of Cr(III) and Cr(VI) were 0.4 and 12%, respectively, and fetal concentrations increased with gestational age.

15.4 TOXICOLOGICAL EFFECTS Chromium metal is biologically inert and has not been reported to produce toxic or other harmful effects in man or animal. The toxicity of Cr compounds has been largely associated with the Cr(VI) form, whereas Cr(III) is virtually inactive in vivo. Following acute exposure of rats to Cr(VI) by various routes of administration, the main target organs affected included the liver and the kidneys (U.S. EPA, 1998a). The main toxic effects in the kidney were necrosis and desquamation of the epithelium of the convoluted tubules. Red blood cells were also found in the intertubular spaces. In rabbits, the effects of intraperitoneal administration of 2 mg Cr/kg (as K2Cr2O7 or Cr(NO3)3) for a period of three or six weeks were largely relegated to alterations in the brain (Mathur et al., 1977). After a period of three weeks, these changes included occasional neuronal degeneration of the cerebral cortex, marked chromatolysis, and nuclear changes in the neurons. Six weeks of exposure resulted in marked degeneration of the cerebral cortex, accompanied by neuronophagic neuroglial proliferation and meningeal congestion. Hepatic changes have also been reported in a separate study using rabbits treated with these same compounds at similar doses (Tandon et al., 1978). Soluble salts of chromates (CrO2 4 ) are highly toxic when administered parenterally, with an LD50 of 10–50 mg/kg as compared to LD50 values of 200–350 and 1500 mg/kg obtained with dermal and oral exposure, respectively (Carson et al., 1986). Large oral doses of chromate administered to rats primarily caused gastric corrosion (U.S. EPA, 1998a). Hexavalent CrO3 given orally was found to be quite toxic in mice and rats with LD50 values of 137–177 and 80–114 mg/kg, respectively. Symptoms of acute toxicity included diarrhea, cyanosis, tail necrosis, and gastric ulcers; death occurred within 3–35 h after dosing. Conversely, oral administration of Cr(III) compounds was relatively nontoxic. From this and other studies, oral LD50 values of 1.87, 11.26, and 3.25 g/kg were calculated for CrCl3, Cr(CH3COO)3, and Cr(NO3)3, respectively (Smyth et al., 1969). Cases of acute systemic poisoning are rare; however, they may follow deliberate or accidental ingestion. The oral LD50 of Na2Cr2O7 in humans has been reported to be 50 mg/kg (NIOSH, 1979). Other effects of Cr(VI) poisoning include gastric distress, olfactory sense impairment, nosebleeds, liver damage, and yellowing of the tongue and

536

CHROMIUM

teeth. Systemic toxicity may occur with both the oxidation states (mainly due to increased absorption of Cr through the broken skin) resulting in renal chromate toxicosis, liver failure, and eventually, death. In humans, the primary effects of Cr exposure occur during and after its inhalation. Hexavalent Cr is highly corrosive and can cause chronic ulcerations and perforation of the nasal septum, although ulcerations of other skin surfaces occur (Carson et al., 1986; U.S. EPA, 1998a). These degenerative responses occur rapidly and are independent of the dose and any hypersensitivity reactions. As industrial hygiene practices (i.e., better worksite ventilation, increased usage of personal breathing masks, etc.) have improved, the reported incidence of nasomucosal ulceration/perforation by workers has decreased (Bidstrup, 1989). The possibility of similar ulcerative events occurring in non-occupational environments is considered to be negligible; even in the heavily Cr-contaminated areas of northern New Jersey, there has yet to be any documentation of these pathologies. Chromium compounds are also responsible for a wide range of respiratory effects. Prolonged inhalation of chromate dusts causes irritation of the respiratory tract, resulting in manifestation such as congestion and hyperemia, chronic rhinitis, congestion of the larynx, polyps of the upper respiratory tract, chronic inflammation of the lung, emphysema, chronic bronchitis, and bronchopneumonia. As with most metals, the solubility of the Cr(VI) agents impacts upon toxicity following inhalation. Insoluble Cr(VI) agents tend to have a greater retention in the lungs than do the soluble forms; however, with repeated inhalation over increasing periods of time, lung burdens eventually become roughly equivalent regardless of solubility (Cohen et al., 1997, 2003, 2006). Contact dermatitis occurs as a result of exposure to both Cr(III) and Cr(VI), although as noted above, ulcerative events are exclusively related to Cr(VI). Among the various Cr (VI) compounds, chromic acid is one of the more potent skin irritants (Adams, 1990); the majority of Cr(III) agents are not sensitizing under normal exposure conditions due to their poor solubility and low permeation of the dermis. However, if the concentration of the Cr(III) agent is high enough, and the exposure period prolonged enough, sensitization can be induced. Because of the disparity between the two valence states in inducing allergic contact dermatitis, only Cr(VI) compounds are utilized for patch-testing for Cr sensitivity in exposed workers and residents of Cr-contaminated areas. Using standard patch test techniques, it was shown that only
10% of all occupationally Cr-exposed workers eventually developed allergic contact dermatitis (Peltonen and Fraki, 1983; Lee and Goh, 1988). Among the non-occupationally exposed populace, the incidence of Cr sensitization is far lower; as of 1991, it was estimated that the percentage of the American population sensitized to Cr (by contact with Cr in the environment and/or due to prolonged contact with leatherware) was
2.2% (Sheehan et al., 1991). To date, it is still unclear what is the threshold dose of Cr(VI) needed to induce sensitization in a previously non-Cr-exposed individual. 15.4.1

Immunotoxicity

Allergic contact dermatitis due to Cr is most commonly observed during occupational contact with low to moderate levels of chromates. This hypersensitivity usually occurs in the presence of other metals (i.e., nickel or cobalt); however, the coexisting hypersensitivities are not due to immunologic cross-reactivities, but rather, to concomitant host sensitization (van Everdingen and van Joost, 1982; Polak, 1983). The elicited contact sensitivity is a fourstage response that depends on T-lymphocyte activation rather than on formation of

TOXICOLOGICAL EFFECTS

537

antibodies against any Cr-containing allergen (Arfsten et al., 1997). In the first phase (i.e., refractory period), after the initial contact with the Cr, Cr(VI) ions penetrate cell membranes and are reduced; resulting Cr(III) ions then bind to cellular proteins to form Cr-protein complexes. If a level of damage is sufficient enough to cause cell death, the damaged/dead cell is engulfed and processed by resident antigen-presenting cells (APC); similarly, APC can engulf Cr-protein complexes if Cr-induced cell lysis occurs and Cr-protein complexes are released. The APC then present the Cr-modified proteins to naive T-lymphocytes and initiate the expansion and proliferation of effector and memory lymphocytes specific for individual Cr-bearing proteins/peptide complexes. Any subsequent exposure to Cr will then induce a hypersensitivity response characterized by both induction and elicitation (Haines and Nieboer, 1988). Induction occurs after the APC present Cr-protein/peptide complexes to memory T-lymphocytes. Elicitation arises from subsequent activated T-lymphocyte release of lymphokines that stimulate chemotaxis, inflammation, and edema. This cascade of cellular events also enhances further Cr-peptide/protein-specific effector T-lymphocyte proliferation. The final phase, persistence, is achieved through continuous renewal of memory T-lymphocytes specific for each APC-expressed Cr-protein/peptide complex. That allergic contact dermatitis due to Cr exposure even occurs is peculiar in that factors about Cr, including: (1) a lack of universal sensitivity in spite of widespread environmental Cr distribution; (2) a relatively weak allergenic potency for Cr itself; (3) variations in skin penetrability by different Cr compounds of equal or different valences; and (4) the long periods of exposure required for clinical manifestations to become evident, all need to be overcome for the response to manifest. While concentrations of Cr needed to induce sensitization are often only slightly greater than physiologic levels, Cr at very low or very high concentrations or after repeated exposure has been shown to induce states of immunologic unresponsiveness (Polak et al., 1973; Vreeburg et al., 1984). Penetration of Cr(VI) through the epidermis is inversely concentration dependent (Spruit and van Neer, 1966); however, once under the dermal layer(s), Cr(VI) is reduced to Cr(III) and Cr(III)– protein conjugate hapten formation occurs. Precisely, which protein is conjugated is uncertain, but serum albumin, heparin, and glycosaminoglycans have been suggested as potential allergens (Rytter and Haustein, 1982). Hosts with Cr-dependent allergic contact dermatitis also display increased levels of serum IgM and IgA antibodies, Cr-induced lymphocyte transformation and proliferation, and immediate (E) rosette formation as well as decreased suppressor index values that reflect changes in the relative numbers of CD4þ helper-T-lymphocytes (TH) and CD8þ suppressor T-lymphocytes (TS) (Al-Tawil et al., 1985; Janeckova et al., 1989). An overall reduction in TS cell activity (through a decrease in cell number or via Cr-mediated alterations in function) is thought responsible, at least in part, for the increases in levels of antibodies and immune complexes (Picardo et al., 1986). While the Cr-induced lymphocyte proliferation was found to be monocyte dependent, it is not clear if monocytes (or mature macrophages) themselves, or inflammation-associated polymorphonuclear leukocytes, were affected by Cr in ways that might contribute to the onset/development of the allergic response. Because inhalation of Cr is the primary route of Cr exposure in industrial settings, studies have examined the impact of Cr compounds upon the cells critical to maintaining lung immunocompetence, that is, lung macrophages (reviewed in Cohen, 2004, 2006). Morphologically, macrophages recovered from the lungs of animals following inhalation of either Cr(VI) or Cr(III) compounds display an increase in Cr-filled cytoplasmic inclusions, enlarged lysosomes, surface smoothing, and a decrease in membrane blebs utilized in mobility and for target contact (Johansson et al., 1986, 1987). Functionally, these cells

538

CHROMIUM

display reductions in phagocytic activity, rates of oxygen consumption following stimulation, and production of reactive oxygen intermediates used for target cell killing (Johansson et al., 1986; Galvin and Oberg, 1984; Glaser et al., 1985). The majority of the effects of Cr on macrophage structure and function have also been reproduced in vitro in alveolar macrophages from a variety of hosts. However, unlike in vivo, Cr(III) compounds are ineffective; this is most likely the result of valence-dependent differences in Cr ion entry into cells. Immunotoxic effects arising from Cr exposure are also observed in lymphocytes. Lymphocytes exposed to Cr(VI) in vivo or in vitro display an increased incidence of chromosomal aberrations (Elias et al., 1989; Gao et al., 1992) (including DNA strand breaks, gaps, and interchanges) and increased levels of DNA-protein complex formation (Coogan et al., 1991; Toniolo et al., 1993). Although the implications from these defects are not certain, it has been suggested that changes in lymphocyte proliferation in vivo or under experimental conditions might arise as a result of the genetic alterations/damage to DNA integrity. Functionally, lymphocytes recovered from Cr-exposed hosts display altered mitogenic responsiveness (Kucharz and Sierakowski, 1987; Borella et al., 1990). At low concentrations, soluble Cr(VI) was slightly stimulatory, yet became inhibitory with increasing concentration; soluble Cr(III) was universally ineffective. An in vitro study using rat splenocytes in mixed lymphocyte cultures or in combination with B-/T-lymphocyte-specific mitogens also indicated a very narrow concentration-dependent biphasic (stimulatory/ inhibitory) effect with Cr(VI) (Snyder, 1991). However, the mitogenic responsiveness of peripheral blood lymphocytes from Cr-exposed rats was enhanced overall, with even greater responsiveness when exogenous Cr was added. The basis for the discrepancies between the in vitro and in vivo studies may be: (1) that Cr added to naive splenocyte cultures reacted with cell surface proteins (i.e., surface mitogen receptors) to block the proliferative effect, while (2) extended periods of in vivo exposure to Cr may have resulted in host sensitization and, ultimately, selection of lymphocyte populations that proliferate in the presence of Cr ions or Cr-conjugated protein haptens (as occurs during allergic contact dermatitis). Other effects upon macrophages/lymphocytes induced by Cr include changes in the production/release of agents required for proper immune cell function and for induction of cellular activation critical to immunocompetency. These include decreased levels of circulating antibody in response to viral antigens (Figoni and Treagan, 1975); formation of interferons in response to viruses/antigenic stimuli (Hahon and Booth, 1984; Christensen et al., 1992); and production of interleukin-2 (Treagan, 1975; Kucharz and Sierakowski, 1987) required for B-lymphocyte proliferation/differentiation during humoral immune responses. A disturbed immune cell intercommunication likely serves as the basis for Cr-induced reductions in cell-mediated and humoral immunity in vivo, and subsequently, for the increased incidence/severity of infectious diseases and, possibly, cancers in hosts exposed to Cr compounds over extended periods of time. Another critical health outcome that arises from inhalation exposure to Cr is pulmonary inflammation (reviewed in Cohen, 2004, 2006). Rats exposed to Cr(III) oxide (Cr2O3) or sulfate (Cr2(SO4)3) had changes in bronchial and mediastinal lymphatic tissues consisting of increases in Cr-laden AM, lymphoid hyperplasia, and interstitial (with Cr2O3) or granulomatous (with Cr2(SO4)3) inflammation. Inflammatory effects were also noted after one instillate; exposure caused granuloma formation in the entire airways and an increased presence of alveoli with progressive fibrotic changes. In a study of roles of Cr solubility on inflammatory effects, exposure (two or four weeks) of rats to soluble or insoluble chromates induced differing levels of PMN and monocyte infiltration. While both agents caused significant increases in total cell numbers, the soluble form increased lavageable PMN and

TOXICOLOGICAL EFFECTS

539

monocytes after only two weeks; after the added two weeks of exposure, both cell types were still elevated but trending lower. Insoluble Cr had no effect on the levels of PMN or monocytes (Cohen et al., 1997). One further interesting immunologically based health effect in Cr toxicity is the prevalence of occupational asthma, subtypes of which (early and late) may have immunologic origins (Leikauf, 2002). Early asthma is mediated by antigen binding to IgE-bound mast cells and rapid mast cell degranulation and the release of mediators of bronchoconstriction. Late asthma depends on proliferating T-lymphocytes secreting lymphokines that promote chemotaxis, bronchoconstriction, and mucous secretion, generally hours after exposure. Both types of asthma have been reported in workers exposed to dichromates, ammonium bichromate, chromic acid, chromite ore, chromate pigments, and welding fumes. In some cases, hypersensitivity to Cr was confirmed by diagnostic patch testing but not all (suggesting immunologic and nonimmunologic origins). Though likely related to dermal hypersensitivity associated with Cr exposure, the underlying mechanisms of these pulmonary reactions remain woefully unexplored. 15.4.2

Carcinogenicity and Teratogenic Effects

The carcinogenicity of Cr in experimental animals is well documented (reviewed in IARC, 1990; Cohen et al., 1993; Cohen and Costa, 2006). These studies also support the hypothesis that some of the most potent carcinogens are the slightly insoluble Cr(VI) agents. An inhalation study with Wistar rats showed an increase in the incidences of lung cancers after long-term exposure to relatively low levels (i.e., 100 mg/m3) of Na2Cr2O7 (Glaser et al., 1986). The three major forms of lung tumors that developed at this level of exposure included two adenomas and one adenocarcinoma, although a malignant tumor of the pharynx was also observed in one rat. No tumors were observed in the control group. In the group exposed to Cr(III), only one case of a primary adenoma of the lung was observed. Positive results were also obtained from an life-time inhalation study employing mice (Nettesheim et al., 1971). The tumors obtained in this study were described as alveologenic adenomas and adenocarcinomas. From this and other studies, it was concluded that hexavalent Cr was a potent carcinogen. However, other variables such as exposure routes and choice of an animal model provided conflicting results regarding the absolute carcinogenicity of Cr compounds. Studies using different routes of administration, such as the implantation of stainless steel wire mesh pellets containing chromate salts, demonstrated the inducibility of squamous cell carcinomas and adenocarcinomas in the lungs of rats exposed to CaCrO4 (Laskin et al., 1970; Kuschner and Laskin, 1971). Similar studies using intrabronchial pellet implants showed positive carcinogenicities for CaCrO4, SrCrO4, and ZnCrO4, whereas negative responses were obtained with chromite ores, Cr2O3, CrO3, Na2CrO4, and Na2Cr2O7, as well as with BaCrO4 and PbCrO4 (Levy and Venitt, 1975a,b). The major drawback in the latter study with largely negative results was the use of only one dose (2 mg/kg) of each compound. In addition, malignant tumors of the respiratory tract of rabbits, guinea pigs, rats, or mice were not produced by the administration of these various chromate salts either by inhalation and/or intratracheal injection (Steffee and Baetjer, 1965); studies with Cr(CH3COO)3 also failed to induce tumors in rats (Schroeder et al., 1965). Despite this conflicting evidence regarding the overall carcinogenicity of Cr it can be concluded from these studies that some chromate and several Cr(VI) compounds are quite potent carcinogens. Epidemiological studies of the incidence of cancer in occupationally exposed individuals (extensively reviewed in U.S. EPA, 1998a; Federal Register, 2004; Cohen and Costa, 2006)

540

CHROMIUM

have indicated that cancer mortality rates in the workers were 5–40 times higher than expected. An excess incidence of lung cancers has been reported in workers in the chromate producing industry and in pigment manufacturing plants. Cancer of the nasal cavities as well as of the larynx has been reported with a greater frequency in chromate workers. In addition, gastric cancers have also been associated with chromate exposure, although only five cases were reported in a small exposure population. A survey of three chromate pigment plants showed an increased risk of cancer in only one of the plants (Norseth, 1981). An increased risk of gastric cancer has been observed among electroplaters as well as in those employed in the ferrochromium plants (five incidences instead of three cases expected) (Langard et al., 1980). Other studies have shown that other types of cancer are elevated in Cr(VI)-exposed workers (Costa, 1997). The teratogenicity of both Cr(III) and Cr(VI) has been demonstrated in animal studies. A study with Syrian golden hamster dams exposed to 5, 7.5, 10 or 15 mg CrO3/kg on day 8 of gestation showed increased incidences of resorption and cleft palates in surviving pups in all treatment groups except for the lowest dosage (Gale, 1978). In addition to the craniofacial defects, the primary internal abnormalities were hydrocephaly as well as a wide range of skeletal defects. In a study employing CrCl3, a dose of 19.5 mg Cr/kg administered intraperitoneally to mice on day 7, 8, or 9 of gestation resulted in increased anomalies in the litters of dams exposed on day 8 and 9 (Matsumato et al., 1976). Malformations included exencephaly and open eyelids as well as increased incidences of skeletal defects. To date, the teratogenicity of Cr has not been demonstrated in humans. 15.4.3

Genotoxicity and Mutagenicity

The genotoxicity of Cr compounds have been well documented. The Cr(VI) ion is readily taken up into eukaryotic cells by anion-carrying proteins, after which it is reduced to Cr(III) by a number of cytoplasmic reducing agents. During this reduction process, unstable intermediates of Cr(V) and Cr(IV) are formed by interacting with reduced glutathione. The final cellular form of Cr, Cr(III), gets trapped intracellularly because it possesses low cell membrane permeability. This shift from Cr(VI) to Cr(III) allows a concentration gradient to be established such that a continual influx of Cr(VI) ions raises intracellular Cr levels until lethal burdens are achieved. The reduction of Cr(VI) to Cr(III) causes the generation of oxygen radicals in cells that can produce DNA damage. Additionally, the Cr(III) that is eventually formed can be adducted to the DNA. Recent studies have shown that Cr(VI) is very potent in forming DNAprotein crosslinks (DPC) that involve the binding of Cr(III) to the phosphate backbone of DNA crosslinking a protein to the DNA. Formation of these lesions is time and dose dependent (Cohen et al., 1990). These crosslinks are prevalent in cells exposed to Cr(VI) and are highly stable. They are likely to lead to mutagenic consequences and are probably more significant in determining the mutagenicity of Cr than the oxidative DNA damage produced by oxygen radicals generated during the reduction of Cr(VI) to Cr(III). The predominant proteins in Cr-induced DPCs include nuclear proteins such as lamins A, B, and C, several cytokeratins, and actin (Miller and Costa, 1988). Among these, actin seems to be most readily crosslinked, with complexing occurring at sublethal concentrations of Cr (VI) or Cr(III). The differences in crosslinking potentials among these and numerous other nuclear proteins demonstrate the dependence on the primary structure of the protein(s), their proximity to the DNA, accessibility to the Cr ions, chromatin conformation, and the DNA strandedness in the formation of the lesion.

TOXICOLOGICAL EFFECTS

541

In addition to DPCs, free amino acids (AA), such as cysteine, histidine, and glutamic acid are also crosslinked to DNA by Cr(III). However, this is less likely to occur in cells in that levels of AA are much smaller compared to those of intact proteins in the cell or its nucleus. In contrast, a small peptide like glutathione (GSH), which exists at higher levels in the cell than any given collection of free AA, is a very good participant in linkage formation and can readily cause shifting in the degree of the adduct formation by the larger proteins. In its adducts, GSH most often gives rise to stable tridentate (i.e., Cr(III)–(GSH)2) complexes. Still, either as free forms or as components of di-, tri-, or polypeptides, AA display a wide spectrum of abilities to form stable complexes with the Cr and DNA. Those AA with the strongest associations have been found to be Cys, Glu, Gln, and His (Zhitkovich et al., 1995, 1996). In many bacterial cells, almost all Cr(VI) compounds tested in either forward mutation or reversion assays demonstrated a mutagenic potential. In several Salmonella typhimurium his-mutant strains, Cr(VI) caused basepair substitutions or frameshift mutations that resulted in the recovery of histidine production (Lofroth and Ames, 1978; Nakamura et al., 1987; DeFlora et al., 1990). However, negative results were obtained with several chromate salts (K2CrO7, Na2CrO7, CrO3, and CaCrO4) in a spot test using S. typhimurium strains TA98, TA100, TA1537, and TA1538 (Kanematsu et al., 1980). In contrast, positive results were obtained with K2CrO7 in the same tester strains using a plate test, while expected negative results were seen with trivalent KCr(SO4)2, Cr(NO3)3, and CrCl3 agents (Petrelli and DeFlora, 1977, 1978; Gava et al., 1989). The positive mutagenic effects of Cr(VI) were demonstrated in the Escherichia coli strain WP2 (try-) reversion assay (Venitt and Levy, 1974). Similar results were observed with the E. coli WP2 uvrA (lacking error-prone excision repair mechanisms) strain (Nishioka, 1975) and in standard and fluctuation assays with K2CrO4 (Venitt and Bosworth, 1983). The Bacillus subtillis Rec-assay using both rec and recþ strains yielded positive results with K2CrO7 and K2CrO4 (Kanematsu et al., 1980; Nakamura and Sayato, 1981), but not with CrCl3 (Nishioka, 1975). The zone of inhibition that developed with Cr(VI) compounds was greater in the rec strain than in recþ cells, indicating greater amounts of unrepaired DNA damage and cell death. Positive Rec-assay results with K2CrO4 were diminished by pretreatment of the host cells with the reducing agent Na2SO4; this suggests that the Cr(VI) oxidative state was necessary for DNA damage. While DNA damage was indicated by the Rec-assay, not all test agents displayed equal potencies; the overall order of mutagenic reactivity in this assay was K2CrO7 > K2CrO4 > CrO3 > Cr(CH3COO)3 > Cr(NO3)3 (Nakamura and Sayato, 1981). In mammalian cells, while most Cr(VI) salts were mutagenic, Cr(III) compounds produced negative responses. Soluble Cr salts such as K2Cr2O7 and ZnCrO4 have been shown to directly induce gene mutations in Chinese hamster V79 cells (Newbold et al., 1979). The loss of function of target genes after Cr exposure, such as those for resistance to 6thioguanine, ouabain, and 8-azaguanine, has also been documented (Rainaldi et al., 1982). Overall, Cr(III) compounds are relatively nonmutagenic (Langerwerf et al., 1985). However, one study indicated that CrCl3 induced weak mutations in the human fibroblast 6-thioguanine resistance locus, but that this effect was only noted with the insoluble (nonhydrated) form (Biedermann and Landolph, 1990). Determining the degree/mechanism of mutagenicity associated with certain types of Cr– DNA damage (such as DPCs or other adducts that form (mono, binary, ternary)) has been advanced by use of shuttle vectors. Results from these types of studies have shown that while coordination between the phosphodiester backbone of DNA and Cr(III)–GSH or Cr(III)–AA

542

CHROMIUM

complexes created significant premutagenic events (reviewed in Zhitkovich, 2005), there were substantive differences among the types of mutagenic change and the degree of mutagenicity produced. For example, formation of His–Cr(III)–DNA and Cys–Cr(III)– DNA adducts most often resulted in transitions (i.e., GC ! AT) and transversions (GC ! TA); in contrast, Cr(III)–GSH crosslinks primarily produced base substitutions (GC ! TA). Overall, the mutation frequencies induced by the ternary DNA adducts were much greater (GSH > His > Cys > Cr(III) alone) than by binary adducts, and even more so than with monoadducts. Overall, shuttle vector studies have been useful to demonstrate the preferential generation of sequence deletions and point mutations (GC ! AT/TA) in DNA by several reactive Cr species. A study on BHK cells exposed to K2Cr2O7 indicated an inducible inhibition of DNA synthesis. Cr(III)-bound DNA contains novel secondary and tertiary structures that alter DNA polymerase processivity or affect the recognition of bases, thereby resulting in misincorporation (Snow and Xu, 1989; Snow, 1994). This increase in polymerase activity and decreased replicative accuracy was attributed to a Cr(III)-dependent stimulation of polymerase–DNA binding. This effect was more pronounced than the Cr-induced inhibition of either RNA or protein synthesis (Levis et al., 1978). In addition, Cr compounds such as CrO3 and CrCl3 also affected the fidelity of DNA replication in vitro and in intact cells (Sirover and Loeb, 1976; Tsapakos and Wetterhahn, 1983) so that the trivalent Cr-induced introduction of an erroneous base into a replicated strand has little chance of being proofread, excised, and replaced. Besides impacting upon DNA replication and repair mechanisms in vitro, DNA damage in the form of DNA intrastrand breaks and crosslinks as well as DPCs and Cr-DNA adducts have been observed in rats and chick embryo tissues following in vivo exposure to Cr(VI) and not to Cr(III) (reviewed in Standeven and Wetterhahn, 1989). Thus, like Cr(III), Cr(VI) ions can also decrease DNA polymerase replication and fidelity, but the mechanisms are due to a different form of enzymatic competitive inhibition. Unlike Cr(III) ions that complex with nucleotides to form altered substrates that inhibit polymerase activity (Beyersmann and Koster, 1987), Cr(VI) ions directly bind to thiol groups in the enzyme or impart oxidative damage to the enzyme. A number of studies have shown Cr to be capable of inducing chromosomal aberrations and enhancing cell transformation. The morphological transformation of BHK21 cells exposed to Cr(VI) was monitored by the loss of anchorage-independent growth (Hansen and Stern, 1985; Lanfranchi et al., 1988). Similar results were obtained in primary hamster embryo cells treated with K2CrO7 (Hansen and Stern, 1985). In addition, Cr(VI) salts increased the transformation of golden Syrian hamster embryo cells following exposure either in vivo or in vitro. Besides being the cause of direct transformation, Cr(VI) compounds (i.e., CaCrO4, K2CrO4, and ZnCrO4) altered the cell susceptibility to virally induced transformations (Casto et al., 1979). A number of assays have shown that chromosomal aberrations can be induced by both valence states of Cr. Significant increases in the aberrations were observed in cultured BALB/c mouse and Chinese hamster V79 cells exposed to a number of Cr(III) or Cr(VI) salts (Tsuda and Kato, 1977; Newbold et al., 1979; Leonard and Deknudt, 1981; Loprieno et al., 1985). However, sister chromatid exchange was produced in cultured lymphocytes with Cr(VI) salts exclusively (Ohno et al., 1982; Stella et al., 1982). An increase in the aberration frequency in cells obtained from occupationally exposed individuals in Cr plating plants paralleled the observations from the in vitro studies (Bigaliev et al., 1977; Stella et al., 1982). While both valence states of Cr are able to interact with DNA, Cr(III) ions are responsible for decreasing the fidelity of DNA replication. In addition, both Cr(III) and Cr(VI) exhibit a

EXPOSURE GUIDELINES AND STANDARDS

543

clastogenic potency; however, Cr(VI) possesses the greater activity and is also a powerful mutagen in many prokaryotic and eukaryotic cell systems. These properties of Cr(VI) supports the claim that hexavalent compounds are likely to be active carcinogens, although it is more likely that the ultimate species responsible for the carcinogenic/mutagenic effects observed in vivo is the intracellularly derived trivalent form.

15.5 EXPOSURE GUIDELINES AND STANDARDS Since Cr deficiency results in impaired glucose and lipid metabolism, the United States Department of Agriculture (USDA) recommended that a dietary intake of 50–200 mg Cr/day would be safe and adequate in adults (Food and Nutrition Board, NRC, 1980). This range was based on the absence of any symptoms associated with Cr deficiency in a population known to consume an average of 60 mg Cr daily. However, metabolic studies estimated daily intake to be <50 mg Cr/day (Andersen and Kozlovsky 1985; Offenbaucer et al., 1986). Today, Cr (III) picolinate is widely advertised as a supplement to enhance muscle mass. The picolinate form greatly enhances the absorption of Cr(III) into the body and into the cells. The mechanism of an essential or pharmacological action of Cr(III) remains to be elucidated, although it seems to have a role in enhancing the effects of insulin on glucose transport. In diabetics that respond poorly to their insulin, Cr(III) picolinate helps control erratic blood glucose levels, but this effect could just as well be classified as pharmacological (i.e., Cr(III)) and not essential. The chemistry of Cr(III), with its ability to form tight kinetically inert bonds, makes it a more unusual essential element compared to others. If it has an essential function, it may be based upon a structural role. The Occupational Safety and Health Administration (OSHA) has established permissible exposure limits (PEL) for Cr and its compounds. This was critical in light of the fact that the most recent National Occupational Exposure Study (from the late 1980s) estimated that the number of workers exposed daily to Cr ranged from 300,000–550,000 (ATSDR 2000). For a typical 40 h workweek, the PEL for chromic acid and soluble chromates was set at 0.05– 0.10 mg Cr/m3. However, differences due to oxidation states and solubilities led to establishment of PELs of 0.5 mg Cr/m3 for Cr(II) and Cr(III) compounds, and 1.0 mg Cr/m3 for Cr metal dust and insoluble Cr(VI) salts. The threshold limit values (TLV) established by the American Conference of Government Industrial Hygienists (ACGIH) are 0.05 mg Cr/m3 for chromate ore processing and water-soluble Cr(VI) compounds, 0.5 mg Cr/m3 for Cr metal and Cr (III) agents, and 0.01 mg Cr/m3 for insoluble Cr(VI) compounds (i.e., zinc chromate). These PEL and TLV values, while originally established to protect workers from irritation of the respiratory system as well as against renal and hepatic damage, are also intended to reduce the carcinogenic risk from exposure to Cr(VI) agents to acceptable levels (ACGIH, 1989; OSHA, 1989). It is interesting to note that permissible levels in many of the individual states within the United States have been set substantively lower, that is, at fractions of a mg/m3, than those established by the Federal agencies (NIOSH, 1990; OSHA, 1997; see Cohen and Costa, 2006). The recommendations for acceptable exposure levels for citizens of Europe are even more stringent. In their 2000 Air Quality Guidelines for Europe, the World Health Organization stated that “When assuming a linear dose–response relationship between exposure to chromium(VI) compounds and lung cancer, no safe level of chromium(VI) can be recommended” and “The concentrations of chromium(VI) associated with an excess lifetime risk of 1 : 10,000, 1:100,000, and 1:1000,000 are 2.5, 0.25, and 0.025 ng/m3, respectively” (WHO, 2000).

544

CHROMIUM

Nevertheless, in light of the ongoing studies of the health effects associated with exposure to various Cr agents, as recently as October, 2004, OSHA petitioned to establish one firm PEL of 1.0 mg Cr/m3 for all Cr(VI) compounds: “Chromium (VI) (hexavalent chromium or Cr(VI)) means chromium with a valence of positive six, in any form or chemical compound in which it occurs. This term includes Cr(VI) in all states of matter, in any solution or other mixture, even if encapsulated by another or several other substances. The term also includes Cr(VI) when created by an industrial process, for example, when welding of stainless steel generates Cr(VI) fume. For regulatory purposes, OSHA is treating Cr(VI) generically, instead of addressing specific compounds individually. This is based on OSHA’s preliminary determination that the toxicological effect on the human body is similar from Cr(VI) in any of the substances covered under the scope of this standard, regardless of the form or compound in which it occurs” (Federal Register, 2004). REFERENCES ACGIH; American Conference of Government Industrial Hygienists (1989) Documentation of the Threshold Limit Values and Biological Exposure Indices, 5th edn, Cincinnati: American Conference of Government Industrial Hygienists. Adams RM (1990)Allergic contact dermatitis. In: Adams RM, editor. Occupational Skin Disease, 2nd edn, Philadelphia: W.B. Saunders. pp.26–31. Al-Tawil NG, Marcusson JA, Moller E (1985) HLA-class II restriction of the proliferative T-lymphocyte responses to nickel, cobalt, and chromium compounds. Tissue Antigens 25:163–172. Andersen RA, Kozlovsky AS (1985) Chromium intake, absorption and excretion of subjects consuming self-selected diets. Am. J. Clin. Nutr. 41:1173–1183. Arfsten DP, Aylward LL, Karch NJ (1997)Immunotoxicity of chromium. In: Zelikoff JT, Thomas P, editors. Immuno-toxicology of Environmental and Occupational Metals. London: Taylor and Francis Publishers. (In Press). ATSDR; Agency for Toxic Substances and Disease Registry (2000) Toxicological Profile for Chromium (Update). Atlanta: U.S. Department of Public Health and Human Services. Beyersmann D, Koster A (1987) On the role of trivalent chromium in chromium genotoxicity. Toxicol. Environ. Chem. 14:11–22. Bidstrup PL (1989) Perspective on safety: Personal opinions. Am. Ind. Hyg. Assoc. J. 50:505–509. Biedermann KA, Landolph JR (1990) Role of valence state and solubility of chromium compounds on induction of cytotoxicity, mutagenesis, and anchorage independence in diploid human fibroblasts. Cancer Research. 50:7835–7842. Bigaliev AB, Turebaev MN, Biganieva RK, Elemesova MSH (1977) Cytogenetic examination of workers engaged in chrome production. Genetika 13:545–547. Borella P, Manni S, Giardino A (1990) Cadmium, nickel, chromium, and lead accumulate in human lymphocytes and interfere with PHA-induced proliferation. J. Trace Elem. Electrolytes Health Dis. 4:87–95. Calabrese EJ, Barnes R, Stanek EJ, Pastides H, Gilbert CE, Veneman PV, Wang X, Lasztity A, Kostechi PT (1989) How much soil do young children ingest: An epidemiology study. Regul. Toxicol. Pharmacol. 10:123–127. Carson BL, Ellis HV, McCann JL (1986) Toxicology and Biological Monitoring of Metals in Humans: Including Feasibility and Need. Chelsea, Michigan: Lewis Publishers Inc. Casto BC, Meyer J, Di Paolo JA (1979) Enhancement of viral transformation for evaluation of the carcinogenic or mutagenic potential of inorganic metal salts. Cancer Res. 39:193–198.

REFERENCES

545

Chen SZ (2003) Direct analysis of trace elements in tobacco using electrothermal vaporization ICPAES. Spectrosc. Spectral Anal. 23:358–360. Christensen MM, Ernst E, Ellerman-Eriksen S (1992) Cytotoxic effects of hexavalent chromium in cultured mouse macrophages. Arch. Toxicol. 66:347–353. Cohen MD (2004) Pulmonary immunotoxicology of select metals: Aluminum, arsenic, cadmium, chromium, copper, manganese, nickel, vanadium, and zinc. J. Immunotoxicol. 1:39–70. Cohen MD (2006) Pulmonary Immunotoxicology.. In: Gardner D, editor. Toxicology of the Lung, 4th edn, Boca Raton: Taylor and Francis/CRC Press. pp. 351–420. Cohen MD, Chen LC, Zelikoff JT, Schlesinger RB (1997) Pulmonary retention and distribution of inhaled chromium: Effects of particle solubility and ozone co-exposure. Inhal. Toxicol. 9: 843–65. Cohen MD, Costa M (2006) Chromium compounds. In: Rom WN, editor. Environmental and Occupational Medicine, 4th edn, Philadelphia: Lippincott–Raven Publishers. Cohen MD, Miller CA, Xu LS, Snow ET, Costa M (1990) A blotting method for monitoring the formation of chemically induced DNA-protein complexes. Anal Biochem 186:1–7. Cohen MD, Kargacin B, Klein CB, Costa M (1993) Mechanisms of chromium carcinogenicity and toxicity. CRC Crit. Rev. Toxicol. 23:255–281. Cohen MD, Sisco M, Baker K, Bowser D, Chen LC, Schlesinger RB (2003) Impact of co-exposure to ozone on the carcinogenic potential of inhaled chromium. I. Effects on retention and on extra- and intracellular distribution. J. Toxicol. Environ. Health 66:39–55. Cohen MD, Prophete C, Sisco M, Chen L, Zelikoff J, Smee J, Holder A, Crans D (2006) Pulmonary immunotoxic potentials of metals are governed by select physicochemical properties: Chromium agents. J. Immunotoxicol. (In Press). Coogan TP, Motz J, Snyder CA, Squibb KS, Costa M (1991) Differential DNA-protein crosslinking in lymphocytes and liver following chronic drinking water exposure of rats to potassium chromate. Toxicol. Appl. Pharmacol. 109:60–72. Costa M (1997) Toxicity and carcinogenicity of Cr(VI) in animal models and humans. CRC Crit. Rev. Toxicol. 27:431–442. Cowherd D, Muleski GE, Englehart PJ, Gillette DA (1985) Rapid Assessment of Exposure to Particulate Emissions from Surface Contamination Sites. U.S. E.P.A., Office of Health and Environmental Assessment, EPA/600-8-85/002. Danielsson BR, Hassoun E, Dencker L (1982) Embryotoxicity of chromium: Distribution in pregnant mice and effects on embryonic cells in vitro. Arch. Toxicol. 51:233–245. DeFlora S, Bagnasco M, Serra D, Zanachhi P (1990) Genotoxicity of chromium compounds. A review. Mutat. Res. 238:99–172. Elias Z, Mur JM, Pierre F, Gilgenkrantz S, Schneider P, Baruthio F, Daniere MC, Fontana JM (1989) Chromosome aberrations in peripheral blood lymphocytes of welders and characterization of their exposure by biological sample analysis. J. Occup. Med. 31:477–483. ESE; Environmental Science and Engineering (1989) Remedial Investigation for Chromium Sites in Hudson County, New Jersey. Trenton, NJ: New Jersey Department of Environmental Protection. Federal Register: 29 CFR Parts (1910, 1915, 1917, 1918, 1926) Occupational Exposure to Hexavalent Chromium. proposed Rule. October 4,(2004);69(#191):59305–59474. Figoni RA, Treagan L (1975) Inhibitory effect of nickel and chromium upon antibody response of rats to immunization with T-1 phage. Res. Commun. Chem. Pathol. Pharmacol. 11: 335–338. Fishbein L (1981) Sources, transport, and alterations of metal compounds: An overview. I. Arsenic, beryllium, cadmium, chromium, and nickel. Environ. Health Perspect. 40:43–64.

546

CHROMIUM

Food and Nutrition Board, National Research Council (1980) Recommended Dietary Allowances, 9th edn, Washington, DC: National Academy of Science. pp. 159–160. Freund H, Atamian S, Fischer JE (1979) Chromium deficiency during total parenteral nutrition. J. Am. Med. Assoc. 241:496–498. Gale TF (1978) Embryotoxic effects of chromium trioxide in hamster. Environ. Res. 16:101–109. Galvin JB, Oberg SG (1984) Toxicity of hexavalent chromium to the alveolar macrophage in vivo and in vitro. Environ. Res. 33:7–16. Gao M, Binks SP, Chipman JK, Levy LS, Braithwaite RA, Brown SS (1992) Induction of DNA strand breaks in peripheral lymphocytes by soluble chromium compounds. Human Exp. Toxicol. 11: 77–82. Gava CR, Costa R, Zordan M, Venier P, Bianchi V, Levis AG (1989) Induction of gene mutations in Salmonella and Drosophila by soluble Cr(VI) compounds: Synergistic effects of nitriloacetic acid (NTA). Toxicol. Environ. Chem. 22:27–38. Gendreau PL, Vitaro F (2005) The unbearable lightness of “light” cigarettes: A comparison of smoke yields in six varieties of Canadian “light” cigarettes. Can. J. Pub. Health. 96:167–172. Glaser U, Hochrainer D, Kloppel H, Kuhnen H (1985) Low level chromium(VI) inhalation effects on alveolar macrophages and immune functions in Wistar rats. Arch. Toxicol. 57:250–256. Glaser U, Hochrainer D, Kloppel H, Oldiges H (1986) Carcinogenicity of sodium dichromate and chromium (VI/III) oxide aerosols inhaled by male Wistar rats. Toxicology 4:219–232. Goyer RA (1986) Toxicity of metals. In: Klassen CD, Amdur MO, Doull J, editors. Toxicology. New York: Macmillan Publishing Company. pp. 588–591. Grant CN, Lalor GC, Vutchkov MK (2004) Trace elements in Jamaican tobacco. West Ind. Med. J. 53:66–70. Gray SJ, Sterling K (1950) Tagging of red cells and plasma proteins with radioactive chromium. J. Clin. Invest. 29:1604–1613. Guthrie BE (1982) The nutritional role of chromium. In: Vangard R, editor. Biological and Environmental Aspects of Chromium. Amsterdam: Elsevier Biomedical Press. pp. 117–148. Hahon N, Booth JA (1984) Effect of chromium and manganese particles on the interferon system. J. Interferon Res. 4:17–27. Haines AT, Nieboer E (1988) Chromium hypersensitivity. In: Nriagu JO, Nieboer E, editors. Chromium in the Natural and Human Environments. New York: Wiley Interscience. pp. 497–532. Hansen K, Stern RM (1985) Welding fumes and chromium compounds in cell transformation assays. J. Appl. Toxicol. 5:306–314. Harris DC (1977) Different metal-binding properties of the two sites of human transferrin. Biochemistry 16:560–564. Hertel RF (1986) Sources of exposure and biological effects of chromium. IARC Monographs 71: 63–77. Hopkins LL (1965) Distribution in the rat of physiological amounts of ingested 51 CrðIIIÞ with time. Am. J. Physiol. 209:731–735. IARC; International Agency for Research on Cancer (1990) Chromium, Nickel, and Welding. IARC Monographs on the Evaluation of Carcinogenic Risks of Chemicals to Humans, Volume 49, Lyon: IARC, World Health Organization. pp. 49–256. Janeckova V, Znojemska S, Korcakova L, Wagnerova M, Kalensky J, Svobodova J (1989) The immune profile of contact allergy patients. J. Hyg. Epidemiol. Microbiol. Immunol. 33:121–127. Jeejeebhoy KN, Chu RC, Marliss EB, Greenberg GR, Bruce-Robertson A (1977) Chromium deficiency glucose intolerance and neuropathy reversed by chromium supplementation in a patient receiving long-term total parenteral nutrition. Am. J. Clin. Nutr. 30:531–538. Jenning ME, Howard ME (1980) Chromium, wine and ischaemic heart disease. Lancet 2:90–91.

REFERENCES

547

Johansson A, Wiernik A, Jarstrand J, Camner P (1986) Rabbit alveolar macrophages after inhalation of hexa- and trivalent chromium. Environ. Res. 39:372–385. Johansson A, Robertson B, Curstedt C, Camner P (1987) Alveolar macrophage abnormalities in rabbits exposed to low concentrations of trivalent chromium. Environ. Res. 4:279–293. Kabata-Pendias A, Pendias H, editors. (1984) Trace Elements in Soils and Plants. Boca Raton: CRC Press. Kanematsu N, Hara M, Kada T (1980) Rec-assay and mutagenicity studies on metal compounds. Mutat. Res. 77:109–116. Kiovistoinen P (1982) Mineral element composition of Finnish food. Acta Agric. Scand., Suppl. 22:171. Kucharz EJ, Sierakowski SJ (1987) Immunotoxicity of chromium compounds: Effect of sodium dichromate on the T-cell activation in vitro. Arh. Hig. Rada Toksikol. 38:239–243. Kuschner M, Laskin S (1971) Experimental models in environmental carcinogenesis. Am. J. Pathol. 64:183–196. Kuykendall JR, Kerger BD, Jarvi EJ, Corbett GE, Paustenbach DJ (1996) Measurement of DNAprotein crosslinks in human leukocytes following acute ingestion of chromium in drinking water. Carcinogenesis 17:1971–1977. Lanfranchi G, Paglialunga S, Levis AG (1988) Mammalian cell transformation induced by chromium (VI) compounds in the presence of nitrilotriacetic acid. J. Toxicol. Environ. Health 24: 251–260. Langard S (1982) Absorption, transport and excretion of chromium in man and animals In: Langard S, editor. Biological and Environmental Aspects of Chromium, Topics in Environmental Health, Vol. 5, Amsterdam: Elsevier Biomedical Press. pp. 149–169. Langard S, Norseth T (1979) Chromium. In: Friberg L, Nordberg GF, Vouk VB, editors. Handbook on the Toxicology of Metals. Amsterdam: Elsevier\North-Holland Biomedical Press. pp. 383–398. Langard S, Andersen A, Gylseth B (1980) Incidence of cancer among ferrochromium and ferrosilicon workers. Br. J. Ind. Med. 37:114–119. Langerwerf JS, Bakkeren HA, Jongen WM (1985) A comparison of the mutagenicity of soluble trivalent chromium compounds with that of potassium chromate. Ecotoxicol. Environ. Safety 9:92–100. Laskin S, Kuschner M, Drew RT (1970) Studies in pulmonary carcinogenesis. In: Hanna MG, Nettesheim P, Gilbert JR, editors. Inhalation Carcinogenesis. Oak Ridge, TN: U.S. Atomic Energy Commission. pp. 321–351. Lee HE, Goh CL (1988) Occupational dermatitis among chrome platers. Contact Dermatitis 18:89– 93. Leikauf GD (2002) Hazardous air pollutants and asthma. Environ Health Perspect. 110:505–526. Leonard A, Deknudt GH (1981) Mutagenicity test with chromium salts in mouse. Mutat. Res. 80:287. Levis AG, Buttignol M, Buchni V, Sponza G (1978) Effects of potassium dichromate on nucleic acid and protein syntheses and on precursor uptake in BHK fibroblasts. Cancer Res. 38:110–116. Levy LS, Venitt S (1975a) Carcinogenic and mutagenic activity of chromium containing materials. Br. J. Cancer 32:254–255. Levy LS, Venitt S (1975b) Carcinogenicity and mutagenicity of chromium compounds: Association between bronchial metaplasia and neoplasia. Carcinogenesis 7:831–835. Lofroth G, Ames BN (1978) Mutagenicity of inorganic compounds in Salmonella typhimurium: Arsenic, chromium and selenium. Mutat. Res. 53:65–66. Loprieno N, Boncristiani G, Venier P, Montaldi A, Majone F, Bianchi V, Paglialunga S, Levis AG (1985) Increased mutagenicity of chromium compounds by nitrilotriacetic acid. Environ. Mutagen. 8:571–577.

548

CHROMIUM

Mathur AK, Datta KK, Tandon SK, Dikshith TS (1977) Effect of nickel sulfate on male rats. Bull. Environ. Contam. Toxicol. 17:241–248. Matsumato N, Iijima S, Katsunuma H (1976) Placental transfer of chromic chloride and its teratogenic potential in embryonic mice. J. Toxicol. Sci. 12:1–13. Mertz W (1969) Chromium occurrence and function in biological systems. Physiol. Rev. 49:163–239. Mertz W, Schwarz K (1955) Impaired intravenous glucose tolerance as an early sign of dietary necrotic liver degeneration. Arch. Biochem. Biophys. 58:504–508. Miller CA, Costa M (1988) Characterization of DNA-protein complexes induced in intact cells by the carcinogen chromate. Mol. Carcinogen. 1:125–133. Mutti A, Pedroni C, Arfinim G, Franchini I, Minoia C, Micoli G, Baldi C (1984) Biological monitoring of occupational exposure to different chromium compounds at various valence states. J. Environ. Anal. Chem. 17:35–41. Nakamura K, Sayato Y (1981) Comparative studies of chromosomal aberration induced by trivalent and pentavalent arsenic. Mutat. Res. 88:73–80. Nakamura S, Oda Y, Shimada T, Oki I, Sugimoto K (1987) SOS-Inducing activity of chemical carcinogens and mutagens in Salmonella typhimurium TA1535/pSk1002: Examination with 151 chemicals. Mutat. Res. 192:239–246. Nettesheim P, Hanna MG, Doherty DG, Newell RI, Hellman A (1971) Effect of calcium chromate dust, influenza virus and 100R whole body X-radiation on lung tumor incidence in mice. J. Natl. Cancer Inst. 47:1129–1144. Newbold RF, Ames J, Connell JR (1979) The cytotoxic, mutagenic and clastogenic effects of chromium-containing compounds on mammalian cells in culture. Mutat. Res. 67:55–63. Nieboer E, Jusy AA (1986) In: Nriagu JO, Nieboer E, editors. Chromium in the Natural and Human Environment. New York: Wiley Interscience. pp. 21–79. NIOSH; National Institute for Occupational Safety and Health (1979) Registry of Toxic Effects of Chemical Substances (RTECS). Washington, DC. NIOSH; National Institute for Occupational Safety and Health (1990) National Occupational Exposure Survey, U.S. Department of Health and Human Services, PHS, CDC, NIOSH, Cincinnati, pp. 89–103. Nishioka H (1975) Mutagenic activities of metal compounds in bacteria. Mutat. Res. 31: 185–189. Norseth T (1981) The carcinogenicity of chromium. Environ. Health Perspect. 40:121–130. Nriagu JO, Nieboer E, editors. (1988) Chromium in the Natural and Human Environment. New York: Wiley Interscience. Nriagu JO, Pacyna JM, Milford JB, Davison CI (1988) Distribution and characteristic features of chromium in the atmosphere. In: Nriagu JO, Nieboer E, editors. Chromium in the Natural and Human Environment. New York: Wiley Interscience. pp. 125–172. Offenbaucer EG, Spencer H, Dowling HJ, Pi-Sunyer FX (1986) Metabolic chromium balance in men. Am. J. Clin. Nutr. 44:77–82. Ohno H, Hanaok F, Yanada M (1982) Inducibility of sister-chromatid exchanges by heavy-metal ions. Mutat. Res. 104:141–145. OSHA; Occupational Safety and Health Agency (1989) Final Rule Air Contaminants - Permissible Exposure Limits. Title 29, Code of Federal Regulations, Part 1910.1000. OSHA; Occupational Safety and Health Agency (1997) Occupational Safety and Health Standards. Code of Federal Regulations, Part 1910.100, Subpart Z - Toxic and Hazardous Substances. Washington, D.C. 1997. U.S. Government Printing Office. Pacyna TM (1986) Emission factors of atmospheric elements. In: Nriagu JO, Davidson C, editors. Toxic Metals in the Atmosphere. New York: John Wiley and Sons. pp. 1–32.

REFERENCES

549

Paustenbach DJ (1987) Assessing the potential environment and human health risks of contaminated soil. Comments Toxicol. 1:185–220. Paustenbach DJ, Sheehan PJ, Paull JM, Wisser LM, Finley BL (1992) Review of the allergic contact dermatitis hazard posed by chromium-contaminated soil: Identifying a “safe” concentration. J. Toxicol. Environ. Health 37:177–207. Peltonen L, Fraki J (1983) Prevalence of dichromate sensitivity. Contact Dermatitis 9:190–194. Petrelli FL, DeFlora S (1977) Toxicity and mutagenicity of hexavalent chromium in Salmonella typhimurium. Appl. Environ. Microbiol. 33:805–809. Petrelli FL, DeFlora S (1978) Oxidation of inactive trivalent chromium to mutagenic hexavalent form. Mutat. Res. 58:167–178. Picardo M, Santucci B, Pastore R, Valesini G, Verducchi D, Bravi D (1986) Immune complexes in patients with contact dermatitis. Dermatologica 172:52–53. Polak L (1983) Immunology of chromium. In: Burrows D, editor. Chromium Metabolism and Toxicity. Boca Raton, FL: CRC Press. pp. 51–136. Polak L, Turk JL, Frey JR (1973) Studies on contact hypersensitivity to chromium compounds. Prog. Allergy 17:145–226. Rai, D, Zachara, JM, Eary, LE, Girvin, DC, Moore, DA, Resch, CT, Sass, BM, Schmidt, RL (1986) Geochemical Behavior of Chromium Species. Interim Report. EPRI, EA-4544, Palo Alto: Electric Power Research Institute. pp. 3–1 – 3–16. Rai D, Zachara JM, Eary LE, Ainsworth CC, Amonette JE, Cowan CE, Szelmeczka RW, Resch CT, Girvin DC, Smith SC (1988) Chromium Reactions in Geologic Materials. Interim Report. E.P.R.I. EA-5741. Palo Alto: Electric Power Research Institute. Rainaldi G, Colella CM, Piras A, Mariani T (1982) Thioguanine resistance and sister chromatid exchanges in V79/AP4 Chinese hamster cells treated with potassium dichromate. Chem. Biol. Interact. 42:45–51. Rytter M, Haustein UF (1982) Hapten conjugation in the leukocyte migration inhibition test in allergic contact eczema. Br. J. Dermatol. 106:161–168. Samitz MH, Katz S (1964) A study of the chemical reactions between chromium and skin. J. Invest. Dermatol. 45:35–43. Saner G, editor. (1980) Chromium in Nutrition and Disease - Current Topics in Nutrition and Disease, Volume 2, New York: Alan R Liss. pp. 26–29. Schroeder HA, Balassa JJ, Vinton WH (1965) Chromium, cadmium, and lead in rats: Effects on life span, tumors and tissue levels. J. Nutr. 86:51–66. Schwarz K, Mertz W (1959) Chromium(III) and glucose tolerance factor. Arch. Biochem. Biophys. 85:292–295. Sheehan PJ, Meyer DM, Sauer MM, Paustenbach DJ (1991) Assessment of the human health risks posed by exposure to chromium-contaminated soils. J. Toxicol. Environ. Health 32:161–201. Sirover MA, Loeb LA (1976) Infidelity of DNA synthesis in vitro: Screening for potential metal mutagens or carcinogens. Science 194:1434–1436. Smith CJ, Livingston SD, Doolittle DJ (1997) An international literature survey of “IARC Group I carcinogens” reported in mainstream cigarette smoke. Food Chem. Toxicol. 35:1107–1130. Smyth HF, Carpenter CP, Weil CS, Pozzani VC, Striegel JA, Nycum JS (1969) Range finding toxicity data: List III. Amer. Ind. Hyg. Assoc. J. 30:470. Snow ET (1994) Effects of chromium on DNA replication in vitro. Environ. Health Perspect. 102 (S3):41–44. Snow ET, Xu LS (1989) Effects of chromium(III) on DNA replication in vitro. Biol. Trace Elem. Res. 21:61–71. Snyder CA (1991) Immune function assays as indicators of chromate exposure. Environ. Health Perspect. 92:83–86.

550

CHROMIUM

Spruit D, van Neer FC (1966) Penetration rate of Cr(III) and Cr(VI). Dermatologica 132:179–82. Standeven AW, Wetterhahn KE (1989) Chromium (VI) toxicity: Uptake reduction and DNA damage. J. Am. Cell Toxicol. 8:1275–1283. Steffee CH, Baetjer AM (1965) Histopathological effects of chromate chemicals. Report of studies in rabbits, guinea pigs, rats and mice. Arch. Environ. Health 11:66–75. Stella M, Mentaldi A, Rossi R, Rossi G, Levis AG (1982) Clastogenic effects of chromium on human lymphocytes in vitro and in vivo. Mutat. Res. 101:151–164. Stern RM, Thomsen E, Furst A (1984) Cr(VI) and other metallic mutagens in fly ash and welding fumes. Toxicol. Environ. Chem. 8:95–108. Tandon SK, Saxena DK, Gaur JS, Chandra SV (1978) Comparative toxicity of trivalent and hexavalent chromium. Alterations in blood and liver. Environ. Res. 15:90–99. Toniolo P, Zhitkovich A, Costa M (1993) Development and utilization of a new simple assay for DNAprotein crosslinks as a biomarker of exposure to welding fumes. Int. Arch. Occup. Environ. Health 65:S87–S89. Treagan L (1975) Metals and the immune response, a review. Res. Commun. Chem. Pathol. Pharmacol. 12:189–220. Tsapakos MJ, Wetterhahn KE (1983) The interaction of chromium with nucleic acids. Chem. Biol. Interact. 46:265–277. Tsuda H, Kato K (1977) Chromosomal aberrations and morphological transformation in hamster embryonic cells treated with potassium dichromate in vitro. Mutat. Res. 46:87–94. U.S. EPA (1984) Health Assessment Document for Chromium. United States Environmental Protection Agency. Research Triangle Park, NC: Environmental Criteria and Assessment Office. EPA/600/8-83-014F. NTIS PB 85-115905. U.S. EPA (1998a) Toxicological Review of Hexavalent Chromium. Washington, DC: United States Environmental Protection Agency. CAS No. 18540-29-9. U.S. EPA (1998b) Toxicological Review of Trivalent Chromium. Washington, DC: United States Environmental Protection Agency. CAS No. 16065-83-1. Underwood EJ, editor. (1977) Trace Elements in Human and Animal Nutrition. New York: Academic Press. van Everdingen JE, van Joost T (1982) Hypersensitivity for nickel, chromium and cobalt: a continuous problem. Neditor. T. Geneesk 126:1088–1092. Venitt S, Bosworth D (1983) The development of anaerobic methods for bacterial mutation assays: aerobic and anaerobic fluctuation test of human fecal extracts and reference mutagens. Carcinogenesis 4:339–345. Venitt S, Levy LS (1974) Mutagenicity of chromate in bacteria and its relevance to chromate carcinogenesis. Nature 250:493–495. Vreeburg KJ, deGroot K, von Blomberg BM, Scheper RJ (1984) Induction of immunological tolerance by oral administration of nickel and chromium. J. Dent. Res. 63:124–128. WHO; World Health Organization (1988) Chromium. Environmental Health Criteria No. 61. Geneva: World Health Organization. WHO; World Health Organization (2000) Chromium. In Air Quality Guidelines for Europe, 2nd Edn, Copenhagen: World Health Organization Regional Publications, European Series, No. 91. pp. 139–142. Zhitkovich A (2005) Importance of chromium-DNA adducts in mutagenicity and toxicity of chromium(VI). Chem. Res. Toxicol. 18:3–11. Zhitkovich A, Voitkun V, Costa M (1996) Formation of the amino acid-DNA complexes by hexavalent and trivalent chromium in vitro: importance of trivalent chromium and the phosphate group. Biochemistry 35:7275–7282. Zhitkovich A, Voitkun V, Costa M (1995) Glutathione and free amino acids form stable complexes with DNA following exposure of intact mammalian cells to chromate. Carcinogenesis 16:907–913.

16 DIESEL EXHAUST Joe L. Mauderly and Eric Garshick

16.1 HISTORICAL OVERVIEW The compression ignition (diesel) engine was patented by Rudolf Diesel in 1892. The chief difference between diesel engines and spark-ignition gasoline engines is that the fuel–air mixture in diesel engines is ignited by the heat of compression alone. Diesel engines have several advantages over their gasoline counterparts. They are generally more efficient in converting fuel energy to work because they operate at higher compression ratios and temperatures and burn fuel that is higher in specific energy content. Diesel fuel is heavier and less volatile than gasoline; indeed, Rudolph Diesel showed that compression ignition engines could burn a variety of low-grade fuels including coal dust. Diesel engines generally have greater durability than gasoline engines. It is not uncommon to find 30-year old engines in regular operation in local commercial fleets, and contemporary long-haul truck engines go multiple hundreds of thousands of miles between overhauls. Because diesel fuel has a lower vapor pressure than gasoline, it contributes less to organic air pollution from evaporative emissions and presents a lower explosive hazard. On the contrary, diesel engines have historically been heavier and noisier than equivalent spark-ignition engines, and were characterized by more vibration and slower acceleration at low speeds. Diesel engines of the past were also characterized by much more visible and malodorous tailpipe emissions than equivalent gasoline engines. This difference was accentuated when on-road gasoline-powered vehicles were required to be fitted with catalytic converters. These characteristics, coupled with historically low gasoline prices, have limited the use of diesel engines in passenger cars and other light- and medium-duty vehicles in the United States. However, diesel engines have been prevalent in applications in which fuel economy and durability offset the negative factors, such as in heavy-duty trucks, buses, off-road equipment, and railroad locomotives. Diesel-powered trucks were introduced into the western U.S. trucking fleet in the 1940s and into the rest of the country during the 1950s and 1960s. Diesel trucks constituted the majority of heavy-duty truck sales for the

Environmental Toxicants, Edited by Morton Lippmann Copyright
2009 John Wiley & Sons, Inc.

551

552

DIESEL EXHAUST

first time in 1961 (Motor Vehicles Manufacturers Association, 1962), and are now used in essentially all heavy- and medium-duty trucking, construction, and agricultural applications. The dieselization of railroad locomotives occurred primarily after World War II. The approximate midpoint of dieselization was 1952, and by 1959, 95% of locomotives in the United States were diesel powered (U.S. Department of Labor, 1972). Some railroads incorporated diesel locomotives earlier; the Baltimore and Ohio railroad first used diesels in 1935 (Kaplan, 1959). The penetration of diesel engines into above-ground mining applications paralleled their penetration into the trucking and construction industries. The introduction of diesel engines into underground mines was more recent; approximately 1000 units were being used in coal mines in 1983 (Daniel, 1984). Diesel engines are much more prevalent among light-duty applications in Europe than in the United States, largely because of the much higher fuel prices. During the past decade, European concerns for emissions focused more on global warming than on particles, and diesels emit less carbon dioxide (CO2) than equivalent gasoline engines. As a result, the development of light-duty diesel engines with performance characteristics acceptable to consumers occurred primarily in Europe. During the past decade, there has been a substantial penetration of diesel engines into the light truck market in the U.S. (3/4 ton and larger). Advances in reducing noise, vibration, and emissions and increasing acceleration of mid- to small-sized diesel engines, and the potential suitability of diesel engines for the internal combustion component of hybrid vehicles bode well for increased penetration of diesels into the U.S. light-duty fleet. Although still called “diesels,” new technology compression ignition engines hardly resemble diesel engines of the past. There have been remarkable engineering advances in compression ignition engine technology in the U.S. during the past decade, driven both by market competition for fuel efficiency, durability, and performance and by progressively more stringent emission standards for particulate matter (PM) and nitrogen oxides (NOx). Through a combination of cleaner fuels, advanced engine technology, and the introduction of exhaust after-treatment, tailpipe emissions are approaching the levels of those from gasoline engines, while fuel economy and durability are retained, and public acceptability continues to increase. There are continued concerns for the potential adverse health effects of diesel exhaust because (1) it still contains trace amounts of toxic compounds; (2) human exposures are common; (3) epidemiologists find associations between proximity to traffic and adverse health outcomes; and (4) toxicologists demonstrate effects of (typically high) exposures in animals. Diesel exhaust is a ubiquitous component of air pollution, and the particulate fraction has been especially of interest in light of the recent intense focus on ambient fine particulate matter (PM2.5) (EPA, 2004). All people living in developed countries are exposed frequently to diesel exhaust at some concentration, although average exposures in the U.S. are low. Although the potential for diesel exhaust to present a health hazard has been known since the 1950s, the reporting in the late 1970s that organic extracts from diesel soot were mutagenic to bacteria launched an international research effort. Research from that time through the 1990s focused almost exclusively on the potential contribution of diesel exhaust to human lung cancer risk. Research since 2000 has focused more heavily on a broad range of potential noncancer hazards. Interestingly, although gasoline emissions contain a similar range of potentially toxic species and present qualitatively similar hazards, there has been very little research on gasoline emissions (McDonald et al., 2007).

COMPOSITION OF DIESEL EXHAUST

553

Diesel exhaust is undoubtedly the most thoroughly studied complex environmental pollutant mixture in history. Nonetheless, it remains remarkably true today, as it was in 1982 (Williams, 1982), that “the history of diesel emission studies is still being written.” The history of diesel exhaust health concerns, research, and regulatory quandary is an excellent case study of the difficulties attendant to understanding and controlling environmental health risks. Despite decades of research, important questions remain to be answered (Mauderly, 2001). The purpose of this chapter is to provide an update on our understanding of the health effects of diesel exhaust from that contained in the second edition of this book (Mauderly, 2000). Other good sources of information include the 1995 report by the Health Effects Institute (HEI, 1995), the health assessment documents developed by the California EPA (Cal EPA, 1998) and U.S. EPA (EPA, 2002), and reviews by Lloyd and Cackette (2001), Bunn et al. (2004), and Hesterberg et al. (2005).

16.2 COMPOSITION OF DIESEL EXHAUST Diesel exhaust is a complex mixture of gases, vapors, and particles that contains a very large number of elements and compounds. Exhaust composition can vary markedly with fuel composition, engine type, operating conditions, after-treatment devices (e.g., particle filters and catalysts), and environmental conditions. Moreover, because (1) engines range from small one-cylinder to very large multicylinder types; (2) on-road and off-road fuels and engines differ; and (3) both the amount and nature of emissions have been changing progressively over the past few decades, it is implausible to conceive of diesel emissions as a single mixture. This chapter does not describe the composition of diesel exhaust in detail. McDonald et al. (2004a) present an informative breakdown of the composition of emissions from a popular 2000 model medium-duty on-road engine fueled with petroleum-based certification fuel and operated on a variable duty cycle for laboratory exposures. Fig. 16.1 below from that manuscript lists the many physical-chemical classes of emissions and illustrates their relative proportions. Because the measurements were made in animal exposure chambers, the figure also illustrates the small variations in the composition of exposure atmospheres that occur from introducing different dilutions of exhaust into exposure chambers containing the same number of animals, and thus the same background of emissions from the animals. Several other compilations of exhaust composition and factors contributing to variations in composition have been published (Johnson, 1988; International Agency for Research on Cancer IARC, 1989; HEI, 1995; Norbeck and Truex, 1998; EPA, 2002). The complete combustion of petroleum fuel produces primarily carbon dioxide, water, and nitrogen; the other diesel exhaust emissions result largely from incomplete combustion and pyrosynthesis (Scheepers and Bos, 1992). Because the air entering diesel engines is not throttled, the engines can operate at air–fuel ratios other than that required for stoichiometric combustion. Fuel is injected under pressure into the combustion chamber in variable amounts to achieve different engine speeds and power outputs. Conditions promoting incomplete combustion are exacerbated before a new steady state is reached at each power setting, contributing to increases in emissions during load changes. The fuel is aerosolized under pressure by injection nozzles, and the air–fuel mixture is self-ignited by compression. Less than ideal injection timing, fuel aerosolization and distribution, and combustion chamber shape and temperature also contribute to incomplete combustion. The products of incomplete combustion include carbon monoxide, unburned fuel, and lubricants, together with their

554

DIESEL EXHAUST

additives. Nitrogen dioxide is formed primarily by high-temperature oxidation of nitrogen in the intake air, rather than by incomplete fuel combustion, but it is still an emission of concern. Few, if any, of the elements and compounds contained in the gas, vapor, and particle phases of diesel exhaust are unique; most are also present in exhaust from engines burning gasoline (Zielinska et al., 2004; McDonald et al., 2008) or natural gas (Kado et al., 2005; Seagrave et al., 2005), as well as in emissions from other sources of combustion or pyrolysis of organic matter, such as wood smoke (McDonald et al., 2006a). The compositions of emissions from diesel engines fueled with low-sulfur fuel (McDonald et al., 2004b; Kado et al., 2005) or water emulsion petroleum-based fuels (Reed et al., 2005, 2006a) or biodiesel fuel (Finch et al., 2002) also overlap. It is primarily the total mass and relative proportions of the constituents that vary among these emissions. The few direct comparisons to date have suggested that the health hazards of diesel, gasoline, and natural gas exhausts also overlap (Seagrave et al., 2003, 2005; Kado et al., 2005). The overlaps in composition and effect, along with the lack of direct measures of exposure in past epidemiological studies and the lack of clearly unexposed “control” populations, have contributed to the difficulty of assessing the impact of environmental exposures to diesel exhaust on human health. The complexity and variability of diesel exhaust makes it difficult to associate adverse health effects with any single compound or class of compounds. In describing recent animal exposures, for example, McDonald et al. (2004a and Fig. 16.1) listed 6 gases, 14 classes of particle components, and 32 classes of gas- and vapor-phase organic compounds, without listing the numerous individual species within those physical-chemical classes. More than

FIGURE 16.1 Composition by weight fraction of diesel exhaust and control exposure atmospheres in animal chambers (McDonald et al., 2004a).

COMPOSITION OF DIESEL EXHAUST

555

20 years ago, Opresko et al. (1984) reported that over 450 organic compounds had been identified in diesel exhaust. Moreover, some of the gases and vapors in tailpipe emissions are also released from vehicles in engine crankcase emissions. Gases and vapors build up in the crankcases of all internal combustion engines from heating of air in the crankcase, volatilization of compounds from hot oil, and blow-by of combustion emissions from the cylinders. Crankcase emissions from new on-road diesel engines are now being vented to the engine intake air, as has been done for on-road gasoline engines for many years. Crankcase emissions from off-road and older on-road diesel engines have been vented directly to the exterior, adding to tailpipe emissions as sources of human exposure (Easter et al., 2006) and further complicating associations between vehicle emissions and health. Adding to the difficulty of associating exhaust composition with health hazards is the fact that few studies of biological effects have included careful characterization of the exposure. Until recently, most laboratory studies of inhaled diesel exhaust only reported the mass concentrations of PM and a few gases such as carbon monoxide (CO) and nitrogen oxides (NOx). Exposures have frequently been described only by the particle mass concentration, implying an assumption that either particles are solely responsible for observed effects or that particles serve as an adequate surrogate for the causal components of the mixture. Even the particle exposures have typically been poorly described. Some papers reporting laboratory studies of diesel particles have not reported the source; fewer have reported the engine and fuel type or operating condition; and very few have reported the particle composition. Estimates of health risks from epidemiological studies have also usually relied on estimated or measured concentrations of diesel PM (DPM). However, a growing body of information, discussed in later sections, demonstrates that the nonparticle components can account for the majority of many noncancer effects. The relative proportions of the different components of diesel exhaust are frequently misunderstood by health scientists and the public. The most common misconception is that DPM comprises a majority of the mass. This misconception results, in part, from the overwhelming focus of health-related attention to the PM phase, and in part from the larger values for mg/m3 used to describe PM mass than the smaller values for ppm and ppb used to describe the gas and vapor fractions. In fact, DPM comprises a small minority of the mass of typical emissions diesel engines (even older models). The report by McDonald et al. (2004a) from which the above figure was taken presents a good example. At a dilution containing 109 mg DPM/m3, sulfur dioxide (SO2) was 8.9 ppb, CO was 3.4 ppm, and NOx was 5.6 ppm. When expressed in mass concentrations, however, SO2 was 141 mg/m3, CO was 3600 mg/m3, and NOx was 6500 mg/m3. At 256 mg/m3, the mass of nonmethane volatile organic carbon (VOC) was twice that of DPM. In the case of this contemporary, but not most recent technology engine, DPM comprised only 1% of the total mass, even disregarding the mass of CO2 and water vapor. This does not mean that DPM is unimportant; however, an accurate concept of the total exposure to emissions may be helpful in forming concepts about doses of toxic materials and their potential health impacts. 16.2.1

Composition of the Particle Phase

Concern for the toxicity of diesel exhaust has focused primarily on the PM phase and until recently, primarily on the potential carcinogenicity of organic hydrocarbons associated with “soot” particles. The following general description of the PM fraction of diesel exhaust is presented with the caveat that, not only is the composition of DPM diverse under all conditions and highly variable among conditions, but evolving fuel specifications and new

556

DIESEL EXHAUST

after-treatment devices will also alter both the amount and composition of DPM emitted from future engines. The majority of DPM mass consists of “soot,” which consists of aggregates of spherical primary particles of 20–30 mm diameter that form in the combustion chamber due to incomplete combustion of fuel (Mathis et al., 2005), grow by agglomeration, and are emitted as clusters having volume median diameters in the 0.05–1.0 mm range (McDonald et. al., 2004a). Particle size can continue to grow by agglomeration at high concentrations, resulting in mass median aerodynamic diameters of over 0.2 mm at high levels in historic animal studies (Cheng et al., 1984). The size of diesel soot makes it readily respirable. Approximately 20% of inhaled 0.1 mm particles would be expected to deposit in the pulmonary region of humans (EPA, 2004). The health risks lie in the small, invisible or poorly visible particles; the larger soot agglomerates comprising most of the visible smoke from older engines during acceleration are too large to be readily respirable and few would be expected to penetrate to the deep lung. The elemental carbon (EC) core of diesel soot has a high specific surface area (30–50 m2/g; Frey and Corn, 1967) and serves as a nucleus for condensation of organic compounds from unburned or incompletely burned fuel, or from crankcase oil volatilized from cylinder walls. As emitted, the portion of the mass of diesel soot consisting of adsorbed organic matter can range from 5 to 90% (Johnson, 1988), with average values of 20–40%, representative of modern engines under most operating conditions. This adsorbed organic matter can be extracted from the EC core by solvents; however, heat and ultrasonic energy are required to thoroughly separate the organic and inorganic carbon. Not all DPM is soot. Historically, soot constituted most of the particulate mass, but nonsoot “nanoparticles” 5–50 nm in size comprised a large portion of the particle number count. Contemporary on-highway measurements show particle number concentrations in the range of 200–240 thousand/cc in the immediate vicinity of fresh diesel exhaust (Kittelson et al., 2004). Many of these particles consist largely of organic matter condensed on sulfate nuclei and do not have elemental carbon or ash cores, as indicated by their high-temperature volatility and chemical fingerprints (Kittelson et al., 2004; Tobias et al., 2001). Although some may exist in the tailpipe, many form as a “cloud” immediately post-emission as the heavier volatilized compounds mix with cooler air. Measuring and characterizing nonelemental carbon-based nanoparticles is technically challenging; thus, although this class of material has undoubtedly long been present in engine emissions, it has only begun to be characterized in the past decade. There is speculation, with some supporting data (Bagley, 1996), that this class of material might increase as soot emissions are reduced. In part, this effect could result from the diminished soot surface on which semivolatile material can condense. If after-treatment devices such as particle traps do not remove organic vapors, this class of emission could persist even under the stringent 2010 DPM mass standards. Fortunately, preliminary indications suggest that catalytic strategies likely to be implemented to enhance carbon burn-out of traps and to reduce NOx emissions are also likely to reduce organic vapor emissions. Although the health hazards of the soot fraction have been studied intensely (described in following sections), almost nothing is known about the health implications of the condensate nanoparticle fraction of DPM. There is much current speculation about the potential health hazards of nanoparticles in general (Oberd€ orster et al., 2005), but little research has focused on this class of engine emissions. To the extent that these particles are soluble in fluids lining the respiratory tract or spread on the surface of the fluid, tissues may not be exposed to particles per se, but rather to a chemical change in the surface environment. The important

COMPOSITION OF DIESEL EXHAUST

557

factor in this case would be the deposited mass, not particle number, and the mass would be extremely small. To the extent that exhaust nanoparticles remain intact in the respiratory tract, they might be transported via blood or the olfactory tract like solid ultrafine and nanoparticles (Elder et al., 2006). Regardless, this material has always been a component of exposures to whole exhaust, and thus the effects of the nanoparticle component could be no different or worse than the effects of whole exhaust. There has been no way to isolate nanoparticles from whole exhaust, but it is possible to generate aerosols that mimic the material in the absence of other components. The technical hurdles to generating aerosols for laboratory exposures have only recently been overcome. Very recent preliminary results suggest that inhaled nanoparticle aerosols generated from used diesel crankcase oil may cause little lung irritation, but may have systemic effects, such as reducing the response of splenic lymphocytes to proliferative stimuli (McDonald et al., 2006b). 16.2.2

Composition of the Gas and Vapor Phases

As discussed previously, the majority of the mass of diesel exhaust consists of gases and vapors. The technical distinction between the two is not as important as their composition and toxicity. Technically, species present as vapors could be liquefied by compression alone, while species present as gases require both reduced temperature and increased pressure. The majority of gases and vapors in diesel are CO2, water vapor, CO, NOx (primarily nitric oxide [NO] and nitrogen dioxide [NO2]), SO2, and organic (hydrocarbon-based) gases and vapors. Emissions of CO are typically lower than for equivalent gasoline engines, and NOx emissions range from similar to higher than those from comparable gasoline engines. Unfortunately, the engine design and operating factors that reduce emissions of NO2 tend to increase emissions of PM and vice versa. Considerable effort is being expended to optimize this engineering trade-off and achieve reductions of both pollutants. Meeting the new onroad emission standards will undoubtedly require after-treatment devices to control both species. The non-PM fraction of diesel exhaust reflects the chemical composition of the fuel, as is generally true for other combustion emissions. The diesel portion of petroleum distillate consists of a complex mixture of hydrocarbons from approximately C8–C25. Aromatic compounds such as benzene, methyl substituted benzenes, and polycyclic aromatic hydrocarbons (PAHs) are present also. The emissions of organic gases and vapors occur as emissions of unburned and partially burned fuel, and to a lesser extent lubrication oil. Schauer et al. (1999) analyzed the C1–C30 emissions from two medium-duty diesel trucks and reported that the majority of the gas-phase emissions consisted of carbonyl compounds. McDonald et al. (2004a) analyzed the composition of the gas phase of exhaust from a 5.9-L diesel engine burning U.S. certification fuel and operated on a variation of the heavy-duty certification cycle, and also reported that carbonyls constituted the largest fraction of nonmethane gaseous compounds (Fig. 16.2). The remainder was alkenes, alkanes, alkynes, aromatics, and organic acids, as shown in Fig. 16.2. The large portion of oxidized organics can be attributed to the oxidative nature of the combustion process. The vapor phase contains larger molecular weight semivolatile organic compounds (SVOCs), compounds that typically exist in both the gas and particle phases in fresh emissions. As reported by Schauer et al. (1999) and McDonald et al. (2004a), the majority of the mass of vapor-phase (and particle-phase) organics have not been completely identified due to the difficulty of chromatographic resolution. This contrasts with the gas phase, for which a nearly complete mass balance can be achieved. The known vapor-phase organics are

558

DIESEL EXHAUST

FIGURE 16.2 Composition of the gaseous organic compounds in fresh diesel engine exhaust reported by McDonald et al. (2004a).

primarily straight chain and branched aliphatics and olefins or their oxidation products (e.g., acids). Smaller amounts of PAHs also exist as vapors and comprise the most thoroughly characterized portion. The PAHs range from 2 to over 7 rings (typical analyses only measure up to 7-ring compounds), with the majority of the PAH mass in the gas and vapor phases, as described below. Understanding the partitioning of organic species between the gas/vapor phase and the PM phase of diesel exhaust is conceptually important when considering their potential health effects, and especially in fresh emissions near sources. Many species are SVOC and exist in both phases in ratios that depend on the vapor pressure, temperature, and concentration of each species (Hampton et al., 1983; Schuetzle, 1983). While much of the VOC and SVOC mass will migrate into the DPM or other ambient fine PM classes with time and chemical reactions, the composition of fresh emissions is important to exposures that occur in close proximity to engines (e.g., on or near roadways). It is often incorrectly assumed that PAHs exist only in the particle phase. Fig. 16.3 illustrates the relative mass concentrations of different classes of PAHs in fresh diesel exhaust (the same source cited in Fig. 16.1). Approximately 75% of the total PAH mass consisted of naphthalenes, which were entirely in

Naphthalenes Biphenyls Phenanthrenes Pyrenes Other PAH

FIGURE 16.3 Relative mass concentrations of different classes of PAHs in fresh diesel exhaust (from McDonald et al., 2004a).

EXPOSURES TO DIESEL EXHAUST

559

2-MP 80

% of Mass

60

2,7-DMP

F

40

P

20

0

Vapor phase

Particle phase

FIGURE 16.4 Vapor-DPM partitioning of four semivolatile organic compounds in fresh diesel exhaust at a DPM concentration of 44 mg/m3. 2-MP: 2-methylphenanthrene, 2,7-DNP: 2,7-dimethylphenanthrene, F: fluoranthene, P: pyrene (from McDonald et al., 2004a).

the vapor phase. The majority of the mass of biphenyls and phenanthrenes were also vapors. Even species as large as the pyrenes exist partially as vapors in fresh exhaust. Fig. 16.4 illustrates the vapor-particle partitioning of four semivolatile PAHs in fresh exhaust diluted to a DPM concentration of 44 mg/m3 and cooled to 25 C (the same source cited in Figs. 16.1 and 16.3). As one would expect, the portion partitioned into the DPM phase increased with molecular weight; however, more than one-third of the mass of pyrene was present in the vapor phase. Partitioning into DPM increases somewhat with increasing exhaust concentration, but 23% of pyrene was in vapor form in this study even at a DPM concentration of 1005 mg/m3.

16.3 EXPOSURES TO DIESEL EXHAUST 16.3.1

Emission Standards

Tailpipe emission standards for diesel engines have become progressively more stringent since the first “smoke standard” was promulgated in the United States in 1968 (Merrion, 2002). A near 30-year series of progressively more stringent standards has achieved remarkable reductions in PM and NOx emissions from on-road vehicles. The standards have been met by a combination of changing fuel composition, by improving fuel injection, combustion chamber design, valve timing, exhaust gas recirculation, turbocharging, and by adding after-treatment devices such as PM traps and reduction catalysts. Reductions in emissions have undoubtedly resulted in health benefits, although both the assessment of benefits and characterization of health hazards and risks have been complicated by the rapid evolution of emissions. Many health scientists are not well aware that emissions from contemporary vehicles are quite different from those of vehicles sold even 10 years ago, much less those associated with the human exposures and animal studies of the more distant past. On-road U.S. emission standards for PM were first set at 0.6 g/mile for light-duty diesel cars and trucks in 1982; no standards were set at that time for heavy-duty vehicles. In 1987, the standards became 0.2 g/mile for cars and 0.26 g/mile for light-duty trucks. A NOx

560

DIESEL EXHAUST

standard of 10.7 g/brake horsepower-hour (bhp-h) was set for heavy-duty vehicles in 1985, and a PM standard of 0.6 g/bhp-h was set in 1988. In 1991, the PM standard became 0.1 g/bhp-h for urban buses and 0.26 g/bhp-h for other heavy-duty vehicles. In 1994, the PM standard became 0.10 g/bhp-h for all heavy-duty vehicles, and sales-weighted averaging of emission rates among the engine families produced by each manufacturer was allowed. For 2010, the PM and NOx standards for all diesel on-road vehicles are set at 0.01 and 0.2 g/bhp-h, respectively. Roughly parallel progressive PM and NOx standards have been established in Europe and Japan, and most other developed countries have established some form of emission standards. Standards for off-road fuel composition and emissions have lagged well behind those for on-road vehicles. Because of higher fuel sulfur content and unregulated emissions, off-road vehicles and equipment became responsible for a progressively increasing portion of human exposure to diesel emissions as on-road emissions were suppressed. In its 1996 assessment of the health hazards of air toxics (EPA, 1996), the EPA estimated that off-road engines were the source of the majority of human exposure to diesel PM, even in Northeastern cities. Progressive off-road fuel and emission standards have now begun in the U.S. and, along with turnover of both the on-road fleet and off-road equipment, should markedly lower human exposures in the future. Reductions in fleet emissions lag considerably behind the implementation of standards, due to the longevity of diesel engines and thus the fleet turnover rate. As an extreme example, locomotive emission standards did not take effect until 2000, and EPA estimates that a typical locomotive engine may be in use for 40 years (EPA, 2002). In view of this lag, some state and local agencies are pressuring for the retrofit of older vehicles with emissions reduction technologies. 16.3.2

Exposure Levels

The environmental concentrations of airborne pollutants derived from diesel emissions are not known with accuracy. Few, if any, components or atmospheric reaction products of diesel emissions are unique; there are multiple common sources of the same types of particles, gases, and vapors. This is especially true for routinely measured PM2.5 mass, NOx, and CO. The pollutants that might provide more specific source profiles are not routinely measured. There are only few actual data for diesel soot (EC-based PM) concentrations in specific environmental locations and only rough estimates of average exposures. EC is the surrogate marker most commonly used to estimated DPM concentrations, but gasoline (Zielinska et al., 2004) and other emissions also contain EC. Even in special studies aimed at source apportionment, it remains difficult to distinguish diesel soot from small carbonaceous particles emitted from other sources. Although diesel soot constitutes a minority of the PM2.5 in most urban settings (EPA, 2004), it constitutes a majority of the PM2.5 from on-road vehicles (HEI, 1995). However, even in urban and suburban areas intersected by interstate highways and containing considerable cross-country and local truck traffic, emissions from gasoline vehicles can predominate PM2.5 (Lawson and Smith, 1998). In 2002, EPA reviewed estimates of environmental concentrations of DPM derived from measurements of EC and other markers and generated estimates using chemical mass balance, positive matrix factorization, dispersion, and other source apportionment and exposure modeling strategies (EPA, 2002). U.S. average urban and rural exposure concentrations from on-road vehicles in the mid-1990s were estimated at approximately 0.8 and 0.5 mg/m3, respectively. Contributions from off-road sources and variations in local emissions and meteorology can result in much higher local and regional concentrations. Concentrations

HEALTH EFFECTS

561

in cities commonly range from 1 to 5 mg/m3, and concentrations of over 40 mg/m3 have been measured in local “hotspots” such as near busy city bus stops. California estimated that the statewide population-weighted mean concentration would be approximately 2.3 mg/m3 by 2010 (Cal EPA, 1998). Although these estimates involve a number of uncertainties, they suggest that the majority of the U.S. population is probably still being exposed to average concentrations in the range of 1–3 mg/m3. There are few historical data for actual occupational exposures to DPM, although some occupational settings are known to present high exposures. Most estimates are based on filter samples of PM2.5, or on analysis of EC concentrations adjusted to estimated soot concentrations on the basis of an assumed carbon-soot ratio. Estimated exposure concentrations for miners, truck drivers, vehicle operators, maintenance shop workers, and other workers thought to have high exposures have ranged from approximately 4–1700 mg/m3 (EPA, 2002). Workers in enclosed spaces, and particularly in mines, had the highest exposures, not uncommonly near or above 1000 mg/m3. Estimates of exposure derived from measurements of EC do not include other forms of DPM. Because neither PM2.5 nor EC is unique to diesel exhaust, the accuracy of the estimates of soot exposures depends on the extent to which DPM predominates EC in the workers’ environments. Exposures of railroad workers and truck drivers have been of particular interest because those occupations have provided the majority of epidemiological data from which cancer risks have been estimated. Woskie et al. (1988a) used personal air samplers to measure concentrations of total respirable particles in different work environments of four northern U.S. railroads. They measured the nicotine content of the collected PM and subtracted the estimated contribution of cigarette smoke, but could not adjust for other nondiesel sources of PM2.5. The smoke-adjusted geometric mean values for respirable particles were 17 mg/m3 for office clerks, 39–73 mg/m3 for engineers and firers, 52–92 mg/m3 for brakemen and conductors, and 114–134 mg/m3 for locomotive shop workers. Zaebst et al. (1991) used personal samplers to measure exposures of truck drivers to EC. They reported geometric mean values of approximately 4 mg/m3 for both city and intercity highway drivers measured in 1989–1990. This would imply soot exposures of approximately 7 mg/m3 based on their assumption that soot was approximately 60% EC. Interestingly, the exposures of the drivers were not much higher than roadside background levels, suggesting that their exposures resulted more from their presence in a roadway environment than from their position in truck cabins adjacent to exhaust stacks. More recent measurements in the same segment of the trucking industry have been reported (Smith et al., 2006). Geometric mean values of EC measured in cabins of trucks in cities and on intercity highways were 1.09 and 1.12 mg/m3, respectively. A comparison of these and earlier data suggests that exposures to traffic-related EC have decreased.

16.4 HEALTH EFFECTS 16.4.1 16.4.1.1

Cancer Plausibility of Cancer Hazard

Mutagenicity Considerable effort has been expended to characterize the potential carcinogenic hazard of compounds contained in diesel soot extracts since Kotin et al. (1955) first reported that the extracts were carcinogenic in the mouse skin assay. Most work has focused on mutagenicity in bacteria and mammalian cells. “Bio-directed fractionation”

562

DIESEL EXHAUST

strategies (Schuetzle and Lewtas, 1986) were developed to identify the active fractions and compounds after Huisingh et al. (1978) reported that diesel soot extracts were mutagenic in the Ames salmonella typhimurium assay and that the mutagenicity was “direct acting” and did not require metabolic activation. The extracts are also direct-acting mutagens in in vitro mammalian cell systems (Brooks et al., 1984; Morimoto et al., 1986; Enya et al., 1997; Liu et al., 2005). All samples of DPM that have been tested to date are mutagenic to some degree in the salmonella reverse mutation assay. This includes DPM from biodiesel exhaust (Finch et al., 2002), as well as DPM from engines burning both earlier petroleum-based fuels (reviewed above) and more recent low-sulfur fuels (Kado et al., 2005). Of course, emissions from engines burning gasoline (Seagrave et al., 2002) and natural gas (Kado et al., 2005) are also mutagenic. The degree of mutagenicity and the classes of compounds responsible vary with engine, fuel type, and operating conditions (Clark et al., 1981, 1984; Claxton, 1983; Brooks et al., 1984; DeMarini et al., 2004). The mutagenic potency of soot extracts (mutations per unit mass) can even vary more than four-fold among different segments of a single variable-load operating cycle (Bechtold et al., 1984). The compounds responsible for the mutagenicity of diesel soot extract have been a subject of much research. Summary lists of compounds found in soot extract and summaries of research efforts can be found in Johnson (1988) and IARC (1989). Most research has focused on the PAHs, a class that includes numerous known and suspected mutagens and carcinogens and the class suspected by Kotin et al. (1955) to be responsible for the mouse skin tumors. Early attention focused largely on benzo(a)pyrene [B(a)P]. Researchers then focused on the more than 50 nitrated PAHs, such as the nitropyrenes that are present in lower concentrations but appear to be responsible for a large part of the direct-acting mutagenic activity (Manabe et al., 1985; Howard et al., 1990). In 1997, Enya et al. (1997) identified 3-nitrobenzanthrone as a potent mutagen in diesel soot extract. Although research will probably continue to identify specific molecular species, it still appears that nitrated PAHs are the most predominant bacterial mutagens. An important issue concerning the implication of bacterial mutagenicity for potential carcinogenicity is the probability that at least some of the mutagenic compounds in soot samples are formed as artifacts during collection of exhaust particles. Arey et al. (1988) first noted the formation of nitroaromatic compounds during high-volume sampling of environmental PM. More recently, Khalek (2004) demonstrated that nitroaromatic compounds were formed on filters doped with PAHs through which fresh emissions from diesel engines equipped with particle filters were passed. This finding provided evidence that potentially mutagenic compounds can be formed from nonmutagenic or poorly mutagenic substrate compounds during filter collections of exhaust particles. The contribution of this artifact to the mutagenicity of collected soot extracts remains to be tested directly. Bioavailability of Mutagenic Compounds The extent to which the soot-borne organic mutagens and carcinogens are available for interaction with DNA in cells in the lung, gastrointestinal tract, or elsewhere in the body remains uncertain. The action of strong organic solvents and heat used to extract the organic component of soot in the laboratory has little parallel in the lung. Early work suggested that little mutagenic activity was released from soot in biological fluids and that the binding of reactive compounds to proteins or other non-DNA molecules in tissues tended to reduce interactions with DNA (Brooks et al., 1981; King et al., 1981; Li, 1981). However, it was also shown that cultured alveolar macrophages take up diesel soot in vitro and release metabolites back into the medium (Bond et al., 1984). It was also shown that substantial portions of the metabolites of inhaled soot-borne and

HEALTH EFFECTS

563

nitropyrene are released from the soot, metabolized, and either bind to pulmonary tissues or are excreted (Sun et al., 1984; Bond et al., 1986). Keane et al. (1991) demonstrated that diesel soot dispersed in simulated pulmonary surfactant had mutagenic activity. Gerde et al. (2001) found that a portion of B(a)P bound to denuded diesel soot in the laboratory and deposited by aerosol into the lungs of dogs was desorbed from the particles and rapidly appeared in circulating blood. Although the portion of soot-borne mutagenic activity that is bioavailable in the lung under environmental exposure conditions remains uncertain, it appears that some level of in vivo mutagenic hazard is plausible. It was once thought that the formation of DNA adducts in lungs of rats exposed chronically to diesel exhaust demonstrated not only the bioavailability of bioreactive compounds, but also a likely mechanism by which carcinogenicity was initiated. Rats exposed repeatedly to diesel exhaust had increased levels of total DNA adducts in the respiratory tract (Wong et al., 1986; Bond et al., 1990a, 1990b, 1990c, 1990d; Wolff et al., 1990). However, subsequent results showed that although total adduct levels may be increased during chronic exposure, the increases occurred in “natural” adducts (types that were also present in controls) and did not increase progressively with exposure time (Randerath et al., 1995). Moreover, quantitatively and qualitatively similar increases in DNA adducts occurred in rats exposed to diesel exhaust or to mutagen-poor carbon black (Gallagher et al., 1994; Randerath et al., 1995) and titanium dioxide (Gallagher et al., 1994). The only suggestion of a diesel-specific adduct effect was the report by Gallagher et al. (1994) of one DNA adduct in diesel-exposed rats that was not found in rats exposed in parallel to carbon black or titanium dioxide. To date, there has been no reported followup of this finding. Coupled with the finding that the organic fraction of diesel soot apparently plays little, if any, role in the carcinogenicity of soot in rats (described later), the DNA adduct information from rat studies does not clarify the bioavailability of soot-borne organic mutagens. Responses in Chronic Inhalation Carcinogenicity Bioassays In the absence of definitive data from humans, estimates of carcinogenic risks typically start with dose-response hazard data from animals, and then derive human risk factors by applying adjustments for comparative dosimetry, sensitivity, safety factors, and so on. An increased tumor incidence in animals, especially if it occurs in more than one species, is accepted as an indication of a potential carcinogenic hazard for humans. However, extrapolating the animal response to quantitative estimates of environmental cancer risk requires confidence that (1) the mechanisms by which cancer occurred in animals are likely to also operate in humans; and (2) the exposure-dose-response relationship observed in animals at high levels of exposure can be extrapolated downward to the much lower levels of human exposure. The history of carcinogenicity bioassays studies of diesel exhaust is interesting because despite the repeated demonstration of a reproducible exposure-related lung tumor response in rats, current knowledge indicates that this is a species-specific threshold response that is not useful for estimating human lung cancer risk. Detailed reviews of the animal bioassay experience have been published in earlier editions of this text (Mauderly, 2000) and other reports (Valberg and Crouch, 1999; Hesterberg et al., 2005). Few additional studies have been published, and little has changed in this field during the past decade. The following information summarizes and updates the findings. Rats Studies of pulmonary carcinogenicity in rats exposed chronically to whole diesel exhaust are summarized in Table 16.1. With the exception of one study each using a dieselpowered generator and a mine diesel engine, all studies used test stand-mounted on-road

564

Wistar

F344

F344

F344

Heinrich et al. (1986)

Iwai et al. (1986)

Takemoto et al. (1986)

Mauderly et al. (1987a)

5

17

MþF

7

8–10

8

18

F

F

F

M

Kaplan et al. (1983)

White et al. (1983)

M

Karagianes et al. Wistar (1981)

F344

Sex

Age at Start (weeks)

Animals

220

15

24

96

30

6

Number per Groupa Variable speed and load

Oldsmobile 5.7 L

Yanmar 0.27 L

2.4-L truck engine

Volkswagen 1.6 L

Variable, U.S. FTP

Constant idle

Constant speed

Variable, U.S. FTPd

Oldsmobile 5.7 L Constant speed

3-cyl, 43-hp electrical generator

Engine

Operating Mode

Lung Tumors

24

44

30

24 (6)

87

75

32

19  5

0

15 (8)b

20  7

0 0.35 3.5 7.1

0 2–4

0 4.9

0 4.2

0.25 0.75 1.5

0 8.3

20

65

0.9 1.3 3.6 12.8

0 0

4.5 42.1

0 15.8

3.3 10.0 3.3

0

0 16.7

þ þ

þ

þ

NRc NR NR

Soot Percentage Concentration with Hours/Day  (mg/m3) Tumors p < 0.05 Days/Week Months

Exposure

Studies of Lung Cancer in Rats Exposed Chronically to Whole Diesel Exhaust

Strain

References

TABLE 16.1

565

F344

F344

F344

F344

F344

F344

Wistar

F344

Ishihara (1988)

Ishihara (1988)

Brightwell et al. (1989)

Lewis et al. (1989)

Mauderly et al. (1990a)

Mauderly et al. (1986)

Heinrich et al. (1995)

Nikula et al. (1995)

123

144

MþF 5

M þ F 6–8

7

7–9

MþF

19

MþF F

18

M

210–214

100–200

80

34

M þ F (Weanling) 180

123

MþF 5

GM LH6 6.2 L

Volkswagen 1.6 L

Oldsmobile 5.7 L

Oldsmobile 5.7 L

U.S. FTP

U.S. FTP

Variable, U.S. FTP

Variable, U.S. FTP

24

24 (6)

18  5

16  5

30

75

24

24 (6)

30

30

75

16  5

16  6

16  6

24

Variable, U.S. FTP

Constant speed

Constant speed

3304 Caterpillar Variable, 75 7.0 L with mine cycle water scrubber

Volkswagen 1.5 L

11-L, 6 cyl, heavy duty

1.8-L 4-cyl, light duty

0 2.4 6.3

0 0.8 2.5 7.0

0 3.5

0 3.5

0 1.95

0 0.7 2.2 6.6

0 0.5 1.0 1.8 3.7

0 0.1 0.4 1.1 2.3

þ þ

þ

þ

þ þ

þ

(continued)

1.4 6.2 17.9

0.5 0 2.0 9.0

0 6.5

0 2.9

3.3 4.4

1.2 0.7 9.7 38.5

0.8 0.8 0 3.3 6.5

3.3 2.4 0.8 4.1 2.4

566 8

102

48

Number per Groupa

Volkswagen 1.6 L

“Light duty”

Engine

b

Number of rats examined for lung tumors. Value in parentheses is number of months rats were observed after cessation of exposure. c NR: not reported. d FTP: Federal Test Procedure, EPA urban certification cycle (US-72).

a

Wistar M þ F 6

Stinn et al. (2005)

F

Sex

Age at Start (weeks)

Animals

F344

Strain

(Continued)

Iwai et al. (2000)

References

TABLE 16.1

U.S. FTP

Constant speed

Operating Mode

Lung Tumors

67

17  3

24 (6)

30 3 (27) 6 (24) 9 (21) 12 (18)

0 3.0 10.0

0 3.5 3.5 3.5 3.5

2.0 23.0 46.0

2.0 0 14.0 40.4 22.7

þ þ

þ

Soot Percentage Concentration with Hours/Day  (mg/m3) Tumors p < 0.05 Days/Week Months

Exposure

HEALTH EFFECTS

567

engines ranging from early 1980s to mid-1990s models. No published chronic inhalation bioassay has evaluated recent, low-emissions fuel, engine, or after-treatment technologies. For studies described in multiple publications, the reference given here is the most complete description. The experimental details are presented only briefly, both because they were not always reported in detail and because variables other than DPM concentration and cumulative exposure time have not proven to strongly influence the outcome. Nine studies involved exposures of 24 months or longer and used groups of 50 or more rats, the minimum number generally considered adequate for testing carcinogenicity. Heinrich et al. (1986) exposed 96 rats/group, 19 h/day, 5 days/week for 32 months to exhaust at 4.2 mg soot/m3, resulting in a 15.8% incidence of lung tumors in contrast to none in controls. A key finding was that a parallel group (not listed in Table 16.1) exposed to the same concentration of exhaust with the soot removed by filtration had no increase in lung tumor incidence. Mauderly et al. (1987a) exposed 220 rats/group, 7 h/day, 5 days/week for 30 months at 0.35, 3.5, and 7.1 mg soot/m3, resulting in lung tumor incidences of 1.3%, 3.6%, and 12.8%, respectively, in contrast to 0.9% among controls. The increases in tumor incidence were significant for the two higher concentrations. In another study conducted later using identical exposures, Mauderly et al. 1986, 1990b exposed 80 rats/group, 7 h/day, 5 days/week for 30 months at 3.5 mg soot/m3 and observed a 6.5% lung tumor incidence in contrast to none among controls. Ishihara (1988) conducted concurrent studies of rats exposed 16 h/day, 6 days/week for 30 months to exhaust from light-duty and heavy-duty engines. The heavy-duty exhaust was administered at 0.5, 1.0, 1.8, and 3.7 mg soot/m3, resulting in lung tumor incidences of 3.3% and 6.5% at the two highest levels, respectively. The highest tumor incidence was significantly elevated above the 0.8% incidence among controls. Brightwell et al. (1989) exposed 144 rats/group, 16 h/day, 5 days/week for 24 months to exhaust at 0.7, 2.2, and 6.6 mg soot/m3 and observed the rats for an additional 6 months. The lung tumor incidences at the two highest levels, 9.7% and 38.5%, were significantly increased above the 1.2% incidence among controls. In agreement with the Heinrich et al. (1986) study mentioned previously, parallel groups of rats exposed to the two higher concentrations of exhaust with the particles removed by filtration (not listed in Table 16.1) had no increase in lung tumor incidence. Lewis et al. (1989) exposed 180 rats/group, 7 h/day, 5 days/week for 24 months to water-scrubbed exhaust from a mine engine at 1.95 mg soot/m3 and observed a slight but insignificant increase in lung tumor incidence. Heinrich et al. (1995) exposed 100– 220 rats/group 18 h/day, 5 days/week for 24 months to exhaust at 0.8, 2.5, and 7.0 mg soot/m3 and observed the surviving rats for an additional 6 months. The lung tumor incidence was increased significantly at the highest exposure level. Nikula et al. (1995) exposed 210– 214 rats/group 16 h/day, 5 days/week for 24 months to exhaust at 2.4 and 6.3 mg soot/m3 and observed a dose-related increase in lung tumor incidence that was statistically significant at both exposure levels. Stinn et al. (2005) reported the most recent bioassay involving exposures of substantive groups of rats for 24 months. They exposed Wistar rats 6 h/day, 7 days/week to exhaust from a 1994 model engine operated on the same urban cycle used in many previous studies, at dilutions producing 3.0 or 10.0 mg DPM/m3. The lung tumor incidence was significantly increased in an exposure-related manner in rats examined at 24 months or held for an additional 6 months without exposure. Only two of the above nine studies did not yield statistically significant increases in lung tumor incidence, the light-duty engine study by Ishihara (1988) and the mine engine study by Lewis and colleagues (1989). Interestingly, these two studies also yielded the highest incidences (3.3%) of lung tumors in control rats; control incidences in the other studies

568

DIESEL EXHAUST

ranged from 0 to 2%. It is doubtful that a lower control incidence would have influenced the statistical outcome of the study by Ishihara and colleagues, but a lower control incidence might have yielded significant increases in the study by Lewis and colleagues. The highest exposure level in both of these studies was approximately 2 mg soot/m3, which has proven to be just below the approximate threshold for a tumor response. Results to date are coherent in demonstrating that the soot fraction of diesel exhaust is a pulmonary carcinogen in rats exposed in sufficient numbers at sufficiently high concentrations for sufficiently long times. The aggregate exposure-response relationship from the nine most robust studies is illustrated in Fig. 16.5, in which the net (exposed minus control) tumor incidences are compared on the basis of the exposure rate or weekly concentrationtime product (mg
h/m3). The data generally fall into three exposure-response groupings and strongly suggest a threshold. Exposure rates below approximately 100 mg
h/m3 produced no suggestion of a tumor response. Exposure rates between approximately 100 and 250 mg
h/m3 produced an intermediate zone of variable response, including some significant responses, some insignificantly elevated responses, and one group with no increase at all. All exposure rates above approximately 250 mg
h/m3 produced significant increases in tumor incidence. The apparent response threshold was explored more rigorously by Valberg and Crouch (1999) who reported a meta-analysis of these data, except for the most recent Stinn et al. (2005) study. The Valberg and Crouch (1999) analysis indicated a response threshold in the range of lifetime average exposures of 200–600 mg DPM/m3. No significant exposure-response relationship was found for data for exposures below 600 mg DPM/m3 average exposure. Mice There are six reports of carcinogenicity results from mice exposed chronically by inhalation to diesel exhaust (Table 16.2). Two studies used strains (Sencar and Strain A) that have high background incidences of lung tumors and were developed for their sensitivity to chemical carcinogens. Using exposures of only 7.5–15 months, these studies are a different

FIGURE 16.5 Net lung tumor incidence (exposed-control) versus exposure rate for nine studies in which groups of 50 or more rats were exposed for 24 months or longer (Heinrich et al., 1986, 1995; Mauderly et al., 1987a, 1990b; Ishihara, 1988; Brightwell et al., 1989; Lewis et al., 1989; Nikula et al., 1995; Stinn et al., 2005). Filled circles represent exposed groups with tumor incidences significantly above controls; open circles represent exposed groups with tumor incidences not significantly above controls.

569

80–87

237–250

MþF 6

MþF 6

M

F

M

Strong-A

Jackson-A

Sencar

Sencar

38–44

Nissan 3.2 L

Nissan 3.2 L

Nissan 3.2 L

Nissan 3.2 L

Nissan 3.2 L

Nissan 3.2 L

Engine

In uteroc 101–105 Nissan 3.2 L

In uteroc 104–111

6

368–403

Strong-A

6

56–58

M

6

Strong-A

Sex

No. per Groupa

F

Strain

Age at Start (weeks)

Animals

Variable, federal short cycle

Variable, federal short cycle

Variable, federal short cycle

Variable, federal short cycle

Variable, federal short cycle

Variable, federal short cycle

Variable, federal short cycle

Operating Mode

7.5

87

87

15

15

10.5

87

87

7.5

87 (Dark)b

7.5

7.5

87

Lung Tumors

0 6/12d

0 6/12d

0 12.0

0 12.0

0 12.0

0 6.0

0 6.0

þ

þ

þ

þ

þ

þ

(continued)

3.8 5.9

7.2 16.3

57.9 25.0

24.9 8.8

24.1 12.5

18.1 17.9

6.9 25.0

Soot Percentage Concentration with (mg/m3) Tumors p < 0.05 Months

87

Hours/Day  Days/Week

Exposure

Studies of Lung Tumors in Mice and Syrian Hamsters Exposed Chronically to Whole Diesel Exhaust

Mice Pepelko and Strong-A Peirano (1983)

References

TABLE 16.2

570

NMRI

Heinrich et al. (1986)

Heinrich et al. (1995) F

F

NMRI

NMRI

7

7

7

80

120

120

Volkswagen 1.6 L

U.S. FTP

Constant idle

Constant idle

Volkswagen 1.6 L Variable, U.S. FTP

96–105 Yanmar 0.27 L

84–93

M þ F Birth

F

Operating Mode

388–399 Oldsmobile 5.7 L Constant speed

59–188 Yanmar 0.27 L

8–10

8

M þ F Birth

F

M

Sex

Age at Start No. per (weeks) Groupa Engine

Animals

C57BL/6N

Takemoto et al. C57BL/6N (1986) IRC/Jc1

Jackson-A

Strain

Kaplan et al. (1983)

References

TABLE 16.2 (Continued)

30.0 32.1

13.5 (9.5)f 0 7.0

5.1 8.5

1.7 9.0 7.3 13.3

13.0 32.0

33.5 33.8 27.3 25.0

30.0 23.0

0 4.5

0 2–4 0 2–4

0 4.2

0 0.25 0.75 1.50

NR

NRe

þ

– – þ

Soot Percentage Concentration with (mg/m3) Tumors p < 0.05

Lung Tumors

0 4.5

24

18  5

23

28

44

28

28

19  5

44

8

20  7

Hours/Day  Days/Week Months

Exposure

571

Brightwell et al. Syrian golden M þ F 6–8 (1989)

Volkswagen 1.6 L

Oldsmobile 5.7 L

Daimler-Benz 2.4 L

3 cyl, 43 hp electrical generator

203–410 Volkswagen 1.5 L

96

30

48

102

155–186 Oldsmobile 5.7 L

Variable, U.S. FTP

Variable, U.S. FTP

b

16  5

19  5

24

0 6.6h

0.2 0

0 0

Lifetimeg 0 4.2

0 0

0 0

13.4 14.6 9.7 7.5

0 0 0 0

0 3.9

0 7.3

0 0.35 3.5 7.1

0 0.25 0.75 1.50

15

Constant speed 20  7

20

24

24

65

75

Constant speed 7–8  5 and load

Variable speed and load

U.S. FTP

Number examined for lung tumors. Light cycle altered for exposure during dark period. c Parents mated and offspring born in exposure atmospheres. d Exposed to 6 mg/m3 to 12 weeks of age, then to 12 mg/m3. e NR: not reported. f Value in parentheses is number of months mice were observed after cessation of exposure. g Maximum possible exposure was 28 months, longest exposure of hamsters not reported. h Exposures at 0.7 and 2.2 mg/m3 were also conducted, but detailed tumor results were not published.

a

Syrian golden M þ F 8–10

8

Heinrich et al. (1986)

M

Syrian golden

8

Kaplan et al. (1983)

F

12

Syrian golden

M

M þ F 16–18

Heinrich et al. (1982)

Syrian hamsters Syrian Cross et al. (1978) golden

Mauderly et al. CD-1 (1996)

572

DIESEL EXHAUST

type of carcinogenicity bioassay than the others conducted using conventional strains of mice, rats, and Syrian hamsters. The other four studies used longer-term exposures of conventional strains, and are more useful for interspecies comparisons. Heinrich et al. (1986) exposed female NMRI mice 19 h/day, 5 days/week for 28 months to exhaust at 4.2 mg soot/m3 and observed a significant increase in lung tumors. Interestingly, parallel exposures of mice to the same dilution of exhaust with the soot removed by filtration (not shown) also increased the lung tumor incidence, in contrast to the finding of no increased carcinogenicity in rats exposed to filtered exhaust in the parallel study. Takemoto et al. (1986) exposed male and female C57BL/6N and ICR/Jcl mice 4 h/day, 4 days/week for 28 months to exhaust at 2–4 mg soot/m3 (mean concentration not reported) and observed modest increases in lung tumor incidence; the significance of the increases was not reported. Heinrich et al. (1995) exposed female NMRI and C57BL/6N mice 18 h/day, 5 days/week for 23 (NMRI) or 24 (C57BL/6N) months to exhaust at 4.5 mg soot/m3. The lung tumor incidence of exposed NMRI mice was lower than that of controls, and that of exposed C57BL/6N mice was slightly, but not significantly, higher than that of controls. They also exposed female NMRI mice 18 h/day, 5 days/week to exhaust at 7.0 mg soot/m3 for 13.5 months followed by a 9.5-month observation period. The lung tumor incidences in exposed and control mice were nearly identical. Mauderly et al. (1996) exposed male and female CD-1 mice 7 h/day, 5 days/week for 24 months to exhaust at 0.35, 3.5, and 7.1 mg soot/m3, concurrent with the rat study reported previously (Mauderly et al., 1987a). The lung tumor incidence at the low level was slightly (insignificantly) higher than that of controls, and the incidences at the higher two levels were lower than that of the controls. The studies of Pepelko and Peirano (1983) yielded mixed results using Strain A and Sencar mice. Exposures of Strong-A mice 8 h/day, 7 days/week for 7.5 months at a soot concentration of 6.0 mg/m3 yielded a significantly positive response in a group of females, but no increase in a parallel group of males. In contrast, two other combined male-female groups exposed at 12 mg soot/m3 yielded significantly reduced lung tumor incidences. An exposure of male Jackson-A mice on the same weekly schedule for 10.5 months at 12 mg soot/m3 yielded a significantly reduced lung tumor incidence. Exposures of Sencar mice from birth on the same weekly schedule for 15 months at 6 mg soot/m3 for the first 12 weeks and then 12 mg soot/m3 for thereafter yielded a significantly positive response in females but not in males. These results indicate that mice have, at most, an equivocal lung tumor response to diesel exhaust. The positive results obtained in female NMRI mice by Heinrich et al. (1986) were not reproduced in their later study (Heinrich et al., 1995). All other results in conventional bioassay strains were negative. The results from genetically susceptible strains were mixed. Overall, the consistently positive response lung tumor of rats was not reproduced in mice. Syrian Hamsters There are five reported studies of Syrian golden hamsters exposed chronically to diesel exhaust (Table 16.2). Groups of 30–410 male and female hamsters have been exposed for times ranging from 15 months to lifetime to exhaust at concentrations ranging from 0.25 to 7.3 mg DPM/m3. No lung tumors have been observed in diesel exhaustexposed hamsters, including under conditions that caused increased lung tumors in parallel groups of rats. Usefulness of Rat Data for Estimating Human Lung Cancer Hazard The rat lung tumor data were used in several early estimates of unit risks for human lung cancer (reviewed in Mauderly, 1992; Pepelko and Chen, 1993; HEI, 1995; Cal EPA, 1998), in the absence of epidemiological studies for which the exposure levels and durations were known.

HEALTH EFFECTS

573

Knowledge gained after the initial series of rat bioassays were completed by the mid-1980s, however, indicated that the characteristic response of rats to extreme exposures should not be used for estimating unit lung cancer risk for humans exposed to environmental levels of diesel exhaust (McClellan, 1986; Mauderly, 1997, 2000; EPA, 2002; Hesterberg et al., 2005), and probably should not be used for estimating risks from even much higher occupational exposures. Understandably, this became a hotly debated issue until scientific consensus developed. Rats are now known to respond with lung tumors to chronic inhalation of high concentrations of diverse solid respirable particles, even those containing no chemical mutagens or carcinogens. Benign and malignant tumors of the epithelium of the lower respiratory tract appear to be a characteristic, species-specific accompaniment to chronic, progressive inflammatory and fibrotic lung disease that occurs when the rate of deposition of particles exceeds the capacity of clearance pathways, resulting in the build-up of substantial lung burdens of particles. The presence of chemical mutagens or carcinogens is not necessary to elicit the response. Indeed, two separate studies demonstrated that the exposure-tumor response relationship was identical for rats exposed to either whole diesel exhaust or to mutagen-free carbon black at the same particle mass concentration (Heinrich et al., 1995; Nikula et al., 1995). The tissue responses of rats accompanying an acceleration of the build-up of particles in the lung after clearance mechanisms were overwhelmed came to be known generally as “lung overload” (Mauderly and McCunney, 1996). This is a somewhat unfortunate term because clearance can also be overloaded in species that do not respond with tumors like rats. As described previously for DPM and also found for other particles, this nonspecific tumor response is not characteristic of the responses of mice and Syrian hamsters similarly exposed and having similar accumulations of particles. It was also found that the pattern of particle sequestration and cellular responses differs between rats and nonhuman primates exposed similarly (Nikula et al., 1997). Moreover, lung cancer is not a predominant outcome among humans accumulating large amounts of nonfibrous particles in the lung (e.g., coal miners) except for tobacco smoke and dusts having high silica content. After much debate, both the California EPA (1998) and the U.S. EPA (2002) decided against using the rat data as a starting point for estimating human lung cancer risks from diesel exhaust. Other scientific review groups have also determined that the rat lung tumor response derived by extreme exposures under overloading conditions is not reliable for estimating human lung cancer risks. An expert group convened to review exposure criteria to be used in the National Toxicology Program (NTP) inhalation bioassays came to the conclusion that “overloading” exposure rates should be avoided (Lewis et al., 1989). An expert workshop convened by the International Life Sciences Institute to review the issue (ILSI, 2000) recommended against using the rat lung tumor response to poorly soluble particles of low inherent toxicity for estimating human risks. Although some agencies view the rat tumor response as supporting the plausibility of a cancer hazard, it has become a consensus view among inhalation toxicologists and inhalation hazard assessment experts that the rat study results cannot be extrapolated to human unit cancer risks. Summary of Laboratory Evidence for Plausibility of Cancer Hazard The aggregate results of laboratory studies provide a modest level of support for the plausibility of a human cancer hazard from repeated inhalation exposure to exhaust from diesel engines. Extracts of DPM are mutagenic in bacteria and mammalian cells and carcinogenic in animal test systems. Some fraction of the DPM-borne organic matter is released in the lung. Repeated exposures of rats increase lung DNA adduct levels (Randerath et al., 1995). Chronic

574

DIESEL EXHAUST

inhalation exposures have increased lung tumors in some, but not all, assays using mice that are genetically susceptible to chemical carcinogens. Other laboratory results diminish the plausibility of cancer hazard. The fraction of DPMborne organic compounds released in the lung cannot be equal to the fraction extracted by strong organic solvents. The lung adenoma and micronucleus assays in strain A/J mice were negative for subchronic exposures to diesel exhaust (Reed et al., 2004). Among conventional bioassay strains of rodents exposed chronically by inhalation, only rats responded with increased lung tumors. This species-specific response was not related to DPM-borne organics, had a threshold well above plausible equivalent environmental exposures, required exposures that overwhelmed particle clearance mechanisms, was also caused by other particle types, and was accompanied by patterns of particle sequestration and tissue changes that were not paralleled in nonhuman primates. The lung DNA adducts elevated in chronically exposed rats were of types also present in controls and were also induced by carbon black, which simulated diesel soot without mutagens or carcinogens (Randerath et al., 1995). Studies from multiple laboratories demonstrated that lung tumor risks for mice or Syrian hamsters could not be predicted from the rat response. 16.4.1.2 Epidemiological Evidence for Lung Cancer Risk Contemporary estimates of cancer risk from diesel exhaust are based on epidemiological evidence from occupational populations having known or presumed high exposure to DPM. It was initially hoped that the results of animal studies would provide insight into human risk; but as described above, results from animal studies have not served that purpose well. Regulatory agencies have instead relied on epidemiological studies of occupational groups having presumed high exposures (Cal EPA, 1998; EPA, 2002). The goal has been to use epidemiologic data to estimate exposure-response relationships and determine risk per unit of exposure that would permit the extrapolation of risk to the lower environmental exposures of the general public. A major limitation in assessing the degree of human lung cancer risk has been the lack of direct information regarding exposure. In the few studies including measurements, there was uncertainty regarding linkage of those data to an epidemiologic database that would permit a quantitative or semiquantitative estimation of risk. In an epidemiologic study, one would like to directly measure the putative carcinogenic agent. However, the mechanism of carcinogenesis attributable to diesel exhaust in humans is uncertain. It is not clear if concern regarding the potential for diesel exhaust to cause lung cancer is unique to diesel exhaust or if it is common to PM2.5 or ultra-fine particles (PM0.1) that are also contributed by other sources of fossil fuel combustion, such as spark-ignition vehicles, or by other sources of PM with or without associated organics. Given this complexity, it has been difficult to measure exposure that specifically targets the constituent of diesel exhaust mechanistically related to lung cancer risk. Measurements that reflect exposure to the EC core of diesel soot have been used in some air pollution and occupational studies. Although it will take years for new emissions reduction technologies to fully saturate the market, the large differences in emissions from diesel and spark-ignition engines will be progressively reduced. More recent results of population-based air pollution prospective cohort studies have also related lung cancer risk to general environmental PM2.5 (Dockery et al., 1993; Pope et al., 2002; Laden et al., 2006a), and city-based air pollution studies have related risk to exposures to vehicle emissions in general based on modeling NO2 (Nyberg et al., 2000; Nafstad et al., 2003). In addition, organic compounds in the vapor and soot phases of diesel exhaust, and their metabolites, might be transferred from the lung to other organs. There is also a probable

HEALTH EFFECTS

575

association, for example, between bladder cancer and occupations having high exposure to diesel exhaust (Boffetta and Silverman, 2001). Numerous epidemiological studies (>40) addressing the relationship between diesel exhaust exposure and lung cancer have been reported, if one considers both the studies focusing specifically on diesel exhaust and those including occupations that may be presumed to have received substantial exposures to diesel exhaust based on job title (Lipsett and Campleman, 1999; Bhatia et al., 1998; HEI, 1995). There are only two series of published epidemiological studies where an exposure assessment accompanied the analysis of lung cancer risk. These studies include U.S. railroad workers (Garshick et al., 1987, 1988, 2004, 2006; Larkin et al., 2000; Laden et al., 2006b) and trucking company workers (Steenland et al., 1990, 1992). In both these series of studies, an assessment of current exposure was used to validate the exposure assignments used in the epidemiological analysis. Exposure classifications in other studies were derived from job history by interview of subject or family, job history from employment records, general occupational history from interview, general occupational history from death certificate, and record of membership in a trade union. In the absence of an exposure assessment, limitations in interpreting results from such studies arise from uncertainties regarding the linkage between actual job duties and the extent of diesel exhaust exposure. Although concern has been raised regarding confounding by cigarette smoking in estimates of lung cancer risk, a similarly increased risk is observed in occupational studies whether or not smoking is accounted for, and the air pollution studies assessing lung cancer risk have adjusted for smoking habits. The extent that unmeasured cigarette smoking may confound the assessment of lung cancer risk depends on the association between measures of exposure and smoking. The studies and issues regarding adjusting for smoking were reviewed by the Health Effects Institute in 1995 (HEI, 1995) and in two meta-analyses published in the late 1990s (Lipsett and Campleman, 1999; Bhatia et al., 1998). This section presents results from studies with either large numbers of lung cancer cases or those using the strongest indices of diesel exhaust exposure to provide the reader with an overview of the body of literature that supports an association between lung cancer and diesel exhaust exposure. The studies are summarized in tabular form and divided into studies based on general population registries, occupational studies, and air pollution studies. Selected study characteristics and reported relative risks for lung cancer are presented, and for reports listing multiple study groups or analyses, only the results considered most robust are presented. The 95% confidence intervals (CI) for relative risks are given if they were reported. Hospital, Registry, or General Population-Based Studies American Cancer Society Cohort Boffetta et al. (1988) (Table 16.3) examined the relationship between lung cancer and occupational exposure to diesel exhaust using data from a prospective mortality study begun in 1982 by the American Cancer Society. Living volunteer subjects from across the United States were enrolled in this study by completing a questionnaire that included, among many other items, information on smoking, asbestos exposure, occupation, job held for the longest period, and exposures to diesel exhaust. Cancer Society volunteers checked the status of enrollees every 2 years, and death certificates were obtained for decedents. The analysis was limited to men 40–79 years old at enrollment, whose status was recorded at the end of the first 2-year followup (September 1984). Information was obtained for 11,044 decedents (1266 lung cancer

576

2-year prospective cohort of 461,981 U.S. males aged 40–79 followed over 2 years

French hospital case–control study

Case–control study of FL, LA, NJ hospital and general populations

Benhamou et al. (1988)

Hayes et al. (1989)

Design

Occupation by questionnaire

Occupation by questionnaire

1,444

Occupation by questionnaire

Exposure Assessment

1,260

1,266

Number of Lung Cancer Cases

Hospital, Registry, or General Population Based Studies

Boffetta et al. (1988)

References

TABLE 16.3

þ

þ

174

þ

10

112

128

157

48

14

15 5

Number of Cases with Exposure

Control for Smoking

1.5, truck drivers 10 years 2.1, heavy equipment operator 10 years

1.35, transport operators 1.42, motor vehicle operators

1.05, 1–15 years of exposure 1.21, 16þ years of exposure 2.67, miner 2.60, heavy equipment operator 1.59, railroad worker 1.24, truck driver

Relative Risk

0.6–7.1

1.1–2.0

1.07–1.89

1.05–1.75

0.93–1.66

0.94–2.69

1.63–4.37 1.12–6.06

0.94–1.56

0.80–1.39

95% Confidence Interval

577

Occupation by questionnaire

3,498

Population based case–control study, East and West Germany

Population-based case–control study, Stockholm County, Sweden

Bruske-Hohlfeld et al. (1999)

Gustavsson et al. (2000)

Occupational title

Census classification of occupation

28,744

Hansen et al. (1998) Case–control study, Danish cancer registry

70

Occupation by questionnaire by participant or surrogate

3,792

Case–control study, Detroit cancer registry

Swanson et al. (1993)

Occupation by questionnaire and self-reported exposure

2,584

Case–control study, 18 U.S. hospitals

Boffetta et al. (1990)

þ

þ

þ

200

412

716

277 1002

972

121

38

78

114 12

210 0.64–1.09 0.87–6.57

0.75–1.12

0.76, >0–9 years of exposure 1.21, 10–29 years of exposure 1.38, 30þ years of exposure

1.43, all drivers/machine operators 1.84, 10–20 years 1.62, 20–30 years 1.32, 30þ years 1.44, truck, bus, taxi driver West Germany

1.31, truck and bus drivers 1.64, taxi drivers 1.39, unspecified drivers

(continued)

0.97–1.97

0.88–1.65

0.51–1.13

1.34–2.52 1.16–2.24 0.95–1.93 1.18–1.76

1.23–1.67

1.22–2.19 1.30–1.51

1.17–1.46

1.4, heavy truck 0.8–2.4 driver 1–9 years 1.6, heavy truck 0.8–3.5 drivers 10–19 years 2.5, heavy 1.4–4.4 truck drivers 20þ years

0.92, all “probably exposed” 0.83, truck drivers 2.39, self-reported exposure 31 years

578 Number of Lung Cancer Cases Exposure Assessment

Retrospective cohort study; cancer incidence 1971–1995 in Finland

33,664

Census classification of occupation, job exposure matrix

Census 6,266 cases Retrospective classification out of 7,400,000 cohort study; of occupation, person-years cancer incidence job exposure follow-up 1971–1989 in matrix Sweden linked to cancer registry

Design

(Continued)

Guo et al. (2004)

Boffetta et al. (2001)

References

TABLE 16.3 Control for Smoking

0.98, lowest exposure in men 1.04, middle exposure in men 0.95, highest exposure in men 0.99, any exposure in men 1.22, any exposure in women

2,436 758 220

3,414 32

1,058

1.2, high probability of exposure 1.3, high intensity of exposure

Relative Risk

1,841

Number of Cases with Exposure

0.96–1.03 0.85–1.73

0.83–1.10

0.97–1.12

0.94–1.03

1.26–1.42

1.10–1.21

95% Confidence Interval

HEALTH EFFECTS

579

cases) among 461,981 men, including 174 lung cancer cases among 378,622 men with exposure to diesel exhaust. The asbestos- and smoking-adjusted relative risk for lung cancer of 1.18 (95% CI ¼ 0.97–1.44) was increased among all men with self-reported exposure to diesel exhaust. When the data for all exposed men were stratified by length of exposure to diesel exhaust, the adjusted relative risk for lung cancer was 1.05 (95% CI ¼ 0.80–1.39) for men with 1–15 years of exposure, with a suggestion of an increase in risk among men with at least 16 years of exposure (1.21; 95% CI ¼ 0.94–1.56). Analysis by occupation demonstrated elevated smoking-adjusted relative risks for lung cancer among miners (2.67, 95% CI ¼ 1.63–4.37) and heavy equipment operators (2.60, 95% CI ¼ 1.12– 6.06), railroad workers (1.59; 95% CI ¼ 0.94–2.69), and truck drivers (1.24; 95% CI ¼ 0.93–1.66). Analysis restricted to truck drivers demonstrated no significant difference between relative risks for lung cancer among men reporting diesel exhaust exposure (1.22; 95% CI ¼ 0.77–1.95) and those reporting no exposure (1.19; 95% CI ¼ 0.74–1.89). Analysis by duration of diesel exhaust exposure among truck drivers yielded a positive time-response trend with relative risks of 0.87 for 1–15 years of exposure and 1.33 for 16 years or more of exposure. These results suggest a small positive association between lung cancer risk and occupations with high presumed exposure to diesel exhaust and a trend toward increasing lung cancer risk with time in those occupations. French Case–Control Study Benhamou et al. (1988) reported a case–control study of occupational risk factors among the French population. A total of 1260 male cases observed between 1976 and 1980 were matched by age, hospital of admission, and interviewer with 2084 controls, and information on occupation and smoking was obtained by interview. The smoking-adjusted relative risk for lung cancer among 285 cases and 391 controls was found to be significantly increased (p ¼ 0.01) to 1.35 for transport operators (95% CI ¼ 1.05–1.75) and to 1.42 for motor vehicle drivers (95% CI ¼ 1.07–1.89). National Cancer Institute Pooled Case–Control Study Hayes et al. (1989) reported a case–control study of lung cancer in motor-exhaust related occupations that used data from National Cancer Institute hospital- and population-based studies in Florida (1976–1979), Louisiana (1979–1983), and New Jersey (1980–1981). The study included a total of 1444 male cases and 1893 controls for which occupation and smoking information was obtained by interview. Among 122 cases and 113 controls, the smoking-adjusted odds ratio (OR) for lung cancer was elevated for truck drivers for 10 years or longer (1.5, 95% CI ¼ 1.1–2.0) and heavy equipment operators for 10 years or longer (2.1, 95% CI ¼ 0.6–7.1). Although less specific for diesel exhaust, the OR for lung cancer after 10 years or more of employment in all vehicle exhaust-related jobs was increased to 1.5. In general among job categories, the risk was greater among persons employed 10 or more years as compared to workers employed less than 10 year. American Health Foundation Case–Control Study Bofetta and coworkers (1990) also conducted a case–control study using data from 18 U.S. hospitals that included 2584 male cases of confirmed lung cancer and 5099 controls matched for age, date, and hospital of admission 1977–1987. Information on usual occupation and smoking was obtained by interview, and later questions inquired specifically about diesel exhaust exposure. Data were divided into occupations with no, possible, and probable diesel exhaust exposure, and truck

580

DIESEL EXHAUST

drivers were examined as a subgroup within the probably exposed group. The smokingadjustedORratioforlungcanceramongthegroupwithprobableexposure(210cases)was0.92 (95% CI ¼ 0.75–1.12), and that among truck drivers (114 cases) was 0.83 (95% CI ¼ 0.64– 1.09). The OR increased with duration of service in occupations with probable exposure, reaching 1.49 for 31 years or longer, but the trend did not reach significance (p ¼ 0.18). There wasnosuggestionofanincreasingORratiowithlengthofserviceamongtruckdrivers.Boththis study and the previous American Cancer Society Cohort study by Boffetta and coworkers, although large, were likely to have considerable misclassification of diesel exhaust exposure because classification was based on self-report, reducing the ability to detect an effect of exposure on lung cancer risk. Detroit Area Case–Control Study Swanson et al. (1993) reported lung cancer results from an occupational cancer incidence study of men in the Detroit area using a cancer registry for the period of 1984–1987. Their study involved a total of 3792 cases and 1966 colon/rectum cancer controls, in which work and tobacco use histories were obtained by interview of the subjects or close surrogates. Although numerous occupational groups were listed in the report, those likely to have had the greatest exposure to diesel exhaust were the drivers of heavy trucks. This category included 325 white and 71 black male cases, matched with 164 white and 41 black male controls. The relative risk for lung cancer among white males was related to length of occupation as a driver of heavy-duty trucks, ranging from 1.4 (95% CI ¼ 0.8–2.4) for 1–9 years to 2.5 (95% CI ¼ 1.4–4.4) for 20 years or more. These results suggest an exposure-response relationship, but the results from black men (albeit with fewer cases) did not, although overall lung cancer risk was elevated. Professional Drivers in Denmark Hansen et al. (1998) identified 28,744 men born in 1897–1966 in whom a primary lung cancer was diagnosed in 1970–1989 as identified through the Danish Cancer Registry. Past employment was ascertained by record linkage with a nationwide pension fund that included the dates of starting and stopping work at a particular company. Job titles were retrieved from the Danish Central Population registry. Controls were matched based on year of birth and sex, and had to be alive and employed without cancer before the case was diagnosed. Adjusting for socioeconomic status based on occupational title, the OR for truck and bus drivers (972 cases and 668 controls) was 1.31 (95% CI ¼ 1.17–1.46), for taxi drivers (277 cases and 149 controls) OR ¼ 1.64 (95% CI ¼ 1.22–2.19), and for unspecified drivers (1002 cases and 598 controls) OR ¼ 1.39 (95% CI ¼ 1.30–1.51). For both truck and bus drivers and taxi drivers, there was an increasing risk of lung cancer with greater years of work. Data on smoking in these occupations was available indirectly from national surveys conducted in Denmark in 1972 and 1983, and smoking rates were similar among drivers and nondrivers in working men, suggesting that the results were not confounded by smoking. Occupational Exposure to Diesel Engine Emissions in Germany Bruske-Hohlfeld et al. (1999) studied the association between occupation and lung cancer in diesel exposed workers in a pooled analysis of two case–control studies that included 3498 male cases and 3541 population controls. Information regarding occupational exposure and smoking was obtained by questionnaire, and jobs were divided into four groups: professional drivers of trucks, buses, and taxies; other exposed jobs, including diesel locomotive and diesel forklift truck drivers; machine operators, including bulldozer, grader, and excavator drivers; and tractor drivers. There were 1146 men occupationally exposed to diesel exhaust (716 cases and

HEALTH EFFECTS

581

430 controls). For all jobs, the crude risk of lung cancer was 1.91, which was reduced to 1.43 (95% CI ¼ 1.23–1.67) by adjusting for smoking and asbestos exposure. There was an increased lung cancer risk with greater years of exposure through 10–20 years (OR ¼ 1.84; 95% CI ¼ 1.34–2.52), but risk slightly decreased with 20–30 years (OR ¼ 1.62; 95% CI ¼ 1.16–2.24) and greater than 30 years of exposure (OR ¼ 1.35; 95% CI ¼ 0.95–1.93). Among those with other diesel exposed jobs (99 cases, data not shown) lung cancer risk was also elevated (OR ¼ 1.53; 95% CI ¼ 1.04–2.24), with evidence of greater risk with greater years of exposure; for tractor drivers (52 cases) there was evidence of increasing risk with increasing duration of employment, and there was an overall increased risk in heavy equipment operators (81 cases; OR ¼ 2.32; 95% CI ¼ 1.44–3.70). Differences in the risk for truck, bus, and taxi drivers were examined separately for East Germany and West Germany. In West Germany (412 cases), the overall risk for professional drivers of trucks, buses, and taxis (data not shown) was 1.44 (95% CI ¼ 1.18–1.76), whereas in East Germany (122 cases) it was 0.83 (95% CI ¼ 0.60–1.14). In West Germany, there was a suggestion of an increase with greater driving hours, and a greater risk if one started to drive 1946, and in particular, 1956 (OR ¼ 1.60; 95% CI ¼ 1.32–1.96), which most likely represented times with the greatest numbers of vehicles on the rounds, particularly diesel cars and trucks. The differences in between East Germany and West Germany were attributable to differences in traffic density, which was estimated to be five times greater in West Germany. Occupational Factors and Lung Cancer Risk in Sweden Gustavsson et al. (2000) studied lung cancer identified in cancer registries in men aged 40–75 who were residents of Stockholm County, Sweden, between 1985 and 1990, and who had lived outside the county for no more than 5 years between 1950 and 1990. Referents were chosen at random from the general population, and included mortality-matched referents. Smoking and occupational histories were obtained from the subject if alive, or from next of kin. Occupations were coded into an exposure intensity and probability matrix, and job-specific historical values of exposure to NO2 were used to estimate occupational exposure to diesel exhaust. Residential exposures to NO2 were modeled using a historical inventory linked to a traffic grid. Adjusting for smoking and ambient NO2, there was increased lung cancer in the highest cumulative occupational diesel exhaust exposure category (1.63; 95% CI ¼ 1.14–2.33), and for 30 years of exposure relative risk ¼ 1.38 (95% CI ¼ 0.97–1.97) (data not shown). Swedish Cancer Environment Registry Study Boffetta et al. (2001) investigated the risk of cancer among workers exposed to diesel exhaust using the Swedish Environment Register III that contains nationwide data on cancer incidence for 1971–1989 and linked this to occupation and industry of employment as reported in the 1960 census. Using a jobexposure matrix, exposures were graded based on intensity and probability of exposure to diesel exhaust. Mortality was ascertained between 1971 and 1989 by linkage to the Swedish Cancer Registry and Register of Causes of Death. There were a total of 28 million person-years of observation, for which 26% was in people classified as exposed. The analysis of lung cancer risk in men (6266 cases overall) with a high probability of exposure was increased (1.2; 95% CI ¼ 1.10–1.21), and there was an increasing risk with increasing exposure intensity (high intensity relative risk ¼ 1.3; 95% CI ¼ 1.26– 1.42). There were fewer cases (n ¼ 57) and less exposure in women, and lung cancer risk was not increased. Although there was no specific information regarding smoking, it was noted that other smoking-related tumors were not increased based on diesel exposure, suggesting no confounding by smoking.

582

DIESEL EXHAUST

Finnish Worker Study Guo and coworkers (2004) studied the association between lung cancer and occupation during 1971–1995 in Finland by linking census occupation reported in 1970 to lung cancer mortality. A job-exposure matrix was used to estimate exposures to gasoline and diesel exhaust based on occupation. The authors used estimates of NO2 exposure as surrogates of diesel exposure and estimates of CO exposures as measures of gasoline engine exhaust exposures. There were 33,664 cases of lung cancer identified. Based on exposure to NO2, the overall relative risk for lung cancer was 0.99 (95% CI ¼ 0.96–1.03) among men and 1.22 (95% CI ¼ 0.85–1.73) among women. Based on estimated CO exposures (data not shown), the relative risk of lung cancer was 1.05 (95% CI ¼ 1.01– 1.09) among men and 1.61 (95% CI ¼ 1.23–2.12) among women. Smoking information was obtained from general population surveys and was incorporated into the analysis, but the specific methods were not stated. Results based on job title were not reported. The authors attributed the generally negative results to low exposures due to the operation of diesel vehicles and other equipment in rural areas with a low population density. It is also possible that considerable misclassification was introduced by the quantitative estimation of exposures. Occupational Case–Control And Cohort Studies Swedish Dockworker Studies Gustafsson et al. (1986) (Table 16.4) compared the incidence of lung cancer among male Swedish dock workers to that among the Swedish male population. Diesel trucks were introduced into Swedish ports in the late 1950s and became prevalent during the 1960s. The cohort consisted of 6071 men employed for a minimum of 6 months before 1974 and followed from 1961 to 1981. Twenty percent of the cohort had 30 or more years of service, and only 10% had less than 5 years of service. There were 70 cases of lung cancer among the 1062 cohort deaths. The relative risk for lung cancer among the dock workers was found to be significantly increased to 1.32 (95% CI ¼ 1.05–1.66). Similarly, Emmelin et al. (1993) reported increasing relative risks for lung cancer with exposure time for 50 cases among nonsmoking Swedish dock workers ranging from 1.0 (no increase) for the lowest category to 2.9 for the highest. Exposure to diesel exhaust was estimated based on diesel fuel consumption and number of workers in each Swedish port, and the cases and controls were selected from male dock workers employed for at least 6 months during 1950–1974, with case ascertainment starting in 1960 through 1982. Based on 50 cases and 154 referents with complete information available and adjusting for smoking (yes/no), there was an increase in the OR ratio for lung cancer with increasing exposure for three indices of exposure (years since diesel equipment was used in a port, estimates of cumulative fuel consumption, and years that fuel use was above a minimum level in a port), consistent with an exposure-response relationship. U.S. Teamster Case–Control Study Steenland et al. (1990, 1992) reported a study of lung cancer among truck drivers in the Central States Teamsters’ Union. The study population included a total of 996 cases and 1085 controls, for whom death certificates were obtained, and information on work history, smoking, and asbestos exposure was obtained from next of kin. Job history information was also available from the worker’s retirement applications. The subjects died in 1982–1983 and were receiving pensions, which required a minimum of 20 years of union membership. Covariates included in the analysis were age, smoking, asbestos exposure, and jobs with diesel exposure. Heavy-duty diesel

583

Retospective cohort of 6,071 male dockworkers 1961–1981

Nested case–control study in dockworker cohort

Case–control study of mortality based in Teamster Pension fund

Emmelin et al. (1993)

Steenland et al. (1990)

Design

Use of diesel equipment and fuel consumption Occupation by questionnaire next of kin questionnaire, retirement application

996

Job title

Exposure Assessment

50

70

Number of Lung Cancer Cases

Occupational Case–Control and Cohort Studies

Gustafsson et al. (1986)

References

TABLE 16.4

162

þ

37

36

213

1,002

228

12 19 19

70

Number of Cases with Exposure

þ

Control for Smoking

1.06, 1–11 years intercity truck driver 1.41, 12–17 years intercity truck driver 1.55, 18þ years, intercity truck driver 1.11, 1–11 years city truck driver 1.15, 12–17 years city truck driver 1.79, 18þ years, city truck driver Years of work based on years after 1959, the date diesel trucks were introduced

1.0, reference 1.6, medium exposure 2.9, highest exposure

Standardized mortality ratio ¼ 1.32

Relative Risk

(continued)

0.94–3.42

0.80–4.19

0.94–3.42

0.97–2.47

0.90–2.21

0.68–1.70

0.83–1.10 0.5–5.1 0.8–10.7

1.05–1.66

95% Confidence Interval

584 Occupation by questionnaire

122

Retrospective cohort study of 389,000 male Swedish construction workers 1971–1993

Jarvholm and Silverman (2003)

Occupational title

38

Retrospective cohort mortality study of 5,536 German Potash miners 1970–1994

Saverin et al. (1999)

Employment in cohort

Exposure Assessment

Occupational title

Nested case–control study in 1998 cohort

Soll-Johanning et al. (2003)

473

Number of Lung Cancer Cases

153

Retrospective cohort mortality study of 18,174 bus drivers and other tramway employees in Copenhagen 1990–1994

Design

(Continued)

Soll-Johanning et al. (1998)

References

TABLE 16.4

Indirect assessment

Indirect assessment

þ

Control for Smoking

61

61

38

11

153

473

Number of Cases with Exposure

0.87, heavy equipment operator 1.29, truck driver Risk compared electricians/carpenters

2.17, exposed production workers 1.68, based on highest category of cumulative exposure

0.97, years of work as bus driver

1.2, compared to population rates

Relative Risk

0.99–1.65

0.66–1.11

0.49–5.8

0.79–5.99

0.96–0.99

1.1–1.3

95% Confidence Interval

585

Retrospective cohort mortality study in 43,826 Canadian railroad retirees

Case–control study of all active and retired U.S. railroad workers (650,000 workers). Deaths collected 1981–1982

Retrospective cohort mortality study of 54,973 railroad workers 1959–1996

Retrospective cohort mortality study of railroad workers 1959– 1996, includes 39,388 deceased workers

Howe et al. (1983)

Garshick et al. (1987)

Garshick et al. (1988, 2004)

Garshick et al. (2006)

Yearly job title from retirement board. Next of kin smoking history Yearly job title

Yearly job title

4,351

4,055

Job title at retirement

1,256

933

Indirect adjustment; imputation of smoking history using information from 1987 case– control study

Indirect adjustment

þ

1.40, train crews, unadjusted for smoking Estimated smoking adjusted relative risk 1.17–1.27 1.35, train crews, unadjusted for smoking 1.22, train crews, smoking adjusted

2,358

1.41, for 20 years of work in an exposed job

1.20, possible exposure 1.35, probably exposed

2,479

335

374 306

1.24–1.46 1.12–1.32

1.30–1.51

1.06–1.88

p ¼ 0.012 p < 0.001

586

DIESEL EXHAUST

trucks for inter-city use were introduced in the industry during the 1950s. Using retirement application job history, drivers of inter-city trucks (552 cases and 604 controls) with the greatest duration of work after 1959 (18þ years) had the greatest lung cancer risk (OR ¼ 1.55; 95% CI ¼ 0.97–2.47) with a significant linear trend with years of work. However, for city drivers who drove gasoline-powered heavy-duty trucks (113 cases and 135 controls), the risk of lung cancer was similarly elevated with 18þ years of work after 1959 (OR ¼ 1.79; 95% CI ¼ 0.94–3.42). City drivers with shorter job duration had lower risks. These data are consistent with the exposure assessment that accompanied the study (Zaebst et al., 1991; Steenland et al., 1992), indicating that local city truck drivers and inter-city truck drivers had similar EC exposures that reflected background exposure from traffic rather than exposure from their own truck. Danish Urban Bus and Tramway Study Soll-Johanning et al. (1998) conducted a retrospective cohort mortality study of 18,174 bus drivers or tramway employees in Copenhagen during 1990–1994 and reported an increased risk of lung cancer based on linkage to the Danish Cancer Registry. Among workers employed for 3 or more years, the relative risk of lung cancer compared to Copenhagen rates was 1.2 (95% CI ¼ 1.1–1.3) based on 473 cases. The same group (Soll-Johanning et al., 2003) conducted a nested case–control study using these cases. There were 257 lung cancer cases where information regarding smoking history was potentially available from the wife or case file. This information was obtained for 153 cases, and the study also included 351 controls. An increased risk of lung cancer was not related to duration of employment, and there was a decreased risk of lung cancer with increasing years or work as a bus driver. There was some limited information about driving route, and driving in areas of high air pollution was not associated with risk. Lung Cancer Mortality in Potash Miners Saverin et al. (1999) studied lung cancer mortality in Potash miners in Germany. Diesel equipment was introduced into Potash mines in 1969–1970, and in 1991, the mines closed. Workers had medical examination every other year, and records on smoking were maintained through 1982. There were 5536 men who had worked underground for at least 1 year after 1969, and mortality was ascertained for 1970– 1994. Estimates of diesel exposure were obtained in 1992 and expressed as total carbon in respirable dust, and because technology had not changed, these levels were assumed to be representative of previous exposure. Although medical records were found to classify former smokers as nonsmokers in 28% of cases when compared to a personal interview, smoking was not associated with exposure so was not considered to be a confounder in the analysis. The exposed workers were the production workers, and the relative risk of lung cancer in production workers who had worked underground for at least 10 years (11 cases) compared to other workers (6 cases) was elevated but imprecise (relative risk ¼ 2.17; 95% CI ¼ 0.79– 5.99). Results using categories of cumulative exposure gave similar results (relative risk ¼ 1.68; 95% CI ¼ 0.49–5.8). Although miners typically have much higher diesel exhaust exposure levels than other occupations, the cohort was small and exposure for many was of short duration, potentially limiting the detection of lung cancer risk. Swedish Truck Driver and Construction Vehicle Case–Control Study Jarvholm and Silverman (2003) studied lung cancer in 389,000 male Swedish construction workers and identified truck drivers and heavy equipment operators. Workers were identified based on health examinations during 1971–1993 and linked to the Swedish National Cancer Registry and National Death Registry and lung cancer cases identified through 1995. Subgroups of

HEALTH EFFECTS

587

exposure were created based on whether a cabin was on the construction equipment. There were 14,364 heavy equipment operators (61 lung cancer cases), 6364 truck drivers (61 lung cancer cases), and 119,984 carpenter/electrician referents (512 lung cancer cases). Eighty percent of the heavy equipment operators operated the same machine between the first and last examination, indicating high job stability. Heavy equipment operators had a lower incidence of lung cancer compared to electricians/carpenters (standardized incidence ratio; SIR ¼ 0.87; 95% CI ¼ 0.66–1.11) and the general population (SIR ¼ 0.76; 95% CI ¼ 0.58–0.97), whereas truck drivers had a greater incidence of lung cancer compared to electricians/carpenters (SIR ¼ 1.29; 95% CI ¼ 0.99–1.65) and the general population (SIR ¼ 1.14; 95% CI ¼ 0.87–1.46), and mortality ratios were similar. Information on smoking habits was available from the examination in a subset of workers, and smoking rates were similar among the jobs, making it unlikely that differences in smoking rates accounted for the finding. When the heavy equipment operators were categorized based on use of cabins, lung cancer risk for never in a cabin was 0.86 (95% CI ¼ 0.5–1.6); for sometimes in a cabin, SIR ¼ 0.71 (95% CI ¼ 0.5–1.0); and for always in a cabin, SIR ¼ 0.50 (95% CI ¼ 0.20–0.70). This trend (p < 0.001) was suggested that working inside a cabin while on a construction vehicle was associated with a lower lung cancer risk. Railroad Worker Case–Control and Cohort Studies Howe et al. (1983) conducted a retrospective cohort study of lung cancer among 43,826 male employees of the Canadian National Railway retired and alive in 1965 or retiring between 1965 and 1977. The total of 16,812 deaths included 933 deaths from lung cancer. The subjects were classified by job at the time of retirement into nonexposed, possibly exposed, and probably exposed to diesel exhaust. A highly significant relationship was found between relative lung cancer risk and the presumed level of exposure:nonexposed ¼ 1.00, possibly exposed ¼ 1.20, and probably exposed ¼ 1.35. As noted earlier, the U.S. railroad industry converted from steam to diesel powered locomotives mainly starting in the late 1940s, and by 1959, 95% of the locomotives in service were diesel powered. Garshick and coworkers (1987) collected death statistics over 1 year (1981–1982) from a population base of 650,000 active and retired male U.S. railroad workers with 10 years or more of service, using records from the Railroad Retirement Board. Their study included a total of 1256 exposed cases and 2385 controls, assigned on the basis of job records and contemporary measurements (early 1980s) of diesel exhaust concentrations in similar job environments (Woskie et al., 1988a, 1988b). Lung cancer cases and controls were matched by birth and death date. The cases and controls were classified by age, length of service, smoking (next of kin history), and likely exposure to asbestos. Exposed workers included train crews and locomotive shop workers; unexposed workers included workers not in these job groups. The cases were divided at age 64 into younger and older groups, with the younger group presumed to have more years of diesel exposure because of the dates of railroad dieselization, and years of work starting in 1959 in a diesel-exposed job was used as a continuous exposure variable. After adjustment for smoking (pack-years) and asbestos exposure (yes/no), the odds ratio for lung cancer among 335 cases and 637 controls for working 20 years in an exposed job was 1.41 (95% CI ¼ 1.06–1.88) (data not shown). After similar adjustments, the odds ratio for workers with 20 or more years of diesel exposure was 1.64 (95% CI ¼ 1.18–2.29), and in analyses examining mortality in the train crews with 20 or more years of exposure was 1.55 (95% CI ¼ 1.09–2.21). To exclude the effects of recent diesel exhaust exposure, the data were analyzed excluding exposures during the 5 years preceding death, and the relative risk for lung cancer remained similarly elevated.

588

DIESEL EXHAUST

Garshick et al. (1988, 2004) also conducted a retrospective cohort study of lung cancer mortality among 54,973 white male U.S. railroad workers, 40–64 years old in 1959, who had begun work 10–20 years earlier. In the original publication, mortality was assessed through 1980 (Garshick et al., 1988), and then later updated through 1996 (Garshick et al., 2003). The cohort was selected on the basis of job title in 1959 using records from the U.S. Railroad Retirement Board. As in the case–control study, exposed workers included train crews and locomotive shop workers, and the unexposed group included clerks, ticket and station agents, and signalmen. Jobs with the most likely exposure to asbestos were excluded. There were 45,593 deaths over the 38 years of followup, including 4351 lung cancer deaths. Workers on operating trains (train crews) had a relative risk of lung cancer mortality of 1.40 (95% CI ¼ 1.30–1.51). There was no increase in lung cancer mortality with greater years of work that was attributed to a healthy worker survivor effect and insufficient information regarding historical changes in railroad exposures. Locomotive shop workers did not have an increased lung cancer risk, but it was later noted that the job titles of the workers included were not specific for diesel locomotive shops, thereby reducing the ability to detect an effect of exposure. Although there was no specific cigarette smoking history information available from the workers in the cohort, smoking information was available from surveys of railroads workers, including the 1987 case–control study conducted by Garshick et al., 1987; Larkin et al., 2000). This information was used to estimate an effect of smoking that reduced the relative risk to 1.17–1.27 (Garshick et al., 2004). The authors also conducted an additional analysis to assess whether differences in smoking behavior between diesel exposed and unexposed workers influenced the risk of lung cancer (Garshick et al., 2006). A simulation of smoking behavior using cause of death, birth cohort, age, and job-specific smoking prevalence from the 1987 lung cancer case–control study was conducted for 39,388 deceased railroad workers. The risk of lung cancer among exposed workers unadjusted for smoking was 1.35 (95% CI ¼ 1.24–1.46), and after adjustment an excess risk remained (1.22; 95% CI 1.12–1.32). In order to improve the estimation of historical exposures during the transition from steam to diesel locomotives, historical information on diesel locomotives used by each railroad was obtained (Laden et al., 2006b). Starting in 1945, annual railroad-specific weighting factors for the probability of diesel exposure were calculated. Among workers hired after 1945, as diesel locomotives were introduced, the relative risk of lung cancer for any exposure was 1.77 (95% CI ¼ 1.50–2.09), and there was evidence of an exposureresponse relationship with exposure duration. Air Pollution Cohort Studies Diesel exhaust contributes to ambient NO2 exposures, and two studies have used estimated values of NO2 as an index of overall traffic exposure and have assessed its relationship with lung cancer. Nyberg et al. (2000) conducted a case– control study among men 40–75 years old that included all cases of lung cancer 1985–1990 in residents of Stockholm County, and smoking histories were obtained from next of kin for 1042 cases and 2364 controls. Geographic Information System techniques were used to estimate residential exposures to NO2 as an index of traffic exposure by linking a regional emission database to a road network. Adjusting for smoking, age, occupational exposures to diesel exhaust, socioeconomic group, and asbestos, the relative risk of lung cancer for the highest decile of average NO2 exposure lagged 20 years was 1.44 (95% CI ¼ 1.05–1.99). Nafstad et al. (2003) conducted a similar study in Oslo, Norway, linking 16,209 participants

HEALTH EFFECTS

589

in a prospective health study in 1972–73 to estimates of outdoor NO2 levels at their residential address. Adjusting for age, smoking, and education, there was a significant increase in lung cancer risk (1.08; 95% CI ¼ 1.02–1.15) for every 10 mg/m3 increase in NO2. In contrast, a study of cancer incidence and residential traffic density in Amsterdam 1989– 1997 did not find a consistent association between distance from a roadway, traffic density, and lung cancer (Visser et al., 2004). Two air pollution studies have related fine PM (PM2.5) to lung cancer mortality but were not able to examine more specific markers of traffic or diesel exposure. In the American Cancer Society Prevention II follow-up study, 1.2 million adults enrolled in 1982 had mortality ascertained through 1998 (Pope et al., 2002). Using national networks of monitors, average PM2.5 levels were linked using zip code to 319,000 people in 51 metropolitan areas. Adjusting for age, sex, race, smoking, education, and diet, there was an increased risk of dying of lung cancer (1.14; 95% CI ¼ 1.04–1.23) for every 10 mg/m3 increase in PM2.5. Similar results were obtained from the Harvard Six Cities Study where lung cancer mortality in 8096 participants starting in the 1970s through 1990 was related to average PM2.5 levels (relative risk ¼ 1.27; 95% CI ¼ 0.96–1.69), adjusting for smoking and multiple other covariates (Laden et al., 2006a). Summary of Epidemiological Evidence for Lung Cancer Despite limitations in the assessment of exposure, when considered together, the weight of the human epidemiologic evidence reviewed above supports a small increased risk of lung cancer associated with diesel exhaust exposure. This increased risk was observed in workers with long-term employment in a variety jobs involving exposures to diesel exhaust. The studies indicating statistically significant increases gave estimates of increases ranging from approximately 20% (relative risk of 1.2) to approximately two-fold increases (relative risk of 2.0). The level of confidence with which one can draw conclusions from the epidemiological studies of workers regarding cancer risks from lower exposures is limited due to lack of historical exposure estimates. Historical exposures were probably higher than current exposures, and there is also uncertainty regarding whether the available measurements were representative of industry-wide exposures. The existing data do not allow determination of either (1) the magnitude of increase per unit exposure; or (2) a description of the exposure-response relationship, with a high level of confidence. The EPA and others have presumed that the health risk is present at ambient levels because exposures experienced by some occupations overlap with general population exposures, such as in professional drivers. Results from general air pollution studies also support the plausibility of a lung cancer risk from ambient PM2.5 exposures that include DPM as a minor mass component. The dose required, the specific carcinogenic agent in diesel exhaust, and the extent to which the cancer risk is unique to diesel exhaust remain uncertain. Two ongoing occupational epidemiological studies may provide improved knowledge about the exposure-response relationship for lung cancer. A national exposure assessment of the trucking industry exposures was recently performed, and the assessment was specifically designed to be linked to an epidemiologic database examining lung cancer mortality (Smith et al., 2006; Davis et al., 2006). The analyses in this study are still in progress. A study of the relationship between exposure and lung cancer among workers in metal and nonmetal mines conducted by the National Cancer Institute (NCI) and the National Institute for Occupational Safety and Health (NIOSH) is also still in progress (Monforton, 2006). At this time, because lung cancer takes years to develop, and in the absence of a long-term prospective environmental

590

DIESEL EXHAUST

epidemiological study accompanied by measurements of exposure to diesel exhaust, estimates of lung cancer risk due to diesel exhaust or other traffic emissions remain dependent on historical exposure estimates. 16.4.1.3 Epidemiological Evidence for Bladder Cancer The bladder is the only organ, other than the lung, for which there has been significant concern for cancer associated with diesel exhaust exposure. There have been several studies of the incidence of bladder cancer among populations presumed to be heavily exposed to diesel exhaust. A few studies focused primarily on the relationship between bladder cancer and diesel exhaust exposure, but most included the bladder in the context of broader cancer surveys. Most of the difficulties described above for the studies of lung cancer are also inherent in the studies of bladder cancer. These include lack of quantification of exposure, lack of uniqueness of the exposure materials to diesel exhaust, the presence of several confounding factors, and potential exposure misclassification. Information on the association between diesel exhaust exposure and bladder cancer was reviewed by Boffetta and Silverman (2001), who identified 35 studies and performed a meta-analysis by job group. Among 15 studies where there was occupational exposure as a truck driver, relative risk ¼ 1.17 (95% CI ¼ 1.06–1.29), among 10 bus driver studies, relative risk ¼ 1.33 (95% CI ¼ 1.22–1.45), among heavy equipment operators in 5 studies, relative risk ¼ 1.37 (95% CI ¼ 1.05–1.81), and using a job-exposure matrix in 10 studies, relative risk ¼ 1.13 (95% CI ¼ 1.00–1.27). In contrast, Boffetta et al. (2001) also investigated the risk of bladder cancer attributable to diesel exhaust exposure in a study where national data on cancer incidence during 1971–1989 in Sweden were linked to occupation and industry as reported in the 1960 census. Using a job-exposure matrix, exposures were graded based on intensity and probability of exposure to diesel exhaust. Mortality was ascertained between 1971 and 1989 by linkage to the Swedish Cancer Registry and Register of Causes of Death. This study included 3669 cases of bladder cancer classified as low, medium, or high intensity exposure and 12,287 cases without exposure. In this very large study, no association with diesel exhaust exposure was observed. A similar study was conducted in Finland (Guo et al., 2004) that included 771 cases with exposure and 4314 cases without exposure and no association was observed. The weight of the present evidence suggests that there may be a small positive risk for bladder cancer among truck drivers and other long-term workers in occupations presumed to be exposed to diesel exhaust, but as illustrated by the latter two studies, the epidemiologic results are not as consistent as for lung cancer. 16.4.1.4 Current Classifications of Cancer Risk Several health and regulatory organizations have reviewed the evidence for human carcinogenesis from inhaled diesel exhaust and issued classifications. Some of these classifications occurred as long as two decades ago, and many have not been reviewed in light of the most recent information. In 1988, NIOSH declared diesel exhaust a “potential occupational carcinogen” (NIOSH, 1988). In 1989, the IARC classified diesel exhaust in Group 2A, “probably carcinogenic to humans, assessing the epidemiological evidence as limited and the animal evidence as sufficient” (IARC, 1989). In 1995, the HEI noted that, although the weight of epidemiological evidence suggested an association between occupational exposure to diesel exhaust and lung cancer, several uncertainties precluded judging the level of risk with confidence (HEI, 1995). In 1996, the World Health Organization’s International Programme on Chemical Safety (IPCS) classified diesel exhaust as “probably carcinogenic to humans” but noted that current

HEALTH EFFECTS

591

epidemiological data were insufficient for estimating risk quantitatively (IPCS, 1996). The California EPA reviewed the carcinogenicity of diesel exhaust in view of its potential listing as a “toxic air contaminant” under state law. California considered the epidemiological data adequate to support a causal association between occupational exposures and lung cancer, and identified “diesel exhaust particulate matter,” in contrast to diesel exhaust per se, as a toxic air contaminant (Cal EPA, 1998). In 1998, the NTP reviewed “diesel exhaust particulates” for listing in its Report on Carcinogens (NTP, 1998). The Carcinogens Subcommittee of the NTP Board of Scientific Counselors voted to list diesel exhaust particulates as “reasonably anticipated to be a human carcinogen” but considered the evidence insufficient for the alternative listing as “known to be a human carcinogen.” In 2002, the U.S. EPA categorized diesel exhaust as “likely to be carcinogenic to humans” (EPA, 2002). In aggregate, the above designations portray a general consensus that diesel exhaust, and most probably DPM, poses some level of human lung cancer risk at some exposure level. There is also continuing concern for potential cancer risk among the general population from environmental exposures that overlap, in some locations such as those associated with heavy traffic, with occupational exposure levels. Most agencies and scientists agree that the present data do not allow estimation of unit cancer risks for humans with high confidence. It should be noted that all of the above assessments were based on results from studies to date; there are no data from either animals or humans from which to estimate carcinogenic hazards or risks from exhaust from the most recent fuel, engine, or after-treatment technologies. 16.4.2

Noncancer Health Effects

There has also been concern for noncancer health effects of diesel exhaust, both because of direct exposures and because of its contribution to general air pollution. These concerns include nonmalignant respiratory effects, such as chronic obstructive pulmonary disease, asthma, allergic sensitization, respiratory symptoms, and effects in other systems. Because DPM is ubiquitous and comprises a variable portion of ambient PM, it may contribute to these health risks in the general population, particularly in close proximity to traffic. Recent evidence suggests that several types of effects are at least plausible, and these topics will be reviewed here. 16.4.2.1 Amplification of Respiratory Allergic Responses It has been hypothesized that diesel emissions have contributed to the increased incidence of asthma and allergic rhinitis in developed countries over the past several decades (Pandya et al., 2002; Riedl and Diaz-Sanchez, 2005). In part, this concern is related to a broader question about links between air pollution and increased asthma; however, air pollution levels have fallen in most countries over the same period that asthma has increased. The specific role of diesel emissions was initially questioned in Japan, where a marked increase in allergy to cedar pollen following widespread planting of cedar trees during post-war reconstruction was accompanied by increased public exposure to diesel exhaust. Research in Japan, and later in the U.S., found that high exposures to diesel exhaust or DPM could enhance allergic responses, raising the hypothesis that DPM could act as an adjuvant. Although Wade and Newman (1993) used the term “diesel asthma,” the airway hyperreactivity and reversible airflow limitation they described in three railroad workers was very likely not due to allergic sensitization. The syndrome developed after extremely high exposures, suggesting a nonspecific reaction of the airways to irritants in high concentration. The findings

592

DIESEL EXHAUST

reviewed below support the plausibility of a contribution of environmental exposures to diesel exhaust to respiratory allergies, but whether or not exhaust at environmentally relevant concentrations acts as an adjuvant in the development or exacerbation of allergies remains uncertain. Effects of Diesel Particles Research in Japan and the U.S. on the potential for DPM to amplify allergic responses was stimulated by the finding of Muranaka et al. (1986) that the IgE antibody response to ovalbumin injected intraperitoneally in mice was amplified by mixing DPM with the antigen. This study is also cited for the source of DPM used in several subsequent studies, including those described in the following paragraph. The DPM was generated by a 1980 Nissan automobile with an LD-28 2.97-L engine operated at speeds of 20–80 km/h on a chassis dynamometer. Particles were collected on filters from a constantvolume dilution tunnel. The PM composition was not reported, although a later paper (Tsien et al., 1997) indicated that the mass was 41% extractable by dichloromethane. Following this initial finding, Takafuji et al. (1987) elicited a similar amplification of allergic response in the respiratory tract by instilling ovalbumin and DPM simultaneously into the noses of mice (Takafuji et al., 1987). Maejima et al. (1997) compared the effects of different particles by dropping suspensions of DPM onto the nares of mice followed by inhalation exposure to Japanese cedar pollen weekly for several weeks. Kanto loam dust, DPM, carbon black, and coal fly ash all amplified antigen-specific IgE to a similar degree above the level in mice given pollen alone. Hao et al. (2003) sensitized BALB/c mice to ovalbumin by intraperitoneal injection with alum adjuvant, and then challenged them with aerosolized ovalbumin, either together with aerosolized DPM or followed by DPM. Treatment with DPM in either sequence enhanced the inflammatory response to antigen, but did not increase expression of antigen-specific antibodies. Researchers at the University of California Los Angeles (UCLA) have conducted numerous studies by instilling DPM intranasally in human subjects, typically at a dose of 300 mg. Japanese sources of DPM have been cited, but the composition of the material has not been reported. They first found that instillation of DPM increased IgE in nasal washings (Diaz-Sanchez et al., 1994). This was followed by demonstration that DPM also stimulated proinflammatory and proallergic cytokine production (Diaz-Sanchez et al., 1996). They found that combining DPM with ragweed allergen challenge markedly enhanced nasal ragweed-specific IgE and shifted cytokine production toward a Th2 pattern in ragweedsensitive subjects (Diaz-Sanchez et al., 1997). They inferred that soot-borne PAHs were the causal component class from finding that either a dichloromethane extract of DPM (incorrectly termed “PAH” by the authors) or pure phenanthrene increased IgE production in human B lymphocytes dosed in vitro (Takanaka et al., 1995; Tsien et al., 1997). They found that combined dosing with ragweed antigen and DPM caused IgE isotype switching (Fujieda et al., 1998). They found that pretreatment with topical fluticasone proprionate did not reduce the IgE or cytokine response to DPM, but did reduce the IgE and cytokine response to ragweed challenge (Diaz-Sanchez et al., 1999a). They instilled keyhole limpit hemocyanin (KLH), into na€ıve atopic subjects with or without pretreatment with DPM, and found that only the DPM-treated subjects developed KLH-specific IgE (Diaz-Sanchez et al., 1999b). This finding suggested that not only can DPM enhance responses in preallergic subjects, but it may also enhance development of allergies. Using a crossover study design, they found considerable individual consistency in the amplification by DPM of responses to ragweed challenge, and concluded that susceptibility to the adjuvant effect was an intrinsic trait (Bastain et al., 2003).

HEALTH EFFECTS

593

In addition to the work described above, both the UCLA group and other investigators have reported numerous studies of the cellular mechanisms by which DPM might alter allergic responses. Most of these have been in vitro studies using epithelial or endothelial cells, lymphocyte subpopulations (e.g., Mamessier et al., 2006), or basophils. Rather than attempting a complete review, suffice it to note that enough mechanisms have been implicated to support the plausibility that, at some dose, DPM can initiate, alter, and amplify steps in allergic response pathways. Because other combustion-derived particles have not received the same scrutiny, the extent to which DPM might be unique in this respect is not known. Effects of Laboratory-Based Exposures to Diesel Exhaust Human Studies Investigators at the University of Umea in Sweden have conducted experimental exposures of normal and asthmatic subjects to exhaust from diesel vehicles operated at steady state. The compositions of the exposure atmospheres have not been described in detail; most exposures were described solely by DPM mass concentration. Exposure of atopic asthmatics for 1 h at 300 mg DPM/m3 increased airway resistance, airway hyperresponsiveness, and proinflammatory cytokines (Nordenhall et al., 2001). Exposures of normal and asthmatic subjects for 2 h at 100 mg DPM/m3 caused similar increases in airway resistance, but exposure did not enhance the eosinophilic inflammation in asthmatics, and the neutrophilic inflammatory response was less in asthmatics than in normal (Stenfors et al., 2004). Exposure of normal subjects for 1 h at 300 mg DPM/m3 caused an increase in IL-13 in bronchial mucosa, which is consistent with promotion of a TH2 inflammatory response (Pourazar et al., 2004). The ability of exposure to exacerbate allergic responses to antigen was not tested directly. However, the induction of inflammatory responses and proallergic cytokines in normal airways and increased reactivity of asthmatic airways indicate that acute high-level exposures can aggravate asthma and may promote allergic responses. Indirect evidence for a relationship between allergic airway disorders and inhaled diesel exhaust comes from studies of lung function in children in the Netherlands attending schools with different proximity to busy roadways. Brunekreef et al. (1997) reported a relationship between proximity of schools to busy roadways and reduced lung function, and found a more significant correlation with truck than with automobile traffic counts. Subsequently, Janssen et al. (2003) observed similar relationships with chronic respiratory symptoms, but found that the effect was almost exclusively expressed in children with bronchial hyperresponsiveness and/or sensitization to common allergens. These results do not confirm that diesel exhaust amplifies allergic responses, but they suggest that the effects of truck (nearly exclusively diesel) exhaust on lung function are expressed most strongly in children with asthma or allergies. Animal Studies Inhalation exposures of rodents to high concentrations of diesel exhaust have been shown to alter sensitization to antigens in some, but not all studies. Takano et al. (1998) exposed ICR mice 12 h/day, 7 days/week for 40 weeks to exhaust from a 2.7-L automobile engine operated at constant speed and load at DPM concentrations of 300, 1000, or 3000 mg/m3 (NOx ¼ 3.5, 10.1, and 23.1 ppm) or to clean air. The mice were sensitized to ovalbumin by intraperitoneal injection, and at 3-week intervals during the last 24 weeks, they were also exposed briefly to aerosols of ovalbumin. Exhaust exposure caused a concentration-related increase in neutrophils, eosinophils, and interleukin-5

594

DIESEL EXHAUST

in bronchoalveolar lavage fluid, but only a very slight increase in serum antigen-specific IgE. Maejima et al. (2001) exposed female BDF1 mice 16 h/day, 5 days/week for 24 weeks to exhaust from a 6.9-L truck engine operated at constant speed and load at a DPM concentration of 3200 mg/m3 (NOx ¼ 30.4 ppm), diesel exhaust at the same dilution with DPM removed by filtration, Kanto loam dust at 3300 mg/m3, or clean air. On 2 days each week, Japanese cedar pollen was added to the exposure atmosphere at a concentration of 550,000 grains/m3. At the end of exposure, the group mean antigen-specific serum IgE levels were 4–5 fold higher in the diesel exhaust, filtered exhaust, and loam particle-exposed groups than in the pollen-only group. All three exposures significantly increased the percentages of mice having elevated titers. This finding indicated that either whole diesel exhaust or exhaust without DPM, and a mineral particle at the same concentration as DPM in whole exhaust, all amplified the sensitization of mice to pollen. A study by Watanabe and Ohsawa (2002) demonstrated the potential importance of exposures before and soon after birth in the later development of allergies. They exposed F344 rats 6 h/day for 19 days to exhaust from a 0.3-L single-cylinder engine operated at constant speed at a dilution yielding a DPM concentration of 1730 mg/m3 or to filtered exhaust at the same dilution. Exposures occurred over one of three periods: gestation day 7 to 3 days of age (in utero), 4–22 of age (postnatal), or 23–41 days of age (weanlings). At 49 days of age, all rats were sensitized to Japanese cedar pollen, and the resulting antigen-specific IgE titers were measured. The in utero and postnatal exposures increased IgE production to a similar degree, but the exposure after weaning did not. Filtered and unfiltered exhaust caused the same effect. No enhancement of allergic sensitization in mice was found by Barrett et al. (2002) using more environmentally relevant concentrations of exhaust. They exposed male BALB/c mice 6 h/day, 7 days/week for 8 weeks to exhaust from 5.9-L 2000 Cummins engines operated on U.S. certification fuel on the heavy-duty certification transient cycle at dilutions producing DPM concentrations of 30, 100, 300, and 1000 mg/m3 (NOx ¼ 2.2, 5.6, 16.9, and 49.3 ppm). The mice were also exposed at weeks 4 and 6 and after completion of exhaust exposure to aerosolized ovalbumin. Exposure to exhaust did not increase the sensitization of mice to ovalbumin, as measured by lung inflammation, proallergic cytokines, serum antigenspecific IgE, or airway responsiveness to methacholine. 16.4.2.2

Reduced Resistance to Respiratory Infection

Evidence from Humans There have been no direct studies of the impact of experimental exposures to diesel exhaust on the resistance of human subjects to respiratory infections. Present evidence is indirect, and derives from studies of relationships between air pollution or traffic and respiratory infections in children. In adults, robust associations between PM2.5 and hospital admissions for respiratory illnesses have been reported, including respiratory tract infections such as bronchitis and pneumonia obtained from Medicare claims data (Dominici et al., 2006). Kim et al. (2004) conducted a survey of approximately 1100 children in grades 3–5 in the San Francisco Bay area in 2001 whose schools were near a traffic corridor, and measured PM2.5, black carbon (as an index of combustion soot), and NO2 and calculated average exposure values. Adjusting for multiple other potential determinants of illness in children, such as a current smoker in the home and mold, there were small increases in the risk of bronchitis and asthma (2–4%

HEALTH EFFECTS

595

increase per inter-quartile range of each pollutant). In a birth cohort of nearly 3000 children in the Netherlands (Brauer et al., 2002), an inter-quartile increase in soot concentration was associated with an increased risk of ear, nose, or throat infection (OR ¼ 1.15; 95% CI ¼ 1.00–1.33) and doctor-diagnosed flu or colds (OR ¼ 1.09; 95% CI ¼ 0.98–1.21), findings that approached statistical significance. The addresses of 11,484 patients from the National Cystic Fibrosis registry 1999–2000 were linked to the EPA Aerometric Information Retrieval Database (Goss et al., 2004). The risk of an exacerbation was significantly related to increases in PM2.5 but not NO2, adjusting for age, sex weight, and airway colonization with Pseudomonas. However, in a study of 19,901 infants discharged from a hospital after an episode of bronchiolitis in California during 1995–2000, there was no association with ambient CO, NO2, or PM2.5 (Karr et al., 2006). Taken together, these epidemiologic studies indicate that it is plausible that diesel exhaust exposure contributes to an increased susceptibility of respiratory infections in adults and children. Evidence from Animals Studies of rodents exposed to diesel exhaust by inhalation or DPM by intratracheal instillation have yielded mixed evidence for reducing resistance to bacterial respiratory infection, but generally support the plausibility that at least heavy exposures can do so. Campbell et al. (1981) exposed mice 8 h/day, 7 days/week to diesel exhaust at a dilution containing 6000 mg DPM/m3 and followed by challenge with Streptococcus pyogenes. Exposures ranging from 2 h to 321 days increased mortality from infection. Hatch et al. (1985) found no effect on mortality of intratracheal instillations of 100 mg DPM followed by S. pyogenes infection. Yang et al. (2001) found that intratracheal instillation of 5000 mg DPM/kg body weight into Sprague–Dawley rats depressed killing of Listeria monocytogenes instilled 3 days later, and that carbon black did not have the same effect. Harrod et al. (2004) exposed C57BL/6 mice 6 h/day for either 7 days or 6 months to exhaust from 5.9-L 2000 Cummins engines operated on U.S. certification fuel on the heavy-duty certification transient cycle at dilutions producing DPM concentrations of 30, 100, 300, and 1000 mg/m3 (NOx ¼ 2.2, 5.6, 16.9, and 49.3 ppm), followed by intratracheal instillation of Pseudomonas aeruginosa. Although clearance of bacteria was slowed and infection-related histopathology was increased in a generally concentration-dependent manner at both exposure times, the effect was only significant for the 7-day exposure. Experimental exposures have also produced mixed results for effects on resistance to respiratory viral infection. Hahon et al. (1985; also described in Castranova et al., 2001) exposed CD-1 mice 7 h/day 5 days/week for 1, 3, or 6 months to exhaust at a DPM concentration of 2000 mg/m3 followed by intranasal instillation of Ao/PR/8/34 influenza virus. Although mortality was not affected by exposure, exposure for 3 or 6 months increased viral replication in the lung and lung histopathology and suppressed the production of interferon. Harrod et al. (2003) exposed C57BL/6 mice for 7 days to exhaust as described in the preceding paragraph, but only at the 30 and 1000 mg DPM/m3 levels, followed by intratracheal instillation of Respiratory Syncytial Virus (RSV). In that pilot study, viral clearance was reduced, and infection-related histopathology was increased at both exposure levels. In a follow-up study using the same exposure but all four exposure levels, Reed and Berger (2006b) found no effect on viral clearance or histopathology at any level. The reason that the effects in the pilot study were not reproduced later was not confirmed, but experience has shown that the assay produces variable results depending on the specific viral culture used and the viral titering methods (both differed between the studies).

596

DIESEL EXHAUST

Potential Mechanisms The mechanisms by which exposure to diesel exhaust might reduce resistance to respiratory infections are not well-defined, but two factors are suggested by previous work: suppression of macrophage function and suppression of systemic immune responses. Macrophage cytotoxicity has been associated with oxidative stress produced by DPM extracts (Hiura et al., 1999; Li et al., 2000), and it has been postulated that quinones might be an important chemical class (Li et al., 2000; Kumagai et al., 2002). Although allergic responses might be amplified by exposure to diesel exhaust as discussed above, protective immune responses might be suppressed. Evidence that diesel exhaust can suppress systemic immune responses is discussed in a following section. 16.4.2.3 Respiratory Tract Inflammation and Noncancer Respiratory Disease Studies of occupational groups having high exposure to diesel exhaust show acute effects on lung function from high exposures and higher incidences of noncancer lung disorders. Several studies of experimentally exposed humans have shown that acute exposures to concentrations within the high occupational range can induce inflammation, reduce lung function, and increase airway reactivity. Short- and long-term exposures of animals at high concentrations produce inflammatory responses. Chronic, extreme exposures of rats produce chronic active lung inflammation, an overwhelming of particle clearance, progressive particle sequestration, fibrosis, and epithelial hyperplasia, metaplasia, and neoplasia. Effects in Humans Experimental Exposures Battigelli (1965) exposed subjects for 1 h to exhaust from a single-cylinder diesel engine and detected no decrements of airflow resistance. Although the DPM concentrations were not given, the highest concentrations of 55 ppm CO, 4.2 ppm NO2, and 1 ppm SO2 suggest that the highest soot concentration was in the range of a few mg/m3. Ulfvarson et al. (1987) exposed subjects for 3.7 h to exhaust from a 3.7-L engine at 600 mg DPM/m3 and detected no decrement in pulmonary function. In a series of experiments conducted by Swedish investigators, Rudell et al. (1996, 1999 exposed healthy, never-smoking subjects to diesel exhaust from an idling truck for 1 h. PM mass concentration was not specified, but eye irritation, nasal irritation, and an unpleasant smell was reported during exposure. Although airway resistance increased, there was no significant difference in FVC, FEV1, or other flows, but there was an increase in neutrophils and decreased phagocytosis by macrophages in bronchoalveolar lavage fluid. In additional studies by the same Swedish group, healthy subjects were exposed to diesel exhaust at 300 mg DPM/m3 for 1 h while riding a stationary bicycle. As before, there was no change in spirometry, but there was an increase in neutrophils and B-lymphocytes in bronchoalveolar lavage fluid. There was also an increase in neutrophils; mast cells; CD4þ, and CD8þ T lymphocytes; with upregulation of vascular endothelial cell adhesion molecules and in bronchial biopsies and evidence of increased IL-8 gene transcription (Salvi et al., 1999, 2000). In peripheral blood sampled 6 h after exposure, there was a significant increase in neutrophils and platelets (Salvi et al., 1999), and also greater numbers of neutrophils in expectorated sputum and IL-6 compared to subjects exposed to air (Nordenhall et al., 2000). In a similar study in Britain, subjects were exposed to either air or diesel exhaust at 200 mg DPM/m3 for 2 h while at rest (Nightingale et al., 2000). There were no changes in spirometry, but there was an increase in expectorated sputum neutrophils and myeloperoxidase 4 h after exposure with no changes in peripheral blood neutrophils. More recently, the Swedish group

HEALTH EFFECTS

597

exposed volunteers to diesel exhaust at 100 mg DPM/m3 for 2 h, and at 18 h post-exposure bronchoalveolar lavage and bronchial biopsies were obtained (Behndig et al., 2006). Increased bronchial mucosa neutrophils, mast cells, and increased neutrophils, IL-8, and myeloperoxidase were noted in the lavage fluid. These findings indicate that exposure of healthy subjects to diesel exhaust at levels of 100 mg DPM/m3 can cause a pulmonary inflammatory response, influence macrophage function, and result in the expression of airway cytokines, without causing significant changes in ventilatory function measured using standard clinical tests. There may also be changes in vascular function and inflammatory changes in peripheral blood, suggesting a link to systemic effects occurring after diesel exhaust inhalation. Effects from a Single Workshift There are six published evaluations of the effects of diesel exhaust exposure on respiratory function during a single workshift. These studies are complicated by the numerous other materials inhaled during the workshift by the miners, garage workers, stevedores, and ferryboat crewmen, and not all studies included control groups not exposed to diesel exhaust. In addition, characterization of the personal exposures to diesel soot and other exhaust components varied considerably among the studies. All of these studies used spirometry to assess respiratory function. Ames et al. (1982) found that workshift changes in respiratory function did not differ between workers in mines where diesel engines were used or not used. Both J€ orgensen and Svensson (1970) and Gamble et al. (1978) reported small workshift decrements in forced expiratory volumes and flow rates among miners. Gamble et al. (1987) measured small workshift decrements in function, and found a stronger association with PM than with NO2. Ulfvarson et al. (1987) found small, but significant workshift decrements in function among stevedores on roll-on, roll-off ships exposed to DPM at 130–590 mg/m3, but not in bus garage or car ferry workers exposed to DPM at 100–460 mg/m3. Ulfvarson and Alexandersson (1990) detected small workshift decrements of function among stevedores exposed to PM at 120 mg/m3. Overall, these findings support the conclusion that reversible changes in respiratory function can occur in humans exposed occupationally, although it is not possible to relate these changes to a specific level of exposure. Long-Term Effects on Respiratory Function and Symptoms Evaluations of the effects of longer-term occupational exposures to diesel exhaust on respiratory function and symptoms have yielded mixed results. Most studies found that exposure was associated with small increases in respiratory symptoms, such as dyspnea, cough, and phlegm (J€orgensen and Svensson, 1970; Attfield et al., 1982; Reger et al., 1982; Gamble et al., 1983; Purdam et al., 1987), but some did not (Battigelli et al., 1964; Ames et al., 1984). There was no consistent effect on respiratory function, but it is possible that the lack of a clear relationship was due to including few workers with long-term exposures and by only including active workers in these studies. In a more recent study, respiratory symptoms in 20,898 farmers were assessed in relation to occupational exposures (Hoppin et al., 2004). Adjusting for age, smoking, asthma history, and atopy, driving diesel tractors was associated with an elevated OR for wheeze (1.31; 95% CI ¼ 1.13–1.52), and there was an increasing risk with increasing years of tractor driving. For farmers driving gasoline tractors, the overall risk was also significantly elevated but of lower magnitude (1.11; 95% CI ¼ 1.02–1.21). Although these results support an association between diesel exposure and respiratory symptoms, the interpretation is hampered by lack of actual exposure information. Jacobson et al. (1988) found in a cohort of 19,901 British coal miners investigated over a 5-year period

598

DIESEL EXHAUST

that there was an increased work absence due to self-reported chest illness in underground workers exposed to diesel exhaust, as compared to surface workers. This finding is consistent with the impairment of alveolar macrophage function or may be due to the effects of exposure on airway inflammation noted following acute exposure described in the previous sections. Traffic Studies There is a large body of emerging literature, primarily in children, relating proximity to traffic to chronic respiratory symptoms. It has not been possible to exclusively implicate diesel exhaust in the occurrence of symptoms. Venn and coworkers (2001) reported that the risk of wheeze among children living in Nottingham, England, within 150 meters of a main road increased 8% for every 30 meters closer in distance, with most of the risk within 90 meters of the road. In the Netherlands, a doubling of the risk of wheeze was reported in children living within 100 meters of a roadway (Van Vliet et al., 1997). For children living within 300 meters, measurements of truck traffic density (mostly diesel) but not automobile traffic density was significantly related to lower values of pulmonary function (Brunekreef et al., 1997). Self-reports of truck traffic volume were also significantly related to wheezing (Weiland et al., 1994; Duhme et al., 1996) and recurrent respiratory illness (Ciccone et al., 1998). An association between ambient NOx, whose major source is diesel vehicle emissions, and asthma prevalence among Taiwan middle-school students was also noted (Guo et al., 1999). The International Study of Asthma and Allergies in Childhood study in Munich Germany assessed nearby residential traffic counts for 7509 children (Nicolai et al., 2003). Current asthma, wheeze, and cough were significantly related to traffic counts. There are fewer traffic studies for adults. Oosterlee et al. (1996) found no association with chronic cough, chronic phlegm, and wheeze in adults living on busy streets compared to those living in neighborhoods with little traffic. In Tokyo, three cross-sectional studies were conducted in three separate groups of over 1500 women in 1979, 1982, and 1983, and conflicting results were obtained (Nitta et al., 1993). Garshick et al. (2003) studied male U.S. veterans drawn from the general population of southeastern Massachusetts. Information on respiratory symptoms and potential risk factors were collected by questionnaire, and residential addresses were related to distance from and traffic density of major roadways. Adjusting for cigarette smoking, age, and occupational exposure to dust, subjects living within 50 meters from a major roadway were more likely to report persistent wheeze (OR ¼ 1.31; 95% CI ¼ 1.00–1.71) compared to those >400 meters away. The risk was only observed for those living within 50 meters of heavily trafficked roads (10,000 vehicles/ 24 h): OR ¼ 1.71, 95% CI ¼ 1.22–2.40). The risk of chronic phlegm on heavily trafficked roads was of borderline significance (OR ¼ 1.40, 95% CI ¼ 0.97–2.02). These results suggest that residential exposure to vehicular emissions near busy roadways (to which diesel emissions contribute, but not exclusively) results in chronic respiratory disease symptoms in adults and children and may be associated with asthma. As evidence that a reduction in traffic results in an improvement in asthma morbidity, during the Atlanta Olympic games in 1996 when traffic was restricted, there were lower rates of childhood asthma hospitalizations, Medicaid claims, and emergency room utilization (Friedman et al., 2001). Nonmalignant Respiratory Disease Mortality There have been few studies by which the relationship between long-term occupational exposure to diesel exhaust and mortality from noncancer respiratory disease can be assessed (HEI, 1995). Boffetta et al. (1988) not only described lung cancer mortality in the American Cancer Society Cohort as described

HEALTH EFFECTS

599

earlier, but also examined mortality attributable to respiratory diseases. Adjusting for smoking and age, chronic obstructive pulmonary disease (COPD) was not significantly increased, but mortality attributable to pneumonia and influenza was significantly increased (p < 0.05; relative risk ¼ 1.97). Wong et al. (1985) assessed mortality in 34,156 operating engineer union members during 1964–1978, and a greater emphysema mortality risk was associated with a greater duration of union membership. In the 38-year mortality followup of U.S. railroad workers, Garshick et al. (2004) observed an elevated mortality risk from all respiratory system diseases, including COPD and allied conditions (relative risk ¼ 1.31; 95% CI ¼ 1.21–1.42), and from COPD and allied conditions alone (1.41; 95% CI ¼ 1.27–1.55. Because there was no direct information regarding smoking, the same authors conducted a case–control study using deaths collected during 1981–1982 with smoking histories from next of kin (Garshick et al., 1987; Hart et al., 2006). In that study there were 536 cases with COPD or allied conditions and 1525 controls with causes of death not related to diesel exhaust or fine particle exposure. After adjustment for age, race, smoking, U.S. Census region of death, and vitamin C use, engineers and conductors (occupations with diesel exposure from operating trains) had an increased risk of COPD mortality. Mortality increased with years of work in jobs with exposure, and for 16þ years of exposure after 1959, OR ¼ 1.61 (95% confidence interval ¼ 1.12–2.30). Overall, the risk for mortality from noncancer chronic respiratory disease associated with occupational exposure to diesel exhaust appears to be on the same order of magnitude as the risk for lung cancer, but the data supporting the association are fewer. Effects in Animals Repeated exposure of animals to diesel exhaust induces concentration-related effects on respiratory function and lung structure that have been reviewed previously (Mauderly, 1994a, 1996, 2000; HEI, 1995; EPA, 2002). Recent studies provide additional detail (e.g., Kato et al., 2000), but are consistent with earlier findings. Near lifetime repeated exposures of rats at concentrations of DPM over approximately 1000 mg/m3 overwhelms the ability of normal particle clearance pathways and results in a progressive accumulation of DPM in the lung. This accumulation is accompanied by persistent inflammation, focal epithelial proliferation and metaplasia, and fibrosis (Mauderly, 1996). The progressive structural changes are reflected by a progressive impairment of respiratory function that includes lung stiffening (loss of compliance), reduced lung volumes, uneven intrapulmonary gas distribution, and impaired alveolar-capillary gas exchange (Mauderly et al., 1988). This structure-function syndrome also occurs in rats exposed heavily to other solid, respirable particles (Mauderly, 1994a). Of importance for estimating hazard for humans, no significant alterations of particle clearance (Wolff et al., 1987), inflammation, fibrosis (Henderson et al., 1988) or respiratory function or structure (Mauderly et al., 1988) resulted from chronic exposures of rats at 350 mg DPM/m3, even though small amounts of DPM accumulated in the lungs. This dose-response information from rats was used by EPA to estimate a reference (safe) concentration for noncancer effects in humans of 5 mg DPM/m3 lifetime exposure (EPA, 2002), after adjustments for interspecies dosimetry and safety factors. A study by Kato et al. (2000) was consistent with the earlier findings in detecting no significant tissue effects of exposure of rats for 24 months at 210 mg DPM/m3. Under exposure conditions producing the above effects in rats, mice accumulate similar amounts of DPM in their lungs (Henderson et al., 1988), but the inflammatory, fibrotic (Henderson et al., 1988), and histopathological (Mauderly et al., 1996) responses are less than those in rats. Small reductions in lung volumes and compliance have also been observed

600

DIESEL EXHAUST

in diesel-exposed Syrian (Heinrich et al., 1986) and Chinese (Vinegar et al., 1981) hamsters. Mice and hamsters have also been shown to have lesser functional and structural responses than rats to other solid respirable particles (Mauderly, 1994b). A smaller, but perhaps more relevant, body of information comes from nonrodent species that have respiratory bronchioles (absent in rodents) and other lung features more similar to humans. Only two such species have been chronically exposed to diesel exhaust. The EPA exposed male cats chronically to exhaust at 6000 mg DPM/m3 for 61 weeks, and then at 12,000 mg DPM/m3 for the remainder of 27 months, followed by a 6-month recovery period (Pepelko and Peirano, 1983). A restrictive functional impairment with decreased lung volumes and uneven intrapulmonary gas distribution was observed at the end of the exposure (Moorman et al., 1985). Histopathology at the end of exposure included peribronchiolar fibrosis and epithelial metaplasia in terminal and respiratory bronchioles (Plopper et al., 1983). Interestingly, while the epithelial changes lessened during the 6-month recovery period, the fibrosis progressed. Lewis et al. (1989) exposed cynomolgus monkeys to diesel exhaust at 2000 mg DPM/m3 and reported that the forced expiratory flow rates were reduced at the end of exposure (Lewis et al., 1986). The lung histopathology of the monkeys differed from that of rats exposed concurrently (Nikula et al., 1997). Soot was present in approximately the same tissue concentration in both species, but was located predominantly in interstitial compartments in monkeys and in alveolar lumens in rats. The species had similar increases in pulmonary macrophages. The most striking difference was in the degree of epithelial proliferation, which was characteristically prevalent near accumulations of soot in rats but essentially absent in monkeys. Although the data base is small, these results suggest that nonrodent species can develop fibrosis and epithelial responses under extreme exposure conditions but exhibit little structural response from chronic exposures at 2000 mg DPM/m3. 16.4.2.4 Cardiovascular Effects There has been increasing attention to the potential cardiovascular effects of inhaled diesel exhaust, concurrent with increasing evidence for the cardiovascular effects of ambient PM (EPA, 2004). Similar to the effects described for other organ systems, it is now clear that exposures to high concentrations of exhaust have potential for altering heart and vascular function in both humans and animals, but the hazards and risks from typical environmental exposures are not yet known. The plausibility of an effect from environmental exposures to DPM is supported by evidence for the effects of ambient PM2.5, to which DPM contribute. In 1995, Seaton and coworkers suggested that lung inflammation caused by inhaled ultrafine PM in might provoke myocardial infarction (MI) as a result of mediator release (Seaton et al., 1995). In recent years it has been recognized that atherosclerosis and coronary artery disease are chronic inflammatory diseases and that blood markers of systemic inflammation are related to clinical cardiovascular outcomes, including MI (Ross, 1999; Pearson et al., 2003). Elevated blood levels of cardiovascular inflammatory markers such as C-reactive protein are associated with increases in particulate air pollution (Schwartz, 2001; Peters et al., 2001; van Eeden et al., 2001). Laboratory studies also suggest a link between PM2.5 and vascular changes, including atherosclerosis. Suwa et al. (2002) instilled urban PM from Ottawa into the nasopharynx of rabbits genetically prone to atherosclerosis and observed a progression of atherosclerotic lesions accompanied by an increase in circulating neutrophils. Sun et al. (2005) observed an enhancement of atherosclerotic changes in the aortas of ApoE / mice exposed to concentrated Northeastern regional PM2.5 6 h/day, 5 days/week for 6 months.

HEALTH EFFECTS

601

Other studies of humans and animals have shown associations between ambient PM and heart rate control, arrhythmias, and abnormalities of cardiac depolarization as manifested by ST segment depression. Although none of the ambient PM results are specific to DPM, DPM must be assumed to contribute some portion in lieu of better information on the PM species causing the effects. Studies in Humans Information from humans derives from several study designs. There is indirect evidence from occupational and population-based epidemiologic studies linking adverse cardiovascular outcomes to traffic emissions, evidence from panel studies where smaller groups were monitored and physiologic observations were recorded and related to environmental exposures, and direct evidence from laboratory studies involving experimental exposures. Experimental and Panel Studies of Traffic, Ambient PM, and Black Carbon There is evidence that inhaled ultrafine carbon particles, similar to the EC core of diesel soot, can pass through the lung into the systemic circulation in humans. Nemmar et al. (2002) administered 99m Technetium-labeled carbon particles via inhalation to five healthy volunteers. There was a rapid increase in radioactivity over the liver, and radioactivity counts increased in blood samples and urine. Pekkanen et al. (2002) assessed ST segment depression during exercise in 342 tests of 45 subjects with coronary disease in Helsinki during the winters of 1998–1999. Air pollution was monitored at a central site, and levels of particulate air pollution 2 days before the test were associated with an increased risk of ST segment depression. The central site recorded the number of ultrafine particles (10–100 nm), number of particles 100–1000 nm (accumulation mode), and particle mass as PM2.5. Each PM measure was associated with an increased risk of ST-segment depression, but the effects of ultrafine PM and PM2.5 were independent, suggesting separate sources. In a subsequent analysis of the same data, filter absorbance was assessed as a measure of EC or black carbon (Lanki et al., 2006). Filter samples were also analyzed for elemental composition using X-ray fluorescence spectrometry and principal component analysis to apportion PM mass to sources. Traffic- and diesel-related PM as indicated by the extent of filter absorbance were associated with the greatest risk of ST depression (Lanki et al., 2006). Gold et al. (2005) studied ST segment changes in 24 active Boston residents 61–88 years of age, each monitored up to 12 times from June through September 1999. Black carbon level in the previous 12 h and the level 5 h before testing predicted ST-segment depression. These results suggest an adverse effect of traffic-related emissions, including diesel exhaust, on people likely to be at risk for cardiac ischemic events. Peters and coworkers (2004) in Augsburg, Southern Germany, found a significant association between exposure to traffic and the onset of a MI 1 h later (OR ¼ 2.92; 95% CI ¼ 2.22–3.83). Because Europe has large numbers of light-duty diesel vehicles, it was speculated that diesel exhaust significantly contributed to traffic emissions in that study. In an additional study in Erfurt, Germany, ambient PM was measured continuously during the winter of 2000–2001 (Ruckerl et al., 2006). Ultrafine and accumulation mode PM were both significantly related to increased levels of C-reactive protein (CRP) with a 2-day lag in 57 male patients with coronary heart disease. Associations between CRP and EC were weaker, but EC was strongly associated with increased levels of intercellular adhesion molecule-1, an indicator of vascular endothelial cell activation. In a study of nine North Carolina State troopers observed over 4 days, PM2.5 measured inside the patrol cars was associated with

602

DIESEL EXHAUST

CRP measured post-shift, further suggesting an association between traffic-related exposures and biomarkers of cardiovascular disease (Riediker et al., 2004). There have also been a series of studies assessing the relationship between ambient air pollution and ventricular and supraventricular ectopy. The results indicate positive associations between ectopy and air pollution, but results based on specific pollutants have varied, and associations with black carbon as an index of traffic or diesel exposure have been inconsistent. Sarnat et al. (2006) assessed 32 nonsmoking older adults on a weekly basis for 24 weeks during the summer and autumn of 2000 in Steubenville, Ohio, using a standardized 30-min protocol that included continuous electrocardiogram measurements. There were associations between supraventricular ectopy and PM2.5, sulfate, and ozone concentrations but not with EC, and there were no associations between any pollutant and ventricular ectopy. In a study of 56 people with implantable defibrillators in St Louis (Rich et al., 2006a), a significant increase in risk of ventricular arrhythmias was associated with SO2 and an insignificantly increased risk was associated with increases in NO2 and EC in the 24 h before the arrhythmia. In a similar study in Boston, Rich et al. (2006b) found a statistically significant positive association between episodes of paroxysmal atrial fibrillation and increased ozone concentration in the hour before the arrhythmia, and positive but not statistically significant risks were associated with PM2.5, NO2, and black carbon. In another study in Boston (Rich et al., 2005), PM2.5 and ozone were associated increased risks of ventricular arrhythmia, and Dockery et al. (2005) found an increased risk of ventricular arrhythmia associated with a 2-day mean exposure for PM2.5, CO, NO2, and black carbon that were not statistically significant. However, statistically significant associations were noted in people who had an arrhythmia 3 days earlier. Peters et al. (2000) found associations with defibrillator discharges in Boston between ventricular arrhythmias and NO2. In a study assessing heart rate variability, a measure of autonomic heart rate regulation in elderly Boston residents, Schwartz et al. (2005) found the strongest associations between decreased variability and black carbon. Taken together, these results indicate an association between arrhythmias with air pollution that is suggestive of traffic-related sources. Laboratory Exposures to Diesel Emissions Mills et al. (2005) exposed 30 healthy men to diesel exhaust at 300 mg DPM/m3 or clean air while riding a bicycle ergometer for 1 h in a double-blind manner. Exhaust was generated by an idling 1991 model off-road engine. Forearm blood flow was monitored, and the endothelium-dependent vasodilators bradykinin and acetylcholine, and the endothelium-independent vasodilator nitropusside were administered 2–4 h after exposure. The vasodilators caused an increase in forearm blood flow that was blunted following exposure, suggesting that diesel exhaust inhibited regulation of vascular tone. Exposure also suppressed the increase in plasma tissue plasminogen activator following infusion of bradykinin. Because tissue plasminogen activator is an endogenous fibrinolytic agent, it was postulated that impaired release could impair the ability of diesel-exposed people to dissolve endogenous intravascular thrombi. Impaired endogenous fibrinolysis and impaired vascular tone could contribute to a greater risk for cardiovascular events. The group also repeated measurements of forearm blood flow and plasma markers at 24 h after exposure (T€ ornqvist et al., 2007). At that time, endotheliumdependent vasodilation was still reduced, but endothelium-independent vasodilation was not. Increases in plasma cytokines indicated that mild systemic inflammation persisted to 24 h.

HEALTH EFFECTS

603

Mills et al. (2007) followed the above studies with an evaluation of effects on men with prior myocardial infarction. Twenty men with stable coronary artery disease underwent the same exposure protocol used in the preceding studies. Myocardial ischemia was evaluated by analysis of the ST segment of the electrocardiogram during exposure; plasma markers and responses to vasodilators were measured at 6 h after exposure, using the previous methods. A greater exercise-induced ST segment depression after diesel exposure than after sham exposure indicated a proischemic effect. The responses to the vasodilators were not altered by exposure, in contrast to the above findings with healthy men. The release of endothelial plasminogen activator was reduced by exposure. Together, the results from this group indicated that exposure to a high concentration of diesel emissions can cause acute cardiovascular changes, some of which may persist for 24 h. Occupational Studies There have been a number of studies examining cardiac outcomes in people exposed occupationally to diesel exhaust and other traffic-related pollutants. Taken as a whole, these studies indicate that professional drivers are at greater risk for cardiovascular disease than other occupations. The magnitude of the contribution of traffic emissions is uncertain because these studies have lacked exposure measurements. Tuchsen and Endahl (1999) described an excess of ischemic heart disease in Danish bus drivers. Employed men in Denmark aged 20–59 years old in 1981, 1986, and 1991 were classified according to occupation, and followed from 1981 to 1985, 1986 to 1990, and 1991 to 1993. All hospitalizations were identified using national records and ischemic heart disease rates in bus drivers were compared to rates for all employed men. In 1981–1985, the ischemic heart disease admission risk for bus drivers was significantly increased (standardized hospital admission ratio ¼ 1.41; 95% CI ¼ 1.20–1.65), and the risk increased throughout the study period. Hannerz and Tuchsen (2001) subsequently examined hospital admission rates for all Danish male professional drivers through 1997 and reported elevated ischemic heart disease rates for both taxi and bus drivers. Bigert et al. (2004) also conducted a study examining trends in the incidence of MI among professional drivers (1183 cases and 6072 controls) in Stockholm County, Sweden, during 1977–1996. They used registers of hospital discharges and deaths, and controls selected randomly from the general population (20,364 cases and 136,342 controls) that were linked to national occupational register job titles. Between 1977 and 1984, the risk among bus, taxi, and truck drivers was increased compared to manual workers, but risks subsequently declined among all three through 1996. Finkelstein et al. (2004) assessed mortality due to ischemic heart disease (1416 cases, ICD9 codes 410–414) in seven Ontario construction unions, and compared deaths among heavy equipment operators to deaths in other occupations. The construction workers had an increased risk of ischemic heart disease mortality (OR ¼ 1.32; 95% CI ¼ 1.13–1.55). Although there was no information regarding smoking or other risk factors in these studies, the comparison groups were other workers, making it unlikely that cardiovascular risk factors substantially differed among case and referent groups. The majority of other epidemiologic studies describing an association between cardiovascular disease and professional drivers were conducted in Sweden. Edling et al. (1987) followed 694 men in five Swedish bus companies during 1951–1983. Standardized mortality ratios were computed for clerks, bus drivers, and bus garage workers, and no increase in mortality was noted for cardiovascular diseases; however, there were relatively few deaths.

604

DIESEL EXHAUST

In contrast, Gustavsson and coworkers (1996) studied cases of MI in professional drivers aged 30–74 in Stockholm County (4105 cases and 13,066 referents) during 1976–1984. Incident cases of first MI were identified using a central register. Referents were selected from the general population and matched based on age, year of admission, and death. National census data on occupation and industry was used to identify long-term drivers. The relative risk of MI was 1.53 (95% CI ¼ 1.15–2.05) for bus drivers (70 cases), 1.65 (95% CI ¼ 1.30–2.11) for taxi drivers (101 cases), and 1.31 (95% CI ¼ 1.05–1.64) for longdistance truck drivers (112 cases). Gustavsson et al. (2001) addressed specific occupational and personal risk factors for MI as part of the Stockholm Heart Epidemiology Program during 1992–1993 for men and during 1992–1994 for women in living in Stockholm County, and free of previous MI (1335 cases and 1658 controls). Cases surviving at least 28 days post infarction were identified using hospital data, and referents were identified through a computerized population register and matched on age, year of enrollment, and hospital service area. Participants completed mail questionnaires to obtain lifetime occupational histories and other personal life-style factors, with followup by telephone. Exposures were assessed using a job-exposure matrix designed to assess the intensity of exposure to motor exhaust using estimated CO levels found in various occupational categories. After adjusting for smoking, alcohol, body mass index, diabetes, hypertension, and physical activity, there was an increase in risk of MI for participants having medium exhaust cumulative exposure levels (154 cases and 136 controls; RR ¼ 1.32; 95% CI ¼ 1.01–1.73) and with high cumulative exhaust exposures (155 cases and 137 controls; RR ¼ 1.21; 95% CI ¼ 0.91–1.59). These data were extended by Bigert et al. (2003) who assessed MI risks among male bus, truck, and taxi drivers who had their first MI in 1992 or 1993. There were 1067 cases and 1482 controls, including 146 cases and 129 controls who had ever worked in a driver category. Adjusting for smoking, alcohol, physical activity during leisure time, body mass index (BMI), diabetes, hypertension, socioeconomic status, and an index of job stress, the MI risk among truck drivers (45 cases and 31 controls) was not elevated (OR ¼ 1.07; 95% CI ¼ 0.77–1.50). For bus drivers (45 cases and 31 controls; OR ¼ 1.46; 95% CI ¼ 0.89– 2.41) and taxi drivers (94 cases and 84 controls; OR ¼ 1.32; 95% CI ¼ 0.81–1.50) risk estimates were elevated (although not significantly so) with suggestion of an increased risk with greater years of work. Epidemiological Studies of Ambient PM There are associations between PM2.5 and short-term hospital admissions for cerebrovascular disease, peripheral vascular disease, ischemic heart disease, heart failure, and heart rhythm disturbances using Medicare claims data (Dominici et al., 2006). Pope et al. (2004) assessed cardiovascular mortality over a 16year period in the American Cancer Society cohort, and linked residential address to PM2.5 data for U.S. metropolitan areas for over 300,000 people. Long-term PM exposures were significantly associated with mortality attributable to ischemic heart disease, dysrhythmias, heart failure, and cardiac arrest. A followup of the Harvard Six City study assessed mortality during 1974–1998, and also found an association between cardiovascular mortality and average PM2.5 (Laden et al., 2006a). Laden et al. (2000) also examined daily mortality in these cities during 1979–1988 and found it to be significantly related to mobile source emissions using particle lead as a marker of exposure to gasoline vehicle emissions. Kunzli et al. (2005) used baseline health data from two studies assessing carotid intima-media thickness in 798 people in the Los Angeles area. Residential addresses were geocoded

HEALTH EFFECTS

605

to assign annual mean PM2.5 values using a regional exposure model. There was a significant relationship between carotid intima-media thickness and PM2.5, adjusting for age, sex, smoking, blood pressure, and treatment for hypertension or with lipid lowering agents, thereby supporting an association between a direct measure of atherosclerosis and PM2.5. Although not specifically implicating diesel exhaust, these studies support a relationship between cardiovascular disease and PM2.5 to which diesels contribute. Other air pollution studies have included exposure markers that more specifically assessed a potential contribution from diesel exhaust exposure. London daily air pollution data were analyzed in relation to daily deaths during 1992–1994 (Bremner et al., 1999). Black smoke with a 1-day lag was a significant predictor of all cardiovascular mortality. Le Tertre et al. (2002) studied the relationship between airborne particles and hospital admissions for cardiac causes in eight European cities. These data included black smoke for in five cities (Barcelona, 1994–1996; Birmimgham, 1992–1995; London, 1992–1994; Netherlands, 1989–1995; Paris, 1992–1996). Black smoke with 0- and 1-day lags was associated with a 1.1% increase in cardiac deaths (95% CI ¼ 0.4–1.8) that was independent of total PM exposure. Hoek et al. (2002) studied a sample of 5000 people from the Netherlands Cohort Study on Diet and Cancer during 1986–1994. Cardiopulmonary mortality was associated with living near a major road (RR ¼ 1.95; 95% CI ¼ 1.09–3.52), defined as 100 m of a freeway or within 50 m of a major urban road, adjusting for smoking, education, BMI, and diet. Nafstad et al. (2004) found a significant association between estimated residential NO2 levels and ischemic heart disease in a population-based prospective study in Oslo, Norway, adjusting for smoking and other potential confounding factors. Rosenlund et al. (2006) conducted analyses using data from the Stockholm Heart Epidemiology Program. Addresses were geocoded and emission databases used to obtain estimates of annual mean levels of NO2, CO, PM10, PM2.5, and SO2. There was no overall association between long-term air pollution exposures and overall MI incidence, including NO2 exposures as an index of traffic-related exposures. Results were adjusted for age, catchment area, smoking, physical activity, diabetes, BMI, and occupational exposures. However, there was a suggestion of an increase in fatal cases attributable to NO2, (272 cases; OR ¼ 1.51; 95% CI ¼ 0.96–2.37), and risk was particularly elevated in those who died out of the hospital (84 cases; OR ¼ 2.17; 95% CI ¼ 1.05–4.51). Studies in Animals and In Vitro Systems Ikeda et al. (1995) found that incubating sections of rat aortas with suspensions of 1–100 mg DPM/mL in saline reduced acetylcholine-induced relaxation and that superoxide dismutase partially abolished the effect. They interpreted the results as suggesting that the DPM inhibited the production or effect of endothelial nitric oxide. Toda et al. (2001) injected 12,000 or 120,000 mg DPM/kg intravenously into rats and monitored blood pressure. The higher dose caused a transient drop in blood pressure that was blocked by atropine but not propanol. These results suggest that the small amounts of DPM that might circulate in the blood at any given time would not be expected to affect blood pressure in rats. Campen et al. (2003) found that repeated exposures altered the electrocardiograms of genetically hypertensive rats. Rats were monitored by telemetry before, during, and after exposure to diesel exhaust at 30, 100, 300, and 1000 mg DPM/m3 for 6 h/day for 7 days (exposure described in detail in McDonald et al., 2004a, and Figs. 16.1, 16.2, 16.3, and

606

DIESEL EXHAUST

16.4). The mean heart rate was increased during the exposure period, largely because the heart rate of exposed rats did not tend to decline during the study as in controls and because the normal diurnal reduction during daytime was less in exposed rats. The effect was clear at even the lowest exposure level in males, but not females. The PQ interval was increased slightly by exposure, but significantly at the highest levels. This effect was also greater in males. Campen et al. (2005) demonstrated that the non-PM components of diesel exhaust had effects on the electrocardiogram by exposing C57BL/6 mice or atherosclerosis-prone ApoE / mice (pre-fed high fat diet) by inhalation. The mice were exposed 6 h/day for 3 days to whole exhaust from a single-cylinder engine operated under steady state load on certification fuel (McDonald et al., 2004c) at dilutions containing 500 or 3600 mg DPM/m3 or to filtered exhaust at the same dilutions. Both levels of whole and filtered exhausts caused identical concentration-related bradycardia in ApoE mice. Significant reductions of the T-wave area were caused by both whole and filtered exhaust at the high, but not low, concentration in ApoE mice, but not in C576BL/6 mice. They also perfused isolated coronary arteries from ApoE mice in vitro with saline through which the exhaust had been bubbled and found increased flow resistance and decreased response to vasodilator. Although the absence of particles in the perfusate was not confirmed (it had been filtered at 5 mm), they hypothesized that the effect was caused by the substantial content of organic compounds determined to be present. 16.4.2.5 Depression of Systemic Immune Responses There are indications from animal studies that diesel exhaust, like other combustion emissions, may suppress systemic immune function at high levels of exposure. The strongest evidence suggests suppression of T lymphocyte-mediated immunity. It is not known if environmentally relevant exposures cause this effect or which component of exhaust is responsible. The effect is not unique to diesel exhaust; it has also been demonstrated with inhaled tobacco smoke (Kalra et al., 2000) and wood smoke (Burchiel et al., 2005), all of which preferentially impact T lymphocyte function. Yang et al. (2003) instilled NIST 1650 DPM intratracheally into female B6C3F1 mice three times every 2 weeks for a 2-week or 4-week period at doses ranging from 50 to 15,000 mg/kg body weight, and examined effects in splenic lymphocytes. In mice injected with sheep red blood cells (SRBC), the number of anti-SRBC antibody-forming cells was decreased by 2 weeks of exposure in a dose-related manner. The effect was less after 4 weeks of exposure, and only significant at the highest exposure levels. The numbers of CD4þ and CD8þ T lymphocytes were reduced in the spleen by exposure, but the numbers of B lymphocytes were not reduced. The proliferative response of T lymphocytes to mitogenic stimulus was also reduced. Overall, the lowest dose causing a significant effect was 1000 mg/kg. Burchiel et al. (2004) exposed A/J mice 6 h/day, 7 days/week for 6 months to exhaust from 5.9-L 2000 Cummins engines operated on U.S. certification fuel on the heavy-duty certification transient cycle at dilutions producing DPM concentrations of 30, 100, 300, and 1000 mg/m3, followed by evaluation of effects on splenic T and B lymphocyte responses to mitogens. They found the proliferative response of T lymphocytes to be reduced similarly at all exposure levels. The response of B lymphocytes was actually increased at the lowest exposure level but not affected at other levels. The finding of suppressed T lymphocyte mitogenic response at 30 mg/m3 suggests that the potential for environmental exposures to combustion emissions to alter systemic immune function is worthy of further exploration.

HEALTH EFFECTS

607

16.4.2.6 Altered Development and Growth There are indications, from both epidemiology and animal studies, that exposure to diesel exhaust might affect organ development and growth. Epidemiology provides indirect evidence that exposure to components of diesel exhaust has the potential to alter organ development, birth weight, and growth, by linking those effects to ambient pollutants that are contributed in part by diesels. The effects of exhaust have been examined directly in animal studies, but only one study has included exposure levels within the environmental “hotspot” range. Epidemiology Gauderman et al. (2000) studied 3035 children in 12 communities in southern California in the fourth, seventh, or tenth grade who had at least two spirometric evaluations over 4 years in the 1990s. In the fourth grade cohort, deficits in lung function growth were related to PM and NO2. The estimated growth rate deficit for children in the most polluted compared to the least polluted city was a reduction of 3.4% in FEV1. They obtained similar results in a follow-up study of a second cohort of 1678 children who were enrolled in the fourth grade in 1996 (Gauderman et al., 2002). There are several birth cohorts of children under followup for evaluation of the effects of prenatal exposure to urban pollutants. In New York City, inner-city mothers were enrolled during pregnancy and the children underwent prenatal assessment for exposure to PAH (Perera et al., 2006). Mothers wore a personal monitor to collect vapors and PM2.5, and umbilical cord blood was collected. In a cohort of 183 children, prenatal PAH exposure was significantly associated with a lower mental development index at age 3 and cognitive developmental delay (Perera et al., 2006). Results in 303 children from the same study indicated a combined effect of prenatal PAH exposure and postnatal exposures to environmental tobacco smoke (ETS) on cough and wheeze at age 12 months and respiratory symptoms and asthma at 24 months (Miller et al., 2004). In offspring of mothers exposed to PAHs near the World Trade Center as indicated by cord blood adducts, in combination with in utero exposure to ETS, there was a 3% reduction in head circumference and a reduction of birth weight of 8% (Perera et al., 2005). In Krakow, Poland, DNA–PAH adducts were measured in cord blood as a marker of PAH exposure, and in 333 children adduct levels were significantly related to cough, sore throat, and wheeze during the first year of life (Jedrychowski et al., 2005). In 1397 children in the Czech Republic, prenatal exposure to PM2.5 and PAHs were related to altered lymphocyte subtypes at birth in the children (HertzPicciotto et al., 2005). Ambient PM has been linked to low birth weight in other studies. In Nova Scotia, regional levels of PM10, and SO2 were linked to home address and found to be associated with lower birth weight in 74,284 births during 1988–2000 (Dugandzic et al., 2006). Other associations have been observed between PM and preterm birth (Hansen et al., 2006) and increased postnatal mortality in Southern California (Woodruff et al., 2006). None of these studies has specifically examined diesel exhaust exposure, but diesels undoubtedly contributed to the exposures. Studies in Animals Studies in animals of the effects of inhaled diesel exhaust on organ development and growth have produced mixed results. Although results to date suggest the plausibility of developmental effects, few conclusions can be drawn from the scanty research done to date. Mauderly et al. (1987b) exposed F344 rats 6 h/day 5 days/week to exhaust from 1980 5.7-L Oldsmobile engines operated on U.S. certification fuel on an urban transient duty cycle, diluted to a DPM concentration of 3500 mg/m3 (NOx ¼ 5 ppm). One group was exposed beginning with conception until 6 months of age, and a parallel group was exposed

608

DIESEL EXHAUST

as adults between 6 and 12 months of age. Measurements after exposure included bronchoalveolar lavage indicators of inflammation and cytotoxicity, immune responses SRBC in pulmonary lymph node cells, respiratory function, clearance of inhaled tracer particles, lung tissue collagen, lung histopathology, and lung burdens of DPM at the end of exposure and after a 6-month recovery period. The percentage of females bearing litters was higher in the exposed group than in controls, and the number of live pups per litter was similar between the groups. Most effects of exposure were identical in the two groups. Rats exposed as adults had increased lung weight, retarded clearance of tracer particles, and considerable intra-alveolar aggregation of retained DPM. Lung weight was not increased, and tracer particle clearance was not slowed in rats exposed during development, and retained DPM was more scattered in the lung. Moreover, retained DPM was cleared more rapidly from the lung during recovery in the younger group. These results did not suggest that exposure to a high concentration of exhaust during breeding and gestation affected fertility or that exposure during lung development had greater impact than exposure during adulthood; indeed, differences tended in the opposite direction. Researchers at the Tokyo Metropolitan Institute of Public Health have conducted studies of the effects of in utero exposures to high concentrations of diesel exhaust on development of nonrespiratory organs of rats. Initial studies revealed that exposures could alter bone mass balance in growing rats (Watanabe and Nakamura, 1996; Watanabe, 1998). Attention then turned to reproductive organs. Watanabe and Oonuki (1999) exposed F344 rats 6 h/day 5 days/week for 3 months beginning at birth to whole exhaust from a 0.3-L diesel engine operated at constant speed and at a dilution containing 5630 mg DPM/m3 and 12.2 ppm NOx. A parallel group was exposed to filtered exhaust. Exposure to both whole and filtered exhaust increased serum testosterone and estradiol and decreased follicle-stimulating hormone and spermatogenesis. The only difference in responses to whole and filtered exhaust was a greater depression of lutenizing hormone by whole exhaust. Watanabe and Kurita (2001) then exposed pregnant F344 rats 6 h/day between days 7 and 20 of gestation using the same whole and filtered exhaust generation system and concentrations, and examined effects in maternal hormone levels and organ development in the near-term fetuses. Exposure to both whole and filtered exhaust increased the anogenital difference in both male and female fetuses (an indicator of “masculinization”) and retarded development of the testis, ovary, and thymus. Maternal testosterone and progesterone levels were increased by whole exhaust and decreased by filtered exhaust. Watanabe (2005) then examined effects of in utero exposures on the numbers of sperm and Sertoli cells in mature rats, using the same exhaust generation system. Pregnant F344 rats were exposed 6 h/day to whole exhaust at DPM concentrations of 1710 or 170 mg/m3 or to filtered exhaust at the same dilutions, and the testis and epididymis of the male offspring were examined at 96 days of age. Testis and epididymis weights were unaffected by exposure, but the sperm, spermatid, and Sertoli cell counts were reduced similarly by both whole and filtered exhaust at both exposure levels. This is the lowest concentration reported to cause significant effects to date and suggests that repeated exposures during pregnancy to diesel exhaust at high occupational or environmental “hot spot” concentrations might affect spermatogenesis in offspring of reproductive age. Although not a study of development, Tsukue et al. (2001) also found effects of high exposures on male reproductive organs of rats. Exposure of adult (13-month-old) F344 rats for 8 months to whole exhaust at dilutions containing 300, 1000, or 3000 mg DPM/m3 did not alter testicular weight, but even the lowest concentration reduced prostate and coagulating gland weights. The highest level increased prostate, coagulating gland, and seminal vesicle

CURRENT ISSUES

609

weights. Changes in serum hormone levels were also found. These findings support the hypothesis that occupational levels of exposure might affect male reproductive function.

16.5 CURRENT ISSUES It could be speculated in view of 60 years of continuous research that studies of the health impacts of diesel emissions could be brought to a close. A review 5 years ago (Mauderly, 2001) pointed to several issues remaining in question, and these have only been partially resolved. Moreover, recent results have raised health issues receiving little previous attention, and evolving emission standards, fuel, engine, and after-treatment technologies, and air pollutant causality issues raise additional questions. It appears that the era of diesel exhaust research is not yet completed. Four issues requiring continued study are raised by the information presented in preceding sections. The results of the new epidemiological studies of miners and truck drivers need to be completed, analyzed, and reported in detail. The incorporation of better estimates of exposure than possible in earlier studies may provide a greater confidence in estimates of unit risks for lung cancer and other health outcomes, especially if the risks derived from the two populations are similar. The cardiovascular effects of diesel exhaust require further exploration. The range of potential effects has not been explored, and better information is needed on the levels of exposure required to cause the effects. The effects of exposure on organ development and reproductive function warrant further scrutiny; again to explore the range of outcomes and whether exposure-response relationships extend down into plausible occupational and environmental exposure ranges. Finally, the relative effects of emissions from different combustion sources are still an open issue. Diesel exhaust has received attention second only to cigarette smoke, but the scanty comparisons to date suggest that diesels present few, if any, hazards that are not also presented by exhaust from gasoline or compressed natural gas engines, wood smoke, and other combustion sources. In addition to the preceding continuing concerns, the following three issues are among those begging further study. 16.5.1

Causal Components

The components of the complex mixture that cause the diverse health effects hypothesized or demonstrated to be caused by diesel exhaust remain an important issue. Determining whether or not the causal components are eliminated is of increasing importance as fuel, engine, and after-treatment technologies evolve to meet emission standards for DPM mass and NOx. For example, it has often been assumed (and sometimes explicitly stated without demonstration) that DPM caused the effects of whole exhaust, yet a growing body of knowledge indicates that this is often not true. The contributions of some subclasses of emissions are largely untested, such as the SVOCs or nonelemental carbon-based “nanoparticle” condensates. Parsing the effects among the different components is difficult and requires direct comparisons for confirmation. Demonstrating that a component can cause an effect at some dose is a useful starting point, but comparisons to whole exhaust or other components are necessary to determine whether that component predominates the effect of whole exhaust. Current knowledge of causal components is largely limited to information from studies comparing the effects of whole and filtered exhaust.

610

DIESEL EXHAUST

The earliest studies of filtered versus unfiltered diesel exhaust focused on carcinogenicity in animals chronically exposed at extreme concentrations. Both Heinrich et al. (1986) and Brightwell et al. (1989) found that removing DPM eliminated the lung tumor response of rats and markedly reduced noncancer inflammation, fibrosis, and epithelial alterations as well. These findings are consistent with the understanding that developed later that the rat lung carcinogenicity was a nonspecific response to overloading with retained particles. More recent studies have addressed noncancer effects in humans and animals and have clearly shown that the nonparticle components also present health hazards. Rudell et al. (1999) found that adding a ceramic particle trap only reduced lung inflammatory responses of humans to single exposures by approximately 20%, although most of the DPM mass was removed. In contrast, Campen et al. (2005) found that filtering exhaust reduced the lung inflammatory response in mice exposed for 3 days. They found, however, that the nonparticle components accounted for all of the electrocardiographic abnormalities in the mice. Multiple Japanese studies of rodents exposed to whole or filtered exhaust, described in preceding sections, have shown that all, or the majority, of effects on amplification of allergic responses from in utero or postnatal exposures (Maejima et al., 2001; Watanabe and Ohsawa, 2002) and organ development (Watanabe and Oonuki, 1999; Watanabe and Kurita, 2001; Watanabe, 2005) were attributable to nonparticulate emissions. There have been few other studies including comparisons that point to causal components. Seagrave et al. (2001) measured inflammatory responses in rat lungs after instillation of combined PM and SVOC or the two fractions separately, collected in the heavy-duty bore of a traffic tunnel. They found that the SVOC contributed more to the response than the PM, even though there was less SVOC mass. Although both fractions caused inflammation, the SVOC fraction was four-fold more potent per unit mass than the PM. Using a similar animal response model and a multivariate statistical approach McDonald et al. (2004d) determined that hopanes and steranes, markers of engine oil, were the chemical species that covaried most closely with the inflammatory and cytotoxic effects of instilled PM and SVOC collected directly from emissions of normal- and high-emitting diesel and gasoline vehicles. These species existed primarily in the PM phase. Overall, the results to date demonstrate the fallacy of assuming, without confirmation, that any one physical-chemical fraction of diesel exhaust is responsible for the effects of whole exhaust. Results to date also show that different fractions are likely responsible for different effects. There are questions about the hazards presented by the smallest portion of DPM, falling into the loosely defined size range broadly termed “nanoparticle.” Although PM of this size has always been present in diesel exhaust and other combustion emissions, this fraction has received increased attention due to the recent upsurge of “nanotechnologies” in the manufacturing, personal product, and pharmaceutical fields, along with accompanying concerns for potential health hazards (Borm et al., 2006). Although combustion-derived nanoparticles have been discussed as a subcomponent of the more general issue (Donaldson et al., 2005), most health researchers have focused only on small (singlet) units of the ECbased soot agglomerates. It is not widely appreciated by the health community that many exhaust nanoparticles are condensates of SVOCs that have no EC core. The behavior of this material in the lung is unknown. Indeed, to the extent that the material disperses on the surfactant-rich liquid surface layer of the lung, it may not persist in particulate form, and any hazard would likely be driven by bulk chemistry in which the delivered mass would be the critical dose determinant. The delivered mass, in contrast to particle number, would be very small. There are yet no published studies of the toxicity of this class of DPM or evaluating the relative contributions of this class compared to other DPM. The issue is important because

CURRENT ISSUES

611

emissions of this material could persist if heavy organic vapors are not eliminated by new emission control strategies. Indeed, as EC is removed, this class could become a predominant form of DPM (although at very low mass emission rates). 16.5.2

Health Implications of Aged Exhaust

With very few exceptions, studies of the health effects of diesel exhaust have addressed fresh emissions. This is especially true for laboratory studies, and is also largely true for epidemiological studies of occupational exposures. While this case is important for on-road and near-road (or near-engine) exposures, people are exposed in the environment to exhaust and its atmospheric reaction products of various ages up to several days. Exhaust components are transformed by condensation of semivolatile fractions onto particles, oxidation reactions, chemical reactions with other air contaminants in the presence of temperature changes and sunlight, and other processes (reviewed by Zielinska, 2005). These changes are at least partially understood from studies involving atmospheric reaction (“smog”) chambers, but it is undoubtedly true that the full range of transformations is not known. There have been few attempts to link smog chamber studies to health assays, although the mutagenicity of DPM extracts has been shown to be altered by aging in such systems (Claxton and Barnes, 1981). In simpler reaction systems, the exposure of DPM to ozone has been shown to increase the lung inflammatory response of rats dosed by instillation (Madden et al., 2000) and production of inflammatory cytokines in cultured human airway epithelial cells (Kafoury and Kelley, 2005). The few studies of animals exposed to aged exhaust, however, have not suggested an increase in toxicity. Pepelko and Peirano (1983) exposed strain A mice 20 h/day, 7 days/week for 8 weeks to “raw” or irradiated and aged exhaust at 6000 mg DPM/m3, followed by a 9-month recovery period and evaluation of lung tumor incidence. Both the percentage of mice with tumors and the number of tumors per mouse were lower in those exposed to irradiated exhaust. Haussmann et al. (1998) exposed Wistar rats 6 h/day, 7 days/week for 13 weeks to fresh diesel exhaust or exhaust aged for 30 min and evaluated indices of inflammation and cytotoxicity in the lung. Both the numbers of inflammatory cells and the concentration of lactate dehydrogenase in lung lavage fluid were more than 50% lower in rats exposed to the aged exhaust. Questions about the effect of aging and atmospheric transformations on the health hazards from diesel exhaust blend quickly into the larger issue of the physical-chemical species responsible for the health effects of air pollution in general. However, a careful assessment of the health impacts of diesel exhaust must consider that effects of fresh and aged exhaust might differ. 16.5.3

Health Impacts of Reduction of Emissions and New Technologies

Nearly all of our information on the health impacts of diesel exhaust pertains to fuel, engine, and after-treatment technologies that are already or are rapidly becoming outdated. The information from laboratory studies also results from a wide spectrum of on-road and offroad engines spanning 25 model years and an assortment of fuels (many not specified). While it is true that the longevity of diesel engines ensures that older technology systems will remain in use for some time, it is also true that emissions from new vehicles and equipment bear scant resemblance to emissions from even 10 years ago. There is little doubt that compression ignition will continue to comprise a portion of internal combustion engines for the foreseeable future. These engines will undoubtedly continue to be called “diesels,” but

612

DIESEL EXHAUST

the marked changes in technology and resulting emissions warrant a distinction between emissions from systems marketed today for on-road use (and rapidly including off-road systems as well) and those of the past. There is yet no straightforward way to demarcate old engine versus new engine emissions, although Hesterberg et al. (2005) coined the term “new technology diesel” (NTD) to describe a combination of advanced engine design, ultra-low sulfur fuel, specialized lubricants, and catalyzed particle traps. While it is logical, and undoubtedly correct, that “less is better” when it comes to health hazards from engine emissions, the health benefits of tightened emission standards and the residual hazards that may remain are worthy of continued evaluation. At this time, there is only one published study directly comparing the health effects of inhalation exposure to conventional diesel and NTD exhaust. McDonald et al. (2004b) exposed C57BL/6 mice 6 h/day for 7 days to the same dilution of exhaust from a 0.4-L singlecylinder engine operated in steady state at full load under two conditions. Conventional conditions were simulated by fueling the engine with contemporary average on-road fuel (certification fuel at 371 ppm sulfur), and NTD conditions were simulated by using ultra-low sulfur fuel (ECD1, British Petroleum, 14 ppm sulfur) and adding an appropriately sized catalyzed ceramic particle trap. The DPM mass, CO, and VOC emissions were markedly reduced under the NTD condition, but NOx emissions were unchanged (no NOx reduction after-treatment was used). Evaluations of health response included lung histopathology, clearance of Respiratory Syncytial Virus, and inflammatory cytokines and indicators of oxidative stress in lung homogenate. All indicators were significantly changed by conventional exposure, but no significant differences from control were induced by NTD exposure. These results suggest that reduction of DPM, VOC, and CO emissions by improved fuel and after-treatment will likely have substantial health benefit, even if the benefits, which will only be implemented gradually, will be difficult or impossible to confirm by epidemiology. It is also useful to know that these benefits were achieved by fuel and after-treatment strategies that can be retrofitted to existing conventional systems. A related issue is whether or not alternative fuels or advanced after-treatment strategies might introduce unintended consequences in the form of health hazards. Numerous new liquid fuels are entering the diesel fuel portfolio; some are already in use, and others are on the way. All pundits forecast at least partial displacement of conventional petroleum diesel stocks by these fuels. These include petroleum-water emulsion blends, vegetable and animal oil-derived biodiesel, grain and biomass-derived alcohols, North American tar sand petroleum, gas-to-liquid and coal-to-liquid fuels, dimethyl ether, and various blends of these with conventional petroleum fuel. Few of these fuels have been tested for health hazards, and there have been even fewer comparisons to conventional petroleum-based fuel. In compliance with EPA requirements for certifying new fuels and fuel additives, 90-day inhalation studies of rats exposed to multiple levels of exhaust from engines burning soybean oil-derived biodiesel (100% soy methyl ester) (Finch et al., 2002), petroleum diesel–water emulsion (Reed et al., 2005), and the same emulsion containing methanol (Reed et al., 2006a) were conducted with numerous health endpoints. Few significant effects were found for any of the fuels even at the highest exposure level (they were subsequently approved for commercial use), but two issues remain. First, none of the studies included comparison to conventional diesel fuel. Second, the regulatory test protocol does not address several health outcomes that have resulted from high exposures to conventional diesel exhaust, such as cardiovascular effects, effects on allergic responses or systemic immune function, or effects on organ development. Exhaust from engines burning the other new fuels has received almost no scrutiny. The scanty emissions data available and the fact that the new fuels will be used in

REFERENCES

613

combination with new engine and after-treatment technologies suggest that new health issues are unlikely; however, some attention to comparative toxicity screening is certainly warranted.

ACKNOWLEDGMENTS Mauderly’s effort was supported by the National Environmental Respiratory Center, a joint government-industry program funded by the EPA Office of Research and Development (CR831455-01-0), the U.S. Department of Energy (DOE) FreedomCar and Vehicle Technology Program (DE-FC04-96AL76406), the DOE Office of Fossil Energy (DE-FC2605NT42304), and several corporations in the automotive, electrical utility, engine, and petroleum sectors. This chapter has not been reviewed by sponsors, and is not intended to reflect the views of any sponsor. Garshick’s effort was supported by the Office of Research and Development, Department of Veterans Affairs, National Institutes of Health National Cancer Institute (R01 CA90792), and National Institute of Environmental Health Science (R21 ES013726).

REFERENCES Ames RG, Attfield MD, Hankinson JL, Hearl FJ, Reger RB (1982) Acute respiratory effects of exposure to diesel emissions in coal miners. Am. Rev. Respir. Dis. 125:39–42. Ames RG, Reger RB, Hall DS (1984) Chronic respiratory effects of exposure to diesel emissions in coal mines. J. Occup. Med. 39:389–394. Arey J, Zielinska B, Atkinson R, Winer AM (1988) Formation of nitroarenes during ambient highvolume sampling. Environ. Sci. Technol. 22:457–462. Attfield MD, Trabant GD, Wheeler RW (1982) Exposure to diesel fumes and dust at six potash miners. Ann. Occup. Hyg. 26:817–831. Bagley ST (1996) Characterization of fuel and after-treatment device effects on diesel emissions. HEI Report No. 76. Barrett EG, Reed MD, Espindola T, Henson R (2002) Diesel exhaust exposure in conjunction with periodic allergen challenge does not lead to allergic sensitization. Am. J. Respir. Crit. Care Med. 165 (8):A303. Bastain TM, Gilliland FD, Li YF, Saxon A, Diaz-Sanchez D (2003) Intraindividual reproducibility of nasal allergic responses to diesel exhaust particles indicates a susceptible phenotype. Clin. Immunol. 109(2):130–136. Battigelli MC (1965) Effects of diesel exhaust. Arch. Environ. Health 10:165–167. Battigelli MC, Mannella RJ, Hatch TF (1964) Environmental and clinical investigation of workmen exposed to diesel exhaust in railroad engine houses. Ind. Med. Surg. 3:121–124. Bechtold WE, Dutcher JS, Mokler BV, Lopez JA, Wolf I, Li AP, Henderson TR, McClellan RO (1984) Chemical and biological properties of diesel exhaust particles collected during selected segments of a simulated driving cycle. Fundam. Appl. Toxicol. 4:370–377. Behndig AF, Mudway IS, Brown JL, Stenfors N, Helleday R, Duggan ST, Wilson SJ, Boman C, Cassee FR, Frew AJ, Kelly FJ, Sandstrom T, Blomberg A (2006) Airway antioxidant and inflammatory responses to diesel exhaust exposure in healthy humans. Eur. Respir. J. 27:359–365 Benhamou S, Benhamou E, Flamant R (1988) Occupational risk factors of lung cancer in a French case–control study. Br. J. Ind. Med. 45:231–233.

614

DIESEL EXHAUST

Bhatia R, Lopipero P, Smith AH (1998) Diesel exhaust exposure and lung cancer. Epidemiology 9(1):84–91. Bigert C, Gustavsson P, Hallqvist J, Hogstedt C, Lewne M, Plato N, Reuterwall C, Scheele P (2003) Myocardial infarction among professional drivers. Epidemiology 14(3):333–339. Bigert C, Klerdal K, Hammar N, Hallqvist J, Gustavsson P (2004) Time trends in the incidence of myocardial infarction among professional drivers in Stockholm 1977–96. Occup. Environ. Med. 61 (12):987–991. Boffetta P, Silverman DT (2001) A meta-analysis of bladder cancer and diesel exhaust exposure. Epidemiology 12(1):125–130. Boffetta P, Stellman SD, Garfinkel L (1988) Diesel exhaust exposure and mortality among males in the American Cancer Society prospective study. Am. J. Ind. Med. 14:403–415. Boffetta P, Harris RE, Wynder EL (1990) Case–control study on occupational exposure to diesel exhaust and lung cancer risk. Am. J. Ind. Med. 15:577–591. Boffetta P, Dosemeci M, Gridley G, Bath H, Moradi T, Silverman D (2001) Occupational exposure to diesel engine emissions and risk of cancer in Swedish men and women. Cancer Causes Control 12:365–374. Bond JA, Butler MM, Medinsky MA, Muggenburg BA, McClellan RO (1984) Dog pulmonary macrophage metabolism of free and particle-associated [14 C]benzo[a]pyrene. J. Toxicol. Environ. Health 14:181–189. Bond JA, Sun JD, Medinsky MA, Jones RD, Yeh HC (1986) Deposition, metabolism, and excretion of 1-[14 C]nitropyrene and 1-[14 C]nitropyrene coated on diesel exhaust particles as influenced by exposure concentration. Toxicol. Appl. Pharmacol. 85:102–117. Bond JA, Wolff RK, Harkema JR, Mauderly JL, Henderson RF, Griffith WC, McClellan RO (1990a) DNA adduct distribution in the respiratory tract of rats exposed to diesel exhaust. In: Pelroy RA, editor. Multilevel Health Effects Research: From Molecules to Man, DOE Symposium Series. Richland, WA:Battelle Pacific Northwest Laboratory, pp.349–355. Bond JA, Harkema JR, Henderson RF, Mauderly JL, McClellan RO, Wolff RK (1990b)The role of DNA adducts in diesel exhaust-induced pulmonary carcinogenesis. In: Mutation and the Environment. New York: Wiley-Liss, pp.259–269. Bond JA, Mauderly JL, Wolff RK (1990c) Concentration- and time-dependent formation of DNA adduct in lungs of rats exposed to diesel exhaust. Toxicology 60:127–135. Bond JA, Johnson NF, Snipes MB, Mauderly JL (1990d) DNA adduct formation in rats alveolar type II cells–cells potentially at risk for inhaled diesel exhaust. Environ. Mol. Mutagen. 16: 64–69. Borm PJ, Robbins D, Haubold S, Kuhlbusch T, Fissan H, Donaldson K, Schins RP, Stone V, Kreyling W, Lademann J, Krutmann J, Warheit D, Oberdorster E (2006) The potential risks of nanomaterials: a review carried out for ECETOC. Part. Fibre Toxicol. 3(1):11. Brauer M, Hoek G, Van Vliet P, Meliefste K, Fischer PH, Wijga A, Koopman LP, Neijens HJ, Gerritson J, Kerkhof M, Heinrich J, Bellander T, Brunekreef B (2002) Air pollution from traffic and the development of respiratory infections and asthmatic and allergic symptoms in children. Am. J. Respir. Crit. Care Med. 166(8):1092–1098. Bremner SA, Anderson HR, Atkinson RW, McMichael AJ, Strachan DP, Bland JM, Bower JS (1999) Short-term associations between outdoor air pollution and mortality in London 1992–1994. Occup. Environ. Med. 56(4):237–244. Brightwell I, Fouillet X, Cassano-Zoppi A-L, Bernstein D, Crawley E, Duchosal E, Gatz R, Perczel S, Pfeifer H (1989) Tumours of the respiratory tract in rats and hamsters following chronic inhalation of engine exhaust emissions. J. Appl. Toxicol. 9:23–31. Brooks AL, Wolff RK, Royer RE, Clark CR, Sanchez A, McClellan RO (1981) Deposition and biological availability of diesel particles and their associated mutagenic chemicals. Environ. Int. 5:263–267.

REFERENCES

615

Brooks AL, Li AP, Dutcher JS, Clark CR, Rothenberg SJ, Kiyoura R, Bechtold WE, McClellan RO (1984) A comparison of genotoxicity of automotive exhaust particles from laboratory and environmental sources. Environ. Mutagen. 6:651–668. Brunekreef B, Janssen NA, de Hartog J, Harssema H, Knape M, Van Vliet P (1997) Air pollution from truck traffic and lung function in children living near motorways. Epidemiology 8(3):298–303. Bruske-Hohlfeld I, Mohner M, Ahrens W, Pohlabein H, Heinrich J, Kreuzer M, Jockel KH, Wichmann HE (1999) Lung cancer risk in male workers occupationally exposed to diesel motor emissions in Germany. Am. J. Ind. Med. 36(4):405–414. Bunn WB, Valberg PA, Slavin TJ, Hart G, Lapin CA (2004) A reevaluation of the literature regarding the health assessment of diesel engine exhaust. Inhal. Toxicol. 16:889–900. Burchiel SW, Lauer FT, McDonald JD, Reed MD (2004) Systemic immunotoxicity in AJ mice following 6-month whole body inhalation exposure to diesel exhaust. Toxicol. Appl. Pharmacol. 196:337–345. Burchiel SW, Lauer FT, Dunaway SL, Zawadzki J, McDonald JD, Reed MD (2005) Hardwood smoke alters murine splenic T cell responses to mitogens following a six-month whole body inhalation exposure. Toxicol. Appl. Pharmacol. 202(3):229–236. Cal EPA (California Environmental Protection Agency) (1998) Health Risk Assessment for Diesel Exhaust Part B: Proposed Identification of Diesel Exhaust as a Toxic Air Contaminant. Sacramento, CA: Office of Environmental Health Hazard Assessment, California Air Resources Board. Campbell KI, George EL, Washington IS Jr (1981) Enhanced susceptibility to infection in mice after exposure to dilute exhaust from light duty diesel engines. Environ. Int. 5:377–382. Campen MJ, McDonald JD, Gigliotti AP, Seilkop SK, Reed MD, Benson JM (2003) Effects of inhaled diesel exhaust in spontaneously hypertensive rats. Cardiovasc. Toxicol. 3(4):353–361. Campen MJ, Babu NS, Helms GA, Pett S, Wernly J, Mehran R, McDonald JD (2005) Nonparticulate components of diesel exhaust promote constriction in coronary arteries from ApoE / mice. Toxicol. Sci. 88:95–102. Castranova V, Ma JYC, Yang H-M, Antonini JM, Butterworth L, Barger MW, Roberts J, Ma JKH (2001) Effect of exposure to diesel exhaust particles on the susceptibility of the lung to infection. Environ. Health Perspect. 109(Suppl. 4):609–612. Cheng YS, Yeh HC, Mauderly JL, Mokler BV (1984) Characterization of diesel exhaust in a chronic inhalation study. Am. Ind. Hyg. Assoc. J. 45:547–555. Ciccone G, Forastiere F, Agabiti N, Biggeri A, Bisanti L, Chellini E, Corbo G, Dell’Orco V, Dalmasso P, Volante TF, Galassi C, Piffer S, Renzoni E, Rusconi F, Sestini P, Viege G , and the SIDRIA Collaborative Group (1998) Road traffic and adverse respiratory effects in children. Occup. Environ. Med. 55(11):771–778. Clark CR, Royer RE, Brooks AlMcClellan RO, Marshal WE, Naman TM, Seizinger DE (1981) Mutagenicity of diesel exhaust particle extracts: Influence of car type. Fundam. Appl. Toxicol. 1:260–265. Clark CR, Dutcher JS, Henderson TR, McClellan RO, Marshal WE, Naman TM, Seizinger DE (1984) Mutagenicity of automotive particulate exhaust: influence of fuel extenders, additives and aromatic content. Adv. Mod. Environ. Toxicol. 6:109–122. Claxton LD (1983) Characterization of automotive emissions by bacterial mutagenesis bioassay: a review. Environ. Mutagen. 5:609–631. ClaxtonLD,BarnesHM(1981)Themutagenicityofdiesel-exhaustparticleextractscollectedundersmogchamber conditions using the Salmonella typhimurium test system. Mutat. Res. 88(3):255–272. Cross ET, Palmer RE, Filipy RE, Busch RH, Stuart BO (1978) Study of the combined effects of smoking and inhalation of uranium ore dust, radon daughters, and diesel exhaust fumes in hamsters and dogs. Report PNL-2744: UC-48. Richland, WA: Pacific Northwest Laboratory.

616

DIESEL EXHAUST

Daniel JH (1984) The use of diesel-powered equipment in U.S. underground coal operations. Presented at the American Mining Congress International Coal Show, Chicago, May 3. Davis ME, Smith TJ, Laden F, Hart JE, Ryan LM, Garshick E (2006) Modeling particle exposure in US trucking terminals. Environ. Sci. Technol. 40:4226–4232. DeMarini DM, Brooks LR, Warren SH, Kobayashi T, Gilmour MI, Singh P (2004) Bioassay-directed fractionation and salmonella mutagenicity of automobile and forklift diesel exhaust particles. Environ. Health Perspect. 112:814–819. Diaz-Sanchez D, Dotson AR, Takenaka H, Saxon A (1994) Diesel exhaust particles induce local IgE production in vivo and alter the pattern of IgE messenger RNA isoforms. J. Clin. Invest. 94 (4):1417–1425. Diaz-Sanchez D, Tsien A, Casillas A, Dotson AR, Saxon A (1996) Enhanced nasal cytokine production in human beings after in vivo challenge with diesel exhaust particles. J. Allergy Clin. Immunol. 98(1):114–123. Diaz-Sanchez D, Tsien A, Fleming J, Saxon A (1997) Combined diesel exhaust particulate and ragweed allergen challenge markedly enhances human in vivo nasal ragweed-specific IgE and skews cytokine production to a T helper cell 2-type pattern. J. Immunol. 158 (5):2406–2413. Diaz-Sanchez D, Tsien A, Fleming J, Saxon A (1999a) Effect of topical fluticasone propionate on the mucosal allergic response induced by ragweed allergen and diesel exhaust particle challenge. Clin. Immunol. 90(3):313–322. Diaz-Sanchez D, Garcia MP, Wang M, Jyrala M, Saxon A (1999b) Nasal challenge with diesel exhaust particles can induce sensitization to a neoallergen in the human mucosa. J. Allergy Clin. Immunol. 104(6):1183–1188. Dockery DW, Pope AC 3rd, Xu X, Spengler JD, Ware JH, Fay ME, Ferris BG, Speizer FE Jr (1993) An association between air pollution and mortality in six U.S. cities. N. Engl. J. Med. 329 (24):1753– 1759. Dockery DW, Luttmann-Gibson H, Rich DQ, Link MS, Mittleman MA, Gold DR, Koutrakis P, Schwartz JD, Verrier RL (2005) Association of air pollution with increased incidence of ventricular tachyarrhythmias recorded by implanted cardioverter defibrillators. Environ. Health Perspect. 113 (6):670–674. Dominici F, Peng RD, Bell ML, Pham L, McDermott A, Zeger SL, Samet JM (2006) Fine particulate air pollution and hospital admission for cardiovascular and respiratory diseases. JAMA 295 (10):1127–1134. Donaldson K, Tran L, Jimenez LA, Duffin R, Newby DE, Mills N, MacNee W, Stone V (2005) Combustion-derived nanoparticles: a review of their toxicology following inhalation exposure. Part. Fibre Toxicol. 2:10. Dugandzic R, Dodds L, Stieb D, Smith-Doiron M (2006) The association between low level exposures to ambient air pollution and term low birth weight: a retrospective cohort study. Environ. Health 5:3. Duhme H, Weiland SK, Keil U, Kraemer B, Schmid M, Stender M, Chambless L (1996) The association between self-reported symptoms of asthma and allergic rhinitis and self-reported traffic density on street of residence in adolescents. Epidemiology 7(6):578–582. Easter MD, Ireson R, Liu L, Lawson D, Zielinska B, Ondov J, Weaver C, Davey M, Lapin C, Hesterberg T (2006) Estimating source contribution inside school buses. Toxicologist 90:175. Edling C, Anjou C-G, Axelson O, Kling H (1987) Mortality among personnel exposed to diesel exhaust. Int. Arch. Occup. Environ. Health 59:559–565. Elder A, Gelein R, Silva V, Feikert T, Opanashuk L, Carter J, Potter R, Maynard A, Ito Y, Finkelstein J, Oberd€orster G (2006) Translocation of inhaled ultrafine manganese oxide particles to the central nervous system. Environ. Health Perspect. 114:1172–1178.

REFERENCES

617

Emmelin A, Nystr€om L, Wall S (1993) Diesel exhaust exposure and smoking: a case-referent study of lung cancer among Swedish dock workers. Epidemiology 4:237–244. Enya T, Suzuki H, Watanabe T, Hirayama T, Hisamatsu Y (1997) 3-Nitrobenzanthrone, a powerful bacterial mutagen and suspected human carcinogen found in diesel exhaust and airborne particulates. Environ. Sci. Technol. 31(10):2772–2776. EPA (U.S. Environmental Protection Agency) (1996) 1996 National scale air toxics assessment. National Center for Environmental Assessment, Office of Research and Development. Washington, DC: U.S. Environmental Protection Agency. Available at http://epa.gov/ttn/atw/nata/index. html. EPA (U.S. Environmental Protection Agency) (2002) Health assessment document for diesel engine exhaust. EPA 600/8-90/057F. National Center for Environmental Assessment, Office of Research and Development. Washington, DC: U.S. Environmental Protection Agency. EPA (U.S. Environmental Protection Agency) (2004) Air quality criteria for particulate matter. EPA/ 600/P99/002aF. National Center for Environmental Assessment, Office of Research and Development. Washington, DC: U.S. Environmental Protection Agency. Finch GL, Hobbs CH, Blair LF, Barr EB, Hahn FF, Jaramillo RJ, Kubatko JE, March TH, White RK, Krone JR, Menache MG, Nikula KJ, Mauderly JL (2002) Effects of subchronic inhalation exposure of rats to emissions from a diesel engine burning soybean oil-derived biodiesel fuel. Inhal. Toxicol. 14:101–132. Finkelstein MM, Verma DK, Sahai D, Stefov E (2004) Ischemic heart disease mortality among heavy equipment operators. Am. J. Ind. Med. 46 (1):16–22. Frey JW, Corn M (1967) Diesel exhaust particulates. Nature 216:615–616. Friedman MS, Powell KE, Hutwagner L, Graham LM, Teague WG (2001) Impact of changes in transportation and commuting behaviors during the 1996 Summer Olympic Games in Atlanta on air quality and childhood asthma. JAMA 285(7):897–905. Fujieda S, Diaz-Sanchez D, Saxon A (1998) Combined nasal challenge with diesel exhaust particles and allergen induces in vivo IgE isotype switching. Am. J. Respir. Cell Mol. Biol. 19(3):507–512. Gallagher J, Heinrich U, George M, Hendee L, Phillips DH, Lewtas J (1994) Formation of DNA adducts in rat lung following chronic inhalation of diesel emissions, carbon black and titanium dioxide particles. Carcinogenesis 15(7):1291–1299. Gamble J, Jones W, Hudak J, Merchant J (1978) Acute changes in pulmonary function in salt miners. In: Proceedings of an American Council of Governmental Industrial Hygiene Topical Symposium: Industrial Hygiene for Mining and Tunneling, November 6–7, 1978. Cincinnati, OH: American Council of Governmental Industrial Hygiene. Gamble J, Jones W, Hudak J (1983) An epidemiological study of salt miners in diesel and nondiesel mines. Am. J. Ind. Med. 4:435–458. Gamble J, Jones W, Minshall S (1987) Epidemiological-environmental study of diesel bus garage workers: chronic effects of diesel exhaust on the respiratory system. Environ. Res. 44:6–17. Garshick E, Shenker MB, Munoz A, Segal M, Smith TJ, Woskie SR, Hammond SK, Speizer FE (1987) A case–control study of lung cancer and diesel exhaust exposure in railroad workers. Am. Rev. Respir. Dis. 135:1242–1248. Garshick E, Shenker MB, Munoz A, Segal M, Smith TJ, Woskie SR, Hammond SK, Speizer FE (1988) A retrospective cohort study of diesel exhaust exposure and lung cancer in railroad workers. Am. Rev. Respir. Dis. 137:820–825. Garshick E, Laden F, Hart JE, Caron A (2003) Residence near a major road and respiratory symptoms in US veterans. Epidemiology 14:728–736. Garshick E, Laden F, Hart JE, Rosner B, Smith TJ, Dockery DW, Speizer FE (2004) Lung cancer in railroad workers exposed to diesel exhaust. Environ. Health Perspect. 112:1539–1543.

618

DIESEL EXHAUST

Garshick E, Laden F, Hart JE, Smith TJ, Rosner B (2006) Smoking imputation and lung cancer in railroad workers exposed to diesel exhaust. Am. J. Ind. Med. 49:709–718. Gauderman WJ, McConnell R, Gilliland F, London S, Thomas D, Avol E, Vora H, Berhane K, Rappaport EB, Lurmann F, Margolis HG, Peters J (2000) Association between air pollution and lung function growth in southern California children. Am. J. Respir. Crit. Care Med. 162(4 Pt. 1):1383– 1390. Gauderman WJ, Gilliland GF, Vora H, Avol E, Stram D, McConnell R, Thomas D, Lurmann F, Margolis HG, Rappaport EB, Berhane K, Peters JM (2002) Association between air pollution and lung function growth in southern California children: results from a second cohort. Am. J. Respir. Crit. Care Med. 166(1):76–84. Gerde P, Muggenburg BA, Lundborg M, Dahl AR (2001) The rapid alveolar absorption of diesel sootadsorbed benzo(a)pyrene: bioavailability, metabolism, and dosimetry of an inhaled particle-borne carcinogen. Carcinogenesis 22:101–109. Gold DR, Litonjua AA, Zanobetti A, Coull BA, Schwartz J, MacCallum G, Verrier RL, Nearing BD, Canner MJ, Suh H, Stone PH (2005) Air pollution and ST-segment depression in elderly subjects. Environ. Health Perspect. 113(7):883–887. Goss CH, Newsom NA, Shildcrout JS, Sheppard L, Kaufman JD (2004) Effect of ambient air pollution on pulmonary exacerbations and lung function in cystic fibrosis. Am. J. Respir. Crit. Care Med. 169 (7):816–821. Guo YL, Lin YC, Sung FC, Haung SL, Ko YC, Lai JS, Su HJ, Shaw CK, Lin RS, Dockery DW (1999) Climate, traffic-related air pollutants, and asthma prevalence in middle-school children in Taiwan. Environ. Health Perspect. 107(12):1001–1006. Guo J, Kauppin T, Kyyronen P, Lindbohm ML, Heikkila P, Pukkala (2004) Occupational exposure to diesel and gasoline engine exhausts and risk of lung cancer among Finnish workers. Am. J. Ind. Med. 45(6):483–490. Gustafsson L, Wall S, Larsson L-G, Skog B (1986) Mortality and cancer incidence among Swedish dock workers—a retrospective cohort study. Scand. J. Work Environ. Health 12:22–26. Gustavsson P, Alfreditorsson L, Brunnberg H, Hammar N, Jakobsson R, Reuterwall C, Ostlin P (1996) Myocardial infarction among male bus, taxi, and lorry drivers in middle Sweden. Occup. Environ. Med. 53(4):235–240. Gustavsson P, Jakobsson R, Nyberg F, Pershagen G, Jarup L, Scheele P (2000) Occupational exposure and lung cancer risk: a population-based case-referent study in Sweden. Am. J. Epidemiol. 152 (1):32–40. Gustavsson P, Plato N, Hallqvist J, Hogstedt C, Lewne M, Reuterwall C, Scheele P (2001) A population-based case-referent study of myocardial infarction and occupational exposure to motor exhaust, other combustion products, organic solvents, lead, and dynamite. Stockholm Heart Epidemiology Program (SHEEP) Study Group. Epidemiology 12(2):222–228. Hahon N, Booth JA, Green F, Lewis TR (1985) Influenza virus infection in mice after exposure to coal dust and diesel engine emissions. Environ. Res. 37:44–60. Hampton CV, Pierson WR, Schuetzle D, Harvey TM (1983) Hydrocarbon gases emitted from vehicles on the road. 2. Determination of emission rates from diesel and spark-ignition vehicles. Environ. Sci. Technol. 17:699–708. Hannerz H, Tuchsen F (2001) Hospital admissions among male drivers in Denmark. Occup. Environ. Med. 58(4):253–260. Hansen J, Raaschou-Nielsen O, Olsen JH (1998) Increased risk of lung cancer among different types of professional drivers in Denmark. Occup. Environ. Med. 55:115–118 Hansen C, Neller A, Williams G, Simpson R (2006) Maternal exposure to low levels of ambient air pollution and preterm birth in Brisbane, Australia. BJOG 113(8):935–941.

REFERENCES

619

Hao M, Comier S, Wang M, Lee JJ, Nel A (2003) Diesel exhaust particles exert acute effects on airway inflammation and function in murine allergen provocation models. J. Allergy Clin. Immunol. 112 (5):905–914. Harrod KS, Jaramillo RJ, Rosenberger CL, Wang SZ, Berger JA, Reed MD (2003) Increased susceptibility to RSV infection by exposure to inhaled diesel engine emissions. Am. J. Respir. Cell Mol. Biol. 28:451–463. Harrod KS, Jaramillo RJ, Berger JA, Gigliotti AP, Seilkop SK, Reed MD (2004) Inhaled diesel engine emissions reduce bacterial clearance and exacerbate lung disease to Pseudomonas aeruginosa infection. Toxicol. Sci. 83:155–165. Hart JE, Laden F, Shenker MB, Garshick E (2006) Chronic obstructive pulmonary disease mortality in diesel exposed railroad workers. Environ. Health Perspect. 114:1013–1017. Hatch GE, Boykin E, Graham JA, Lewtas J, Pott F, Loud K, Mumford JL (1985) Inhalable particles and pulmonary host defense: in vivo and in vitro effects of ambient air and combustion particles. Environ. Res. 37:67–80. Haussmann HJ, Anskett E, Kuhl P, Rustemeier K, Schepers G, Schnell P, Terpstra P (1998)The influence of diesel engine exhaust age on lung soot burden and related responses of rats. In: Mohr U, editor. Relationships Between Respiratory Disease and Exposure to Air Pollution.Washington, DC: International Life Sciences Institute, ILSO Press. pp.338–342. Hayes RB, Thomas T, Silverman DT, Vineis P, Blot WJ, Mason TJ, Pickle LW, Correa P, Fontham ETH, Schoenberg JB (1989) Lung cancer in motor exhaust occupations. Am. J. Ind. Med. 16:685– 695. HEI (Health Effects Institute)(1995) Diesel exhaust: a critical analysis of emissions, exposure, and health effects. A Special Report of the Institute’s Diesel Working Group. Cambridge, MD: Health Effects Institute. Heinrich U, Peters L, Funcke W, Potts E, Mohr U, St€ober W (1982)Investigations of toxic and carcinogenic effects of diesel exhaust in long-term inhalation exposure of rodents. In: Lewtas J, editor. Toxicological Effects of Emissions from Diesel Engines.Amsterdam:Elsevier, pp.225– 242. Heinrich U, Muhle H, Takenaka S, Ernst E, Fuhst R, Mohr U, Pott E, St€ober W (1986) Chronic effects on the respiratory tract of hamsters, mice and rats after long-term inhalation of high concentrations of filtered and unfiltered diesel engine emissions. J. Appl. Toxicol. 6:383–395. Heinrich U, Fuhst R, Rittinghausen S, Creutzenberg O, Bellmann B, Koch W, Levsen K (1995) Chronic inhalation exposure of Wistar rats and two different strains of mice to diesel engine exhaust, carbon black, and titanium dioxide. Inhal. Toxicol. 7:533–556. Henderson RF, Pickrell JA, Jones RD, Sun JD, Benson JM, Mauderly JL, McClellan RO (1988) Response of rodents to inhaled diluted diesel exhaust: biochemical and cytological changes in bronchoalveolar lavage fluid and in lung tissue. Fundam. Appl. Toxicol. 11:546–567. Hertz-Picciotto I, Herr CE, Yap PS, Dostal M, Shumway RH, Ashwood P, Lipsett M, Joad JP, Pinkerton KE, Sram RJ (2005) Air pollution and lymphocyte phenotype proportions in cord blood. Environ. Health Perspect. 113(10):1391–1398. Hesterberg TW, Bunn WB, McClellan RO, Hart GA, Lapin CA (2005) Carcinogenicity studies of diesel engine exhausts in laboratory animals: a review of past studies and a discussion of future research neditors. Crit. Rev. Toxicol. 35(5):379–411. Hiura TS, Kaszubowski MP, Li N, Nel AE (1999) Chemicals in diesel exhaust particles generate reactive oxygen radicals and induce apoptosis in macrophages. J. Immunol. 163:5582–5591. Hoek G, Brunekreef B, Goldbohm S, Fishcer P, Van den Brandt PA (2002) Association between mortality and indicators of traffic-related air pollution in the Netherlands: a cohort study. Lancet 360(9341):1203–1209.

620

DIESEL EXHAUST

Hoppin JA, Umbach DM, London SJ, Alavanja MC, Sandler DP (2004) Diesel exhaust, solvents, and other occupational exposures as risk factors for wheeze among farmers. Am. J. Respir. Crit. Care Med. 169(12):1308–1313. Howard PC, Hecht SS, Beland FA (1990) Nitroarenes, Occurrence, Metabolism, and Biological Impact. New York: Plenum Press. Howe GR, Fraser D, Lindsay J, Presnal B, Yu SZ (1983) Cancer mortality (1965–77) in relation to diesel fume and coal exposure in a cohort of retired railway workers. J. Natl. Cancer Inst. 70:1015– 1019. Huisingh JL, Bradow R, Jungers R, Claxton L, Zweidinger R, Tejada S, Bumgarner J, Duffield F, Waters M, Simmon FVF, Hare C, Rodiguez C, Snow L (1978)Application of bioassay to the characterization of diesel particle emissions. In: Waters MD, Nesnow S, Huisingh JL, Sandhu SS, Claxton L, editors. Application of Short-Term Bioassays in the Fractionation and Analysis of Complex Environmental Mixtures. New York: Plenum Press, pp. 381–418. IARC (International Agency for Research on Cancer)(1989) IARC Monographs on the Evaluation of Carcinogenic Risks to Humans, Vol. 46.Lyons:IARC. Ikeda M, Suzuki M, Watarai K, Sagai M, Tomita T (1995) Impairment of endothelium-dependent relaxation by diesel exhaust particles in rat thoracic aorta. Jpn. J. Pharmacol. 68 (2):183–189. ILSI (ILSI Risk Science Institute Workshop Participants)(2000) The relevance of the rat lung response to particle overload for human risk assessment: a workshop consensus report. Inhal. Toxicol. 12:1– 17. IPCS (International Programme on Chemical Safety)(1996) Environmental Health Criteria 171: Diesel Fuel and Exhaust Emissions.Geneva:World Health Organization. Ishihara T (1988) Diesel exhaust and health risks. Final Report of HERP Studies. Tsukuba, Japan: Health Effects Research Program. Iwai K, Udagawa T, Yamagishi M, Yamada H (1986)Long-term inhalation studies of diesel exhaust on F344 SPF rats. Incidence of lung cancer and lymphoma. In:Ishinishi N, Koizumi A, McClellan RO, St€ober W, editors. Carcinogenic and Mutagenic Effects of Diesel Engine Exhaust.Amsterdam: Elsevier, pp.349–360. Iwai K, Adashi S, Takahashi M, M€oller L, Udagawa T, Mizuno S, Sugawara I (2000) Early oxidative DNA damages and late development of lung cancer in diesel exhaust-exposed rats. Environ. Res. A 84:255–264. Jacobson M, Smith TA, Hurley JF, Robertson A, Roscrow R (1988) Respiratory infections in coal miners exposed to nitrogen oxides. Research Report No. 18. Cambridge, MA: Health Effects Institute. Janssen NAH, Brunekreef B, Van Vliet P, Aarts F, Meliefste K, Harssema H, Fishcer P (2003) The relationship between air pollution from heavy traffic and allergic sensitization, bronchial hyperresponsiveness, and respiratory symptoms in Dutch schoolchildren. Environ. Health Perspect. 111:1512–1518. Jarvholm B, Silverman D (2003) Lung cancer in heavy equipment operators and truck drivers with diesel exhaust exposure in the construction industry. Occup. Environ. Med. 60(7):516–520. Jedrychowski W, Galas A, Pac A, Flak E, Camman D, Rauh V, Perera F (2005) Prenatal ambient air exposure to polycyclic aromatic hydrocarbons and the occurrence of respiratory symptoms over the first year of life. Eur. J. Epidemiol. 20(9):775–782. Johnson JH (1988)Automotive emissions. In:Watson AY, Bates RR, Kennedy D, editors. Air Pollution, The Automobile, and Public Health.Washington, DC:National Academy Press, pp.39–75. ˚ (1970) Studies on pulmonary function and respiratory tract symptoms of J€orgensen H, Svensson A workers in an iron ore mine where diesel trucks are used underground. J. Occup. Med. 12:348–354. Kado NY, Okamoto RA, Kuzmicky PA, Kobayashi R, Ayala A, Gebel ME, Rieger RL, Maddox C, Zafonte L (2005) Emissions of toxic pollutants from compressed natural gas and low sulfur

REFERENCES

621

diesel-fueled heavy-duty transit buses tested over multiple driving cycles. Environ. Sci. Technol. 39:7638–7649. Kafoury R, Kelley J (2005) Ozone enhances diesel exhaust particles (DEP)-induced interleukin-8 (IL-8) gene expression in human airway epithelial cells through activation of nuclear factors-kappaB (NF-kappaB) and IL-6 (NF-IL6). Int. J. Environ. Res. Public Health 2(3– 4):403–410. Kalra R, Singh SP, Savage SM, Finch GL, Sopori ML (2000) Effects of cigarette smoke on immune response: chronic exposure to cigarette smoke impairs antigen-mediated signaling in T cells and depletes IP3-sensitive Ca(2þ) stores. J. Pharmacol. Exp. Ther. 293(1):166–171. Kaplan I (1959) Relationship of noxious gases to carcinoma of the lung in railroad workers. JAMA 171:97–101. Kaplan HL, Springer KJ, MacKenzie WF (1983) Studies of potential health effects of long-term exposure to diesel exhaust emissions. Final Report No. 01-0750-103 (SWRI) and No. 1239 (SFRE). San Antonio, TX: Southwest Research Institute. Karagianes MT, Palmer RF, Busch RH (1981) Effects of inhaled diesel emissions and coal dust in rats. Am. Ind. Hyg. Assoc. J. 42:382–391. Karr C, Lumley T, Shepherd K, Davis R, Larson T, Ritz B, Kaufman J (2006) A case–crossover study of wintertime ambient air pollution and infant bronchiolitis. Environ. Health Perspect. 114 (2):277–281. Kato A, Nagai A, Kagawa J (2000) Morphological changes in rat lung after long-term exposure to diesel emissions. Inhal. Toxicol. 12:469–490. Keane MJ, Xing S-G, Harrison JC, Ong T, Wallace WE (1991) Genotoxicity of diesel-exhaust particles dispersed in simulated pulmonary surfactant. Mutat. Res. 260:233–238. Khalek IA (2004) Nitro-PAH artifact investigation. Final Report to National Renewable Energy Laboratory, SwRI Report No. 10738, November 2004. San Antonio, TX: Southwest Research Institute. Kim JJ, Smorodinsky S, Lipsett M, Singer BC, Hodgson AT, Ostro B (2004) Traffic-related air pollution near busy roads: The East Bay Children’s Respiratory Health Study. Am. J. Respir. Crit. Care Med. 170(5):520–526. King LC, Kohan MJ, Austin AC, Claxton LD, Lewtas J, Huisingh JL (1981) Evaluation of the release of mutagens from diesel particles in the presence of physiological fluids. Environ. Mutagen. 3:109– 121. Kittelson DB, Watts WF, Johnson JP, Remerowski ML, Ische EE, Oberdoster G, Gelein RM, Elder A, Hopke PK, Kim E, Zhao W, Zhou L, Jeong CH (2004) On-road exposure to highway aerosols. I. Aerosol and gas measurements. Inhal. Toxicol. 16(Suppl. 1):31–39. Kotin P, Falk HL, Thomas M (1955) Aromatic hydrocarbons. III. Presence in the particulate phase of diesel-engine exhausts and the carcinogenicity of exhaust extracts. Arch. Ind. Health 11:113–120. Kumagai Y, Koide S, Tauchi K, Endo A, Nakai Y, Yoshikawa T, Shimojo N (2002) Oxidation of proximal protein sulfhydryls by phenanthraquinone, a component of diesel exhaust particles. Chem. Res. Toxicol. 15(4):483–489. Kunzli N, Jerrett M, Mack WJ, Beckerman B, LaBree L, Gilliland F, Thomas D, Peters J, Hodis HN (2005) Ambient air pollution and atherosclerosis in Los Angeles. Environ. Health Perspect. 113 (2):201–206. Laden F, Neas LM, Dockery DW, Schwartz J (2000) Association of fine particulate matter from different sources with daily mortality in six U.S. cities. Environ. Health Perspect. 108(10):941– 947. Laden F, Schwartz J, Speizer FE, Dockery DW (2006a) Reduction in fine particulate air pollution and mortality: extended follow-up of the Harvard Six Cities study. Am. J. Respir. Crit. Care Med. 173 (6):667–672.

622

DIESEL EXHAUST

Laden F, Hart JE, Eschenroeder A, Smith TJ, Garshick E (2006b) Historical estimation of diesel exhaust exposure in a cohort study of U.S. railroad workers and lung cancer. Cancer Causes Control 17:911–919. Lanki T, De Hartog J, Heinrich J, Hoek G, Janssen NA, Peters A, Stolzel M, Timonen KL, Vallius M, Vanninen E, Pekkanen J (2006) Can we identify sources of fine particles responsible for exerciseinduced ischemia on days with elevated air pollution? The ULTRA study. Environ. Health Perspect. 114(5):655–660. Larkin EK, Smith TJ, Stayner L, Rosner B, Speizer FE, Garshick E (2000) Diesel exhaust exposure and lung cancer: adjustment for the effect of smoking in a retrospective cohort study. Am. J. Ind. Med. 38:399–409. Lawson DR, Smith RE (1998) The northern front range study. Final report, Colorado State University. Available at www.nfraqs.coloradostate.edu/nfraqs/default.html. Le Tertre A, Medina S, Samoli E, Forsberg B, Michelozzi P, Boumghar A, Vonk JM, Bellini A, Atkinson R, Ayres JG, Sunyer J, Schwartz J, Katsouyanni K (2002) Short-term effects of particulate air pollution on cardiovascular diseases in eight European cities. J. Epidemiol. Community Health 56(10):773–779. Lewis TR, Green FHY, Moorman WJ, Burg JR, Lynch DW (1986)A chronic inhalation toxicity study of diesel engine emissions and coal dust, alone and combineditor. In:Ishinishi N, Koizumi A, McClellan RO, St€ober W, editors. Carcinogenic and Mutagenic Effects of Diesel Engine Exhaust. Amsterdam:Elsevier, pp.361–380. Lewis TR, Green FHY, Moorman WJ, Burg JR, Lynch DW (1989) A chronic inhalation toxicity study of diesel engine emissions and coal dust, alone and combineditor. J. Am. Coll. Toxicol. 8:345–375. Li AP (1981) Antagonistic effects of animal sera, lung and liver cytosols, and sulfhydryl compounds on the cytotoxicity of diesel exhaust particle extracts. Toxicol. Appl. Pharmacol. 57:55–62. Li N, Venkatesan MI, Miguel A, Kaplan R, Gujuluva C, Alam J, Nel A (2000) Induction of heme oxygenase-1 expression in macrophages by diesel exhaust particle chemicals and quinines via the antioxidant-responsive element. J. Immunol. 165(6):3393–3401. Lipsett M, Campleman S (1999) Occupational exposure to diesel exhaust and lung cancer: a metaanalysis. Am. J. Public Health 89(7):1009–1017. Liu Y-Q, Keane M, Ensell M, Miller W, Kashon M, Ong T, Mauderly J, Lawson D, Gautam M, Zielinska B, Whitney K, Wallace W (2005) In vitro genotoxicity of exhaust emissions of diesel and gasoline engine vehicles operated on a unified driving cycle. J. Environ. Monitoring 7:60–66. Lloyd AC, Cackette TA (2001) Diesel engines: environmental impact and control. J. Air Waste Manage. Assoc. 51:809–847. Madden MC, Richards JH, Dailey LA, Hatch GE, Ghio AJ (2000) Effect of ozone on diesel exhaust particle toxicity in rat lung. Toxicol. Appl. Pharmacol. 168(2):140–148. Maejima K, Tamura K, Taniguchi Y, Nagase S, Tanaka H (1997) Comparison of the effects of various fine particles on IgE antibody production in mice inhaling Japanese cedar pollen allergens. J. Toxicol. Environ. Health 52:231–248. Maejima K, Tamura K, Nakajima T, Taniguchi Y, Saito S, Takenaka H (2001) Effects of the inhalation of diesel exhaust, Kanto loam dust, or diesel exhaust without particles on immune responses in mice exposed to Japanese cedar (Cryptomeria japonica) pollen. Inhal. Toxicol. 13:1047–1063. Mamessier E, Nieves A, Vervloet D, Magnan A (2006) Diesel exhaust particles enhance T-cell activation in severe asthmatics. Allergy 61(5):581–588. Manabe Y, Kinouchi T, Ohnishi Y (1985) Identification and quantification of highly mutagenic nitroacetoxypyrenes and nitrohydroxypyrenes in diesel-exhaust particles. Mutat. Res. 158 (1–2):3–18. Mathis U, Mohr M, Kaegi R, Bertola A, Boulouchos K (2005) Influence of diesel engine combustion parameters on primary soot particle diameter. Environ. Sci. Technol. 39(6):1887–1892.

REFERENCES

623

Mauderly JL (1992) Diesel exhaust. In: Lippmann M, editor. Environmental Toxicants—Human Exposures and Their Health Effects. New York, NY: Van Nostrand Reinhold, pp.119–162. Mauderly JL (1994a) Toxicological and epidemiological evidence for health risks from inhaled engine emissions. Environ. Health Perspect. 102(Suppl. 4):165–171. Mauderly JL (1994b) Non-cancer pulmonary effects of chronic inhalation exposure of animals to solid particles. In: Mohr U, Dungworth DL, Mauderly JL, Oberdoerster G, editors. Toxic and Carcinogenic Effects of Solid Particles in the Respiratory Tract. Washington, DC: ILSI Press, pp.43–56. Mauderly JL (1996) Lung overload: the dilemma and opportunities for resolution. Inhal. Toxicol. 8 (Suppl.):1–28. Mauderly JL (1997) Relevance of particle-induced rat lung tumors for assessing lung carcinogenic hazard and human lung cancer risk. Environ. Health Perspect. 105(Suppl. 5):1337–1346. Mauderly JL (2000) Chapter 7. Diesel exhaust. In:Lippmann M, editor. Environmental Toxicants: Human Exposures and Their Health Effects, 2nd edn. New York, NY: John Wiley & Sons, Inc, pp.193–241. Mauderly JL (2001) Diesel emissions: is more health research still needed?Toxicol. Sci. 62:6–9. Mauderly JL, McCunney RJ (1996) Particle Overload in the Rat Lung and Lung Cancer: Implications for Human Risk Assessment.Washington, DC:Taylor & Francis. Mauderly JL, Barr EB, Bice DE, Eidson AF, Henderson RF, Jones RD, Pickrell JA, Wolff RK (1986) Inhalation exposure of rats to oil shale dust and diesel exhaust. In: Inhalation Toxicology Research Institute Annual Report, DOE Research and Development Report LMF-115. Springfield, VA: National Technical Information Service, pp. 273–278. Mauderly JL, Jones RD, Griffith WC, Henderson RF, McClellan RO (1987a) Diesel exhaust is a pulmonary carcinogen in rats exposed chronically by inhalation. Fundam. Appl. Toxicol. 9:208– 221. Mauderly JL, Bice DE, Carpenter RL, Gillett NA, Hahn FF, Henderson RF, Pickrell JA, Wolff RK (1987b) Effect of inhaled NO2 and diesel exhaust on developing lung. Research Report No. 8, May 1987. Cambridge, MA: Health Effects Institute. Mauderly JL, Gillett NA, Henderson RF, Jones RD, McClellan RO (1988) Relationships of lung structural and functional changes to accumulation of diesel exhaust particles. Ann. Occup. Hyg. 32 (Suppl. 1):659–669. Mauderly JL, Bice DE, Cheng YS, Gillett NA, Griffith WC, Henderson RF, Pickrell JA, Wolff RK (1990a) Influence of pre-existing pulmonary emphysema on susceptibility to chronic inhalation exposure to diesel exhaust. Am. Rev. Respir. Dis. 141:1333–1341. Mauderly JL, Cheng YS, Snipes MB (1990b) Particle overload in toxicological studies: friend or foe? J. Aerosol Med. 3:S-169–S-187 Mauderly JL, Banas DA, Griffith WC, Hahn FF, Henderson RF, McClellan RO (1996) Diesel exhaust is not a pulmonary carcinogen in CD-1 mice exposed under conditions carcinogenic to F344 rats. Fundam. Appl. Toxicol. 30:233–242. McClellan RO (1986) Health effects of diesel exhaust: a case study in risk assessment. Am. Ind. Hyg. Assoc. J. 47(1):1–13. McDonald JD, Barr EB, White RK, Chow JC, Schauer JJ, Zielinska B, Grosjean E (2004a) Generation and characterization of four dilutions of diesel engine exhaust for a subchronic inhalation study. Environ. Sci. Technol. 38(9):2513–2522. McDonald JD, Harrod KS, Seagrave JC, Seilkop SK, Mauderly JL (2004b) Effects of low sulfur fuel and a catalyzed particle trap on the composition and toxicity of diesel emissions. Environ. Health Perspect. 112:1307–1312. McDonald J, Barr EB, White RK (2004c) Design, characterization, and evaluation of a small-scale diesel exhaust exposure system. Aerosol Sci. Technol. 38:62–78.

624

DIESEL EXHAUST

McDonald JD, Eide I, Seagrave JC, Zielinska B, Whitney K, Lawson DR, Mauderly JL (2004d) Relationship between composition and toxicity of engine emission samples. Environ. Health Perspect. 112:1527–1538. McDonald JD, White RK, Barr EB, Zielinska B, Chow JC, Grosjean E (2006a) Generation and characterization of hardwood smoke inhalation exposure atmospheres. Aerosol Sci. Technol. 40:573–584. McDonald JD, Seagrave JC, Mitchell L, Gigliotte APiMauderly JL (2006b) Pulmonary and systemic immune response to inhaled oil condensates. Proceedings of Conference on Diesel Engine Emissions Reduction, Detroit, MI, August 23, 2006. Available atwww.Eere.energy.gov/vehiclesandfuels/resources/conferences/deer/index.html. McDonald JD, Reed MD, Barrett EG, Campen MJ, Seagrave JC, Mauderly JL (2007) Health effects of inhaled gasoline engine emissions. Inhal. Toxicol. 19:107–116. McDonald JD, Barr EB, White RK, Kracko D, Chow JC, Zielinska B, Grosjean E (2008) Generation and characterization of gasoline engine exhaust inhalation exposure atmospheres. Inhal. Toxicol. 20:1157–1168. Merrion DF (2002)Heavy duty diesel emissions, fifty years: 1960 to 2010. In: Design, Application, Performance and Emissions of Modern Internal Combustion Engine Systems and Components, ICE, Vol. 39. American Society of Mechanical Engineers. pp.1–15. Miller RL, Garfinkel R, Horton M, Camann D, Perera FP, Whyatt RM, Kinney PL (2004) Polycyclic aromatic hydrocarbons, environmental tobacco smoke, and respiratory symptoms in an inner-city birth cohort. Chest 126(4):1071–1078. Mills NL, T€ornqvist H, Robinson SD, Gonzalez M, Darnley K, MacNee W, Boon NA, Donaldson K, Blomberg A, Sandstr€om T, Newby DE (2005) Diesel exhaust inhalation causes vascular dysfunction and impaired endogenous fibrinolysis. Circulation 112(25):3930–3936. Mills NL, T€ornqvist H, Gonzalez MC, Vink E, Robinson SD, S€oderberg S, Boon NA, Donaldson K, Sandstr€om T, Blomberg A, Newby DE (2007) Ischemic and thrombotic effects of dilute diesel exhaust inhalation in men with coronary artery disease. N. Engl. J. Med. 357 (11):1075– 1082. Monforton C (2006) Weight of evidence or wait for the future? Protecting underground miners from diesel particulate matter. Am. J. Pub. Health 96:271–276. Moorman WJ, Clark JC, Pepelko WE, Mattox J (1985) Pulmonary function responses in cats following long-term exposure to diesel exhaust. J. Appl. Toxicol. 5:301–305. Morimoto K, Kitamura M, Kondo H, Koizumie A (1986)Genotoxicity of diesel exhaust emissions in a battery of in-vitro short-term and in-vivo bioassays. In:Ishinishi N, Koizumi A, McClellan RO, St€ober W, editors. Carcinogenic and Mutagenic Effects of Diesel Engine Exhaust.Amsterdam: Elsevier, pp.85–101. Motor Vehicle Manufacturers Association (1962) Facts and Figures. Detroit:Motor Vehicle Manufacturers Association. Muranaka M, Suzuki S, Koizumi K, Takafuji S, Miyamoto T, Ikemori R, Tokiwa H (1986) Adjuvant activity of diesel-exhaust particulates for the production of IgE antibody in mice. J. Allergy Clin. Immunol. 77(4):616–623. Nafstad P, Haheim LL, Oftedal B, Gram F, Holme I, Hjermann I, Leren P (2003) Lung cancer and air pollution: a 27 year follow up of 16 209 Norwegian men. Thorax 58(12):1071–1076. Nafstad P, Haheim LL, Wisloff T, Gram F, Oftedal B, Holme I, Hjermann I, Leren P (2004) Urban air pollution and mortality in a cohort of Norwegian men. Environ. Health Perspect. 112 (5):610– 615. Nemmar A, Hoet PH, Vanquickenborne B, Dinsdale D, Thomeer M, Hoylaerts MF, Vanbilloen H, Mortelmans L, Nemery B (2002) Passage of inhaled particles into the blood circulation in humans. Circulation 105(4):411–414.

REFERENCES

625

Nicolai T, Carr D, Weiland SK, Duhme H, Von Ehrenstein O, Wagner C, Von Mutius E (2003) Urban traffic and pollutant exposure related to respiratory outcomes and atopy in a large sample of children. Eur. Respir. J. 21(6):956–963 Nightingale JA, Maggs R, Cullinan P, Donnelly LE, Rogers DF, Kinnersley R, Chung KF, Barnes PJ, Ashmore M, Newman-Taylor A (2000) Airway inflammation after controlled exposure to diesel exhaust particulates. Am. J. Respir. Crit. Care Med. 162(1):161–166. Nikula KJ, Snipes MB, Barr EB, Griffith WC, Henderson RF, Mauderly JL (1995) Comparative pulmonary toxicities and carcinogenicities of chronically inhaled diesel exhaust and carbon black in F344 rats. Fundam. Appl. Toxicol. 25:80–94. Nikula KJ, Griffith WC, Avila KJ, Mauderly JL (1997) Lung tissue responses and site of particle retention differ between rats and cynomolgus monkeys exposed chronically to diesel exhaust and coal dust. Fundam. Appl. Toxicol. 37:37–53. NIOSH (National Institute of Occupational Safety and Health). (1988) Carcinogenic Effects of Exposure to Diesel Exhaust. NIOSH Current Intelligence Bulletin 50, DHHS Publication 88–116. Atlanta, GA: Centers for Disease Control. Nitta H, Sato T, Nakai S, Maeda K, Aoki S, Ono M (1993) Respiratory health associated with exposure to automobile exhaust. I. Results of cross-sectional studies in 1979, 1982, and 1983. Arch. Environ. Health 48(1):53–58. Norbeck JM, Truex TD (1998) Evaluation of factors that affect diesel toxicity. Final Report for Contract No. 94-312. Sacramento, CA: California Air Resources Board. Nordenhall C, Pourazar J, Blomberg A, Levin JO, Sandstrom T, Adelroth E (2000) Airway inflammation following exposure to diesel exhaust: a study of time kinetics using induced sputum. Eur. Respir. J. 15(6):1046–1051. Nordenhall C, Pourazar J, Ledin MC, Levin JO, Sandstrom T, Adelroth E (2001) Diesel exhaust enhances airway responsiveness in asthmatic subjects. Eur. Respir. J. 17(5):909–915. NTP (National Toxicology Program) (1998) Draft report on carcinogens: background document for diesel exhaust particulates. Research Triangle Park, NC: U.S. Department of Health and Human Services (NTP). Nyberg F, Gustavsson P, Jarup L, Bellander T, Berglind N, Jakobsson R, Pershagen G (2000) Urban air pollution and lung cancer in Stockholm. Epidemiology 11(5):487–495. Oberd€orster G, Oberd€orster E, Oberd€orster J (2005) Nanotoxicology: an emerging discipline evolving from studies of ultrafine particles. Environ. Health Perspect. 113:823–839. Oosterlee A, Drijver M, Lebret E, Brunekreef B (1996) Chronic respiratory symptoms in children and adults living along streets with high traffic density. Occup. Environ. Med. 53 (4):241–247. Opresko DM, Holleman JW, Ross RH, Carroll JW (1984) Problem definition study on emission byproduct hazards from diesel engines for confined space army workplaces. ORNL Report No. 6017. Oak Ridge, TN: Oak Ridge National Laboratory. Pandya RJ, Solomon G, Kinner A, Balmes JR (2002) Diesel exhaust and asthma: hypotheses and molecular mechanisms of action. Environ. Health Perspect. 110(Suppl. 1):103–112. Pearson TA, Mensah GA, Alexander RW, Anderson JL, Cannon RO 3rd, Criqui M, Fadi YY, Fortmann SP, Hong Y, Myers GL, Rifai N, Smith SC Jr, Taubert K, Tracy RP, Vinicor F, Centers for Disease Control and Prevention; American Heart Association. (2003) Markers of inflammation and cardiovascular disease: application to clinical and public health practice: a statement for healthcare professionals from the Centers for Disease Control and Prevention and the American Heart Association. Circulation 107(3):499–511. Pekkanen J, Peters A, Hoek G, Tittanen P, Brunekreef B, De Hartog J, Heinrich J, Ibald-Mulli A, Kreyling WG, Lanki T, Timonen KL, Vanninen E (2002) Particulate air pollution and risk of ST-segment depression during repeated submaximal exercise tests among subjects with coronary

626

DIESEL EXHAUST

heart disease: the exposure and risk assessment for fine and ultrafine particles in ambient air (ULTRA) study. Circulation 106(8):933–938. Pepelko WE, Peirano WB (1983) Health effects of exposure to diesel engine emissions. A summary of animal studies conducted by the US Environmental Protection Agency’s Health Effects Research Laboratory at Cincinnati, OH. J. Am. Coll. Toxicol. 2:253–306. Pepelko WE, Chen C (1993) Quantitative assessment of cancer risk from exposure to diesel engine emissions. Reg. Toxicol. Pharmacol. 17:52–65. Perera FP, Tang D, Rauh V, Lester K, Tsai WY, Tu YH, Weiss L, Hoepner L, King J, Del Priore G, Lederman SA (2005) Relationships among polycyclic aromatic hydrocarbon–DNA adducts, proximity to the World Trade Center, and effects on fetal growth. Environ. Health Perspect. 113(8):1062–1067. Perera FP, Rauh V, Whyatt RM, Tsai WY, Tang D, Diaz D, Hoepner L, Barr D, Tu YH, Camann D, Kinney P (2006) Effect of prenatal exposure to airborne polycyclic aromatic hydrocarbons on neurodevelopment in the first 3 years of life among inner-city children. Environ. Health Perspect. 114(8):1287–1292. Peters A, Liu E, Verrier RL, Schwartz J, Gold DR, Mittleman M, Baliff J, Oh JA, Allen G, Monahan K, Dockery DW (2000) Air pollution and incidence of cardiac arrhythmia. Epidemiology 11(1):11– 17. Peters A, Frohlich M, Doring A, Immervoll T, Wichmann HE, Hutchinson WL, Pepys MB, Koenig W (2001) Particulate air pollution is associated with an acute phase response in men; results from the MONICA-Augsburg Study. Eur. Heart J. 22(14):1198–1204. Peters A, Von Klot S, Heier M, Trentinaglia I, Hormann A, Wichmann HE, Lowel H, Cooperative Health Research in the Region of Augsburg Study Group (2004) Exposure to traffic and the onset of myocardial infarction. N. Engl. J. Med. 351(17):1721–1730. Plopper CG, Hyde DM, Weir AJ (1983) Centriacinar alterations in lungs of cats chronically exposed to diesel exhaust. Lab. Invest. 49:391–399. Pope CA 3rd, Burnett RT, Thun MJ, Calle EE, Krewski D, Ito K, Thurston GD (2002) Lung cancer, cardiopulmonary mortality, and long-term exposure to fine particulate air pollution. JAMA 287 (9):1132–1141. Pope CA 3rd, Burnett RT, Thurston GD, Thun MJ, Calle EE, Krewski D, Godleski JJ (2004) Cardiovascular mortality and long-term exposure to particulate air pollution: epidemiological evidence of general pathophysiological pathways of disease. Circulation 109(1):71–77. Pourazar J, Frew AJ, Blomberg A, Helleday R, Kelly FJ, Wilson S, Sandstrom T (2004) Diesel exhaust exposure enhances the expression of IL-13 in the bronchial epithelium of healthy subjects. Respir. Med. 98(9):821–825. Purdam JT, Holness DL, Pilger CW (1987) Environmental and medical assessment of stevedores employed in ferry operations. Appl. Ind. Hyg. 2:133–139. Randerath K, Putman KL, Mauderly JL, Williams PL, Randerath E (1995) Pulmonary toxicity of inhaled diesel exhaust and carbon black in chronically exposed rats: II DNA adducts. Research Report No. 68. Cambridge, MA: Health Effects Research Institute. Reed MD, Gigliotti AP, McDonald JD, Seagrave JC, Seilkop SK, Mauderly JL (2004) Health Effects of Subchronic Exposure to Environmental Levels of Diesel Exhaust. Inhal. Toxicol. 16:177–193. Reed MD, Blair LF, Burling K, Daly I, Gigliotti AP, Gudi R, Mercieca MD, McDonald JD, Naas DJ, O’Callaghan JP, Seilkop SK, Ronsko NL, Wagner VO, Kraska RC (2005) Health effects of subchronic inhalation of diesel–water emulsion emissions. Inhal. Toxicol. 17(14):851–870. Reed MD, Blair LF, Burling K, Daly I, Gigliotti AP, Gudi R, Mercieca MD, McDonald JD, O’Callaghan JP, Seilkop SK, Ronsko NL, Wagner VO, Kraska RC (2006a) Health effects of subchronic exposure to diesel–water–methanol emulsion emissions. Toxicol. Ind. Health 22 (2):65–85.

REFERENCES

627

Reed MD, Berger JA (2006b) Real time RT-PCR assessment of clearance of respiratory syncytial virus altered by exposure to diesel exhaust and hardwood smoke. Toxicologist 90:232. Reger R, Hancock J, Hankinson J, Hearl F, Merchant J (1982) Coal miners exposed to diesel exhaust emissions. Ann. Occup. Hyg. 26:799–815. Rich DQ, Schwartz J, Mittleman MA, Link M, Luttmann-Gibson H, Catalano PJ, Speizer FE, Dockery DW (2005) Association of short-term ambient air pollution concentrations and ventricular arrhythmias. Am. J. Epidemiol. 161(12):1123–1132. Rich DQ, Kim MH, Turner JR, Mittleman MA, Schwartz J, Catalano PJ, Dockery DW (2006a) Association of ventricular arrhythmias detected by implantable cardioverter defibrillator and ambient air pollutants in the Saint Louis, Missouri metropolitan area. Occup. Environ. Med. 63 (9):591–596. Rich DQ, Mittleman MA, Link MS, Schwartz J, Luttmann-Gibson H, Catalano PJ, Speizer FE, Gold DR, Dockery DW (2006b) Increased risk of paroxysmal atrial fibrillation episodes associated with acute increases in ambient air pollution. Environ. Health Perspect. 114(1):120–123. Riediker M, Cascio WE, Griggs TR, Herbst MC, Bromberg PA, Neas L, Williams RW, Devlin RB (2004) Particulate matter exposure in cars is associated with cardiovascular effects in healthy young men. Am. J. Respir. Crit. Care Med. 169(8):934–940. Riedl M, Diaz-Sanchez D (2005) Biology of diesel exhaust effects on respiratory function. J. Allergy Clin. Immunol. 115(2):221–228. Rosenlund M, Berglind N, Pershagen G, Hallqvist J, Jonson T, Bellander T (2006) Long-term exposure to urban air pollution and myocardial infarction. Epidemiology 17(4):383–390. Ross R (1999) Atherosclerosis—an inflammatory disease. N. Engl. J. Med. 340(2):115–126. Ruckerl R, Ibald-Mulli A, Koenig W, Schneider A, Woelke G, Cyrys J, Heinrich J, Marder V, Frampton M, Wichmann HE, Peters A (2006) Air pollution and markers of inflammation and coagulation in patients with coronary heart disease. Am. J. Respir. Crit. Care Med. 173(4):432–441. Rudell B, Ledin MC, Hammarstrom U, Stjernberg N, Lundback B, Sandstrom T (1996) Effects on symptoms and lung function in humans experimentally exposed to diesel exhaust. Occup. Environ. Med. 53(10):658–662. Rudell B, Blomberg A, Helleday R, Ledin MC, Lundback B, Stjernberg N, Horstedt P, Sandstrom T (1999) Bronchoalveolar inflammation after exposure to diesel exhaust: comparison between unfiltered and particle trap filtered exhaust. Occup. Environ. Med. 56(8):527–534. Salvi S, Blomberg A, Rudell B, Kelly F, Sandstrom T, Holgate ST, Frew A (1999) Acute inflammatory responses in the airways and peripheral blood after short-term exposure to diesel exhaust in healthy human volunteers. Am. J. Respir. Crit. Care. Med. 159(3):702–709. Salvi SS, Nordenhall C, Blomberg A, Rudell B, Pourazar J, Kelly FJ, Wilson S, Sandstrom T, Holgate ST, Frew AJ (2000) Acute exposure to diesel exhaust increases IL-8 and GRO-alpha production in healthy human airways. Am. J. Respir. Crit. Care Med. 161(2 Pt. 1):550–557. Sarnat SE, Suh H, Coull BA, Schwartz J, Stone PH, Gold DR (2006) Ambient particulate air pollution and cardiac arrhythmia in a panel of older adults in Steubenville, Ohio. Occup. Environ. Med. 63 (10):700–706. Saverin R, Braunlich A, Dahmann D, Enderlein G, Heuchert G (1999) Diesel exhaust and lung cancer mortality in potash mining. Am. J. Ind. Med. 36(4):415–422. Schauer JJ, Kleeman MJ, Cass GR, Simoneit BRT (1999) Measurement of emissions from air pollution sources. 2. C1 through C30 organic compounds from medium duty diesel trucks. Environ. Sci. Technol. 33:1578–1587. Scheepers PT, Bos RP (1992) Combustion of diesel fuel from a toxicological perspective. I. Origin of incomplete combustion products. Int. Arch. Occup. Health 64:149–161. Schuetzle D (1983) Sampling of vehicle emissions for chemical analysis and biological testing. Environ. Health Perspect. 47:65–80.

628

DIESEL EXHAUST

Schuetzle D, Lewtas J (1986) Bioassay-directed chemical analysis in environmental research. Anal. Chem. 58:1060A–1075A. Schwartz J (2001) Air pollution and blood markers of cardiovascular risk. Environ. Health Perspect. 109(Suppl. 3):405–409. Schwartz J, Litonjua A, Suh H, Verrier M, Zanobetti A, Syring M, Nearing B, Verrier R, Stone P, MacCallum G, Speizer FE, Gold DR (2005) Traffic related pollution and heart rate variability in a panel of elderly subjects. Thorax 60(6):455–461. Seagrave JC, Berger J, Zielinska B, Sagabiel J, Rogers CF, McDonald J, Mauderly JL (2001) Comparative acute toxicities of particulate matter (PM) and semi-volatile organic compound (SVOC) fractions of traffic tunnel air. Toxicologist 60(1):192. Seagrave JC, McDonald JD, Gigliotti AP, Nikula KJ, Seilkop SK, Gurevich M, Mauderly JL (2002) Mutagenicity and in vivo toxicity of combined particulate and semi-volatile organic fractions of gasoline and diesel engine emissions. Toxicol. Sci. 70:212–226. Seagrave JC, Seilkop SK, Mauderly JL (2003) In Vitro relative toxicity screening of combined particulate and semi-volatile organic fractions of gasoline and diesel engine emissions. J. Toxicol. Environ. Health 66:1113–1132. Seagrave JC, Gigliotti A, McDonald JD, Seilkop SK, Whitney KA, Zielinska B, Mauderly JL (2005) Composition, toxicity, and mutagenicity of particulate and semivolatile emissions from heavyduty compressed natural gas-powered vehicles. Toxicol. Sci. 87:232–241. Seaton A, MacNee W, Donaldson K, Godden D (1995) Particulate air pollution and acute health effects. Lancet 345(8943):176–178. Smith TJ, Davis ME, Reaser P, Natkin J, Hart JE, Laden F, Heff A, Garshick E (2006) Overview of particulate exposures in the US trucking industry. J. Environ. Monitor. 8:711–720. Soll-Johanning H, Bach E, Olsen JH, Tuchsen F (1998) Cancer incidence in urban bus drivers and tramway employees: a retrospective cohort study. Occup. Environ. Med. 55(9):594–598. Soll-Johanning H, Bach E, Jensen SS (2003) Lung and bladder cancer among Danish urban bus drivers and tramway employees: a nested case–control study. National Institute of Occupational Health, Lerso Parkalle 105, DK-2100 Copenhagen, Denmark. Occup. Med. (Lond). 5 (1):25–33. Steenland NK, Silverman DT, Hornung RW (1990) Case–control study of lung cancer and truck driving in the Teamsters’ Union. Am. J. Public Health 80:670–674. Steenland K, Silverman D, Zaebst D (1992) Exposure to diesel exhaust in the trucking industry and possible relationships with lung cancer. Am. J. Ind. Med. 21(6):887–890. Stenfors N, Nordenhall C, Salvi SS, Mudway I, Soderberg M, Blomberg A, Helleday R, Levin JO, Holgate ST, Kelly FJ, Frew AJ, Sandstrom T (2004) Different airway inflammatory responses in asthmatic and healthy humans exposed to diesel. Eur. Respir. J. 23(1):82–86. Stinn W, Teredesai A, Anskeit E, Rustemeier K, Schepers G, Schnell P, Haussmann H-J, Carchman RA, Coggins CRE, Reininghaus W (2005) Chronic nose-only inhalation study in rats, comparing room-aged sidestream cigarette smoke and diesel engine exhaust. Inhal. Toxicol. 17:549–576. Sun JD, Wolff RK, Kanapilly GM, McClellan RO (1984) Lung retention and metabolic fate of inhaled benzo(a)pyrene associated with diesel exhaust particles. Toxicol. Appl. Pharmacol. 73:48–59. Sun Q, Wang A, Jin X, Natanzon A, Duquaine D, Brook RD, Aguinaldo J-GSFayad ZA, Fuster V, Lippmann M, Chen LC, Rajagopalan S (2005) Long-term air pollution exposure and acceleration of atherosclerosis and vascular inflammation in an animal model. J. Am. Med. Assoc. 294 (23):3003–3010. Suwa T, Hogg JC, Quinlan KB, Ohgami A, Vincent R, Van Eeden SF (2002) Particulate air pollution induces progression of atherosclerosis. J. Am. Coll. Cardiol. 39(6):935–942.

REFERENCES

629

Swanson GM, Lin C-S, Burns PB (1993) Diversity in the association between occupation and lung cancer among black and white men. Cancer Epidemiol. Biomarkers Prev. 2:313–320. Takafuji S, Suzuki S, Koizumi K, Tadokoro K, Miyamoto T, Ikemori R, Muranaka M (1987) Dieselexhaust particulates inoculated by the intranasal route have an adjuvant activity for IgE production in mice. J. Allergy Clin. Immunol. 79(4):639–645. Takanaka H, Zhang K, Diaz-Sanchez D, Tsien A, Saxon A (1995) Enhanced human IgE production results from exposure to the aromatic hydrocarbons from diesel exhaust: direct effects on B-cell IgE production. J. Allergy Clin. Immunol. 95(1 Pt 1):103–115. Takano H, Ichinose T, Miyabara Y, Shibuya T, Lim H-B, Yoshikawa T, Sagai M (1998) Inhalation of diesel exhaust enhances allergen-related eosinophil recruitment and airway hyperresponsiveness in mice. Toxicol. Appl. Pharmacol. 150:328–337. Takemoto K, Yoshimura H, Katayama H (1986)Effects of chronic inhalation exposure to diesel exhaust on the development of lung tumors in di-isopropanol-nitrosamine-treated F344 rats and newborn C57BL and ICR mice. In:Ishinishi N, Koizumi A, McClellan RO, St€ober W, editors. Carcinogenic and Mutagenic Effects of Diesel Engine Exhaust.Amsterdam:Elsevier, pp.311–327. Tobias HJ, Beving DE, Ziemann PJ, Sakurai H, Zuk M, McMurry PH, Zarling D, Waytulonis R, Kittelson DB (2001) Chemical analysis of diesel engine nanoparticles using a nano-EMA/thermal desorption particle beam mass spectrometer. Environ. Sci. Technol. 35:2233–2243. Toda N, Tsukue N, Tsubone H, Sagai M, Biurmachi J, Suzuki AK (2001) Effects of diesel exhaust particles on blood pressure in rats. J. Toxicol. Environ. Health A 63(6):429–435. T€ornqvist H, Mills NL, Gonzalez M, Miller MR, Robinson SD, Megson IL, MacNee W, Donaldson K, S€oderberg S, Newby DE, Sandstr€omBlomberg A (2007) Persistent endothelial dysfunction in humans after diesel exhaust inhalation. Am. J. Respir. Crit. Care Med. 176:395–400. Tsien A, Diaz-Sanchez D, Ma J, Saxon A (1997) The organic component of diesel exhaust particles and phenanthrene, a major polyaromatic hydrocarbon constituent, enhances IgE production by IgEsecreting EBV-transformed human B cells in vitro. Toxicol. Appl. Pharmacol. 142(2):256–263. Tsukue N, Toda N, Tsubone H, Sagai M, Jin WZ, Watanabe G, Taya K, Biurmachi J, Suzuki AK (2001) Diesel exhaust (DE) affects the regulation of testicular function in male Fischer 344 rats. J. Toxicol. Environ. Health A 63(2):115–126. Tuchsen F, Endahl LA (1999) Increasing inequality in ischaemic heart disease morbidity among employed men in Denmark 1981–1993: the need for a new preventive policy. Int. J. Epidemiol. 28 (4):640–644. Ulfvarson U, Alexandersson R (1990) Reduction in adverse effect on pulmonary function after exposure to filtered diesel exhaust. Am. J. Ind. Med. 17(3):341–347. Ulfvarson U, Alexandersson R, Aringer L, Svensson E, Hedenstierna G, Hogstedt C, Holmberg B, Rosen G, Sorsa M (1987) Effect of exposure to vehicle exhaust on health. Scand. J. Work Environ. Health 13:505–512. U.S. Department of Labor, Bureau of Labor Statistics (1972) Railroad Technology and Manpower in the 1970s.Washington, DC:USGPO. Valberg PA, Crouch EA (1999) Meta-analysis of rat lung tumors from lifetime inhalation of diesel exhaust. Environ. Health Perspect. 107(9):693–699. van Eeden SF, Tan WC, Suwa T, Mukae H, Terashima T, Fujii T, Qui D, Vincent R, Hogg JC (2001) Cytokines involved in the systemic inflammatory response induced by exposure to particulate matter air pollutants (PM10). Am. J. Respir. Crit. Care. Med. 164(5):826–830. Van Vliet P, Knape M, De Hartog J, Janssen N, Harssema H, Brunekreef B (1997) Motor vehicle exhaust and chronic respiratory symptoms in children living near freeways. Environ. Res. 74 (2):122–132. Venn AJ, Lewis SA, Cooper M, Hubbard R, Britton J (2001) Living near a main road and the risk of wheezing illness in children. Am. J. Respir. Crit. Care Med. 164(12):2177–2180.

630

DIESEL EXHAUST

Vinegar A, Carson AI, Pepelko WE, Orthoefer JG (1981) Effects of six months of exposure to two levels of diesel exhaust on pulmonary function of Chinese hamsters (abstract). Feditor. Proc. 40:593. Visser O, van Wijnen JH, van Leeuwen FE (2004) Residential traffic density and cancer incidence in Amsterdam, 1989–1997. Cancer Causes Control. 15(4):331–339 Wade JF III, Newman LS (1993) Diesel asthma: reactive airways disease following overexposure to locomotive exhaust. J. Occup. Med. 35(2):149–154. Watanabe N (1998) Inhalation of diesel exhaust for twelve months retarded the increase in bone mass in rats. Connect. Tissue 30:201–206. Watanabe N (2005) Decreased number of sperms and Sertoli cells in mature rats exposed to diesel exhaust as fetuses. Toxicol. Lett. 155:51–58. Watanabe N, Nakamura T (1996) Inhalation of diesel engine exhaust increases bone mineral concentrations in growing rats. Arch. Environ. Contam. Toxicol. 3:407–411. Watanabe N, Oonuki Y (1999) Inhalation of diesel engine exhaust affects spermatogenesis in growing male rats. Environ. Health Perspect. 107:539–544. Watanabe N, Kurita M (2001) The masculinization of the fetus during pregnancy due to inhalation of diesel exhaust. Environ. Health Perspect. 109:111–119. Watanabe N, Ohsawa M (2002) Elevated serum immunoglobulin E to Cryptomeria japonica pollen in rats exposed to diesel exhaust during fetal and neonatal periods. BMC Pregnancy Childbirth 2: 2–11. Weiland SK, Mundt KA, Ruckmann A, Keil U (1994) Self-reported wheezing and allergic rhinitis in children and traffic density on street of residence. Ann. Epidemiol. 4(3):243–247. White H, Vostal JJ, Kaplan HL, MacKenzie WF (1983) A long-term inhalation study evaluates the pulmonary effects of diesel emissions. J. Appl. Toxicol. 3:332. Williams RL (1982)Diesel particulate emissions: composition, concentration, and control. In: Lewtas J, editor. Toxicological Effects of Emissions from Diesel Engines. New York: Elsevier, pp.15–32. Wolff RK, Henderson RF, Snipes MB, Griffith WC, Mauderly JL, Cuddihy RG, McClellan RO (1987) Alterations in particle accumulation and clearance in lungs of rats chronically exposed to diesel exhaust. Fundam. Appl. Toxicol. 9:154–166. Wolff RK, Bond JA, Henderson RE, Harkema JR, Mauderly JL (1990) Pulmonary inflammation and DNA adducts in rats inhaling diesel exhaust or carbon black. Inhal. Toxicol. 2:241–254. Wong O, Morgan RW, Kheifets L, Larson SR, Whorton MD (1985) Mortality among members of a heavy construction equipment operators union with potential exposure to diesel emissions. Br. J. Ind. Med. 42:435–448. Wong D, Mitchell CE, Wolff RK, Mauderly JL, Jeffrey AM (1986) Identification of DNA damage as a result of exposure of rats to diesel engine exhaust. Carcinogenesis 7:1595–1597 Woodruff TJ, Parker JD, Schoendorf K (2006) Fine particulate matter (PM2.5) air pollution and selected causes of postneonatal infant mortality in California. Environ. Health Perspect. 114 (5):786–790. Woskie SR, Smith TJ, Hammond SK, Shenker MB, Garshick E, Speizer FE (1988a) Estimation of the diesel exhaust exposures of railroad workers. I. Current exposures. Am. J. Ind. Med. 13:381– 394. Woskie SR, Smith TJ, Hammond SK, Shenker MB, Garshick E, Speizer FE (1988b) Estimation of the diesel exhaust exposures of railroad workers. II. National and historical exposures. Am. J. Ind. Med. 15:395–404. Yang H-M, Antonini JM, Barger MW, Butterworth L, Roberts JR, Ma JKH, Castranova V, Ma JYC (2001) Diesel exhaust particles suppress macrophage function and slow the pulmonary clearance of Listeria monocytogenes in rats. Environ. Health Perspect. 109:515–521.

REFERENCES

631

Yang H-M, Butterworth L, Munson AE, Jean Meade B (2003) Respiratory exposure to diesel exhaust particles decreases the spleen IgM response to a T cell-dependent antigen in female B6C3F1 mice. Toxicol. Sci. 71:207–216. Zaebst DD, Clapp DE, Blade LM, Marlow DA, Steenland K, Hornung RW, Scheutzle D, Butler J (1991) Quantitative determination of trucking industry workers’ exposures to diesel exhaust particles. Am. Ind. Hyg. Assoc. J. 52(12):529–541. Zielinska B (2005) Atmospheric transformation of diesel emissions. Exp. Toxicol. Pathol. 57(Suppl. 1):31–42. Zielinska B, Sagebiel J, McDonald J, Whitney K, Lawson D (2004) Emission rates and comparative chemical composition from selected in-use diesel and gasoline-fueled vehicles. J. Air Waste Manage. Assoc.

17 DIOXINS AND DIOXIN-LIKE CHEMICALS Michael A. Gallo and Morton Lippmann

17.1 INTRODUCTION Dioxins, a subset of the polyhalogenated aromatic hydrocarbons (PHAHs), are ubiquitous in the environment at low concentrations. At much higher concentrations, they are known to produce a variety of adverse health effects in laboratory animals, and interspecies modeling suggests, to some public health professionals, that they pose exceptionally high health risks to people exposed to background levels. The extents of the public health risks have been the subject of a series of risk assessments conducted by the United States Environmental Protection Agency (EPA) since the mid-1980s (NRC, 2006), but EPA has not yet issued a formal risk assessment. In 1995, and again in 2001, EPA’s Science Advisory Board (SAB) issued reports that evaluated draft EPA risk assessments for dioxin and dioxin-like compounds but did not endorse them as being scientifically valid (SAB, 1995, 2001). The most recent EPA draft reassessment was reviewed by a National Research Council (NRC) committee, which also failed to fully endorse its methodology and its quantitative estimates of public health risks (NRC, 2006). This chapter includes an update of the chapter authored by Michael J. DeVito and Michael A. Gallo for the second edition of this book on the science of dioxin and dioxin-like chemicals (DLCs), along with a summary of the findings and conclusions of the NRC Committee and a review of some recent papers on the associations of exposures to dioxin and DLCs in human populations and health effects. As noted above, dioxins and DLCs are a subset of the PHAHs. The most well studied and the most toxic of the class is 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD). These chemicals are present in a variety of environmental media as well as low-level contaminants of the food supply. The dioxin-like PHAHs consist in part of the polyhalogenated dibenzo-p-dioxin, dibenzofurans, and biphenyls (PCBs and PBBs). These chemicals are sparingly soluble in water and are highly lipophilic. They are persistent in both environmental and biological samples, with half-lives in humans ranging from 1 to over 20 years (Flesch-Janys et al.,

Environmental Toxicants, Third Edition Edited by Morton Lippmann Copyright
2009 John Wiley & Sons, Inc.

633

634

DIOXINS AND DIOXIN-LIKE CHEMICALS

1995). The dioxin-like chemicals induce similar toxicities in experimental animals. These toxicities are initiated, in part, by the binding and activation of these chemicals to an intracellular protein called the aryl hydrocarbon receptor (AhR) The health effects of dioxin and related chemicals in humans and wildlife are hotly debated with the intensity of the debate based as much on the uncertainties in risk assessment as it is for political, social, and economic reasons. Despite being one of the most studied class of chemicals, risk assessments for dioxin by various governmental health and regulatory agencies throughout the world have resulted in tolerable daily intakes or virtually safe doses that range over almost three orders of magnitude. This large discrepancy is due to the use of either threshold or linear models in cancer risk estimates. The political and social issues revolve, in part, around the widespread background exposure to the “most toxic man-made chemical” as well as to the use of dioxin-contaminated herbicides during the Vietnam War. In addition, there have been a number of industrial accidents that resulted in human and wildlife exposure to these chemicals, and attempts to resolve the exposures and effects of these accidents have been widely criticized from all sides. The debate on the health effects of dioxin and related chemicals is not likely to be resolved in the near future.

17.2 SOURCES One of the problems in estimating potential health effects of dioxin is the coexposure to numerous dioxin-like chemicals originating from different sources. There are 75 different polychlorinated dibenzo-p-dioxin (CDDs), 135 dibenzofurans (CDFs), and 209 biphenyls (PCBs), depending upon the number and position of the chlorine substitutions. Fortunately, only a subset of these chemicals produces dioxin-like toxicities in experimental animals. The 2,3,7,8-substituted CDDs and CDFs are considered dioxin-like. Only 7 of the 75 CDDs and 10 of the 135 CDFs are dioxin-like (Van den Berg et al., 1998). Of the 209 PCB congeners, only 12 have dioxin-like activity (Safe 1994; Van den Berg et al., 1998). In addition, brominated analogues have been found in the environment as well as in human tissues. The data available on the brominated analogues demonstrates that these compounds induce dioxin-like biochemical effects, toxicities, and similar structure–activity relationships with their chlorinated analogues (Safe 1990). Because dioxins consist of a broad class of chemicals with varying potencies, sources, and exposures, health risk assessments have used the toxic equivalency factor (TEF) methodology. This methodology is a relative potency scheme that compares the toxicity of all dioxins to the most potent, that is, TCDD). Congeners are assigned relative potency values called TEFs (Van den Berg et al., 1998). The concentration of a congener in a mixture is multiplied by its TEF, and this product is the TCDD or toxic equivalents (TEQ) for that congener. The TEQs for all chemicals in the mixture are summed to produce a TEQ for the entire mixture. It is assumed that the mixture will behave as if it contained the TEQ concentration of TCDD alone. At present, only 29 chemicals are considered in the TEF scheme, and they consist of the 2,3,7,8-substituted CDDs/CDFs and 12 out of 209 PCBs (Van den Berg et al., 1998). TEF values for the brominated analogues have not been adopted by regulatory agencies. The development of this methodology is described in greater detail later in this chapter. The CDDs and CDFs are unwanted products of several industrial processes. Combustion systems are a primary source for the production of CDDs and CDFs. Included in this category are waste incinerators, such as municipal solid waste, medical waste, sewage sludge, and hazardous waste incinerators. The burning of fuel, such as coal, wood, and petroleum, also

SOURCES

635

produce CDDs and CDFs. Other high-temperature sources, such as cement kilns, also produce significant amounts of dioxins. Iron ore sintering, steel production, and scrap metal recovery operations also produce and release CDDs/CDFs into the air. In some parts of the United States, open burning of household trash is a common practice, and it results in the formation of CDDs/CDFs. The contribution of open burning of household trash to total emissions of dioxin into the environment cannot be quantified with any certainty (USEPA, 1997). Several chemical manufacturing processes result in the formation of CDDs/CDFs as by-products, such as the production of herbicides, such as Agent Orange or 2,4,5-trichlorophenoxy acetic acid, and chlorinated phenols such as pentachlorophenol. These processes have either been altered to eliminate dioxin production or the products have been discontinued. The manufacture of chlorine-bleached wood pulp results in the production of trace quantities of CDDs/CDFs, of which the octa- and heptachlorinated congeners predominate. It should be noted that many pulp and paper mills have reengineered their processes and have decreased CDD/CDF production and emissions in these facilities by approximately 90% (Cleverly et al., 1998). The production of ethylene dichloride or vinyl chloride produces CDDs/CDFs; however, the data are insufficient to quantify emission estimates for these processes. Several natural processes can result in the production of CDDs/CDFs. The predominant congeners are hepta- and octachlorodibenzo-p-dioxin produced in forest fires. Under certain environmental conditions, such as composting, microorganisms may produce CDDs/CDFs from chlorinated phenolic compounds. Ball clay deposits in western MS, KY, and TN were found to contain CDDs and CDFs. The CDDs/CDFs in these clay deposits were approximately 90% by weight 2,3,7,8-tetrachlorodibenzo-p-dioxin and 1,2,3,7,8-pentachlorodibenzo-p-dioxin. While this congener pattern is similar to that found in contaminated herbicides, the origin of these chemicals in the clay has not been determined and natural occurrence is but one possibility. Another source of CDDs/CDFs is redistribution and circulation of reservoirs of previously released CDDs/CDFs (USEPA, 1994). Contaminated sediments, soils, and pentachlorophenol-treated wood are also considered reservoirs. Retrainment of these reservoirs may contribute significantly to overall exposure, but its exact contribution is uncertain. While there are natural sources for CDDs/CDFs production, several lines of evidence indicate that recent emissions and exposures are due to anthropogenic sources. Analysis of sediment cores in the United States and Europe shows consistent patterns of CDD/CDF concentration changes over time. CDD/CDF concentrations in sediment tend to increase starting around the 1920s–1930s, peaking between the 1960s and the 1970s, and then beginning to decrease in the later 1970s through the 1980s (Czuczwa et al., 1985; Cleverly et al., 1996). The increasing trends are consistent with the increase in general industrial activity in the 1930s, and the current decreasing trends are attributable to the promulgation of environmental regulations since the 1970s (Czuczwa et al., 1985). CDD/CDF concentrations are higher in human tissues from industrialized countries compared to those from underdeveloped nations (Schecter et al., 1994b, 1994c). There are also data comparing present human tissue concentrations with those tissues taken from preserved 140–400-year-old human remains that show almost the complete absence of CDD/CDFs compared to tissues of modern man (Schecter et al., 1994b, 1994c). Finally, no known large natural sources of CDD/CDFs have been identified. Estimates of all emission sources suggest that forest fires are a minor source compared to anthropogenic sources (Cleverly et al., 1998).

636

DIOXINS AND DIOXIN-LIKE CHEMICALS

The EPA developed an inventory of CDDs/CDFs sources for the years 1987 and 1995 (Cleverly et al., 1998). The year 1987 was chosen because prior to this year few potential sources in the United States had been characterized, and 1995 was chosen as the latest year for which significant data were available. This analysis indicates that the dominant source of dioxin releases to air in the United States is combustion. The estimate of releases in 1987 was 12 kg TEQ (range 5–30 kg) and from 1995 was 3 kg TEQ (range 1–8 kg). While there was uncertainty in these estimates, it was clear that emissions significantly decreased between 1987 and 1995. The reductions in emissions were due primarily to reductions in air emissions from municipal and medical waste incinerators (Cleverly et al., 1998). An updated EPA inventory for releases into the environment of PCDDs, PCDFs, and TCDD-like PCBs for the United States for 2000 was published in 2005 and was included in the NRC (2006) review. There was a reduction of 89% from a 1987 best estimate. Some dioxin-like chemicals were synthesized and sold commercially, such as the polychlorinated and polybrominated biphenyls, PCBs and PBBs, respectively. The PCBs were used as heat transfer fluids; flame-retardants; paint additives; and dielectric fluids for capacitors, transformers, and in several other industrial processes (Devoogt and Brinkman, 1989). The PBBs were used predominantly as flame-retardants in the early 1970s. Since 1929, approximately 1.5 million metric tons of PCBs were produced and sold. Numerous commercial mixtures of PCBs were manufactured and sold worldwide, including Arochlor, Clophen, Fenclor, Kanechlor, Phenoclor, and Pyralene among others. These mixtures were sold as blends based on their chlorine content. For example, Aroclor 1242, the most widely produced Aroclor, contained 42% chlorine while Aroclor 1254 contained 54% chlorine. In the United States, the sale of PCBs was banned in 1977. Since 1977, the disposal of PCBs has been strictly regulated under the Toxic Substances Control Act (TSCA). If disposals of PCBs following TSCA guidelines are strictly followed, environmental release should be minimal. The majority of current PCB releases would appear to be from rerelease of these compounds from reservoir sources. Other current sources of PCBs are most likely due to leaks and spills of still in-service PCBs, from transformers, for example, or from illegal disposal of PCBs. Despite the ban on the production of PCBs in the United States and Western Europe in the 1970s, PCBs were not banned in the former Soviet Union until the 1990s. 17.2.1

Environmental Fate and Transport

Much of the information on the fate and transport of CDDs/CDFs is based on the data for TCDD. Because CDDs/CDFs share many physical and chemical properties, the fate and transport of these chemicals should be qualitatively similar. CDDs/CDFs enter the terrestrial food chain via atmospheric deposition (Fries and Paustenbach, 1990). Recent airborne sources of dioxin are dominated by the combustion of wastes and fuels. The CDDs/CDFs emitted from combustion sources are predominately bound to particulate matter (PM), although there are some in the vapor phase. Once airborne, the CDDs/CDFs deposit on plants, soil, or water. Plant contamination by dioxins is mainly due to the deposition of contaminated PM on leaves. While there is evidence that CDDs/CDFs in soil can be absorbed directly by the roots of plants, it is highly unlikely that this results in significant concentrations of the dioxins in the above-ground plant (Insensee and Jones, 1971; Jensen et al., 1983). CDDs/CDFs also accumulate in soil through pesticide application and leakage from waste sites, although these pathways are more important for localized areas with specific problems. Once deposited on the plants or soil, the dioxins either enter the food chain or degrade. The deposition of dioxins on plants and soil enter the food chain through direct ingestion of the

SOURCES

637

plants or the incidental ingestion of soil by animals, resulting in the bioaccumulation of these chemicals in livestock (Fries and Paustenbach, 1990). Concentrations of CDDs/CDFs are higher in cow’s milk collected from the vicinity of a municipal waste incinerator than in commercial cow’s milk (Rappe et al., 1987). Dioxins can be degraded in the environment. However, this appears to be a relatively slow process. Photolysis is the main degradation pathway for CDDs/CDFs in the environment and requires ultraviolet (UV) light and an organic hydrogen donor (Crosby et al., 1971). Because of the requirement for UV light, dioxins that are on soil surfaces have shorter half-lives than those that are deeper in the soil. For example, the half-life of TCDD at the soil surface is estimated at 1–3 years, while it is 10–12 years in the subsurface soil (di Domenico et al., 1982; Kimbrough et al., 1984). Dioxins are sparingly soluble in water, and the concentrations of these chemicals in water are extremely low. Despite their low solubility, aquatic environments have significant amounts of dioxin as a result of their adsorption to sediments. Sediments of surface waters are thought to be the ultimate sink (or environmental reservoir) of CDDs/CDFs (Hutzinger et al., 1985), and the persistence of these compounds in water bodies results in bioaccumulation in aquatic organisms (Insensee and Jones, 1975). While deposition of PM contaminated with dioxin appears to be the major source of water contamination, local events such as industrial effluents and herbicide runoff may also contribute to the dioxin burdens. The PCBs and PBBs have similar physical and chemical properties as the CDDs/CDFs, which result in qualitatively similar environmental fates and transports (Safe, 1994). Atmospheric concentrations of PCBs in urban centers (1–10 ng/m3) were approximately an order of magnitude higher than in nonurban areas (0.1–0.5 ng/m3) (Atlas et al., 1986). Numerous factors including local sources, source emission strengths, and meteorological conditions influence ambient concentrations of PCBs. Surface waterways appear to be a major reservoir for PCBs (Tanabe and Tatsukawa, 1986), and the National Academy of Sciences (NAS, 1979) estimated that 50–80% of the PCBs in the environment are contained within the waterways of the North Atlantic. Similar to the CDDs/CDFs, PCBs are predominantly found in the sediments of these waterways. PCBs are also found in soil to varying degrees from 0.01 to over 2000 ppm (Tatsukawa, 1976). Photolysis and photoxidation appear to be the major pathways for destruction of PCBs in the environment. All PCBs photodechlorinate, and the photolysis rate increases as the chlorine content increases (Tiedje et al., 1993). In addition to photolysis, PCBs are also sensitive to biological degradation. There are over two dozen strains of aerobic bacteria and fungi that are capable of degrading most PCB congeners with five or fewer chlorines (Tiedje et al., 1993). Many of these organisms are of the genus Pseudomonas or Alcaligenes, and they are widely distributed in the environment. The higher chlorinated PCBs are more resistant to biodegradation than are the lower chlorinated congeners. In addition, PCBs substituted in two or more of the ortho positions are resistant to biodegradation. 17.2.2

Human Exposure

Exposure to dioxin can occur through occupational exposure, accidental exposures, or environmental exposures to the general population. The predominant hypothesis for environmental exposure to the general population focuses on the air to plant to animal hypothesis. This hypothesis focuses on the deposition of PM onto plants and soil. The deposited dioxins then enter the food chain by ingestion of the contaminated substrates by either livestock or aquatic life where they bioaccumulate. Eventually, the livestock or aquatic lives are consumed by humans.

638

DIOXINS AND DIOXIN-LIKE CHEMICALS

Estimates of daily human intake in the United States in the 1990s were approximately 3–6 pg TCDD equivalents/kg/d (Pinsky and Lorber, 1998; Schecter et al., 1994a), and these values are consistent with those reported from Western Europe (Schecter et al., 1994b, 1994c). Approximately half of the TCDD equivalents arose from PCBs, with only 10% attributable to TCDD itself (Schecter et al., 1994a, 1994b, 1994c). These daily intakes resulted in serum concentrations of approximately 40–60 parts per trillion (ppt) TEQ, and body burdens of approximately 8–13 ng TEQ/kg (DeVito et al., 1995). In general, dioxin body burdens increase with age, and this is thought to be the reflection of higher exposure of both during the 1940s through the 1960s compared to more recent exposures (Pinsky and Lorber, 1998), as well as decreased metabolism of dioxin with age (Flesch-Janys et al., 1995). A major route of elimination of dioxin in female humans, experimental animals, and mammalian wildlife is through breast milk. Breast milk has a high concentration of fat, and the dioxin distribute to fatty tissues. In women, nursing for at least seven months can decrease maternal serum concentrations of dioxin by up to 50% (Schecter et al., 1996). Exposures to nursing infants are much higher than maternal exposure and is estimated at 30–100 pg TEQ for the first year (Schecter et al., 1994a, 1994b, 1994c, 1996). These daily intakes are higher than all tolerable daily intakes defined by regulatory agencies throughout the world. In nursing infants, body burdens of dioxin at 12 months are two to four times higher than maternal burdens (Schecter et al., 1996). The potential health risks associated with these background exposures are uncertain; however, it should be noted that the benefits of breastfeeding are well documented and far outweigh the potential risks from background dioxin exposures (Rogan and Gladen 1993). There have been numerous incidents throughout the world where small populations have potentially been highly exposed to dioxin through industrial accidents, occupational exposures, wartime use of herbicides, or environmental pollution. One of the most wellcharacterized exposures occurred in Seveso, Italy in 1976. Approximately 1 kg of TCDD was released following an explosion at a trichlorophenol manufacturing plant. In 1976, there were no validated methodologies available to determine serum concentrations of TCDD. Despite this fact, a team of physicians, lead by Dr. Paolo Mocarelli, collected and stored blood from over 30,000 patients in the area, both exposed and unexposed (Bertazzi and di Domenico, 1995). Initial chemical analysis of the serum from several children from the most highly exposed area found serum concentrations of dioxin up to 50,000 ppt (Bertazzi and di Domenico, 1995). Based on the initial exposure estimates, the Seveso region was divided into three areas A, B, and R. Region Awas thought to be more highly exposed than region B, and region R was the unexposed area. More recent characterization of the exposures in Seveso indicates that the average serum concentrations in regions A are lower than the initial studies indicated (Bertazzi and di Domenico, 1995). Several industrial cohorts have been examined for dioxin exposure. Most of these workers are either farm workers spraying phenoxy herbicides or workers manufacturing herbicides or trichlorophenol (Fingerhut et al., 1991; Flesch-Janys et al., 1995). Workers in these studies have had serum concentrations of dioxin ranging 100–5000 ppt TEQ on a lipid-adjusted basis. TCDD is the predominant congener in the occupational exposures, in contrast to the low-level background exposure of the general population, where TCDD contributes approximately 10% of the total TEQ. Spraying of herbicides contaminated with TCDD during the Vietnam War exposed military personnel from both sides of the conflict, as well as Vietnamese civilians. Exposures of members of the U.S. Armed Forces are the best characterized from these groups. Despite

SOURCES

639

initial fears of high exposure during the war, it appears that, with the exception of those directly involved in the formulation and the spraying, most ground troops had limited exposure to these chemicals. The spraying occurred predominantly in the south of Vietnam. Civilians from South Vietnam had approximately 10 times the serum concentrations of TCDD compared to those in North Vietnam (Schecter et al., 1994b, 1994c). Concentrations of dioxin were similar between residents of South Vietnam, the United States, and Western Europe (Schecter et al., 1994b, 1994c). Accidental exposures to dioxins and dioxin-like compounds have occurred following consumption of contaminated rice oils. In 1968, a mass poisoning, called Yusho or oil disease, occurred in Fukuoka and Nagasaki prefectures in Japan due to ingestion of rice oil contaminated with Kanechlor-400, a commercial PCB mixture produced in Japan. It was subsequently discovered that the Kanechlor-400 mixture was contaminated with CDFs as well as polychlorinated tetraphenyls (Masuda, 1995). A similar incident occurred in Yucheng, Taiwan, in 1979 (Hsu et al., 1995). These incidents are known as Yusho and Yu-cheng, respectively. The concentrations of dioxin in the Yusho population shortly after the accident were approximately 20 times higher than controls, while PCB concentrations were 20–50 times the background (Masuda, 1995). Similar exposures were observed in the Yu-cheng incidents as well. While these populations were intensively studied, it has been difficult to determine which of the effects seen were due to the dioxin, the non-dioxin-like PCBs, or to their coexposures. 17.2.2.1 Pharmacokinetics The dioxin-like toxicity of these chemicals is due to the parent compound, and for the CDDs and the CDFs, metabolism is a detoxification step. Dioxins have relatively long half-lives in biological systems. In rats, the half-life of dioxin ranges from approximately 1 day for TCDF and 3,30 ,4,40 -tetrachlorobiphenyl to greater than 6 months for octachlorodibenzo-p-dioxin (OCDD) (Van den Berg et al., 1994). TCDD has a half-life ranging from 9 days in mice (Birnbaum, 1986; Gasiewicz and Rucci, 1984; Gasiewicz et al., 1983) to approximately 1 year in rhesus monkeys (Bowman et al., 1989). In humans the half-life of the CDDs and CDF range from months to over 20 years. Estimates of the half-life of TCDD range from 5.7 to 11.3 years, with an average of about 8 years (Flesch-Janys et al., 1998; Pirkle et al., 1984). The difference in the half-life between mice, rats, and humans is approximately 80–400-fold. Such large differences in half-life among species have significant impact on the dose metric used in comparing species sensitivity to these chemicals. Many of the earlier risk assessments were based on comparisons of daily dose or administered dose and did not adequately correct for the difference in the kinetics of these chemicals among species. More recent comparisons have used steady-state body burdens as the dose metric, which provided a more accurate comparison, although it has limitations depending upon the end point of comparison (DeVito et al., 1995; Wang et al., 1997). The term half-life must be used cautiously with dioxin for several reasons. First, in humans there is a relationship between percent body fat and half-life. The greater the percent body fat the longer the half-life (Flesch-Janys et al., 1995, 1998). In addition, there is nonlinear kinetics of these chemicals that precludes the use of a single half-life. The disposition of dioxin is dose dependent due to hepatic sequestration (DeVito et al., 1997; Andersen et al., 1993). The liver contains CYP1A2, which is inducible by dioxin. CYP1A2 binds TCDD and sequesters dioxin in the liver. In CYP1A2 knockout mice, neither TCDD nor 4-PeCDF is sequestered in hepatic tissue (Diliberto et al., 1997). In the wild-type mice, the induction of CYP1A2 results in the sequestration of these chemicals in the liver.

640

DIOXINS AND DIOXIN-LIKE CHEMICALS

The dose-dependent hepatic sequestration has been demonstrated in rats, mice, hamsters, guinea pigs, and humans (Van den Berg et al., 1994; Carrier et al., 1995). While TCDD is sequestered in the liver, other dioxins are sequestered to a greater degree. Because of these nonlinearities in the kinetics of dioxin, using a single half-life value has limitations, is best used as an indication of the persistence of these chemicals, and may not be suitable for quantitative determination of exposures. Intensive efforts were made to develop physiologically based pharmacokinetics models for dioxin. Most of these models focused on TCDD (Andersen et al., 1993, 1997; Kedderis et al., 1993); however, a few have examined other congeners such as 2,3,7,8-tetrabromodibenzo-p-dioxin, dibenzofurans, and PCBs (De Jongh et al., 1993; Kedderis et al., 1993). These models not only describe the pharmacokinetics of dioxin but are also the basis for response models examining gene transcription and hepatocarcinogensis (Kohn et al., 1996; Conolly and Andersen 1997; Portier et al., 1996). These models have potential use in risk assessment but require further validation. Congener specific pharmacokinetics information for PCBs in humans is also available. Those chemicals that have been examined are persistent, with half-lives in humans ranging from months to years (GE data). However, there are significant differences between the metabolism and disposition of PCBs compared to CDDs and CDFs. While metabolism for the CDDs and CDFs is a detoxifying pathway, metabolism of some of the dioxin-like PCBs produces bioactive metabolites. For example, hydroxylation of PCBs 77, 105, and 118 results in metabolites that bind to transthyretin, one of the thyroxine-binding proteins in serum. It has been hypothesized that the binding to transthyretin displaces thyroxine and increases its elimination (Brouwer et al., 1998). Other hydroxylated PCBs bind to utero globulin and accumulate in tissues expressing this protein, particularly the lung. Similar to the CDDs and CDFs, PCBs 77, 126, and 169 are sequestered in hepatic tissue. In contrast, the mono-ortho dioxin-like PCBs do not accumulate in the liver (Devitor et al., 1998). There is also evidence of pharmacokinetics interactions between CDDs, CDFs, and PCBs (van der Plas et al., 1998; Van den Berg et al., 1994), and extrapolations of animal data on single congeners to human exposures must be viewed with caution.

17.3 TOXICOLOGICAL EFFECTS AND MECHANISMS OF ACTION In experimental animals, dioxins induce numerous toxicities including immunotoxicity, reproductive and developmental toxicities, and carcinogenicity (DeVito and Birnbaum, 1995; Safe, 1990; Pohjanvirta and Tuomisto, 1994). The lethal effects of dioxin are unique in that death occurs weeks after the initial exposure and is preceded by a wasting syndrome. In acute exposure studies, the time to death appears independent of the dose, and increasing the dose does not decrease the time to death. Studies by Rozman and coworkers determined that the lethal effects of dioxin are time dependent (Viluksela et al., 1997). Animals initially die from the wasting syndrome within weeks following the initial exposures. However, animals that survive the initial wasting syndrome may eventually die from other causes, such as anemia, months after the initial exposure. The time to death, while independent of dose, can be altered by hypophysectomy. Lethal effects of TCDD can be observed within the first 24 h in hypophysectomized mice (DeVito et al., 1992a). While the dose response and time course for TCDD-induced wasting syndrome and lethality have been well characterized, the exact cause of death is uncertain. The wasting syndrome is thought to be due to alterations in the body weight set point. While the wasting syndrome can be severe, it does not appear to be

TOXICOLOGICAL EFFECTS AND MECHANISMS OF ACTION

641

a direct cause of death. Pair-fed controls with the same weight loss as the TCDD-treated animals did not exhibit mortality (Seefeld et al., 1984). 17.3.1

Carcinogenicity

The carcinogenicity of TCDD has been examined in rats, mice, hamsters, and Japanese medaka and is positive in all four species (Huff et al., 1994). Several studies in rats and mice indicated that TCDD is carcinogenic when administered for the lifetime of an animal. Tumors were observed in both sexes at multiple sites including liver, thyroid, lung, and several other tissues (Huff et al., 1994). In hamsters, the species most resistant to the lethal effects of TCDD, epidermal tumors were observed after dosing once a month for 6 months by either oral gavage or subcutaneous injections (Rao et al., 1988). Few of the related dioxins have been examined for carcinogenicity in 2-year bioassays. A mixture of HxCDDs induced liver tumors in female rats and thyroid tumors in the males. The Aroclors 1016, 1242, 1254, and 1260 mixtures have been tested for carcinogenicity, and all mixtures induce liver tumors while 1242, 1254, and 1260 also increase the incidence of thyroid tumors (Mayes et al., 1998). TCDD is negative in several short-term mutagenicity assays. In addition, using methods that can detect one DNA adduct in 1011 nucleotides, no TCDD-derived adducts have been detected (Turtletaub et al., 1990). Carcinogenesis is a multistage process that requires discrete steps involving genetic alterations of cells that clonally expand and progress into tumors. In this multistage process, TCDD clearly acts as a tumor promoter, although it may act on multiple stages of this process. TCDD is one of the most potent tumor promoters (Pitot et al., 1980; Poland et al., 1982; Lucier et al., 1991). All of the dioxin-like chemicals tested in hepatic tumor promotion models have tested positive, including 1,2,3,7,8-pentaCDD, 2,3,4,7,8-pentaCDF, and PCBs 126, 169, 118, 105, and 156 (Hemming et al., 1993, 1995; Buchmann et al., 1994; Maronpot et al., 1993; Haag-Gronlund et al., 1997, 1998; Schrenk et al., 1994). Much of the research on the carcinogenic effects of dioxin has examined the liver tumors in rats. In rats, the development of hepatic tumors occurs only in the female (Kociba et al., 1978; NTP, 1982). Tumor promotion studies in ovariectomized rats indicate that TCDD does not promote liver tumors in rats without a functioning ovary (Lucier et al., 1991). However, in these studies, the ovariectomized rats developed lung tumors, while the intact rats only developed hepatic tumors (Lucier et al., 1991). It has been hypothesized that TCDD increases the metabolism of estrogens and results in the production of catechol estrogen metabolites (Lucier et al., 1991). These metabolites are thought to redox cycle, resulting in the production of oxygen free radicals that then produce DNA damage (Tritscher et al., 1996). While the role of estrogen is critical in the TCDD-induced liver tumor in female rats, it should be noted that hepatic tumors were not observed in hamsters, and in mice males were more responsive to the hepatic carcinogenic effects of TCDD than were females (NTP, 1982). In rats and mice, TCDD is an extremely potent carcinogen. The LOAEL for hepatic adenomas in rats is 10 ng/kg/d, while the NOEAL is 1 ng/kg/d (Kociba et al., 1978; NTP, 1982). These doses seem rather high compared to the human intake of 1–6 pg TEQ/kg/d. To make an appropriate comparison, the large difference in half-life between species must be taken into account. One method is to express dose as steady-state body burdens (ng/kg). Daily intake of a chemical will eventually result in a steady-state condition in which the amount ingested equals the amount eliminated. With persistent chemicals such as dioxin,

642

DIOXINS AND DIOXIN-LIKE CHEMICALS

small concentrations ingested daily can result in large accumulation of the chemical in the body. In rats, 1 ng/kg/d results in a body burden of approximately 30 ng TCDD/kg body weight. In mice the LOAEL for hepatocarcinogenesis is 71 ng/kg/d (NTP, 1982), with a resulting steady-state body burden of 944 ng/kg (DeVito et al., 1995). 17.3.2

Developmental Toxicity

TCDD and several other congeners induce cleft palate and hydronephrosis in mice. Cleft palate can be induced in rats and hamsters, but only at doses that result in significant fetal mortality (Olson and McGarrigle, 1991). Studies in cultured developing palates indicate that mouse palates are approximately 100–1000 times more sensitive to the effects of TCDD than are human and rat palates (Abbott and Birnbaum, 1991; Abbott et al., 1994). In addition, human and rat palates are equally sensitive to the effects of TCDD, suggesting that humans are unlikely to develop cleft palate from dioxin. Other more subtle effects of dioxin have also been observed in developing animals. The developing reproductive system of rats and hamsters is extremely sensitive to the actions of TCDD and other AhR agonists. Prenatal exposure to TCDD decreases epididymal sperm counts in mice and epididymal and ejaculated sperm counts in rats and hamsters (Theobald and Peterson, 1997; Mably et al., 1992a, 1992b; Gray et al., 1995). Female rats and hamsters exposed to TCDD in utero and lactationally develop malformations of the phallus, clitoris, and incomplete opening of the vaginal orifice (Gray et al., 1995, 1997). These developmental alterations of the reproductive systems can occur at doses as low as 0.05 mg/kg when administered on gestational day 15 (Gray et al., 1995). In a multigenerational study, doses as low as 1 ng/kg/d decrease fertility in the F1 and F2 generation (Murray et al., 1979). The 1 ng/ kg/d dose in the multigenerational study resulted in steady-state body burdens of approximately 30 ng/kg (DeVito et al., 1995). Dioxin-like chemicals are also developmental neurotoxicants. Prenatal exposure to TCDD produces a permanent low-frequency auditory deficient in rats (Goldey et al., 1996). Similar effects were observed in animals prenatally treated with either Aroclor 1254 or PCB 126 (Goldey et al., 1996; Crofton and Rice, 1999). The development of the auditory system depends upon thyroid hormones (Goldey et al., 1996). TCDD and other dioxin decrease circulating thyroid hormones by inducing uridine diphosphate glucuronsyl transferase (UDPGT) (Henry and Gasiewicz, 1987). In rats, pre- and postnatal exposure to TCDD and other dioxins decreases circulating thyroid hormone concentrations during the period of cochlear development, particularly the regions of the cochlea responsible for lowfrequency hearing (Goldey et al., 1996; Goldey and Crofton, 1998; Crofton and Rice, 1999). The auditory deficits in rats produced by TCDD is a relatively high-dose phenomena requiring at least 1 mg/kg on gestational day 18, which is approximately 20 times higher than the doses needed to alter the developing reproductive tract. The developmental neurotoxicity of TCDD is also expressed as a permanent change in regulated body temperature in both rats and hamsters (Gordon et al., 1995, 1996). Several laboratories have demonstrated behavioral changes in animals exposed prenatally to dioxinlike chemicals. Schantz and Bowman (1989) examined the neurological developmental effects of TCDD in rhesus monkeys. Rhesus monkeys were exposed prenatally and lactationally to TCDD with dams receiving as little either 5 or 25 ppt of TCDD in the diet. The offspring had alterations in object learning at the low dose, and at the high dose few of the dams were able to maintain pregnancy and only one offspring from seven dams survived (Schantz and Bowman, 1989). The rhesus monkeys fed a diet of 5 ppt TCDD averaged

MECHANISMS OF ACTION

643

151 pg/kg/d (Bowman et al., 1989), and this diet resulted in a body burden of approximately 42 ng/kg (DeVito et al., 1995). 17.3.3

Immunotoxicity

TCDD and related chemicals are immunotoxicants in several species (Harper et al., 1993). The immunotoxicity of dioxin is difficult to characterize, and there does not appear to be a unique dioxin-like immune response. The responses affected by TCDD depend on the species studied and the model used to examine the immune response. For example, the suppression of the plaque forming cell response to sheep red blood cells in mice is one of the most consistent findings, with an ED50 of approximately 0.7 mg TCDD/kg (Smialowicz et al., 1994). Yet in rats, TCDD enhances the response to sheep red blood cells (Smialowicz et al., 1994). In contrast, exposures to the same doses of TCDD that enhance the response to sheep red blood cells suppress host resistance to Trichinella spiralis (Luebke et al., 1994, 1995). In influenza models, doses of TCDD as low as 10 ng/kg increase mortality in mice following exposure to influenza virus (Burleson et al., 1996). Other low-dose effects on the immune system by TCDD are altered lymphocyte subsets in marmoset monkeys exposed to 0.3 pg/kg/week for 24 weeks (Neubert et al., 1992). This dose results in body burdens of approximately 10 ng/kg (DeVito et al., 1995). The immunotoxicity is thought to be associated with changes in proliferation and differentiation in a variety of cell types in the immune system (Kerkvliet, 1995).

17.4 MECHANISMS OF ACTION 17.4.1

Role of the Ah Receptor in the Biological Effects of Dioxin

Binding to and activating the AhR is the initial step in the biological and toxicological effects of dioxin. Most, if not all, of the effects are mediated by this protein (Birnbaum 1994). The unliganded AhR is found in either the cytosol or nucleus as a multimeric complex that includes two molecules of a 90 kDa heat shock protein and several other smaller molecular weight proteins (Whitlock et al., 1996). Upon ligand binding, the AhR dissociates from this complex and binds to the aryl hydrocarbon nuclear translocator (ARNT). The transformed AhR–ARNT complex then binds to specific dioxin-responsive enhancer (DRE) sequences located in the promotor region of the CYP1A gene and several other TCDD responsive genes. The binding of the activated AhR complex to the DREs alters chromatin structure and enhances the association of other components of the transcriptional machinery resulting in the initiation of transcription. This transformation of the AhR into a nuclear binding form can be attenuated by protein kinase C inhibitors, and phosphorylation events appear to be important in regulating the activity of these proteins (Whitlock et al., 1996). While the exact role of the AhR in normal biochemical and physiological processes is uncertain, there are several lines of evidence that suggest that it plays an important function in developmental and homeostatic functions. AhR, a member of the beta-helix-loop-helixPer-ARNT-Sim (bHLH-PAS) superfamily, is a ligand-activated transcription factor. The AhR is a highly conserved protein present in all mammalian species examined and has been found in all vertebrate classes examined, including modern representatives of early vertebrates, such as cartilaginous and jawless fish (Hahn, 1998). In addition, a possible

644

DIOXINS AND DIOXIN-LIKE CHEMICALS

AhR homolog has been identified in Caenorhabditis elegans (Powell-Coffman et al., 1998). The bHLH-PAS superfamily consists of at least 32 proteins found in diverse organisms from Drosophila, C. elegans, and humans. These proteins are transcription factors and appear to require dimerization, either homo-or heterodimers, for functional effects. These proteins regulate circadian rhythms (per and clock), steroid receptor signaling (SRC-1, TIF2, RAC3), or are involved in sensing oxygen tension (HIF-1, EPAS-1/HLF) (Hahn, 1998). It has been proposed that understanding the function of the bHLH-PAS family of proteins and the phylogenetic evolution of the AhR may lead to an understanding of the role of this protein in normal processes. Other lines of evidence indicate that the AhR has important physiological functions based on its spatial and temporal expression in developing embryos. The AhR is expressed in tissue- and cell-specific manner during development (Abbott et al., 1994, 1995, in press). It is highly expressed in the neural epithelium, which forms the neural crest during development (Abbott et al., 1995). AhR knockout mice have been produced using a targeted disruption of the Ahr locus (Fernandez-Salguero et al., 1997). While the results from the two laboratories are not identical, it is apparent that the AhR is not essential to life, since the knockout mice survive and reproduce. However, the Ahr
/
mice develop numerous lesions with age (Fernandez-Salguero et al., 1997). Mortality begins to increase at about an age of 20 weeks, and by 13 months 46% of the mice either died or were ill. Cardiovascular alterations consisting of cardiomyopathy with hypertrophy and focal fibrosis, hepatic vascular hypertrophy and mild fibrosis, gastric hyperplasia, T-cell deficiency in the spleen, and dermal lesions were apparent in these mice, and the incidence and severity increased with age (Fernandez-Salguero et al., 1997). While male and female AhR-/-mice are fertile, the females have difficulty maintaining conceptuses during pregnancy, surviving pregnancy and lactation, and rearing pups to weaning. In one study, only 46% of 39 pregnant AhR-/-females successfully raised pups to weaning (Abbott et al., in press). Other groups have developed an ARNT knockout mouse. In contrast to the Ahr
/
mice, the lack of its dimerization partner, ARNT, results in fetal mortality. Unlike the AhR, ARNT has several other dimerization partners including HIF-1 alpha (Hahn, 1998). It has been suggested that some of the toxicity of AhR ligands such as TCDD may be due to sequestration of ARNT by the AhR, decreasing ARNTs availability for other members of the PAS family (Hahn, 1998). The importance of the AhR in the toxicity of dioxin stems from several lines of evidence. First, structure–activity relationships indicate a correlation between receptor binding affinity and in vivo toxicity in mice (Safe, 1990). Second, there is genetic evidence in mice that is consistent with the AhR mediating the toxic effects of dioxin. The C57BL mice are the most sensitive to the toxic effects of dioxin while the DBA/2J is the most resistant. The binding affinity of the AhR is approximately 10–20 times lower in the C57BL compared to the DBA, and the C57BL mice are approximately 10–20 times more sensitive than the DBA mice (Poland and Glover, 1980). In fact, the sensitivity to dioxin in mice segregates nicely with several different alleles and the binding affinity of dioxin to these gene products (Poland and Glover, 1990). Finally, the most powerful evidence is that the AhR
/
mice are resistant to the toxicities of TCDD (Mimura et al., 1997; Fernandez-Seguaro et al., 1996). While the initial step in the toxicity of dioxin is the binding to the AhR, this binding is clearly not the sole determinant in the toxicity of these chemicals. In the mouse, the different AhR alleles have different binding affinities to dioxin, and these differences can explain the differences in sensitivity to dioxin between these strains (Poland and Glover, 1980). The biochemical and biophysical properties of the AhR are similar between species, with only

MECHANISMS OF ACTION

645

slight differences in molecular mass and subunit stability (Gasiewicz and Rucci, 1984). Yet for some end points, such as lethality, there are significant differences in the ED50 values between species (Pohjanvirta and Tuomisto, 1994). While the ED50 values for some end points, such as fetal mortality and enzyme induction, are similar between the species (Olson and McGarrigle, 1991). Overall, these data indicate that the steps after AhR binding and activation are critical in understanding the species differences in response to dioxin. 17.4.2

Dioxins as Growth Dysregulators and Endocrine Disruptors

While dioxins appear to alter numerous systems, there are several underlying commonalities of the toxicity of these chemicals. Most of the toxicities are associated with alterations in proliferation and/or differentiation, such as cancer, immunotoxicity, and chloracne. Hyperplastic responses are observed in gastric mucosa and bile ducts in monkeys (McConnell and Moore, 1979), urinary bladder in guinea pigs, and hepatic and dermal hyperplasia occurs in several species. Hypoplasia occurs in lymphoid tissues in all mammalian species examined and in the gonads of mice, rats, rabbits, and mink. Squamous metaplasia is induced by TCDD in ceruminous and sebaceous glands in monkeys (Moore et al., 1979). Dysplastic responses of the nails and teeth have been observed in humans and nonhuman primates following prenatal exposure to dioxin (Moore et al., 1979; Masuda, 1995). Dioxins are potent growth dysregulators. Dioxins alter the normal homeostatic processes by disrupting cell signaling pathways through multiple mechanisms. Cells maintain homeostasis through a complex processes involving the release of paracrine or autocrine hormones or growth factors. These hormones and growth factors interact with specific receptors that can be localized to the cell membrane, cytosol, or nucleus. Activation of these receptors initiates a cascade of events that affect cell functioning. Dioxins alter cell signaling and homeostasis by altering hormones and growth factors, and their receptors and by altering the signaling of the activated receptors. These effects of dioxin are often tissue and developmental stage specific (DeVito and Birnbaum, 1995). TCDD alters many of the signaling pathways involved in the endocrine system as well and can be considered an endocrine disruptor. Dioxin alters serum concentrations of several pituitary hormones (Moore et al., 1989) and steroid hormones including androgens (Moore et al., 1985), glucocorticoids (Lin et al., 1991), and thyroid hormones (Henry and Gasiewicz, 1987; Kohn et al., 1996). In cell culture, TCDD increases estradiol metabolism by almost 100-fold (Spink et al., 1990). Estrogen receptors are decreased in several tissues including liver and uteri in mice and rats administered dioxin (DeVito et al., 1992a, 1992b; Safe et al., 1991). Dioxins have frequently been described as antiestrogens because they decrease estrogen and estrogen receptor concentrations. However, this characterization should be viewed cautiously, since dioxins also increase the incidence and severity of endometriosis in monkeys (Reier et al., 1993) and increase endometriotic lesions in mice and rats (Cummings et al., 1996; Johnson et al., 1997). Endometriosis is dependent upon estrogens for growth and is often treated with drugs that downregulate ovarian function. Once again, this highlights the tissue-specific effects of these chemicals. 17.4.3

Human Health Effects Associated with Dioxin Exposure

The epidemiological data examining the potential carcinogenic effects of dioxin have focused mostly on several industrial cohorts of pesticide manufacturers (Kogevinas

646

DIOXINS AND DIOXIN-LIKE CHEMICALS

et al., 1997), American veterans of Vietnam, and the Seveso, Italy, cohorts. It has proved difficult to obtain a consensus on the interpretation of these studies. IARC evaluated the evidence for carcinogenicity of TCDD and other dioxins (McGregor et al., 1998). The IARC working group focused on the four cohorts with the highest exposure (Fingerhut et al., 1991; Hooiveld et al., 1998; Ott and Zober 1996; Flesch-Janys et al., 1995; Zober et al., 1994). In these studies, the overall risk of developing cancer was significantly greater (approximately 1.4-fold) in the exposed populations (McGregor et al., 1998). However, few site-specific cancer risks were consistent across the cohorts, but evidence of increased risk of lung cancer was significant in all the four studies (Fingerhut et al., 1991; Hooiveld et al., 1998; Ott and Zober 1996; Flesch-Janys et al., 1995; Zober et al., 1994). One of the problems with interpretation of these studies is that an increase in risk for cancers at many sites was observed for only a few cases, as is the case for smoking exposure to ionizing radiation in atomic bomb survivors (McGregor et al., 1998). In contrast to TCDD, both smoking and ionizing radiation also increase site-specific tumors. Because TCDD is the only chemical that increased risk for all cancers without consistently increasing the risk for specific tumors suggests that the epidemiological data should be viewed with caution. The IARC workgroup considered the evidence for carcinogenicity as limited. Another problem with the interpretation of the cancer data for the most highly exposed occupational cohorts is the lack of a consistent exposure–response relationship. This can be seen in Fig. 1 from a paper by Starr (2002) that shows the data from the studies of workers in Hamburg (Flesch-Janys et al., 1998), the NIOSH study (Fingerhut et al., 1991), and for the BASF cohort (Ott and Zober, 1996). It can be seen that the SMRs above unity in almost all of the exposed groups show little, if any, association with TCDD body burden over a range covering three orders of magnitude. It is likely that the excess cancers in these groups of chemical plant production workers were attributable to their occupational exposures to many other toxic chemicals that were produced in those plants or to their tobacco smoking. Bodner et al. (2003) extended the analysis of the 2187 Dow Chemical Company workers in the NIOSH cohort by 9–30 years of follow-up. This cohort included workers with high exposures, as evidenced by 11% of them having had clinically confirmed chloracne. There was no excess for all cancers, and an SMR for lung cancer of 0.8. Also, there was no increased cancer risk with increasing exposure. In a recent study of all workers in the NIOSH cohort at eight U.S. plants with sufficient data to estimate their exposure to TCDD, Cheng et al. (2006) sought to determine whether the association between TCDD and cancer was due to, or modified by, other occupational exposures. They used both untransformed and ln-transformed TCDD ppt-years, lagged 15 years in their analyses, as well as in their analyses of smoking-related cancers and other cancers. They reported that their estimated incremental lifetime mortality risk at age 75 from all cancers was 6–10 times lower than previous estimates for this cohort that did not consider the age and concentration dependence of TCDD elimination. In view of the relatively weak epidemiological data on cancer in humans, the IARC workgroup also considered mechanistic information to support their overall evaluation that TCDD is carcinogenic in humans (McGregor et al., 1998). In 2005, after completion of its dioxin reassessment, EPA revised its cancer guidelines and charged the NRC Committee to use the new guidelines in addressing the scientific evidence for classifying TCDD as a human carcinogen. In its report (NRC, 2006), there was a split among the committee members on whether the evidence met all of the criteria necessary for definitive classification of TCDD as “carcinogenic to humans,” although they could all agree that it should be considered “likely to be carcinogenic to humans.”

MECHANISMS OF ACTION

647

Considering the somewhat equivocal evidence of TCDD carcinogenicity among highly exposed industrial workers, and the currently much lower and still dropping levels of exposure in the general population, the cancer risks to the general public from contemporary environmental levels of dioxin and DLCs are extremely small. In fact, there is suggestive evidence that very low levels of dioxin exposure may be associated with reduced risks of cancer based on the observed rates in the NIOSH occupational cohorts (Kayajanian, 2002). The results are not inconsistent with an examination of the association between benign prostatic hyperplasia (BPH) and serum dioxin concentrations that was made for participants in the National Health and Nutritional Examination Survey 1999–2000 (NHANES) by Gupta et al. (2006). After age adjustment, men without BPH had 21% (CI: 5–39%) higher TEQs compared to men with BPH. The population was relatively small, but it was a national probability sample. Immune effects of dioxin are observed in all species examined (Kerkvliet, 1995). However, human health has not been as consistently altered. Some of the best evidence comes from developmental studies in humans, which have suggested that there are developmental effects following exposure to CDDs CDFs, and PCBs (Weisglas-Kuperus et al., 1995). In addition, neurodevelopmental effects have been reported for a number of cohorts (Rogan and Gladen, 1991, 1992; Jacobson and Jacobson, 1997; Jacobson et al., 1990). Recent follow-up studies on people heavily exposed to TCDD in Seveso reported that plasma IgG levels decreased with increasing plasma TCDD, while there was no association between TCDD and IgM, IgA, C3, or C4 levels (Baccarelli et al., 2002). Also, for women who were heavily exposed to TCDD in Seveso, there was no clear evidence of alterations in ovarian function. The Dutch studies of a cohort of infants has demonstrated relationships between high CDDs, CDFs, and PCBs exposure and decreases in thyroid hormones and delayed neurodevelopment (Koopman-Esseboom et al., 1994; Weisglas-Kuperus et al., 1995). While these studies indicate that exposure to this class of chemicals may be associated with alterations in the developing immune and neurological systems, these effects were subclinical, and extrapolation to the population at large is uncertain. However, it should be stressed that in the Dutch studies, this cohort is of women and infants with no known high exposures to PHAHs other than background exposures. The associations found in this cohort, while subclinical, warrant further research and concern. 17.4.4

Toxic Equivalency Factors

Estimating the risk associated with dioxin exposure can be problematic. These chemicals are present as trace contaminants of a complex mixture containing numerous dioxin-like chemicals. Initially, risk assessments focused on TCDD alone, and all other dioxins were excluded. However, experimental evidence clearly indicates that these other dioxin-like chemicals cannot be ignored. If risk assessments are to include other dioxins, then they must either assume that all dioxin-like chemicals are as potent as TCDD or they must use some sort of relative potency scaling methodology. The experimental data clearly demonstrate that assuming all dioxin are equally potent to TCDD would greatly overestimate the potential risk. Hence, a relative potency scheme described as toxic equivalency factors was proposed to estimate the potential health risk following exposure to dioxin-like chemicals (Eadon et al., 1986). The toxicological basis for the TEF methodology is the shared mechanism of action of these chemicals, namely, activation of the AhR (Safe, 1990; Birnbaum and DeVito, 1995). The relative potency of a dioxin-like chemical is compared to TCDD, one of the most

648

DIOXINS AND DIOXIN-LIKE CHEMICALS

potent AhR ligands studied. The relative potency for a chemical from a particular study is described as a REP. A TEF is assigned by an expert group, using all the available REP values for a particular chemical. The data available for assigning a TEF value for a chemical varies widely depending upon the congener. For example, there are over 25 studies that examined the relative potency of PCB 126 compared to TCDD with end points ranging from in vitro studies examining binding affinity and biochemical alterations to in vivo subchronic studies examining toxicological and biochemical effects (Van den Berg et al., 1998). For other chemicals, such as PCB 80, there are only in vitro studies suggesting a dioxin-like effect (Van den Berg et al., 1998). In determining the TEF value, the varying types of data are weighted according to relevancy. For example, data on binding affinity are weighted less than data from acute in vivo studies, which are weighted less than data from subchronic studies examining toxicological effects (Van den Berg et al., 1998). TEFs are considered order-of-magnitude estimates for several reasons, including the quality and quantity of the data available, the variability of the data, and the limited information on the species extrapolation of the relative potency of these chemicals (Van den Berg et al., 1998). The relative potency of a chemical compared to TCDD is a function of its binding affinity to the AhR and its comparable pharmacokinetics properties, including absorption, disposition, metabolism, and elimination (DeVito et al., 1997;). One example of the importance of pharmacokinetics in the relative potency of these chemicals is OCDD. Originally, OCDD was assigned a TEF value of 0 because it demonstrated no toxicological effects in acute studies in rodents and binding data were unavailable due to the compound’s insolubility in aqueous solutions. However, in subchronic studies in rats, OCDD demonstrated significant dioxin-like effects and was assigned a TEF value of 0.001. In short-term studies, very little OCDD was absorbed by the animals. However, because of its persistence, in longer-term studies OCDD will accumulate and eventually attain concentrations in the animal sufficient to produce toxicological effects (Birnbaum and Couture, 1988; Couture et al., 1988). The TEF methodology assumes that the biochemical and toxicological effects of mixtures of dioxin can be predicted using a dose addition methodology. Several investigators have examined the TEF methodology, using either simple mixtures containing 2, 3, or 4 congeners, or more complex mixtures from environmental samples or laboratory prepared mixtures. The TEF methodology adequately predicted the immunotoxicity (Silkworth et al., 1989), lethality (Eadon et al., 1986), or biochemical changes (Van den Berg et al., 1989) of complex environmental mixtures. Schrenck and coworkers used the TEF methodology to predict the tumor promotion potency of a laboratory prepared mixture of 49 different chlorinated dioxins. Rozman and coworkers examined the effects of a quarternary mixture of chlorinated dioxin and found that the TEF methodology predicted both biochemical and toxicological responses of the mixture (Viluksela et al., 1998a, 1998b). While there are a number of studies demonstrating additivity and supporting the TEF methodology, there are still some unanswered questions. Recently, the TEF methodology was expanded to include the coplanar PCBs and the mono-ortho-substituted PCBs. These chemicals, while demonstrating dioxin-like toxicities, have significant non-dioxin-like effects. For example, TCDD and other chlorinated dibenzo-p-dioxins decrease thyroxine concentrations by only 40–50% in rats. In contrast, PCBs 77 and 118 can produce a 90% decrease in thyroxine (Crofton et al., 1998). In a complex mixture consisting of two chlorinated dibenzo-p-dioxins, four chlorinated dibenzofurans, and six chlorinated biphenyls, the TEF methodology predicted the immunotoxicity and enzyme inducing effects of

MECHANISMS OF ACTION

649

the mixture but underpredicted the thyroxine decreases and porphyrin accumulation in rats administered the mixture compared to rats receiving TCDD. One of the problems with the TEF methodology is the confusion as to what this method actually does. The TEF methodology only examines dioxin-like chemicals and assumes that there are no interactions between the dioxin and the nondioxin present in the mixture. In essence, the method allows one to estimate the potential health risk of a complex mixture by assuming that only TCDD is present. This method neither makes attempt to predict interactions of dioxin and other classes of chemicals nor the potential for toxicity unrelated to the AhR-binding cascade. Hence, the method ignores potential nonadditive interactions, both antagonistic and greater than additive interactions. While there is uncertainty in the use of the TEF methodology, one question to ask is whether using this method decreases or increases our uncertainty in the overall risk assessment. As described above, either excluding all other dioxin-like chemicals or considering other dioxin-like chemicals as potent as TCDD increases uncertainty in risk assessments. At a workshop on the use of TEFs in ecological risk assessment sponsored by the USEPA, the participants clearly agreed that the use of the TEF methodology decreases uncertainty. However, quantifying the uncertainty in the TEF methodology, and its effects on the uncertainty of the overall risk assessment, remains elusive (National Research Council 2006). With no alternative methodology, one can argue that the TEF method decreases uncertainty in the overall risk assessment. Until alternative methodologies are developed, the TEF methodology is the most appropriate method to estimate the potential health risks of exposure to dioxin-like chemicals. Significant efforts should be made to clarify the uncertainty of this methodology to support risk assessors and managers until more accurate methods are developed. 17.4.5

Risk Characterization

Dioxins induce numerous toxicities in experimental animals. The toxicity of these chemicals is mediated through an interaction with the Ah receptor. The Ah receptor is a highly conserved protein throughout evolution. While the exact role of the Ah receptor is uncertain, inappropriate activation of this receptor by dioxins produces tissue-specific alterations in differentiation, proliferation, and apoptosis. These processes are critical in numerous biological systems including development, the immune and endocrine systems, and carcinogenesis, to name a few. While there may be outliers, either resistant or sensitive, to certain dioxin toxicities, most species respond at similar exposures (i.e., within an order of magnitude for dose). The available data suggest that for biochemical and some toxic responses, humans have similar sensitivities to experimental animals. For example, the steady-state body burdens in rodents that result in cancers are between 300 and 900 ng/kg (DeVito et al., 1995). In the epidemiological studies, the high end of exposures have body burdens ranging 400–5000 ng/kg (DeVito et al., 1995). There is no firm evidence to suggest that humans are less sensitive to the potential effects of dioxin than are laboratory animals. Because of the consistency of effects observed in both humans and experimental animals, and the evolutionary conservation of the structure and function of the Ah receptor, efforts should be made to limit exposure to this class of chemicals. Since the available human data have not been considered to provide an adequate basis for quantitative risk assessments for dioxins in the past, there has been a reliance on the toxicity of dioxin in laboratory animals as a basis for understanding and estimating the risks

650

DIOXINS AND DIOXIN-LIKE CHEMICALS

associated with exposure to these chemicals. In the past, regulators have relied on either linearized models for cancer risk assessments or on safety factor methods. EPA has used a margin of exposure (MOE) method to examine potential risks from these chemicals (DeVito et al., 1995). Recent human background exposure was approximately 8–13 ng TEQ/kg. Responses in experimental animals have been observed at body burdens ranging 5–900 ng/ kg or more. In the low dose range, immunotoxicity in rodents and monkeys, reproductive effects in multigeneration studies, and endometriosis in rhesus monkeys occurred at body burdens between 5 and 64 ng/kg. If one divides the body burden in animals where effects are observed, by the average human body burden one obtains the MOE. From the above data, the MOE ranges from less than 1 to approximately 10 (an order of magnitude). The MOE does not imply that effects are occurring in the population. However, it compares the difference in present human exposure to that of animals exposed to toxic doses of a chemical. Typically, the EPA considers a MOE of 100 reasonable for noncarcinogens. The World Health Organization (WHO) has also assessed the potential human health effects of dioxin. The WHO experts suggested that the tolerable daily intake be set at 1–4 pg TEQ/kg/d. They also noted that present exposures throughout the world range 2–6 pg TEQ/ kg/d. The WHO experts emphasized the Dutch cohorts and stressed that these data suggest that background exposures are associated with effects. In addition, the working group also noted that concentrations of dioxin are decreasing in the environment over the last two decades. The risk characterizations for dioxins conducted by EPA have indicated significant public health risks from environmental exposures, especially from dietary sources. Because of the seriousness of its concerns, and those of the public, EPA has requested reviews of its risk assessments for this class of chemicals from its External Science Advisory Board (SAB, 1995, 2001) and more recently from the National Research Council (NRC, 2006). All three of these in-depth reviews by broadly constituted panels with expertise covering the various scientific and medical fields have withheld endorsements of the risk estimates contained in the successive EPA risk assessments. The basic problem has been that the risks associated with exposure to dioxin have been based on (1) studies conducted with animal models; (2) require interspecies extrapolations; (3) high-to-low dose extrapolations; (4) inclusion of multiple safety factors; and/or (5) reliance on overly conservative risk models for carcinogenesis. Thus, their usefulness for risk characterization remains uncertain, at best. As noted by NRC (2006), a major issue in the designation of carcinogenicity is that “analysts must extrapolate well below the doses observed in the studies to consider typical human exposure levels. This extrapolation involves two critical decisions: (1) selecting a point of departure (POD), which corresponds to the lowest dose associated with observable adverse effects within the range of data of a study, and (2) selecting the mathematical model used to extrapolate risk from typical human exposures that are well below the POD.” The NRC Committee concluded that EPA’s decision to rely solely on a default linear extrapolation model lacked scientific support. It recommended that EPA provide risk estimates using both nonlinear and linear methods to extrapolate below PODs and that it communicate the scientific strengths and weaknesses of both approaches so that the full range of uncertainty generated by modeling of the data is conveyed in their reassessment. With regard to noncancer effects, the NRC Committee faulted the decision of the EPAs not to derive a reference dose (RfD), its traditional metric for such effects. An RfD provides (1) estimates of the proportion of the population with intakes above the RfD; (2) detailed assessment of population groups, such as those with elevated exposures; and (3)contributions of the major food and other environmental sources for those with high intakes. Without such data, the

REFERENCES

651

NRC Committee found the risk characterization difficult to follow and recommended that EPA substantially revise it to include a more comprehensive risk characterization and discussion of the uncertainties surrounding key assumptions and variables. However, since some studies have suggested that dioxin intakes remain within the range where nonclinical responses could occur in the population, it is prudent to continue monitoring this class of chemicals and to support studies of their health-related effects. In the meantime, we can take comfort in that concentrations of dioxin in the population are decreasing. REFERENCES Abbott BD, Birnbaum LS (1991) TCDD exposure of human embryonic palatal shelves in organ culture alters the differentiation of medial epithelial cells. Teratology 43:119–132. Abbott BD, Probst MR, Perdew GH (1994) Immunohistochemical double-staining for Ah receptor and ARNT in human embryonic palatal shelves. Teratology 50:361–366. Abbott BD, Birnbaum LS, Perdew GH (1995) Developmental expression of two members of a new class of transcription factors: I. Expression of aryl hydrocarbon receptor in the C57BL/6N mouse embryo. Dev. Dyn. 204:133–143. Abbott BD, Buckalew AJ, Diliberto JJ, Wood CR, Held GA, Pitt J, Schmid JE , Adverse reproductive outcomes in the transgenic ah receptor-deficient mouse. J. Am. Med. Assoc. (In press). Andersen ME, Mills JJ, Gargas ML, Kedderis L, Birnbaum LS, Neubert D, Greenlee WF (1993) Modeling receptor-mediated processes with dioxin: implications for pharmacokinetics and risk assessment. Risk Anal. 13:25–36. Andersen ME, Birnbaum LS, Barton HA, Eklund CR (1997) Regional hepatic CYP1A1 and CYP1A2 induction with 2,3,7,8- tetrachlorodibenzo-p-dioxin evaluated with a multicompartment geometric model of hepatic zonation. Toxicol. Appl. Pharmacol. 144:145–155. Atlas E, Bidleman T, Giam CS (1986) Atmospheric transport of PCBs to the oceans. In:Waid JS, editor. PCBs and the Environment. Vol 1.Boca Raton, FL:CRC Press. pp.79–100. Baccarelli A, Mocarelli P, Patterson DG Jr, Bonzini M, Pesatori AC, Caporaso N, Landi MT (2002) Immunologic effects of dioxin: new results from Seveso and comparison with other studies. Environ. Health Perspect. 110:1169–1173. Bertazzi PA, di Domenico A (1995) Chemical, environmental, and health aspects of the Seveso, Italy, accident. In:Schecter A, editor. Dioxins and Health. New York:Plenum Press. pp.587–632. Birnbaum LS (1994) Evidence for the role of the Ah receptor in response to dioxin. Prog. Clin. Biol. Res. 387:139–154. Birnbaum LS (1995) Workshop on perinatal exposure to dioxin-like compounds. V. Immunologic effects. Environ. Health Perspect. 103(Suppl. 2):157–160. Birnbaum LS (1986) Distribution and excretion of 2,3,7, 8-tetrachlorodibenzo-p-dioxin in congenic strains of mice which differ at the Ah locus. Drug Metab. Dispos. 14:34–40. Birnbaum LS, Couture LA (1988) Disposition of octachlorodibenzo-p-dioxin (OCDD) in male rats. Toxicol. Appl. Pharmacol. 93:22–30. Birnbaum LS, DeVito MJ (1995) Use of toxic equivalency factors for risk assessment for dioxins and related compounds. Toxicology 105:391–401. Bodner KM, Collins JJ, Bloemen LJ, Carson ML (2003) Cancer risk for chemical workers exposed to 2,3,7,8-tetrachlorodibenzo-p-dioxin. Occup. Environ. Med. 60:672–675. Bowman RE, Schantz SL, Weerasinghe NCA, Gross ML, Barsotti DA (1989) Chronic dietary intake of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) at 5 or 25 parts per trillion in the monkey: TCDD kinetics and dose-effect estimate of the reproductive toxicity. Chemosphere 18:243–252.

652

DIOXINS AND DIOXIN-LIKE CHEMICALS

Brouwer A, Morse DC, Lans MC, Schuur AG, Murk AJ, Klasson-Wehler E, Bergman A, Visser TJ (1998) Interactions of persistent environmental organohalogens with the thyroid hormone system: mechanisms and possible consequences for animal and human health. Toxicol. Ind. Health 14: 59–84. Buchmann A, Stinchcombe S, Korner W, Hagenmaier H, Bock KW (1994) Effects of 2,3,7,8tetrachloro- and 1,2,3,4,6,7,8-heptachlorodibenzo-p-dioxin on the proliferation of preneoplastic liver cells in the rat. Carcinogenesis 15:1143–1150. Burleson GR, Lebrec H, Yang YG, Ibanes JD, Pennington KN, Birnbaum LS (1996) Effect of 2,3,7,8tetrachlorodibenzo-p-dioxin (TCDD) on influenza virus host resistance in mice. Fundam. Appl. Toxicol. 29:40–47. Carrier G, Brunet RC, Brodeur J (1995) Modeling of the toxicokinetics of polychlorinated dibenzop-dioxins and dibenzofurans in mammalians, including humans. I. Nonlinear distribution of PCDD/PCDF body burden between liver and adipose tissues. Toxicol. Appl. Pharmacol. 131: 253–266. Cheng H, Aylward L, Beall C, Starr TB, Brunet RC, Carrier G, Delzell E (2006) TCDD exposure– response analysis and risk assessment. Risk Anal. 26:1059–1071. Cleverly D, Monetti M, Phillips L, Cramer P, Heit M, McCarthy S, O’Rourke K, Stanley J, Winters D (1996) A time–trends study of the occurrences and levels of CDDs, CDFs, and dioxin–like PCBs in sediment cores from 11 geographically distributed lakes in the United States. Organohalogen Compd. 28:77–82. Cleverly D, Schaum J, Winters D, Schweer G, O’Rourke K (1998) The inventory of sources of dioxin in the United States. Organohalogen Compd. 36:1–6. Conolly RB, Andersen ME (1997) Hepatic foci in rats after diethylnitrosamine initiation and 2,3,7,8tetrachlorodibenzo-p-dioxin promotion: evaluation of a quantitative two- cell model and of CYP 1A1/1A2 as a dosimeter. Toxicol. Appl. Pharmacol. 146:281–293. Couture LA, Elwell MR, Birnbaum LS (1988) Dioxin-like effects observed in male rats following exposure to octachlorodibenzo-p-dioxin (OCDD) during a 13-week study. Toxicol. Appl. Pharmacol. 93:31–46. Crofton KM, Rice DC (1999) Low-frequency hearing loss following perinatal exposure to 3,30 ,4,40 5pentachlorobiphenyl (PCB 126) in rats. Neurotoxicol. Teratol. 21:299–301. Crosby DG, Wong AS, Plimmer JR, Woolson AE (1971) Photodecomposition of chlorinated dibenzop-dioxins. Science 173:748–749. Cummings AM, Metcalf JL, Birnbaum L (1996) Promotion of endometriosis by 2,3,7,8-tetrachlorodibenzo-p-dioxin in rats and mice: time–dose dependence and species comparison. Toxicol. Appl. Pharmacol. 138:131–139. Czuczwa JM, Neissen F, Hites RA (1985) Historical record of polychlorinated dibenzo-p-dioxins in Swiss lake sediments. Chemosphere 14:1175–1179. De Jongh J, Nieboer R, Schroders E, Seinen W, Van den Berg M (1993) Toxicokinetic mixture interactions between chlorinated aromatic hydrocarbons in the liver of the C57BL/6J mouse: 2. Polychlorinated dibenzo-p-dioxins (PCDDs), dibenzofurans (PCDFs) and biphenyls (PCBs). Arch. Toxicol. 67:598–604. DeVito MJ, Birnbaum LS (1995) The importance of pharmacokinetics in determining the relative potency of 2,3,7,8-tetrachlorodibenzo-p-dioxin and 2,3,7,8-tetrachlorodibenzofuran. Fundam. Appl. Toxicol. 24:145–148. DeVito MJ, Thomas T, Martin E, Umbreit TH, Gallo MA (1992a) Antiestrogenic action of 2,3,7,8tetrachlorodibenzo-p-dioxin: tissue-specific regulation of estrogen receptor in CD1 mice. Toxicol. Appl. Pharmacol. 113:284–292. DeVito MJ, Thomas T, Umbreit TH, Gallo MA (1992b) Multi-site regulation of estrogen receptor by 2,3,7,8-tetrachlorodibenzo-p-dioxin. Prog. Clin. Biol. Res. 374:321–336.

REFERENCES

653

DeVito MJ, Birnbaum LS, Farland WH, Gasiewicz TA (1995) Comparisons of estimated human body burdens of dioxinlike chemicals and TCDD body burdens in experimentally exposed animals. Environ. Health Perspect. 103:820–831. DeVito MJ, Diliberto JJ, Ross DG, Menache MG, Birnbaum LS (1997) Dose–response relationships for polyhalogenated dioxins and dibenzofurans following subchronic treatment in mice. I. CYP1A1 and CYP1A2 enzyme activity in liver, lung, and skin. Toxicol. Appl. Pharmacol. 147: 267–280. Devitor MJ, Ross DG, Dupuy AE Jr, Ferrario J, McDaniel DL, Birnbaum S (1998) Dose–response relationships for disposition and hepatic sequestration polyhalogenated dibenzo-p-dioxin, dibenzofurans, and biphenyls following subchronic treatment in mice. Toxicol. Sci. 46: 223–234. Devoogt P, Brinkman UA (1989) Production, properties and usage of polychlorinated biphenyls. In: Kimbrough RD, Jensen AA, editors. Halogenated Biphenyls, Terphenyls, Naphthalenes, Dibenzodioxins and Related Products. Amsterdam, NY:Elsevier, pp.3–45. di Domenico A, Viviano G, Zapponi G (1982) Environmental persistence of 2,3,7,8-TCDD at Seveso. In:Hutzinger O, Frei RW, Merian E, Pocchiari F, editors. Chlorinated Dioxins and Related Compounds. Impact on the Environment. Oxford:Pergamon Press, pp.105–114. Diliberto JJ, Burgin D, Birnbaum LS (1997) Role of CYP1A2 in hepatic sequestration of dioxin: studies using CYP1A2 knock-out mice. Biochem. Biophys. Res. Commun. 236:431. Eadon G, Kaminsky L, Silkworth J, Aldous K, Hilker D, O’Keefe P, Smith R, Gierthy J, Hawley J, Kim N, DeCaprio A (1986) Calculation of 2,3,7,8-TCDD equivalent concentrations of complex environmental contaminant mixtures. Environ. Health Perspect. 70:221–227. Fernandez-Salguero PM, Ward JM, Sundberg JP, Gonzalez FJ (1997) Lesions of aryl-hydrocarbon receptor-deficient mice. Vet. Pathol. 34:605–614. Fingerhut MA, Halperin WE, Marlow DA, Piacitelli LA, Honchar PA, Sweeney MH, Greife AL, Dill PA, Steenland K, Suruda AJ (1991) Cancer mortality in workers exposed to 2,3,7,8-tetrachlorodibenzo-p-dioxin. N. Engl. J. Med. 324:212–218. Flesch-Janys D, Berger J, Gurn P, Manz A, Nagel S, Waltsgott H, Dwyer JH (1995) Exposure to polychlorinated dioxins and furans (PCDD/F) and mortality in a cohort of workers from a herbicide-producing plant in Hamburg, Federal Republic of Germany. Am. J. Epidemiol. 142: 1165–1175. Flesch-Janys D, Steindorf K, Gurn P, Becher H (1998) Estimation of the cumulated exposure to polychlorinated dibenzo-p-dioxins/furans and standardized mortality ratio analysis of cancer mortality by dose in an occupationally exposed cohort. Environ. Health Perspect. 106(Suppl 2): 655–662. Fries GF, Paustenbach DJ (1990) Evaluation of potential transmission of 2,3,7,8-tetrachlorodibenzop-dioxin-contaminated incinerator emissions to humans via foods. J. Toxicol. Environ. Health 29: 1–43. Gasiewicz TA, Rucci G (1984) Cytosolic receptor for 2,3,7,8-tetrachlorodibenzo-p-dioxin. Evidence for a homologous nature among various mammalian species. Mol. Pharmacol. 26:90–98. Gasiewicz TA, Geiger LE, Rucci G, Neal RA (1983) Distribution, excretion, and metabolism of 2,3,7,8-tetrachlorodibenzo-p-dioxin in C57BL/6J, DBA/2J, and B6D2F1/J mice. Drug Metab. Dispos. 11:397–403. Goldey ES, Crofton KM (1998) Thyroxine replacement attenuates hypothyroxinemia, hearing loss, and motor deficits following developmental exposure to Aroclor 1254 in rats. Toxicol. Sci. 45:94–105. Goldey ES, Lau C, Kehn LS, Crofton KM (1996) Developmental dioxin exposure: disruption of thyroid hormones and ototoxicity. 35th Annual Meeting of the Society of Toxicology, Anaheim, CA. The Toxicologist. 30:225.

654

DIOXINS AND DIOXIN-LIKE CHEMICALS

Gordon CJ, Gray LE Jr, Monteiro-Riviere NA, Miller DB (1995) Temperature regulation and metabolism in rats exposed perinatally to dioxin: permanent change in regulated body temperature? Toxicol. Appl. Pharmacol. 133:172–176. Gordon CJ, Yang Y, Gray LE Jr (1996) Autonomic and behavioral thermoregulation in golden hamsters exposed perinatally to dioxin. Toxicol. Appl. Pharmacol. 137:120–125. Gray LE Jr, Kelce WR, Monosson E, Ostby JS, Birnbaum LS (1995) Exposure to TCDD during development permanently alters reproductive function in male Long Evans rats and hamsters: reduced ejaculated and epididymal sperm numbers and sex accessory gland weights in offspring with normal androgenic status. Toxicol. Appl. Pharmacol. 131:108–118. Gray LE, Wolf C, Mann P, Ostby JS (1997) In utero exposure to low doses of 2,3,7,8-tetrachlorodibenzo-p-dioxin alters reproductive development of female Long Evans hooded rat offspring. Toxicol. Appl. Pharmacol. 146:237–244. Gupta A, Schecter A, Aragaki CC, Roehrborn CG (2006) Dioxin exposure and benign prostatic hyperplasia. J. Occup. Environ. Med. 48:708–714. Haag-Gronlund M, Kato Y, Fransson-Steen R, Scheu G, Warngard L (1997) Promotion of enzyme altered foci in female rat 2,3,30 ,4,40 ,5-hexachlorobiphenyl. Toxicol. Appl. Pharmacol. 147:46–55. Haag-Gronlund M, Johansson N, Fransson-Steen R, Hakansson H, Scheu G, Warngard L (1998) Interactive effects of three structurally different polychlorinated biphenyls in a rat liver tumor promotion bioassay. Toxicol. Appl. Pharmacol. 152:153–165. Hahn ME (1998) The aryl hydrocarbon receptor: a comparative perspective. Comp. Biochem. Physiol. Part C Pharmacol. Toxicol. Endocrinol. 121:23–53. Harper N, Connor K, Safe S (1993) Immunotoxic potencies of polychlorinated biphenyl (PCB), dibenzofuran (PCDF) and dibenzo-p-dioxin (PCDD) congeners in C57BL/6 and DBA/2 mice. Toxicology 80:217–227. Hemming H, Flodstrom S, Warngard L, Bergman A, TKronevi T, Nordgren I, Ahlborg UG (1993) Relative tumour promoting activity of three polychlorinated biphenyls in rat liver. Eur. J. Pharmacol. 248:163–174. Hemming H, Bager Y, Flodstrom S, Nordgren I, Kronevi T, Ahlborg UG, Warngard L (1995) Liver tumour promoting activity of 3,4,5,30 ,40 -pentachlorobiphenyl and its interaction with 2,3,7,8tetrachlorodibenzo-p-dioxin. Eur. J. Pharmacol. 292:241–249. Henry EC, Gasiewicz TA (1987) Changes in thyroid hormones and thyroxine glucuronidation in hamsters compared with rats following treatment with 2,3,7,8-tetrachlorodibenzo-p-dioxin. Toxicol. Appl. Pharmacol. 89:165–174. Hooiveld M, Heederik DJ, Kogevinas M, Boffetta P, Needham LL, Patterson DG Jr, Bueno-deMesquita HB (1998) Second follow-up of a Dutch cohort occupationally exposed to phenoxy herbicides, chlorophenols, and contaminants. Am. J. Epidemiol. 147:891–901. Hsu C-C, Yu M-LM, Chen Y-CJ, Gou Y-LL, Rogan WL (1995) The Yu-cheng rice oil poisoning incident. In: Schecter A, editor. Dioxins and Health. New York:Plenum Press, pp.661–684. Huff J, Lucier G, Tritscher A (1994) Carcinogenicity of TCDD: experimental, mechanistic, and epidemiologic evidence. Annu. Rev. Pharmacol. Toxicol. 34:343–372. Hutzinger O, Blumich MJ, van den Berg M, Olie K (1985) IARC Monographs on the Evaluation of Carcinogenic Risks to Humans. Overall Evaluations of Carcinogenicity: An Updating of IARC Monographs Volumes 1 to 42, Suppl. 7. Lyons: IARC. Insensee AR, Jones GE (1971) Absorption and translocation of root and foliage applied 2,4dichlorophenol, 2,7-dichlordibenzo-p-dioxin and 2,3,7,8-tetrachlorodibenzo-p-dioxin. J. Agric. Food Chem. 19:1210–1214. Insensee AR, Jones GE (1975) Distribution of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in an aquatic model ecosystem. Environ. Sci. Technol. 9:668–672.

REFERENCES

655

Jacobson JL, Jacobson SW (1997) Evidence for PCBs as neurodevelopmental toxicants in humans. Neurotoxicology 18:415–424. Jacobson JL, Jacobson SW, Humphrey HE (1990) Effects of in utero exposure to polychlorinated biphenyls and related contaminants on cognitive functioning in young children. J Pediatr. 116: 38–45. Jensen DJ, Getzendaner ME, Hummel RA, Turley J (1983) Residues studies for (2,4,5-trichlorophenoxy) acetic acid and 2,3,7,8-tetrachlorodibenzo-p-dioxin in grass and rice. J. Agric. Food Chem. 31:118–122. Johnson KL, Cummings AM, Birnbaum LS (1997) Promotion of endometriosis in mice by polychlorinated dibenzo-p- dioxins, dibenzofurans, and biphenyls. Environ. Health Perspect. 105:750–755. Kayajanian GM (2002) The J-shaped dioxin dose response curve. Ecotoxicol. Environ. Saf. 51:1–4. Kedderis LB, Mills JJ, Andersen ME, Birnbaum LS (1993) A physiologically based pharmacokinetic model for 2,3,7,8-tetrabromodibenzo-p-dioxin (TBDD) in the rat: tissue distribution and CYP1A induction. Toxicol. Appl. Pharmacol. 121:87–98. Kerkvliet NI (1995) Immunological effects of chlorinated dibenzo-p-dioxins. Environ. Health Perspect. 103(Suppl 9):47–53. Kimbrough RD, Falk H, Stehr P, Fries G (1984) Health implications of 2,3,7,8-tetrachlorodibenzo-pdioxin (TCDD) contamination of residential soil. J. Toxicol. Environ. Health 14:47–93. Kociba RJ, Keyes DG, Beyer JE, Carreon RM, Wade CE, Dittenber DA, Kalnins RP, Frauson LE, Park CN, Barnard SD, Hummel RA, Humiston CG (1978) Results of a two-year chronic toxicity and oncogenicity study of 2,3,7,8-tetrachlorodibenzo-p-dioxin in rats. Toxicol. Appl. Pharmacol. 46: 279–303. Kogevinas M, Becher H, Benn T, Bertazzi PA, Boffetta PH, Bueno-de-Mesquita B, Coggon D, Colin D, Flesch-Janys D, Fingerhut M, Green L, Kauppinen T, Littorin M, Lynge E, Mathews JD, Neuberger M, Pearce N, Saracci R (1997) Cancer mortality in workers exposed to phenoxy herbicides, chlorophenols, and dioxins. An expanded and updated international cohort study. Am. J. Epidemiol. 145:1061–1075. Kohn MC, Sewall CH, Lucier GW, Portier CJ (1996) A mechanistic model of effects of dioxin on thyroid hormones in the rat. Toxicol. Appl. Pharmacol. 136:29–48. Koopman-Esseboom C, Morse DC, Weisglas-Kuperus N, Lutkeschipholt IJ, Van der Paauw CG, Tuinstra LG, Brouwer A, Sauer PJ (1994) Effects of dioxins and polychlorinated biphenyls on thyroid hormone status of pregnant women and their infants. Pediatr. Res. 36:468–473. Lin FH, Stohs SJ, Birnbaum LS, Clark G, Lucier GW, Goldstein JA (1991) The effects of 2,3,7,8tetrachlorodibenzo-p-dioxin (TCDD) on the hepatic estrogen and glucocorticoid receptors in congenic strains of Ah responsive and Ah nonresponsive C57BL/6J mice. Toxicol. Appl. Pharmacol. 108:129–139. Lucier GW, Tritscher A, Goldsworthy T, Foley J, Clark G, Goldstein J, Maronpot R (1991) Ovarian hormones enhance 2,3,7,8-tetrachlorodibenzo-p-dioxin-mediated increases in cell proliferation and preneoplastic foci in a two-stage model for rat hepatocarcinogenesis. Cancer Res. 51:1391– 1397. Luebke RW, Copeland CB, Diliberto JJ, Akubue PI, Andrews DL, Riddle MM, Williams WC, Birnbaum LS (1994) Assessment of host resistance to Trichinella spiralis in mice following preinfection exposure to 2,3,7,8-TCDD. Toxicol. Appl. Pharmacol. 125:7–16. Luebke RW, Copeland CB, Andrews DL (1995) Host resistance to Trichinella spiralis infection in rats exposed to 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD). Fundam. Appl. Toxicol. 24:285–289. Mably TA, Bjerke DL, Moore RW, Gendron-Fitzpatrick A, Peterson RE (1992a) In utero and lactational exposure of male rats to 2,3,7,8-tetrachlorodibenzo-p-dioxin. 3. Effects on spermatogenesis and reproductive capability. Toxicol. Appl. Pharmacol. 114:118–126.

656

DIOXINS AND DIOXIN-LIKE CHEMICALS

Mably TA, Moore RW, Goy RW, Peterson RE (1992b) In utero and lactational exposure of male rats to 2,3,7,8-tetrachlorodibenzo-p-dioxin. 2. Effects on sexual behavior and the regulation of luteinizing hormone secretion in adulthood. Toxicol. Appl. Pharmacol. 114:108–117. Maronpot RR, Foley JF, Takashi K, Goldsworthy T, Clark G, Tritscher A, Portier C, Lucier G (1993) Dose response for TCDD promotion of hepatocarcinogenesis in rats initiated with DEN: histologic, biochemical, and cell proliferation endpoints. Environ. Health Perspect. 101: 634–642. Masuda Y (1995) The Yusho rice oil poisoning incident. In:Schecter A, editor. Dioxins and Health. New York:Plenum Press, pp.633–660. Mayes BA, McConnell EE, Neal BH, Brunner MJ, Hamilton SB, Sullivan TM, Peters AC, Ryan MJ, Toft JD, Singer AW, Brown JF Jr, Menton RG, Moore JA (1998) Comparative carcinogenicity in Sprague-Dawley rats of the polychlorinated biphenyl mixtures Aroclors 1016, 1242, 1254, and 1260. Toxicol. Sci. 41:62–76. Mimura J, Yamashita K, Nakamura K, Morita MT, Takagi TN, Nakao K, Ema M, Sogawa K, Yasuda M, Katsuki M, Fujii-Kuriyama Y (1997) Loss of teratogenic response to 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in mice lacking the Ah (dioxin) receptor. Genes Cells 2:645–654. McConnell EE, Moore JA (1979) Toxicopathology characteristics of the halogenated aromatic hydrocarbons. Ann. N.Y. Acad. Sci. 320:138–150. McGregor DB, Partensky C, Wilbourn J, Rice JM (1998) An IARC evaluation of polychlorinated dibenzo-p-dioxin and polychlorinated dibenzofurans as risk factors in human carcinogenesis. Environ. Health Perspect. 106(Suppl 2):755–760. Moore JA, McConnell EE, Dalgard DW, Harris MW (1979) Comparative toxicity of three halogenated dibenzofurans in guinea pigs, mice and monkeys. Ann. N.Y. Acad. Sci. 320:151–163. Moore RW, Potter CL, Theobald HM, Robinson JA, Peterson RE (1985) Androgenic deficiency in male rats treated with 2,3,7,8-tetrachlorodibenzo-p-dioxin. Toxicol. Appl. Pharmacol. 79:99–111. Moore RW, Parsons JA, Bookstaff RC, Peterson RE (1989) Plasma concentrations of pituitary hormones in 2,3,7,8-tetrachlorodibenzo-p-dioxin-treated male rats. J. Biochem. Toxicol. 4:165– 172. Murray FJ, Smith FA, Nitschke KD, Humiston CGR, Kociba J, Schwetz BA (1979) Three-generation reproduction study of rats given 2,3,7,8- tetrachlorodibenzo-p-dioxin (TCDD) in the diet. Toxicol. Appl. Pharmacol. 50:241–252. National Academy of Sciences (NAS) (1979) PCB transport throughout the environment. In Polychlorinated Biphenyls, Washington, DC:National Academy Press. National Research Council (2006) Health Risks from Dioxin and Related Compounds.Washington, DC:National Academy Press. National Toxicology Program/National Cancer Institute (1982) Carcinogenesis bioassay of 2,3,7,8tetrachlorodibenzo-p-dioxin (CAS No 1746-091-6) in Osborne-Mendel rats and B6C3F1 mice (gavage study). Technical Report 209, U.S. Department of Health, Education and Welfare, Public Health Service, National Institutes of Health, Bethesda, MD. Neubert R, Golor G, Stahlmann R, Helge H, Neubert D (1992) Polyhalogenated dibenzo-p-dioxin and dibenzofurans and the immune system. 4. Effects of multiple dose treatment with 2,3,7,8tetrachlorodibenzo-p-dioxin (TCDD) on peripheral lymphocyte subpopulations of a non-human primate (Callithrix jacchus). Arch. Toxicol. 66:250–259. Olson JR, McGarrigle BP (1991) Comparative developmental toxicity of 2,3,7,8-tetrachlorodibenzop-dioxin. Chemosphere 25:71–74. Ott MG, Zober A (1996) Cause specific mortality and cancer incidence among employees exposed to 2,3,7,8-TCDD after a 1953 reactor accident. Occup. Environ. Med. 53:606–612. Pinsky PF, Lorber MN (1998) A model to evaluate past exposure to 2,3,7,8-TCDD. J. Expo. Anal. Environ. Epidemiol. 8(2):187–206.

REFERENCES

657

Pitot HC, Goldsworthy T, Campbell HA, Poland A (1980) Quantitative evaluation of the promotion by 2,3,7,8-tetrachlorodibenzo-p-dioxin of hepatocarcinogenesis from diethylnitrosamine. Cancer Res. 40:3616–3620. Pohjanvirta R, Tuomisto J (1994) Short-term toxicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin in laboratory animals: effects, mechanisms, and animal models. Pharmacol. Rev. 46:483–549. Poland A, Glover E (1980) 2,3,7,8-Tetrachlorodibenzo-p-dioxin: segregation of toxicity with the Ah locus. Mol. Pharmacol. 17:86–94. Poland A, Glover E (1990) Characterization and strain distribution pattern of murine A receptor specified by the Ahd and Ahb alleles. Mol. Pharmacol. 38:306–312. Poland A, Phalen D, Glover E (1982) Tumour promotion by TCDD in skin of HRS/J hairless mice. Nature 300:271–273. Portier CJ, Sherman CD, Kohn M, Edler L, Kopp-Schneider A, Maronpot RM, Lucier G (1996) Modeling the number and size of hepatic focal lesions following exposure to 2,3,7,8-TCDD. Toxicol. Appl. Pharmacol. 138:20–30. Powell-Coffman JA, Bradfield CA, Wood WB (1998) Caenorhabditis elegans orthologs of the aryl hydrocarbon receptor and its heterodimerization partner the aryl hydrocarbon receptor nuclear translocator. Proc. Natl. Acad. Sci. USA 95:2844–2849. Rao MS, Subbaro V, Psada JD, Scarpelli DC (1988) Carcinogenicty of 2,3,7,8-tetrachlorodibenzop-dioxin in the Syrian hamster. Carcinogensesis 9:1677–1679. Rappe C, Nygren M, Lindstrom G, Buser HR, Blaser O, Wuthrich C (1987) Polychlorinated dibenzofurans and dibenzo-p-dioxins and other chlorinated contaminants in cow milk from various locations in Switzerland. Environ. Sci. Technol. 21:964–970. Reier SE, Martin DC, Bowman RE, Dmowski WP, Becker JL (1993) Endometriosis in Rhesus monkeys (Macca mulatta) following chronic exposure to 2,3,7,8-tetrachlordibenzo-p-dioxin. Fundam. Appl. Toxicol. 21:433–441. Rogan WJ, Gladen BC (1991) PCBs, DDE, and child development at 18 and 24 months. Ann. Epidemiol. 1:407–413. Rogan WJ, Gladen BC (1992) Neurotoxicology of PCBs and related compounds. Neurotoxicology 13: 27–35. Rogan WJ, Gladen BC (1993) Breast-feeding and cognitive development. Early Hum. Dev. 31:181–193. Safe S (1990) Polychlorinated biphenyls (PCBs), dibenzo-p-dioxins (PCDDs), dibenzofurans (PCDFs), and related compounds: environmental and mechanistic considerations which support the development of toxic equivalency factors (TEFs). Crit. Rev. Toxicol. 21:51–88. Safe SH (1994) Polychlorinated biphenyls (PCBs): environmental impact, biochemical and toxic responses, and implications for risk assessment. Crit. Rev. Toxicol. 24:87–149. Safe S, Astroff B, Harris M, Zacharewski T, Dickerson R, Romkes M, Biegel L (1991) 2,3,7,8Tetrachlorodibenzo-p-dioxin (TCDD) and related compounds as antiestrogens: characterization and mechanism of action. Pharmacol. Toxicol. 69:400–409. Schantz SL, Bowman RE (1989) Learning in monkeys exposed perinatally to 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD). Neurotoxicol. Teratol. 11:13–19. Schecter A, Startin J, Wright C, Kelly M, Papke O, Lis A, Ball M, Olson JR (1994a) Congener-specific levels of dioxins and dibenzofurans in U.S. food and estimated daily dioxin toxic equivalent intake. Environ. Health Perspect. 102:962–966. Schecter A, Stanley J, Boggess K, Masuda Y, Mes J, Wolff M, Furst P, Furst C, Wilson-Yang K, Chisholm B (1994b) Polychlorinated biphenyl levels in the tissues of exposed and nonexposed humans. Environ. Health Perspect. 102(Suppl 1):149–158. Schecter A, Furst P, Furst C, Papke O, Ball M, Ryan JJ, Hoang DC, Le CD, Hoang TQ, Cuong HQ et al. (1994c) Chlorinated dioxins and dibenzofurans in human tissue from general populations: a selective review. Environ. Health Perspect. 102(Suppl 1):159–171.

658

DIOXINS AND DIOXIN-LIKE CHEMICALS

Schecter A, Papke O, Lis A, Ball M, Ryan JJ, Olson JR, Li L, Kessler H (1996) Decrease in milk and blood dioxin levels over two years in a mother nursing twins: estimates of decreased maternal and increased infant dioxin body burden from nursing. Chemosphere 32:543–549. Schrenk D, Buchmann A, Dietz K, Lipp HP, Brunner H, Sirma H, Munzel P, Hagenmaier H, Gebhardt R, Bock KW (1994) Promotion of preneoplastic foci in rat liver with 2,3,7,8-tetrachlorodibenzo-pdioxin, 1,2,3,4,6,7,8-heptachlorodibenzo-p-dioxin and a defined mixture of 49 polychlorinated dibenzo-p-dioxins. Carcinogenesis 15:509–515. Science Advisory Board (1995) Re-evaluating Dioxin: Science Advisory Board’s Review of EPA’s Reassessment of Dioxin and Dioxin-Like Compounds. EPA-SAB-EC-95-021. USEPA: Washington, DC. Science Advisory Board (2001) Dioxin Reassessment—An SAB Review of the Office of Research and Development’s Reassessment of Dioxin. EPA-SAB-EC-01-006. USEPA: Washington, DC. Seefeld MD, Corbett SW, Keesey RE, Peterson RE (1984) Characterization of the wasting syndrome in rats treated with 2,3,7,8-tetrachlorodibenzo-p-dioxin. Toxicol. Appl. Pharmacol. 73:311–322. Silkworth JB, Cutler DS, Sack G (1989) Immunotoxicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin in a complex environmental mixture from the Love Canal. Fundam. Appl. Toxicol. 12:303–312. Smialowicz RJ, Riddle MM, Williams WC, Diliberto JJ (1994) Effects of 2,3,7,8-tetrachlorodibenzop-dioxin (TCDD) on humoral immunity and lymphocyte subpopulations: differences between mice and rats. Toxicol. Appl. Pharmacol. 124:248–256. Spink DC, d Lincoln DW, Dickerman HW, Gierthy JF (1990) 2,3,7,8-Tetrachlorodibenzo-p-dioxin causes an extensive alteration of 17 beta-estradiol metabolism in MCF-7 breast tumor cells. Proc. Natl. Acad. Sci. USA 87:6917–6921. Starr TB (2002) Significant shortcomings of the U.S. Environmental Protection Agency’s latest draft risk characterization for dioxin-like compounds. Toxicol. Sci. 64:7–13. Tanabe S, Tatsukawa R (1986) Distribution, behavior, and load of PCBs in the oceans. In:Waid JS, editor. PCBs and the Environment, Vol. 1.Boca Raton, FL:CRC Press, pp.143–161. Tatsukawa R (1976) PCB pollution of the Japanese environment. In:Higuchi K, editor.PCB Poisoning and Pollution.Tokyo:Kodansha, pp.147–179. Theobald HM, Peterson RE (1997) In utero and lactational exposure to 2,3,7,8-tetrachlorodibenzop-dioxin: effects on development of the male and female reproductive system of the mouse. Toxicol. Appl. Pharmacol. 145:124–135. Tiedje JM, Quensen JF 3rd, Chee-Sanford J, Schimel JP, Boyd SA (1993) Microbial reductive dechlorination of PCBs. Biodegradation 4:231–240. Tritscher AM, Seacat AM, Yager JD, Groopman JD, Miller BD, Bell D, Sutter TR, Lucier GW (1996) Increased oxidative DNA damage in livers of 2,3,7,8-tetrachlorodibenzo-p-dioxin treated intact but not ovariectomized rats. Cancer Lett. 98(2):219–225. Turtletaub KW, Felton JS, Gledhill BL, Vogel JS, Southon JR, Caffee MW, Finkel RC, Nelson DE, Proctor ID, David JC (1990) Accelerator mass spectrometry in biomedical dosimetry: Relationship between low-level exposure and covalent binding of heterocyclic amine carcinogens to DNA. Proc. Natl. Acad. Sci. USA 87:5288–5292. United States Environmental Protection Agency (1994) Estimating exposures to dioxin-like compounds. Vol II; Properties, sources, occurrence and background levels. External review draft. Office of Research and Development, USEPA, Washington, DC. EPA/66/6-88/005Cb. United States Environmental Protection Agency (1997) Evaluation of emissions from the open burning of household waste in barrels. Volume 1. Technical Report. USEPA, Washington, DC. EPA-600/r-97-134a. Van den Berg M, van Wijnen J, Wever H, Seinen W (1989) Selective retention of toxic polychlorinated dibenzo-p-dioxins and dibenzofurans in the liver of the rat after intravenous administration of a mixture. Toxicology 55:173–182.

REFERENCES

659

Van den Berg M, De Jongh J, Poiger H, Olson JR (1994) The toxicokinetics and metabolism of polychlorinated dibenzo-p-dioxins (PCDDs) and dibenzofurans (PCDFs) and their relevance for toxicity. Crit. Rev. Toxicol. 24:1–74. Van den Berg M, Birnbaum L, Bosveld ATC, Brunstrom B, Cook P, Feeley M, Giesy JP, Hanberg A, Hasegawa R, Kennedy SW, Kubiak T, Larsen JC, van Leeuwen FXR, Liem AKD, Nolt C, Peterson RE, Poellinger L, Safe S, Schrenk D, Tillitt D, Tysklind M, Younes M, Wærn F, Zacharewski T (1998) Toxic equivalency factors (TEFs) for PCBs, PCDDs, PCDFs for humans and wildlife. Environ. Health Perspect. 106:775–792. van der Plas SA, de Jongh J, Faassen-Peters M, Scheu G, van den Berg M, Brouwer A (1998) Toxicokinetics of an environmentally relevant mixture of dioxin-like PHAHs with or without a non-dioxin-like PCB in a semi-chronic exposure study in female Sprague Dawley rats. Chemosphere 37:1941–1955. Viluksela M, Stahl BU, Birnbaum LS, Schramm KW, Kettrup AA, Rozman KK (1997) Subchronic/ chronic toxicity of 1,2,3,4,6,7,8-heptachlorodibenzo-p- dioxin (HpCDD) in rats. Part I. Design, general observations, hematology, and liver concentrations. Toxicol. Appl. Pharmacol. 146:207– 216. Viluksela M, Stahl BU, Birnbaum LS, Schramm KW, Kettrup A, Rozman KK (1998a) Subchronic/ chronic toxicity of a mixture of four chlorinated dibenzo-p-dioxins in rats. I. Design, general observations, hematology, and liver concentrations. Toxicol. Appl. Pharmacol. 151:57–69. Viluksela M, Stahl BU, Birnbaum LS, Rozman KK (1998b) Subchronic/chronic toxicity of a mixture of four chlorinated dibenzo-p-dioxins in rats. II. Biochemical effects. Toxicol. Appl. Pharmacol. 151:70–78. Wang X, Santostefano MJ, Evans MV, Richardson VM, Diliberto JJ, Birnbaum LS (1997) Determination of parameters responsible for pharmacokinetic behavior of TCDD in female SpragueDawley rats. Toxicol. Appl. Pharmacol. 147:151–168. Weisglas-Kuperus N, Sas TC, Koopman-Esseboom C, van der Zwan CW, De Ridder MA, Beishuizen A, Hooijkaas H, Sauer PJ (1995) Immunologic effects of background prenatal and postnatal exposure to dioxins and polychlorinated biphenyls in Dutch infants. Pediatr. Res. 38:404–410. Whitlock JP Jr, Okino ST, Dong L, Ko HP, Clarke-Katzenberg R, Ma Q, Li H (1996) Cytochromes P450 5: induction of cytochrome P4501A1: a model for analyzing mammalian gene transcription. FASEB J. 10:809–818. Zober A, Ott MG, Messerer P (1994) Morbidity follow up study of BASF employees exposed to 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) after a 1953 chemical reactor incident. Occup. Environ. Med. 51:479–486.

18 ENDOCRINE ACTIVE CHEMICALS: BROADENING THE SCOPE* Kathryn R. Mahaffey, Shirlee W. Tan, K. Christiana Grim Jessica C. Meiller and Donald R. Bergfelt

18.1 INTRODUCTION In the early-to-mid 1990s, significant evidence that chemicals were affecting the endocrine system of wildlife drew attention from both the public and scientific communities. Because of this, intensive laboratory-based studies were designed to evaluate chemicals for potential endocrine activity across taxonomic groups (among others see U.S. EPA, 1998; Reiter et al., 1998; Cooper and Kavlock, 1997; Colborn et al., 1993). Initial research emphasized the potential of chemical agents to alter the reproductive process that primarily reflected earlier experiences in reproductive toxicology (Cooper and Kavlock, 1997). More than a decade later, attention has expanded to a broader domain of endocrine effects such as cellular and molecular effects, multigenerational effects, and many of the physiological/pathological responses such that endocrine disrupting chemicals have become elucidated. Increasingly, it is recognized that the investigation of environmental chemicals cannot rely solely on utilization of in vitro systems (e.g., isolated hormone receptors), but rather it is the use of these data with real-world exposure, epidemiological, and in vivo testing data from whole animals that depicts a more accurate picture of the impact of chemical exposures on the endocrine system. The endocrine system is an elegant regulatory network that is necessary for maintaining physiological homeostasis, and which reacts to internal (e.g., blood pressure) and external cues (e.g., daylight through the pineal gland). It is composed of organ systems that interact with one another through hormone signals, which are, in turn, regulated through highly intricate feedback mechanisms. Generally, the components of the endocrine system include neuroendocrine/endocrine regulatory feedback mechanisms and hormone production, *

The findings and conclusions in this chapter are those of the authors and not of the U.S. Environmental Protection Agency. The chapter is not being formally disseminated by U.S. EPA, so it should not be construed as representing any Agency determination or policy.

Environmental Toxicants, Third Edition Edited by Morton Lippmann Copyright
2009 John Wiley & Sons, Inc.

661

662

ENDOCRINE ACTIVE CHEMICALS: BROADENING THE SCOPE

Regulatory/feedback mechanisms Neuroendocrine system

}

}

Peripheral endocrine system

A Hypothalamus Releasing factors

B Pituitary

-Interfere with regulatory mechanisms (positive and negative feedback) of circulating hormone concentrations -Interfere with potentiators of hormone regulation *Proper function of the neuroendocrine system requires the same components as the peripheral system: hormone production, transport, conversion, recognition, internalization, action, and metabolism. Therefore, endocrine interference can occur within the neuroendocrine system through similar mechanisms as those listed below.

Stimulating factors

C Hormone production

D Hormone transport

E Hormone conversion

-Interfere with signal to initiate hormone synthesis -Upregulate or downregulate hormone synthesis -Alter normal hormone structure hence altering function -Interfere with storage and release of hormones

-Interfere with production of binding/transport proteins -Change binding affinities to serum and cytoplasmic transport proteins -Alter ratio of specific binding proteins available to natural hormones

-Interfere with normal conversion of hormones to active/inactive forms -Interfere with conversion of hormones to other hormones (e.g., steroidogenesis pathway)

-Compete with membrane and nuclear receptors of hormones -Up-/downregulate production of hormone receptors -Act as a mimic or antagonist of natural hormones -Change in hormone receptor structure and function F -Activation/deactivation of postreceptor pathways Hormone recognition, internalization and Action -Interfere with postreceptor pathways (DNA transcription/translation and protein synthesis) and action of endogenous hormones -Epigenetic changes (e.g., changes in methylation patterns that influence gene expression) to target cells increasing or decreasing sensitivity of cells to natural hormones, or permanently changing genetic material passed to subsequent generations resulting in latent effects

G

Metabolism and elimination

-Interfere with hormone biotransformation, conjugation, and elimination

H Latent and population effects

FIGURE 18.1

Points of potential endocrine activity of chemical compounds.

transport, conversion, receptor recognition, and internalization, target organ action, and metabolism and elimination (Fig. 18.1). These endocrine components are vastly conserved over vertebrate taxa. Hence, research and information on hormonal activity, regulation or disruption in one vertebrate species can generally be applied to other vertebrate species.

INTRODUCTION

663

Concerning endocrine disruption, often the most definitive information available to make causal relationships between environmentally induced endocrine activity and physiological alteration is elucidated through controlled studies in laboratory animals and wildlife. These animal studies, coupled with the high homology between endocrine function and regulation across vertebrate species, allow greater confidence when making causal relationships between exposure to endocrine toxicants and disease in humans. While it is convenient to break down the endocrine system into distinct entities for specific functions for comprehension, such as the hypothalamic–pituitary–thyroid (HPT) axis, the hypothalamic–pituitary–gonadal (HPG) axis, or the hypothalamic–pituitary–adrenal (HPA) axis, the reality is that all components of the endocrine system are integrative, and all react and respond to one another. For instance, regulation of the reproductive system is dominated by the hypothalamic–pituitary–gonadal axis, which controls sexual maturation, gamete maturation, ovulation and pregnancy, to name a few. However, thyroid hormone is integral to regulating reproductive cyclicity in some species (Ben Saad and Maurel, 2004), hence interference in thyroid hormone action can ultimately affect the “central” reproductive axis and subsequently the reproductive capacity of an individual. In addition, several hormones have multiple and supportive functions and may be called to action simultaneously to regulate several different processes to accomplish an end result. These interrelationships within and outside of the endocrine system are important to consider when assessing the effects of endocrine active chemicals (EACs), or chemicals that interact with and influence endocrine function, either directly or indirectly on clinical disease. Human and animals are exposed to chemicals through several pathways. These exposure pathways are no different for EACs than for other chemicals. Examples of EAC exposure pathways include chemicals that travel through air such as mercury and dioxin; agrochemicals or veterinary use chemicals and hormones applied on farms and in concentrated animal feeding operations (CAFOs); and the use of pharmaceuticals and personal care products that enter our sewage treatment plants, effluents, surface waters, seawaters, groundwater, and some drinking waters (Fent et al., 2006). Much of the initial focus on EACs was on estrogen or estrogen mimicking chemicals (Segner et al., 2003), and some of the most potent of the EACs are considered to be natural and artificial steroid estrogens of anthropogenic origin, which are commonly detected in sewage effluents (Johnson et al., 2007). The biological potency of these differ from, and are influenced by, various biodegradation processes (D’Ascenzo et al., 2003). Depending on the type of sewage treatment, estrone and estradiol can be removed from as little as 5% to 14% to as much as 85% to 96% (Braga et al., 2005). These hormones, along with pharmaceuticals and personal care products, are introduced into the aquatic environment primarily by both untreated and treated sewage (Daughton and Ternes, 1999). Given the abundance of possible exposure pathways in the environments of humans and animals, it is critical that methods of detection for potential effects of these chemicals be developed and utilized. However, such a task is not straightforward. For example, the endocrine effects produced by pharmaceuticals in water typically involve species not targeted by the original investigators who developed the chemicals. Consequently, evaluation of wildlife species, ex situ or in situ has often not been part of the original assessment (Fent et al., 2006). Also problematic is that many chemicals were tested only in in vitro systems. Serious drawbacks of such systems, as well as those of applied quantitative structure–activity relationships (QSARs) have been noted (Fent et al., 2006). Likewise, the problem of mixtures is rarely addressed. Widespread environmental contamination has been documented in several other countries including France, Germany, Canada, Brazil, and the United States (Cargouet et al., 2004; Ternes et al.,

664

ENDOCRINE ACTIVE CHEMICALS: BROADENING THE SCOPE

1999; Boyd et al., 2003). Thus, the major task is to understand how EACs affect species and how such effects can be detected. EACs can interfere at several levels within an organism (Fig. 18.1), either directly (e.g., xenoestrogens interacting with estrogen receptors) or indirectly (e.g., through cytotoxicity of nonendocrine organs leading to endocrine dysfunction), potentially with the same outcome. They can also induce latent, irreversible effects that are manifested in adulthood according to the concepts of developmental origins of health and disease (DOHaD; Lau and Rogers, 2004; and in subsequent generations through transgenerational effects (Campbell and Perkins, 1988; Anway and Skinner, 2006). The distinction between the latent effects of chemical exposure on the offspring associated with DOHaD and transgenerational modes of action (MOA) is that, in the former, exposure of the pregnant parent results primarily in prenatal exposure through an in utero/in ovo route whereas, in the latter exposure of one or both parents occurs before mating or spawning such that the effects are thought to be passed along to the offspring through genomic or epigenomic routes. Such effects are difficult to study because they challenge the conventional approach to toxicology and risk assessment. They can be permanent, self-sustaining, are not necessarily dose-dependent, nor influenced by threshold levels of EACs, and they may not be expressed until years, or generations after the insult occurred, making conventional risk assessment difficult and regulation reliant on reproducible assay systems. Latent and transgenerational manifestations of disease associated with DOHaD, and the role that epigenomics plays in understanding causal relationships between exposure to endocrine toxicants and latent disease, are emerging areas of research within the field of endocrine toxicology. In this review, we (1) provide an overview of potential end points and clinical signs associated with exposure to EACs in myriad species using thyroid hormone as an example; (2) provide an example of a classic environmental contaminant, in this case inorganic lead, and its influence on a clinically important endocrine system that is rarely considered in a paradigm that focuses on reproductive biology, specifically the vitamin D/endocrine system; and (3) address the new and exciting disciplines of DOHaD and epigenetic and transgenerational effects of EACs using current examples from humans, laboratory species, and wildlife. 18.2 BIOMARKERS: TERMINOLOGY FROM VARIOUS DISCIPLINES There are myriad disciplines that are involved in understanding the effects of EACs in humans and wildlife. Due to the multidisciplinary nature of this field, there are also differences in terminology, especially as related to the term biomarker. Interestingly, in the literature associated with human epidemiology (among others, see discussion by Munoz and Gange, 1998), medicine, and health, the term biomarker (especially biomarkers of exposure, biomarkers of effect, and biomarkers of susceptibility) has arisen from terminology used by the World Health Organization (for a summary report, see WHO, 2001) and by various committees organized under the National Research Council of the National Academy of Sciences (NRC/ NAS) in the United States (among other see NAS, 2006). The term is defined as follows: Biomarker of exposure: The chemical or its metabolite or the product of an interaction between a chemical and some target molecule or cell that is measured in a compartment in an organism. Biomarker of effect: A measurable biochemical, physiologic, behavioral, or other alteration in an organism that, depending on the magnitude, can be recognized as associated with an established or possible health impairment or disease.

BIOMARKERS: TERMINOLOGY FROM VARIOUS DISCIPLINES

665

Biomarker of susceptibility: An indicator of an inherent or acquired ability of an organism to respond to the challenge of exposure to a specific chemical substance. This terminology is often used in studies evaluating an association between exposures to environmental chemicals and health outcomes (typically adverse) that could include, but are in no manner limited to, endocrine effects. Also, older literature exists that utilized biomarkers of exposure (e.g., blood lead monitoring) and principally reports on assessments conducted as part of occupational and public health surveillance. Thus, the general terms for biomarkers, as these terms are used in public health, environmental epidemiology (Munoz and Gange, 1998), and environmental health refer to biological indicators (NAS, 2006) including behavior, biochemical, molecular, genetic, immunologic, or physiologic signals of events in biologic systems. These events are considered (NAS, 2006) to follow a continuum between indicators of external exposure to a chemical and resultant clinical effect (Schulte and Perera, 1993). In addition, MOA is not necessarily a critical feature of biomarkers, as used in human public health, epidemiology, and surveillance. Alternatively, there are other definitions for the term biomarkers in fields of study that focus on the ecological effects of environmental chemicals on populations of nondomesticated animals and wildlife. In the ecological toxicology field, biomarkers are generally defined as biochemical, physiological, or histological indicators of either exposure or effects of xenobiotic chemicals (Forbes et al., 2006; Adams and Rowland, 2003). As discussed in the Handbook of Toxicology: Indicators or biomarkers can be defined at any level of biological organization, including changes manifested as enzyme content or activity, DNA adducts, chromosomal aberrations, histopathological alterations, immune-system effects, reproductive effects, physiological effects, and fertility at the molecular and individual level, as well as size distributions, diversity in disease, and functional parameters at the population and ecosystem level.

Another definition of environmental biomarkers that has been used in ecotoxicology is that proposed by Shugart et al. (1992): “. . . a xenobiotically induced variation in cellular or biochemical components or processes, structures, or functions that is measurable in a biological system or sample.” Older biomarkers were focused on measures of whole animal physiology or biochemistry, whereas the newer biomarkers include measures of changes in gene expression. The utility of biomarkers in ecotoxicology is controversial, depending on the intended outcome of their use. In the ecological literature, ecological biomarkers of exposure used in environmental biomonitoring in the field are important for tracking substances and mixtures of concern (Hutchinson et al., 2005) and can provide a linkage between field and laboratory studies with a variety of organisms. Usage of these terms in ecological studies differs from the way these are used to report human public health, epidemiology, and biomonitoring trends. For additional information see also Kendall et al., 1998. Environmental biomarkers are expected to be mechanistically relevant, which may not necessarily be the basis for those used in public health, and reproducible (Hutchinson et al., 2005). The term biomarker has historically been used differently by field biologists and by laboratory scientists due to the desired outcome from their respective work. To accommodate both fields, many criteria have been developed to encompass all potential characteristics of a biomarker (for review see Forbes et al., 2006). The terms biomarker and biomonitoring differ

666

ENDOCRINE ACTIVE CHEMICALS: BROADENING THE SCOPE

within wildlife and ecology fields and are distinct from those used in human health disciplines so that no one definition is easily consistent across disciplines. This likely stems from a differential focus on individuals (humans) versus populations (wildlife). Furthermore, in the laboratory the term biomarker is rarely used, rather end point or clinical sign is more common. In addition, for many endocrine-relevant end points, the specific MOAs have not yet been identified, and hence these end points cannot be considered biomarkers of either exposure or effect. They are more general indicators possibly involving multiple MOAs in the endocrine system. Therefore, for the purpose of this chapter and the discussion on methods used to detect chemical exposures and effects, we have chosen to use end point as a general term to relay a measure of detection that is more quantitative (e.g., hormone concentration, fecundity). In this sense, an end point is any measurable parameter that is indicative of EAC exposure or effect. The term clinical sign is used to describe less quantitative clinical observations (e.g., head tilt).

18.3 END POINTS AND CLINICAL SIGNS ASSOCIATED WITH ENDOCRINE ACTIVITY End points available to assess EACs in the laboratory, field, and clinic vary across species and endocrine organ systems with respect to availability, accessibility, reliability, sensitivity, specificity, and general ease of use. Some such end points can be used to indicate changes in the endocrine system, whereas other end points are more general and not specific to the endocrine system but rather serve as indicators or clinical signs that can be linked to abnormalities of the endocrine system. Some general indicators in mammals of adverse effects from EACs include irregular cyclicity (in females), the formation of nonfunctional gametes (i.e., abnormal sperm), and infertility. Similar end points for endocrine activity in other taxa groups (i.e., fish, birds, amphibians, and invertebrates) include survival and the size of adults, fecundity (number of eggs spawned) or fertilization success (number of fertile eggs produced), hatchability or hatching success (number of fertile eggs that produce offspring), and offspring survivability (e.g., occurrence of malformations in hatched animals) (Hutchinson, 2004). For birds and fish, the sex ratio of offspring, the fertility of offspring at maturity, and the number of eggs produced by offspring can be indicative of altered reproductive viability of the offspring (Touart, 2004). Also, a number of end points (i.e., sex ratios, secondary sex characteristics, sexual behavior, and growth during early life stages) can be used across taxa (Hutchinson et al., 2005; Ankley and Johnson, 2004). Although some end points may differ between humans and other taxa, in this chapter, where possible, end points are compared across species to illustrate common markers of endocrine activity. Throughout this section, end points or studies that involve invertebrate species are described. The authors recognize that some biological systems in invertebrates are not conserved in vertebrates. However, many similarities have been established between invertebrates and vertebrates in the area of endocrinology such as developmental and neurohormonal effects and end points. Still, more work needs to be done on the extrapolation of endocrine effects from one species to another and certainly from invertebrate to vertebrate species, as well as the identification of critical differences between species that may make one species more sensitive to certain chemicals than others. At present, it is difficult to make definitive causal links between taxa. A chemical’s effects in one species are not always translatable to other species, and this is an area where more research and data collection across taxa would be helpful.

END POINTS AND CLINICAL SIGNS ASSOCIATED WITH ENDOCRINE ACTIVITY

667

Some end points used to assess endocrine activity are general and can sometimes be caused by acting on a nonendocrine organ, leading to a downstream response by the endocrine system (i.e., an indirect effect). For this reason, general information is useful for population modeling; however, more diagnostic indicators of endocrine responses specific to MOAs can also accompany and hopefully be correlated with such general measures. Specific biomarkers of exposures to EACs include circulating levels of proteinaceous and steroid hormones, associated enzymes, and binding proteins. Also, the expression of secondary sex characteristics, histopathology of endocrine tissues, and vitellogenin (VTG) concentrations are considered to be specific end points that can be associated with exposure to EACs (Hutchinson et al., 2005). Some measurements for such end points can be accomplished with protein concentration measures and enzyme activity, while others are accomplished by measuring gene and protein expression. 18.3.1

Molecular, Cellular and Biochemical End Points

The following section describes examples of, but certainly not all, end points relevant to exposure to EACs in humans and other taxa groups. The end points discussed have been used to assess endocrine effects; however we emphasize that other end points of endocrine activity exist that are not listed here. A variety of molecular level assays exist to assess endocrine responses to chemical exposures (see Table 18.1). These assays often measure, for example, enzyme activities, mRNA synthesis, hormone concentrations, and changes in cellular structure. One enzyme class that provides a useful measure of function following chemical exposure is the hydroxysteroid dehydrogenases (HSDs), key enzymes in the steroidogenesis pathway that are involved in estradiol production. HSDs are good biomarkers because there is little natural variation in their activity levels (Kumar et al., 2000); however, different chemicals often result in a range of affinities for different HSDs and they may therefore be chemical specific (Rotchell and Ostrander, 2003). Another way to detect abnormal gene expression after exposure to chemicals is to measure mRNA production of key proteins. The mRNA synthesis of gonadotropins, which are derived from the pituitary gland and control gonad development and sex steroid synthesis, is disrupted by EACs, which in turn results in disrupted gonadal development (Harries et al., 2000). Like mRNA levels, changes in concentrations of key proteins within an organism can also be useful indicators of how a chemical affects homeostasis. Levels of neurohormones, such as vasotocin, gonadotropinreleasing hormone (GnRH), serotonin, corticotropin releasing factor (CRF), are relevant to effects following exposure to EACs. Some tests that are run routinely to assess thyroid activity in humans include tests for total and free thyroid hormones thyroxine (T4) and triiodothyronine (T3), and thyroid stimulating hormone (TSH) or thyrotropin; antithyroid peroxidase antibodies; and a marker for iodine status (e.g., urine iodine/creatinine). Altered steroid biosynthesis (e.g., aromatase) usually precedes histopathological changes and can therefore provide an early warning of endocrine effects (Rotchell and Ostrander, 2003; Guiguen et al., 1999). However, many controlled studies of exposure are needed to provide a range of compounds to build a profile of the catalytic properties for each biosynthesis enzyme for selected species of interest. Diagnostic tests that assess the hypothalamic–pituitary–gonadal axis in humans have varying degrees of sensitivity and specificity. Measurements of serum concentrations of follicle-stimulating hormone (FSH), luteinizing hormone (LH), free and total testosterone, and inhibin in men are used clinically, in part, to diagnose the basis of reduced spermatogenesis. In women, levels of sulfated metabolite of dehydroepiandrosterone (DHEA-S) are

668

Major key hormones: progesterone, testosterone, estradiol Major key enzymes: HSDs, aromatise Thyroid hormone

C

C D

Major key hormones: T4 (thyroxine), T3 (triiodothyronine) in thyroid gland Major key enzyme: thyroid peroxidase Other hormone production measures: e.g., vitamin D in kidney Hormone Transport Serum/plasma hormone-binding proteins (e.g., alpha-fetoprotein; AFP and sex hormone binding globulin, SHBG) Thyroid hormone transport proteins: thyroxin-binding globulin (75%), transthyretin (15%), serum albumin Adrenal: corticosteroid-binding globulin (CBG), cortisol-binding protein (transcortin), serum albumin

Steroidogenesis

C

B

Neurohormones: vasotocin, GnRH, catecholamines, CRF, serotonin levels; ecdysteroid (molting hormone) receptor binding (invertebrates); ecdysone receptor (EcR) and juvenile growth hormone (invertebrates); Ala-Pro-Gly-Trp (ARGW) amide levels Pituitary Hormones: prolactin, LH, FSH, TSH, ACTH levels

Examples of Potential Endocrine End Points and Clinical Signs

Molecular and Cellular End Points

A, B

Letter Linked to Fig. 18.1

TABLE 18.1

(Fraser and Kodicek, 1970) (Davidson et al., 2006; OECD, 2006; Meulenberg, 2002)

(OECD, 2006; Bernal, 2005; Braverman et al., 2005; Eskiocak et al., 2005; Mukhi et al., 2005; Zaki et al., 2004; Gray et al., 2004; Laws et al., 2000; Stoker et al., 2000)

(Eskiocak et al., 2005; Zaki et al., 2004; Gray et al., 2004; Langer et al., 2003; Laws et al., 2000; Stoker et al., 2000) (Lavado et al., 2006a, 2006b; Ottinger et al., 2005; Gray et al., 2004; Rotchell and Ostrander, 2003; Odum and Ashby, 2002; McMaster et al., 2001; Kumar et al., 2000; Stoker et al., 2000; Gray, 1998; Soto et al., 1992)

(Ottinger et al., 2005; DeFur, 2004; Gagne and Blaise, 2003; Khan et al., 2001; Gao et al., 2000)

Reference(s)

669

Estrogen, progesterone, thyroid, androgens receptor binding Hormone Internalization and Action:

F

(Rotchell and Ostrander, 2003; Kester et al., 2002)

(Eidem et al., 2006; Ghekiere et al., 2006; Puinean and Rotchell, 2006; Nilsen et al., 2004; Tatarazako et al., 2004; Inui et al., 2003; Islinger et al., 2003; Rotchell and Ostrander, 2003; Hemmer et al., 2002; Oberdorster, 2001; Kanamori, 2000; Kumar et al., 2000; Panter et al., 1999; Celius and Walther, 1998; Arukwe et al., 1997; Flouriot et al., 1995)

(Gray et al., 2004; Vetillard et al., 2003; Bowman et al., 2002; Gray, 1998; Flouriot et al., 1995; Pakdel et al., 1991)

(Peeters et al., 2005; Fenske and Segner, 2004; Gray et al., 2004; Rotchell and Ostrander, 2003; Stoker et al., 2000)

The letters (A–H) in the table correspond to the letters (A–H) associated with the components of the endocrine system as illustrated in Fig. 18.1. The end points and clinical signs listed in the table are examples of end points that have been used in some endocrine assays; the list is not intended to be comprehensive or to include all end points of endocrine activity. This table combines human end points with those from other taxa groups.

G

Enzymes: deiodinases, aromatases, HSDs Hormone Recognition

F

VTG (vitellogenin, yolk protein) mRNA expression; VEP, ZR, and ZP, protein synthesis/detection; gcl and PCNA gene expression Metabolism/Metabolic Disruption: cytochrome P450s, glucoronidases, sulfotransferases, insulin, glucose, vitamin D metabolite (125s) levels

Hormone Conversion

E

670

ENDOCRINE ACTIVE CHEMICALS: BROADENING THE SCOPE

helpful, in addition to testosterone levels, for androgen testing that may be associated with infertility. Although commercially available hormone assay kits are relatively reliable, it is important to note the timing of the menstrual/estrous cycle when the test was done to aid in the interpretation of the results. The use of several valuable biomarkers has been demonstrated in taxa other than mammals. These include VTG, a major egg-yolk protein precursor produced by female oviparous vertebrates and some invertebrates, and vitelline envelope proteins (VEPs), glycoproteins that are components of the egg envelope that form the chorion, or the outer envelope surrounding the embryo, in the developing egg (Celius and Walther, 1998). VTG and VEPs are used to detect estrogenic and antiestrogenic exposures in fish, amphibians, birds and invertebrates (Hutchinson et al., 2005; Nilsen et al., 2004; Panter et al., 2002; Arukwe et al., 2000; Thorpe et al., 2000; Jobling et al., 1998; Matthiessen, 1998; Arukwe et al., 1997; Sumpter and Jobling, 1995). Also, the production of the chorion proteins zona radiata (ZR) and zona pellucida (ZP) is regulated by estradiol and can be more sensitive than VTG at low doses of estrogen exposure. (For a review of VTG, ZR, and aromatase in fish and invertebrates, see Rotchell and Ostrander, 2003.) Another protein that can be upregulated in response to EAC exposure is the estrogen receptor itself (Bowman et al., 2002; Pakdel et al., 1991). Commonly used end points for thyroid status are the thyroid hormones, T4 and T3, and the pituitary hormone, TSH (see Table 18.1). The majority of T4 is produced in the thyroid gland, but 80% of T3 is produced external to the thyroid through deionization of T4 (Sapin and Schlienger, 2003). In serum, the vast majority of T4 and T3 is protein bound (Sapin and Schlienger, 2003). Measurements for TSH receptor antibodies and for thyroid peroxidase and thyroglobulin antibodies are also available (Squire, 2006). In humans, individual variance in serum T4 and TSH is substantially narrower than the variance for the population (Andersen et al., 2002, 2003). Thyroid hormones are narrowly regulated around a “set point” that is predominantly controlled by genetics (Hansen et al., 2004). To illustrate, serum T4 and TSH are highly correlated between monozygotic twins (Hansen et al., 2004); in fact, to a significantly greater extent than between dizygotic twins. Population variability is a summation of individual variability. National estimates of the distribution of the biomarker end points of thyroid function have been published for several countries including the United States (Hollowell et al., 2002) and China (Teng et al., 2006); and Reference Intervals have been established for areas of Germany (Volzke et al., 2005) and Norway (Bjoro et al., 2000). National differences in distributions can reflect many factors including nutritional status (e. g., iodine intake) and potential exposure to environmental chemicals, as well as differences in genetic composition of populations. 18.3.2

Individual and Population-Level End Points and Clinical Signs

Individual and population-level end points and clinical signs used to assess endocrine activity range from general indicators (e.g., behavior and organ weights) to specific biomarker end points (e.g. secondary sex characteristics) (Table 18.2). Certain organs are individually examined and weighed following exposure to suspected EACs to identify abnormal increases or decreases in dry or wet weight as well as relative organ-to-body weight ratios (see Table 18.2). Increased weight of ventral prostate and seminal vesicles has been used to assess androgen exposure in rats (Gray et al., 2004). Many end points can elicit population-level effects as a result of exposure to EACs (see Fig. 18.2 as a case study of how end points associated with thyroid dysfunction can be

671

F

F F

F

F

F

F

F

Letter Linked to Fig. 18.1

TABLE 18.2

Histopathology: thyroid, brain, pituitary, adrenal glands, gross lesions, reproductive organs; and tissues include: testes, epididymes, seminal vesicles and coagulating glands, prostate, levator ani muscle plus the bulbocavernosus and glans penis, vagina, cervix, uterus, ovaries Mitotic index of epithelial lining of endometrium Target Organ Weight: organ-to-body weight ratio (i.e., GSI,gonadal somatic index; his, hepatic somatic index); thyroid weight/volume; organ-to-brain weight; absolute weight of organ(s) including: liver, kidneys, adrenal glands, pituitary, thyroid/parathyroid, ovaries, uterus, testes, seminal vesicle with coagulating glands, prostate, epididymis, levator ani and bulbocavernosus, Cowper’s gland, glans penis, brain, heart, spleen, and thymus Immunity: cytokine production, proliferation, T-cell function, induction of immunoglobulins, and autoantibody production by B1 cells Fertilization Success: fertility/infertility/# fertilized eggs/conception rates, embryo/larval/fetal/neonatal survival, # still and live births, time to brood release, hatchability (days to hatch), appearance of larvae, and premature labor Fecundity: spawning activity/frequency (group and pair breeding), number eggs produced Egg Integrity: shell thickness, # eggs cracked (birds) Growth: measured by body weight, hind limb length, body length, wing and bone length (birds), 28d growth LOEC (lowest observed effect concentration) (fish), strength Development: morphological development examples include: hypospadias, anogenital distance (AGD), undescended testes (cryptorchidism), urethral–vaginal distance, premature puberty; time to first foam, time to first egg, size cloacal gland/protuberance (birds), and molt frequency (invertebrates)

Examples of Potential Endocrine End Points and Clinical Signs

Whole Animal- and Population-Level End Points and Clinical Signs

(continued)

(Gray et al., 2004; Halldin et al., 2003; Laws et al., 2000; Gray, 1998)

(Kamata et al., 2006) (Mukhi et al., 2005; Gould et al., 1997)

(Thorpe et al., 2006; US EPA, 2002)

(Silva et al., 2005; Yurino et al., 2004; Nielsen and Hultman, 2002; Hultman et al., 1992) (Lawn et al., 2005; Ottinger et al., 2005; US EPA, 2002)

(Eskiocak et al., 2005; Tajtakova et al., 2005; Zaki et al., 2004; Gray et al., 2004; Langer et al., 2003; Rotchell and Ostrander, 2003; Laws et al., 2000; Stoker et al., 2000)

(Eskiocak et al., 2005; Zaki et al., 2004; Gray et al., 2004; Weber et al., 2002; Zillioux et al., 2001; Laws et al., 2000; Stoker et al., 2000)

Reference(s)

672

H

H

F. H

F, H

F, H

F

Letter Linked to Fig. 18.1

TABLE 18.2 Examples of Potential Endocrine End Points and Clinical Signs

Altered Reproductive Viability of Offspring: embryo/larval/juvenile viability, time to sexual maturation, fertility of offspring at maturity, number of eggs produced by offspring, sex ratio of offspring

Males: coloration/banding, nuptial tubercles, dorsal nape pad; preputial separation (PPS), nipple retention Females: size of ovipositor, urogenital papilla, vaginal opening, precocious breast development Male and female: plumage length, plumage dimorphism (birds); hirsutism (abnormal hair growth) (mammals) Sexual/Reproductive Behavior: territorial aggressiveness/defense, nesting (spawning), courtship, egg protection/care, nest attentiveness (birds), mounting, libido, erectile dysfunction (mammals) Formation of Functional Gametes: sperm number, motility, and morphology; gamete maturation (production, final oocytes maturation, sperm motility test) Sex Ratios: intersex, imposex, skewed sex ratios

Estrous/Menstral Cyclicity: irregular menses, amenorrhea, premature menopause Secondary Sex Characteristics:

(Continued)

(Maack and Segner, 2003; Orn et al., 2003; Rotchell and Ostrander, 2003; Seki et al., 2002; van Aerle et al., 2002; Oberdorster, 2001; Parks et al., 2001; Gronen et al., 1996, 1999) (Touart, 2004; Rotchell and Ostrander, 2003; Oberdorster, 2001)

(Nice, 2005; Horiguchi et al., 2002; Swan et al., 1997; Carlsen et al., 1992)

(Denslow and Larkin, 2006; Halldin et al., 2005; Ottinger et al., 2005; Oliva et al., 2002; Whelan et al., 1996)

(Gray et al., 2004; Rotchell and Ostrander, 2003; Quinn et al., 2002; Ankley et al., 2001; Harries et al., 2000; Laws et al., 2000; Stoker et al., 2000; Gray, 1998; Bortone and Davis, 1994; Bortone et al., 1989; Howell et al., 1980)

(McLachlan et al., 2006; Laws et al., 2000)

Reference(s)

END POINTS AND CLINICAL SIGNS ASSOCIATED WITH ENDOCRINE ACTIVITY

673

Tissue indicators Clinical signs or toxicological effects (histopathology)

Biochemical indicators

Molecular indicators

Polychlorinated biphenyls

None

Maternal T3, T4 with Infant TSH Anti-TPO and abnormal TSH (males only) Minor changes in tT4, fT4 and TSH

Thyroid gland volume Frequency of echogenicity

Inconsistent association between PCB exposure and thyroid hormone homeostasis

TH responsive gene expression of RC3/ Neurogranin TH responsive gene expression of myelin basic protein Serum tT4, fT4 and plasma tT4 Serum tT3, fT3 and plasma tT3

Colloid area

Hypothyroidism Neurological deficits Hearing loss

None

Plasma T4 (chicken embryos) Plasma T3 (mallards, kestrels) EROD (mallards) PROD (mallards) Hepatic type I monodeiodinases (chicken embryos) Thyroid gland weight (mallards)

Growth (femur length in chicken embryos)

FIGURE 18.2 Generalized effects of PCBs on thyroid end points.

indicative of various forms of clinical disease). Some such measures are visible only at the population level and not at the individual level (e.g., thyroid disease in human population as a result of iodine deficiency (WHO, 1997)) (see Table 18.2). These effects can bring serious long-term consequences to an entire population, although the individual health consequences are less well documented. Information on distribution of end points for the population remains an important contribution that generally has not been made. 18.3.3

Histology End Points

A variety of tissues are regularly examined using histology to detect potential pathological effects following exposure to EACs (see Table 18.2). The occurrence of testicular and ovarian tissue in the same gonad, apoptosis of gonad tissues, increased apoptosis of spermatocytes, Sertoli and Leydig cell hypertrophy or hyperplasia, stage or success of spermatogenesis/maturation of oocytes, status of seminiferous tubules, structural integrity and sperm production, and the number of uterine implantation sites/scars are all indications of altered reproductive tissues that can result from exposure to EACs. The thyroid is also often examined histologically in fish, birds, and laboratory rodents. Follicular cell hypertrophy or hyperplasia, characteristics of colloid, vascular supply, density/size/shape of thyroid follicles, and angiogenesis are all indicators of altered thyroid status. 18.3.4

End Points for HPT Effects

Size or volume of the thyroid gland and pathologies are often examined using ultrasound imaging (Wiesner et al., 2006; Braverman et al., 2005; Tajtakova et al., 2005; Langer et al.,

674

ENDOCRINE ACTIVE CHEMICALS: BROADENING THE SCOPE

2003). Sonography is frequently used to determine thyroid volume in screening and is recommended (WHO, 1997) especially to assess iodine status of populations (Wiersinga et al., 2001; Brahmbhatt et al., 2000). Long-term iodine sufficiency influences thyroid volume creating difficulties in applying reference standards for screening with sonography developed on populations with marginal iodine sufficiency (WHO, 1997) to iodine-replete populations (Hess and Zimmermann, 2000; Foo et al., 1999; Xu et al., 1999). The same approaches for evaluating thyroid status are used in other species, especially dogs (Bromel et al., 2006; Reese et al., 2005; Peterson et al., 1997) and cats (Peterson, 2006; Daniel et al., 2002; Norsworthy et al., 2002; Mooney, 2001). Several examples are well described in literature documenting population-level effects of dietary components on endocrine status of domestic livestock and humans. The interrelationship between diet and contaminants is one that is often overlooked. Such examples may not readily come to the attention of persons accustomed to reading the environmental health literature. These are, nonetheless, important examples of population-level effects of a component of the environment (i.e., diet) on endocrine status. Although iodine deficiency is the most well-understood environmental cause of thyroid disease (Delange, 1994), worldwide additional food sources contain chemicals that produce clinically significant thyroid diseases including severe endemic goiter (Moreno-Reyes et al., 1993). Generally, the mechanisms of action appear to involve disruption of iodine utilization. Major examples of such antithyroid chemicals include hydrogen cyanide in cassava root (Teles, 2002), a major energy source for humans and domestic livestock that interferes with thyroid peroxidase activity; C-glycosylflavones from millet (Gaitan et al., 1989), a major energy source in semiarid tropical areas that inhibits thyroid peroxidases (Gaitan et al., 1989); and isoflavones from soy-containing foods and dietary supplements (Doerge and Chang, 2002; Doerge and Sheehan, 2002b). Although the mechanism of action of soy-containing foods in adversely affecting the thyroid is not fully understood, it may dispose individuals with iodine deficiency to hypothyroidism (Doerge and Sheehan, 2002b). Many additional plants contain chemicals having antithyroid activity (among others, see Chandra et al., 2004). Perchlorate, a chemical known to block the sodium iodide symporter (NIS) responsible for iodine uptake into the thyroid gland, has been found to be widespread in human milk that can serve as a biomarker for exposure to perchlorate (Kirk et al., 2005). Genetic defects in the iodination process of thyroid synthesis combined with environmental exposures and suboptimal iodine status have the potential to be clinically significant (Scinicariello et al., 2005). 18.3.5

End Points for HPG Effects

In species with prominent secondary sex characteristics (e.g., many fish species), skewed sex ratios can be easily and quickly identified at sexual maturation; however, identification of intersex gonads, such as the presence of testicular oocytes, requires histopathological analysis (Hutchinson et al., 2005; Ankley et al., 2003; Ankley et al., 2001; Harries et al., 2000). The early onset of puberty in humans can be considered a secondary sex characteristic end point of exposure. Observational studies in humans have identified end points as potential biomarkers of early developmental exposure and susceptibility to EACs (Ouyang et al., 2005; Colon et al., 2000). There is accumulating evidence that the early onset of puberty results in longer life exposure to estrogens and, thereby, may increase the risk of breast cancer (Lippman et al., 2001). Identifying such end points is critical in research with EACs in order to provide linkage to population-level relevant effects. Longitudinal studies

ENVIRONMENTAL CHEMICALS AND END POINTS: CASE EXAMPLES

675

from early childhood to adulthood are required to identify clear biomarker end points of effect regarding the risk associated with developmental estrogen exposure, breast cancer in women, and, perhaps, other developmental malformations and cancers in both women and men. 18.3.6

End Point and Clinical Signs Summary

As discussed above, several end points and clinical signs exist, ranging from behavior and molecular indices in individuals to population-level markers that can indicate effects from exposure to EACs. Some of the end points and clinical signs discussed above and mentioned in Tables 18.1 and 18.2 are of high ecological importance (i.e., fecundity, fertility, altered sexual behavior, secondary sex characteristics, and sex ratio). However, many of these end points and signs can be affected by changes in the environment (e.g., temperature, food availability, age, and reproductive status), including, but not limited to, exposure to contaminants, and hence do not offer much information on the potential MOA of the chemicals of interest (Denslow and Larkin, 2006). If these end points are utilized in concert with mechanistically relevant assays, they can elucidate more information about MOA or pathway of such effects, while still making the link to the population level. Molecular and cellular in vitro assays often offer more sensitive end points and may, therefore, be able to provide early warning of adverse effects in individuals and at much lower chemical concentrations than in vivo approaches (Denslow and Larkin, 2006). The usefulness of molecular markers greatly enhences if they can be used to make the link to adverse effects on the individual and/or population. Recently, efforts have been made using “systems toxicology” (Waters and Fostel, 2004) to describe changes in gene expression to toxicants from a whole animal-level approach. Also, “phenotypic anchoring” (Tennant, 2002) stresses the correlation of changes in gene expression with traditional end points that lead to adverse effects. These approaches are notable, but still more research is needed to create linkages between molecular approaches and adverse individual and population effects. Clinical signs that are currently related to endocrine effects in humans as a result of exposure to potential EACs are useful, but continued research is needed to develop more sensitive end points that can serve as relevant biomarkers to link the exposure of potential EACs to endocrine disruption. In this regard, value and insight have been and will continue to be gained in studying various taxa in the laboratory and ecoenvironment. The hope is that the development of better technological tools will result in more specific end points that can be used as biomarkers to help in diagnosis and treatment and, ultimately, in the prevention of overexposure to potential EACs that may result in adverse effects on human and animal individuals and populations.

18.4 ENVIRONMENTAL CHEMICALS AND END POINTS: CASE EXAMPLES Case 1. Thyroid Hormone and Three Environmental Chemicals with Different Mechanisms of Action The endocrine actions of thyroid hormone can be disrupted at multiple levels such as synthesis, iodination, release of stored hormone, transport by plasma proteins, peripheral activation, target organ metabolism (e.g., brain, liver), deiodination, and excretion of

676

ENDOCRINE ACTIVE CHEMICALS: BROADENING THE SCOPE

breakdown products. Recently, the role of the deiodinases, in local control of thyroid hormone at the tissue level, and implications for altering whole animal physiological status have been more thoroughly described (reviewed by Bianco and Kim, 2006). Environmental chemicals can affect multiple and different steps in this process. A description of three chemicals that affect thyroid hormone synthesis, metabolism, and excretion is provided in Fig. 18.2a–c. The chemicals, lithium, perchlorate, and polychlorinated biphenyls (PCBs), chosen have distinctly divergent MOAs on thyroid hormone as well as different environmental characteristics. Lithium. Lithium, the so-called ‘hard’ ion that binds to oxygen-donors better than do the hard metal ions, which are utilized biologically (Na, K, Mg, and Ca) is in the ‘a’ class of hard metals (Frau´sto da Silva and Williams, 1993). Lithium is perhaps best known for its pharmaceutical use, especially to treat psychiatric conditions. Data from the National Health and Nutrition Examination Survey (1999 through 2001) have provided population estimates for the United States that indicate pharmaceuticals containing lithium were taken by approximately 0.1% of the population, representing around 220 000 people in the United States (Aoki et al., 2006). Lithium is found at low concentrations in the major rivers in the United States (Kszos Lakowicz and Stewart, 2003). Although best known for pharmacological properties, lithium has a number of industrial uses (see Moore, 1995 for summary) including as a component of batteries; research is going on to produce nanoparticles for improved lithium ion battery cathodes (Liu et al., 2006). It is also used as a catalyst in chemical reactors and as a cell additive in aluminum production. Developmental and reproductive toxicity of lithium to mammalian species is well established. Moore and the Institute for Evaluating Health Risks Expert Scientific Committee (IEHRSC) assessed the reproductive and developmental toxicology literature for lithium utilizing an expert evaluative process concluding that the data were sufficient to indicate that lithium at therapeutic doses can cause developmental toxicity in humans and that lithium causes developmental toxicity in both rats and mice based on data from studies using both prenatally and postnatally dosed animals (based on studies that included the period of lactation to weaning and beyond) (Moore, 1995). The teratogenic effects of lithium on nonmammalian species are so pronounced that lithium salts are used as a laboratory “standard” to create teratogenicity in frog species. As evaluated by Moore and the IEHRSC, some of the reproductive studies in rodents observed changes with lithium dosing near clinical therapeutic concentrations, although these studies were not used for quantitative risk assessment. This work did not include avian, amphibian, or fish species and noted that no data specifically on mammalian male reproductive toxicity in the literature were found (Moore, 1995). In a more recent study in nonrodent species (Banerji et al., 2001), specifically birds (i.e., finches), lithium treatment decreased testicular weight and seminiferous tubule diameter and induced severe degenerative changes in germ cells at serum lithium concentrations in the therapeutic range of plasma lithium for humans (0.5 to 1.5 mEq/L or 3.5 mg/L to 10.5 mg/L). Although comparison of responses to lithium across species, as illustrated in Fig. 18.2a, indicates several similarities with respect to thyroid, the reproductive tracts of male birds appear to be especially sensitive to the effects of lithium exposure. Clearly, the sensitivity of different tissue types to the effects of lithium differs across species. Whether concentrations of lithium producing teratogenic or reproductive effects are obtained in nature is a wider topic than can be addressed in this brief review. Background concentrations of lithium in soil average 30 mg/kg to which lithium from

ENVIRONMENTAL CHEMICALS AND END POINTS: CASE EXAMPLES

677

sources such as volcanic activity, fossil fuels, and disposal of lithium-based batteries are added (Rydh and Svard, 2003). The effects identified in the nonmammalian species are severe. Exposures that produce subtle, adverse effects appear to be unknown. Moore and the IEHRSC recommended that studies be undertaken on developmental toxicology that included long-term assessment of kidney, heart, thyroid, and central nervous system (CNS) structure and function (Moore, 1995). Data are available to observe some of the effects of lithium on thyroid in multiple species, including nonrodent species. Lithium also produces a variety of adverse effects on the that are summarized in Fig. 18.2a for humans (Bocchetta and Loviselli, 2006; Lazarus, 1998; Loviselli et al., 1997; Bocchetta et al., 1996; Mizukami et al., 1995; Perrild et al., 1990; Bagchi et al., 1978; Rosser, 1976); rats (Allagui et al., 2006; Frankenfeld et al., 2002; Chanoine et al., 1993; Chatterjee et al., 1990; Etling et al., 1987; Bagchi et al., 1978; Berens et al., 1970); and birds (Downie et al., 1977). Perchlorates. Perchlorates are widely distributed environmental contaminants, not bioaccumulative, and are also used as drugs. Perchlorates form naturally in the atmosphere and are present in precipitation (Dasgupta et al., 2005), concentrate in particular regions (Dasgupta et al., 2005), and can be greatly increased through use of various products including explosives, pyrotechnics, and solid rocket propellant (Miediratta et al., 1996). Blount and colleagues reported that among women who had been participants during the years 2001 and 2002 in the National Health and Nutrition Examination Survey in the United States whose urinary iodine was less than 100 mg/L (but not those whose urinary iodine was more than 100 mg/L), increased perchlorate was associated with lower production of T4 and an increase in TSH to stimulate additional T4 production (Blount et al., 2006). This is the first time that effects of environmental levels of perchlorate have been associated with thyroid hormone decrements in a general population study at ambient contamination levels. The general trends observed in the perchlorate literature for humans, rodents, amphibians, and birds are represented in Fig. 18.2b. These studies demonstrate generalized trends across taxonomic groups for the thyroid hormones and thyroid gland effects. In human females and all rodents, serum T4 declines and TSH increases following exposure to perchlorate, while in birds T4 and T3 were shown to decline and in frogs only T3 was demonstrated to decline. Similarly, thyroid gland weight or volume increased in human, rodent and bird examples. Many of the effects listed are taxa specific and represent a limited number of studies. Therefore, it is important to note that the effects outlined in Fig. 18.2 demonstrate generalized trends seen in the literature. These figures are intended to demonstrate the importance of observing effects across species, demonstrating the value of a weight-of-evidence approach for use in risk assessment. The studies utilized in Fig. 18.2 represent general trends observed and do not demonstrate the effects of each study in the literature due to differences in experimental design such as exposure times, exposure duration, chemical administration, and sex and age of the animal studied. Studies summarized in Fig. 18.2b include those for humans (Blount et al., 2006; Kirk et al., 2005; Greer et al., 2002), rodents (OECD, 2006), birds (McNabb et al., 2004a, 2004b), and frogs (Zhang et al., 2006; Tietge et al., 2005; Fort et al., 2000). Polychlorinated Biphenyls. The PCBs make up a mixture of different biphenyl compounds that can be chlorinated in different ways to include 209 separate chemical structures or congeners. The PCBs were designed for use in transformers, as lubricants in electrical equipment, as plasticizers, in surface coatings and inks, in carbonless duplicating paper, and as an extender in pesticide mixtures (http://www.epa.gov/ttn/atw/hlthef/polychlo.html).

678

ENDOCRINE ACTIVE CHEMICALS: BROADENING THE SCOPE

The PCBs are highly lipid soluble, bioaccumulative, and persistent and continue to exist in the environment decades after their use was discontinued in the 1970s. Commercial production of the PCBs was banned in 1979 due to their ability to accumulate in the environment (Ross, 2004; Carpenter, 1998). These congeners have been detected in many environmental media including air, water, aquatic and marine sediments, fish, wildlife, and human adipose tissue, serum, and milk (Safe, 1993). It is thought that exposures were highest in the late 1970s and early 1980s, declining after manufacturing ceased and regulatory controls were enacted (Ross, 2004). As a chemical class, the PCBs have multiple toxic effects and are well studied as antithyroidal agents (Carpenter, 1998). Various PCB congeners are known to interact directly with the thyroid gland and alter the ability of T4 to bind to the serum binding protein, transthyretin (TTR), and they have also been shown to alter the expression of liver glucuronidase, and thus the normal metabolism of the thyroid hormones (OECD, 2006). Exposure of rats to PCBs can produce thyroid changes. The National Toxicology Program bioassay of PCB 153 across a wide range of doses showed mild follicular cell hypertrophy at the highest dose. PCB 153 caused nonneoplastic lesions on the thyroid gland in female rats (NTP, 2006). Similar results are reported for PCB 126, and data support a MOA for PCB 126 involving induction of hepatic uridine diphosphate glucuronyl transferases activity (Fisher et al., 2006). In other nonrodent species, increasing PCB concentrations in blubber of harbor seals is inversely associated with serum total T4, free T4, total T3, and free T3, as well as seal thyroid hormone receptor (TR) activity (Tabuchi et al., 2006). PCBs are known to affect brain development through their effects on thyroid (reviewed by Koibuchi and Iwasaki, 2006; Roegge and Schantz, 2006; Zoeller et al., 2000), although the mechanism is not well characterized. In humans, PCBs reduce circulating T4 concentrations that could selectively damage the cerebellum during in utero development (potential mechanisms reviewed by Roegge and Schantz, 2006). Exposure to PCBs has been associated with lower levels of total T4 and free T4 in women (Persky et al., 2001) and a significant negative association between T3, T4, and TSH with PCBs has been reported for adult men (Turyk et al., 2006). These associations are, however, not consistent across populations (Hagmar et al., 2001a; Hagmar et al., 2001b; Osius et al., 1999). The generalized effects of PCBs on the thyroid systems of humans, rodents, and birds are summarized in Table 18.2c. Figure 18.2c shows a general trend across species of a decline in thyroid hormone levels following exposure to PCBs. Specific effects are noted for each taxonomic group based on literature reviews for humans, rodents, and birds. The effects represented in Fig. 18.2c show general trends summarized in the literature, but by no means represent either every study and exposure scenario or every congener: humans (Langer et al., 1998, 2005; Hagmar, 2003; Longnecker et al., 2000; Koopman-Esseboom et al., 1994); rodents (OECD, 2006; Zoeller et al., 2000; Brouwer et al., 1998; Goldey et al., 1995); and birds (Smits et al., 2002; Gould et al., 1999; Fowles et al., 1997). Observing the three examples presented in Fig. 18.2a–c, it is clear that these three, highly evaluated chemicals can disrupt thyroid hormone function at multiple and diverse steps in the synthesis and metabolic processes. Likewise, information obtained by using different taxa is complimentary, providing useful information in that responses not identified in one species may be apparent in another. Relying on in vitro assays or in vivo assays, in isolation, using single species, can produce only a partial picture of the endocrine activity of a chemical. Likewise, it is essential to note species differences (e.g., see Fig. 18.2a–c), so that these differences can be interpreted to most accurately assess the effects of a chemical on the environment.

ENVIRONMENTAL CHEMICALS AND END POINTS: CASE EXAMPLES

679

Case 2. Lead and the Vitamin D/Endocrine System Much interest, research, and programmatic activity have focused on EACs that disturb the estrogenic, androgenic, and thyroid systems. This reflects the origins of the work in reproductive toxicology and observations among wildlife species (among others see EPA, 1998; Reiter et al., 1998; Cooper and Kavlock, 1997; Colborn et al., 1993). Major, critically important areas of endocrine function have largely been omitted from consideration, including, for example, the adrenals (Harvey and Everett, 2003), the pancreas, and other systems such as the vitamin D/endocrine system. Adverse health outcomes following exposures to inorganic lead have been very well documented as an environmental health problem over the past 40 years. Risk assessments for inorganic lead have addressed the following organ systems: hematopoiesis, the kidney, and the nervous system (see chapter by Mahaffey et al., 2000). Endocrine effects of inorganic lead exposures have never been routinely explored, so have never been a central part of risk assessments. The example that follows describes how inorganic lead affects the vitamin D/ endocrine system and refers to the steps described in Fig. 18.1. Vitamin D is a prohormone produced in the skin through ultraviolet irradiation of 7-dihydrocholesterol (DeLuca, 2004), as well as being a provisionally required nutrient. More than 90% of the vitamin D requirement for most people is met by exposure to sunlight rather than diet (Holick, 2004). The physiological role of vitamin D in regulating calcium and phosphorus metabolism received the earliest attention following observations that rickets was associated with lack of exposure to sunlight (reviewed by Holick, 2004). With advancing experimental methods, elucidation of the cellular biology of the vitamin D receptor and genetic control of the vitamin D–endocrine pathways has advanced and been the topic of numerous reviews (among others, see Kiraly et al., 2006; Dusso et al., 2005; DeLuca, 2004; Holick, 2004; Prosser and Jones, 2004; Fleet, 2004). The biochemical and physiological changes caused by the vitamin D/endocrine system have been reviewed for multiple organ systems asuch as the central nervous system (Kiraly et al., 2006), the kidney (Dusso et al., 2005), and the immune system (DeLuca, 2004; Holick, 2004), as well as the pathogenesis of diabetes (Reis et al., 2005; Mathieu and Badenhoop, 2005; Ogunkolade et al., 2002). Hormone Production. One of the toxic effects of inorganic lead is interference with the production of 1, 25-dihydroxyvitamin D, which is, in turn, known to affect the storage and release of lead (Onalaja and Claudio, 2000)(see Fig. 18.1, Box C). Hormonal vitamin D (i.e., 1,25-dihydroxyvitamin D) interacts with nuclear vitamin D receptor (VDR), an ancient member of nuclear receptors for steroid hormones (Dusso et al., 2005), regulating the production of calcium-binding proteins (Onalaja and Claudio, 2000). Decreased production of the metabolically active form (1,25-dihydroxy-vitamin D) due to lead exposure affects the amount of hormonal vitamin D available to bind and activate the VDR, and thus ultimately affecting production of calcium-binding proteins. Hormone Conversion. Preformed vitamin D2 and vitamin D3 are obtained directly from the diet. 7-dehydrocholesterol (present in skin) can be converted into vitamin D3 by UV sunlight (Holick et al., 1977). These preformed chemicals are metabolically converted into hormonally active chemicals in the skin, known as prohormones (see Fig. 18.1). These prohormones are then metabolized in the liver into the steroid hormone, 25-hydroxyvitamin D3. The hepatic cytochrome P-450 CYP2R1 appears to be the critical 25-hydroxylase involved in vitamin D metabolism (reviewed by Dusso et al., 2005). The second step of vitamin D activation is conversion of 25-hydroxyvitamin D to 1,25-dihydroxyvitamin D,

680

ENDOCRINE ACTIVE CHEMICALS: BROADENING THE SCOPE

occurring under physiological conditions primarily in the kidney (Fraser and Kodicek, 1970). Fig. 18.1 describes these steps as hormone conversion (Fig. 18.1, Box E). Inorganic lead interferes with the production of the metabolically active form of vitamin D. Reduced plasma 1,25-dihydroxyvitamin D occurred in lead-fed rats (Smith et al., 1981). Lead-fed chicks had reduced conversion of 25-hydroxyvitamin D to 1,25-dihydroxyvitamin D (Edelstein et al., 1984). In epidemiological studies children with blood lead levels varying from 12 to 120 mg Pb/dL, serum concentrations of the vitamin D hormone, 1,25-dihydroxyvitamin D, were strongly correlated (r ¼ 0.88) and reduced to levels found in patients with metabolic bone disease (Mahaffey et al., 1982). In lead-poisoned children, 1,25dihydroxyvitamin D concentrations returned to normal shortly after chelation therapy, whereas, serum 25-hydroxy vitamin D levels did not change with treatment (Rosen et al., 1980). Hormone Recognition, Internalization, and Action. All biological responses of vitamin D arise following its metabolism into secosteroids with two principal metabolically active forms: 1, 25-a-dihydroxyvitamin D3 and 24R, 25-dihydroxyvitamin D3. With regard to Fig. 18.1, these processes would be included as part of metabolism and elimination (Fig. 18.1, Box G). 1,25 –a-dihydroxyvitamin D3 is the dominant metabolite producing a wide array of biological responses by interacting with the vitamin D nuclear receptor that regulates gene transcription in over 30 target organs with a cell membrane receptor that mediates rapid biological responses (Norman et al., 2002). Vitamin D metabolites are best known as regulators of ionized calcium homeostasis through actions on the intestines, the kidneys and bone, collectively called the vitamin D endocrine system (MacDonald et al., 1994). In the schematic shown in Fig. 18.1, these steps would be in the area of hormone recognition, internalisation, and action (Fig. 18.1, Boxes F and G). In addition to affecting calcium metabolism, vitamin D exerts a number of nonclassical effects (as reviewed by Dusso et al., 2005): suppression of cell growth, regulation of apoptosis, modulation of immune response, control of keratinocyte differentiation and function in the skin, and control of other organ systems including the renin– angiotensin system, insulin secretion, muscle function, and neurological function. Whether or not there are effects of lead on metabolism and elimination of the metabolically active form of vitamin D, as shown in Fig. 18.1 (Box G), has not been entirely clarified; however, because exposure to inorganic lead can interfere with conversion of vitamin D to its active metabolite, it can be surmised that this effect could ultimately impact the function of the organ systems reliant upon active form of vitamin D. Investigations into this form of endocrine disruption are warranted and should potentially be considered for risk assessment. Latent and Population-Level Effects. The last step identified in Fig. 18.1 addresses latent and population-level effects of the interaction of the environmental chemical and the endocrine system (Fig. 18.1, Box H). Broadly speaking interactions between environmental agents, diet, genes, and disease are areas of great interest in advancing disease prevention (Willett, 2002). Combinations of such factors are known to increase risk of disease. Interactions between lead and vitamin D, as previously described, have been investigated experimentally over several decades. In recent years, it has become clear that there are genetic differences in the population distribution of various forms of the receptors and in the biology of the various vitamin D receptor polymorphisms (Fleet, 2004; Uitterlinden et al., 2004). These, as well as the nature of human microsomal vitamin D3 -25-hydroxylase remain areas of active investigation. Vitamin D receptors are anticipated to affect lead storage and/or release (Onalaja and Claudio, 2000). The influence of VDR gene polymorphisms on VDR protein function and

DEVELOPMENTAL ORIGINS OF HEALTH AND DISEASE

681

signaling is largely unknown (Uitterlinden et al., 2004). These differences could help assess the population-level impact (see Fig. 18.1, Box H) of interactions between exposure to inorganic lead and vitamin D effects and emphasize that there may be sensitive populations. To illustrate population-level effects, the role of genetic differences has been studied. Three vitamin D receptor gene polymorphisms (BB, tt, and AA genotypes) were associated with low mineral density (Fountas et al., 1999). One of the best-characterized restriction fragment length polymorphisms, Bsml (Uitterlinden et al., 2004) has also been studied with lead (Onalaja and Claudio, 2000). Schwartz and colleagues found that among former organolead workers, bone lead was higher in workers with the BB genotype, intermediate in the heterozygous, and lower in the bb homozygous workers (Schwartz et al., 2000). However, results across studies have been variable. Recent investigations of modifications by polymorphisms in genes encoding the VDR and the association between lead exposure and biomarkers of lead exposure and effect have either found no association (Weaver et al., 2003) or found that a particular polymorphism of the VDR gene (VDR B allele rather than the VDR bb genotype) was associated with high lead burden (Theppeang et al., 2004) and that renal outcomes were more adverse in younger lead workers with the variant VDR B allele (Weaver et al., 2006). Among children, the VDR gene polymorphism VDR-Fok1 has been found to modify the association between exposure to inorganic lead and blood lead concentrations based on observations in young children 6–24 months of age (Haynes et al., 2003). Children with the FF genotype had a greater increase in blood lead concentration than did children with the Ff genotype. Based on a different VDR marker (VDRBB and VDRBb), Shi and colleagues reported that 5–6-year-old children in a Chinese kindergarten who had the VDR B allele reported significantly higher blood lead levels than those with the VDR bb genotype (Shi et al., 2003). Overall, at the latent or population level (see Fig. 18.1, Box H) the associations between vitamin D receptor gene polymorphisms and bone disease are complex. A recent metaanalysis of data from more than 26,000 participants involving 9 European research teams concluded that the FokI, BsmI, ApaI, and TapI vitamin D receptor polymorphisms are not associated with metabolic bone disease or fractures, but the Cdx2 polymorphism may be associated with the risk of vertebral fracture (Uitterlinden et al., 2006). Similarly, MacDonald and colleagues found no significant association between common polymorphisms for the VDR and bone mass, bone loss, or fracture (Macdonald et al., 2006). In view of the limited impact of lead on bone mineral through vitamin D receptor, identifying an effect of lead on bone density or bone mass would not be expected to raise substantial questions about the population-level impact of this association.

18.5 DEVELOPMENTAL ORIGINS OF HEALTH AND DISEASE After describing some of the end points that are indicative of endocrine activity and providing an example of how a widely distributed environmental contaminant, inorganic lead, affects a relatively less well-known endocrine system, we now turn to two important areas that are of comparatively recent interest. A research area that has become a major focus is the study of how EACs may act through the concept of Developmental Origins of Health and Disease (DOHaD). This is the most recent terminology used to describe how the onset of adult diseases may be induced in utero/in ovo during embryo/fetal development, remain latent during neonatal and pubertal development, and emerge as an acute or chronic disease

682

ENDOCRINE ACTIVE CHEMICALS: BROADENING THE SCOPE

in adulthood. In the late 1980s, the “Barker hypothesis” was introduced, which was primarily based on human epidemiological data relating low birth size and poor maternal nutrition during fetal development to an increased risk of noncommunicable diseases (e.g., coronary heart disease, type-2 diabetes, osteoporosis, and metabolic and endocrine dysfunction) in the adult offspring (Barker, 2003). This led to the development of the paradigm, “Fetal Origins of Adult Disease” or “Fetal Basis of Adult Disease” (FeBAD) (Lau and Rogers, 2004). The paradigm was initially criticized, but has since gained support. Now FeBAD no longer remains restricted to just maternal nutritional effects on the fetus but also applies to chemical exposure and to other life stages, pre- and postnatally. For example, in animal studies, the effects of maternal exposure to environmental toxicants during pregnancy have led to permanent morphological and physiological changes in the developing embryo and neonate that apparently predispose the F1 offspring to diseases later in adult life (Lau and Rogers, 2004), some of which are similar to those observed in adult humans as discussed in the next section on transgenerational effects, which do not necessarily involve in utero exposure. The role of EACs in DOHaD is difficult to decipher in humans due, in part, not only to moral and ethical constraints but also to the latency of effects and complexity and scope that the endocrine system has alone and in concert with other major physiological systems (e.g., neurological and immunological). Nonetheless, a classic example of environmental endocrine exposure in utero and the dire consequences in the F1 offspring thereafter is the effect of diethylstilbestrol (DES). A historical account of the development and use of DES as an alternative therapeutic source of estrogen for women during pregnancy has been reviewed (Newbold and Jefferson, 2006). Briefly, a clinical report in 1971 indicated that a rare form of reproductive tract cancer, vaginal adenocarcinoma, was detected in adolescent daughters of women who had taken the drug while pregnant. Although the US Food and Drug Administration immediately restricted the use of DES during pregnancy, it has since been observed that daughters of mothers who had taken the drug while pregnant have had various other abnormalities, including reproductive organ dysfunction, abnormal pregnancies, decreased fertility, immune system disorders, and behavioral problems. In addition, DES sons were found to have structural and functional abnormalities, including hypospadias, microphallus, retained testes, inflammation, and decreased fertility. Notably, the deleterious effects of DES have been related to the time of exposure in utero (i.e., more adverse outcomes are associated with exposure earlier during gestation than later). Many of these developmental abnormalities observed in humans have been observed in mice following prenatal exposure to experimental doses of DES (Newbold et al., 2006). Rodent studies clearly suggest that DES exposure to pregnant dams increases susceptibility of the F1 offspring to malignant tumors of the reproductive tract (for review, see Newbold and Jefferson, 2006). In addition, numerous studies in pregnant rodents have indicated an increased incidence of mammary tumors (Fenton, 2006; Walker, 1992; Rothschild et al., 1987) and increased prostate-specific androgen receptor binding and organ weight when respective F1 offspring of DES-exposed dams reached adulthood (Gupta, 2000; vom Saal et al., 1997). It is equivocal, however, whether breast tumor development in daughters and prostate tumor development in sons of DES-exposed mothers are comparable to that observed in laboratory animals. Apart from DES, no other estrogenic chemicals have been documented to be directly linked to adult diseases in humans. However, there are numerous studies with pregnant animals that have shown in utero exposure of the embryo/fetus to synthetic and natural environmental compounds with estrogen-like activity (e.g., bisphenol A, atrazine, dioxin, and genistein) accelerates mammary gland development (Birnbaum and Fenton, 2003) and

DEVELOPMENTAL ORIGINS OF HEALTH AND DISEASE

683

prostate size in the F1 offspring. In this regard, accentuated growth may lead to an increased susceptibility of the adult animal to chemicals with estrogenic activity and, perhaps, a predisposition to breast and prostate cancers. For example, in female rats, prenatal or perinatal exposure through mothers (or externally during the perinatal period) to bisphenol A, a chemical compound used in epoxy resins and polycarbonate plastics, stimulated early mammary gland development in the F1 offspring that was associated with an increased risk of mammary gland tumor development as the animals reached adulthood (Munoz-de-Toro et al., 2005; Markey et al., 2001; Colerangle and Roy, 1997). In male rats, exposure of the pregnant mother to bisphenol Awas associated with an increased sensitivity to estrogen and increased prostate weight in the male F1 offspring with age. Dioxin, an industrial waste byproduct, and genistein, a phytoestrogen found in soy-based products, increased the susceptibility of exposed rodent F1 offspring to carcinogens in adulthood (Birnbaum and Fenton, 2003). Perinatal exposure to atrazine, a herbicide, has been associated with increased prostate gland inflammation in adult rats (Stoker et al., 1999). Atrazine and dioxin exposure in utero delayed mammary gland development in early postnatal life, as early as postnatal day (PND) 4 in rat F1 offspring. Both compounds alter development of the neonatal mammary fat pad. Thus, prolonging maturation of the mammary gland during adolescence lengthens the period in which mammary tissue is vulnerable to potential carcinogens that may result in mammary gland tumors in adults (Birnbaum and Fenton, 2003). More recently, exposure of pregnant rats to experimental doses of vinclozolin or methoxychlor during the period of embryo gonadal differentiation in utero resulted in decreased spermatogenic capacity and increased male infertility in F1 and subsequent generations of adults (Anway et al., 2006). The concept of DOHaD has also been documented in nonmammalian species. For example, latent effects of EACs have been repeatedly demonstrated in Japanese quail (Coturnix japonica). A single in ovo injection of a relatively low dose of DES before sexual differentiation altered reproductive behavior of F1 adult males (Vigiletti-Panzica et al., 2005; Halldin et al., 2005). Elements of copulatory behavior necessary for successful mating that include neck grab, mount attempts, successful mounts, and cloacal contact movements were significantly altered; however, gonadosomatic index, testes weight asymmetry, serum testosterone concentration, and cloacal gland area were not significantly affected (Halldin et al., 2005). In a comparable study with a higher dose of DES administered in ovo prior to sexual differentiation (Vigiletti-Panzica et al., 2005), male copulatory behavior of the F1 adult was again altered and was also associated with a significant decrease in density of vasotocin immunoreactivity in the medial preoptic nucleus, bed nucleus of the stria terminalis and lateral septum of the central nervous system (CNS). The results of these studies demonstrate that direct in ovo exposure to an estrogenic compound can exert permanent effects on sexually dimorphic regions of the CNS that are not expressed until the individual reaches adulthood. Despite a wealth of observational or epidemiological data on humans and empirical information on various mammalian and nonmammalian animal models, a direct link between in utero/ovo exposure to EACs and potential adverse effects in the F1 generation of adults has not been established, apart from the effects of DES. More research is necessary to understand the mechanisms (e.g., genomic and epigenomic) and effects of exposure at various developmental life stages (e.g., embryo, fetal, and neonatal) as well as the influence of various lifestyles (e.g., nutrition, physical activity, occupation, and addictions) between the time of exposure during pregnancy and the time at which disease, initially induced in utero/ovo, is expressed in the F1 generation of adults.

684

ENDOCRINE ACTIVE CHEMICALS: BROADENING THE SCOPE

18.6 TRANSGENERATIONAL EFFECTS To bring this chapter to a close, we now turn to another research area of substantial interest in understanding the mechanisms of action of EACs, specifically transgenerational effects. These effects are defined differently within various groups among the scientific community. Some scientists define a transgenerational effect as one that is simply transferred from the parent generation to the next or first (F1) generation of offspring (Campbell and Perkins, 1988), whereas others define transgenerational effects as having been elicited minimally through the third (F3) generation (Anway and Skinner, 2006), which would likely be due to a genetic (DNA) or an epigenetic (e.g., DNA methylation) alteration. According to the first definition, trangenerational effects may be applicable to the concept of DOHaD via maternal transfer. For the purposes of this review, we do not differentiate between these two definitions. As was described in the previous section, exposure of pregnant mothers to DES is capable of causing latent developmental effects on the F1 generation following in utero or in ovo exposure. These latent effects of DES, however, may be distinct from the concept of DOHaD since F2 and, perhaps, F3 generations of sons from grandmothers initially exposed to DES have been observed with an increased risk of hypospadias (Klip et al., 2002). This study should be viewed with caution, however, as only a small number of affected grandsons were observed (Hernandez-Diaz, 2002). Nonetheless, researchers continue to study subsequent generations of sons and daughters whose grandparents were initially exposed to DES to determine if a transgenerational MOA of DES occurs in humans. Rodent studies have also reported DES alterations to germ cells that were then passed on to later generations of offspring who were never directly exposed to DES (for review see Newbold and Jefferson, 2006). Distinct from the concept of DOHaD, trangenerational effects of EACs have also been observed in nonmammalian species, specifically in fish exposed to DES prior to spawning. In these studies, the F1 generation was not exposed directly to DES, but they experienced effects due to their parent’s exposure before spawning. In one of these studies, a fish lifecycle test conducted on the Chinese rare minnow (Gobiocypris rarus) observed transgenerational effects of DES on the F1 generation, including decline in the survival of F1 fry, decrease in testosterone in male progeny, and a slight elevation of estrogen levels in female progeny (Zhong et al., 2005). Other estrogenic chemicals have been demonstrated to have similar transgenerational effects on fish. In another study, Schwaiger and colleagues showed that the surfactant nonylphenol, an alkylphenol with weak estrogenic activity, leads to hormonal imbalances in the offspring of adult rainbow trout (Oncorhynchus mykiss) exposed to DES before spawning (Schwaiger et al., 2002). F1 generation offspring in this study had increased plasma levels of vitellogenin, estradiol, and testosterone in females and increased plasma estradiol in males. The fish data demonstrate that transgenerational effects occur in multiple taxa, and this phenomenon is thus not exclusive to any one species. The mechanisms required to induce transgenerational effects have not been thoroughly elucidated, and this is currently an exciting area of research. One mechanism described that could lead to transferable effects across multiple generations is epigenetic inheritance, also referred to as imprinting. Although epigenetic inheritance has not been investigated in most of the transgenerational studies in literature, several groups have begun to demonstrate epigenetic changes, specifically in methylation patterns in germ cell lines, which could lead to changes in exposed individuals that would be passed to their offspring (Anway et al., 2005, 2006; Anway and Skinner, 2006). Some of the chemicals studied

TRANSGENERATIONAL EFFECTS

685

include DES, vinclozolin, methoxychlor, and Cr (III) chloride. These studies are described in more detail below. Epigenetic inheritance is the transmission of information from an individual to its progeny without modification of the DNA sequences of genes, but rather DNA characteristics are changed, such as the DNA methylation patterns (Anway and Skinner, 2006). DNA methylation is the most studied form of epigenetics, and this methylation helps up- or downregulate several molecular events including gene transcription, X-chromosome inactivation, chromosome positioning, and repression of parasitic DNAs and imprinted genes. Imprinting, which transfers parent-specific information to progeny, involves DNA methylation of the carbon 5 position of a cytosine residue positioned next to a guanine residue, called CpG dinucleotides or islands, and is considered a major form of epigenetic modification. Imprinted genes have been hypothesized as one method for the developing animal to respond to environmental pressures (Swales and Spears, 2005). When this is considered in the context of toxicant exposures, it is proposed that chemicals can alter DNA methylation, and the consequences of this methylation are just beginning to be investigated. Epigenetic reprogramming of the genome during preimplantation may be altered due to an exogenous insult (MacPhee, 1998), resulting in permanent alteration of DNA methylation patterns (Guerrero-Bosagna et al., 2005). DNA methylation occurs at two points in development:(1) during gastrulation, which affects somatic cell development following fertilization, and (2) during gonad development, which leads to changes in methylation patterns that are sex specific and can alter the heritable epigenetic information (Anway and Skinner, 2006). As described for the mouse (Swales and Spears, 2005), mature, hypermethylated gene sequences of the parental male and female gametes converge during fertilization to form the zygote. During early embryo development, imprinted genes of the somatic and primordial germ cells (PGC) retain the parental imprints, but nonimprinted genes are demethylated. Within two weeks following fertilization, inherited parental imprinted genes of PGCs undergo demethylation; however, somatic cells maintain their imprints throughout embryo/fetal development and into adulthood. Hence, embryonic alteration of DNA methylation patterns can ultimately be expressed in adult life. Developmental changes in imprinted genes have been demonstrated in mice exposed to estrogen and other estrogenic chemicals such as DES (Newbold and Jefferson, 2006; Crews and McLachlan, 2006; McLachlan, 2001). Lactoferrin expression is regulated by estrogen in the uterine epithelial cells of adult mice, and this gene is abnormally expressed following neonatal exposure to DES. The gene continues to be abnormally expressed into adulthood, even if the animals are ovariectomized, implying that gene expression that was once estrogen dependent is now independent of estrogen regulation. The lactotransferrin gene was monitored in mice for methylation or demethylation of the regulatory elements of the gene following exposure to estrogen or DES during different periods of life (Li et al., 1997). A region upstream of the estrogen response element (ERE) of the lactoferrin gene promoter is abnormally demethylated at one CpG site in mature uteri of mice exposed to DES as neonates. The control mice, on the other hand, had normal methylation patterns. A one base pair change in the methylation pattern of the DNA can have a strong effect on gene expression, influencing which genes are transcribed or repressed, and the methylation/ demethylation patterns observed relative to the lactoferrin gene in mice exposed to DES may be linked to its ability to promote tumor formation in adult animals initially exposed as neonates (Swales and Spears, 2005; Li et al., 1997, 2003; McLachlan, 2001). Some diseases associated with imprinting errors include certain cancers and immunodeficiencies,

686

ENDOCRINE ACTIVE CHEMICALS: BROADENING THE SCOPE

centromeric region instability, facial anomalies syndrome, and Rett syndrome (Swales and Spears, 2005). Methylation pattern change can also lead to a decrease in DNA methylation. Undermethylation occurred on the 45S ribosomal RNA gene in the sperm of mice exposed to Cr (III) chloride. These changes were then associated with neoplastic and thyroid hormone (T3) changes in the offspring of exposed adults (Cheng et al., 2004). Although changes beyond the F1 generation were not investigated, these initial findings demonstrate a transgenerational effect and a chemical’s ability to potentially alter epigenetic inheritance, which could be transferred up to and beyond the F3 generation. Epigenetic transgenerational inheritance has recently been documented in the offspring of rodents exposed to EACs during gestation (Anway et al., 2005). Exposure of pregnant rats to high doses of vinclozolin (a fungicide with antiandrogenic activity) or methoxychlor (a pesticide with estrogenic, antiestrogenic, and antiandrogenic activity), during the period of embryo gonadal differentiation, resulted in decreased spermatogenic capacity and increased male infertility in the first generation of adults. These effects were transferred to the next three generations of offspring (F1–F3) as they reached adulthood in the vinclozolin treated line and through the F2 generation with the methoxychlor treated line. Moreover, the effects of vinclozolin appeared to be mediated by alterations in DNA methylation patterns through the male germlines during reprogramming. Although the doses were high, the 2005 study by Anway and colleagues provides evidence that environmental chemicals can have latent effects that are caused by changes in the epigenetic manner of inheritance in mammals. Furthermore, when the F1, F2, and F3 generations were allowed to grow and develop between 6 and 14 months of age, a greater than normal number of illnesses occurred in the mice including tumor development, prostate disease, kidney disease, testes abnormalities, immune system defects, and blood abnormalities (Anway and Skinner, 2006). The frequency of these disease phenotypes remain consistent for four generations and indicate that they are linked to the changes in methylation patterns observed in the germ cells. Several of the genes methylated in the vinclozolin exposure studies were identified and confirm an epigenetic alteration in the male germline (Chang et al., 2006). Continued work is needed to understand how these changes are associated with the transgenerational effects observed in these studies. The ultimate impacts of environmental chemicals on transgenerational effects, especially through epigenetic inheritance, are difficult to predict at this time. However, the results of the few studies described above warrant a greater need for research in this area with special consideration for hazard identification and risk assessment, especially if these effects prove to be underlying factors for common human and wildlife diseases.

18.7 CONCLUSION As discussed in this chapter, the endocrine system is integrative in nature and can be affected directly or indirectly, at many levels, in response to chemical exposures. These effects can be immediately detrimental to the organism (e.g., thyroid status during in utero development) or can be expressed later in life or in subsequent generations. Key components of the endocrine system serve to send and receive hormonal signals to and from target tissues of endocrine as well as nonendocrine organs until they are eliminated through metabolism and excretion. Among the topics chosen for inclusion in this chapter are emerging areas in endocrine research considered pivotal to assessing the full range of endocrine-related effects. Emphasis

REFERENCES

687

has also been given to different biological levels of organization ranging from molecular and cellular to whole animal, developmental, and transgenerational effects, which at the higher levels are indicative of what is happening in the real world. Due to the inherent difficulty in making causal relationships without filling in all of the links in between (i.e., altered gene expression from EACs relating to clinical disease), more research is needed in this area. We have also illustrated relevant end points and clinical signs indicative of endocrine action related to Fig. 18.1 through each of the components of the endocrine system (Table 18.1). As research findings are revealed, our understanding of how we respond to chemicals that affect the endocrine system will advance. Ultimately, such information will reveal the hazards and risks of EACs for humans and wildlife. Although the approach to risk assessment in humans focuses on the individual, and that in wildlife primarily emphasizes the population, except for endangered or charismatic species, due to the conserved nature of the endocrine system across taxa, many of the end points used to detect exposure to EACs are similar and readily translated from nonhuman animals to humans. Also, it is prudent to remember that wildlife data can be useful to estimate risk for human health. As demonstrated by this review, understanding the effects of EACs is a complicated task, because it includes myriad events from DNA regulation to population-level effects. It is our hope that this review stimulates interest in the study of consequences of exposure to EACs in humans and wildlife, encouraging scientists from different disciplines to think about the research issues facing this field and to possibly take a more global approach in this line of research.

REFERENCES Adams WJ, Rowland CD (2003) Aquatic Toxicology Test Methods. In:Hoffman DJ, Rattner BA, Burton GA, Jr, Cairns J, Jr, editors. Handbook of Ecotoxicology, 2nd edn. CRC Press LLC, Baca Raton, FL. 2(19–43). Allagui MS, Hfaiedh N, Vincent C, Guermazi F, Murat JC, Croute F, El Feki A (2006) Changes in growth rate and thyroid- and sex-hormones blood levels in rats under sub-chronic lithium treatment. Hum. Exp. Toxicol. 25(5):243–250. Andersen S, Pedersen KM, Bruun NH, Laurberg P (2002) Narrow individual variations in serum T(4) and T(3) in normal subjects: a clue to the understanding of subclinical thyroid disease. J. Clin. Endocrinol. Metab. 87(3):1068–1072. Andersen S, Bruun NH, Pedersen KM, Lau berg P (2003) Biologic variation is important for interpretation of thyroid function tests. Thyroid 13(11):1069–1078. Ankley GT, Johnson RD (2004) Small fish models for identifying and assessing the effects of endocrine-disrupting chemicals. Inst. Lab. Animal Res. 45(4):469–483. Ankley GT, Jensen KM, Kahl MD, Korte JJ, Makynen EA (2001) Description and evaluation of a short-term reproduction test with the fathead minnow (Pimephales promelas). Environ. Toxicol. Chem. 20(6):1276–1290. Ankley GT, Jensen KM, Makynen EA, Kahl MD, Korte JJ, Hornung MW, Henry TR, Denny JS, Leino RL, Wilson VS, Cardon MC, Hartig PC, Gray L (2003) Effects of the androgenic growth promoter 17-beta-trenbolone on fecundity and reproductive endocrinology of the fathead minnow. Environ. Toxicol. Chem. 22(6):1350–1360. Anway MD, Skinner MK (2006) Epigenetic transgenerational actions of endocrine disruptors. Endocrinol. 147:(6 Suppl):S43–S49. Anway MD, Cupp AS, Uzumcu M, Skinner MK (2005) Epigenetic transgenerational actions of endocrine disruptors and male fertility. Science 308(5727):1466–1469.

688

ENDOCRINE ACTIVE CHEMICALS: BROADENING THE SCOPE

Anway MD, Memon MA, Uzumcu M, Skinner MK (2006a) Transgenerational effect of the endocrine disruptor vinclozolin on male spermatogenesis. J. Androl. 27(6)868–879. Aoki Y, Belin RR, Phillips L, Clickner R, Jeffries R, Mahaffey KR (2006) Serum TSH and total T4 in the United States population and their association with participant characteristics: National Health and Nutrition Examination Survey (NHANES 1999–2002). Thyroid. 17(12):1211–1223. Arukwe A, Knudsen FR, Gogsoyr A (1997) Fish zona radiata (eggshell) proteins: a sensitive biomarker for environmental estrogens. Environ. Health Perspect. 105:418–422. Arukwe A, Male R, Johnson L, Peck-Miller K, Collier T, Gogsoyr A (2000) Fish zona radiata (eggshell) proteins: evaluation and validation as a biomarker for xenoestrogen monitoring. Mar. Environ. Res. 50(1–5):195–196. Bagchi N, Brown RE, Mack RE (1978) Studies on the mechanism of inhibition of thyroid function by lithium. Biochim. Biophys. Acta 542(1):163–169. Banerji TK, Maitra SK, Dey M, Hawkins HK (2001) Gametogenic responses of the testis in spotted munia (Lonchura punctulata; Aves) to oral administration of lithium chloride. Endocr. Res. 27 (3):345–356. Barker DJ (2003) The developmental origins of adult disease. Eur. J. Epidemiol. 18(8):733–736. Ben Saad MM, Maurel DL (2004) Reciprocal interaction between seasonal testis and thyroid activity in Zembra Island wild rabbits (Oryctolagus cuniculus): effects of castration, thyroidectomy, temperature, and photoperiod. Biol. Reprod. 70(4):1001–1009. Berens SC, Bernstein RS, Robbins J, Wolff J (1970) Antithyroid effects of lithium. J. Clin. Invest. 49 (7):1357–1367. Bernal J (2005) Thyroid hormones and brain development. Vitam. Horm. 71:95–122. Bianco AC, Kim BW (2006) Deiodinases: implications of the local control of thyroid hormone action. J. Clin. Invest. 116(10):2571–2579. Birnbaum LS, Fenton SE (2003) Cancer and developmental exposure to endocrine disruptors. Environ. Health Perspect. 111(4):389–394. Bjoro T, Holmen J, Kruger O, Midthjell K, Hunstad K, Schreiner T, Sandnes L, Brochmann H (2000) Prevalence of thyroid disease, thyroid dysfunction and thyroid peroxidase antibodies in a large, unselected population. The Health Study of Nord-Trondelag (HUNT). Eur. J. Endocrinol. 143 (5):639–647. Blount BC, Pirkle JL, Osterloh JD, Valentin-Blasini L, Caldwell KC (2006) Urinary perchlorate and thyroid hormone levels in adolescent and adult men and women living in the United States. Environ. Health Perspect. 114(12):1865–1871. Bocchetta A, Cherchi A, Loviselli A, Mossa P, Velluzzi F, Derai R, Del Zompo M (1996) Six-year follow-up of thyroid function during lithium treatment. Acta Psychiatr. Scand. 94 (1):45–48. Bocchetta A, Loviselli A (2006) Lithium treatment and thyroid abnormalities. Clin. Pract. Epidemol. Ment. Health. 2:23. Bortone S, Davis W (1994) Fish intersexuality as indicator of environmental stress. Bioscience 44:165–172. Bortone S, Davis W, Bundrick CM (1989) Morphological and behavioral characters in mosquitofish as potential bioindication of exposure to Kraft mill effluent. Bull. Environ. Contam. Toxicol. 43:370–377. Bowman CJ, Kroll KJ, Gross TG, Denslow ND (2002) Estradiol-induced gene expression in largemouth bass (Micropterus salmoides). Mol. Cell Endocrinol. 196(1–2):67–77. Boyd GR, Reemtsma H, Grimm DA, Mitra S (2003) Pharmaceuticals and personal care products (PPCPs) in surface and treated waters of Louisiana, USA and Ontario, Canada. Sci. Total Environ. 311(1–3):135–149.

REFERENCES

689

Braga O, Smythe GA, Schafer AI, Feitz AJ (2005) Steroid estrogens in primary and tertiary wastewater treatment plants. Water Sci. Technol. 52(8):273–278. Brahmbhatt S, Brahmbhatt RM, Boyages SC (2000) Thyroid ultrasound is the best prevalence indicator for assessment of iodine deficiency disorders: a study in rural/tribal schoolchildren from Gujarat (Western India). Eur. J. Endocrinol. 143(1):37–46. Braverman LE, He X, Pino S, Cross M, Magnani B, Lamm SH, Kruse MB, Engel A, Crump KS, Gibbs JP (2005) The effect of perchlorate, thiocyanate, and nitrate on thyroid function in workers exposed to perchlorate long-term. J. Clin. Endocrinol. Metab. 90(2):700–706. Bromel C, Pollard RE, Kass PH, Samii VF, Davidson AP, Nelson RW (2006) Comparison of ultrasonographic characteristics of the thyroid gland in healthy small-, medium-, and large-breed dogs. Am. J. Vet. Res. 67(1):70–77. Brouwer A, Ahlborg UG, van Leeuwen FX, Feeley MM (1998) Report of the WHO working group on the assessment of health risks for human infants from exposure to PCDDs, PCDFs and PCBs. Chemosphere 37(9–12):1627–1643. Campbell JH, Perkins P (1988) Transgenerational effects of drug and hormonal treatments in mammals: a review of observations and ideas. Prog. Brain Res. 73:535–553. Cargouet M, Perdiz D, Mouatassim-Souali A, Tamisier-Karolak S, Levi Y (2004) Assessment of river contamination by estrogenic compounds in Paris area (France). Sci. Total Environ. 324(1–3):55– 66. Carlsen E, Giwercman A, Keiding N, Skakkebaek NE (1992) Evidence for decreasing quality of semen during past 50 years. Brit. Med. J. 305(6854):609–613. Carpenter DO (1998) Polychlorinated biphenyls and human health. Int. J. Occup. Med. Environ. Health 11(4):291–303. Celius T, Walther BT (1998) Oogenesis in Atlantic salmon (Salmo salar L.) occurs by zonagenesis preceding vitellogenesis in vivo and in vitro. J. Endocrinol. 158:259–266. Chandra AK, Mukhopadhyay S, Lahari D, Tripathy S (2004) Goitrogenic content of Indian cyanogenic plant foods & their in vitro anti-thyroidal activity. Indian J. Med. Res. 119(5):180–185. Chang HS, Anway MD, Rekow SS, Skinner MK (2006) Transgenerational epigenetic imprinting of the male germ-line by endocrine disruptor exposure during gonadal sex determination. Endocrinol. 147(12):5524–5541. Chanoine JP, Braverman LE, Farwell AP, Safran M, Alex S, Dubord S, Leonard JL (1993) The thyroid gland is a major source of circulating T3 in the rat. J Clin. Invest 91(6):2709–2713. Chatterjee S, Roden K, Banerji TK (1990) Morphological changes in some endocrine organs in rats following chronic lithium treatment. Anat. Anz. 170(1):31–37. Cheng RY, Hockman T, Crawford E, Anderson LM, Shiao YH (2004) Epigenetic and gene expression changes related to transgenerational carcinogenesis. Mol. Carcinog. 40(1):1–11. Colborn T, vom Saal FS, Soto AM (1993) Developmental effects of endocrine-disrupting chemicals in wildlife and humans. Environ. Health Perspect. 101(5):378–384. Colerangle JB, Roy D (1997) Profound effects of the weak environmental estrogen-like chemical bisphenol A on the growth of the mammary gland of Noble rats. J. Steroid Biochem. Mol. Biol. 60 (1–2):153–160. Colon I, Caro D, Bourdony CJ, Rosario O (2000) Identification of phthalate esters in the serum of young Puerto Rican girls with premature breast development. Environ. Health Perspect. 108 (9):895–900. Cooper RL, Kavlock RJ (1997) Endocrine disruptors and reproductive development: a weight-ofevidence overview. J. Endocrinol. 152(2):159–166. Crews D, McLachlan JA (2006) Epigenetics, evolution, endocrine disruption, health, and disease. Endocrinol. 147:(6 Suppl):S4–S10.

690

ENDOCRINE ACTIVE CHEMICALS: BROADENING THE SCOPE

D’Ascenzo G, Di Corcia A, Gentili A, Mancini R, Mastropasqua R, Nazzari M, Samperi R (2003) Fate of natural estrogen conjugates in municipal sewage transport and treatment facilities. Sci. Total Environ. 302(1–3):199–209. Daniel GB, Sharp DS, Nieckarz JA, Adams W (2002) Quantitative thyroid scintigraphy as a predictor of serum thyroxin concentration in normal and hyperthyroid cats. Vet. Radiol. Ultrasound 43 (4):374–382. Dasgupta PK, Martinelango PK, Jackson WA, Anderson TA, Tian K, Tock RW, Rajagopalan S (2005) The origin of naturally occurring perchlorate: the role of atmospheric processes. Environ. Sci. Technol. 39(6):1569–1575. Daughton CG, Ternes TA (1999) Pharmaceuticals and personal care products in the environment: agents of subtle change? Environ. Health Perspect. 107 (Suppl 6):907–938. Davidson JS, Bolland MJ, Croxson MS, Chiu W, Lewis JG (2006) A case of low cortisol-binding globulin: use of plasma free cortisol in interpretation of hypothalamic-pituitary–adrenal axis tests. Ann. Clin. Biochem. 43(Pt 3):237–239. DeFur PI (2004) Use and role of invertebrate models in endocrine disruptor research and testing. Inst. Lab. Animal Res. 45(4):484–493. Delange F (1994) The disorders induced by iodine deficiency. Thyroid 4(1):107–128. DeLuca HF (2004) Overview of general physiologic features and functions of vitamin D. Am. J. Clin. Nutr. 80(6 Suppl):1689S–1696S. Denslow ND, Larkin P (2006) Utilizing molecular technologies for bioindicator research. Environ. Bioind. 1:40–59. Doerge DR, Chang HC (2002) Inactivation of thyroid peroxidase by soy isoflavones, in vitro and in vivo. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 777(1–2):269–279. Doerge DR, Sheehan DM (2002b) Goitrogenic and estrogenic activity of soy isoflavones. Environ. Health Perspect. 110(Suppl 3):349–353. Downie SE, Wasnidge C, Floto F, Robinson GA (1977) Lithium-induced inhibition of 125I accumulation by thyroids and growing oocytes of Japanese quail. Poult. Sci. 56(4):1254–1258. Dusso AS, Brown AJ, Slatopolsky E (2005) Vitamin D. Am. J. Physiol. Renal Physiol. 289(1):F8–28. Edelstein S, Fullmer CS, Wasserman RH (1984) Gastrointestinal absorption of lead in chicks: involvement of the cholecalciferol endocrine system. J. Nutr. 114(4):692–700. Eidem JK, Kleivdal H, Kroll D, Denslow ND, van Aerle R, Tyler CR, Panter GH, Hutchinson TH, Gogsoyr A (2006) Development and validation of a direct homologous quantitative sandwich ELISA for fathead minnow (Pimephales promelas) vitellogenin. Aquat. Toxicol. 78(2):202–206. Eskiocak S, Dundar C, Basoglu T, Altaner S (2005) The effects of taking chronic nitrate by drinking water on thyroid functions and morphology. Clin. Exper. Med. 5(2):66–71. Etling N, Levy M, Fouque F (1987) Thyroid hormones in rats receiving increasing doses of lithium. Ann. Endocrinol (Paris). 48(6):452–456. Fenske M, Segner H (2004) Aromatase modulation alters gonadal differentiation in developing zebrafish (Danio rerio). Aquat. Toxicol. 67(2):105–126. Fent K, Weston AA, Caminada D (2006) Ecotoxicology of human pharmaceuticals. Aquat Toxicol 76 (2):122–159. Fenton SE (2006) Endocrine-disrupting compounds and mammary gland development: early exposure and later life consequences. Endocrinol. 147(6 Suppl):S18–S24. Fisher JW, Campbell Muralidhara S, Bruckner JV, Ferguson D, Mumtaz M, Harmon B, Hedge JM, Crofton KM, Kim H, Almekinder TL (2006) Effect of PCB 126 on hepatic metabolism of thyroxine and perturbations in the hypothalamic-pituitary–thyroid axis in the rat. Toxicol. Sci. 90(1):87–95. Fleet JC (2004) Genomic and proteomic approaches for probing the role of vitamin D in health. Am. J. Clin. Nutr. 80(6 Suppl):1730S–1734S.

REFERENCES

691

Flouriot G, Pakdel F, Ducouret B, Valotaire Y (1995) Influence of xenobiotics on rainbow trout liver estrogen receptor and vitellogenin gene expression. J. Mol. Endocrinol. 15(2):143–151. Foo LC, Zulfiqar A, Nafikudin M, Fadzil MT, Asmah AS (1999) Local versus WHO/International Council for Control of Iodine Deficiency Disorders-recommended thyroid volume reference in the assessment of iodine deficiency disorders. Eur. J. Endocrinol. 140(6):491–497. Forbes VE, Palmqvist A, Bach L (2006) The use and misuse of biomarkers in ecotoxicology. Environ. Toxicol. Chem. 25(1):272–280. Fort DJ, Rogers RL, Morgan LA, Miller MF, Clark PA, White JA, Paul RR, Stover EL (2000) Preliminary validation of a short-term morphological assay to evaluate adverse effects on amphibian metamorphosis and thyroid function using Xenopus laevis. J. Appl. Toxicol. 20 (5):419–425. Fountas L, Moutsatsou P, Kastanias I, Tamouridis N, Tzanela M, Anapliotou M, Sekeris CE (1999) The contribution of vitamin D receptor gene polymorphisms in osteoporosis and familial osteoporosis. Osteoporos. Int. 10(5):392–398. Fowles JR, Fairbrother A, Trust KA, Kerkvliet NI (1997) Effects of Aroclor 1254 on the thyroid gland, immune function, and hepatic cytochrome P450 activity in mallards. Environ. Res. 75 (2):119–129. Frankenfeld TG, Correa DC, Nascimento-Saba CC, Ortiga-Carvalho TM, Santos RM, Lisboa PC, Carvalho DP, Rosenthal D (2002) Thyroid and pituitary thyroxine-50 -deiodinase activity and thyrotrophin secretion in lithium-treated rats. J. Endocrinol. 174(2):331–334. Fraser DR, Kodicek E (1970) Unique biosynthesis by kidney of a biological active vitamin D metabolite. Nature 228(5273):764–766. Frau´sto da Silvaa JJR, Williams RJP (1993) The Biological Chemistry of the Elements: The Inorganic Chemistry of Life. Oxford: Oxford University Press. Gagne F, Blaise C (2003) Effects of municipal effluents on serotonin and dopamine levels in the freshwater mussel Elliptio complanata. Comp. Physiol. Biochem. C 136(2):117–125. Gaitan E, Lindsay RH, Reichert RD, Ingbar SH, Cooksey RC, Legan J, Meydrech EF, Hill J, Kubota K (1989) Antithyroid and goitrogenic effects of millet: role of C-glycosylflavones. J. Clin. Endocrinol. Metab. 68(4):707–714. Gao X, Petroff BK, Rozman KK, Terranova PF (2000) Gonadotropin-releasing hormone (GnRH) partially reverses the inhibitory effect of 2,3,7,8-tetrachlorodibenzo-p-dioxin on ovulation in the immature gonadotropin-treated rat. Toxicol. 147(1):15–22. Ghekiere A, Verslycke T, Janssen CR (2006) Effects of methoprene, nonylphenol, and estrone on the vitellogenesis of the mysid Neomysis integer. Gen. Comp. Endocrinol. 147:190–195. Gimeno S, Gerritsen A, Bowmer T, Komen H (1996) Feminization of male carp. Nature 384 (6606):221–222. Goldey ES, Kehn LS, Lau C, Rehnberg GL, Crofton KM (1995) Developmental exposure to polychlorinated biphenyls (Aroclor 1254) reduces circulating thyroid hormone concentrations and causes hearing deficits in rats. Toxicol. Appl. Pharmacol. 135(1):77–88. Gould JC, Cooper KR, Scanes CG (1997) Effects of polychlorinated biphenyl mixtures and three specific congeners on growth and circulating growth-related hormones. Gen. Comp. Endocrinol. 106(2):221–230. Gould JC, Cooper KR, Scanes CG (1999) Effects of polychlorinated biphenyls on thyroid hormones and liver type I monodeiodinase in the chick embryo. Ecotoxicol. Environ. Saf. 43(2):195–203. Gray LE Jr (1998) Tiered screening and testing strategy for xenoestrogens and antiandrogens. Toxicol. Lett. 102–103:677–680. Gray LE, Jr, Wilson V, Noriega N, Lambright C, Furr J, Stoker TE, Laws SC, Goldman J, Cooper RL, Foster PM (2004) Use of the laboratory rat as a model in endocrine disruptor screening and testing. Inst. Lab. Animal Res. 45(4):425–437.

692

ENDOCRINE ACTIVE CHEMICALS: BROADENING THE SCOPE

Greer MA, Goodman G, Pleus RC, Greer SE (2002) Health effects assessment for environmental perchlorate contamination: the dose response for inhibition of thyroidal radioiodine uptake in humans. Environ. Health Perspect. 110(9):927–937. Gronen S, Denslow N, Manning S, Barnes S, Barnes D, Brouwer M (1999) Serum vitellogenin levels and reproductive impairment of male Japanese Medaka (Oryzias latipes) exposed to 4-tertoctylphenol. Environ. Health Perspect. 107(5):385–390. Guerrero-Bosagna C, Sabat P, Valladares L (2005) Environmental signaling and evolutionary change: can exposure of pregnant mammals to environmental estrogens lead to epigenetically induced evolutionary changes in embryos? Evol. Dev. 7(4):341–350. Guiguen Y, Baroiller JF, Ricordel MJ, Iseki K, McMeel OM, Martin SA, Fostier A (1999) Involvement of estrogens in the process of sex differentiation in two fish species: the rainbow trout (Oncorhynchus mykiss) and a tilapia (Oreochromis niloticus). Mol. Reprod. Dev. 54:154–162. Gupta C (2000) Reproductive malformation of the male offspring following maternal exposure to estrogenic chemicals. Proc. Soc. Exp. Biol. Med. 224(2):61–68. Hagmar L (2003) Polychlorinated biphenyls and thyroid status in humans: a review. Thyroid 13 (11):1021–1028. Hagmar L, Bjork J, Sjodin A, Bergman A, Erfurth EM (2001a) Plasma levels of persistent organohalogens and hormone levels in adult male humans. Arch. Environ. Health 56(2):138–143. Hagmar L, Rylander L, Dyremark E, Klasson-Wehler E, Erfurth EM (2001b) Plasma concentrations of persistent organochlorines in relation to thyrotropin and thyroid hormone levels in women. Int. Arch. Occup. Environ. Health 74(3):184–188. Halldin K, Holm L, Ridderstrale Y, Brunstrom B (2003) Reproductive impairment in Japanese quail (Coturnix japonica) after in ovo exposure to o,p0 -DDT. Arch. Toxicol. 77(2):116–122. Halldin K, Axelsson J, Brunstrom B (2005) Effects of endocrine modulators on sexual differentiation and reproductive function in male Japanese quail. Brain Res. Bull. 65(3):211–218. Hansen PS, Brix TH, Sorensen TI, Kyvik KO, Hegedus L (2004) Major genetic influence on the regulation of the pituitary–thyroid axis: a study of healthy Danish twins. J. Clin. Endocrinol. Metab. 89(3):1181–1187. Harries JE, Runnalls T, Hill E, Harris CA, Maddix S, Sumpter JP, Tyler CR (2000) Development of a reproductive performance test for endocrine disrupting chemicals using pair-breeding fathead minnow (Pimephales promelas). Environ. Sci. Technol. 34(14):3003–3011. Harvey PW, Everett DJ (2003) The adrenal cortex and steroidogenesis as cellular and molecular targets for toxicity: critical omissions from regulatory endocrine disrupter screening strategies for human health ? J. Appl. Toxicol. 23(2):81–87. Haynes EN, Kalkwarf HJ, Hornung R, Wenstrup R, Dietrich K, Lanphear BP (2003) Vitamin D receptor Fok1 polymorphism and blood lead concentration in children. Environ. Health Perspect 111(13):1665–1669. Hemmer M, Bowman CJ, Hemmer BL, Friedman SD, Marcovich D, Kroll K, Denslow ND (2002) Vitellogenin mRNA regulation and plasma clearance in male sheepshead minnows (Cyprinodon variegatus) after cessation of exposure to 17 beta-estradiol and p-nonylphenol. Aquat. Toxicol. 58:99–112. Hernandez-Diaz S (2002) Iatrogenic legacy from diethylstilbestrol exposure. Lancet 359 (9312):1081–1082. Hess SY, Zimmermann MB (2000) Thyroid volumes in a national sample of iodine-sufficient Swiss school children: comparison with the World Health Organization/International Council for the Control of Iodine Deficiency Disorders normative thyroid volume criteria. Eur. J. Endocrinol. 142 (6):599–603. Holick MF (2004) Sunlight and vitamin D for bone health and prevention of autoimmune diseases, cancers, and cardiovascular disease. Am. J. Clin. Nutr. 80(6 Suppl):1678S–1688S.

REFERENCES

693

Holick MF, Frommer JE, McNeill SC, Richtand NM, Henley JW, Potts JT Jr (1977) Photometabolism of 7-dehydrocholesterol to previtamin D3 in skin. Biochem. Biophys. Res. Commun. 76(1):107– 114. Hollowell JG, Staehling NW, Flanders WD, Hannon WH, Gunter EW, Spencer CA, Braverman LE (2002) Serum TSH, T(4), and thyroid antibodies in the United States population (1988 to 1994): National Health and Nutrition Examination Survey (NHANES III). J. Clin. Endocrinol. Metab. 87 (2):489–499. Horiguchi T, Kojima M, Kaya M, Matsuo T, Shiraishi H, Morita M, Adachi Y (2002) Tributyltin and triphenyltin induce spermatogenesis in ovary of female abalone, Haliotis gigantea. Mar. Environ. Res. 54(3–5):679–684. Howell WM, Black DA, Bortone SA (1980) Abnormal expression of secondary sex characters in a population of mosquitofish, Gambusia affinis holbrooki: evidence for environmentally-induced masculinization. Copeia 4:676–681. Hultman P, Bell LJ, Enestrom S, Pollard KM (1992) Murine susceptibility to mercury. I. Autoantibody profiles and systemic immune deposits in inbred, congenic, and intra-H-2 recombinant strains. Clin. Immunol. Immunopathol. 65(2):98–109. Hutchinson TH (2004) Biomarkers of Endocrine Disruption in OECD Fish Species: Background Document on the Gonadosomatic Index (GSI). No. BL7508/A.Brixham, UK:AstraZeneca Brixham Environmental Laboratory. Hutchinson TH, Ankley GT, Segner H, Tyler CR (2005) Screening and testing for endocrine disruption in fish: biomarkers as “signposts,” not “traffic lights,” in risk assessment. Environ. Health Perspect. 114(Supp. 1):106–114. Inui M, Adachi T, Takenaka S, Inui H, Nakazawa M, Ueda M, Watanabe H, Mori C, Iguchi T, Miyatake K (2003) Effect of UV screens and preservatives on vitellogenin and choriogenin production in male medaka. Toxicol. 194:43–50. Islinger M, Williamski D, Voikl A, Braunbeck T (2003) Effects of 17a-ethinylestradiol on the expression of three estrogen responsive genes and cellular unltrastructure of liver and testes in male zebrafish. Aquat. Toxicol. 62:85–103. Jobling S, Nolan M, Tyler CR, Brighty G, Sumpter JP (1998) Widespread sexual disruption in wild fish. Environ. Sci. Technol. 32(17):2498–2506. Johnson AC, Williams RJ, Simpson P, Kanda R (2007) What difference might sewage treatment performance make to endocrine disruption in rivers? Environ Pollut. 147:194–202. Kamata R, Takahashi S, Shimizu A, Shiraishi F (2006) Avian transgenerational reproductive toxicity test with in ovo exposure. Arch. Toxicol. 80(12):846–856. Kanamori A (2000) Systematic identification of genes expressed during early oogenesis in medaka. Mol. Reprod. Dev. 55:31–36. Kester MHA, Bulduk S, van Toor H, Tibboel D, Meinl W, Glatt H, Falany CN, Coughtrie MWH, Schuur AG, Brouwer A, Visser TJ (2002) Potent inhibition of estrogen sulfotransferase by hydroxylated metabolites of polyhalogenated aromatic hydrocarbons reveals alternative mechanism for estrogenic activity of endocrine disrupters. J. Clin. Endocrinol. Metab. 87:1142–1150. Khan IA, Mathews S, Okuzawa K, Kagawa H, Thomas P (2001) Alterations in the GnRH-LH system in relation to gonadal stage and Aroclor 1254 exposure in Atlantic croaker. Comp. Biochem. Physiol. B: Biochem. Molec. Biol. 129(2–3):251–259. Kiraly SJ, Kiraly MA, Hawe RD, Makhani N (2006) Vitamin D as a neuroactive substance: review. Scient. World J. 6:125–139. Kirk AB, Martinelango PK, Tian K, Dutta A, Smith EE, Dasgupta PK (2005) Perchlorate and iodide in dairy and breast milk. Environ. Sci. Technol. 39(7):2011–2017. Klip H, Verloop J, van Gool JD, Koster ME, Burger CW, van Leeuwen FE (2002) Hypospadias in sons of women exposed to diethylstilbestrol in utero: a cohort study. Lancet 359(9312):1102–1107.

694

ENDOCRINE ACTIVE CHEMICALS: BROADENING THE SCOPE

Koibuchi N, Iwasaki T (2006) Regulation of brain development by thyroid hormone and its modulation by environmental chemicals. Endocr. J. 53(3):295–303. Koopman-Esseboom C, Morse DC, Weisglas-Kuperus N, Lutkeschipholt IJ, Van der Paauw CG, Tuinstra LG, Brouwer A, Sauer PJ (1994) Effects of dioxins and polychlorinated biphenyls on thyroid hormone status of pregnant women and their infants. Pediatr. Res. 36(4):468–473. Kszos Lakowicz JR, Stewart AJ (2003) Review of lithium in the aquatic environment: distribution in the United States, toxicity and case example of groundwater contamination. Ecotox. 12(5):439– 447. Kumar RS, Ijiri S, Trant JM (2000) Changes in the expression of genes encoding steroidogenic enzymes in the channel catfish (Ictalurus punctatus) ovary throughout a reproductive cycle. Biol. Reprod. 63:1676–1682. Langer P, Tajtakova M, Fodor G, Kocan A, Bohov P, Michalek J, Kreze A (1998) Increased thyroid volume and prevalence of thyroid disorders in an area heavily polluted by polychlorinated biphenyls. Eur. J. Endocrinol. 139(4):402–409. Langer P, Kocan A, Tajtakova M, Petrik J, Chovancova J, Drobna B, Jursa S, Pavuk M, Koska J, Trnovec T, Sebokova E, Klimes I (2003) Possible effects of polychlorinated biphenyls and organochlorinated pesticides on the thyroid after long-term exposure to heavy environmental pollution. J. Occup. Environ. Med. 45(5):526–532. Langer P, Kocan A, Tajtakova M, Petrik J, Chovancova J, Drobna B, Jursa S, Pavuk M, Trnovec T, Sebokova E, Klimes I (2005) Human thyroid in the population exposed to high environmental pollution by organochlorinated pollutants for several decades. Endocr. Regul. 39(1):13–20. Lau C, Rogers JM (2004) Embryonic and fetal programming of physiological disorders in adulthood. Birth Defects Res. C. Embryo. Today 72(4):300–312. Lavado R, Janer G, Porte C (2006a) Steroid levels and steroid metabolism in the mussel Mytilus edulis: The modulating effect of dispersed crude oil and alkylphenols. Aquat. Toxicol. 78 (Suppl 1):S65– S72. Lavado R, Sugni M, Candia C, Porte C (2006b) Triphenyltin alters androgen metabolism in the sea urchin Paracentrotus lividus. Aquat. Toxicol. 79:247–256. Lawn J, Cousens S, Zupan J (2005) 4 million neonatal deaths: When? Where? Why? Lancet 365 (9462):891–900. Laws SC, Ferrell JM, Stoker TE, Schmid J, Cooper RL (2000) The effects of atrazine on female wistar rats: an evaluation of the protocol for assessing pubertal development and thyroid function. Toxicol. Sci. 58(2):366–376. Lazarus JH (1998) Is thyroxine during lithium therapy necessary? J. Endocrinol. Invest 21(11):784– 786. Li S, Washburn KA, Moore R, Uno T, Teng C, Newbold RR, McLachlan JA, Negishi M (1997) Developmental exposure to diethylstilbestrol elicits demethylation of estrogen-responsive lactoferrin gene in mouse uterus. Cancer Res. 57(19):4356–4359. Li S, Hursting SD, Davis BJ, McLachlan JA, Barrett JC (2003) Environmental exposure, DNA methylation, and gene regulation: lessons from diethylstilbesterol-induced cancers. Ann. N.Y. Acad. Sci. 983:161–169. Lippman ME, Krueger KA, Eckert S, Sashegyi A, Walls EL, Jamal S, Cauley JA, Cummings SR (2001) Indicators of lifetime estrogen exposure: effect on breast cancer incidence and interaction with raloxifene therapy in the multiple outcomes of raloxifene evaluation study participants. J. Clin. Ocol. 19(12):3111–3116. Liu HK, Wang GX, Guo Z, Wang J, Konstantinov K (2006) Nanomaterials for lithium-ion rechargeable batteries. J Nanosci. Nanotechnol. 6(1):1–15. Longnecker MP, Gladen BC, Patterson DG, Jr, Rogan WF (2000) Polychlorinated biphenyl (PCB) exposure in relation to thyroid hormone levels in neonates. Epidemiol. 11(3):249–254.

REFERENCES

695

Loviselli A, Bocchetta A, Mossa P, Velluzzi F, Bernardi F, Del Zompo M, Mariotti S (1997) Value of thyroid echography in the long-term follow-up of lithium-treated patients. Neuropsychobiol. 36 (1):37–41. Maack G, Segner M (2003) Morphological development of the gonads in zebrafish. J. Fish Biol. 62:895–906. MacDonald PN, Dowd DR, Haussler MR (1994) New insight into the structure and functions of the vitamin D receptor. Semin. Nephrol. 14(2):101–118. Macdonald HM, McGuigan FE, Stewart A, Black AJ, Fraser WD, Ralston S, Reid DM (2006) Large-scale population-based study shows no evidence of association between common polymorphism of the VDR gene and BMD in British women. J. Bone Miner. Res. 21 (1):151–162. MacPhee DG (1998) Epigenetics and epimutagens: some new perspectives on cancer, germ line effects and endocrine disrupters. Mutat. Res. 400(1–2):369–379. Mahaffey KR, McKinney J, Routt Reigart J (2000) Lead and compounds. In:Lippman M, editor. Environmental Toxicants: Human Exposures and their Health Effects, 2nd edn, Chichester, UK: Wiley–Interscience. Mahaffey KR, Rosen JF, Chesney RW, Peeler JT, Smith CM, DeLuca HF (1982) Association between age, blood lead concentration, and serum 1,25-dihydroxycholecalciferol levels in children. Am. J. Clin. Nutr. 35(6):1327–1331. Markey CM, Luque EH, Munoz DT, Sonnenschein C, Soto AM (2001) In utero exposure to bisphenol A alters the development and tissue organization of the mouse mammary gland. Biol. Reprod. 65 (4):1215–1223. Mathieu C, Badenhoop C (2005) Vitamin D and type 1 diabetes mellitus: state of the art. Trends Endocrinol. Metab. 16(6):261–266. Matthiessen P (1998) Effects on fish of estrogenic substances in English rivers. In:Kendall RJ, Dickerson RL, Giesy JP, Suk WP, editors. Principles and Processes for Evaluating Endocrine Disruption in Wildlife. Pensacola, FL:SETAC Press. pp.239–247. McLachlan JA (2001) Environmental signaling: what embryos and evolution teach us about endocrine disrupting chemicals. Endocr. Rev. 22(3):319–341. McLachlan J, Simpson E, Martin M (2006) Endocrine disruptors and female reproductive health. Best Pract. Res. Clin. Endocrinol. Metab. 1:63–75. McMaster ME, Jardine JJ, Ankley GT, Benson WH, Greeley J, Gross TS, Guillette J, MacLatchy DL, Orlando EF, Van Der Kraak GJ, Munkittrick KR (2001) An interlaboratory study on the use of steroid hormones in examining endocrine disruption. Environ. Toxicol. Chem. 20 (9):2081–2087. McNabb FM, Jang DA, Larsen CT (2004a) Does thyroid function in developing birds adapt to sustained ammonium perchlorate exposure? Toxicol. Sci. 82(1):106–113. McNabb FM, Larsen CT, Pooler PS (2004b) Ammonium perchlorate effects on thyroid function and growth in bobwhite quail chicks. Environ. Toxicol. Chem. 23(4):997–1003. Meulenberg EP (2002) A new test to identify endocrine disruptors using sex hormone-binding globulins from human serum. Eur. J. Lipid Sci. Tech. 104(2):131–136. Miediratta SK, Dotson RI, Brooker RT (1996)Perchloric acid and perchlorates. In:Kroschwitz JI, Howe-Grant M,editors. Kirk–Othmer Encyclopedia of Chemical Technology, 4th edn, Vol. 18. New York: John Wiley & Sons. pp.157–170. Mizukami Y, Michigishi T, Nonomura A, Nakamura S, Noguchi M, Takazakura E (1995) Histological features of the thyroid gland in a patient with lithium induced thyrotoxicosis. J Clin. Pathol. 48 (6):582–584. Mooney CT (2001) Feline hyperthyroidism. Diagnostics and therapeutics. Vet. Clin. North Am. Small Anim. Pract. 31(5):963–983, viii.

696

ENDOCRINE ACTIVE CHEMICALS: BROADENING THE SCOPE

Moore JA (1995) An assessment of lithium using the IEHR Evaluative Process for Assessing Human Developmental and Reproductive Toxicity of Agents. IEHR Expert Scientific Committee. Reprod. Toxicol. 9(2):175–210. Moreno-Reyes R, Boelaert M, el Badawi S, Eltom M, Vanderpas JB (1993) Endemic juvenile hypothyroidism in a severe endemic goitre area of Sudan. Clin. Endocrinol. (Oxf) 38 (1):19–24. Mukhi S, Carr JA, Andersen TA, Patino R (2005) Novel biomarkers of perchlorate exposure in zebrafish. Environ. Toxicol. Chem. 24(5):1107–1115. Munoz A, Gange SJ (1998) Methodological issues for biomarkers and intermediate outcomes in cohort studies. Epidemiol. Rev 20(1):29–42. Munoz-de-Toro M, Markey CM, Wadia PR, Luque EH, Rubin BS, Sonnenschein C, Soto AM (2005) Perinatal exposure to bisphenol-A alters peripubertal mammary gland development in mice. Endocrinol. 146(9):4138–4147. NAS (2006) Human Biomonitoring for Environmental Chemicals. Washington, DC: The National Academies Press. Newbold RR, Padilla-Banks E, Jefferson WN (2006) Adverse effects of the model environmental estrogen diethylstilbestrol are transmitted to subsequent generations. Endocrinol. 147(6 Suppl): S11–S17. Newbold RR, Jefferson W (2006)Developmental and reproductive abnormalities associated with environmental estrogens: diethylstilbestrol (DES) as an example. In:Naz Rajesh K,editor. Endocrine Disruptors: Effects on male and female reproductive systems, 2nd edn.CRC Press. pp.47–65. Nice HE (2005) Sperm motility in the Pacific oyster (Crassostrea gigas) is affected by nonylphenol. Mar. Pollut. Bull. 50(12):1668–1674. Nielsen JB, Hultman P (2002) Mercury-induced autoimmunity in mice. Environ. Health Perspect. 110:(Suppl 5):877–881. Nilsen BM, Berg K, Eidem JK, Kristiansen S-V, Brion F, Porcher J-M, Gogsoyr A (2004) Development of quantitative vitellogenin-ELISAa for fish test species used in endocrine disruptor screening. Anal. Bioanal. Chem. 378:621–633. Norman AW, Okamura WH, Bishop JE, Henry HL (2002) Update on biological actions of 1alpha,25 (OH)2-vitamin D3 (rapid effects) and 24R,25(OH)2-vitamin D3. Mol. Cell Endocrinol. 197(1–2): 1–13. Norsworthy GD, Adams VJ, McElhaney MR, Milios JA (2002) Relationship between semi-quantitative thyroid palpation and total thyroxine concentration in cats with and without hyperthyroidism. J. Feline. Med. Surg. 4(3):139–143. NTP (2006) NTP technical report on the toxicology and carcinogenesis studies of 2,20 ,4,40 ,5,50 hexachlorobiphenyl (PCB 153) (CAS No. 35065-27-1) in female Harlan Sprague-Dawley rats (Gavage studies). National Toxicology Program Technical Report Series 529, pp. 4–168 Oberdorster E (2001) Peptide hormones versus steroid hormones: case studies from snail and turtle populations. Ann. N.Y. Acad. Sci. 948:75–79. Odum J, Ashby J (2002) Detection of aromatase inhibitors in vitro using rat ovary microsomes. Toxicol. Lett. 129(1–2):119–122. OECD. 8-2-2006. Detailed Review Paper (DRP) on Thyroid Hormone Disruption Assays. No. 57. Ogunkolade BW, Boucher BJ, Prahl JM, Bustin SA, Burrin JM, Noonan K, North BV, Mannan N, McDermott MF, DeLuca HF, Hitman GA (2002) Vitamin D receptor (VDR) mRNA and VDR protein levels in relation to vitamin D status, insulin secretory capacity, and VDR genotype in Bangladeshi Asians. Diabetes 51(7):2294–2300. Oliva A, Giami A, Multigner L (2002) Environmental agents and erectile dysfunction: a study in a consulting population. J. Androl. 23:546–550.

REFERENCES

697

Onalaja AO, Claudio L (2000) Genetic susceptibility to lead poisoning. Environ. Health Perspect. 108 (Suppl 1):23–28. Orn S, Holbech H, Madsen T, Norrgren L, Petersen G (2003) Gonad development and vitellogenin production in zebrafish (Danio rerio) exposed to ethinylestradiol and methyltestosterone. Aquat. Toxicol. 65(4):397–411. Osius N, Karmaus W, Kruse H, Witten J (1999) Exposure to polychlorinated biphenyls and levels of thyroid hormones in children. Environ. Health Perspect. 107(10):843–849. Ottinger MA, Quinn J, Lavoie E, Abdelnabi MA, Thompson N, Hazelton JL, Wu JM, Beavers J, Jaber M (2005) Consequences of endocrine disrupting chemicals on reproductive endocrine function in birds: establishing reliable end points of exposure. Dom. Animal Endocrinol. 29(2):411–419. Ouyang F, Perry MJ, Venners SA, Chen C, Wang B, Yang F, Fang Z, Zang T, Wang L, Xu X, Wang X (2005) Serum DDT, age at menarche, and abnormal menstrual cycle length. Occup. Environ. Med. 62(12):878–884. Pakdel F, Feon S, Le Gac F, Le Menn F, Valotaire Y (1991) In vivo estrogen induction of hepatic estrogen receptor mRNA and correlation with vitellogenin mRNA in rainbow trout. Mol. Cell Endocrinol. 75(3):205–212. Panter GH, Thompson RS, Beresford N, Sumpter J (1999) Transformation of a non-oestrogenic steroid metabolite to an oestrogenically active substance by minimal bacterial activity. Chemosphere 38 (15):3579–3596. Panter G, Hutchinson T, Lange R, Lye C, Sumpter J, Zerulla M, Tyler C (2002) Utility of a juvenile fathead minnow screening assay for detecting (anti-)estrogenic substances. Environ. Toxicol. Chem. 21(2):319–326. Parks LG, Lambright CS, Orlando EF, Guillette LJ, Jr Ankley GT, Gray LE Jr (2001) Masculinization of female mosquitofish in Kraft mill effluent-contaminated Fenholloway River water is associated with androgen receptor agonist activity. Toxicol. Sci. 62(2):257–267. Peeters RP, Wouters PJ, van Toor H, Kaptein E, Visser TJ, van den Berghe G (2005) Serum 3,30 ,50 triiodothyronine (rT3) and 3,5,30 -triiodothyronine/rT3 are prognostic markers in critically ill patients and are associated with postmortem tissue deiodinase activities. J. Clin. Endocrinol. Metab. 90(8):4559–4565. Perrild H, Hegedus L, Baastrup PC, Kayser L, Kastberg S (1990) Thyroid function and ultrasonically determined thyroid size in patients receiving long-term lithium treatment. Am. J. Psychiatry 147 (11):1518–1521. Persky V, Turyk M, Anderson HA, Hanrahan LP, Falk C, Steenport DN, Chatterton R, JrFreels S (2001) The effects of PCB exposure and fish consumption on endogenous hormones. Environ. Health Perspect. 109(12):1275–1283. Peterson ME (2006) Diagnostic tests for hyperthyroidism in cats. Clin. Tech. Small Anim. Pract. 21 (1):2–9. Peterson ME, Melian C, Nichols R (1997) Measurement of serum total thyroxine, triiodothyronine, free thyroxine, and thyrotropin concentrations for diagnosis of hypothyroidism in dogs. J. Am. Vet. Med. Assoc. 211(11):1396–1402. Prosser DE, Jones G (2004) Enzymes involved in the activation and inactivation of vitamin D. Trends Biochem. Sci. 29(12):664–673. Puinean A, Rotchell J (2006) Vitellogenin gene expression as a biomarker of endocrine disruption in the invertebrate. Mytilus edulis. Mar. Environ. Res. 62:(Suppl 1):S211–S214. Quinn MJ, JrFrench JB, Jr McNabb FM, Ottinger MA (2002) The effects of polychlorinated biphenyls (Aroclor 1242) on thyroxine, estradiol, molt, and plumage characteristics in the American kestrel (Falco sparverius). Environ. Toxicol. Chem. 21(7):1417–1422. Reese S, Breyer U, Deeg C, Kraft W, Kaspers B (2005) Thyroid sonography as an effective tool to discriminate between euthyroid sick and hypothyroid dogs. J. Vet. Intern. Med. 19(4):491–498.

698

ENDOCRINE ACTIVE CHEMICALS: BROADENING THE SCOPE

Reis AF, Hauache OM, Velho G (2005) Vitamin D endocrine system and the genetic susceptibility to diabetes, obesity and vascular disease. A review of evidence. Diabetes Metab. 31(4 Pt 1):318–325. Reiter LW, DeRosa C, Kavlock RJ, Lucier G, Mac MJ, Melillo J, Melnick RL, Sinks T, Walton BT (1998) The U.S. federal framework for research on endocrine disruptors and an analysis of research programs supported during fiscal year 1996. Environ. Health Perspect. 106(3):105–113. Roegge CS, Schantz SL (2006) Motor function following developmental exposure to PCBS and/or MEHG. Neurotoxicol. Teratol. 28(2):260–277. Rosen JF, Chesney RW, Hamstra A, DeLuca HF, Mahaffey KR (1980) Reduction in 1,25-dihydroxyvitamin D in children with increased lead absorption. N. Engl. J. Med. 302(20):1128–1131. Ross G (2004) The public health implications of polychlorinated biphenyls (PCBs) in the environment. Ecotoxicol. Environ. Saf. 59(3):275–291. Rosser R (1976) Thyrotoxicosis and lithium. Br. J. Psychiatry 128:61–66. Rotchell JM, Ostrander GK (2003) Molecular markers of endocrine disruption in aquatic organisms. J. Toxicol. Environ. Health, Part B 6:453–495. Rothschild TC, Boylan ES, Calhoon RE, Vonderhaar BK (1987) Transplacental effects of diethylstilbestrol on mammary development and tumorigenesis in female ACI rats. Cancer Res. 47 (16):4508–4516. Rydh CJ, Svard B (2003) Impact on global metal flows arising from the use of portable rechargeable batteries. Sci. Total Environ. 302(1–3):167–184. Safe S (1993) Toxicology, structure–function relationship, and human and environmental health impacts of polychlorinated biphenyls: progress and problems. Environ. Health Perspect. 100:259– 268. Sapin R, Schlienger JL (2003) Thyroxine (T4) and tri-iodothyronine (T3) determinations: techniques and value in the assessment of thyroid function. Ann. Biol. Clin. (Paris) 61(4):411–420. Schulte PA, Perera FP, editors. (1993) Molecular Epidemiology: Principles and Practices.San Diego, CA: Academic Press. Schwaiger J, Mallow U, Ferling H, Knoerr S, Braunbeck T, Kalbfus W, Negele RD (2002) How estrogenic is nonylphenol? A transgenerational study using rainbow trout (Oncorhynchus mykiss) as a test organism. Aquat. Toxicol. 59(3–4):177–189. Schwartz BS, Stewart WF, Kelsey KT, Simon D, Park S, Links JM, Todd AC (2000) Associations of tibial lead levels with BsmI polymorphisms in the vitamin D receptor in former organolead manufacturing workers. Environ. Health Perspect. 108(3):199–203. Scinicariello F, Murray HE, Smith L, Wilbur S, Fowler BA (2005) Genetic factors that might lead to different responses in individuals exposed to perchlorate. Environ. Health Perspect. 113 (11):1479–1484. Segner H, Caroll K, Fenske M, Janssen CR, Maack G, Pascoe D, Schafers C, Vandenbergh GF, Watts M, Wenzel A (2003) Identification of endocrine-disrupting effects in aquatic vertebrates and invertebrates: report from the European IDEA project. Ecotoxicol. Environ Saf. 54(3):302–314. Seki M, Yokota H, Matsubara H, Tsuruda Y, Maeda M, Tadokoro H, Kobayashi K (2002) Effect of ethinylestradiol on the reproduction and induction of vitellogenin and testis–ova in medaka (Oryzias latipes). Environ. Toxicol. Chem. 21(8):1692–1698. Shi KL, Guo RR, Wang WY, Ma H, Yuan PF (2003) Association between vitamin D receptor gene polymorphism and the susceptibility of children to lead. Zhonghua Er. Ke. Za Zhi. 41(10):751– 754. Shugart LR, McCarthy JF, Halbrook RS (1992) Biological markers of environmental and ecological contamination: an overview. Risk Anal. 12(3):353–360. Silva IA, Graber J, Nyland JF, Silbergeld EK (2005) In vitro HgCl2 exposure of immune cells at different stages of maturation: effects on phenotype and function. Environ. Res. 98(3):341–348.

REFERENCES

699

Smith CM, DeLuca HF, Tanaka Y, Mahaffey KR (1981) Effect of lead ingestion on functions of vitamin D and its metabolites. J. Nutr. 111(8):1321–1329. Smits JE, Fernie KJ, Bortolotti GR, Marchant TA (2002) Thyroid hormone suppression and cellmediated immunomodulation in American kestrels (Falco sparverius) exposed to PCBs. Arch. Environ. Contam. Toxicol. 43(3):338–344. Soto AM, Silvia RM, Sonnenschein C (1992) A plasma-borne specific inhibitor of the proliferation of human estrogen-sensitive breast-tumor cells (estrocolyone-I). J. Steroid Biochem. Mol. Biol. 43 (7):703–712. Squire CR (2006) Methods for the investigation of thyroid function. Methods Mol. Biol. 324:91–108. Stoker TE, Robinette CL, Cooper RL (1999) Maternal exposure to atrazine during lactation suppresses suckling-induced prolactin release and results in prostatitis in the adult offspring. Toxicol. Sci. 52 (1):68–79. Stoker TE, Laws SC, Guidici DL, Cooper RL (2000) The effect of atrazine on puberty in male wistar rats: an evaluation in the protocol for the assessment of pubertal development and thyroid function. Toxicol. Sci. 58(1):50–59. Sumpter JP, Jobling S (1995) Vitellogenesis as a biomarker for estrogenic contamination of the aquatic environment. Environ. Health Perspect. 103(Suppl 7):173–178. Swales AK, Spears N (2005) Genomic imprinting and reproduction. Reproduction 130(4):389–399. Swan SH, Elkin EP, Fenster L (1997) Have sperm densities declined? A reanalysis of global trend data. Environ. Health Perspect. 105(11):1228–1232. Tabuchi M, Veldhoen N, Dangerfield N, Jeffries S, Helbing CC, Ross PS (2006) PCB-related alteration of thyroid hormones and thyroid hormone receptor gene expression in free-ranging harbor seals (Phoca vitulina). Environ. Health Perspect. 114(7):1024–1031. Tajtakova M, Semanova Z, Tomkova Z, Szokeova E, Majoros J, Radikova Z, Sebokova E, Klimes I, Langer P (2005) Increased thyroid volume and frequency of thyroid disorders signs in schoolchildren from nitrate polluted area. Chemosphere 62(4):559–564. Tatarazako N, Koshiro M, Hori H, Morita M, Iguchi T (2004) Validation of an enzyme-linked immunosorbent method for vitellogenin in medaka (Oryzias latipes). J. Health Sci. 50:301–308. Teles FF (2002) Chronic poisoning by hydrogen cyanide in cassava and its prevention in Africa and Latin America. Food Nutr. Bull. 23(4):407–412. Teng W, Shan Z, Teng X, Guan H, Li Y, Teng D, Jin Y, Yu X, Fan C, Chong W, Yang F, Dai H, Yu Y, Li J, Chen Y, Zhao D, Shi X, Hu F, Mao J, Gu X, Yang R, Tong Y, Wang W, Gao T, Li C (2006) Effect of iodine intake on thyroid diseases in China. N. Engl. J. Med. 354(26):2783–2793. Tennant RW (2002) The National Center for Toxicogenomics: using new technologies to inform mechanistic toxicology. Environ. Health Perspect. 110(1):A8–A10. Ternes TA, Stumpf M, Mueller J, Haberer K, Wilken RD, Servos M (1999) Behavior and occurrence of estrogens in municipal sewage treatment plants–I. Investigations in Germany, Canada and Brazil. Sci. Total Environ. 225(1–2):81–90. Theppeang K, Schwartz BS, Lee BK, Lustberg ME, Silbergeld EK, Kelsey KT, Parsons PJ, Todd AC (2004) Associations of patella lead with polymorphisms in the vitamin D receptor, deltaaminolevulinic acid dehydratase and endothelial nitric oxide synthase genes. J. Occup. Environ. Med. 46(6):528–537. Thorpe KL, Hutchinson TH, Hetheridge MJ, Sumpter JP, Tyler CR (2000) Development of an in vivo screening assay for estrogenic chemicals using juvenile rainbow trout (Oncorhynchus mykiss). Environ. Toxicol. Chem. 19(11):2812–2820. Thorpe KL, Benstead R, Hutchinson TH, Tyler CR (2006) Fecundity dynamics in pair-breeding fathead minnows (Pimephales promelas). Biol. Reprod. 114(S-1):106–114.

700

ENDOCRINE ACTIVE CHEMICALS: BROADENING THE SCOPE

Tietge JE, Holcombe GW, Flynn KM, Kosian PA, Korte JJ, Anderson LE, Wolf DC, Degitz SJ (2005) Metamorphic inhibition of Xenopus laevis by sodium perchlorate: effects on development and thyroid histology. Environ. Toxicol. Chem. 24(4):926–933. Touart LW (2004) Factors considered in using birds for evaluating endocrine-disrupting chemicals. Inst. Lab. Animal Res. 45(4):462–468. Turyk ME, Anderson HA, Freels S, Chatterton R, Jr, Needham LL, Patterson DG, Jr, Steenport DN, Knobeloch L, Imm P, Persky VW (2006) Associations of organochlorines with endogenous hormones in male Great Lakes fish consumers and nonconsumers. Environ. Res. 102(3):299–307. U.S. EPA (1998) EDSTAC Final Report. No. Uitterlinden AG, Fang Y, Van Meurs JB, Pols HA, Van Leeuwen JP (2004) Genetics and biology of vitamin D receptor polymorphisms. Gene 338(2):143–156. Uitterlinden AG, Ralston SH, Brandi ML, Carey AH, Grinberg D, Langdahl BL, Lips P, Lorenc R, Obermayer-Pietsch B, Reeve J, Reid DM, Amidei A, Bassiti A, Bustamante M, Husted LB, Diez-Perez A, Dobnig H, Dunning AM, Enjuanes A, Fahrleitner-Pammer A, Fang Y, Karczmarewicz E, Kruk M, Van Leeuwen JP, Mavilia C, Van Meurs JB, Mangion J, McGuigan FE, Pols HA, Renner W, Rivadeneira F, van Schoor NM, Scollen S, Sherlock RE, Ioannidis JP (2006) The association between common vitamin D receptor gene variations and osteoporosis: a participant-level meta-analysis. Ann. Intern. Med. 145(4):255–264. U.S. EPA (2002) A short-term test method for assessing the reproductive toxicity of endocrinedisrupting chemicals using the fathead minnow (Pimephales promelas). No. EPA/600/R-01/067. van Aerle R, Pounds N, Hutchinson TH, Maddix S, Tyler CR (2002) Window of sensitivity for the estrogenic effects of ethinylestradiol in early life-stages of fathead minnow, Pimephales promelas. Ecotoxicol. 11(6):423–434. Vetillard A, Atteke C, Saligaut C, Jego P, Bailhache T (2003) Differential regulation of tyrosine hydroxylase and estradiol receptor expression in the rainbow trout brain. Mol. Cell Endocrinol. 199:37–47. Vigiletti-Panzica C, Montoncello B, Mura E, Pessatti M, Panzica G (2005) Organizational effects of diethylstilbestrol on brain vasotocin and sexual behavior in male quail. Brain Res. Bull. 65:225–233. Volzke H, Alte D, Kohlmann T, Ludemann J, Nauck M, John U, Meng M (2005) Reference intervals of serum thyroid function tests in a previously iodine-deficient area. Thyroid 15(3):279–285. vom Saal FS, Timms BG, Montano MM, Palanza P, Thayer KA, Nagel SC, Dhar MD, Ganjam VK, Parmigiani S, Welshons WV (1997) Prostate enlargement in mice due to fetal exposure to low doses of estradiol or diethylstilbestrol and opposite effects at high doses. Proc. Natl. Acad. Sci. USA 94(5):2056–2061. Walker BE (1992) No effect of adult dietary fat on tumors induced prenatally by diethylstilbestrol. Nutr. Cancer 17(2):161–169. Waters MD, Fostel JM (2004) Toxicogenomics and systems toxicology: aims and prospects. Nat. Rev. Genet. 5(12):936–948. Weaver VM, Lee BK, Todd AC, Ahn KD, Shi W, Jaar BG, Kelsey KT, Lustberg ME, Silbergeld EK, Parsons PJ, Wen J, Schwartz BS (2006) Effect modification by delta-aminolevulinic acid dehydratase, vitamin D receptor, and nitric oxide synthase gene polymorphisms on associations between patella lead and renal function in lead workers. Environ. Res. 102(1):61–69. Weaver VM, Schwartz BS, Ahn KD, Stewart WF, Kelsey KT, Todd AC, Wen J, Simon DJ, Lustberg ME, Parsons PJ, Silbergeld EK, Lee BK (2003) Associations of renal function with polymorphisms in the delta-aminolevulinic acid dehydratase, vitamin D receptor, and nitric oxide synthase genes in Korean lead workers. Environ. Health Perspect. 111(13):1613–1619. Weber LP, Kiparissis Y, Hwang GP, Niimi AJ, Janz DM, Metcalfe CD (2002) Increased cellular apoptosis after chronic aqueous exposure to nonylphenol and quercetin in adult medaka (Oryzias latipes). Comp. Physiol. Biochem. C. 131:51–59.

REFERENCES

701

Whelan E, Grazewski B, Wild D, Schnorr T, Alderfer R (1996) Evaluation of reproductive function among men ocupationally exposed to a stilbene derivative: II. Percieved libido and potency. Am. J. Ind. Med. 29:59–65. WHO (1997) Recommended normative values for thyroid volume in children aged 6–15 years. World Health Organization & International Council for Control of Iodine Deficiency Disorders. Bull. World Health Organ. 75:95–97. WHO (2001) Biomarkers in Risk Assessment: Validity and Validation. Environmental Health Criteria. No. 222. Geneva:World Health Organization. Wiersinga WM, Podoba J, Srbecky M, van Vessem M, van Beeren HC, Platvoet-Ter Schiphorst MC (2001) A survey of iodine intake and thyroid volume in Dutch schoolchildren: reference values in an iodine-sufficient area and the effect of puberty. Eur. J. Endocrinol. 144(6):595–603. Wiesner W, Engel H, Steinbrich W, Oertli D (2006) Sonography of the thyroid. Schweiz. Rundsch. Med. Prax. 95(15):575–580. Willett WC (2002) Balancing life-style and genomics research for disease prevention. Science 296 (5568):695–698. Xu F, Sullivan K, Houston R, Zhao J, May W, Maberly G (1999) Thyroid volumes in US and Bangladeshi schoolchildren: comparison with European schoolchildren. Eur. J. Endocrinol. 140 (6):498–504. Yurino H, Ishikawa S, Sato T, Akadegawa K, Ito T, Ueha S, Inadera H, Matsushima K (2004) Endocrine disruptors (environmental estrogens) enhance autoantibody production by B1 cells. Toxicol. Sci. 81(1):139–147. Zaki A, Ait Chaoui A, Talibi A, Derouiche AF, Aboussaouira T, Zarrouck K, Chait A, Himmi T (2004) Impact of nitrate intake in drinking water on the thyroid gland activity in male rat. Toxicol. Lett. 147:27–33. Zhang F, Degitz SJ, Holcombe GW, Kosian PA, Tietge J, Veldhoen N, Helbing CC (2006) Evaluation of gene expression end points in the context of a Xenopus laevis metamorphosis-based bioassay to detect thyroid hormone disruptors. Aquat. Toxicol. 76(1):24–36. Zhong X, Xu Y, Liang Y, Liao T, Wang J (2005) The Chinese rare minnow (Gobiocypris rarus) as an in vivo model for endocrine disruption in freshwater teleosts: a full life-cycle test with diethylstilbestrol. Arch. Toxicol. 71(1):85–95. Zillioux EJ, Johnson IC, Kiparissis Y, Metcalfe CD, Wheat JV, Ward SG, Liu H (2001) The sheepshead minnow as an in vivo model for endocrine disruption in marine teleosts: a partial life-cycle test with 17alpha-ethynylestradiol. Environ. Toxicol. Chem. 20(9):1968–1978. Zoeller RT, Dowling AL, Vas AA (2000) Developmental exposure to polychlorinated biphenyls exerts thyroid hormone-like effects on the expression of RC3/neurogranin and myelin basic protein messenger ribonucleic acids in the developing rat brain. Endocrinol. 141(1):181–189.

19 SECONDHAND SMOKE Jonathan M. Samet, Gila I. Neta and Sophia S. Wang

Extensive toxicological, experimental, and epidemiological data, largely collected since the 1950s, have established that active cigarette smoking is the major preventable cause of morbidity and mortality in the United States (U.S. Department of Health and Human Services, 1989; U.S. Department of Health and Human Services, 2004). More recently, since the 1970s, involuntary exposure to tobacco smoke has been investigated as a risk factor for disease and also found to be a cause of preventable morbidity and mortality in nonsmokers. The 1986 report of the Surgeon General on smoking and health and a report by the National Research Council (NRC), also published in 1986, comprehensively reviewed the data on involuntary exposure to tobacco smoke and reached comparable conclusions with significant public health implications (National Research Council and Committee on Passive Smoking, 1986; U.S. Department of Health and Human Services, 1986); both reports concluded that involuntary smoking causes disease in nonsmokers. Subsequently, the Environmental Protection Agency (EPA) reached a similar conclusion in its 1992 risk assessment, which classified environmental tobacco smoke as a class A carcinogen (U.S. Environmental Protection Agency, 1992). These conclusions have already had significant impact on public policy and public health. Subsequently, a now substantial body of evidence (Table 19.1) has continued to identify new diseases and other adverse effects of secondhand smoke (SHS)(California Environmental Protection Agency and Air Resources Board, 2005; California Environmental Protection Agency and Office of Environmental Health Hazard Assessment, 1997; International Agency for Research on Cancer, 2004; Scientific Committee on Tobacco and Health and HSMO, 1998; World Health Organization, 1999). The 2006 U.S. Surgeon General’s report leaves no doubt that any exposure to tobacco smoke is harmful to human health (U.S. Department of Health and Human Services, 2006). The findings on secondhand smoke and disease have been the foundation of the drive for smoke-free indoor environments and for educating parents concerning the effects of their smoking on their children’s health.

Environmental Toxicants, Third Edition Edited by Morton Lippmann Copyright
2009 John Wiley & Sons, Inc.

703

704 Yes/c

Yes/c

Yes/c Yes/c Yes/c

Yes/c

Yes/c Yes/c

Yes/a

Yes/a

Yes/c

Yes/c

Yes/a

Yes/c

Yes/c

Yes/c

Yes/c

Yes/c

Yes/c

Yes/c

WHO 1999

Yes/c

IARC 2004

Yes/c

Yes/c

Yes/c

Yes/c Yes/c Yes/c

Yes/c

Yes/c

Yes/c

Yes/a

Yes/c

Cal/EPAa 2005

Yes/c

Yes/c Yes/c Yes/c

Yes/c

Yes/c

Yes/c

Yes/c

Yes/c

SGR 2006

a

Only effects causally associated with SHS exposure are included.

Yes/a: association; Yes/c: cause. Table adapted from U.S. Department of Health and Human Services (2006). SGR 1984: U.S. Department of Health and Human Services (1984); SGR 1986: U.S. Department of Health and Human Services (1986); EPA 1992: U.S. Environmental Protection Agency (1992); Cal/EPA 1997: California Environmental Protection Agency and Office of Environmental Health Hazard Assessment (1997; UK 1998: Scientific Committee on Tobacco and Health and HSMO (1998); WHO 1999: World Health Organization (1999); IARC 2004: International Agency for Research on Cancer (2004); Cal/EPA 2005: California Environmental Protection Agency and Air Resources Board (2005); SGR 2006: U.S. Department of Health and Human Services (2006).

Yes/c

Yes/a Yes/c

Yes/c

Yes/a

Yes/a

Yes/a

Yes/a

Yes/c

Yes/a

Yes/c

Yes/a

Yes/c

Yes/a

UK 1998

Yes/a

Cal/EPA 1997

Increased prevalence of chronic respiratory symptoms Decrement in pulmonary function Increased occurrence of acute respiratory illnesses Increased occurrence of middle ear disease Increased severity of asthma episodes and symptoms Risk factor for new asthma Risk factor for SIDS Risk factor for lung cancer in adults Risk factor for breast cancer for younger, primarily postmenopausal women Risk factor for nasal sinus cancer Risk factor for heart disease in adults

EPA 1992

SGR 1986

SGR 1984

Adverse Effects from Exposure to Tobacco Smoke

Health Effect

TABLE 19.1

EXPOSURE TO SECONDHAND SMOKE

705

This chapter provides an overview of the evidence on secondhand smoke and its impact on the health of children and adults. It covers the conclusions of the major recent reports that have systematically evaluated the evidence. The chapter describes the findings of some key representative studies, but it is not systematic in approach given the current scope of the literature. Complete reviews are provided by the 2005 Cal/EPA and 2006 Surgeon General’s reports (California Environmental Protection Agency and Air Resources Board, 2005; U.S. Department of Health and Human Services, 2006). We also note that there has been a long-standing campaign by the tobacco industry to discredit the evidence on secondhand smoke and health, in order to maintain an apparent controversy as a basis for slowing tobacco control. These tactics are well described through research based on the industry’s own documents, obtained as a result of litigation (Tobacco Documents Online, 2006). The influence of this campaign has extended to the peer-reviewed literature, including reports on methodologic issues, exposures, epidemiological studies, risk estimates, and control measures. 19.1 EXPOSURE TO SECONDHAND SMOKE 19.1.1

Characteristics of SHS

Nonsmokers inhale SHS, the combination of the sidestream smoke that is released from the burning end of the cigarette and the mainstream smoke exhaled by the active smoker (U.S. Department of Health and Human Services, 2006). This mixture has also been referred to as environmental tobacco smoke (ETS) or SHS. The inhalation of SHS is generally referred to as passive smoking or involuntary smoking. The exposures of involuntary and active smoking differ quantitatively and, to some extent, qualitatively (International Agency for Research on Cancer, 2004). Because of the lower temperature in the burning cone of the smoldering cigarette, most partial pyrolysis products are enriched in sidestream compared to mainstream smoke. Consequently, sidestream smoke has higher concentrations of some toxic and carcinogenic substances than mainstream smoke; however, dilution by room air markedly reduces the concentrations inhaled by the involuntary smoker in comparison to those inhaled by the active smoker. Nevertheless, involuntary smoking is accompanied by exposure to toxic agents generated by tobacco combustion (International Agency for Research on Cancer, 2004; U.S. Department of Health and Human Services, 1986). 19.1.2

Secondhand Smoke Concentrations

Tobacco smoke is a complex mixture of gases and particles that contains myriad chemical species (Guerin et al., 1992; International Agency for Research on Cancer, 2004; U.S. Department of Health and Human Services, 1984). Not surprisingly, tobacco smoking in indoor environments increases levels of respirable particles, nicotine, polycyclic aromatic hydrocarbons, carbon monoxide CO, acrolein, nitrogen dioxide (NO2), and many other substances. The extent of the increase in concentrations of tobacco smoke components varies with the number of smokers, the intensity of smoking, the rate of exchange between the indoor air space and the outdoor air, and the use of air-cleaning devices. Ott (1999) has used mass balance models to characterize factors influencing concentrations of tobacco smoke indoors. Using information on the source strength (i.e., the generation of emissions by cigarettes) and on the air exchange rate, researchers can apply mass balance models to

706

SECONDHAND SMOKE

predict tobacco smoke concentrations. Such models can be used to estimate exposures and to project the consequences of control measures. Several components of cigarette smoke have been measured in indoor environments as markers of the contribution of tobacco combustion to indoor air pollution. Particles have been measured most often because both sidestream and mainstream smoke contain high concentrations of particles in the respirable size range (International Agency for Research on Cancer, 2004; U.S. Department of Health and Human Services, 1986; 2006). Particles are a nonspecific marker of tobacco smoke contamination, however, because numerous sources other than tobacco combustion add particles to indoor air. Other more specific markers have also been measured, including nicotine, solanesol, and ultraviolet light (UV) absorption of particulate matter (Guerin et al., 1992). Nicotine, which is present in the gas phase in secondhand smoke, is usually measured with passive diffusion badges (Guerin et al., 1992; Leaderer and Hammond, 1991; U.S. Department of Health and Human Services, 2006). Nicotine has become the principal marker because of its specificity to cigarette smoke and the ease of measurement with the passive monitors. Many studies of levels of SHS components have been conducted in public buildings; fewer studies have been conducted in homes and offices (U.S. Department of Health and Human Services, 1986, 2006). The contribution of various environments to personal exposure to tobacco smoke varies with the time–activity pattern, namely, the distribution of time spent in different locations. Time–activity patterns may heavily influence lung airway exposures in particular environments for certain groups of individuals. For example, exposure in the home predominates for infants who do not attend day care (Harlos et al., 1987). For adults residing with nonsmokers, the workplace may be the principal location where exposure takes place. A nationwide study assessed exposures of nonsmokers in 16 metropolitan areas of the United States (Jenkins et al., 1996). This study, involving 100 persons in each location, was directed at workplace exposure and included measurements of respirable particulate matter and other markers. The results showed that in 1993 and 1994, exposures to SHS in the home were generally much greater than those in the workplace. The contribution of smoking in the home to indoor air pollution has been demonstrated by studies using personal monitoring and monitoring of homes for respirable particles. In one of the earliest studies, Spengler and Tosteson (1981) monitored homes in six U.S. cities for respirable particle concentrations over several years and found that a smoker of one pack of cigarettes daily contributed about 20 mg/m3 to 24 h indoor particle concentrations. In homes with two or more heavy smokers, this study showed that the pre-1987 24 h National Ambient Air Quality Standard (NAAQS) of 260 mg/m3 for total suspended particles could be exceeded. Because cigarettes are not smoked uniformly over the day, higher peak concentrations must occur when cigarettes are actually smoked. Spengler et al. (1985) measured the personal exposures to respirable particles sustained by nonsmoking adults in two rural Tennessee communities. The mean 24 h exposures were substantially higher for those exposed to smoke at home: 64 mg/m3 for those exposed versus 36 mg/m3 for those not exposed. Nicotine levels have now been measured in multiple homes in the United States, as shown in Fig. 19.1 (Emmons et al., 2001; Hammond et al., 1989; Henderson et al., 1989; Jenkins et al., 1996; Leaderer and Hammond, 1991; Marbury et al., 1993; U.S. Department of Health and Human Services, 2006). In homes with smokers, mean values range about 2–5 mg/m3 in the various studies, but maximum values in some homes are much higher. Additionally, these measures reflect the average concentration across the time of measurement, but not the values when nonsmokers are actually being exposed.

EXPOSURE TO SECONDHAND SMOKE

707

FIGURE 19.1 Concentration of nicotine in homes of U.S. smokers (U.S. Department of Health and Human Services, 2006).

The Total Exposure Assessment Methodology (TEAM) study, conducted by the U.S. Environmental Protection Agency, provided extensive data on concentrations of 20 volatile organic compounds in a sample of homes in several communities (Wallace and Pellizzari, 1987). Cigarette combustion is a strong source of many volatile organic compounds. Indoor monitoring showed increased concentrations of benzene, xylenes, ethylbenzene, and styrene in homes with smokers compared to homes without smokers. Figures 19.2 and 19.3 illustrate that more extensive information is now available on levels of SHS components in public buildings and workplaces of various types (U.S. Department of Health and Human Services, 2006). Monitoring in locations where smoking may be intense, such as bars and restaurants, has generally shown substantial elevations of particles and other markers of smoke pollution while smoking is taking place. For example, Repace and Lowrey (1980) in an early study used a portable piezobalance to sample aerosols in restaurants, bars, and other locations. In the places sampled, respirable particulate levels ranged up to 700 mg/m3, and the levels varied with the intensity of smoking (California Environmental Protection Agency and Office of Environmental Health Hazard Assessment, 1997; Guerin et al., 1992; National Research Council and Committee on Passive Smoking, 1986). Studies summarized in the 2006 report of the Surgeon General (U.S. Department of Health and Human Services, 2006) show the widespread presence of nicotine in workplaces and other locations with smoking allowed (Figs 19.2 and 19.3) and the potential for maximum concentrations to be extremely high. Monitoring studies document the effectiveness of workplace smoking policies for sharply reducing nicotine concentrations. Recent studies indicate low concentrations in many workplace settings, reflecting declining smoking prevalence in recent years and changing practices of smoking in the workplace. Using passive nicotine samplers, Hammond (Hammond, 1999) showed that worksite smoking policies can sharply reduce SHS exposure. Transportation environments may also be polluted by cigarette smoking. Contamination of air in trains, buses, automobiles, airplanes, and submarines has been documented (National Research Council and Committee on Passive Smoking, 1986; U.S. Department of Health and Human Services, 1989). A NRC report (National Research Council and

FIGURE 19.2 Occupational exposures to nicotine among groups of nonsmoking office workers (U.S. Department of Health and Human Services, 2006).

FIGURE 19.3 Average concentrations of nicotine in homes, offices, other workplaces, and restaurants where smoking is permitted (U.S. Department of Health and Human Services, 2006). 708

EXPOSURE TO SECONDHAND SMOKE

709

Committee on Airliner Cabin Environment Safety Committee, 1986) on air quality in airliners summarized studies for tobacco smoke pollutants in commercial aircraft. In one study, during a single flight, the NO2 concentration varied with the number of passengers with a lighted cigarette. In another study, respirable particles in the smoking section were measured at concentrations five or more times higher than in the nonsmoking section. Peaks as high as 1000 mg/m3 were measured in the smoking section. (Mattson et al., 1989) used personal exposure monitors to assess nicotine exposures of passengers and flight attendants. All persons were exposed to nicotine, even if seated in the nonsmoking portion of the cabin. These studies are now of historical interest only as almost all commercial flights worldwide are smoke free. Automobiles, however, are potential sites of high levels of exposure: preliminary data from a study in Greece (Vardavas et al., 2006) show that levels of concentrations can reach excessive heights when a smoker in a car exposes others to secondhand smoke. 19.1.3

Biological Markers of Exposure

Biological markers can be used to describe the prevalence of exposure to secondhand smoke, to investigate the dosimetry of involuntary smoking, and to validate questionnaire-based measures of exposure. In both active and involuntary smokers, the detection of tobacco smoke components or their metabolites in body fluids or alveolar air provides evidence of exposure to tobacco smoke, and levels of these markers can be used to gauge the intensity of exposure. The risk of involuntary smoking has also been estimated by comparing levels of biological markers in active and involuntary smokers. At present, the most sensitive and specific markers for tobacco smoke exposure are nicotine and its metabolite, cotinine (International Agency for Research on Cancer, 2004; Jarvis and Russell, 1984; U.S. Department of Health and Human Services, 1988, 2006). Neither nicotine nor cotinine is usually present in body fluids in the absence of exposure to tobacco smoke, although unusually large intakes of some foods could produce measurable levels of nicotine and cotinine (Idle, 1990). Cotinine, formed by oxidation of nicotine by cytochrome P450, is one of the several primary metabolites of nicotine (U.S. Department of Health and Human Services, 1988). Cotinine itself is extensively metabolized, and only about 17% of cotinine is excreted unchanged in the urine (International Agency for Research on Cancer, 2004). Because the circulating half-life of nicotine is generally shorter than 2 h (Rosenberg et al., 1980), nicotine concentrations in body fluids reflect more recent exposures. Nicotine can be measured in hair, as it is incorporated into the growing hair. By using several centimeters of hair, the level of nicotine reflects exposure over several weeks (Jaakkola and Jaakkola, 1997). In contrast to the short half-life of nicotine in the blood, cotinine has a half-life of about 10 h in the blood or plasma of active smokers (U.S. Department of Health and Human Services, 2006) and of about 20 h in nonsmokers (Kyerematen et al., 1982; U.S. Department of Health and Human Services, 2006), and hence, cotinine levels in blood, urine, or saliva provide information about exposure to tobacco smoke of involuntary smokers over periods of several days (Turner et al., 1987; Wall et al., 1988). Concerns about nonspecificity of cotinine, arising from eating nicotine-containing foods, have been set aside (Benowitz, 1996). Thiocyanate concentration in body fluids, concentration of CO in expired air, and carboxyhemoglobin level distinguish active smokers from nonsmokers but are not as sensitive and specific as cotinine for assessing involuntary exposure to tobacco smoke (Jarvis and Russell, 1984; U.S. Department of Health and Human Services, 2006).

710

SECONDHAND SMOKE

FIGURE 19.4 Serum cotinine geometric means (95% CI) for U.S. nonsmokers by study interval: exposure of nonsmokers in the U.S. population to SHS, 1988–2002. The data are plotted at the approximate midpoint for four separate time intervals: 1988–1991 (NHANES III, phase (1), 1991–1994 (NHANES III, phase (2), 1999–2000, and 2001–2002 (Pirkle et al., 2006).

Cotinine levels have been measured in adult nonsmokers and in children (U.S. Department of Health and Human Services, 2006). In the studies of adult nonsmokers, exposures at home, in the workplace, and in other settings determined cotinine concentrations in urine and saliva. The cotinine levels in involuntary smokers ranged from less than 1% to about 8% of cotinine levels measured in active smokers. Smoking by parents is the predominant determinant of the cotinine levels in their children. In 1988–1991, using liquid chromatography–mass spectrometry as the assay method, it was found that 88% of nonsmokers had a detectable level of serum cotinine (Pirkle et al., 1996, 2006; U.S. Department of Health and Human Services, 2006). Cotinine levels in this national sample increased with the number of smokers in the household and the hours exposed in the workplace. In subsequent phases of NHANES, the proportions of participants with a detectable level of cotinine and the mean level have dropped substantially (Fig. 19.4). The results of studies on biological markers have important implications for research on involuntary smoking and add to the biological plausibility of associations between involuntary smoking and diseases documented in epidemiological studies (Benowitz, 1996). The data on marker levels provide ample evidence that involuntary exposure leads to absorption, circulation, and excretion of tobacco smoke components. The studies of biological markers also confirm the high prevalence of involuntary smoking, as ascertained by questionnaire (Benowitz, 1996; Coultas et al., 1987; Pirkle et al., 1996). The observed correlations between reported exposures and levels of markers suggest that questionnaire methods for assessing recent exposure have some validity. 19.1.4

Exposure Assessment

The information on the health effects of involuntary smoking has been largely derived from observational epidemiological studies. In these studies, exposure to SHS has been estimated primarily by responses to questionnaires concerning the smoking habits of household members or fellow employees; attempts have been made to quantitate exposure by determining the number of cigarettes smoked by family members and the duration of

HEALTH EFFECTS OF INVOLUNTARY SMOKING IN CHILDREN

711

exposure. Biomarkers have also been used in some studies. Limitations of the questionnaire approach were discussed extensively in the 1986 report of the Surgeon General and again in the 2006 report (U.S. Department of Health and Human Services, 1986, 2006). The potential for information bias to introduce positive associations of SHS exposure with disease risk has been a focus of debate. A number of studies have addressed characteristics of questionnaires and biological markers for assessing exposure to SHS. Two studies evaluated the reliability of questionnaires on lifetime exposure (Coultas et al., 1989; Lubin, 1999; Pron et al., 1988). Both showed a high degree of repeatability for questions concerning whether a spouse had smoked but a lower reliability for responses concerning quantitative aspects of exposure. Several studies have assessed the validity of subjects’ reports on smoking by parents and spouses. Sandler and Shore (1986) compared responses on parents’ smoking given by cases and controls with responses given by the parents or siblings of the index subjects. Concordance was high for whether the parents had ever smoked. Responses concerning numbers of cigarettes smoked did not agree as highly. In a follow-up study of a nationwide sample, children’s responses on smoking by their deceased parents closely agreed with the information given 10 years previously by the parents (McLaughlin et al., 1987). A number of studies have shown that people correctly report the smoking habits of their spouses (U.S. Department of Health and Human Services, 1990b). In a study of nonsmokers in Buffalo, index subjects’ reports agreed well with reports from parents or siblings, spouse or children, and coworkers concerning exposure during childhood, at home, and at work, respectively (Cummings et al., 1989). Coghlin et al. (1989) used a passive nicotine monitor as well as a questionnaire and diary approach for characterizing exposure to SHS. In a sample of 19 volunteers, they found a strong correlation between the monitored nicotine exposure and a questionnaire-based index; the sampling lasted only a week, however, and the diary method would be too cumbersome to implement among all participants in a large epidemiological study. Although biological markers have provided important evidence of population exposures, the utility of cotinine as an indicator of individual exposure has been questioned. Idle (1990) has reviewed the complex metabolism of nicotine and the many factors affecting the relationship between exposure to atmospheric nicotine and the concentration of cotinine in body fluids. He cautions against using any single determination of cotinine as a measure of exposure. Several epidemiological studies support this concern about the limited validity of a single measurement of cotinine. Spot cotinine levels are not tightly predicted by questionnaire measures of exposures (Coultas et al., 1989; Cummings et al., 1990), and cotinine levels are highly variable at any particular level of smoking in a household (Coultas et al., 1990). Thus, questionnaires remain the best method for characterizing usual exposure to SHS. However, biological markers and personal monitoring offer complementary approaches for developing more accurate exposure estimates for quantifying dose and judging the extent of misclassification introduced by questionnaires. 19.2 HEALTH EFFECTS OF INVOLUNTARY SMOKING IN CHILDREN 19.2.1

Fetal Effects

Researchers have demonstrated that active smoking by mothers results in a variety of adverse health effects in children. Some of the health effects predominantly result from transplacental exposure of the fetus to tobacco smoke components. Recently, studies have also

712

SECONDHAND SMOKE

investigated and demonstrated associations between adverse health effects in children and exposure to SHS. For example, paternal smoking in the presence of a pregnant mother may lead to perinatal health effects manifested upon birth of the baby, and either maternal or paternal smoking in the presence of a newborn child may lead to postnatal health effects in the developing child. Health effects on the fetus resulting from SHS include fetal growth effects (decreased birth weight, growth retardation, or prematurity), fetal loss (spontaneous abortion and perinatal mortality), and congenital malformations. Health effects on the child postnatally, resulting from SHS exposure either to the fetus or to the newborn child, include sudden infant death syndrome (SIDS) and adverse effects on neuropsychological development and physical growth. Possible longer term health effects of fetal SHS exposure include childhood cancers of the brain, leukemia, and lymphomas, among others. 19.2.2

Biological Plausibility

This topic receives extensive coverage in the 2006 Surgeon General’s report (U.S. Department of Health and Human Services, 2006). Secondhand smoke could plausibly have adverse health effects at one or more steps of the developmental and reproductive processes, particularly during critical periods of susceptibility. Presumably, particular SHS components reach target sites and have toxic effects. For example, fetal exposure to CO and nicotine due to SHS may increase risk for perinatal health effects. CO in SHS may contribute to increased concentrations of CO and carboxyhemoglobin in the fetus, and the fetus may not be able to physiologically compensate for the reduced oxygen delivery (U.S. Department of Health and Human Services, 1980), leading to fetal hypoxia. With regard to SIDS, a number of mechanisms have been identified—arousal failure, inadequate cardiorespiratory compensatory motor responses, and sleep apnea—attributable to developmental abnormalities in the brainstem and autonomic nervous system (U.S. Department of Health and Human Services, 2006). Animal models indicate lasting effects on brain nicotine receptors (Slotkin, 2004). Animal models also indicate neural cellular effects from postnatal exposure that could underlie the link between paternal smoking and increased risk for SIDS (U.S. Department of Health and Human Services, 2006). Association between SHS and childhood cancers is biologically plausible due to the presence of carcinogenic tobacco smoke components or metabolites, such as benzene, nitrosamines, urethane, and radioactive compounds, at organ sites of the cancers. In animal studies, neurogenic tumors as well as other tumors were induced after transplacental exposure to a number of compounds present in tobacco smoke, including several nitrosamines. Smoking metabolites such as thiocyanate have also been found in fetal blood (Bottoms et al., 1982; Coghlin et al., 1991) and amniotic fluid (Andersen et al., 1982; Smith et al., 1982) of nonsmoking women exposed to tobacco smoke. Moreover, Huel et al. (1989) measured aryl hydrocarbon hydroxylase activity in human placenta of involuntary tobacco smokers; levels were increased in placentas of women passively exposed to tobacco smoke. 19.2.3

Nonfatal Perinatal Health Effects

19.2.3.1 Fetal Growth Most studies have used paternal smoking as the exposure measure to assess the association between SHS exposure and nonfatal perinatal health effects, such as reduced fetal growth. Low birth weight was first reported in 1957 to be

HEALTH EFFECTS OF INVOLUNTARY SMOKING IN CHILDREN

713

associated with maternal smoking during pregnancy (U.S. Department of Health and Human Services, 1980). On average, birth weight is reduced by about 250 g for newborns whose mothers smoked during pregnancy (U.S. Department of Health and Human Services, 2001; U.S. Department of Health and Human Services, 2004). Extensive studies have since been conducted to assess SHS exposure and birth weight, with consideration of gestational age at delivery, multiple births, maternal age, race, parity, maternal smoking, socioeconomic status, pregnancy history (U.S. Department of Health and Human Services, 2001, 2006). Exposures have been measured with questionnaires that assess home and work exposure, and in some studies, with the use of biomarkers. Studies generally find lower birth weight for infants of nonsmoking women passively exposed to tobacco smoke during pregnancy (Rubin et al., 1986; U.S. Department of Health and Human Services, 2006). For example, Haddow et al. (1988) used cotinine as a biomarker to measure exposure to SHS; they also adequately controlled for potential confounders. SHS exposure was defined as cotinine levels of 1.1–9.9 ng/mL in the fetus born to a nonsmoking mother. Their study demonstrated a decrease of 100 g in birth weight for fetuses exposed to SHS. Other biomarker studies (Eskenazi and Bergmann, 1995; Eskenazi and Trupin, 1995; Martinez et al., 1994) support the findings of Haddow et al. (1988). Other epidemiologic studies assessed SHS exposure from multiple sources through questionnaire (Mainous and Hueston, 1994; Rebagliato et al., 1995; Roquer et al., 1995). While not using a method as specific or sensitive as cotinine measurements, these studies still demonstrated decreases in mean birth weights after adjustment for confounders (20–40 g). In a 1999 meta-analysis, Windham et al. (1999) found a mean reduction of about 3 g for infants of nonsmoking mothers exposed to SHS during pregnancy. There was also a slightly increased risk for low birth weight. The 2006 report of the U.S. Surgeon General (U.S. Department of Health and Human Services, 2006) concluded that these effects of SHS were causal. 19.2.3.2 Other Effects Other nonfatal perinatal health effects possibly associated with SHS are growth retardation and congenital malformations. Martin and Bracken (1986) demonstrated a strong association with growth retardation in their 1986 study. Later studies (Mainous and Hueston, 1994; Roquer et al., 1995) supported this finding; however, these studies had small sample sizes and did not control for potential confounders. A few studies (Savitz et al., 1991; Seidman et al., 1990; Zhang et al., 1992) have been conducted to assess the association between paternal smoking and congenital malformations. The most consistent associations appear with the central nervous system or neural tube defects. However, due to possible effects of active smoke on the sperm, a causal association between SHS and congenital malformations cannot be concluded. 19.2.4

Fetal Perinatal Health Effects

SHS exposure to the fetus during its development may lead to fatal perinatal health effects such as spontaneous abortion and perinatal mortality. Very few studies have examined the association between SHS exposure and perinatal death. Eight studies have examined neonatal mortality in relation to paternal smoking, and a few supported an increase in risk (Ahlborg and Bodin, 1991; Comstock and Lundin, 1967; Lindbohm et al., 1991; Mau and Netter, 1974).

714

19.2.5

SECONDHAND SMOKE

Postnatal Health Effects

SHS exposure due to maternal or paternal smoking may lead to postnatal health effects related to SIDS, physical development, decrements in cognition and behavior, and cancers. 19.2.5.1 SIDS Sudden infant death syndrome (SIDS) is the sudden, unexplained death of an infant under 1 year of age. SIDS has been causally associated with maternal smoking during pregnancy (U.S. Department of Health and Human Services, 2001, 2004) and now with SHS exposure after birth (California Environmental Protection Agency and Air Resources Board, 2005; U.S. Department of Health and Human Services, 2006). The 2006 Surgeon General’s report cites 13 studies directed at the association between SIDS and postnatal SHS exposure. Ten studies have addressed the association between postpartum maternal smoking and SIDS, and nine studies considered paternal smoking. While maternal smoking during pregnancy has been causally associated with SIDS, these studies measured maternal smoking after pregnancy, along with paternal smoking and household smoking generally. Effects of SHS exposure after birth and maternal smoking during pregnancy cannot be readily separated in many of these studies, but paternal smoking involves SHS exposure with the potential to avoid the complicating consequences of smoking during pregnancy. Previously, maternal smoking during pregnancy had been causally linked to SIDS (U.S. Department of Health and Human Services, 2001), but separating the effects of prenatal exposure from maternal smoking during pregnancy and of postnatal exposure had been difficult. The evidence reviewed in the 2006 Surgeon General’s report (U.S. Department of Health and Human Services, 2006) showed that postnatal maternal smoking and paternal smoking were associated with increased risk of SIDS. Risks were increased by 50% to more than 100%, and several studies showed a dose–response relationship with level of SHS exposure. One study found increased risk for smoking by adults in the same room with the child (Klonoff-Cohen et al., 1995). Both the 2005 Cal/EPA and the 2006 Surgeon General’s reports (California Environmental Protection Agency and Air Resources Board, 2005; U.S. Department of Health and Human Services, 2006) concluded that a causal relationship exists between SHS exposure and SIDS. In reaching their conclusion, the reports noted not only the epidemiological evidence but also the findings of animal models that indicate potential mechanisms. The Cal/EPA report estimates that 10% of SIDS deaths are attributable to SHS exposure. 19.2.5.2 Cognition and Behavior While it is biologically plausible that SHS affects a child’s neuropsychological development, perhaps through nicotine’s effect on the central nervous system and through the effect of chronic exposure to CO, research on this potential consequence of SHS exposure needs to examine this relationship independent of prenatal exposure and maternal active smoking. Research on this topic also needs to carefully take into account potential confounding factors from the correlates of SHS exposure. Furthermore, cognition and behavior are also measured through a variety of tests, making direct comparisons between studies difficult. The 2006 Surgeon General’s report (U.S. Department of Health and Human Services, 2006) cited 12 epidemiologic studies that addressed SHS exposure and cognitive development in children. Of these, eight found associations between measures of involuntary smoking and children’s cognitive development. However, the evidence was heterogeneous and consistency could not be assessed. The evidence was judged to be inadequate to infer the presence or absence of a causal relationship.

HEALTH EFFECTS OF INVOLUNTARY SMOKING IN CHILDREN

19.2.6

715

Childhood Cancers

A causal association between involuntary smoking and childhood cancer derives biological plausibility from evidence of transplacental carcinogenesis in animal studies and in humans. Notably, cancers develop in offspring of rabbits and monkeys when the oncogenic compound ethylnitrosourea is administered to pregnant mothers (Rice et al., 1989; Stavrou et al., 1984), and in humans, transplacental carcinogenesis is well established between diethylstilbesterol exposure in pregnant mothers and development of vaginal clear cell adenocarcinoma among their daughters (Vessey, 1989). Evidence for transplacental carcinogenesis with involuntary smoking includes increased detection levels of smoking metabolites such as thiocyanate in fetal blood (Bottoms et al., 1982; Coghlin et al., 1991) and amniotic fluid (Andresen et al., 1982; Smith et al., 1982) compared to unexposed nonsmoking women. Moreover, genotoxic effects as measured by deletions within the housekeeping gene, hypoxanthine guanine phosphoribosyl transferase (HPRT), have been documented in cord blood of newborns from exposed mothers (Finette et al., 1998; International Agency for Research on Cancer, 1986). Most epidemiologic studies that have evaluated the association between involuntary smoking to the mother and childhood cancer among their offspring are case-control studies that have measured exposure based on the smoking habits of the father; few have included relevant exposures outside the home. Exposure assessment via the father, however, cannot distinguish the resulting cancers as being caused by involuntary smoking exposure in the mother or due to active smoking in the father that has resulted in DNA damage to the father’s sperm (U.S. Department of Health and Human Services, 2006). Further, distinguishing the effects of prenatal and postnatal secondhand smoke is difficult due to their high correlation. Finally, not all studies excluded mothers who were active smokers or accounted for other cancer risk factors such as maternal X-rays, sodium nitrite consumption, and drug use. It should be noted, however, that active maternal smoking during pregnancy has not been established as a causal risk factor for childhood cancer (Boffetta et al., 2000). Early evidence for an association between involuntary smoking in the mother and childhood cancer came from a 1982 report for childhood brain cancer (Preston-Martin et al., 1982). Results from studies published since then, particularly for all cancers, have provided mixed results. One meta-analysis by Sorahan et al. (1997b) reported a 1.2-fold increased risk for all cancers with paternal smoking, but recent reviews have not consistently supported this association (International Agency for Research on Cancer, 2004; Sasco and Vainio, 1999; Tredaniel et al., 1994). A cohort study also did not support an association between paternal smoking and childhood cancer (Seersholm et al., 1997). The evidence for specific cancers, notably leukemia, lymphoma, and central nervous system tumors, are more limited but some reveal more consistent patterns than the evidence for all cancers combined. 19.2.6.1 Brain Tumors Brain tumors are the most extensively studied childhood cancer in relation to involuntary maternal smoking. The association between involuntary maternal smoking and brain tumors is biologically plausible due to endogenously formed N-nitroso precursors found in SHS. Associations between paternal smoking and risk of brain tumors have been demonstrated in a number of studies (Preston-Martin et al., 1982; Filippini et al., 1994; McCredie et al., 1994; Sorahan et al., 1997a, 1997b) with statistically significant odds ratios ranging from 1.5 to 2.2 in some but not in others (Bunin et al., 1994; Gold et al., 1993; Howe et al., 1989; Kuijten et al., 1990; Norman et al., 1996a, 1996b).

716

SECONDHAND SMOKE

19.2.6.2 Leukemia The hypothesized pathway of transplacental carcinogenesis for leukemia resides in the leukemogen benzene, a component of secondhand smoke. Casecontrol studies evaluating the association between childhood leukemia and paternal smoking, however, have not been consistent. Most studies do not support an association (Brondum et al., 1999; Infante-Rivard et al., 2000; John et al., 1991; Magnani et al., 1990; Shu et al., 1996) but two studies (Ji et al., 1997; Sorahan et al., 2001), which distinguished between acute lymphocytic leukemia (ALL) and non-ALL, found increased risks for ALL at the highest levels of paternal smoking prior to conception. Although the study by Sorahan et al. (2001) included both smoking and nonsmoking mothers, results by Ji et al. (1997) were derived from nonsmokers. 19.2.6.3 Lymphomas Although it is suggestive, the evidence for involuntary smoking and childhood lymphoma remains sparse. Ji et al. (1997) present the strongest evidence to date for paternal smoking and childhood lymphoma among nonsmokers. Although Sorahan et al. (2001) and Magnani et al. (1990) also demonstrate increased risks, their studies do not delineate between smoking and nonsmoking mothers. Two studies also report no association between involuntary smoking and childhood lymphoma (Sorahan et al., 1995, 1997b). In summary, the data to date suggest an association between secondhand smoke exposure and childhood cancer and in particular for childhood brain tumors, leukemias, and lymphomas. However, the evidence is not yet sufficient to establish a causal relationship, and the effects of exposures during pregnancy and during infancy cannot be separated at the present time. 19.2.7

Lower Respiratory Tract Illnesses in Childhood

Studies of involuntary smoking, particularly maternal smoking, and lower respiratory illnesses in childhood, including bronchitis and pneumonia, provided some of the earliest evidence on adverse effects of SHS (Colley et al., 1974a, 1974b; Harlap and Davies, 1974). Presumably this association represents an increase in frequency or severity of illnesses that are infectious in etiology and not a direct response of the lung to toxic components of SHS. Investigations conducted throughout the world have demonstrated an increased risk of lower respiratory tract illness in infants with smoking parents (California Environmental Protection Agency and Air Resources Board, 2005; U.S. Department of Health and Human Services, 2006). These studies indicate a significantly increased frequency of bronchitis and pneumonia during the first year of life of children with smoking parents. The 2006 Surgeon General’s report (U.S. Department of Health and Human Services, 2006) includes a quantitative review of this information, combining data from over 50 studies. Overall, there is an approximate 50% increase in illness risk if either parent smoked, with the risk for maternal smoking being somewhat higher (Fig. 19.5). Although the health outcome measures have varied somewhat among the studies, the relative risks associated with involuntary smoking were similar, and dose–response relationships with extent of parental smoking were demonstrable. Although most of the studies have shown that maternal smoking, rather than paternal smoking, underlies the increased risk of lower respiratory tract illnesses, studies from China and elsewhere show that paternal smoking alone can increase incidence of lower respiratory illness (Strachan and Cook, 1997; U.S. Department of Health and Human Services, 2006). In these studies, an effect of passive smoking has not been readily identified after the first year of life. During the first year of life, the strength of its effect may reflect higher exposures consequent to the time–activity patterns of young infants, which place them in close proximity to cigarettes smoked by their mothers.

HEALTH EFFECTS OF INVOLUNTARY SMOKING IN CHILDREN

717

FIGURE 19.5 Odds ratios for the effect of smoking by either parent on lower respiratory illnesses during infancy (U.S. Department of Health and Human Services, 2006).

718

SECONDHAND SMOKE

19.2.8

Respiratory Symptoms and Illness in Children

The evidence on respiratory symptoms and illnesses in children and SHS exposure is now coming from numerous surveys and also from cohort studies. Data from surveys, largely of school children, demonstrate a greater frequency of the most common respiratory symptoms, cough, phlegm, and wheeze, in the children of smokers (California Environmental Protection Agency and Air Resources Board, 2005; California Environmental Protection Agency and Office of Environmental Health Hazard Assessment, 1997; Cook and Strachan, 1997a; U.S. Department of Health and Human Services, 1986, 2006) (Table 19.2). In these studies, the subjects have generally been school children, and the effects of parental smoking TABLE 19.2 Summary of Pooled Random Effects (Odds Ratios) Associated with Parental Smoking Restricted to Studies of Children Aged 11 Years (U.S. Department of Health and Human Services, 2006) Odds Ratio for Smoking (95% CI) Symptom Asthma

Numbers of Studies

Either Parent

13

1.18 (1.06–1.31)

One Parent

Both Parents

Mother Only

Father Only

Insufficient studies 5

1.47 (1.29–1.68)

7

1.31 (1.15–1.50)

4 Wheezea

15

1.13 (0.99–1.29) 1.27 (1.16–1.38)

4

1.21 (1.10–1.45)

5

1.41 (1.16–1.71)

8

1.26 (1.15–1.36)

5 Cough

13 4 5 4

1.10 (1.02–1.20) 1.28 (1.13–1.44) 1.17 (0.84–1.61) 1.85 (1.29–2.64) 1.07 (0.91–1.24)

3

1.12 (0.95–1.38)

Note: The symptoms “phlegm” and “breathlessness” were included in this table because of an insufficient number of studies. a

Excluded the European Communities Study, which had a pooled odds ratio of 1.20.

HEALTH EFFECTS OF INVOLUNTARY SMOKING IN CHILDREN

719

have been examined. Thus, the less prominent effects of passive smoking, in comparison to the studies of lower respiratory illness in infants, may reflect lower exposures to SHS by older children who spend less time with their parents. By the mid-1980s, results from several large studies provided convincing evidence that involuntary exposure to SHS increases the occurrence of cough and phlegm in the children of smokers, although earlier data from smaller studies had been ambiguous. For example, in a study of 10,000 school children in six U.S. communities, smoking by parents increased the frequency of persistent cough in their children by about 30% (Ware et al., 1984). The effect of parental smoking was derived primarily from smoking by the mother. Charlton (1984) conducted a survey on cigarette smoking that included 15,709 English children aged 8–19 years. In the nonsmoking children, the prevalence of frequent cough was significantly higher if either the father or the mother smoked. For the symptom of chronic wheeze, the preponderance of the early evidence also indicated an excess associated with involuntary smoking.Inasurveyof650schoolchildreninBoston,oneofthefirststudiesonthisassociation, persistent wheezing was the most frequent symptom (Weiss et al., 1980); the prevalence of persistent wheezing increased significantly as the number of smoking parents increased. In the SixCitiesStudyofchildren,theprevalenceofpersistentwheezingduringthepreviousyearwas significantly increased if the mother smoked (Ware et al., 1984). 19.2.9

Childhood Asthma

Although involuntary exposure to tobacco smoke has been associated with the symptom of wheeze, evidence for association of involuntary smoking with childhood asthma was initially conflicting. Exposure to SHS might cause asthma as a long-term consequence of the increased occurrence of lower respiratory infection in early childhood or through other pathophysiological mechanisms including inflammation of the respiratory epithelium (Tager et al., 1988; U.S. Department of Health and Human Services, 2006). The effect of SHS may also reflect, in part, the consequences of in utero exposure. Assessment of airways responsiveness shortly after birth has shown that infants whose mothers smoke during pregnancy have increased airways responsiveness compared with those whose mothers do not smoke (U.S. Department of Health and Human Services, 2006). Maternal smoking during pregnancy also reduced ventilatory function measured shortly after birth (Hanrahan et al., 1992). These observations suggest that in utero exposures from maternal smoking may affect lung development, perhaps reducing relative airways size. Additionally, childhood asthma is considered to have a strong genetic basis, and SHS exposure may act to increase or hasten incidence in a genetically predisposed subgroup of the population. While the underlying mechanisms remain to be fully characterized, the epidemiologic evidence linking SHS exposure and childhood asthma is substantial (California Environmental Protection Agency and Air Resources Board, 2005; Cook and Strachan, 1997a, 1997b; U.S. Department of Health and Human Services, 2006). There is evidence relevant to the causation of asthma and to the effect of SHS on the status of children with asthma. The 2006 report of the Surgeon General (U.S. Department of Health and Human Services, 2006) provides a full review of the evidence, considering the large number of cross-sectional studies and the smaller number of cohort studies. The cross-sectional studies cannot directly address SHS exposure as a cause of asthma onset because the existence of prevalent asthma reflects both incidence and maintenance of the asthmatic condition. Nonetheless, the prevalence studies provide firm evidence that prevalent asthma is associated with SHS exposure at home (Table 19.2). The 2006 report considered 41 cross-sectional studies with

720

SECONDHAND SMOKE

quantitative risk information. Overall, if either parent smoked, the pooled odds ratio was 1.23, compared to neither smoking. Household exposure to SHS was also associated with wheeze. The evidence was judged to be sufficient to infer a causal relationship between parental smoking and ever having asthma (U.S. Department of Health and Human Services, 2006). The report separately reviewed the seven cohort studies that addressed asthma incidence and also 21 case-control studies. Interpretation of the cohort study findings is complicated by the array of outcome measures and heterogeneity of effect by age of the children. The quantitative meta-analysis yielded a pooled odds ratio of 1.31, statistically significant, for children during the first 5–7 years of life; for the school years, the estimate was only 1.13. The case-control studies of prevalent asthma showed a 40% increase in association with smoking by either parent, a 50% increase for maternal smoking, but no increase for paternal smoking. Acknowledging the complexities of interpreting the cohort data, the 2006 report concluded that the evidence was suggestive, but not sufficient to infer a causal relationship between SHS exposure from parental smoking and onset of childhood asthma. The report also noted that SHS exposure can exacerbate childhood asthma (U.S. Department of Health and Human Services, 2006). 19.2.10

Lung Growth and Development

From gestation through adolescence, the lung goes through a complex process of maturation and growth that may be adversely affected by environmental agents, including SHS. On the basis of the primarily cross-sectional data available at the time, the 1984 report of the Surgeon General (U.S. Department of Health and Human Services, 1984) concluded that the children of smoking parents in comparison with those of nonsmokers had small reductions of lung function, but the long-term consequences of these changes were regarded as unknown. In the 2 years between the 1984 and the 1986 reports, sufficient longitudinal evidence was accumulated to support the conclusion in the 1986 report (U.S. Department of Health and Human Services, 1986) that involuntary smoking reduces the rate of lung function growth during childhood. Subsequently, substantial additional evidence has been obtained from many cross-sectional studies and some cohort studies. The evidence consistently shows that children exposed to SHS in their homes have reduced ventilatory function, as assessed by spirometry. Findings from cohort studies imply that the reduction comes from a reduced rate of lung growth. Meta-analyses of the cross-sectional data provide an indication of the magnitude of the effect of SHS exposure on lung function (Cook et al., 1998; U.S. Department of Health and Human Services, 2006). The 2006 Surgeon General’s report pooled 26 studies; the effect of SHS exposure was greatest for flow measures (the mid-expiratory and end-expiratory flow rates), approximately 4–5%, and less for the forced expiratory volume in 1 s (FEV1), approximately 1%. The effect of SHS exposure was greatest if both parents smoked and was robust to adjustment for potential confounding factors. The cohort studies show that lung function during childhood is adversely affected by maternal smoking during pregnancy and further impaired by exposure after birth. Studies of lung function shortly after birth show increased airways resistance and airways responsiveness for children exposed in utero (U.S. Department of Health and Human Services, 2006). These in utero effects appear to have implications for later lung growth and development. 19.2.11

SHS and Middle Ear Disease in Children

Otitis media (OM) is one of the most frequent diseases diagnosed in children at outpatient facilities. OM occurs as a result of dysfunction in the eustachian tube; serious OM results when

HEALTH EFFECTS OF INVOLUNTARY SMOKING IN CHILDREN

721

serous fluid effuses into the middle ear, and acute OM results when the serous fluid effused into the middle ear becomes infected. All stages of OM lead to varying degrees of hearing loss. There are four biologically plausible mechanisms by which SHS could lead to middle ear disease in children. First, SHS exposure could lead to decreased mucociliary clearance, increasing possible risk of dysfunction in the eustachian tube. Second, SHS may decrease eustachian tube patency due to adenoidal hyperplasia, a known risk factor for OM. Third, SHS may also decrease patency as a result of SHS-induced mucosal swelling. Fourth, SHS could decrease patency and mucociliary clearance by causing more frequent viral upper respiratory infections. The literature considered in the 2006 report of the Surgeon General included 61 reports based on 59 studies, covering multiple outcomes including acute OM, recurrent OM, middle ear disease, and adenotonsillectomy. The pooled evidence is shown in Fig. 19.6.

FIGURE 19.6 Odds ratios for the effect of smoking by either parent on middle-ear disease in children (U.S. Department of Health and Human Services, 2006).

722

SECONDHAND SMOKE

19.3 HEALTH EFFECTS OF INVOLUNTARY SMOKING IN ADULTS 19.3.1

Lung Cancer

In 1981, published reports from Japan (Hirayama, 1981) and Greece (Trichopoulos et al., 1981) indicated increased lung cancer risk in nonsmoking women married to cigarette smokers. Subsequently, this association has been examined in over 50 investigations conducted in the United States and other countries. A causal association of involuntary smoking with lung cancer derives biological plausibility from the presence of carcinogens in sidestream smoke and the lack of a documented threshold dose for respiratory carcinogens in active smokers (International Agency for Research on Cancer, 1986; U.S. Department of Health and Human Services, 1982, 1986, 2004). Moreover, genotoxic activity had been demonstrated for many components of SHS (Bennett et al., 1999; Claxton et al., 1989; DeMarini, 2004; Lofroth, 1989; Weiss, 1989). Experimental and real-world exposures of nonsmokers to SHS leads to their excreting NNAL, a tobacco-specific carcinogen, in their urine (Carmella et al., 2003; Hecht et al., 1993). Nonsmokers exposed to SHS also have increased concentrations of adducts of tobacco-related carcinogens (Crawford et al., 1994; Maclure et al., 1989). Additionally, Mauderly et al. (2004), using an animal model, found that whole-body exposure in rats to cigarette smoke increases the risk of neoplastic proliferative lung lesions and induces lung cancer. Time trends of lung cancer mortality in nonsmokers have been examined with the rationale that temporally increasing exposure to SHS should be paralleled by increasing mortality rates (Enstrom, 1979; Garfinkel, 1981). These data provide only indirect evidence on the lung cancer risk associated with involuntary exposure to tobacco smoke. Epidemiologists have directly tested the association between lung cancer and involuntary smoking utilizing conventional designs, the case-control and cohort studies. In a casecontrol study, the exposures of nonsmoking persons with lung cancer to SHS are compared to those of an appropriate control group. In a cohort study, the occurrence of lung cancer over time in nonsmokers is assessed in relation to involuntary tobacco smoke exposure. The results of both study designs may be affected by inaccurate assessment of exposure to SHS, by inaccurate information on personal smoking habits that leads to classification of smokers as nonsmokers, by failure to assess and control for potential confounding factors, and by the misdiagnosis of a cancer at another site as a primary cancer of the lung. As stated previously, methodological investigations suggest that accurate information can be obtained by interview in an epidemiological study on the smoking habits of a spouse (i.e., never or ever smoker) (Coultas et al., 1989; Cummings et al., 1989; Lubin, 1999; Pron et al., 1988). However, information concerning quantitative aspects of the spouse’s smoking is reported with less accuracy. Misclassification of current or former smokers as never smokers may introduce a positive bias because of the concordance of spouse smoking habits (Lee, 1998). The extent to which this bias explains the numerous reports of association between spouse smoking and lung cancer has been addressed, and findings indicate that bias does not account for the observed association (Lee, 1988; U.S. Department of Health and Human Services, 2006; U.S. Environmental Protection Agency, 1992; Wald et al., 1986; Wu, 1999). Use of spouse smoking alone to represent exposure to SHS does not cover exposures outside the home (Friedman et al., 1983) or necessarily all exposure inside the home, particularly during the period relevant to the epidemiological studies. Klepeis et al. (2001)

HEALTH EFFECTS OF INVOLUNTARY SMOKING IN ADULTS

723

used data from the National Human Activity Pattern Survey to assess the contribution of the home and other indoor environments to SHS exposures. Overall, the data show that 43% of the time spent with a smoker is in a residence, while 7% is in the workplace, 9% in a vehicle, and 15% in a bar or restaurant. This survey may help to explain the results of the International Agency for Research on Cancer (IARC), which found that the number of cigarettes smoked per day by the husband is only moderately correlated with “actual” exposure of women married to smokers (Saracci and Riboli, 1989). A subsequent IARC study published in 2004 conducted a pooled analysis of data to assess the risk of lung cancer to nonsmokers exposed to spouse and workplace sources of SHS. They found an excess risk of 23% from exposure to spousal smoking and 27% from exposure to workplace sources of SHS (Brennan et al., 2004). A study in the United States examined the contribution of spouse smoking to total exposure to SHS received at home (Sandler et al., 1989b). Using 1963 data from the Washington County (Maryland) study, Sandler et al. found that for nonsmoking women, spouse smoking contributed 88% of the exposure, whereas for nonsmoking men spouse smoking contributed 62% of the exposure. In some countries, including the United States, smoking prevalence varies markedly with indicators of income and education, more recently tending to rise sharply with decreasing educational level and income (U.S. Department of Health and Human Services, 1989, 2004). In general, exposure to SHS follows a similar trend, and critics of the findings on SHS and lung cancer have argued that uncontrolled confounding by lifestyle, occupation, or other factors may explain the association. In fact, current data for the United States do indicate a generally less healthy lifestyle in those with greater SHS exposure (Matanoski et al., 1995). However, other than a few occupational exposures at high levels, as well as indoor radon, risk factors for lung cancer in never smokers that might confound the SHS association cannot be proffered, and the relevance to past studies of these current associations of potential confounders with SHS exposure is uncertain. The first major studies on SHS and lung cancer were reported in 1981. An early report by Hirayama (1981) was based on a prospective cohort study of 91,540 nonsmoking women in Japan. Standardized mortality ratios (SMRs) for lung cancer increased significantly with the amount smoked by the husbands. The findings could not be explained by confounding factors and were unchanged when follow-up of the study group was extended (Hirayama, 1984). On the basis of the same cohort, Hirayama (1984) also reported significantly increased risk for nonsmoking men married to wives smoking 1–19 cigarettes and 20 or more cigarettes daily. In 1981, Trichopoulos et al. (1981) also reported increased lung cancer risk in nonsmoking women married to cigarette smokers. These investigators conducted a case-control study in Athens, Greece, which included cases with a diagnosis other than for orthopedic disorders. The positive findings reported in 1981 were unchanged with subsequent expansion of the study population (Trichopoulos et al., 1983). By 1986, the evidence had mounted, and three reports published in that year concluded that SHS was a cause of lung cancer. The IARC of the World Health Organization (International Agency for Research on Cancer, 1986) concluded that “passive smoking gives rise to some risk of cancer.” In its monograph on tobacco smoking, the agency supported this conclusion on the basis of the characteristics of sidestream and mainstream smoke, the absorption of tobacco smoke materials during involuntary smoking, and the nature of dose–response relationships for carcinogenesis. In the same year, the NRC (National Research Council and Committee on Passive Smoking, 1986) and the U.S.

724

SECONDHAND SMOKE

Surgeon General (U.S. Department of Health and Human Services, 1986) also concluded that involuntary smoking increases the incidence of lung cancer in nonsmokers. In reaching this conclusion, the NRC cited the biological plausibility of the association between exposure to SHS and lung cancer and the supporting epidemiological evidence. Based on a pooled analysis of the epidemiological data adjusted for bias, the report concluded that the best estimate for the excess risk of lung cancer in nonsmokers married to smokers was 25%. The 1986 report of the Surgeon General (U.S. Department of Health and Human Services, 1986) characterized involuntary smoking as a cause of lung cancer in nonsmokers. This conclusion was based on the extensive information already available on the carcinogenicity of active smoking, on the qualitative similarities between SHS and mainstream smoke, and on the epidemiological data on involuntary smoking. In 1992, the EPA (U.S. Environmental Protection Agency, 1992) published its risk assessment of SHS as a Group A carcinogen. The agency’s evaluation drew on the toxicologic evidence on SHS and the extensive literature on active smoking. A metaanalysis of the 31 studies published to that time was central in the decision to classify SHS as a Group A carcinogen—namely, a known human carcinogen. The meta-analysis considered the data from the epidemiologic studies by tiers of study quality and location and used an adjustment method for misclassification of smokers as never smokers. Overall, the analysis found a significantly increased risk of lung cancer in never-smoking women married to smoking men; for the studies conducted in the United States, the estimated relative risk was 1.19 (90% CI: 1.04, 1.35). Critics of the report have raised a number of concerns including the use of meta-analysis, reliance of 90% rather than 95% confidence intervals, uncontrolled confounding, and information bias. The report, however, was endorsed by the Agency’s Science Advisory Board, and its conclusion is fully consistent with the 1986 reports. Subsequent to the 1992 risk assessment, over a dozen additional studies and three major reports have been published that further contribute to the evidence supporting a causal association between secondhand smoke and the risk of lung cancer (California Environmental Protection Agency and Office of Environmental Health Hazard Assessment, 1997; Cardenas et al., 1997; Fontham et al., 1994; International Agency for Research on Cancer, 2004; Kabat et al., 1995; U.S. Department of Health and Human Services, 2006). Among the additional studies, the multicenter study of Fontham et al. (1994) is one of the largest to date, with 651 cases and 1253 controls. It shows a significant increase in overall relative risk (OR ¼ 1.26, 95% CI ¼ 1.04, 1.54). Significant risk was also associated with occupational exposure to SHS. Findings of an autopsy study conducted in Greece also strengthened the plausibility of the lung cancer/SHS association. Trichopoulos et al. (1992) examined autopsy lung specimens from 400 persons 35 years of age and older to assess airways changes. Epithelial lesions were more common in nonsmokers married to smokers than in nonsmokers married to nonsmokers. Hackshaw et al. (1997) carried out a comprehensive meta-analysis that included 37 published studies. They estimated an excess risk of lung cancer for smokers married to nonsmokers as 24% (95% CI: 13%, 36%). Adjustment for potential bias and confounding by diet did not alter the estimate. This meta-analysis was part of the basis for the conclusion by the U.K. Scientific Committee on Tobacco and Health (Scientific Committee on Tobacco and Health and HSMO, 1998) that SHS is a cause of lung cancer. A subsequent IARC (2004) meta-analysis including 46 studies and 6257 cases yielded similar results, 24% (95% CI: 14%, 34%), and incorporating the results from a cohort study with null results overall,

HEALTH EFFECTS OF INVOLUNTARY SMOKING IN ADULTS

725

but only 177 cases (Enstrom and Kabat, 2003), did not change the findings (Hackshaw, 2003). The most recent summaries from the 2006 Surgeon General’s report are provided in Table 19.3. The extent of the lung cancer hazard associated with involuntary smoking in the United States and in other countries remains subject to some uncertainty, however, although estimates have been made that are useful indications of the magnitude of the disease risk (U.S. Department of Health and Human Services, 1986; Weiss, 1986). Risk estimation procedures have been used to describe the lung cancer risk associated with involuntary smoking, but assumptions and simplifications are necessary to apply this method. The estimates of lung cancer deaths attributable to passive smoking have previously received widespread media attention and have figured prominently in the evolution of public policy on passive smoking. In 1990, Repace and Lowrey (1990) reviewed the risk assessments of lung cancer and passive smoking and estimated the numbers of lung cancer cases in U.S. nonsmokers attributable to passive smoking. The range of the nine estimates, covering both never smokers and former smokers, provided by Repace and Lowery was from 58 to 8124 lung cancer deaths for the year 1988, with an overall mean of 4500 or 5000 excluding the lowest estimate of 58. The bases for the individual estimates included the comparative dosimetry of tobacco smoke in smokers and nonsmokers using presumed inhaled dose or levels of nicotine or cotinine, the epidemiological evidence, and modeling approaches. The 1992 estimate of the EPA, based on the epidemiologic data was about 3000, including 1500 and 500 deaths in never-smoking women and men, respectively, and about 100 in long-term former smokers of both sexes (U.S. Environmental Protection Agency, 1992). More recently, Repace et al. (1998) developed a model of risk to workers of lung cancer and heart disease arising from SHS exposure. The pharmacokinetic model incorporated nicotine as an indicator of exposure and cotinine as a measure of dose to estimate risks. The model estimated that 400 lung cancer deaths occur annually from workplace exposure at a prevalence of 28% smoking in the workplace. The California EPA estimates that at least 3423, and perhaps as many as 8866, lung cancer deaths were caused by SHS across the nation in 2003 alone. Of those 3423 deaths, 967 were due to nonspousal exposures to secondhand smoke and 2456 were due to spousal exposure (California Environmental Protection Agency and Air Resources Board, 2005). These calculations illustrate that passive smoking must be considered an important cause of lung cancer death from a public health perspective; exposure is involuntary and not subject to control. The specific risk assessments require assumptions concerning the extent and degree of exposure to SHS, exposure–response relationships, and the lifetime expression of the excess risk associated with passive smoking at different ages. Moreover, the calculations did not consider the potential contributions of other exposures, such as occupational agents and indoor radon. The current decline in the prevalence of active smoking, and the implementation of strong clean indoor air policies, will reduce the relevance of estimates based on past patterns of smoking behavior. 19.3.2

Other Cancers

In adults, active smoking has been linked to a generally increased risk of malignancy and to excess risk at specific sites, including lung, urinary tract, upper aerodigestive tract, liver, stomach, pancreas, and others (U.S. Department of Health and Human Services, 2004; Vineis et al., 2004), reflecting the widespread effects of carcinogens in tobacco smoke. Given

726

SECONDHAND SMOKE

TABLE 19.3 Quantitative Estimate of Lung Cancer with Differing Sources of Exposure to Secondhand Smoke (U.S. Department of Health and Human Services, 2006) Study Hackshaw (1997) Zhong (2000)

Data Source 37 studies

Surgeon General report 2006 Spouse

40 studies (including 37 from Hackshaw study) Case control (448 studies) Cohort (8 studies)

54 studies

Men

Women

United States and Canada Europe Asia Surgeon General report 2006 Workplace 25 studies

Surgeon General report 2006 Childhood 24 studies

Nonsmokers (25 studies) Nonsmoking men (11 studies) Nonsmoking women (25 studies) Nonsmokers United States and Canada (8 studies) Nonsmokers Europe (7 studies) Nonsmokers Asia (10 studies) Men and women Men and women Men and women Women Women United States (8 studies) Europe (6 studies) Asia (10 studies)

Exposure Versus Referent Smoking versus nonsmoking spouse Smoking husband versus nonsmoking husband Smoking versus nonsmoking spouse Smoking versus nonsmoking spouse Smoking wife versus nonsmoking wife Smoking husband versus nonsmoking husband Smoking versus nonsmoking spouse Smoking versus nonsmoking spouse Smoking versus nonsmoking spouse Workplace SHS versus not Workplace SHS versus not Workplace SHS versus not Workplace SHS versus not Workplace SHS versus not Workplace SHS versus not Maternal smoking Paternal smoking Either parent smoking Maternal smoking Paternal smoking Either parent smoking Either parent smoking Either parent smoking

RR

95% CI

1.24

1.13–1.36

1.20

1.12–1.29

1.21

1.13–1.30

1.29

1.125–1.49

1.37

1.05–1.79

1.22

1.13–1.31

1.15

1.04–1.26

1.16

1.03–1.30

1.43

1.24–1.66

1.22

1.13–1.33

1.12

0.86–1.50

1.22

1.10–1.35

1.24

1.03–1.49

1.13

0.96–1.34

1.32

1.13–1.55

1.15

0.86–1.52

1.10 1.11

0.89–1.36 0.94–1.31

1.28 1.17 0.93

0.93–1.78 0.91–1.50 0.81–1.07

0.81

0.71–0.92

1.59

1.18–2.15

HEALTH EFFECTS OF INVOLUNTARY SMOKING IN ADULTS

727

that the same carcinogens are present in both mainstream and sidestream smoke and that these carcinogens have no evident threshold level in active smokers and have a demonstrated uptake by involuntary smokers, there is a compelling rationale for the hypothesis that secondhand smoke exposure increases the risk of cancers that are also caused by active smoking. While there is substantial and sufficient epidemiologic evidence for a causal association between SHS exposure and lung cancer, less data are available to analyze the causal association for other cancer sites. However, the data are mounting for some sites, and controversy has ensued, particularly for the association between SHS exposure and breast cancer. Two early studies assessed the association between SHS exposure and cancer in general (Miller, 1984; Sandler et al., 1985a, 1985b, 1985c). Miller (1984) interviewed surviving relatives of 537 deceased nonsmoking women in western Pennsylvania concerning the smoking habits of their husbands. A significantly increased risk of cancer death (OR ¼ 1.94; P < 0.05) was found in women who were married to smokers and also not employed outside the home. The large number of potential subjects who were not interviewed and the possibility of information bias detract from this report. Sandler et al. (1985a, 1985b, 1985c) conducted a case-control study on the effects of exposure to SHS during childhood and adulthood on the risk of cancer. The 518 cases included cancers of all types other than basal cell cancer of the skin; the cases and the matched controls were between the ages of 15 and 59 years. For all sites combined, significantly increased risk was found for parental smoking (crude OR ¼ 1.6), and for marriage to a smoking spouse (crude OR ¼ 1.5); the effects of these two exposures were independent (Sandler et al., 1985c). Significant associations were also found for some individual sites: for childhood exposure (Sandler et al., 1985b), maternal and paternal smoking increased the risk of hematopoietic malignancy, and for adulthood exposure (Sandler et al., 1985a), spouse’s smoking increased the risk for cancers of the female breast, female genital system, and the endocrine system. The findings are primarily hypothesis generating and require replication. In a case-control study, such as those reported by Sandler et al., information on exposure to SHS may be affected by information bias. One cancer site of particular interest at present is breast cancer. Given the widespread exposure to secondhand smoke, this exposure could potentially be an important avoidable cause of breast cancer. In considering whether passive smoking causes breast cancer, the evidence for active smoking needs to be considered in assessing the plausibility of an association of breast cancer risk with secondhand smoke in nonsmokers. There is some evidence to suggest that an association between tobacco smoke and breast cancer is biologically plausible. Studies have shown that carcinogens in tobacco smoke reach breast tissue (Li et al., 1996; Petrakis et al., 1978, 1988) and are mammary mutagens (Dunnick et al., 1995; El Bayoumy et al., 1995; Nagao et al., 1994). However, other studies using biomarkers have found an association between smoking and decreased levels of estrogen (MacMahon et al., 1982; Michnovicz et al., 1986), which implies that active smoking might decrease risk of breast cancer. Furthermore, the 2001 and 2004 reports of the Surgeon General found that smoking was associated with a decreased risk of endometrial cancer and an earlier age at menopause (U.S. Department of Health and Human Services, 2001, 2004). These antiestrogenic consequences of active smoking have been construed as implying that breast cancer risk would be reduced for active smokers in comparison to never smokers. The evidence is not consistent, however, and uncertainty remains about the effect of smoking on blood estrogen levels. These possibly opposing biological consequences of active smoking may explain why

728

SECONDHAND SMOKE

review of the epidemiologic data has found an overall null effect of active smoking on the risk of breast cancer. Since the 1960s, there have been more than 50 studies investigating the association between active smoking and breast cancer. In 2002, Hamajima et al. (2002) conducted a pooled analysis of data from 53 studies and found a relative risk of 0.99 (95% CI: 0.92, 1.05) for women who were current smokers compared with women who were lifetime nonsmokers. One possible explanation for the null results in Hamajima’s pooled analysis is that the antiestrogenic effects of smoking may offset the potentially carcinogenic effects on the risk of breast cancer. Subsequently, the 2004 reports of the Surgeon General and of IARC concluded that the weight of evidence strongly suggests that there is no causal association between active smoking and breast cancer (U.S. Department of Health and Human Services, 2004; International Agency for Research on Cancer, 2004). One year later, the California EPA concluded that active smoking is a cause of breast cancer, although it did not carry out a full systematic review (California Environmental Protection Agency and Air Resources Board, 2005). Two cohort studies published in 2004 found a significant increase in risk of breast cancer (Al Delaimy et al., 2004; Reynolds et al., 2004). However, sufficient evidence has not accumulated to suggest a causal association between active smoking and breast cancer (U.S. Department of Health and Human Services, 2006). More than 20 epidemiologic studies have been published specifically addressing the association between secondhand smoke and breast cancer. Several major reports, including the IARC report, the California EPA 2005 report, and the Surgeon General 2006 report, have reviewed the evidence for an association between SHS exposure and breast cancer (California Environmental Protection Agency and Air Resources Board, 2005; International Agency for Research on Cancer, 2004; U.S. Department of Health and Human Services, 2006). The Cal/EPA conducted a meta-analysis using six cohort studies and 12 case-control studies that were deemed to provide the “best evidence.” They found an increased risk of 25% (95% CI: 8%, 44%) overall and concluded that there is sufficient evidence for a causal association among premenopausal women (California Environmental Protection Agency and Air Resources Board, 2005). Among postmenopausal women, there was no indication of an association. In 2004, the IARC concluded that the evidence is inconsistent, and although some case-control studies found positive effects, cohort studies overall did not find a causal association (International Agency for Research on Cancer, 2004). Additionally, the lack of a positive dose–response relationship and the lack of association with active smoking weigh against the possibility of an increased risk of breast cancer from SHS exposure. Subsequently, the Surgeon General came to similar conclusions (U.S. Department of Health and Human Services, 2006). Using data from seven prospective cohort studies and 14 casecontrol studies, the Surgeon General’s report conducted a meta-analysis. Sensitivity analyses showed that cohort studies overall found null results and studies that adjusted for potential confounding showed weaker associations (U.S. Department of Health and Human Services, 2006). Furthermore, the Surgeon General’s report evaluated the possibility of publication bias and found that less precise studies tended to have more positive results. Finally, after reviewing all the evidence using the criteria for causality, the Surgeon General’s report found that overall the evidence is inconsistent and concluded that the evidence is suggestive but not sufficient to infer a causal association between SHS exposure and breast cancer. Other studies provide data on passive smoking and cancers of diverse sites. Hirayama (1984) has reported significantly increased mortality from nasal sinus cancers and from brain tumors in nonsmoking women married to smokers in the Japanese cohort. Additionally, two

HEALTH EFFECTS OF INVOLUNTARY SMOKING IN ADULTS

729

case-control studies, one in Japan and one in the United States, found an association between nasal sinus cancer and SHS exposure in nonsmokers living with a smoking spouse (Fukuda and Shibata, 1990; Zheng et al., 1993). However, these case-control studies relied on relatively small sample sizes, using fewer than 170 cases. Nasopharyngeal carcinoma has also been investigated as a potential cancer caused by SHS exposure in three published case-control studies (Cheng et al., 1999; Yu et al., 1990; Yuan et al., 2000). Two of the three studies found null effects among nonsmokers (Cheng et al., 1999; Yu et al., 1990), and the third found an association among women, but nonsignificant effects in men (Yuan et al., 2000). The 2006 Surgeon General’s report concluded that “the evidence is inadequate to infer the presence or absence of a causal relationship” for nasopharyngeal carcinoma (U.S. Department of Health and Human Services, 2006). Cervical cancer, which has been linked to active smoking (U.S. Department of Health and Human Services, 1990b), was associated with duration of involuntary smoking in a casecontrol study in Utah (Slattery et al., 1989) and with cumulative exposure in a case-control study by Sandler et al. (1985c), looking at only 62 cases of cervical cancer. In 2003, Wu et al. found an increased risk of cervical intraepithelial neoplasia among nonsmoking women exposed to SHS exposure at home, suggesting a possible role for secondhand smoke in the etiology of cervical cancer (Wu et al., 2003). Two other case-control studies found borderline statistical significance for an association between SHS exposure and cervical cancer (Coker et al., 1992) and for an association between SHS exposure and abnormal Pap smear results (Scholes et al., 1999). Among the two published cohort studies investigating an association between cervical cancer and SHS exposure in nonsmoking wives married to smokers, Hirayama et al. (Hirayama, 1981) found a nonsignificant increased risk in a Japanese cohort and Jee et al. (2002) found no association in a Korean cohort. Based on a review of the evidence, the 2006 Surgeon General’s report concluded that a causal association between SHS exposure and cervical cancer cannot yet be inferred (U.S. Department of Health and Human Services, 2006). In a case-control study of bladder cancer, involuntary smoke exposure at home and at work did not increase risk (Kabat et al., 1986). Another case-control study by Burch et al. also found null effects for the association between SHS exposure and bladder cancer (Burch et al., 1989). In 2002, Zeegers et al. found a slight but nonsignificant increase in risk of bladder cancer in nonsmokers exposed in the workplace (Zeegers et al., 2002). Although the association between bladder cancer and active smoking has been established (International Agency for Research on Cancer, 1986), there is insufficient evidence to conclude that SHS exposure is causally linked to bladder cancer. In the Washington County (Maryland) study, colorectal cancer incidence rates were significantly increased for male secondhand smokers, but not for female secondhand smokers; incidence rates were significantly reduced for female active smokers (Sandler et al., 1988). This pattern of findings cannot be readily explained. These associations of involuntary smoking with cancer at diverse nonrespiratory sites cannot be readily supported with arguments for biological plausibility. Increased risks at some of the sites, such as female breast cancer, have not been observed in active smokers (International Agency for Research on Cancer, 2004; U.S. Department of Health and Human Services, 1989, 1990a, 2004, 2006). In fact, the IARC has concluded that effects would not be produced in passive smokers that would not be produced to a larger extent in active smokers (International Agency for Research on Cancer, 1986, 2004). Thus, investigation of cancer sites other than the lung should be guided by the data from active smokers and by appropriate toxicological evidence. For example, the plausibility of secondhand smoking

730

SECONDHAND SMOKE

with cervical cancer would be supported by the demonstration of tobacco smoke components in the cervical mucus of exposed nonsmoking women. In investigations of cancer at sites not plausibly linked to secondhand smoke exposure, associations may arise by chance or by the effect of bias. Amassing data on secondhand smoke and all cancers may thus produce noncausal associations, prompting further but possibly unnecessary investigations.

19.4 SHS AND CORONARY HEART DISEASE 19.4.1

Introduction

Causal associations between active smoking and fatal and nonfatal coronary heart disease (CHD) outcomes have long been demonstrated (U.S. Department of Health and Human Services, 2004). This increased risk of CHD morbidity and mortality has been found in younger persons and the elderly, men and women, and ethnically and racially diverse populations. The risk of CHD in active smokers increases with amount and duration of cigarette smoking and decreases during the first year after cessation. Active cigarette smoking is considered to increase the risk of cardiovascular disease by promoting atherosclerosis, affecting endothelial cell functioning, increasing the tendency to thrombosis, causing spasm of the coronary arteries that increases the likelihood of cardiac arrhythmias, and decreasing the oxygen-carrying capacity of the blood (U.S. Department of Health and Human Services, 1990b). It is biologically plausible that exposure to secondhand smoke could also be associated with increased risk for CHD through the same mechanisms considered relevant for active smoking, although the lower exposures to smoke components of the secondhand smoker have raised questions regarding the relevance of the mechanisms cited for active smoking. 19.4.2

Biological Plausibility

In 2005, Barnoya and Glantz (2005) summarized the pathophysiological mechanisms by which secondhand smoke exposure might increase the risk of heart disease. They suggest that passive smoking may promote atherogenesis, increase the tendency of platelets to aggregate and thereby promote thrombosis, impair endothelial cell function, increase arterial stiffness leading to atherosclerosis, reduce the oxygen-carrying capacity of the blood, and alter myocardial metabolism, much as for active smoking and CHD. Several separate experiments involving exposure of nonsmokers to SHS have shown that passive smoking affects measures of platelet function in the direction of increased tendency toward thrombosis (Barnoya and Glantz, 2005; Glantz and Parmley, 1995). In a 2004 study by Rubenstein et al., sidestream smoke was found to be 50% more potent than mainstream smoke in activating platelets (Rubenstein et al., 2004). Glantz and Parmley also proposed that carcinogenic agents such as polycyclic aromatic hydrocarbons found in tobacco smoke promote atherogenesis by effects on cell proliferation (Glantz and Parmley, 1995). Exposure to secondhand smoke may also worsen the outcome of an ischemic event in the heart: animal data have demonstrated that SHS exposure increases cardiac damage following an experimental myocardial infarction. Experiments on two species of animals (rabbits and cockerels) have demonstrated that not only does exposure to SHS at doses similar to exposure to humans accelerate the growth of atherosclerotic plaques through the increase of lipid deposits, but it also induces atherosclerosis. There is also impressive and accumulating evidence that SHS

SHS AND CORONARY HEART DISEASE

731

affects vascular endothelial cell functioning (Celermajer et al., 1996; Otsuka et al., 2001; Sumida et al., 1998). Otsuka et al. found that 30 min of exposure to secondhand smoke in healthy young volunteers compromised coronary artery endothelial function in a manner that was indistinguishable from that of habitual smokers, suggesting that endothelial dysfunction may be an important mechanism by which exposure to secondhand smoke increases CHD risk (Otsuka et al., 2001). In addition to its effects on platelets, SHS exposure affects the oxygen-carrying capacity of the blood. Even small increments, on the order of 1%, in the carboxyhemoglobin, may explain finding that secondhand smoke exposure decreases the duration of exercise of patients with angina pectoris (Allred et al., 1989). This is supported with evidence that cigarette smoking has been shown to increase levels of CO in the spaces where ventilation is low or smoking is particularly intense (U.S. Department of Health and Human Services, 1986). 19.4.3

Epidemiological Studies

Epidemiologic data first raised concern that exposure to secondhand smoke may increase risk for CHD with the 1985 report of Garland et al. (1985) based on a cohort study in southern California. There are now more than 20 studies on the association between secondhand smoke and cardiovascular disease, including 11 cohort and 12 case-control studies, and 1 cross-sectional study. These studies assessed both fatal and nonfatal cardiovascular heart disease outcomes, and most used self-administered questionnaires to assess SHS exposure. They cover a wide range of populations, both geographically and racially. While many of the studies were conducted within the United States, studies were also conducted in Europe (Scotland, Italy, United Kingdom, Sweden), Asia (Japan and China), South America (Argentina), and the South Pacific (Australia and New Zealand). The majority of the studies measured the effect of SHS exposure due to spousal smoking; however, some studies also assessed exposures from smoking by other household members or occurring at work or in transit. Several studies included measurement of biomarkers. One group of studies addresses the promotion of atherosclerosis and SHS exposure, particularly through the mechanism of increased carotid intimal medial thickness (IMT). Three studies that assessed the link between SHS and carotid IMT found an increase in carotid IMT with exposure to SHS of nonsmokers (Diez-Roux et al., 1995; Howard et al., 1994). Using data from the ARIC Study, Howard et al. found an average difference in carotid IMT of 13 mm between exposed and unexposed nonsmokers after adjusting for possible confounders (Howard et al., 1994). They also found a dose–response relationship between exposure to SHS and increased carotid IMT in male nonsmokers. Diez-Roux and colleagues also found an increase in carotid IMTwith exposure to SHS, analyzing data from 2073 ARIC participants who were lifetime nonsmokers. The participants were assessed on two separate occasions in 1975 and 1987–1989. Regardless of whether they were exposed to SHS on the first, second, or both occasions, there was a significant increase in carotid IMT at final assessment (Diez-Roux et al., 1995). These findings suggest that SHS has a long-term effect on carotid IMT and provide further evidence for the association between SHS and atherosclerosis. Subsequently, in 1998 Howard et al. examined the effect of SHS exposure on progression of IMT. They found that nonsmokers exposed to SHS had a 20% increased IMT progression rate compared to those unexposed (Howard et al., 1998). As the evidence has subsequently mounted since the Garland et al. report, it has been reviewed systematically by the American Heart Association (Taylor et al., 1992), the

732

SECONDHAND SMOKE

California EPA (California Environmental Protection Agency and Air Resources Board, 2005; California Environmental Protection Agency and Office of Environmental Health Hazard Assessment, 1997), the Scientific Committee on Tobacco and Health in the United Kingdom (Scientific Committee on Tobacco and Health and HSMO, 1998) and most recently by the U.S. Surgeon General (U.S. Department of Health and Human Services, 2006). Review of the evidence has uniformly led to the conclusion that there is a causal association between exposure to secondhand smoke and risk of cardiovascular disease (California Environmental Protection Agency and Office of Environmental Health Hazard Assessment, 1997; Scientific Committee on Tobacco and Health and HSMO, 1998). The meta-analysis prepared for the 2006 U.S. Surgeon General’s Report, including nine cohort studies and seven case-control studies, estimated the pooled excess risk from SHS exposures as 27% (95% CI: 19–36%) (U.S. Department of Health and Human Services, 2006). The forest plot in Fig. 19.7 summarizes the evidence reviewed in the 2006 Surgeon

FIGURE 19.7 Pooled relative risks of coronary heart disease (CHD) associated with secondhand smoke exposure among nonsmokers in various subgroups (U.S. Department of Health and Human Services, 2006).

SHS AND CORONARY HEART DISEASE

733

General’s report (U.S. Department of Health and Human Services, 2006). Among the cohort studies included in the meta-analysis, seven were conducted in the United States, one in Japan, and one in Scotland. Among the case-control studies, one was conducted in the United States and the others were conducted abroad. Nine of the studies, both cohort and casecontrol, included men and women subjects, while six included only women. Only one study was restricted to men. In all studies, participants were nonsmokers and except for three of the studies, they were lifetime nonsmokers. For those three studies, either the study did not explicitly indicate that participants had formerly smoked or it included former smokers (Butler, 1988; Hirayama, 1984; McElduff et al., 1998). In all but one study, investigators assessed exposure to secondhand smoke in the home from either a spouse or cohabitant through self-report. Four of the studies also assessed exposure to secondhand smoke outside of the home, including workplace exposures. Two of the studies did not specify or distinguish among the different sources of secondhand smoke exposure. Among the cohort studies reporting on the effect of exposure to secondhand smoke on cardiovascular outcomes, five of the cohort studies used fatal CHD as the primary outcome, three used ischemic heart disease (IHD), and one combined fatal CHD and nonfatal acute myocardial infarction. Among the case-control studies, four used nonfatal acute myocardial infarction as the primary outcome, one used nonfatal CHD, one used fatal and nonfatal acute myocardial infarction, and one used nonfatal IHD. While the risk estimates for SHS and CHD outcomes vary in these studies, they range mostly from null to modestly significant increases in risk, with the risk for fatal outcomes generally higher and more significant. Additionally, a prospective cohort study reported in 2004 used serum cotinine levels for exposure classification (Whincup et al., 2004). The study included 4729 men in the United Kingdom who provided baseline blood samples in 1978–1980. After 20 years of follow-up, among the 2105 men who were nonsmokers, the risk of CHD was increased in those with higher serum cotinine concentrations. Compared to men in the lowest quartile of serum cotinine concentration, after adjusting for established CHD risk factors, the risks in the second, third, and fourth quartiles were 1.45, 1.49, and 1.57, respectively. A consistent association was not found between serum cotinine concentration and stroke. However, there is increasing epidemiologic evidence suggestive of a causal association between SHS exposure and stroke. At least seven epidemiologic studies (four case-control, one cohort, and two cross-sectional studies) have been published exploring the association between SHS exposure and stroke (Bonita et al., 1999; Donnan et al., 1989; Howard et al., 1998; Lee et al., 1986; Sandler et al., 1989a; You et al., 1999; Zhang et al., 2005). A large cross-sectional study of 60,377 women in China found an association between prevalent stroke in women and smoking by their husbands (Zhang et al., 2005). The prevalence of stroke increased with greater duration of smoking and with an increasing number of cigarettes smoked daily. Sandler et al. conducted a cohort study of 19,035 lifetime nonsmokers using census data from Washington County, MD. Based on 297 cases among women exposed to SHS, they found a 24% increased risk of stroke compared with unexposed. They found null results for an association in men but were limited to only 33 cases among men (Sandler et al., 1989a). A case-control study in New Zealand did find a twofold increased risk of stroke in men exposed to SHS, looking at 215 cases and 1336 controls (Bonita et al., 1999). 19.4.4

Conclusions

There are strengths and weaknesses to both the case-control and cohort study designs in investigating SHS and CHD outcomes. Many of the case-control studies suffer from

734

SECONDHAND SMOKE

small sample sizes and lack the power to detect significant associations. Furthermore, many studies also lack information on other risk factors for CHD, and therefore they do not adequately adjust for confounders. In contrast, many of the cohort studies have large sample sizes and do adjust for confounders. They also avoid information bias by assessing smoking status and exposure prior to the CHD outcome. However, cohort studies are more susceptible to exposure misclassification due to the cessation or resumption of smoking by the source of exposure; this risk of misclassification increases with the length of follow-up. Although the risk estimates for SHS and CHD outcomes vary, they range mostly from null to modestly significant increases in risk, with the risk for fatal outcomes generally higher and more significant. In their meta-analysis, Law et al. (1997) estimated the excess risk from SHS exposure as 30% (95% CI: 22–38%) at age 65 years. In 1997, the California Environmental Protection Agency (California Environmental Protection Agency and Office of Environmental Health Hazard Assessment, 1997) concluded that there is “an overall risk of 30%” for CHD due to exposure from SHS (California Environmental Protection Agency and Office of Environmental Health Hazard Assessment, 1997). In 2005, Cal/EPA established that 22,700–69,500 deaths from CHD were attributable to SHS in 2000 (California Environmental Protection Agency and Air Resources Board, 2005). The American Heart Association’s Council on Cardiopulmonary and Critical Care has also concluded that secondhand smoke both increases the risk of heart disease and is “a major preventable cause of cardiovascular disease and death” (Taylor et al., 1992). This conclusion was echoed in 1998 by the Scientific Committee on Tobacco and Health, both Cal/EPA reports, and the 2006 report of the Surgeon General (California Environmental Protection Agency and Office of Environmental Health Hazard Assessment, 1997; Scientific Committee on Tobacco and Health and HSMO, 1998; California Environmental Protection Agency and Air Resources Board, 2005; U.S. Department of Health and Human Services, 2006).

19.5 RESPIRATORY SYMPTOMS AND ILLNESSES IN ADULTS Extensive evidence has shown that active smoking causes respiratory symptoms and illnesses (U.S. Department of Health and Human Services, 2004). Active smoking can cause inflammatory injury throughout the respiratory tract, leading to both acute and chronic respiratory symptoms, impaired lung function, and eventually to chronic obstructive pulmonary disease (COPD). The similarity of mainstream and SHS implies that involuntary smoking might also cause inflammation of the respiratory tract. One autopsy study of the lungs of nonsmokers found inflammation associated with SHS exposure (Trichopoulos et al., 1992). The 2006 report of the Surgeon General concluded that there are multiple mechanisms by which SHS exposure may cause respiratory symptoms and illnesses (U.S. Department of Health and Human Services, 2006). Observational studies provide evidence that SHS exposure can cause respiratory effects in adults. In 1986, both the NRC and the Surgeon General’s reports not only concluded that SHS exposure is an irritant (National Research Council and Committee on Passive Smoking, 1986; U.S. Department of Health and Human Services, 1986), but also stated that additional evidence was needed to determine if there is a causal association of exposure with chronic respiratory symptoms and reduced pulmonary function. Since then, many studies and several major reports (California Environmental Protection Agency and Air Resources Board, 2005; California Environmental Protection Agency and Office of Environmental Health Hazard Assessment, 1997; U.S. Department of Health and Human Services, 2006; U.S. Environ-

RESPIRATORY SYMPTOMS AND ILLNESSES IN ADULTS

735

mental Protection Agency, 1992) have been published that considered the association between SHS exposure and respiratory health in nonsmokers. 19.5.1

Respiratory Symptoms

The association between respiratory symptoms in nonsmokers and involuntary exposure to tobacco smoke has largely been investigated in experimental and cross-sectional studies. The experimental evidence primarily comes from studies that assessed acute responses of asthmatics who were exposed to SHS in a chamber. This experimental approach cannot be readily controlled because of the impossibility of blinding subjects to their being exposed to SHS. However, suggestibility does not appear to underlie physiological responses of asthmatics to SHS (Urch et al., 1988). Of the three studies involving exposure of unselected asthmatics to SHS, only one showed a definite adverse effect (Dahms et al., 1981; Hargreave et al., 1981; Murray and Morrison, 1986; Qin et al., 1991; Shephard et al., 1979). Stankus et al. (1988) recruited 21 asthmatics who reported exacerbation with exposure to SHS. With challenge in an exposure chamber at concentrations much greater than that typically encountered in indoor environments, seven subjects experienced a more than 20% decline in FEV1. Among the 13 epidemiologic studies that were reviewed in the 2006 Surgeon General’s report, only a few were longitudinal in their design (Jaakkola et al., 1996; Robbins et al., 1993; Schwartz and Zeger, 1990). Consistent evidence of an effect of SHS exposure on acute respiratory symptoms in adults has been found (U.S. Department of Health and Human Services, 2006). However, the evidence of an effect of SHS on chronic respiratory symptoms has been less consistent (U.S. Department of Health and Human Services, 2006). Overall, symptoms of chronic cough and dyspnea have been more consistently associated with exposure to SHS than have the symptoms of chronic phlegm and wheeze (U.S. Department of Health and Human Services, 2006). Several studies suggest that exposure to SHS may cause acute respiratory morbidity. Analysis of National Health Interview Survey data showed that a pack-a-day smoker increases respiratory restricted days by about 20% for a nonsmoking spouse (Ostro, 1989). In a study of determinants of daily respiratory symptoms in Los Angeles student nurses, it was found that there was a significantly increased risk of an episode of phlegm with a smoking roommate, after controlling for personal smoking (Schwartz and Zeger, 1990). Several studies have addressed chronic respiratory symptoms. Leuenberger et al. (1994) describe associations between passive exposures to tobacco smoke, at home and in the workplace, and respiratory symptoms in 4197 randomly selected never-smoking adults in the Swiss Study on Air Pollution and Lung Diseases in Adults, a multicenter study in eight areas of the country. Exposed subjects were those who reported any exposure during the past 12 months; exposed persons were then asked about workplace exposure and also about the number of smokers and the duration of exposure at home and work together. Involuntary smoke exposure was associated with asthma, dyspnea, bronchitis and chronic bronchitis symptoms, and allergic rhinitis. The increments in risk were substantial, ranging approximately 40–80% for the different respiratory outcome measures. The increments were not reduced by control for educational level, and dose–response relationships were found with the quantitative indicators of exposure. For several of the outcome measures, the dose–response relationships tended to be steeper for those who also reported workplace exposure. In a cross-sectional population study of 1954 women, Baker and

736

SECONDHAND SMOKE

Henderson (1999) found a significant association (OR ¼ 1.73; 95% CI: 1.05, 2.85) of wheeze in nonsmoking mothers living with a smoking partner. No association was found in fathers. Other adverse respiratory effects of involuntary smoking have been found in adults. Robbins et al. (1993) examined predictors of new symptoms compatible with “airway obstructive disease” in a cohort study of 3914 nonsmoking participants in the Adventist Health Study. Significantly increased risk was identified in association with exposure during both childhood and adulthood. In a cross-sectional study, Dayal et al. (1994) found that never-smoking Philadelphia residents with a reported diagnosis of asthma, chronic bronchitis, or emphysema had sustained significantly greater exposure to tobacco smoke than unaffected controls. Several epidemiologic studies have investigated the roles of SHS exposure in the onset of asthma and in exacerbating asthma in adults. Two cross-sectional studies, one case-control study, and two prospective cohort studies all found an association between SHS exposure and asthma morbidity (Jindal et al., 1994, 1999; Mannino et al., 1997; Ostro et al., 1994; Sippel et al., 1999; Tarlo et al., 2000). A large cross-sectional study by Mannino et al., using data from the 1991 National Health Interview Survey (Mannino et al., 1997), found that lifetime nonsmokers exposed to SHS had a 44% increased risk of exacerbated chronic respiratory conditions, compared to unexposed nonsmokers, after adjusting for potential confounders. In a small case-control study, Tarlo et al. (2000) found that asthma patients with an exacerbation were more likely to have been exposed to SHS than asthma patients without an exacerbation. In a cohort study of lifetime nonsmoking asthma patients by Jindal et al., SHS exposure was found to increase the risk of acute episodes and impaired lung function (Jindal et al., 1994). Ostro et al. (1994) also found an increased risk of shortness of breath, cough, and restricted activity among asthmatics exposed to SHS compared to unexposed asthmatics. Sippel et al. (1999) found a significant increase in the use of hospital services, such as urgent care and emergency room visits, among asthma patients exposed to SHS compared to those who were unexposed. In a prospective cohort study of adult nonsmokers admitted to the hospital for asthma, Eisner et al. found a significant association between the severity of asthma symptoms and SHS exposure after controlling for potential confounders (Eisner et al., 2005). In 1999, Weiss, Utell and Samet reviewed the literature on SHS exposure and asthma in adults (Weiss et al., 1999). They found that two prospective cohort studies and a populationbased case-control study found a significant association between SHS exposure and the onset of asthma in adults. However, the authors concluded that because the evidence is scant and has potential problems in study design, “a definitive conclusion cannot be made at this time.” Subsequently, a population-based case-control study by Jaakkola et al. (2003) investigated the association between SHS exposure and the onset of adult asthma in the Pirkanmaa district of Finland. They recruited all incident cases of asthma in the district and selected populationbased controls. After excluding current or previous smokers, there were 239 lifetime nonsmoking cases and a comparison group of 487 lifetime nonsmoking controls. They found a twofold increased risk of asthma among those exposed to SHS in the home and the workplace compared with those who were unexposed. In a study published in 2006, Menzies et al. (2006) reported findings of a beneficial effect of the ban on smoking in Scotland on the health of bar workers. The investigators measured FEV1 level in nonasthmatic and asthmatic nonsmoking bar workers before and after the ban was implemented. They found that the FEV1 level increased after the ban was implemented.

RESPIRATORY SYMPTOMS AND ILLNESSES IN ADULTS

737

They also found a decrease in the prevalence of respiratory symptoms among nonsmoking bar workers. 19.5.2

Lung Function in Adults

With regard to involuntary smoking and lung function in adults, exposure to secondhand smoke has been associated in cross-sectional investigations with reduction of the FEF25–75. White and Froeb (1980) compared spirometric test results in middle-aged nonsmokers with at least 20 years of involuntary smoking in the workplace to the results in an unexposed control group of nonsmokers. The mean FEF25–75 of the exposed group was significantly reduced, by 15% of predicted value in women and by 13% in men. This investigation has been intensely criticized with regard to the spirometric test procedures, the determination and classification of exposures, and the handling of former smokers in the analyses. An investigation in France examined the effect of marriage to a smoker in over 7800 adults in seven cities (Kauffmann et al., 1983). The study included 849 male and 826 female nonsmokers exposed to tobacco smoke by their spouses’ smoking. At age above 40 years, the FEF25–75 was reduced in nonsmoking men and women with a smoking spouse. The investigators interpreted this finding as representing a cumulative adverse effect of marriage to a smoker. In a subsequent report, the original findings in the French women were confirmed, but a parallel analysis in a large population of U.S. women did not show effects of involuntary smoking on lung function (Spengler and Ferris, 1985). The results of an investigation of 163 nonsmoking women in the Netherlands also suggested adverse effects of tobacco smoke exposure in the home on lung function (Brunekreef et al., 1985; Remijn et al., 1985). Cross-sectional analysis of spirometric data collected in 1982 demonstrated adverse effects of tobacco smoke exposure in the home, but in a sample of women, domestic exposure to tobacco smoke was not associated with longitudinal decline of lung function during the period 1965–1982. Svendsen et al. (1987) assessed the effects of spouse smoking on 1400 nonsmoking male participants in the Multiple Risk Factor Intervention Trial (MRFIT). The subjects were aged 35–57 years at enrollment and were at high risk for mortality from coronary artery disease. At the baseline visit, the maximum FEV1 was approximately 3% lower for the men married to a smoker. Masi et al. (1988) evaluated lung function of 293 young adults, using spirometry and measurement of the diffusing capacity and lung volumes. The results varied with gender. In men, reduction of the maximal midexpiratory flow rate was associated with maternal smoking and exposure to SHS during childhood. In women, reduction of the diffusing capacity was associated with exposure to SHS at work. In the study of a general population sample in western Scotland, nonsmokers living with another household member who was a smoker had significantly reduced lung function in comparison to unexposed nonsmokers (Hole et al., 1989); the reduction of FEV1 associated with involuntary smoking was about 5%. Secondhand smokers with higher exposure had greater reduction of FEV1. Masjedi et al. (1990) investigated the effects of exposure to SHS on lung function of 288 nonsmoking volunteers living in Tehran. Ventilatory function was reduced significantly for men exposed at work, although an additional effect of exposure at home was not found. SHS exposure at home and at work did not reduce the lung function of the female subjects.

738

SECONDHAND SMOKE

In a meta-analysis of 15 cross-sectional studies, Carey et al. found a mean deficit of 1.7% in FEV1 level due to SHS exposure (Carey et al., 1999). They also conducted a separate crosssectional investigation of 1623 adults in Britain and found similar results, with a stronger effect in men than in women. Subsequently, Chen et al. found an inverse dose–response relationship between SHS and FEV1 level in 301 adults in Scotland for exposure to SHS at work (Chen et al., 2001). In a study of young Canadian adults, Jaakkola et al. (1995) did not find effects of home and workplace exposures on an 8-year change in lung function. In persons less than 26 years of age at enrollment, workplace SHS exposure was associated with greater decline. In another cohort study of 1391 lifetime nonsmokers and former smokers in California, Abbey et al. found a decrease in the ratio of FEV1 to FVC in both women exposed at home and men exposed at work (Abbey et al., 1998). However, the results were nonsignificant. Using NHANES III data, Eisner (2002) conducted a cross-sectional study to investigate the association between the level of SHS exposure and the pulmonary function in 10,581 nonsmoking adults and 440 nonsmoking adults with asthma. He found that FVC and FEV1 levels were significantly lower in adult females with the highest concentration of serum cotinine levels compared to adult females with lower serum cotinine levels. He did not find a significant association in adult males. Several investigators have reported associations of involuntary smoking with COPD in nonsmokers. In the Japanese cohort study, a nonsignificant trend of increasing mortality from chronic bronchitis and emphysema with increasing passive exposure of nonsmoking women was reported (Hirayama, 1984). Kalandidi et al. (1987) conducted a case-control study of involuntary smoking and chronic obstructive pulmonary disease; the cases were nonsmoking women with obstruction and reduction of the FEV1 by at least 20%. Smoking by the husband was associated with a doubling of risk. Dayal et al., (1994) conducted a casecontrol study of self-reported obstructive lung disease in 219 never-smoking residents of Philadelphia. Household SHS exposure from one or more packs per day was associated with a doubling of risk. In a prospective cohort study of 3914 nonsmoking Adventists, SHS exposure was associated with report of symptoms considered to be reflective of “airway obstructive disease” (Robbins et al., 1993). An association of SHS exposure with COPD seems biologically implausible, however, since only a minority of active smokers develop this disease, and adverse effects of involuntary smoking on lung function in adults have not been observed consistently (U.S. Department of Health and Human Services, 1984). The autopsy study of Trichopoulos et al. (1992) did show, however, that airways of nonsmokers can be affected by SHS. The 2006 report of the Surgeon General states that the evidence is suggestive but not sufficient to infer a causal association for the effects of SHS exposure on lung function in adults (U.S. Department of Health and Human Services, 2006). However, further research is warranted because of widespread exposure in workplaces and homes. 19.5.3

Odor and Irritation

Tobacco smoke contains numerous irritants, including particulate matter and gases (U.S. Department of Health and Human Services, 1986). Both questionnaire surveys and laboratory studies involving exposure to SHS have shown annoyance and irritation of the eyes and upper and lower airways from involuntary smoking. In several surveys of nonsmokers, complaints about tobacco smoke at work and in public places were common

RESPIRATORY SYMPTOMS AND ILLNESSES IN ADULTS

739

(U.S. Department of Health and Human Services, 1986): about 50% of respondents complained about tobacco smoke at work, and a majority were disturbed by tobacco smoke in restaurants. The experimental studies show that the rate of eye blinking is increased by SHS, as are complaints of nose and throat irritation (U.S. Department of Health and Human Services, 1986). In the study of exposure to secondhand smoke on commercial airline flights reported by Mattson et al. (1989), changes in nose and eye symptoms were associated with nicotine exposure. The odor and irritation associated with SHS merit special consideration because a high proportion of nonsmokers are annoyed by exposure to SHS, and control of concentrations in indoor air poses difficult problems in the management of heating, ventilating, and air-conditioning systems. Using a challenge protocol, Bascom et al. (1991) showed that persons characterizing themselves as SHS sensitive have greater responses on exposure than persons considering themselves as nonsensitive. They found similar results in a subsequent study (Bascom et al., 1996). Nowak et al. also found an increase in nose and mouth symptoms after exposing 10 persons with mild asthma to SHS in a chamber (Nowak et al., 1997). In a cross-sectional survey, Cummings et al. found a significant association between SHS exposure and irritation in 723 adult volunteers (Cummings et al., 1991). In 2001, Junker et al. reported findings of an experimental exposure assessment of SHS to determineodorandirritationthreshold levels(Junkeretal.,2001).They exposed people tovery low concentrations of sidestream SHS and measured odor detection, acute sensory symptoms, breathing patterns, and annoyance. They found that thresholds for odor and irritation were significantly lower than previously reported by 100 times and 10 times, respectively. 19.5.4

Total Mortality

Several cohort studies provide information on involuntary smoking and mortality from all causes. In the Scottish cohort study, total mortality was initially reported as increased for women living with a smoker, but not for men (Gillis et al., 1984). On further follow-up, allcause mortality was increased in all secondhand smokers (RR: 1.27; 95% CI: 0.95,1.70). As described previously, total mortality was also increased among nonsmoking participants in MRFIT who lived with smokers (Svendsen et al., 1987). In contrast, mortality was not increased for nonsmoking female subjects in a study in Amsterdam (Vandenbroucke et al., 1984). Neither the study in Scotland nor the study in Amsterdam controlled for other factors that influence total mortality. In the cohort study in Washington County, all-cause mortality rates were significantly increased for men (RR: 1.17) and for women (RR: 1.15) after adjustment for housing quality, schooling, and marital status (Sandler et al., 1989a). Allcause mortality was also increased for secondhand smokers in the Evans County cohort (RR: 1.39, 95% CI: 0.99,1.94). Wells (1988) has made an estimate of the number of adult deaths in the United States attributable to secondhand smoke exposure. The total is about 46,000, including 3000 from lung cancer, 11,000 from other cancers, and 32,000 from heart disease. Additionally, Nurminen and Jaakkola estimated adult mortality attributable to SHS exposures in the workplace in Finland (Nurminen and Jaakkola, 2001). They estimated that 250 deaths were due to exposure to secondhand smoke at work in 1996. Among these deaths, 2.8% were from lung cancer, 1.1% from chronic obstructive pulmonary disease, 4.5% from asthma, 3.4% for ischemic heart disease, and 9.4% for cerebrovascular stroke. The small excesses of all-cause mortality associated with exposure to secondhand smoke in the epidemiological studies parallel the findings for cardiovascular disease, the leading

740

SECONDHAND SMOKE

cause of death in these cohorts. The increased risk of death associated with secondhand smoke has public health significance as an indicator of the overall impact of this avoidable exposure. 19.5.5

Control Measures

Since the 1980s, there has been growing momentum for making public places and workplaces smoke free. The public health basis for this movement lies in the increasingly strong findings on the health risks of SHS (Table 19.1) and on the need to eliminate smoking indoors to fully protect nonsmokers from inhaling SHS. Cigarettes are strong sources of gaseous and particulate emissions, and use of mass balance models implies that concentrations of SHS components could not be controlled by either ventilation or air cleaning (U.S. Department of Health and Human Services, 2006). Such consideration led the American Society of Heating, Refrigerating and Air Conditioning Engineers (ASHRAE) to conclude that ventilation was not a sufficient control measure for SHS (Samet et al., 2005). The 2006 report of the Surgeon General reached a similar conclusion (U.S. Department of Health and Human Services, 2006). There are regulatory and nonregulatory approaches to eliminating smoking indoors. There are an increasing number of local and state ordinances banning smoking in public places and workplaces, and many large companies have policies in place that prohibit smoking indoors. Even some major hotel chains are now smoke free. In recent surveys, the majority of employed people report working under a smoke-free policy, although there is some variation by type of workplace and region of the country (U.S. Department of Health and Human Services, 2006). Blue collar and farm workers are least likely to be covered. The home is not subject to regulation, but increasing numbers of households in the United States have voluntary policies in place (U.S. Department of Health and Human Services, 2006). The majority of households are now smoke free, and increasing numbers of households with smokers have smoke-free policies in place. The effectiveness of these policies is still inadequate, and research is in progress to develop more efficacious approaches to reduce SHS exposure in homes, particularly, for example, for children with asthma who are especially susceptible to the adverse health effects of SHS. Thischaptersummarizestheconvergingandnow extensiveevidenceonthehealtheffectsof involuntary exposure to tobacco smoke. Although the initial research on involuntary smoking addressed respiratory effects, subsequent investigations have examined associations with diverse health effects including nonrespiratory cancers in children and adults, ischemic heart disease, age at menopause, sudden infant death syndrome, and birth weight. The evidence on involuntary exposure to tobacco smoke is now voluminous and consequently this review is selective in its citations. The most recent compilation of the evidence can be found in the 2005 report of the California Environmental Protection Agency, “Health Effects of Exposure to Environmental Tobacco Smoke” (California Environmental Protection Agency and Air Resources Board, 2005) and the 2006 report of the U.S. Surgeon General (U.S. Department of Health and Human Services, 2006).

19.6 SUMMARY The effects of active smoking and the toxicology of cigarette smoking have been comprehensively examined. The periodic reports of the U.S. Surgeon General and other summary

REFERENCES

741

reports have considered the extensive evidence on active smoking; these reports have provided definitive conclusions concerning the adverse effects of active smoking, which have prompted public policies and scientific research directed at prevention and cessation and smoking. Although the evidence on involuntary smoking is not as extensive as that on active smoking, health risks of involuntary smoking have been identified and causal conclusions reached, beginning in the mid-1980s (Table 19.1). The 1986 report of the U.S. Surgeon General (U.S. Department of Health and Human Services, 1986) and the 1986 report of the National Research Council (National Research Council and Committee on Passive Smoking, 1986) both concluded that involuntary exposure to tobacco smoke causes respiratory infections in children, increases the prevalence of respiratory symptoms in children, reduces the rate of functional growth as the lung matures, and causes lung cancer in nonsmokers. These conclusions have been reaffirmed in subsequent reports (California Environmental Protection Agency and Office of Environmental Health Hazard Assessment, 1997; U.S. Department of Health and Human Services, 2006; U.S. Environmental Protection Agency, 1992) and new conclusions added. Involuntary smoking is now considered as a cause of asthma and a factor exacerbating asthma (California Environmental Protection Agency and Office of Environmental Health Hazard Assessment, 1997; U.S. Department of Health and Human Services, 2006; U.S. Environmental Protection Agency, 1992) and as a cause of heart disease (California Environmental Protection Agency and Office of Environmental Health Hazard Assessment, 1997; U.S. Department of Health and Human Services, 2006). The 2006 Surgeon General’s report (U.S. Department of Health and Human Services, 2006) leaves no doubt: secondhand smoke causes premature death and disease in children and adults who do not smoke (p. 11). The adverse effects of involuntary exposure to tobacco smoke have provided a strong rationale for policies directed at reducing and eliminating exposure of nonsmokers to SHS (U.S. Department of Health and Human Services, 1986). Complete protection of nonsmokers in public locations and the workplace may require the banning of smoking, since the 1986 report of the Surgeon General (U.S. Department of Health and Human Services, 1986) concluded that “the simple separation of smokers and nonsmokers within the same air space may reduce, but does not eliminate, the exposure of nonsmokers to environmental tobacco smoke” (p. 7). The 2006 Surgeon General’s report went even further: “the scientific evidence indicates that there is no risk-free level of exposure to secondhand smoke” (p. 11).

REFERENCES Abbey DE, Burchette RJ, Knutsen SF, Mc Donnell WF, Lebowitz MD, Enright PL (1998) Long-term particulate and other air pollutants and lung function in nonsmokers. Am. J. Respir. Crit. Care Med. 158:289–298. Ahlborg G Jr, Bodin L (1991) Tobacco smoke exposure and pregnancy outcome among working women. A prospective study at prenatal care centers in Orebro County, Sweden. Am. J. Epidemiol. 133:338–347. Al Delaimy WK, Cho E, Chen WY, Colditz G, Willet WC (2004) A prospective study of smoking and risk of breast cancer in young adult women. Cancer Epidemiol. Biomarkers Prev. 13:398–404.

742

SECONDHAND SMOKE

Allred EN, Bleecker ER, Chaitman BR, Dahms TE, Gottlieb SO, Hackney JD, Pagano M, Selvester RH, Walden SM, Warren J (1989) Short-term effects of carbon monoxide exposure on the exercise performance of subjects with coronary artery disease. New Engl. J. Med. 321:1426–1432. Andersen P, Pedersen OF, Bach B, Bonde GJ (1982) Serum antibodies and immunoglobulins in smokers and nonsmokers. Clin. Exp. Immunol. 82:467–473. Andresen BD, Ng KJ, Iams JD, Bianchine JR (1982) Cotinine in amniotic fluids from passive smokers (letter). Lancet 82:791–792. Baker D, Henderson J (1999) Differences between infants and adults in the social aetiology of wheeze. The ALSPAC Study Team. Avon Longitudinal Study of Pregnancy and Childhood. J. Epidemiol. Community Health 53:636–642. Barnoya J, Glantz SA (2005) Cardiovascular effects of secondhand smoke: nearly as large as smoking. Circulation 111:2684–2698. Bascom R, Kulle T, Kagey-Sobotka A, Proud D (1991) Upper respiratory tract environmental tobacco smoke sensitivity. Am. Rev. Respir. Dis. 143:1304–1311. Bascom R, Kesavanathan J, Permutt T, Fitzgerald TK, Sauder L, Swift DL (1996) Tobacco smoke upper respiratory response relationships in healthy nonsmokers. Fundam. Appl. Toxicol. 29:86–93. Bennett WP, Alavanja MC, Blomeke B, Vahakangas KH, Castren K, Welsh JA, Bowman ED, Khan MA, Flieder DB, Harris CC (1999) Environmental tobacco smoke, genetic susceptibility, and risk of lung cancer in never-smoking women. J. Natl. Cancer Inst. 91:2009–2014. Benowitz NL (1996) Cotinine as a biomarker of environmental tobacco smoke exposure. Epidemiol. Rev. 18:188–204. Boffetta P, Tredaniel J, Greco A (2000) Risk of childhood cancer and adult lung cancer after childhood exposure to passive smoke: a meta-analysis. Environ. Health Perspect. 108:73–82. Bonita R, Duncan J, Truelsen T, Jackson RT, Beaglehole R (1999) Passive smoking as well as active smoking increases the risk of acute stroke. Tob. Control 8:156–160. Bottoms SF, Kuhnert BR, Kuhnert PM, Reese AL (1982) Maternal passive smoking and fetal serum thiocyanate levels. Am. J. Obstet. Gynecol. 144:787–791. Brennan P, Buffler PA, Reynolds P, Wu AH, Wichmann HE, Agudo A, Pershagen G, Jockel KH, Benhamou S, Greenberg RS, Merletti F, Winck C, Fontham ET, Kreuzer M, Darby SC, Forastiere F, Simonato L, Boffetta B (2004) Secondhand smoke exposure in adulthood and risk of lung cancer among never smokers: a pooled analysis of two large studies. Int. J. Cancer 109:125–131. Brondum J, Shu XO, Steinbuch M, Severson RK, Potter JD, Robison LL (1999) Parental cigarette smoking and the risk of acute leukemia in children. Cancer 85:1380–1388. Brunekreef B, Fischer P, Remijn B, Van Der Lende R, Schouten J, Quanjer P (1985) Indoor air pollution and its effect on pulmonary function of adult nonsmoking women. III. Passive smoking and pulmonary function. Int. J. Epidemiol. 14:227–230. Bunin GR, Buckley JD, Boesel CP, Rorke LB, Meadows AT (1994) Risk factors for astrocytic glioma and primitive neuroectodermal tumor of the brain in young children: a report from the Children’s Cancer Group. Cancer Epidemiol. Biomarkers Prev. 3:197–204. Burch JD, Rohan TE, Howe GR, Risch HA, Hill GB, Steele R, Miller AB (1989) Risk of bladder cancer by source and type of tobacco exposure: a case-control study. Int. J. Cancer 44:622–628. Butler TL (1988) The Relationship of Passive Smoking to Various Health Outcomes Among Seventh Day Adventists in California. Los Angeles: CA University of California Los Angeles. California Environmental Protection Agency (Cal/EPA) and Air Resources Board (2005) Proposed Identification of Environmental Tobacco Smoke as a Toxic Air Contaminant. Sacramento, CA: California Environmental Protection Agency. California Environmental Protection Agency (Cal/EPA) (1997) Office of Environmental Health Hazard Assessment. Health Effects of Exposure to Environmental Tobacco Smoke. California Environmental Protection Agency.

REFERENCES

743

Cardenas VM, Thun MJ, Austin H, Lally CA, Clark WS, Greenberg RS, Heath CW Jr (1997) Environmental tobacco smoke and lung cancer mortality in the American Cancer Society’s Cancer Prevention Study. II (published erratum appears in Cancer Causes Control 1997 Jul;8(4):675). Cancer Cause Control 8:57–64. Carey IM, Cook DG, Strachan DP (1999) The effects of environmental tobacco smoke exposure on lung function in a longitudinal study of British adults. Epidemiology 10:319–326. Carmella SG, Han S, Fristad A, Yang Y, Hecht SS (2003) Analysis of total 4-(methylnitrosamino)-1-(3pyridyl)-1-butanol (NNAL) in human urine. Cancer Epidemiol. Biomarkers Prev. 12:1257–1261. Celermajer DS, Adams MR, Clarkson P, Robinson J, McCredie R, Donald A, Deanfield JE (1996) Passive smoking and impaired endothelium-dependent arterial dilatation in healthy young adults. N. Engl. J. Med. 334:150–154. Charlton A .(1984) Children’s coughs related to parental smoking. Brit. Med. J. 288:1647–1649. Chen R, Tunstall-Pedoe H, Tavendale R (2001) Environmental tobacco smoke and lung function in employees who never smoked: the Scottish MONICA study. Occup. Environ. Med. 58:563–568. Cheng YJ, Hildesheim A, Hsu MM, Chen IH, Brinton LA, Levine PH, Chen CJ, Yang CS (1999) Cigarette smoking, alcohol consumption and risk of nasopharyngeal carcinoma in Taiwan. Cancer Cause Control 10:201–207. Claxton LD, Morin RS, Hughes TJ, Lewtas J (1989) A genotoxic assessment of environmental tobacco smoke using bacterial bioassays. Mutat. Res. 222:81–99. Coghlin J, Hammond SK, Gann PH (1989) Development of epidemiologic tools for measuring environmental tobacco smoke exposure. Am. J. Epidemiol. 130:696–704. Coghlin J, Gann PH, Hammond SK, Skipper PL, Taghizadeh K, Paul M, Tannenbaum SR (1991) 4Aminobiphenyl hemoglobin adducts in fetuses exposed to the tobacco smoke carcinogen in utero. J. Natl. Cancer Inst. 83:274–280. Coker AL, Rosenberg AJ, McCann MF, Hulka BS (1992) Active and passive cigarette smoke exposure and cervical intraepithelial neoplasia. Cancer Epidemiol. Biomarkers Prev. 1:349–356. Colley JR, Holland WW, Corkhill RT (1974a) Influence of passive smoking and parental phlegm on pneumonia and bronchitis in early childhood. Lancet 2:1031–1034. Colley JRT, Holland WW, Corkhill RT (1974b) Influence of passive smoking and parental phlegm on pneumonia and bronchitis in early childhood. Lancet 2:1031–1034. Comstock GW, Lundin FE (1967) Parental smoking and perinatal mortality. Am. J. Obstet. Gynecol. 98:708–718. Cook DG, Strachan DP (1997a) Health effects of passive smoking. 3. Parental smoking and prevalence of respiratory symptoms and asthma in school age children. Thorax 52:1081–1094. Cook DG, Strachan DP (1997b) Parental smoking and prevalence of respiratory symptoms and asthma in school age children. Thorax 52:1081–1094. Cook DG, Strachan DP, Carey IM (1998) Health effects of passive smoking. 9. Parental smoking and spirometric indices in children. Thorax 53:884–893. Coultas DB, Howard CA, Peake GT, Skipper BJ, Samet JM (1987) Salivary cotinine levels and involuntary tobacco smoke exposure in children and adults in New Mexico. Am. Rev. Respir. Dis. 136:305–309. Coultas DB, Peake GT, Samet JM (1989) Questionnaire assessment of lifetime and recent exposure to environmental tobacco smoke. Am. J. Epidemiol. 130:338–347. Coultas DB, Samet JM, McCarthy JF, Spengler JD (1990) Variability of measures of exposure to environmental tobacco smoke in the home. Am. Rev. Respir. Dis. 142:602–606. Crawford FG, Mayer J, Santella RM, Cooper TB, Ottman R, Tsai WY, Simon-Cereijido G, Wang M, Tang D, Perera FP (1994) Biomarkers of environmental tobacco smoke in preschool children and their mothers. J. Natl. Cancer Inst. 86:1398–1402.

744

SECONDHAND SMOKE

Cummings KM, Markello SJ, Mahoney MC, Marshall JR (1989) Measurement of lifetime exposure to passive smoke. Am. J. Epidemiol 89:122–132. Cummings KM, Markello SJ, Mahoney M, Bhargava AK, McElroy PD, Marshall JR (1990) Measurement of current exposure to environmental tobacco smoke. Arch. Environ. Health 45:74–79. Cummings KM, Zaki A, Markello S (1991) Variation in sensitivity to environmental tobacco smoke among adult non-smokers. Int. J. Epidemiol. 20:121–125. Dahms TE, Bolin JF, Slavin RG (1981) Passive smoking: effects on bronchial asthma. Chest 80:530– 534. Dayal HH, Khuder S, Sharrar R, Trieff N (1994) Passive smoking in obstructive respiratory disease in an industrialized urban population. Environ. Res. 65:161–171. DeMarini DM (2004) Genotoxicity of tobacco smoke and tobacco smoke condensate: A review. Mutat. Res. Diez-Roux AV, Nieto FJ, Comstock GW, Howard G, Szklo M (1995) The relationship of active and passive smoking to carotid atherosclerosis 12–14 years later. Prev. Med. 24:48–55. Donnan GA, McNeil JJ, Adena MA, Doyle AE, O’Malley HM, Neill GC (1989) Smoking as a risk factor for cerebral ischaemia. Lancet 89:643–647. Dunnick JK, Elwell MR, Huff J, Barrett JC (1995) Chemically induced mammary gland cancer in the National Toxicology Program’s carcinogenesis bioassay. Carcinogenesis 16:173–179. Eisner MD (2002) Environmental tobacco smoke exposure and pulmonary function among adults in NHANES III: impact on the general population and adults with current asthma. Environ. Health Perspect. 110:765–770. Eisner MD, Klein J, Hammond SK, Koren G, Lactao G, Iribarren C (2005) Directly measured second hand smoke exposure and asthma health outcomes. Thorax 60:814–821. El Bayoumy K, Chae YH, Upadhyaya P, Rivenson A, Kurtzke C, Reddy B, Hecht SS (1995) Comparative tumorigenicity of benzo[a]pyrene, 1-nitropyrene and 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine administered by gavage to female CD rats. Carcinogenesis 16:431–434. Emmons KM, Hammond SK, Fava JL, Velicer WF, Evans JL, Monroe AD (2001) A randomized trial to reduce passive smoke exposure in low-income households with young children. Pediatrics 108:18–24. Enstrom JE (1979) Rising lung cancer mortality among nonsmokers. J. Natl. Cancer Inst. 62:755– 760. Enstrom JE, Kabat GC (2003) Environmental tobacco smoke and tobacco related mortality in a prospective study of Californians, 1960–98. Brit. Med. J. 326:1057. Eskenazi B, Bergmann JJ (1995) Passive and active maternal smoking during pregnancy as measured by serum cotinine, and postnatal smoke exposure. I. Effects on physical growth at age 5 years. Am. J. Public Health 142:S10–S18. Eskenazi B, Trupin LS (1995) Passive and active maternal smoking during pregnancy, as measured by serum cotinine, and postnatal smoke exposure. II. Effects on neurodevelopment at age 5 years. Am. J. Public Health 142:S19–S29. Filippini G, Farinotti M, Lovicu G, Maisonneuve P, Boyle P (1994) Mothers’ active and passive smoking during pregnancy and risk of brain tumours in children. Int. J. Cancer 57:769–774. Filippini G, Farinotti M, Ferrarini M (2000) Active and passive smoking during pregnancy and risk of central nervous system tumours in children. Paediatr. Perinat. Epidemiol. 14:78–84. Finette BA, O’Neill JP, Vacek PM, Albertini RJ (1998) Gene mutations with characteristic deletions in cord blood T lymphocytes associated with passive maternal exposure to tobacco smoke. Nat. Med. 4:1144–1151.

REFERENCES

745

Fontham ETH, Correa P, Reynolds P, Wu-Williams A, Buffler PA, Greenberg RS, Chen VW, Alterman T, Boyd P, Austin DF, Liff J (1994) Environmental tobacco smoke and lung cancer in nonsmoking women: A multicenter study. J. Am. Med. Assoc. 271:1752–1759. Friedman GD, Petitti DB, Bawol RD (1983) Prevalence and correlates of passive smoking. Am. J. Public Health 73:401–405. Fukuda K, Shibata A (1990) Exposure–response relationships between woodworking, smoking or passive smoking, and squamous cell neoplasms of the maxillary sinus. Cancer Cause Control 1:165–168. Garfinkel L (1981) Time trends in lung cancer mortality among nonsmokers and a note on passive smokers. J. Natl. Cancer Inst. 66:1061–1066. Garland C, Barret-Connor E, Suarez L, Criqui MH, Wingard DL (1985) Effects of passive smoking on ischemic heart disease mortality of nonsmokers: a prospective study. Am. J. Epidemiol. 121:645– 650. Gillis CR, Hole DJ, Hawthorne VM, Boyle P (1984) The effect of environmental tobacco smoke in two urban communities in the west of Scotland. Eur. J. Respir. Dis. 65:121–126. Glantz SA, Parmley WW (1995) Passive smoking and heart disease: mechanisms and risk. J. Am. Med. Assoc. 273:1047–1053. Gold EB, Leviton A, Lopez R, Gilles FH, Hedley-Whyte ET, Kolonel LN, Lyon JL, Swanson GM, Weiss NS, West D, et al.(1993) Parental smoking and risk of childhood brain tumors. Am. J. Epidemiol. 137:620–628. Guerin MR, Jenkins RA, Tomkins BA (1992) The Chemistry of Environmental Tobacco Smoke: Composition and Measurement. Chelsea, MI, Lewis Publishers, Inc. Hackshaw A (2003) Passive smoking: paper does not diminish conclusion of previous reports. Brit. Med. J. 327:501–502. Hackshaw AK, Law MR, Wald NJ (1997) The accumulated evidence on lung cancer and environmental tobacco smoke. Brit. Med. J. 315:980–988. Haddow JE, Knight GJ, Palomaki GE, McCarthy JE (1988) Second-trimester serum cotinine levels in nonsmokers in relation to birth weight. Am. J. Obstet. Gynecol. 159:481–484. Hamajima N, Hirose K, Tajima K, Rohan T, Calle EE, Heath CW Jr, Coates RJ, Liff JM, Talamini R, Chantarakul N, Koetsawang S, Rachawat D, Morabia A, Schuman L, Stewart W, Szklo M, Bain C, Schofield F, Siskind V, Band P, Coldman AJ, Gallagher RP, Hislop TG, Yang P, Kolonel LM, Nomura AM, Hu J, Johnson KC, Mao Y, de Sanjose S, Lee N, Marchbanks P, Ory HW, Peterson HB, Wilson HG, Wingo PA, Ebeling K, Kunde D, Nishan P, Hopper JL, Colditz G, Gajalanski V, Martin N, Pardthaisong T, Silpisornkosol S, Theetranont C, Boosiri B, Chutivongse S, Jimakorn P, Virutamasen P, Wongsrichanalai C, Ewertz M, Adami HO, Bergkvist L, Magnusson C, Persson I, Chang-Claude J, Paul V, Skegg DC, Spears GF, Boyle P, Evstifeeva T, Daling JR, Hutchinson WB, Malone K, Noonan EA, Stanford JL, Thomas DB, Weiss NS, White E, Andrieu N, Bremond A, Clavel F, Gairard B, Lansac J, Piana L, Renaud R, Izquierdo A, Viladiu P, Cuevas HR, Ontiveros P, Palet A, Salazar SB, Aristizabel N, Cuadros A, Tryggvadottir L, Tulinius H, Bachelot A, Le MG, Peto J, Franceschi S, Lubin F, Modan B, Ron E, Wax Y, Friedman GD, Hiatt RA, Levi F, Bishop T, Kosmelj K, Primic-Zakelj M, Ravnihar B, Stare J, Beeson WL, Fraser G, Bullbrook RD, Cuzick J, Duffy SW, Fentiman IS, Hayward JL, Wang DY, McMichael AJ, Mcpherson K, Hanson RL, Leske MC, Mahoney MC, Nasca PC, Varma AO, Weinstein AL, Moller TR, Olsson H, Ranstam J, Goldbohm RA, van den Brandt PA, Apelo RA, Baens J, de la Cruz JR, Javier B, Lacaya LB, Ngelangel CA, La Vecchia C, Negri E, Marubini E, Ferraroni M, Gerber M, Richardson S, Segala C, Gatei D, Kenya P, Kungu A, Mati JG, Brinton LA, Hoover R, Schairer C, Spirtas R, Lee HP, Rookus MA, Van Leeuwen FE, Schoenberg JA, McCredie M, Gammon MD, Clarke EA, Jones L, Neil A, Vessey M, Yeates D, Appleby P, Banks E, Beral V, Bull D, Crossley B, Goodill A, Green J, Hermon C, Key T, Langston N, Lewis C, Reeves G, Collins R, Doll R, Peto R, Mabuchi K, Preston D, Hannaford P, Kay C, Rosero-Bixby L, Gao YT, Jin F, Yuan JM, Wei HY, Yun T, Zhiheng C, Berry

746

SECONDHAND SMOKE

G, Cooper BJ, Jelihovsky T, MacLennan R, Shearman R, Wang QS, Baines CJ, Miller AB, Wall C, Lund E, Stalsberg H, Shu XO, Zheng W, Katsouyanni K, Trichopoulou A, Trichopoulos D, Dabancens Martinez L, Molina R, Salas O, Alexander FE, Anderson K, Folsom AR, Hulka BS, Bernstein L, Enger S, Haile RW, Paganini-Hill A, Pike MC, Ross RK, Ursin G, Yu MC, Longnecker PM, Newcomb P, Bergkvist S, Kalache A, Farley TM, Holck S, Meirik O (2002) Alcohol, tobacco and breast cancer–collaborative reanalysis of individual data from 53 epidemiological studies, including 58,515 women with breast cancer and 95,067 women without the disease. Br. J. Cancer 87:1234–1245. Hammond SK (1999) Exposure of U.S. workers to environmental tobacco smoke. Environ. Health Perspect. 107:329–340. Hammond SK, Lewtas J, Mumford J, Henderson FW (1989) Exposures to environmental tobacco smoke in homes. Measurement of Toxic and Related Air Pollutants. Environmental Protection Agency/Air and Waste Management Association International Symposium, Raleigh, NC. Pittsburgh, PA:Air and Waste Management Association, pp.590–595. Hanrahan JP, Tager IB, Segal MR, Tosteson TD, Castile RG, Van Vunakis H, Weiss ST, Speizer FE (1992) The effect of maternal smoking during pregnancy on early infant lung function. Am. Rev. Respir. Dis. 145:1129–1135. Hargreave FE, Ryan G, Thomson NC, O’Bryne PM, Latimer K, Juniper EF, Dolovich J (1981) Bronchial responsiveness to histamine or methacholine in asthma: measurement and clinical significance. J. Allergy Clin. Immunol. 68:347–355. Harlap S, Davies AM (1974) Infant admissions to hospital and maternal smoking. Lancet 1:529–532. Harlos DP, Marbury M, Samet JM, Spengler JD (1987) Relating indoor NO2 levels to infant personal exposures. Atmos. Environ. 21:369–378. Hecht SS, Carmella SG, Murphy SE, Akerkar S, Brunneman KD, Hoffmann D (1993) A tobaccospecific lung carcinogen in the urine of men exposed to cigarette smoke. New Engl. J. Med. 329:1543–1546. Henderson FW, Reid HF, Morris R, Wang OL, Hu PC, Helms RW, Forehand L, Mumford J, Lewtas J, Haley NJ, Hammond SK (1989) Home air nicotine levels and urinary cotinine excretion in preschool children. Am. Rev. Respir. Dis. 140:197–201. Hirayama T (1981) Non-smoking wives of heavy smokers have a higher risk of lung cancer: a study from Japan. Br. Med. J. (Clin. Res. Ed.) 282:183–185. Hirayama T (1984) Cancer mortality in nonsmoking women with smoking husbands based on a largescale cohort study in Japan. Prev. Med. 13:680–690. Hole DJ, Gillis CR, Chopra C, Hawthorne VM (1989) Passive smoking and cardiorespiratory health in a general population in the west of Scotland. Bri. Med. J. 299:423–427. Howard G, Burke GL, Szklo M, Tell GS, Eckfeldt J, Evans G, Heiss G (1994) Active and passive smoking are associated with increased carotid wall thickness. The Atherosclerosis Risk in Communities Study. Arch. Intern. Med. 154:1277–1282. Howard G, Wagenknecht LE, Burke GL, Diez-Roux A, Evans GW, McGovern P, Nieto FJ, Tell GS (1998) Cigarette smoking and progression of atherosclerosis. The Atherosclerosis Risk in Communities (ARIC) Study. J. Am. Med. Assoc. 279:119–124. Howe GR, Burch JD, Chiarelli AM, Risch HA, Choi BC (1989) An exploratory case-control study of brain tumors in children. Cancer Res .49:4349–4352. Huel G, Godin J, Moreau T, Girard F, Sahuquillo J, Hellier G, Blot P (1989) Aryl hydrocarbon hydroxylase activity in human placenta of passive smokers. Environ. Res. 50:173–183. Idle JR (1990) Titrating exposure to tobacco smoke using cotinine—a minefield of misunderstandings. J. Clin. Epidemiol. 43:313–317. Infante-Rivard C, Krajinovic M, Labuda D, Sinnett D (2000) Parental smoking, CYP1A1 genetic polymorphisms and childhood leukemia (Quebec, Canada). Cancer Cause Control 11:547–553.

REFERENCES

747

International Agency for Research on Cancer (IARC).(1986) IARC Monographs on the Evaluation of the Carcinogenic Risk of Chemicals to Humans: Tobacco Smoking. [Monograph 38]. Lyon, France: World Health Organization, IARC. International Agency for Research on Cancer (IARC).(2004) Tobacco smoke and involuntary smoking. IARC [Monograph 83]. Lyon, France: International Agency for Research on Cancer. Jaakkola MS, Jaakkola JJK (1997) Assessment of exposure to environmental tobacco smoke. Eur. Respir. J. 10:2384–2397. Jaakkola MS, Jaakkola JJK, Becklake MR, Ernst P (1995) Passive smoking and evolution of lung function in young adults. An eight-year longitudinal study. J. Clin. Epidemiol. 48:317–327. Jaakkola MS, Jaakkola JJ, Becklake MR, Ernst P (1996) Effect of passive smoking on the development of respiratory symptoms in young adults: an 8-year longitudinal study. J. Clin. Epidemiol. 49:581–586. Jaakkola MS, Piipari R, Jaakkola N, Jaakkola JJ (2003) Environmental tobacco smoke and adult-onset asthma: a population-based incident case-control study. Am. J. Public Health 93:2055–2060. Jarvis MJ, Russell MA (1984) Measurement and estimation of smoke dosage to non-smokers from environmental tobacco smoke. Eur. J. Respir. Dis. (Suppl.) 13368–75. Jee SH, Lee JE, Kim S, Kim JH, Um SJ, Lee SJ, Namkoong SE, Park JS (2002) GSTP1 polymorphism, cigarette smoking and cervical cancer risk in Korean women. Yonsei Med. J. 43:712–716. Jenkins RA, Palausky A, Counts RW, Bayne CK, Dindal AB, Guerin MR (1996) Exposure to environmental tobacco smoke in sixteen cities in the United States as determined by personal breathing zone air sampling. J. Expo. Anal. Environ. Epidemiol. 6:473–502. Ji BT, Shu XO, Linet MS, Zheng W, Wacholder S, Gao YT, Ying DM, Jin F (1997) Paternal cigarette smoking and the risk of childhood cancer among offspring of nonsmoking mothers. J. Natl. Cancer Inst. 89:238–244. Jindal SK, Gupta D, Singh A (1994) Indices of morbidity and control of asthma in adult patients exposed to environmental tobacco smoke. Chest 106:746–749. Jindal SK, Jha LK, Gupta D (1999) Bronchial hyper-responsiveness of women with asthma exposed to environmental tobacco smoke. Indian J. Chest Dis. Allied Sci. 41:75–82. John EM, Savitz DA, Sandler DP (1991) Prenatal exposure to parents’ smoking and childhood cancer. Am. J. Epidemiol. 133:123–132. Junker MH, Danuser B, Monn C, Koller T (2001) Acute sensory responses of nonsmokers at very low environmental tobacco smoke concentrations in controlled laboratory settings. Environ. Health Perspect. 109:1045–1052. Kabat GC, Dieck GS, Wynder EL (1986) Bladder cancer in nonsmokers. Cancer 2:362–367. Kabat GC, Stellman SD, Wynder EL (1995) Relation between exposure to environmental tobacco smoke and lung cancer in lifetime nonsmokers (published erratum appears in Am J Epidemiol 1996 Mar 1; 143(5):527). Am. J. Epidemiol 142:141–148. Kalandidi A, Trichopoulos D, Hatzakis A, Tzannes S, Saracci R (1987) Passive smoking and chronic obstructive lung disease. Lancet 2:1325–1326. Kauffmann F, Tessier JF, Oriol P (1983) Adult passive smoking in the home environment: a risk factor for chronic airflow limitation. Am. J. Epidemiol. 117:269–280. Klepeis NE, Nelson WC, Ott WR, Robinson JP, Tsang AM, Switzer P, Behar JV, Hern SC, Engelmann WH (2001) The National Human Activity Pattern Survey (NHAPS): a resource for assessing exposure to environmental pollutants. J. Expo. Anal. Environ. Epidemiol. 11:231–252. Klonoff-Cohen HS, Edelstein SL, Lefkowitz ES, Srinivasan IP, Kaegi D, Chang JC, Wiley KJ (1995) The effect of passive smoking and tobacco exposure through breast milk on sudden infant death syndrome. JAMA 273:795–798. Kuijten RR, Bunin GR, Nass CC, Meadows AT (1990) Gestational and familial risk factors for childhood astrocytoma: results of a case-control study. Cancer Res .50:2608–2612.

748

SECONDHAND SMOKE

Kyerematen GA, Damiano MD, Dvorchik BH, Vesell ES (1982) Smoking-induced changes in nicotine disposition: application of a new HPLC assay for nicotine and its metabolites. Clin. Pharmacol. Ther. 32:769–780. Law MR, Morris JK, Wald NJ (1997) Environmental tobacco smoke exposure and ischaemic heart disease: an evaluation of the evidence. Brit. Med. J. 315:973–980. Leaderer BP, Hammond SK (1991) Evaluation of vapor-phase nicotine and respirable suspended particle mass as markers for environmental tobacco smoke. Environ. Sci. Technol. 25:770–777. Lee PN (1988) Misclassification of Smoking Habits and Passive Smoking. Berlin,Springer Verlag. Lee PN (1998) Difficulties in assessing the relationship between passive smoking and lung cancer. Stat. Methods Med. Res. 7:137–163. Lee PN, Chamberlain J, Alderson MR (1986) Relationship of passive smoking to risk of lung cancer and other smoking-associated diseases. Brit. J. Cancer 54:97–105. Leuenberger P, Schwartz J, Ackermann-Liebrich U, Blaser K, Bolognini G, Bongard JP, Brandli O, PBraun P, Bron C, Brutsche M, Domenighetti G, Elsasser S, Guldimann P, Hollenstein C, Hufschmid P, Karrer W, Keller R, Keller-Wossidlo H, Kunzli N, Luthi JC, Martin BW, Medici T, Perruchoud AP, Radaelli A, Schindler C, Schoeni MH, Solari G, Tschopp JM, Villiger B, Wuthrich BJ, Zellweger JP, Zemp E (1994) Passive smoking exposure in adults and chronic respiratory symptoms (SAPALDIA Study). Am. J. Respir. Crit. Care Med. 150:1222–1228. Li D, Wang M, Dhingra K, Hittelman WN (1996) Aromatic DNA adducts in adjacent tissues of breast cancer patients: clues to breast cancer etiology. Cancer Res .56:287–293. Lindbohm ML, Sallmen M, Hemminki K, Taskinen H (1991) Paternal occupational lead exposure and spontaneous abortion. Scan. J. Work Environ. Health 17:95–103. Lofroth G (1989) Environmental tobacco smoke: overview of chemical composition and genotoxic components. Mutat. Res. 222:73–80. Lubin JH (1999) Estimating lung cancer risk with exposure to environmental tobacco smoke. Environ. Health Persp. 107:879–883. Maclure M, Katz RB, Bryant MS, Skipper PL, Tannenbaum SR (1989) Elevated blood levels of carcinogens in passive smokersAm. J. Public Health 89:1381–1384. MacMahon B, Trichopoulos D, Cole P, Brown J (1982) Cigarette smoking and urinary estrogens. New Eng. J. Med. 307:1062–1065. Magnani C, Pastore G, Luzzatto L, Terracini B (1990) Parent at occupation and other environmental factors in the etiology of leukemia’s and non-Hodgkins lymphomas in childhood: a case-control study. Tumori 76:413–419. Mainous AG, Hueston WJ (1994) Passive smoke and low birth weight. Evidence of a threshold effect. Arch. Fam. Med. 3:875–878. Mannino DM, Siegel M, Rose D, Nkuchia J, Etzel R (1997) Environmental tobacco smoke exposure in the home and worksite and health effects in adults: results from the 1991 National Health Interview Survey. Tob. Control 6:296–305. Marbury MC, Hammond SK, Haley NJ (1993) Measuring exposure to environmental tobacco smoke in studies of acute health effects. Am. J. Epidemiol. 137:1089–1097. Martin TR, Bracken MB (1986) Association of low birth weight with passive smoke exposure in pregnancy. Am. J. Epidemiol. 124:633–642. Martinez FD, Wright AL, Taussig LM ,Group Health Medical Associates(1994) The effect of paternal smoking on the birthweight newborns whose mothers did not smoke. Am. J. Public Health 84:1489–1491. Masi MA, Hanley JA, Ernst P, Becklake MR (1988) Environmental exposure to tobacco smoke and lung function in young adults. Am. Rev. Respir. Dis. 88:296–299.

REFERENCES

749

Masjedi MR, Kazemi H, Johnson DC (1990) Effects of passive smoking on the pulmonary function of adults. Thorax 45:27–31. Matanoski G, Kanchanaraksa S, Lantry D, Chang Y (1995) Characteristics of nonsmoking women in NHANES I and NHANES I epidemiologic follow-up study with exposure to spouses who smoke. Am. J. Epidemiol. 142:149–157. Mattson ME, Boyd G, Byor D, Brown C, Callahan JF, Corle D, Cullen JW, Greenblatt J, Haley NJ, Hammond K, Lewtas J, Reeves W (1989) Passive smoking on commercial airline flights. J. Am. Med. Assoc. 261:867–872. Mau G, Netter P (1974) The effects of paternal cigarette smoking on perinatal mortality and the incidence of malformations. Dtsch. Med. Wochenschr. 99:1113–1118. Mauderly JL, Gigliotti AP, Barr EB, Bechtold WE, Belinsky SA, Hahn EF, Hobbs CA, March TH, Seilkop SK, Finch GL (2004) Chronic inhalation exposure to mainstream cigarette smoke increases lung and nasal tumor incidence in rats. Toxicol. Sci. 81:280–292. McCredie M, Maisonneuve P, Boyle P (1994) Antenatal risk factors for malignant brain tumors in New South Wales children. Int. J. Cancer 56:6–10. McElduff P, Dobson AJ, Jackson R, Beaglehole R, Heller RF, Lay-Yee R (1998) Coronary events and exposure to environmental tobacco smoke: a case-control study from Australia and New Zealand. Tob. Control 7:41–46. McLaughlin JK, Dietz MS, Mehl ES, Blot WJ (1987) Reliability of surrogate information on cigarette smoking by type of informant. Am. J. Epidemiol. 126:144–146. Menzies D, Nair A, Williamson PA, Schembri S, Al Khairalla MZ, Barnes M, Fardon TC, McFarlane L, Magee GJ, Lipworth BJ (2006) Respiratory symptoms, pulmonary function, and markers of inflammation among bar workers before and after a legislative ban on smoking in public places. J. Am. Med. Assoc. 296:1742–1748. Michnovicz JJ, Hershcopf RH, Naganuma H, Bradlow HL, Fishman J (1986) Increased 2-hydroxylation of estradiol as a possible mechanism for the anti-estrogenic effect of cigarette smoking. N. Engl. J. Med. 86:1305–1309. Miller GH (1984) Cancer, passive smoking and nonemployed and employed wives. West. J. Med. 140:632–635. Murray AB, Morrison BJ (1986) The effect of cigarette smoke from the mother on bronchial responsiveness and severity of symptoms in children with asthma. J. Allergy Clin. Immun. 77:575–581. Nagao M, Ushijima T, Wakabayashi K, Ochiai M, Kushida H, Sugimura T, Hasegawa R, Shirai T, Ito N (1994) Dietary carcinogens and mammary carcinogenesis. Induction of rat mammary carcinomas by administration of heterocyclic amines in cooked foods. Cancer 74:1063–1069. National Research Council (NRC) and Committee on Airliner Cabin Environment Safety Committee (1986) The Airliner Cabin Environment: Air Quality and Safety. Washington, DC:National Academy Press. National Research Council (NRC) and Committee on Passive Smoking (1986) Environmental Tobacco Smoke: Measuring Exposures and Assessing Health Effects. Washington, DC:National Academy Press. Norman MA, Holly EA, Ahn DK, Preston-Martin S, Mueller BA, Bracci PM (1996a) Prenatal exposure to tobacco smoke and childhood brain tumors: Results from the United States West Coast Childhood Brain Tumor Study. Cancer Epidem. Biomar. Prev. 5:127–133. Norman MA, Holly EA, Preston-Martin S (1996b) Childhood brain tumors and exposure to tobacco smoke Cancer Epidem. Biomar. Prev. 5:85–91. Nowak D, Jorres R, Martinez-Muller L, Grimminger F, Seeger W, Koops F, Magnussen H (1997) Effect of 3 hours of passive smoke exposure in the evening on inflammatory markers in bronchoalveolar and nasal lavage fluid in subjects with mild asthma. Int. Arch. Occup. Environ. Health 70:85–93.

750

SECONDHAND SMOKE

Nurminen MM, Jaakkola MS (2001) Mortality from occupational exposure to environmental tobacco smoke in Finland. J. Occup. Environ. Med. 43:687–693. Ostro BD (1989) Estimating the risks of smoking, air pollution, and passive smoke on acute respiratory conditions. Risk Anal .9:189–196. Ostro BD, Lipsett MJ, Mann JK, Wiener MB, Selner J (1994) Indoor air pollution and asthma. Results from a panel study. Am. J. Respir. Crit. Care Med. 149:1400–1406. Otsuka R, Watanabe H, Hirata K, Tokai K, Muro T, Yoshiyama M, Takeuchi K, Yoshikawa J (2001) Acute effects of passive smoking on the coronary circulation in healthy young adults. J. Am. Med. Assoc. 286:436–441. Ott WR (1999) Mathematical models for predicting indoor air quality from smoking activity. Environ. Health Perspect. 107:375–381. Petrakis NL, Gruenke LD, Beelen TC, Castagnoli N Jr, Craig JC (1978) Nicotine in breast fluid of nonlactating women. Science 199:303–305. Petrakis NL, Miike R, King EB, Lee L, Mason L, Chang-Lee B (1988) Association of breast fluid coloration with age, ethnicity, and cigarette smoking. Breast Cancer Res. Treat. 11:255–262. Pirkle JL, Flegal KM, Bernert JT, Brody DJ, Etzel RA, Maurer KR (1996) Exposure of the US population to environmental tobacco smoke: the third National Health and Nutrition Examination Survey 1988 to 1991. J. Am. Med. Assoc. 275:1233–1240. Pirkle JL, Bernert JT, Caudill SP, Sosnoff CS, Pechacek TF (2006) Trends in the exposure of nonsmokers in the U.S. population to secondhand smoke: 1988–2002. Environ. Health Perspect. 114:853–858. Preston-Martin S, Yu MC, Benton B, Henderson BE (1982) N-nitroso compounds and childhood brain tumors: a case-control study. Cancer Res .42:5240–5245. Pron GE, Burch JD, Howe GR, Miller AB (1988) The reliability of passive smoking histories reported in a case-control study of lung cancer. Am. J. Epidemiol. 127:267–273. Qin DX, Li JY, Sun LQ, Gu M, Wang GS, Cong JW, Wang JZ (1991) Sputum occult blood screening for lung cancer. Stage II screening of 14,431 subjects. Cancer 67:1960–1963. Rebagliato M, d Florey C, Bolumar F (1995) Exposure to environmental tobacco smoke in nonsmoking pregnant women in relation to birth weight. Am. J. Epidemiol. 142:531–537. Remijn B, Fischer P, Brunekreef B, Lebret E, Boleij JS, Noij D (1985) Indoor air pollution and its effect on pulmonary function of adult nonsmoking women. I. Exposure estimates for nitrogen dioxide and passive smoking. Int. J. Epidemiol. 14:215–220. Repace JL, Lowrey AH (1980) Indoor air pollution, tobacco smoke, and public health. Science 80:464–472. Repace JL, Lowrey AH (1990) Risk assessment methodologies for passive smoking-induced lung cancer. Risk Anal .10:27–37. Repace JL, Jinot J, Bayard S, Emmons K, Hammond SK (1998) Air nicotine and saliva cotinine as indicators of workplace passive smoking exposure and risk. Risk Anal .18:71–83. Reynolds P, Hurley S, Goldberg DE, Anton-Culver H, Bernstein L, Deapen D, Horn-Ross PL, Peel D, Pinder R, Ross RK, West D, Wright WE, Ziogas A (2004) Active smoking, household passive smoking, and breast cancer: evidence from the California Teachers Study. J. Natl. Cancer Inst. 96:29–37. Rice JM, Rehm S, Donovan PJ, Perantoni AO (1989)Comparative transplacental carcinogenesis by directly acting and metabolism-dependent alkylating agents in rodents and nonhuman primates. In:Napalkov NP, Rice JM, Tomatis L, Yamasaki H, editors. Perinatal and Multigeneration Carcinogenesis. Lyon, International Agency for Research on Cancer. p.17. Robbins AS, Abbey DE, Lebowitz MD (1993) Passive smoking and chronic respiratory disease symptoms in non-smoking adults. Int. J. Epidemiol. 22:809–817. Roquer JM, Figueras J, Botet F, Jimenez R (1995) Influence on fetal growth of exposure to tobacco smoke during pregnancy. Acta Paediatr .84:118–121.

REFERENCES

751

Rosenberg J, Benowitz NL, Jacob P, Wilson KM (1980) Disposition kinetics and effects of intravenous nicotine. Clin. Pharmacol. Ther. 28:517–522. Rubenstein D, Jesty J, Bluestein D (2004) Differences between mainstream and sidestream cigarette smoke extracts and nicotine in the activation of platelets under static and flow conditions. Circulation 109:78–83. Rubin DH, Krasilnikoff PA, Leventhal JM, Weile B, Berget A (1986) Effect of passive smoking on birth-weight. Lancet 2:415–417. Samet JM, Bohanon HR Jr, Coultas DB, Houston TP, Persily AK, Schoen LJ, Spengler J, Callaway CA, ASHRAE’s Environmental Tobacco Smoke Position Document Committee. (2005) ASHRAE position document on environmental tobacco smoke. Atlanta, GA, SHRAE. Sandler DP, Shore DL (1986) Quality of data on parents’ smoking and drinking provided by adult offspring. Am. J. Epidemiol. 124:768–778. Sandler DP, Everson RB, Wilcox AJ (1985a) Passive smoking in adulthood and cancer risk. Am. J. Epidemiol. 85:37–48. Sandler DP, Everson RB, Wilcox AJ, Browder JP (1985b) Cancer risk in adulthood from early life exposure to parents’ smoking. Am. J. Public Health. 74:487–492. Sandler DP, Wilcox AJ, Everson RB (1985c) Cumulative effects of lifetime passive smoking on cancer risk. Lancet 1:312–315. Sandler RS, Sandler DP, Comstock GW, Helsing KJ, Shore DL (1988) Cigarette smoking and the risk of colorectal cancer in women. J. Natl. Cancer Inst. 80:1329–1333. Sandler DP, Comstock GW, Helsing RJ, Shore DL (1989a) Deaths from all causes in non-smokers who lived with smokers. Am. J. Public Health 79:163–167. Sandler DP, Helsing KJ, Comstock GW, Shore DL (1989b) Factors associated with past household exposure to tobacco smoke. Am. J. Epidemiol. 129:380–387. Saracci R, Riboli E (1989) Passive smoking and lung cancer: current evidence and ongoing studies at the International Agency for Research on Cancer. Mutat. Res. 22:117–127. Sasco AJ, Vainio H (1999) From in utero and childhood exposure to parental smoking to childhood cancer: a possible link and the need for action. Hum. Exp. Toxicol. 18:192–201. Savitz DA, Schwingl PJ, Keels MA (1991) Influence of paternal age, smoking, and alcohol consumption on congenital anomalies. Teratology 44:429–440. Scholes D, McBride C, Grothaus L, Curry S, Albright J, Ludman E (1999) The association between cigarette smoking and low-grade cervical abnormalities in reproductive-age women. Cancer Cause Control 10:339–344. Schwartz J, Zeger S (1990) Passive smoking, air pollution, and acute respiratory symptoms in a diary study of student nurses. Am. Rev. Respir. Dis. 14:62–67. Scientific Committee on Tobacco and Health HSMO (1998) Report of the Scientific Committee on Tobacco and Health. 011322124x. The Stationary Office. Seersholm N, Hertz H, Olsen JH (1997) Cancer in the offspring of parents with lung cancer. Eur. J. Cancer 33:2376–2379. Seidman DS, Ever-Hadani P, Gale R (1990) Effect of maternal smoking and age on congenital anomalies. Obstet. Gynecol. 76:1046–1050. Shephard RJ, Collins R, Silverman F (1979) “Passive” exposure of asthmatic subjects to cigarette smoke. Environ. Res. 20:392–402. Shu XO, Ross JA, Pendergrass TW, Reaman GH, Lampkin B, Robison LL (1996) Parental alcohol consumption, cigarette smoking, and risk of infant leukemia: a Childrens Cancer Group study. J. Natl. Cancer Inst. 88:24–31. Sippel J, Pedula K, Vollmer W (1999) Associations of smoking with hospital based care and quality of life in patients with obstructive airway disease. Chest 115:691–696.

752

SECONDHAND SMOKE

Slattery ML, Robison LM, Schuman KL, French TK, Abbott TM, Overall JC Jr, Gardner JW (1989) Cigarette smoking and exposure to passive smoke are risk factors for cervical cancer. J. Am. Med. Assoc. 261:1593–1598. Slotkin TA (2004) Cholinergic systems in brain development and disruption by neurotoxicants: nicotine, environmental tobacco smoke, organophosphates. Toxicol. Appl. Pharmacol. 198:132– 151. Smith N, Austen J, Rolles CJ (1982) Tertiary smoking by the fetus (letter). Lancet 82:1252–1253. Sorahan T, Lancashire R, Prior P, Peck I, Stewart A (1995) Childhood cancer and parental use of alcohol and tobacco. Ann. Epidemiol. 5:354–359. Sorahan T, Lancashire RJ, Hulten MA, Peck I, Stewart AM (1997a) Childhood cancer and parental use of tobacco: deaths from 1953 to 1955. Br. J. Cancer 75:134–138. Sorahan T, Prior P, Lancashire RJ, Faux SP, Stewart AM, Peck IM (1997) Childhood cancer and parental use of tobacco: deaths from 1971 to 1976. Br. J. Cancer 76:1525–1531. Sorahan T, McKinney PA, Mann JR, Lancashire RJ, Stiller CA, Birch JM, Dodd HE, Cartwright RA (2001) Childhood cancer and parental use of tobacco: findings from the inter-regional epidemiological study of childhood cancer (IRESCC). Br. J Cancer 84:141–146. Spengler JD, Ferris BG Jr (1985) Harvard air pollution health study in six cities in the U.S.A. Tokai J. Exp. Clin. Med. 10:263–286. Spengler JD, Tosteson TD (1981) Personal Exposures to Respirable Particles. Alexandria, VA: Society for Industrial and Applied Mathematics. Spengler JD, Treitman RD, Tosteson T, Mage DT, Soczek ML (1985) Personal exposures to respirable particulates and implications for air pollution epidemiology. Environ. Sci. Technol. 19:700–707. Stankus RP, Menan PK, Rando RJ, Glindmeyer H, Salvaggio JE, Lehrer SB (1988) Cigarette smokesensitive asthma: challenge studies. J. Allergy Clin. Immunol. 82:331–338. Stavrou D, Dahme E, Haenichen T (1984)Transplacental carcinogenesis in the rabbit In:Schuller HM, editor. Comparative Perinatal Carcinogenesis. Boca Raton, FL,CRC Press. pp.23–43. Strachan DP, Cook DG (1997) Health effects of passive smoking. 1. Parental smoking and lower respiratory illness in infancy and early childhood. Thorax 52:905–914. Sumida H, Watanabe H, Kugiyama K, Ohgushi M, Matsumura T, Yasue H (1998) Does passive smoking impair endothelium-dependent coronary artery dilation in women?. J. Am. Coll. Cardiol. 31:811–815. Svendsen KH, Kuller LH, Martin MJ, Ockene JK (1987) Effects of passive smoking in the multiple risk factor intervention trial. Am. J. Epidemiol. 126:783–795. Tager IB, Segal MR, Speizer FE, Weiss ST (1988) The natural history of forced expiratory volumes. Effect of cigarette smoking and respiratory symptoms. Am. Rev. Respir. Dis. 138:837–849. Tarlo SM, Broder I, Corey P, Chan-Yeung M, Ferguson A, Becker A, Warren P, Simons FE, Sherlock C, Okada M, Manfreda J (2000) A case-control study of the role of cold symptoms and other historical triggering factors in asthma exacerbations. Can. Respir. J. 7:42–48. Taylor AE, Johnson DC, Kazemi H (1992) Environmental tobacco smoke and cardiovascular disease: a position paper from the council on cardiopulmonary and critical care, American Heart Association. Circulation 86:1–4. Tobacco Documents Online, 2006. Tredaniel J, Boffetta P, Little J, Saracci R, Hirsch A (1994) Exposure to passive smoking during pregnancy and childhood, and cancer risk: The epidemiological evidence. Paediatr. Perinat. Epidemiol. 8:233–255. Trichopoulos D, Kalandidi A, Sparros L, MacMahon B (1981) Lung cancer and passive smoking. Int. J. Cancer 27:1–4.

REFERENCES

753

Trichopoulos D, Kalandidi A, Sparros L (1983) Lung cancer and passive smoking: conclusion of Greek study. Lancet 2:677–678. Trichopoulos D, Mollo F, Tomatis L, Agapitos E, Delsedime L, Zavitsanos X, Kalandidi A, Katsouyanni K, Riboli E, Saracci R (1992) Active and passive smoking and pathological indicators of lung cancer risk in an autopsy study. J. Am. Med. Assoc. 268:1697–1701. Turner RB, Lande AE, Chase P, Hilton N, Weinberg D (1987) Pneumonia in pediatric outpatients: cause and clinical manifestations. J. Pediatr. 111:194–200. Urch RB, Silverman F, Corey P, Shephard RJ, Cole P, Goldsmith LJ (1988) Does suggestibility modify acute reactions to passive cigarette smoke exposure? Environ. Res. 47:34–47. U.S. Department of Health and Human Services (USDHHS) (1980) The health consequences of smoking for women: a report of the Surgeon General. U.S. Department of Health and Human Services, Public Health Service, Office of the Assistant Secretary for Health, Office of Smoking and Health. U.S. Department of Health and Human Services (USDHHS) (1982) The Health Consequences of Smoking: Cancer. A Report of the Surgeon General. DHHS Publication No. (PHS) 82-50179. Washington, DC: U.S. Department of Health and Human Services, Public Health Service, Office on Smoking and Health. U.S. Department of Health and Human Services (USDHHS) (1984) The Health Consequences of Smoking: Chronic Obstructive Lung Disease. A Report of the Surgeon General, Washington, DC: U.S. Government Printing Office. U.S. Department of Health and Human Services (USDHHS) (1986) The Health Consequences of Involuntary Smoking: A Report of the Surgeon General. DHHS Publication No. (CDC) 87-8398. Washington, DC: U.S. Government Printing Office. U.S. Department of Health and Human Services (USDHHS) (1988) The Health Consequences of Smoking: Nicotine Addiction. A Report of the Surgeon General, Washington, DC: U.S. Government Printing Office. U.S. Department of Health and Human Services (USDHHS) (1989) Reducing the Health Consequences of Smoking: 25 years of Progress. A Report of the Surgeon General, Washington, DC: U.S. Government Printing Office. U.S. Department of Health and Human Services (USDHHS) (1990a) Smoking, Tobacco, and Cancer Program, 1985-1989 Status Report. NIH Publication No. 90-3107. Washington, DC: U.S. Government Printing Office. U.S. Department of Health and Human Services (USDHHS) (1990b) The health Benefits of Smoking Cessation. A Report of the Surgeon General. DHHS Publication Number 90-8416. Washington, DC: U.S. Government Printing Office. U.S. Department of Health and Human Services (USDHHS) (2001) Women and Smoking. A Report of the Surgeon General. Rockville, MD: U.S. Department of Health and Human Services. U.S. Department of Health and Human Services (USDHHS) (2004) The Health Effects of Active Smoking: A Report of the Surgeon General. Washington, DC: U.S. Government Printing Office. U.S. Department of Health and Human Services (USDHHS) (2006) The Health Effects of Involuntary Exposure to Tobacco Smoke. Rockville, MD: Centers for Disease Control and Prevention (CDC). U.S. Environmental Protection Agency (EPA) (1992) Respiratory Health Effects of Passive Smoking: Lung Cancer and Other Disorders. EPA/600/006F. Washington, DC: U.S. Government Printing Office. Vandenbroucke JP, Verheesen JH, De Bruin A, Mauritz BJ, van der Heide-Wessel C, van der Heide RM (1984) Active and passive smoking in married couples: results of 25 year follow up. Br. Med. J. (Clin. Res. Ed.) 84:1801–1802. Vardavas CI, Linardakis M, Kafatos AG (2006) Environmental tobacco smoke exposure in motor vehicles: a preliminary study. Tob. Control 15:415.

754

SECONDHAND SMOKE

Vessey MP (1989) Epidemiological studies of the effects of diethylstilbestrol. In:Napalkov NP, Rice JM, Tomatis L, Yamasaki H, (editors) Perinatal and Multigeneration Carcinogenesis. Lyon, International Agency for Research on Cancer. pp.335–348. Vineis P, Alavanja M, Buffler P, Fontham E, Franceschi S, Gao YT, Gupta PC, Hackshaw A, Matos E, Samet J, Sitas F, Smith J, Stayner L, Straif K, Thun MJ, Wichmann HE, Wu AH, Zaridze D, Peto R, Doll R (2004) Tobacco and cancer: recent epidemiological evidence. J Natl. Cancer Inst. 96:99–106. Wald NJ, Nanchahal K, Thompson SG, Cuckle HS (1986) Does breathing other people’s tobacco smoke cause lung cancer?Br. Med. J. (Clin. Res. Ed.) 86:1217–1222. Wall MA, Johnson J, Jacob P, Benowitz NL (1988) Cotinine in the serum, saliva, and urine of nonsmokers, passive smokers, and active smokers. Am. J. Public Health 78:699–701. Wallace LA, Pellizzari ED (1987) Personal air exposures and breath concentrations of benzene and other volatile hydrocarbons for smokers and nonsmokers. Toxicol. Lett. 35:113–116. Ware JH, Dockery DW, Spiro A III (1984) Passive smoking, gas cooking, and respiratory health of children living in six cities. Am. Rev. Respir. Dis. 129:366–374. Weiss ST (1986) Passive smoking and lung cancer. What is the risk?Am Rev. Respir. Dis. 133:1–3. Weiss SJ (1989) Tissue destruction by neutrophils. N. Engl. J. Med. 320:365–376. Weiss ST, Tager IB, Speizer FE, Rosner B (1980) Persistent wheeze: its relation to respiratory illness, cigarette smoking, and level of pulmonary function in a population sample of children. Am Rev. Respir. Dis. 122:697–707. Weiss ST, Utell MJ, Samet JM (1999) Environmental tobacco smoke exposure and asthma in adults. Environ. Health Perspect. 107 (Suppl 6):891–895. Wells AJ (1988) An estimate of adult mortality in the United States from passive smoking. Environ. Int. 14:249–265. Whincup PH, Gilg JA, Emberson JR, Jarvis MJ, Feyerabend C, Bryant A, Walker M, Cook DG (2004) Passive smoking and risk of coronary heart disease and stroke: prospective study with cotinine measurement. Br. Med. J. 329:200–205. White JR, Froeb HF (1980) Small-airways dysfunction in nonsmokers chronically exposed to tobacco smoke. N. Engl. J. Med. 302:720–723. Windham GC, Eaton A, Hopkins B (1999) Evidence for an association between environmental tobacco smoke exposure and birthweight: a meta-analysis and new data. Paediatr. Perinat. Epidemiol. 13:35–37. World Health Organization (1999) International Consultation on Environmental Tobacco Smoke (ETS) and Child Health. Consultation Report. Geneva, OH: World Health Organization. Wu AH (1999) Exposure misclassification bias in studies of environmental tobacco smoke and lung cancer. Environ. Health Perspect. 107:873–877. Wu MT, Lee LH, Ho CK, Liu CL, Wu TN, Wu SC, Lin LY, Cheng BH, Yang CY (2003) Lifetime exposure to environmental tobacco smoke and cervical intraepithelial neoplasms among nonsmoking Taiwanese women. Arch. Environ. Health 58:353–359. You RX, Thrift AG, McNeil JJ, Davis SM, Donnan GA (1999) Ischemic stroke risk and passive exposure to spouses’ cigarette smoking. Melbourne Stroke Risk Factor Study (MERFS) Group. Am. J Public Health 89:572–575. Yu MC, Garabrant DH, Huang TB, Henderson BE (1990) Occupational and other non-dietary risk factors for nasopharyngeal carcinoma in Guangzhou, China. Int. J. Cancer 45:1033–1039. Yuan JM, Wang XL, Xiang YB, Gao YT, Ross RK, Yu MC (2000) Non-dietary risk factors for nasopharyngeal carcinoma in Shanghai, China. Int. J. Cancer 85:364–369. Zeegers MP, Goldbohm RA, van den Brandt PA (2002) A prospective study on active and environmental tobacco smoking and bladder cancer risk (The Netherlands). Cancer Causes Control 13:83–90.

REFERENCES

755

Zhang J, Savitz DA, Schwingl PJ, Cai WW (1992) A case-control study of paternal smoking and birth defects. Int. J. Epidemiol. 21:273–278. Zhang X, Shu XO, Yang G, Li HL, Xiang YB, Gao YT, Li Q, Zheng W (2005) Association of passive smoking by husbands with prevalence of stroke among Chinese women nonsmokers. Am. J. Epidemiol. 161:213–218. Zheng W, McLaughlin JK, Chow WH, Chien HT, Blot WJ (1993) Risk factors for cancers of the nasal cavity and paranasal sinuses among white men in the United States. Am. J. Epidemiol. 138:965– 972. Zong L, Goldberg MS, Parent ME, Hanley JA (2000) Exposure to environmental tobacco smoke and the risk of lung cancer: a meta-analyses. Lung Cancer 27:3–18.

20 LEAD AND COMPOUNDS Lester D. Grant*

20.1 INTRODUCTION Lead (Pb), a heavy metal with numerous useful properties (low melting point, highly malleable, etc.), has been put to many diverse uses by both ancient civilizations (e.g., the Roman Empire) and by modern societies. The extensive expanded use of the metal in modern times (in water distribution systems, in additives to paints and gasoline, for electronics applications, etc.) caused widespread increases in lead exposures for human populations around the world, especially during the twentieth century. Despite its broad usefulness, however, the metal has also been long recognized to be acutely toxic at high-dose exposure (e.g., see Aub et al., 1925; Beechx, 1986). Diagnosis of classically defined acute lead poisoning (often life-threatening) historically typically involved clinical observation in individual medical cases of signs and symptoms of (a) marked impairment of red blood cell formation/function; (b) severe central and/or peripheral nervous system (PNS) damage/functional impairment; and/or (c) notable kidney damage/renal dysfunction. Very importantly, however, extensive new research findings emerging since the 1970s (comprising major advances in our understanding of lead toxicity) indicate that lead can exert toxic effects of concern on many organ systems and at exposure levels far lower than those producing clinically evident signs and symptoms of overt lead intoxication. Such “subclinical” lead toxicity effects are not identifiable through routine clinical examination of any one patient at one point in time, but rather their characterization generally requires more sophisticated methods, for example, identification through longitudinal observation. Also, although many such subclinical effects may be very subtle, their ultimate impact on a population basis can be substantial, as illustrated both for lead (Eckerman et al., 1999) and for other chemicals (Lester et al., 1998). *

Formerly Director (now retired), National Center for Environmental Assessment–Research Triangle Park Division (NCEA-RTP), U.S. Environmental Protection Agency (U.S. EPA). The contents of this chapter have not undergone review or clearance by U.S. EPA and should not be taken to represent views or official positions of the U.S. EPA.

Environmental Toxicants, Third Edition Edited by Morton Lippmann Copyright
2009 John Wiley & Sons, Inc.

757

758

LEAD AND COMPOUNDS

Demonstration during the past 30–40 years of “subclinical” toxic effects at lower and lower exposure levels (and recognition of their potential public health implications) has led to repeated downward revision over the same time period of views regarding lead exposure levels seen as posing unacceptable risks for adverse human health impacts. Stimulated by such changes in viewpoints regarding “safe” lead exposure levels, lead regulatory guidelines/standards have been successively tightened and extensive remedial actions implemented in many parts of the world, producing notable reductions in lead exposure among many population groups. Thus, although clinically evident acute lead intoxication due to high-dose exposures continues to be of much concern in certain geographic areas, subclinical lead toxicity has come to be of much broader public health interest. Accordingly, greater attention is devoted here to lower level lead exposure effects. Also discussed here are factors affecting susceptibility/vulnerability to such effects, some hypotheses and evidence regarding potential mechanisms of action, and progress in developing biokinetic models used to predict likely increased risk of lead-related toxicity among human population groups due to various lead exposure scenarios. The remarkable advances made in our knowledge of lead toxicity during the past 30–40 years have been described in numerous reviews published during the last several decades. The prior edition of this chapter, for example, noted those by Davis and Svendzgaard (1987) and Lippmann (1990), and a number of newer, more recent reviews are cited at various points later in this chapter. Major advances made via evolving lead research during the past 4–5 decades are especially well reflected by the series of intensively peer-reviewed, periodic assessments of the latest available lead-related information that comprises U.S. Environmental Protection Agency (U.S. EPA) Air Quality Criteria for Lead documents and associated Addendum/Supplement materials (U.S. EPA, 1977, 1986a, 1986b, 1990a, 2006) generated to support the setting and periodic review of U.S. National Ambient Air Quality Standards (NAAQS) for Lead. The reader is referred to such EPA documents and other reviews for more extensive, detailed discussions of particular topics. It should also be noted that certain important points made in older classic literature, as well as advances made in our knowledge of lead toxicology through the late 1990s/early 2000s, were delineated well in the previous edition of this chapter (by Mahaffey, McKinney, and Reigart), the contents of which still largely remain germane today. Extensive portions of text from that edition are, therefore, retained here largely intact or with minimal revision. Other text has been updated to reflect more recent findings on a given topic or to provide information on additional topics beyond those covered in the prior edition, with the most recent EPA assessment (U.S. EPA, 2006) of lead exposure and toxicity information being drawn upon as a particularly cogent, readily accessible authoritative source.

20.2 PHYSICAL/CHEMICAL PROPERTIES AND BEHAVIOR OF LEAD AND ITS COMPOUNDS Lead is a member of subgroup IVA of the periodic table and is a typical heavy metal (Greninger et al., 1978) with a relatively high atomic weight. The valence shell of the lead atom in the ground state has two s and two p electrons. Because the element has four electrons in its outer shell, lead would be expected to show a normal valence þ4 in its compounds. The two s electrons of lead do not readily ionize and are thus often referred to as the inert pair. Lead is considered to have a stable oxidation state (Pb2þ) that furnishes a divalent ion. The metabolism of lead in a redox sense does not appear to be particularly important

PHYSICAL/CHEMICAL PROPERTIES AND BEHAVIOR OF LEAD AND ITS COMPOUNDS

759

in determining its biological properties. Bivalent lead has a notable tendency to form well-delineated, often highly crystalline basic salts of both anhydrous and hydrated types, for example, white lead (a pigment of past widespread commercial use). In general, the inorganic chemistry of bivalent lead resembles that of alkaline earth elements. Several lead salts (e.g., lead carbonate, nitrate, and sulfate) are isomorphous with corresponding strontium and barium compounds. Lead forms highly insoluble salts of phosphate, carbonate, and sulfide. Lead can also form salts with organic acids, which is the basis for the use of certain chelating agents for treating lead intoxication. Elemental lead would be readily oxidized in biological systems and is thus not considered as a separate form here. Lead’s position in the periodic chart favors formation of covalent rather than ionic bonds in Pb4þ compounds. This expectation is confirmed by the properties of compounds such as lead tetrachloride and tetraacetate. Predominately covalent bonding is also seen with organolead compounds (with up to four Pb C bonds). An organometallic compound differs fundamentally in both chemical and biological properties from an ionic compound of the same metal. Thus, determining only the total amount of the metal in a biological sample can be very misleading with regard to estimating the potential for toxic effects. In general, inorganic lead has been more extensively studied than organometallic lead. Speciation and movement of lead and other heavy metals in the environment were extensively discussed at a mid-1980s international conference (Landner, 1987), and the biological effects of organolead compounds were reviewed by Grandjean and Grandjean (1984). An appreciation of some general features of metal chemistry is very helpful in understanding specific aspects of lead chemistry (Hanzlik, 1981). Among the most important criteria differentiating metal ions from each other, and from electrophilic organic species, is the chemistry of their bonding to biological ligands. Metal–ligand bonds can be as strong, in a thermodynamic sense, as bonds formed when a reactive epoxide alkylates a nucleophilic group in DNA or protein. Regardless of the mechanism by which a metal ion enters a biological system, complexation undoubtedly plays a role in both its distribution within and elimination from the organism. Metals’ ions are Lewis acids, and one very important determinant of their affinity for ligands is their charge/radius ratios. Increasing the metal’s oxidation state increases its Lewis acidity and its affinity for a given ligand (assuming that it does not ionize the ligand). In an antagonist sense, the relative size of the ion can also be important. In addition to the energetics of the ligation of ions, the energetics of the ionization process itself must also be considered. In this case, complex geometry and ligand exchange rate play important roles. Most metal complexes undergo ligand exchange by processes that involve a dissociative rate-limiting step analogous to the Sn1 solvolysis of alkyl halides. For a given metal, the rates are rather sensitive to the nature of the departing ligand and are essentially independent of the entering group. The dependence of biological activity on ligand exchange rates reflects the fact that the complex must be sufficiently inert to survive long enough in vivo to reach critical reactive target molecules and yet, once having reached those sites, be sufficiently labile to react. A given metal-macromolecular interaction may persist enough to have biological consequences, if the equilibrium constant of the rate for dissociation of the complex is very small. The hard–soft, acid–base dichotomy provides a rationale underlying many features of the behavior of metal systems in chemistry and biology. This parameter is correlated qualitatively with the charge–size ratio of the ion in that large ions of low ionic charge have easily polarizable or deformable (i.e., soft) electrostatic fields about them, whereas small highly

760

LEAD AND COMPOUNDS

charged ions with relatively intense electrostatic fields are hard. Flexibility in hard–soft ligand preferences appears to be a key property underlying the biological activity/toxicity of metals in living systems. Several toxic heavy metal ions (i.e., Pb2þ, Hg2þ, Th1þ, Ni2þ, and Sb3þ) are classified as soft or have borderline properties in this classification scheme. The ligands, analogously, can also be classified by these hard–soft criteria. In this regard, it is not surprising that the rate of transport of a given metal ion across model liquid membranes can be varied by several orders of magnitude simply by altering the anion (ligand) present in the original salt solution (Christensen et al., 1978). Like the other group IVA metals, tin and germanium, lead forms complexes in which the donor atom is chiefly oxygen (Greninger et al., 1978). It also forms stable complexes with sulfur and halogens as the donor. Carbon and nitrogen donors are less common. Lead generally forms complexes with the coordination number six (having octahedral geometrical structure), whereas other geometries are less common. On the basis that hard metal ions prefer to bind ligands and vice versa, one might expect lead to bind halides in the order I 4 Br 4 Cl 4 F. This is consistent with the early use of potassium iodide to enhance lead removal from the body (Aub et al., 1925). Another important aspect of metal chemistry is the potential for metal compounds to act as initiators or catalysts in vivo (Hanzlik, 1981). It is not difficult to envision that inhibition of enzyme molecules by stoichiometric quantities of tightly bound metal ions could reduce the flow of vital metabolites through a pathway and, thus, cause toxicity. In addition to stimulating or inhibiting the synthesis of enzymes, as well as the enzymes’ activities, many simple metal ions and compounds have catalytic activity in their own right. Of much importance here are the electrochemical gradients across biological membranes and the potential of a foreign metal ion to act as an “antimetabolite.” This may be significant in view of possible existence of a mechanism for coupling biological oxidation–reduction pathways to ion transport and the control of membrane potential. In many cases, apparently nonessential metals are absorbed into an organism and not excreted at all; rather, they are simply concentrated and deposited in granular, insoluble complexes with or without accompanying proteinaceous material. There are several ways to express the relationship of lead to other metals, both foreign and endogenous. For example, the resemblance of bivalent lead chemistry to that of the alkaline earth metals in general was mentioned earlier. Of particular note is the similarity to calcium. Lead and calcium both form insoluble carbonates and insoluble phosphates. However, lead phosphate is much more insoluble than lead carbonate, whereas calcium phosphate is more soluble than calcium carbonate. Lead phosphate is one of the few insoluble phosphates that do not react with most chemical reagents. The extreme insolubility of lead phosphate may serve as a driving force for lead to function as a phosphate scavenger in biological systems, which may include inorganic forms of phosphate as well as the various important phosphate esters. The ultimate deposition of lead in the skeleton is consistent with lead’s chemical relationship to calcium and the formation of highly insoluble salts (Aub et al., 1925). Strontium, another alkaline earth metal, can also compete with calcium in bone tissue (Smith et al., 1985), and this chemical similarity of lead and strontium may be related to their ionic radii and the stable 2þ oxidation state relative to that of Ca2þ. Intestinal calcium-binding proteins have been shown to bind lead with high affinities and in preference to calcium (Fullmer et al., 1985). These proteins bind several other cations (notably Sr2þ, Ba2þ, and Cd2þ) in a fashion apparently related to metal ionic radii relative to calcium. Another important factor that can determine metal chemistry and biology is the hard–soft, acid–base property (amphoteric nature). In this regard, lead is similar to iron, copper, zinc,

LEAD IN THE ENVIRONMENT AND HUMAN EXPOSURE

761

mercury, and thallium, among others. The amphoteric property, along with redox cycling, appears to account, in large part, for the importance of iron in biological systems and its adsorption, storage, and transfer in these systems (Hanzlik, 1981). Such effects are attributable to the greater polarizabilities (function of number of electrons) of these cations and subsequently greater covalent character of the bonds they form with donor ligands relative to that expected for alkaline earth cations of the same size. The situation with Pb2þ is very analogous to that of Th1þ (Izatt et al., 1976). Similarities in size and coordination chemistry may be important factors that determine the ability of metals to act antagonistically (Hill and Matrone, 1970). The chemical similarity of lead to certain alkaline earth metals (particularly calcium) and lead’s ability to form highly insoluble salts (particularly of carbonate and phosphate) along with increased affinities to biological donors (enriched in oxygen and possibly nitrogen) due to favorable polarizabilities may well account for much of the relevant biological/toxicological chemistry of lead compounds. In an overall sense, the importance of the ligand exchange chemistry of divalent lead is emphasized in the expression of toxicity. Ligands can include simple anions or more complex donors that can form chelates or organic complexes. The biological activity of a given metal is a consequence of the way in which the metal’s compounds (salts, complexes, etc.) and cells interact. It is this interaction that is governed by intrinsic chemical properties (modulated by certain physical properties) of both the particular metal compound and the cell. In addition, the extent of cellular interaction can be affected by the same or different chemical properties that determine the in vivo absorption, distribution, and elimination of the compounds. Importantly, lead forms highly stable bonds with sulfur and sulfur-containing compounds, but somewhat less stable ones with carboxylic acids (O-based ligands) and imidazoles (N-based ligands) (Claudio et al., 2003), as noted in U.S. EPA (2006). Also, as noted there, lead competes very effectively in biological systems with native or homeostatic metal ions for binding with sulfhydral, carboxyl, and imidazole side chains that comprise enzyme active sites, and this competition leads to inhibition of enzyme activity, the replacement of calcium in bone, and many other deleterious health impacts. Other key features of lead coordination chemistry, their roles in biological systems, and relationships to lead-induced adverse health effects are delineated in the review by Claudio et al. (2003).

20.3 LEAD IN THE ENVIRONMENT AND HUMAN EXPOSURE The previous edition of this chapter noted that trace metals, such as lead, can be present in the environment in various forms (Boline, 1981), such as free hydrated ions; ion pair salts/ complexes; organic complexes/chelates; surface-adsorbed material; and undissolved compound. Although differences may occur in valence states and associated ligands (including mixed ligands) across these various forms, the metal identity is still retained, and the chemical reactivity of the metal is a function of the combined physicochemical properties of the metal and its associated ligands. This chemical reactivity can be modulated by physical properties, such as the surface properties of the metal compound itself. Thus, under certain physical, chemical, and biological conditions, it is possible for a given metal to assume more than one form, which can follow new pathways of chemical reactivity. This same reactive potential contributes to the posing by certain metals of possible toxic threats to the environment and living systems that concentrate them. The range of chemical properties and reactivities associated with various types of metal compounds is thus

762

LEAD AND COMPOUNDS

much greater than that of simple organic compounds. Since the metal is never destroyed, the potential to exert this complex chemistry always exists. Inorganic lead compounds to which humans are likely to be exposed include halides and oxides, sulfides and sulfate, carbonate, and chromate. With the exception of a few sporadic measurements in air, marine fish, sediments, birds, and in human brains, there is relatively little information available on organic lead compounds (Jawrerski et al., 1987). It appears that most organic lead compounds in the environment come from the release of organolead compounds (such as those used as gasoline additives) prior to or during their use rather than being derived from inorganic compounds of lead. In the expanding number of areas restricting the use of lead-based gasoline additives in the past 20–30 years, an increasingly greater proportion of inorganic lead compounds have come to dominate. Thus, the global movement of inorganic lead and its compounds, as well as human exposure to them, are of much greater concern, have been far more extensively studied, and are the main focus here. Calculations of loading rates of lead and other trace metals into various environmental compartments indicate that human activities exert major impacts on global and regional cycles of most trace elements (Nriagu and Pacyna, 1988). Other environmental problems, such as acidic precipitation (Mohnen, 1988), have contributed to mobilization and distribution of metals in the environment as well. The greatly increased circulation of toxic metals through the air, soil, and water, and their ultimate transfer into the human food chain remain important environmental issues and likely entail unknown health risks for future generations. The biogeochemical cycling of lead and routes of human exposure were described earlier by Schlag (1987) and have been more recently extensively discussed in U.S. EPA (2006). Key types of environmental reservoirs of lead can be identified and quantitative estimations made of inputs to them from various natural or human sources in a reasonably straightforward manner, but rates of transfer within and between the reservoirs are generally known only qualitatively or semiquantitatively. For example, as noted in the prior edition of this chapter, various estimates of natural and anthropogenic lead emissions to air and to the oceans have been made on a global basis (e.g., by Nriagu and Pacyna, 1988), and all such estimates indicate that contributions from anthropogenic sources are at least one to two orders of magnitude greater than from the natural ones. Detectable long-term elevations in global lead emissions can be seen starting as of the time of the Roman Empire, followed by a relatively slow increase over many centuries and then a steep rise starting in the eighteenth century, which peaked at around 400,000 tons per year during 1970–1980 and has since declined to about 100,000 tons per year (Nriagu, 1998). This temporal pattern of changing worldwide lead emissions is reflected well by examination of natural historical records of lead accumulated over time in ice packs or other layered natural materials, with analogous peaking in the 1970–1980 period being followed by declining lead levels. For example, the pattern of atmospheric lead deposition in a peat bog in the Swiss Jura Mountains parallels well historical increases in deposition of lead attributable to the introduction in 1947 of leaded gasoline into Switzerland, with later bog layer samples corresponding to 1991 showing a decline in atmospheric lead deposition and a shift in isotopic ratios toward radiogenic values (Shotyk et al., 1998) reflective of subsequent phasing out of leaded gasoline in Europe. Analogous patterns of variations in sediment lead levels (i.e., reaching peak concentrations in layers reflecting highest deposition during the 1970s, followed by ensuing declines) have been observed for a variety of other locations in North America and Europe. For more detailed reviews of natural historical records reflecting temporal

LEAD IN THE ENVIRONMENT AND HUMAN EXPOSURE

763

variations in lead emissions and deposition, see Boutron et al. (1994), Weiss et al. (1999), and Garty (2001). Human lead exposure generally occurs via one or more of four main components of the human environment: inhaled air, soil and dust of various types, drinking water, and food. The prior edition of this chapter noted that the primary medium for widespread dispersal of lead in the ambient (outdoor) environment tends to be air, because lead-containing fine particles (emitted mainly by high-temperature anthropogenic sources) can travel long distances before settling out via wet, dry, or cloud deposition. Lead deposition from air is most intense near a given source, but the zone of readily detectable elevated deposition can extend some distance (even many kilometers) away. Substantial decreases in ambient airborne lead levels have occurred in parallel with the phase down in usage of organic lead compounds as additives in gasoline in many countries, for example, more than 90% decrease in U.S. ambient air lead levels from the mid-1970s to more current urban air concentrations typically falling in the 0.10– 0.25 mg/m3 range by 2000–2002, as noted in U.S. EPA (2006). However, accumulation of lead in roadside soil and other soils due to past deposition of airborne lead from gasoline, smelters, and other sources constitutes a persisting reservoir of anthropogenically generated lead that will likely continue for many decades to contribute to human exposure and health risks. This is partly due to the fact that most lead particles deposited on soil are retained and, eventually, are mixed into the surface layer, with the lead accumulated at the soil surface becoming available to be taken up by plants, grazing animals, or soil microorganisms and, thereby, enter terrestrial food chains. Also, very importantly, direct exposure of children to lead in soil can occur by oral intake of lead-contaminated dust and dirt during normal hand-to-mouth activity. In addition, lead-contaminated soil particles can be resuspended in air and as such become a long-term source of airborne lead exposure for humans via the inhalation route as well. Lead in rivers comes from runoff, erosion, and direct deposition from air. Freshwater generally contains more inorganic and organic suspended particulate material than marine water, and this suspended material has a strong tendency to adsorb any dissolved lead, with lead adsorption potential being much greater for smaller size particles (due to their notably greater surface area per unit weight of lead) than for larger particles (Rhoads and Cahill, 1999). Lead adsorption into sediments can be enhanced by the presence of various substances. For example, lead concentrations in sediment typically increase with humic (organic matter) content (Kiratli and Ergin, 1996; Rhoads and Cahill, 1999), and lead in sediments can be sequestered on iron or manganese oxides (Schintu et al., 1991; Peltier et al., 2003; Gallon et al., 2004) or its adsorption increased by sulfides, especially under anoxic conditions (Kiratli and Ergin, 1996; Perkins et al., 2000). Most of the lead entering the open oceans comes from atmospheric deposition rather than from rivers. The lower concentrations of particulate matter in marine waters, as well as high salt (chloride, bromide, etc.) levels, tend to favor a larger proportion of dissolved lead in the marine water column. Deep ocean sediments may thereby represent a sink for lead, since there is little evidence to suggest notable remobilization from such sediments. On the other hand, it appears that lead in sediments in freshwater lakes and rivers can be remobilized (e.g., see Steding et al., 2000; Hlavay et al., 2001; Peltier et al., 2003; Kurkjian et al., 2004; Gallon et al., 2004), especially under acidic conditions. Thus, in some areas, lead in freshwater sediments may continue to be a potential source of water-related exposure to lead deposited many years earlier from the air, via runoff, and so on. Lead in drinking water supplied by municipal water distribution systems typically derives mainly from corrosion of lead pipes, lead-based solder, or bronze and brass fixtures

764

LEAD AND COMPOUNDS

(e.g., faucets) used as part of plumbing within residences or workplaces, with little lead generally coming from properly buffered utility supplies (Lee et al., 1989; Singley, 1994; Gulson et al., 1994; Isaac et al., 1997). Low pH (acidic) water enhances the leaching of lead from indoor plumbing components, which can be greatly reduced by buffering utility water supplies toward neutral pH conditions. Of much importance, the addition for disinfection purposes of chlorine to drinking water (which can lower the pH) does not generally increase leaching of lead into the water, because of a chlorine reaction with Pb2þ that results in precipitating out of a highly insoluble, red-brown colored lead solid (Edwards and Dudi, 2004). On the other hand, the insoluble lead solid is not formed in the presence of chloramines, and the introduction of chloramine disinfectants in place of chlorine (especially in the absence of other adequate water chemistry adjustments) appears to have contributed to increases in drinking water lead levels seen in some U.S. communities since 2000, as noted in U.S. EPA (2006). Potential human exposure via drinking of lead-contaminated water has received much attention in the past and can still be a nontrivial contributor to lead exposures for some populations. As an example, tap water lead was the main correlate of elevations in maternal blood lead levels in a study of mothers and infants in Glasgow, Scotland (Watt et al., 1996). In a U.S. prospective study, Lanphear et al. (2002) found that children exposed to water with lead levels over 5 ppb had blood lead levels 1.0 mg/dL higher than children exposed to water with lead concentrations less than 5 ppb. Although exposure to lead via drinking water undoubtedly still occurs, it is often difficult to readily determine the importance of this exposure route in contributing to any specific overall toxic insult, due in part to the considerable potential for wide variations in the concentration and bioavailability of lead in water. Furthermore, the bioavailability of lead and other metals in water not only depends on trace metal solubility but also on numerous complex chemical equilibria affected by the presence of other trace inorganic and organic compounds in the water (Jackson and Sheiham, 1986). The bioavailability of lead in soils, food, and inhaled air may depend on similar factors that determine the ligand exchange chemistry once in contact with the biota and aqueous phase. Accordingly, metal speciation analyses and solubility modeling are likely to yield further insights and improved understanding of the potential for toxic insult (Hunt and Creasey, 1980) and, thereby, contribute to improved scientific bases for abatement strategies for lead and other metal pollutants (e.g., reducing the bioavailability of trace metals in water via manipulation of their solubility and aqueous chemistry, as suggested by French and Hunt, 1988). Lead in food derives from plant and animal exposure to contaminated air, soil, and water, and from products used in the processing and storage of foods. The prior edition of this chapter noted that lead in foods in many countries has declined dramatically during the last several decades, from typically 100 to 200 mg/day (Mahaffey, 1977) to typically less than 5 mg/day by the mid-1990s (Bolger et al., 1996). This decline resulted from virtual bans on the use of lead solders in food and beverage containers and to more widely practiced limits on the use of lead glazes in pottery and food storage containers. Lead contamination of raw food resulting from air and water contamination has also notably declined. Marked decreases noted above for lead concentrations in several environmental media reflect, in large part, strong steps taken to reduce environmental lead levels, based on the growing recognition that the best approach for dealing with lead intoxication as a public health issue is to prevent lead exposure. This notable progress in reducing lead in environment media during the last several decades represents a major success story for a number of countries. For example, the prior edition of this chapter highlighted dramatic reductions in

LEAD IN THE ENVIRONMENT AND HUMAN EXPOSURE

765

blood lead levels among U.S. population groups. That is, between NHANES II (conducted 1976–1980) and Phase I of NHANES III (conducted 1988–1991), the geometric mean blood lead level for U.S. persons aged 1 through 74 years declined from 12.8 mg/dL to 2.9 mg/dL, and the prevalence of elevated blood lead levels (i.e., 410 mg/dL) decreased from 77.8% to 4.4% (Mahaffey et al., 1982; Pirkle et al., 1994). Also, between NHANES III Phase II (1988–1991) and Phase III (1991–1994), the comparable geometric mean decreased by 22% (Centers for Disease Control, 1997). It was further noted that the blood lead concentrations of children had been reduced by more than 80% over the prior 2 decades; and U.S. EPA (2006) more recently noted that marked decreases in environmental lead during the 1980s–1990s were paralleled by decreases in concurrent blood lead levels of U.S children (from a geometric mean of 15 mg/dL in 1980 to 1–2 mg/dL in 2004). Such declines in blood lead levels among the U.S. general population and children not only partly reflect changing economic circumstances (e.g., declining numbers of operating U.S. primary or secondary lead smelters) but also, very importantly, the success of highly effective primary interventions (virtual elimination of lead additives from gasoline, shifting to low lead plumbing solder and fixtures, removal of lead solder from food and beverage cans, etc.) and secondary prevention strategies, for example, public health screening/ education programs and improved nutrition. Despite this overall success, however, some U.S children still experience blood lead concentrations above10 mg/dL, and disproportionate numbers of black and low-income children continue to exhibit such elevated blood lead levels. The most extensive remaining source of lead exposure for U.S. children appears to be lead-based paint in housing, especially for those living in older residences, which are more likely to contain lead-based paints (Jacobs, 1995). Although vigorous national efforts to reduce lead exposure from leaded paint in housing still continue, the problem persists, with the CDC (1997) having estimated in the 1990s that 890,000 U.S. children had sufficiently high blood lead levels (10 mg/dL or above) to impair their learning ability. The prior edition of this chapter noted that elevated blood lead concentrations occur occasionally among adults, with the use of “folk” remedies, cosmetics, and lead-glazed pottery typically being reported to be the source in such isolated cases among the general population. Further, the remaining most persistent cause of elevated blood lead concentrations among adults was noted to be occupational exposures. Apropos to this, the National Institute for Occupational Safety and Health (NIOSH) was also noted as maintaining an Adult Blood Lead Epidemiology and Surveillance (ABLES) Program that tracks the laboratory-reported elevated blood lead levels among adults in the United States, and based on data reported from the 27 states, the cumulative number of reports in 1996 was 16,551 adults with blood lead concentrations of 25 mg/dL or higher, with 318 blood lead concentrations greater than or equal to 60 mg/dL (NIOSH, 1998). More recently, the ABLES Web site currently indicates that the geometric mean of all reported U.S adult blood lead concentrations is now less than 3 mg/dL, an average much lower than the 25 mg/dL level that the U.S. Department of Human Service recommends not to be exceeded by adults. Adverse health effects of lead obviously derive from various types and intensities of responses of cells following particular patterns and intensities of lead exposure. The transfer of lead from the environment to the cells and subsequent interactions of lead with cell components occur as functions of the physical/chemical properties of lead and of physiological factors inherent in the organism. The delineation of physiological pathways involved in determining uptake and internal distribution of lead to various tissues, lead’s accumulation in particular types of tissue and internal redistribution, as well as factors affecting uptake, internal distribution, metabolism, or excretion of the metal are all important,

766

LEAD AND COMPOUNDS

especially (a) to help identify tissues and mechanisms underlying different types of toxic effects of lead and (b) to enhance our ability to estimate or predict the likelihood that adverse responses will result from varying lead exposure patterns and intensities. The human interface with the environment that permits lead entry into the body is comprised of the gastrointestinal (GI) system, respiratory system (including the nasal cavity), and skin. The absorption of lead via these portals of entry and the important factors that affect absorption of lead from different exposure routes as well as certain other aspects of the biokinetics of lead are summarized next, as a prelude to discussion of lead-related health effects. More detailed reviews of lead toxicokinetics can be found in Mushak (1991, 1993, 1998) and U.S. EPA (1986a, 1986b, 2006).

20.4 LEAD ABSORPTION 20.4.1

Gastrointestinal Absorption of Lead

The mechanisms involved in the gastrointestinal absorption of lead are only partially understood. Most of the pertinent research on such mechanisms has been carried out in rodents, especially the rat. Findings by Aungst and Fung (1981), Henning and Cooper (1988), and Fullmer (1997) emphasize the complexity of the process and the extent to which the specific research findings depend on various factors, for example, the dose of lead and the physiological state, nutritional condition, and age of the exposed animal. Based on data from several types of experiments, at least two mechanisms for GI lead absorption appear to exist. One exhibits characteristics of energy-dependent, carriermediated, active transport (Aungst and Fung, 1981); as has been reported, whether this process is saturable or not depend at least in part on the dose of lead used (Keller and Doherty, 1980a, 1980b). This absorption mechanism has active transport mechanism characteristics, because intestinal uptake and flux of lead depend on metabolic energy. At a buffer lead concentration on the mucosal side of 0.5 mM, capacity-limited processes contributed nearly 200 times more to the mucosal-to-serosal lead flux than did diffusion (Aungst and Fung, 1981). At a lead concentration three orders of magnitude higher (48.3 mM), diffusion still only accounted for less than 20% of the flux (Aungst and Fung, 1981). Others have concluded that the major control level for GI absorption of lead likely resides in the intestinal mucosal cell and that the interrelationships between the elements may affect lead bioavailability at both luminal and mucosal levels. It has also been suggested that there are three components to the absorptive phase: uptake by the mucosal cell, transfer through the cell, and movement into the plasma (Ragan, 1983). Lead absorption also appears to have a dose-dependent component. In addition to the work of Aungst and Fung (1981), other work includes in situ studies by Barton et al. (1978a, 1978b) in which the percentage of lead absorbed depended on the magnitude of the dose and was increased in iron-deficient animals (Hamilton, 1978) in relation to fed or iron-replete animals. 20.4.2

Effects of Age on Lead Absorption

Age substantially influences absorption of lead in human and nonhuman primates. For example, Willes et al. (1977) reported that infant monkeys at 10 and 150 days of age retained 64.5% and 69.8% of an oral dose of 210 PbðNO3 Þ2 , whereas adult monkeys only retained 3.2% of an oral dose. Similar age-related differences have been observed for humans. Using

LEAD ABSORPTION

767

classic balance study techniques, Kehoe (1961) found the GI absorption of lead by adult males to be 5–10% of ingested lead. This range of lead absorption with usual patterns of food intake (i.e., not fasting) has been confirmed to be in the range of 5–15% of ingested lead, based on studies with short-lived lead radioisotopes (Hursch and Suomela, 1968) or stable lead isotopes (Rabinowitz et al., 1975, 1976). There are few studies of adult female subjects. Only James et al. (1985), whose subjects (age 26–77 years) included both females and males (12 women and 11 men), reported lead absorption from foods and beverages. However, the report did not discuss any sex-related differences observed between the retention rates for radio-labeled lead. Oral exposure uptake rates for children have been much less clearly documented than those for adults. Available absorption coefficients for children have been derived mainly from two mass balance studies with small numbers of children. Alexander et al. (1974) conducted balance studies in eight subjects (aged 3 months–8 years) with lead intakes averaging 10.6 mg lead/kg bw/d (body weight/day). Absorption averaged 53% of intake, and retention averaged 18% of intake. Also, Ziegler et al. (1978) studied lead absorption by 12 infants (aged 14–746 days) whose lead intakes exceeded 5 mg/kg bw/d. These fractional absorption estimates, having been derived from studies in the 1970s (when exposures to lead were many times higher than current levels), may not be directly applicable to current estimates of kinetics. Still, until more data at lower exposure levels become available, it is appropriate to use distinctly higher fractional absorption estimates for infants. It is unclear over what age period in childhood do the lead absorption characteristics become more like those of adults than infants. Although studies specifically evaluating agerelated changes in fractional absorption of lead by children older than infants are not yet available, some insight might be drawn from other studies. Based on analyses of stable lead isotope profiles of nine immigrant children from Eastern Europe living in Australia, Gulson et al. (1997) found the fractional absorption of ingested lead by the children (aged 6–11years) to be comparable to absorption patterns of adult females in the 29–37 years age range. Whether the 40–50% absorption values for ingested lead obtained for such subjects typically under 2 years old apply to children in the 2–6 years age range remains unclear. Lower absorption values for 2–6 year old children are supported by data from Angle et al. (1995), who suggested that absorption of ingested lead among 2–3 year old children was 10–15%. See Mushak (1991) for more detailed discussion of factors (including both physiological and dietary) potentially underlying age-related differences in GI absorption of inorganic lead. 20.4.3

Influence of Nutritional Status and Dietary Factors on Lead Absorption

The influence of nutritional status and dietary factors on blood and on tissue lead distributions have been most clearly observed at low levels of lead exposure, for example, those more likely to exist currently in the post-2000 than the pre-1980 period. Since the mid-1980s, lead exposure has declined markedly in countries that have discontinued the use of lead solder in food and beverage cans and have phased out the use of lead-based gasoline additives. The beneficial effects of optimal nutrition are enhanced under these lower exposure circumstances (Mahaffey, 1995; Bogden et al., 1997), but the primary approach to effectively reduce lead impacts on public health is to limit human exposures. 20.4.3.1 Total Food Intake Adults in the fasting state have been reported to absorb a substantially greater fraction of lead compared with the fraction absorbed in the nonfasting state (Blake and Mann, 1983). However, there appears to be a lack of data regarding

768

LEAD AND COMPOUNDS

comparable information on the effects of fasting state on lead absorption by children or young nonhuman primates. 20.4.3.2 Calcium Dietary calcium can influence gastrointestinal absorption of lead through both acute and long-term effects of low dietary calcium intake. Numerous studies of experimental animals fed low calcium diets have established that calcium deficiency increased both tissue retention and toxicity of lead (Mahaffey-Six and Goyer, 1970). Longterm calcium deficiency produces a number of physiological adaptations. These include increased concentrations of various binding proteins and the stimulation of endocrine systems and regulatory systems, for example, 1,25-dihydroxycholecalciferol and parathyroid hormone. These secondary changes produced by calcium deficiency also affect the biokinetics of lead. Overall, calcium deficiency generally increases lead uptake. Experimental animal studies show that simultaneous ingestion of lead with reduced calcium concentrations in the incubation medium (i.e., comparable to a low calcium meal) enhanced absorption of lead. For example, Barton et al. (1978a), using ligated intestinal loop techniques to measure lead absorption, found that when the concentration of calcium in the incubation medium varied within physiological ranges, lead absorption decreased with increasing calcium concentration. Prior conditioning by low or high calcium diets did not significantly alter the rat’s lead absorption in vivo. Lower lead absorption was observed in rats and chicks (Smith et al., 1980) during studies on the role of vitamin D in lead absorption. Reduced lead uptake from ligated gut loops upon addition of calcium to incubation media was reported by Barltrop and Khoo (1975), and Meredith et al. (1977) found that oral calcium given immediately before lead very effectively decreased lead absorption in rats. Analogously, higher dietary calcium intake decreased lead absorption in humans (Ziegler et al., 1978; Blake and Mann, 1983; Heard and Chamberlain, 1982). Mechanisms that produce changes in lead absorption due to calcium status have become increasingly better understood (Fullmer, 1997). Aungst and Fung (1985) found that the apparent systemic availability of 1, 10, and 100 mg/kg oral lead doses were three- to fourfold greater in calcium-deficient than in control animals. However, the intestinal absorption of 10 kg/mg doses of oral lead was unaffected by calcium supplements. Such differences were thought to reflect the roles in lead and calcium absorption of dietary vitamin D (cholecalciferol) and metabolically active vitamin D (1,25-dihydroxycholecalciferol). Mykkanen et al. (1982) reported that in chicks, both cholecalciferol and 1,25-dihydroxy vitamin D3 affected both the 203 Pb and 47 Ca absorptive processes, but the nature of these responses were not identical, suggesting differences in the transport path or the macromolecular interactions of these metal ions (or both) during the course of absorption. Studies of lead conclusively verified specific, high-affinity binding of lead to several calcium-biding proteins and suggested that it may be a general property of certain intestinal calcium-binding proteins (Fullmer et al., 1985). Fullmer (1997) further investigated time courses and dose–response relationships for these interactions. By feeding five different levels of calcium and five levels of lead to chicks, Fullmer found that lead ingestion and calcium deficiency, either alone or in combination, generally increased serum 1,25-dihydroxy vitamin D levels over most of the range of dietary lead and calcium intake. However, with severe calcium deficiency, consumption of lead produced marked decreases in 1,25-dihydroxy vitamin D levels. Overall similarities in the responses of 1,25-dihydroxy vitamin D, intestinal calcium absorption, and calbindin-D indicate that the predominant interaction between lead and calcium is mediated via changes

LEAD ABSORPTION

769

in circulating 1,25-dihydroxy vitamin D concentrations, rather than directly through the intestine. Kidney and bone lead levels also changed in response to these dietary manipulations, suggesting that added effects occur that do not fully depend on the concentrations of 1,25-dihydroxy vitamin D, although this appears to be the predominant control mechanism for intestinal absorption. 20.4.3.3 Iron Iron deficiency increases lead tissue deposition and toxicity (Mahaffey-Six and Goyer, 1972). Ragan (1983) demonstrated sixfold increases in tissue lead in rats when body iron stores were reduced but before frank iron deficiency developed. Also, Hamilton (1978) and Flannagan et al. (1979) reported significantly increased absorption of lead from the GI tract of iron-deficient animals. Lastly, based on results obtained by in situ ligated gut loop techniques, Barton et al. (1978b) reported that iron deficiency (secondary to bleeding and to iron-deficient diets) increased lead absorption and that iron loading decreased lead absorption. Ferritin has been shown to bind lead both in vivo and in vitro. In rats fed an iron-deficient diet, the ferritin concentration was low, permitting increased transfer of lead to blood rather than retention in the small intestine bound to ferritin. Transferrin is increased in irondeficiency anemia as a result of increased synthesis. Although transferrin binds iron preferentially, transferrin also transports a number of trivalent and divalent cations such as plutonium, americanium, chromium, cobalt, manganese, and copper, among others. A protein that specifically binds lead, as well as iron, was isolated from both the rat and from the human duodenal mucosa (Conrad et al., 1992). The influence of iron status on lead absorption has also been studied in human subjects, with mixed results (Flannagan et al., 1979; Watson et al., 1980). To date, it is not clear whether these mixed results reflect differences in the severity of iron deficiency, differences in analytical approaches, or other undefined factors. Despite the lack of clarity regarding mechanisms, iron therapy is proving to be a valuable adjunct in the treatment of low-level lead toxicity (Granado et al., 1994). 20.4.3.4 Influence of Chemical Forms of Lead on Gastrointestinal Absorption The effect of the chemical forms of lead on gastrointestinal absorption of the metal can be described only in general terms. Lead bound to alkyl compounds is readily absorbed and concentrates in tissues high in lipid, such as the brain. The percentage of absorption and tissue distribution of alkyl lead differs markedly from those of inorganic lead compounds. Among inorganic lead compounds, the particle size of ingested lead plays a major role in determining the fractional absorption. Barltrop and Meek (1979) showed a fivefold enhancement in lead absorption by rats when the particle size of lead was reduced from 196 to 6 mm. Healy et al. (1982) reported that lead sulfide (considered to be one of the least soluble lead compounds) had increased solubility in gastric fluid (apparently as a result of chemical conversion to the more soluble chloride) when the particle size was reduced from 100 to 30 mm (Healy et al., 1982). Information about the influence of the chemical form of lead on its absorption is frequently complicated by limited information on the experimental conditions, including other factors in the diet, particle size of the inorganic lead source, physiological condition of the animal, and so on. In vitro solubility does not appear to predict well the degree of in vivo absorption (Sartorelli et al., 1985). Despite this limitation, the use of multiple in vitro methods to estimate solubility continues. Interpretation of results from these in vitro methods is further complicated by organ-to-organ differences in bioaccumulation of lead. For example, when immature swine were fed two fully characterized soil samples from a western U.S. Superfund

770

LEAD AND COMPOUNDS

site (Casteel et al., 1997), the bioavailability ranged from about 50–90%, depending on the organ system used to express dose (e.g., blood, liver, or renal lead concentrations). In mechanistic terms, lead absorption depends on such factors as chelation, membrane permeability, solubility, and particle size (Brezinski, 1976; Huisingh and Huisingh, 1974). The coordination chemistry of Pb2þ is likely to play an important role in many of these factors. Coordination with proteins may be a determining factor for the availability of lead for absorption and transfer across the mucosal cell. It is important to know in which form(s) lead is (are) available and what ligands exist in mucosal cells that may be vehicles for absorption or inhibition. Metals can precipitate or coagulate proteins in solution. Also, metal salts often show increased solubility in body fluids (Fairhall, 1924), lead carbonate being, for example, about 300 times more soluble in serum than in water. Such metal–protein interactions depend on factors such as the radius, charge, and coordination number of the metal, as well as factors intrinsic to the proteins, such as size and basicity. In a protein-rich environment, the local metal ion power and complexing ability of the food should be considered. 20.4.4

Absorption Following Inhalation

Absorption of lead from the pulmonary system depends, in a major way, on the particle size of the inhaled lead. It is difficult to determine what fraction of lead dust in inhaled air actually gets deposited in the gas exchange airways and taken up by alveolar cells or what fraction is deposited on the conducting airways and is eventually passed out through the trachea and swallowed with mucus from the trachea or nasal passages. It is clear, nevertheless, that the pattern of lead deposition in the respiratory tract is affected by the particle size of the inhaled aerosol and the ventilation rate (Chamberlain, 1983). The rate of absorption of lead from the particles deposited also depends on solubility of the chemical species of lead. In humans, the absorption of lead from the lung is usually rapid and complete within 24 h. Apparently, even relatively insoluble lead compounds can be taken up directly into the general circulation in this way. Ligand exchange ability is also a key factor here, since it relates to physiochemical properties of the available lead species and the surface properties of the particles involved. All species of lead compounds deposited in the deep (alveolar) lung region are thought to be more or less completely absorbed into the blood stream (Morrow et al., 1980), with distinctly greater alveolar deposition typically occurring for particles less than 2.5 mm in diameter than for larger-sized particles. However, as is often seen with various environmental or occupational exposures, it is not unusual for lead dusts in inhaled air to contain lead particles of large enough size (larger than 2.5 mm diameter) so that many are cleared from the conductive airways by mucociliary action. Lead particles larger than 2.5 mm in diameter deposit mainly in the nasopharyngeal and tracheobronchial regions (which constitute the upper repiratory tract) where they can be transferred by mucociliary action and then swallowed. Therefore, because many such particles are swallowed, factors that affect the GI absorption of lead can also often significantly influence the bioavailability of inhaled lead. Thus, depending on the particle size and concentration of lead associated with a particular air exposure source, the digestive tract can also be an important avenue of lead absorption following inhalation. Chemical speciation of lead dust in occupational settings has shown marked variability in the size of particles generated in primary smelters (Spear et al., 1998). Depending on the process performed (e.g., samples from ore storage, sintering, or blast and dross furnaces), the particle size, mineralogy, and extractability of the constituent lead can differ substantially, and, consequently, changes to the ore during processing are thought to influence the biological availability of the lead.

DISTRIBUTION

20.4.5

771

Dermal Absorption of Lead

Skin absorption is not usually considered to be a significant mode of lead uptake (Minot, 1929), unless the metal is present in its more lipid-soluble organic forms (such as tetraethyl lead). Florence et al. (1988) found that inorganic lead can be absorbed through skin and rapidly distributed throughout the body. Of note was the observation that the distribution tends to vary in some ways from that of ingested lead. For example, skin absorption gave rise to increased lead excretion in sweat, although similar increases in blood and urine were not observed.

20.5 DISTRIBUTION Lead is absorbed into blood plasma, where it rapidly equilibrates with extracellular fluid. More slowly, but within minutes, lead is transferred from plasma into blood cells (Chamberlain, 1985; Simons, 1986). The typical concentration of lead in whole blood is about 10 6 M. Because 95–98% of the lead is bound in red blood cells, about 10 8 M is present in the plasma. If the distribution of lead between plasma and cytosol is similar to that of calcium (10,000:1), the cytosolic concentration of lead in exposed individuals should be in the picomolar range. In animal experiments, no constant relationship has been found between the lead concentrations in blood and in soft tissue. Thus, as earlier noted by Kazantzis (1988), controversy exists with regard to whether lead in blood represents biologically active lead, and indeed, the extent to which the two may be linearly related. Improved detection limits for analytical methods have increased our ability to determine concentrations of lead present in plasma. Concerns remain, however, that even very slight hemolysis of erythrocytes during the separation process can transfer lead into the plasma fraction. Consequently, data on plasma lead concentrations must be treated with caution unless the technique can establish that what is thought to be plasma lead does not simply represent the in vitro transfer of lead from erythrocytes. From the blood plasma, absorbed lead is distributed to different organs, with liver and kidney attaining highest concentrations. Of interest, the peripheral nervous system may accumulate much more lead than the central nervous system (CNS). Also, marked variation occurs in lead distribution within various other tissues and organs (Barry, 1975, 1981; Drasch et al., 1987; Drasch, 1974, 1997; Drasch and Ott, 1988). In this regard, there are several important corollary observations. First, lead tends to accumulate wherever high calcium levels are found. Highest lead concentrations are, therefore, found in bone, especially in dense cortical bone. Second, within and among soft tissues, highest lead concentrations seem to accumulate in those tissues and organs having the highest mitochondrial activity. Likewise, within a given organ, the highest concentrations occur in regions with the highest mitochondrial activity (e.g., in the renal tubule and in the choroid plexus and cerebellum of the central nervous system). The skeleton contains more than 90% of the body burden of lead when measured at steady state. However, this pool is neither homogeneous (Kehoe, 1961; Chamberlain, 1985; Rabinowitz et al., 1976) nor is it static (Gulson et al., 1995). This association with bone is related to lead’s similarity to calcium and formation of insoluble lead phosphate. As described subsequently, there are many bone pools of lead. Bone is a very complex organ, having varying density and structure, depending on skeletal site and function. The turnover of tissue lead is high throughout life. Among young adult women, between 50 and 75% of

772

LEAD AND COMPOUNDS

blood lead reportedly comes from tissue stores (i.e., skeletal) rather than the current environment (Gulson et al., 1995, 1996). Under conditions of physiological stress for calcium (including pregnancy and lactation), the release of bone lead becomes even higher (Gulson et al., (1998a, 1998c); that is, the initial source of the lead is environmental lead that has been accumulated in tissues over previous years. Estimates of the fraction of current blood lead derived from bone were also provided by Smith et al. (1996), who reported that the skeleton contributed 40–70% of lead in blood among five subjects who had trabecular bone samples obtained at surgery. There are now thought to be three basic types of bone lead pools:(1) rapidly exchanged lead in very metabolically active portions of bone; (2) lead in trabecular or spongy bone; and (3) lead in dense cortical bone. Lead turnover in these three basic types of pools appears to roughly parallel the relative rates of calcium turnover. Various observations suggest marked variation in the distribution and turnover rates within these pools, at least partially depending on the particular region of the skeleton studied. It is important to note further that although lead concentration is lower in trabecular than in cortical bone, the mass of trabecular bone is, on average, four times that of cortical bone and it is approximately four times as labile. Therefore, trabecular bone may represent a much more metabolically important pool of lead. The distribution of lead in tissues reflects a state of constant, dynamic equilibrium. As noted in the section on excretion, many methods of enhancing lead excretion are also influenced by lead’s redistribution within the body (Cory-Slechta et al., 1987). Clearly, any situation that mobilizes the very large, relatively stable pools of lead within the body, particularly those in the bone, will lead to the redistribution of lead to a variety of tissues. This redistribution is thought to explain the increased symptomatology that is frequently noted in lead-poisoned children during acute illnesses. Redistribution is known to occur during pregnancy and, even under usual circumstances, can result in increased risk to the fetus, particularly in women with prior lead poisoning. There is also some evidence that the osteoporosis of aging may be accompanied by the significant mobilization of lead from bone pools. It is clear that much additional information is needed to clarify more fully the various physiological and pathological conditions of enhanced mobilization and redistribution of lead. 20.5.1

Excretion

Lead is excreted from the body mainly by urinary and fecal routes, with fecal excretion representing the sum of unabsorbed endogenous lead from saliva, bile, and (to a lesser extent) other gastrointestinal secretions, plus the unabsorbed portion of inhaled and ingested lead. Less excretion occurs through sweat and integumentary losses (including skin, hair, and nails), and these routes account for only a small portion of total excretion. Under conditions of fairly constant exposure to low lead concentrations, a steady-state condition evolves, wherein excretion approximates intake (Rabinowitz et al., 1976) and 70% of intake is excreted via urine. Under short-term conditions of low-level increased exposure (Chamberlain, 1985), 60% is retained by the body and 40% excreted. In a 14-day study of human volunteers receiving a single dose of 203 Pb, 18% of the dose was excreted whereas 35% was retained in blood, with a residual of 47% in soft tissues and bone storage. Extrapolating this result would predict an eventual 30% excretion and 70% retention of this single dose. Urinary excretion of lead is quite complex, depending on the situation, but is most likely a function of plasma lead levels. Under most conditions of exposure to relatively low lead

DISTRIBUTION

773

concentrations (e.g., at blood lead levels below 25 mg/dL), the concentration of lead in plasma is very low (about 0.01 mg/dL) and not related to whole blood lead. Above this concentration, plasma lead increases significantly, as does urinary lead excretion. At blood lead concentrations below 25 mg/dL, urinary clearance of lead has been estimated at 1.1% per day. As blood lead rises above this, urinary clearance appears to rise at a rate reasonably related to the increase in plasma lead (Chamberlain, 1985). At very high blood lead concentrations, renal dysfunction may decrease urinary excretion of lead. In considering fecal lead excretion, one must differentiate between fecal lead that is unabsorbed from ingestion or inhalation and fecal lead that truly represents endogenous fecal excretion. Endogenous excretion was measured by Rabinowitz et al. (1976) using stable isotope studies and by Chamberlain (1985) after inhalation and/or parenteral administration of 203 Pb. Endogenous excretion is often estimated by comparing renal clearance to apparent total body clearance. All such estimates suggest a clearance of 0.5% per day at blood lead levels 525 mg/dL. It is likely that this rate of clearance is basically independent of blood lead and so it does not significantly increase with increasing blood or plasma lead. Excretion by all other routes is at a rate of 0.2% per day and is, again, essentially independent of blood lead concentration. The total for all these excretion routes is 1.8% per day at blood lead concentrations less than 25 mg/dL and somewhat greater at higher blood levels because of increased urine lead excretion at higher concentrations. A special form of excretion is that which occurs via breast milk. Various studies of maternal breast milk composition indicate that breast milk lead concentrations appear to correlate well with maternal blood lead concentrations. One report of plasma lead concentrations in mice (Keller and Doherty, 1980a, 1980b) suggests that breast milk lead concentrations are more closely related to plasma lead concentrations and can be as much as 25 times that of plasma lead. This suggests that at lower blood lead concentration in the mother, breast milk would likely represent a minor route of excretion and is usually a minor exposure route for the infant. However, at higher blood lead concentrations with increasing plasma lead concentrations, it is plausible that a significant amount of lead could be mobilized from the maternal skeleton in lactating women and that breast milk could represent an important exposure pathway for breast-fed infants. In fact, using stable isotope methods, Gulson et al. (1998c) demonstrated this mobilization among women with blood lead concentrations less than 10 mg/dL. Breast milk lead appears to be linearly related to blood lead and has concentrations similar to plasma. For blood lead levels in the range of 2– 34 mg/dL, breast milk contains les than 3% of the quantities of lead in blood. The amounts of lead released from the skeleton show much person-to-person variation, suggesting that among women who have had substantial prior lead exposure, it is important to assess this as a possible exposure source for the infant. A variety of specific methods have been used to alter the clearance of lead. Aub et al. (1925) showed that urinary excretion could be enhanced by acidification, presumably by mobilization of lead from relatively stable pools. After some initial enthusiasm for the therapeutic potential of this intervention, it was abandoned because such therapy often also enhanced symptomatology, presumably by redistributing lead to soft tissues. A variety of chelating agents have also been used to enhance the clearance of lead. The majority of chelating agents, including ethylene diamine tetraacetic acid (EDTA), dimercaptosuccinic acid, and d-penicillamine, enhance clearance by binding lead and promoting urinary clearance. Dimercaprol is another reasonably effective chelator of lead, but it predominantly enhances biliary excretion of lead.

774

LEAD AND COMPOUNDS

20.6 KINETICS Basic to an understanding of the effects of lead exposure on animals and humans is an appreciation of the kinetics of lead in living animals. This requires recognition and knowledge of the various phases of lead kinetics, including absorption, distribution, and clearance. Although lead may exist in the body in various ionic forms and compounds, it is not properly considered, as a material, to be “metabolized” by the body. Rather, it is transported by various more or less metabolically active complexes and compounds. Useful terms to describe this distribution are lead kinetics and biokinetics. Several diverse approaches have been used to assess the lead kinetics in mammals. Of interest is the fact that all of the approaches have produced similar conclusions. The most important of these is that delineation of lead kinetics requires a multicompartment model, with some compartments being rather large and relatively stable, whereas other compartments are smaller and comparatively labile. Also, it appears most likely that the kinetics of lead at lower concentrations of exposure may be considered linear, whereas at high exposure concentrations, they appear to be nonlinear. This has potentially important implications in considering the biological effects of lead at varying exposure levels. Several types of approaches have been used to investigate the biokinetics of lead. Such approaches mainly include the following: 1. Exposure of both experimental animals and human subjects to lead orally or via inhalation, wherein increasing amounts of lead are introduced and the resulting accumulation and excretion of lead are measured by a variety of methods. 2. Introduction of radioactive or stable isotopes of lead to determine kinetics without increasing exposure and disturbance of the steady-state situation. 3. Study of the spontaneous clearance of lead in situations where exposure to a high concentration of lead has been terminated. 4. Study of the distribution of lead in various tissues by postmortem examination. The last approach has been very useful in animal studies but little used in studying lead distribution in humans. It is important to remember that there is notable interspecies variation with regard to the kinetics of lead due to many factors, for example, diet, physiology, and relative tissue mass. For these reasons, it is clear that studies of animals other than humans must be viewed with great caution in attempting to understand human kinetics. Since the greatest concern is with human exposure, human studies are emphasized below. The earliest studies of lead balance in humans and animals were by Aub et al. (1925), who used several of the above-noted approaches in a series of studies of exposures of animals to lead by both inhalation and ingestion. Aub concluded from those studies that lead was most readily absorbed from the respiratory tract, particularly by the inhalation of “finely divided particles,” and he noted that lead was somewhat less well absorbed by ingestion. In his animal studies, Aub also measured lead concentrations in a variety of tissues after exposure to determine the distribution of lead in the body. In addition, he performed autopsies on human bodies to evaluate the internal distribution of lead in lead-exposed patients. Aub concluded that especially after termination of lead exposure, all the lead is “permanently” stored in bone and, accordingly, that lead was harmless to a person unless there was “recent absorption from an external source or mobilization of a skeletal store.” He conducted

KINETICS

775

extensive further studies of lead excretion in animal and human subjects and found spontaneous excretion to be very low and variable, but noted that calcium-deficient diets and administration of acids markedly increased lead excretion (especially if both factors are applicable at the same time). Kehoe (1961) subsequently conducted a landmark series of experiments that involved long-term exposure of human volunteers to lead by inhalation and ingestion. The subjects were observed prior to and subsequent to exposure to lead to determine both baseline lead balance and balance during a “recovery” stage from lead exposure. Kehoe’s studies were summarized by Gross et al. (1975), who made several important observations that are relatively consistent with Aub’s earlier findings. For example, there was considerable variability of observations within the subjects during the control period, which appeared to relate, at least in part, to variability in dietary lead exposure. Also, these variations in ingested lead (about a mean of 191 mg per day) were paralleled by variations in fecal and urine lead with an overall net negative lead balance (in most subjects), as calculated by intake versus urine and stool output. It is not clear whether this reflected an actual net negative balance or was simply a result of not being able to measure airborne lead exposures during the control period. A later study by Rabinowitz et al. (1976) substantially clarified our understanding of absorption and distribution of lead in the body. The results of this study, which are among the most carefully obtained data available, have been extensively used by the subsequent kineticists. Rabinowitz et al. (1976) studied “normal” volunteers under standard conditions of diet and activity. These volunteers were fed a low lead diet supplemented to approximate their usual level of lead intake by addition of 204 Pb as a tracer. Since this isotope is rare in most usual sources of exposure, it provided a stable tracer for purposes of defining the kinetics of lead in the body. Lead isotope distributions in samples were determined by mass spectroscopy. These adult volunteers had prestudy blood lead concentrations of 16–25 mg/dL and had intakes of lead between 156 and 215 mg per day during the study, which approximated their prestudy intake. One of the five subjects was studied for 10 days. The other four were studied for longer periods, ranging from 108 to 210 days. The range of absorption observed was 6.5–13.7% of the ingested dose. Of this absorbed lead, 54–78% was excreted in the urine, with most of the rest being excreted via bile and integumentary losses. The translocation of the tracer lead was best described by a three-compartment kinetic model. The first compartment, which included 1.5–2.2 times the amount of lead in blood, contained an average of 1900 mg of lead and was turned over in about 36 days. A second one, which comprised most of the soft tissue lead, contained about 600 mg of lead and was turned over every 40 days. These authors noted, however, that the total of less than 3 mg of lead in these two labile pools was much less than the 10–30 mg found in autopsy studies, suggesting that most of the soft tissue lead must have been in a more stable compartment. The third, large and comparatively stable, compartment was primarily comprised of bone lead. In these subjects, it contained about 200 mg of lead and was turned over approximately every 104 days. Of interest with regard to this pool was the comparison of total lead and tracer lead to total lead ratios in cortical and trabecular bone from the iliac crest. The total lead concentration in the cortical bone by weight was approximately twice that in trabecular bone. However, trabecular bone had a two- to threefold greater ratio of tracer to total lead, suggesting that it was turned over much more rapidly than the cortical bone. Of further interest was their calculation, based on these measurements, that the iliac bone received lead three–seven

776

LEAD AND COMPOUNDS

times more rapidly than did the very stable pool as a whole. These observations make it clear that bone cannot be regarded physiologically as a single pool. Furthermore, Rabinowitz et al. (1976) emphasized that all of the pools are in dynamic equilibrium with each other and, therefore, any changes in the movement of lead from one to another can cause significant changes in the measured amount. Factors that might cause movement of lead from the large stable pool into blood, and hence to target soft tissues, are of particular interest. Steenhout (1982) developed a kinetic model of lead distribution based on data regarding lead in teeth in children and adults in three regions of Belgium. Her model, basically consistent with the Rabinowitz model, suggests that the rate of transfer of lead to teeth is 1.85 ppm/yr/mg per 100 mL blood. According to the model, lead accumulation in teeth and dense (cortical) bone is linear and continuous over age, suggesting very slow loss from dense bone (approximately 0–0.005 ppm/yr/mg per 100 mL blood). In contrast, estimates of lead loss from “porotic” bones, such as ribs and vertebrae, are on the order of 0.06 ppm/yr/mg per 100 mL blood. Steenhout concluded that the apparent nonlinearity of lead transfer in some other studies reflected this relatively rapid loss of lead from such porotic bone. She also suggested that her data support the concept that for dense bone, there is no loss of lead with increasing age and, therefore, the osteoporosis of age should not represent a risk for lead mobilization. Chamberlain (1985) used data from a variety of data sets, including some from other investigators and some he had developed to assess several aspects of lead kinetics in humans. He focused primarily on volunteer feeding experiments to observe the response of blood lead to either airborne exposure or to dietary intake. His discussion concerned largely inorganic lead and relatively short-term studies. He found that lead is absorbed rapidly into plasma and then into extracellular fluid in minutes, based on experiments with the injection of radioactive lead tracers. He noted that in a time frame measured in tens of minutes, lead in plasma, and lead from extracellular fluid via plasma, becomes largely bound to red blood cells. Approximately 58% of a dose of lead was found to be bound to red cells after 20 h. Chamberlain (1985) also found that excretion of lead after a single dose occurs over a month and that lead storage in tissues and bone persists for months to years. He further noted that the accumulation and distribution of lead differ in several ways from those of strontium. Most importantly, the attachment of lead to red cells appears to retard, rather than to promote, the distribution of lead to storage sites. Chamberlain’s autopsy studies showed that relative to “dose,” there is more lead than strontium or calcium in soft tissues. All the studies Chamberlain reviewed agreed that the transfer of lead to excreta from blood occurs over a period of about one month. He also noted that at low-level exposures, urinary excretion is two or more times greater than excretion via stool. In his discussion of transfer of lead to bone, he analyzed both the discrepancies and consistencies among available data sets. Of most importance for kinetic modeling is the observation in some of the data sets that the lead concentration in trabecular bone is similar to that in cortical bone, whereas in other data sets the lead concentration in trabecular bone was much higher than in the cortical bone. Some of this discrepancy may be related to the duration of the studies, since short-term (versus longer) studies may show relatively more lead in the presumably more labile trabecular pool. Chamberlain’s resorption rates from storage in bone are inferred indirectly, that is, based on studies of strontium turnover, given the assumption that the rates do not significantly differ for various trace minerals. However, considering his review of differences between blood, plasma, and tissue distribution of strontium and lead, this assumption may not be entirely valid.

KINETICS

777

Based on these studies, Chamberlain (1985) suggested a mean life of lead to be 12.5 years in trabecular bone and 50 years in cortical bone. His estimate for the mean life of lead in soft tissues, derived from autopsy data, was 500 days. The relationship between urinary clearance and blood lead appeared to be constant in the range of “normal” blood lead concentrations, but to increase proportionately at blood lead levels above 20–25 mg/dL. This implies a decreasing apparent relative uptake with increasing blood lead concentration. Chamberlain reviewed some data on intestinal absorption, noting that uptake of soluble lead tracers is markedly affected by a period of fasting, with an average (in several studies) of 8% uptake when lead was taken with a meal and 60% when taken after an overnight fast, if the fast is continued for several hours after the lead ingestion. He noted that insoluble lead sulfide absorption was less affected by fasting (12% absorbed in fasting versus 6% with meals). He also reviewed data showing that the addition of calcium and phosphorus salts markedly decreases the absorption of soluble lead. Chemical incorporation of lead with foodstuffs did not alter lead absorption below the levels observed when lead was administered with another metal. The discussion of airborne lead exposure by Chamberlain (1985) focused on three major factors of exposure to inorganic lead only: (1) airborne lead concentration; (2) ventilation rate/volume; and (3) fractional distribution of the aerosols. He did not consider airborne exposure to organic lead. The particle size in the inhaled aerosol and the “residence time” in the pulmonary region, determined largely by respiratory rate, appeared to be relatively unimportant. Once retained in the lungs, the deposited lead is essentially completely absorbed into the bloodstream within 24 h. Of note is that many larger particles, deposited higher in the conductive airways, are returned by mucociliary clearance to the pharynx and thereafter ingested, reducing this mainly to the case of ingested inorganic lead. A somewhat more complex model of lead distribution was suggested by Bernard (1977). His “reference man” had a total body burden of lead of 120 mg. Of this total burden, 110 mg was in bone and the rest in soft tissues. His model proposed at least two bone pools, a slow pool in cortical bone and a relatively labile pool in trabecular bone. Bernard further proposed two soft tissue pools, one of which is relatively large and slow and another is quite large and very rapidly turned over. Although this model is logical and based on experimental observations, its actual validity is somewhat subject to question because it is based exclusively on studies in rats and nonhuman primates, which may markedly differ physiologically from humans. Subsequently, Schutz et al. (1987a) observed the decline of blood lead after termination of occupational exposure by studying two separate and somewhat different groups. The first group included workers who no longer worked in the lead industry, whereas the second included those who were removed from work due to a blood lead increase to above 3 mmol/L (60 mg/dL). The first group was older, had longer periods of exposure, and generally had lower mean blood lead levels than the second. The subjects also had bone lead concentrations estimated by the use of X-ray fluorescence (XRF) of the middle phalanx of the left index finger. A two-compartment model was found by Schutz et al. (1987a) to provide a satisfactory fit to his data, with the “fast” compartment having a half-life of 30 days and the “slow” compartment a half-time of 5.6 years. There was notable intersubject variation, which was suggested to represent “considerable variation in risk at a given exposure level.” The “slow” pool was turned over somewhat more rapidly than other reported rates. The bone lead observations by X-ray fluorescence correlated positively with estimates of the slow pool, but the coefficient of correlation was rather low (r ¼ 0.36). Schutz hypothesized that the slow pool may actually consist of a combination of two bone pools, one of trabecular and

778

LEAD AND COMPOUNDS

the other of cortical bone lead. This hypothesis was used to explain why the measurement of bone lead somewhat differed between the two groups. He suggested that this difference may relate to differing proportions of lead in the cortical versus trabecular bone, the concept being that long-term exposure would result in higher relative levels in cortical bone than in trabecular bone. If this is correct, it has implications for considering widespread application of noninvasive methods of bone lead measurement to research and clinical assessments. The biological redistribution of lead pools within the body has been described in a series of reports based on the evaluation of differences between stable isotope profiles of body lead compartments accumulated in separate geographic locations and during different life periods. Two persons had earlier been investigated by Manton (1977, 1985 in initially describing the process. Further research to confirm these changes, as well as to verify their occurrence in a number of subjects, was conducted by Gulson et al. (1995, 1997, who investigated a cohort of adult women who immigrated to Australia from Eastern Europe and Russia during the early 1990s. These subjects had accumulated tissue stores of lead in Europe that had a stable isotope ratio distinctly different from that of lead in Australia. Such differences enabled the researchers, by means of meticulous measurement of stable isotope ratios via thermal ionization mass spectroscopy, to identify the proportion of blood lead from the contemporaneous environment and that mobilized from tissue lead stores accumulated earlier in Europe. The data obtained showed that among these young adult women, 45–70% of lead in blood came from the long-term tissue stores, presumably in bone (Gulson et al., 1995). These proportions occurred at blood lead concentrations that averaged 55 mg/dL as the result of environmental lead exposures typical of developed countries where steps had been taken to restrict lead exposures. The above study also aimed to determine the influence of pregnancy and lactation on this mobilization. During pregnancy, blood lead concentrations in these female subjects increased by about 20%, on average, with individual changes ranging from 14 to þ83% (Gulson et al., 1997). Among those subjects whose blood lead levels increased during pregnancy, the mean increase in mobilization of long-term tissue lead stores varied from 26 to 99%, averaging about 30% (Gulson et al., 1997). Skeletal lead mobilization continued to be elevated after the pregnancy. Observations for infants born to these mothers showed that the long-term tissue stores of lead in the mothers had been transferred to the fetus and that among those infants who were breast-fed, additional transfer of lead continued to occur during breast-feeding (Gulson et al., 1998a, 1998c). The transfer of maternal skeletal lead to the fetus as shown by stable isotope analyses has also been confirmed among nonhuman primates (Franklin et al., 1997). Barry (1985) measured lead concentrations in tissues of 129 subjects at autopsy and presented extensive data on lead content in various tissues from this very large series. He noted that consistent with other studies, the content of lead in bones was much higher than in soft tissues and that the levels of lead in dense bone were much higher than in more cancellous bone. For example, petrous bone had the highest levels and ribs the lowest (by an approximate ratio of 4:1), a finding of much importance in understanding the generally nonhomogeneous distribution of lead in bones. Barry noted that in those soft tissues with the higher lead amounts, the concentrations in males exceeded those in females by about 30%. He stated that soft tissue lead concentrations increased with age only through the second decade of life and were thereafter stable. Children were reported to have soft tissue concentrations similar to adult females, but had much lower bone concentrations. He also stated that in adults, over 90% of the lead was in bone, with more than 70% being in dense bone. Among occupationally exposed adult males, 97% of their total body lead was in bone.

KINETICS

779

Barry noted that the increase of lead in bone with age, with respect to stable soft tissue levels, is consistent with the hypothesis that lead in bone is not available to soft tissues. He also indicated that the lack of decline in the bone lead concentrations in the face of demineralization with increasing age suggested that lead was not mobilizable even under conditions of massive calcium turnover. However, it must be further noted that Barry expressed bone lead changes with age only as concentration and not as total lead, and he ignored the fact that demineralized bone would have decreased total mass. Thus, the apparent constant concentration of lead in bone may actually reflect a markedly decreased total amount of lead in bone. Predicting quantities of lead mobilized from bone requires bone lead concentration data for both cortical and trabecular bone. Data sets providing bone lead concentrations among adults and children described above were obtained during time periods in which environmental lead exposures were much higher than now present in many developed countries. Such lower lead exposures result in distinctly lower bone lead levels. For example, Drasch and coworkers obtained data on bone lead concentrations for cases coming to autopsy in Munich between the early 1970s and 1994 (Drasch, 1974, 1997; Drasch et al., 1987; Drasch and Ott, 1988). These data are for subjects living in the same geographic vicinity in southern Germany. Between 1974 and 1994, trabecular bone lead decreased from 2.5 mg/kg (1974) to 1.7 mg/kg (1984) to 0.7 mg/kg (1994). Compact bone decreased from 5.5 mg/kg (1984) to 2.8 mg/kg (1994). These data are for adults. Changes in bone lead can be expected to be even more dramatic among young children, who (unlike adults) do not have long-term stores of lead accumulated during decades of much higher lead exposures. Drawing upon the above types of advances in regard to lead kinetics and associated modeling thereof, substantial further progress has been made during the last several decades in developing and refining biokinetic model systems to project likely increased risk of lead-related toxicity in human population groups due to various lead exposure scenarios. Such biokinetic models involve stipulation of mathematical relationships among biological processes (e.g., absorption, distribution, redistribution, clearance, elimination) that determine variations in internal concentrations of the metal and associated potential for causing toxic effects. In particular, during recent decades, much progress has been made in developing various biokinetic models for predicting “blood lead” as the most widely accepted internal biomarker (discussed subsequently) traditionally used to index lead exposure/dose and to gauge consequent potential for leadrelated pathophysiological responses to occur. Some salient examples of progress made since the early 1980s in such modeling efforts include those of Marcus (1985a, 1985b, 1985c), in which available data sets were used to derive multicompartment kinetic models for lead, and that of Bert et al. (1989) who developed a compartmental model for adult males. Also, Leggett (1993) published an age-specific biokinetic model for lead that was developed originally for the International Commission on Radiological Protection (ICRP) but was later expanded to include additional features of use in consideration of lead as a chemical toxicant. In it, the transport of lead between compartments was assumed to follow linear, first-order kinetics, provided that the concentrations of lead in red blood cells remained below a nonlinear threshold level but a nonlinear relationship between plasma lead and red blood cell lead was modeled for concentrations above that level. Several other physiologically based models for bone-seeking elements published by O’Flaherty (1993, 1995, 1998 utilize information about age dependence of bone formation rate and take into account increasing localization of bone remodeling activity with age.

780

LEAD AND COMPOUNDS

In addition to the above models, the U.S. Environmental Protection Agency developed a widely used “Integrated Exposure/Uptake/Biokinetic” (IEUBK) Model for Lead in children (U.S. EPA, 1994a, 1994b; White et al., 1998; Hogan et al., 1998). As noted in Chapter 4 of U. S. EPA (2006), the IEUBK Model simulates lead exposure and biokinetics from birth to age 7 years and predicts quasi-steady-state average blood lead concentrations corresponding to daily average lead exposures averaged over periods of one or more years. Comprised of four subcomponent models (an exposure model, an uptake model, a biokinetic model, and a blood lead probability model), the IEUBK Model (1) calculates average daily intakes of lead (mg/ day) for each inputted exposure concentration (or rates) of lead for different multimedia lead exposure routes (via air, water, diet, dust, soil); (2) next converts the media-specific lead intake rates (calculated from the exposure model) into media-specific time-averaged rates of uptake (mg/day) of lead into blood plasma as the central compartment; followed by (3) biokinetic modeling that simulates transfer of absorbed lead between blood and other body tissues (bone, brain, kidney, etc.), lead excretion from the body (via urine, feces, skin, hair, nails), and predicts an average blood lead concentration for the exposure time period of interest; and (4) lastly utilizes a blood lead probability submodel that applies a log-normal distribution (with specific geometric mean/geometric standard deviation parameters) to predict probabilities for the resulting occurrence of a specified blood lead concentration in a population of similarly exposed children. Further efforts are now underway by the U.S. EPA to develop to an “All-Ages Lead Model” (AALM) that (1) aims to simulate lifetime lead exposure and biokinetics of lead in humans from birth to age 90 years and (2) expects at some near-future time to include a pregnancy module that simulates transplacental transfer of lead from mother to fetus. The AALM is expected to be capable of predicting lifetime exposure impacts on (a) the internal distribution of lead to bone and various soft tissues (brain, kidney, etc.) in addition to (b) blood lead distributions for population groups from birth up to 90 years of age. Chapter 4 of U.S. EPA (2006) summarizes the most salient features of the IEUBK Model, the AALM, and other biokinetic models alluded to above, and it notes available evaluations of the relative accuracy of the models in terms of how well their predicted blood lead distributions (or means) approximate actual blood lead distributions (or means) observed for modeled lead exposure scenarios. Models such as those discussed above have had varied success in predicting blood lead concentrations. Typically, such models have been more successful in predicting mean/ median blood lead concentrations than the overall distribution of blood lead levels. Dose-dependent differences in fractional absorption and distribution of lead complicate application of these models. Person-to-person variabilities (intensity of hand-to-mouth activity, nutritional status, etc.) can also modify the relationships between external (environmental) doses of lead and internal (blood, bone, soft tissue) lead concentrations. Among those models that recognize the importance of bone lead contribution to blood lead, the very limited available data on contemporary bone lead concentrations remain an important factor that needs to be more fully addressed. Workshops have been held that have attempted to reconcile differences between the modeled distributions of blood lead data and the observations from epidemiological studies. The availability of pooled analyses of epidemiological data from childhood lead studies in the United States has identified lead-contaminated dust loadings within the residence as a very strong predictor of blood lead among children (Lanphear et al., 1998). Further, the child’s age, race, mouthing behaviors, and study- or site-specific factors are influential in predicting blood lead at a given level of lead exposure (Lanphear et al., 1998).

BIOMARKERS

781

Analysis of available information on the kinetics of lead reveals some problems in using the currently available models to study lead in the living mammalian subject. Many of the models fit observed data for lead absorption, distribution, and excretion reasonably well and often describe observed changes in blood lead quite well. However, currently available models are generally not fully adequate for purposes of anatomic/physiologic description and prediction, thus limiting their usefulness in devising experimental models or developing useful approaches for clinical diagnosis/management of lead intoxication. For example, the assertion that bone lead is a single homogeneous pool, or two relatively homogeneous pools comprised of cortical and trabecular bone, does not comport well with certain available data. The Rabinowitz models and others derived from his data suggest that the most stable pool, thought to be largely bone, had a mean life of 30–50 years (Chamberlain, 1985; Simons, 1989; Kazantzis, 1988; Kehoe, 1961; Rabinowitz et al., 1976; Cory-Slechta et al., 1987). However, when studied directly by bone biopsy (Rabinowitz et al., 1976), the same subjects appeared to have much more rapid trabecular and cortical bone turnover than predicted by their models. The clear conclusion, if both the kinetic model and the biopsy results are accepted as correct, is that either the iliac bone is not part of the stable pool or that there must be exceptionally stable portions of the bone pool that outweigh the relative lability of iliac bone. In either case, it is clear that bone is not a single homogeneous pool nor is either major type of bone (cortical or trabecular) likely to represent a homogeneous pool of lead, this obviously having some implications for utilizing invasive and noninvasive methods of sampling of bone so as to determine its lead content.

20.7 BIOMARKERS An evaluation of potentially useful biomarkers of lead exposure requires consideration of a variety of specific issues. In common with all toxic exposures is the basic dose–response issue. In any such system, if dose–response characteristics are typical, well defined, and predictable, it makes little difference whether one focuses on measuring the dose or prefers to focus on a response variable. In the case of a toxicant such as lead, which has many diverse effects, it is usually necessary to define the response variable(s) of most importance to the investigator. In the case of mammalian (particularly human) studies, it is often useful to try to define the “critical” organ, tissue, or system. This definition presupposes that it is indeed possible to define the most sensitive or most important effect of the toxic agent. In the case of lead toxicity, especially childhood lead toxicity, the nervous system has been most commonly identified as being the “critical” organ or organ system. To the extent that the nervous system is accepted as the critical organ system, then any indicator of lead exposure/dose, whether measurement of lead per se in one or another tissue component or of some closely varying biological response, should closely define or reflect the extent of nervous system exposure to lead. Ideally, any response variable(s) used as the lead exposure/dose marker (s) should be based on nervous system toxicity or, at least, strongly correlate with nervous system toxicity. An important correlate of variation in response to a toxic agent such as lead is that it is often more difficult to detect a response in an individual than it is for groups of individuals. Hence, some particular measures of dose and/or response that may be useful in epidemiological studies may not necessarily aid much in the diagnostic categorization of the individual subject. An in-depth analysis of analytical methods, important considerations in selection of lead biomarkers, and interpretation of data for

782

LEAD AND COMPOUNDS

sensitive populations (e.g., maternal/fetal pairs, women of childbearing age, infants, and young children) was published by the National Research Council’s Committee on Measuring Lead in Critical Populations (1993). Lead exposure occurs via multiple routes, at variable external dose rates, and results in variable absorption depending on route(s) of absorption and other factors described elsewhere in this chapter. It is, therefore, generally very difficult to directly specify the “dose” of lead to which a subject is (or has been) exposed. The exception to this has been studies using stable radioactive lead tracers, wherein the absorbed dose can be quite carefully calculated by isotope dilution methods (Rabinowitz et al., 1976) or through careful measurement of diverse external sources (e.g., see Gulson et al., 1996). With the exception of such tracer methods, approaches to characterizing lead dosage in living subjects have generally been limited to measurement of lead in relatively easily obtained biological samples. All such methods encounter difficulties in sample contamination and analytical technique because of the very small amounts of lead typically found in the samples. Measurement of lead in whole blood has been most widely used, but has been criticized on practical and theoretical grounds. Difficulties potentially arising from contamination or due to technical aspects of measurement have largely been averted when blood lead analyses are done by a competent laboratory, and by the late 1990s, many laboratories could accurately detect blood lead concentrations less than 1 mg/dL. On the other hand, lead in blood represents only a small fraction of the total lead body burden and there is extensive turnover of lead in the body. Thus, the blood lead concentration at any given point of time can be considered to reflect both current and longer term past lead exposure (Mushak, 1989; Gulson et al., 1995, 1997, 1998a, 1998b, 1998c; Smith et al., 1996). It has been posited that lead in blood plasma is a better measure of lead available for internal transport to target tissues (e.g., neural tissues) and that such lead may, therefore, be a preferable measure of lead dose. However, it has been very hard to accomplish measurement of lead in plasma free of contamination, mainly due to the destruction of red cells in preparation of plasma samples. Because more than 95% of lead in blood is in red cells (DeSilva, 1981), even a small degree of contamination by red cell material could markedly alter plasma lead results. Urine lead reflects lead in blood in the sense that their stable lead isotope profiles are highly correlated, but lead concentrations in these two biological fluids are typically only weakly correlated (Gulson et al., 1998b). Thus, urine lead concentrations cannot serve to predict blood lead concentrations, particularly at exposures associated with blood lead values 510 mg/dL. Measurements of lead in urine may pose significant problems for a variety of reasons. First, as noted elsewhere in this chapter, urine lead has a complex relationship with lead dose. Also, the concentration of urine itself is affected by fluid intake. Thus, the variability of urinary lead excretion during the day has been found to be rather problematic when attempting to express lead dose (Gulson et al., 1998b). Variations in the concentration of urine itself (related to fluid intake) and the marked variability of urine lead excretion throughout the day mandate careful quantitative urine collections, which can be quite difficult to accomplish for children and experimental animals. Yet another problem is the contamination of urine samples with feces or other body products. A variant of urine lead measurement, often used diagnostically but infrequently in research studies, is the measurement of the amount of lead in urine after administration of a chelating agent. Such measurements presumably sample a larger pool of lead than unstimulated urinary lead excretion and have been held to define the “chelatable” pool of

BIOMARKERS

783

lead that should be available for further therapeutic chelation therapy. However, the “chelatable” pool varies with the agent used, and it is not particularly clear, in any real sense, as to what the “chelatable” pool represents biologically or how it can be directly related to response measurements. Because, with few exceptions, it is not routinely possible to directly sample body tissues other than blood or urine in living subjects, the measurement of lead in samples of other tissues has generally seen only very limited use. For example, since lead in the dentine of teeth represents a stable pool of lead reflecting cumulative past lead exposure, the measurement of lead in dentine of shed deciduous teeth has been used for some research on the effects of early lead exposure in children (Needleman et al., 1979). Also, bone biopsies have occasionally been used to assess quantities of lead in the more stable and larger pools of lead in the body (Rabinowitz et al., 1976; Aufderheide and Wittmers, 1992). Lastly, although sporadic efforts have also been made to use samples such as hair, nails, or saliva to assess lead exposure or dose in living subjects, all have been found to be unsatisfactory for general use for one reason or another and, at best, have only very limited applicability. Most direct measurements of lead in internal organs and tissues have been done on autopsy subjects (e.g., Barry, 1975, 1981; Drasch et al., 1974, 1997; Drasch et al., 1987; Drasch and Ott, 1988). However, the development of methods for noninvasive measurement of lead in tissues of living subjects has begun to emerge as a potentially viable alternative means for assessing lead levels in internal organs. In particular, the development of approaches to measurement of lead in bone in living subjects has progressed notably during the last decade or so (Nordberg et al., 1991). The human skeleton contains the great majority of body lead burden. The inactivity of the skeletal lead deposits was thought to reflect a very long half-time of lead in bone, and it was generally assumed that bone was homogeneous as a lead compartment and that the very long half-time would greatly delay transfer of lead from bone back to other tissues. Based on data from stable isotope studies, this is no longer a defensible concept. Current evidence indicates that bone is not only a set of compartments for lead deposition but is also a target of lead toxicity itself. Human bone appears to have at least two kinetic compartments for lead. Trabecular (spongy) bone lead is more mobile than lead stored in long, dense, or cortical bone (Skerfving, 1998). Also chelatable lead is well correlated with trabecular but less so with cortical bone (Schutz et al., 1987a, 1987b). In adults, long- term tissue (presumably largely bone) stores of lead contributed between 50 and 75% of the lead present in blood (Gulson et al., 1995, 1997; Smith et al., 1996). Young children, due to constant skeletal turnover during physiological remodeling processes that accompany somatic growth, recycle lead between bone and other tissue compartments. Rosen et al. (1989) reported that cortical bone (e.g., tibia) lead is correlated with and predictive of chelatable lead. During the past few decades, X-ray fluorescence methods have been developed to measure lead in bone noninvasively (Committee on Measuring Lead in Critical Populations,1993). Two general groups of XRF techniques can be distinguished based on their sampling of the fluorescence emitted either by k-shell or by l-shell electrons following radiation from an X-ray machine or other radiation source. Analyses of dosimetry, volume sampled, and precision for these instruments have been provided in the National Research Council’s report “Measuring Lead Exposure in Infants, Children, and Other Sensitive Populations” (1993). Such XRF methods have gained some use in a few epidemiological studies (e.g., see Hu, 1998), having been applied most successfully to groups with high lead exposures, for example, for persons living in high lead exposure environments, those occupationally exposed, or overtly lead-poisoned children. One concern is that the

784

LEAD AND COMPOUNDS

quantitation limits and precision of the XRF instruments may be too high for use in a general population (Rosen and Pounds, 1998). The rate of improvement in quantitation limits and precision of these instruments appears slower than the rate of declining levels of bone burden of lead in many countries. A wide range of methods have been employed, with varying success, to define the biological effects of lead. Such methods range from in vitro biochemical testing to observation and measurement of behavioral attributes of exposed subjects. Among the widely diverse array of biological effects shown to be caused by lead, the impairment of heme synthesis has long been recognized as being a key class of pathophysiological effects highly responsive to variations in lead exposure/dose. With the reliable detection of lead-induced impacts at several points along the heme synthesis pathway in red blood cells being facilitated by sampling of readily available blood in living subjects, certain indicators of lead-induced impairment at salient points in the heme synthesis pathway were used for many years as biochemical measures (i.e., biomarkers) of lead exposure and/or toxicity. Particular emphasis has been placed on the inhibition of two enzymes in the heme biosynthetic pathway, porphobilinogen synthetase and heme synthetase (Sassa et al., 1973; Piomelli, 1973). Porphobilinogen synthetase has been measured by activity levels and characterized by electrophoresis as to its phenotypic variability (Doss et al., 1982). Inhibition of heme synthetase, as a mitochondrial bound and dependent enzyme, has been gauged primarily by the accumulation of its porphyrin precursors, especially photoporphyrin IX. Accumulation of zinc protoporphyrin is strongly and logarithmically correlated with blood lead concentrations in both children (e.g., Piomelli et al., 1973, 1982; Roels et al., 1976) and adults (e.g., Grandjean and Lintrup, 1978; Lilis et al., 1978). Among children, the threshold for response is thought to fall within a blood lead concentration range of 15–20 mg/dL whole blood (Piomelli et al., 1982; Hammond et al., 1985). Analysis of free erythrocyte protoporphyrin (FEP) had previously been used in screening children to identify lead poisoning. However, because accumulation of protoporphyrin is also seen with iron deficiency, increased FEP is not a change specific only to lead, and because iron deficiency is sufficiently common among lead-exposed populations, the measurement of erythrocyte protoporphyrin (EP) poorly distinguishes between iron deficiency and lead excess (Mahaffey and Annest, 1986). This as well as the threshold for EP change being higher than for blood lead levels that are now recognized to be of concern based on neurobehavioral changes has led to curtailing of use of erythrocyte protoporphyrins as a lead exposure screening method. Another area of biochemical investigation, that is, the evaluation of levels of neurochemical mediators in blood and urine, has also been of some interest, but has thus far produced often conflicting results in the hands of different investigators. It appears clear that in consideration of all areas of biochemical investigation, such markers have not produced, to date, reliable and valid measures of response to lead exposure, at least in regard to neurological effects of lead. It remains to be seen if any other types of lead-induced changes in signature indicators of lead-related neurotoxicity, especially any of those resulting from decreasingly lower lead exposure levels, may emerge as useful biomarkers in the future. One possibility might be the use of certain electrophysiological changes, for example, altered brainstem auditory evoked potentials (BAEPs), that are discussed in the next section. Or, perhaps, further advances in the pioneering use of magnetic resonance imaging (MRI) or of magnetic resonance spectroscopy (MRS) methods to detect lead-induced neurotoxic effects (also discussed below) may ultimately generate new results that may prove to be useful in deriving new biomarkers for lead.

HEALTH EFFECTS

785

20.8 HEALTH EFFECTS As noted earlier, it has long been known that high-level lead exposures can cause quite serious health effects in human children and adults. Such effects include severe neurological, renal, and hematological impairments that typify classically defined lead ‘poisoning” or “intoxication” that continued to be of much medical concern well into the latter part of the twentieth century. However, extensive research conducted especially since the 1960s–1970s has led to the recognition that lead exerts notable impacts on many different tissues and organ systems, including some impacts seen at very low exposure levels extending down to only slightly above current U.S. population means. This has been accompanied by parallel shifts in public health protection/medical attention to focus on reducing human lead exposures as the primary approach to dealing with lead as a continuing important public health issue. Reflecting this widespread shift of public health/medical interest in many countries, key emphasis is accordingly placed here on the discussion of health effects induced by low-level lead exposures. Unfortunately, the present space limitations preclude comprehensive discussion here of the full range of the rapidly expanded new information on lead-induced health impacts. Thus, rather than attempting to summarize the entire broad scope of newly documented lead effects here, the ensuing discussion focuses most strongly on neurotoxic impacts that, as an overall class, have come to be seen as key “signature” or “critical” effects of low-level lead exposures. A few examples of newly demonstrated low-level lead exposure impacts on other organ systems (e.g., renal, cardiovascular, immune) are also highlighted, and the reader is referred to other reviews for much fuller discussions of lead impacts on various organ systems and functions. 20.8.1

Neurotoxic Effects of Lead

Among the best known and most widely recognized “subclinical” impacts of low-level lead exposures are decrements in IQ and impacts on other global measures of neurocognitive abilities. Following pioneering work in the 1970s, for example, by Needleman et al. (1979), numerous epidemiological studies during the 1980s and 1990s evaluated the lead effects on the higher order integrated neurological functions (as indexed by lead impacts on intelligence measures, perceptual-motor coordination, and various other neurobehavioral end points). As noted in the prior version of this chapter, such studies (e.g., see Bellinger et al., 1986; Baghurst et al., 1992; Dietrich et al., 1990, 1993; Wasserman et al., 1997) substantiated that lead has adverse neurocognitive effects at very low levels of exposure, especially of the fetus and infant, as demonstrated by investigations in multiple cultures. Several major prospective, longitudinal epidemiological studies were highlighted as having shown impaired intellectual functioning in childhood following increases in blood lead across a range of approximately 10–30 mg/dL whole blood, even after control for social and demographic conditions associated both with exposures to lead and with lowered (slowed) development scores (Bellinger et al., 1986; Dietrich et al., 1993a, 1993b; Baghurst et al., 1992; Wasserman et al., 1997). These studies found an approximate 4–6-point decrease in subsequent IQ (when measured at about age of 6 or 7 years, but not earlier). Of interest, analogous to some earlier reports by Baghurst et al. (1995) and Dietrich et al. (1993) of lead impairment of visualmotor integration at this range of exposures, lead impacts on perceptual-motor skills were also found by Wasserman et al. (1997) and were suggested as possibly being more sensitive to lead exposure than language-related aspects of intelligence. Despite the demonstration of neurocognitive effects across geographic areas, social class, and cultures by the above

786

LEAD AND COMPOUNDS

studies, it should be noted that such findings were not obtained by all other analogous contemporary studies. More recent studies have continued to find evidence of impaired cognitive abilities being associated with low-level lead exposure of children, even at still lower levels than those previously thought to be harmful. Extensive, detailed reviews of such studies and those of other types of neurotoxic effects of lead in children and adults can be found in the recent U.S. EPA (2006) lead assessment and in the overview of neurotoxic effects of lead in humans by Bellinger (2008). For example, as discussed in U.S. EPA (2006), in the largest available new cross-sectional study, Lanphear et al. (2000) found relationships between the blood lead concentrations and the cognitive deficits in a nationally representative sample of 4853 U.S. children (all NHANES III participants), aged 6–16 years (having a geometric mean blood lead value of 1.9 mg/dL, with 97.9% being below 10 mg/dL). In multivariate analyses, significant covariate-adjusted associations were found between the blood lead levels and the two subtests for visual-motor skills and for short-term and working memory for all children and for those with blood lead 510 mg/dL, as well as with the visual-motor subtest for children with blood lead levels 57.5 mg/dL. U.S. EPA (2006) also noted that numerous other recent longitudinal studies have consistently observed effects on IQ in children at blood lead levels 510 mg/dL. Perhaps of most import, a large international pooled analysis of 1333 children from 7 different cohorts (by Lanphear et al., 2005) was highlighted as estimating a 6.2 point decline in full-scale IQ per increase in blood lead concentration from 1 to 10 mg/dL. Also of much importance is an observation across several of the new studies of a nonlinear relationship between blood lead concentrations and IQ or other neurobehavioral outcomes, with larger impacts being seen per unit increase in blood lead levels below 10 mg/dL than above that level. This otherwise nonintuitive dose–response relationship may be plausible, as noted by U.S. EPA (2006), if different underlying biological mechanisms (e.g., early CNS neurodevelopmental processes) are initially affected at relatively lower lead exposures than are other processes that may be disrupted in producing classic indicators of frank lead poisoning, with the dominant mechanisms at low exposure levels perhaps being very rapidly saturated versus less rapidly saturated mechanisms becoming predominant at higher exposure levels. However, it should be noted that although the reported nonlinear relationship between lead effects and neurocognitive functions appears to be gaining wide acceptance, some associated controversy remains, as reflected by the note by Bowers and Beck (2006) and the ensuing series of comments published in Neurotoxicoloy, Vol. 27, 2006, and Vol. 28, 2007. As stated in the prior version of this chapter, other neurocognitive changes and long-term educational, behavioral, and social consequences of low-level lead exposure have also been identified. For example, Bellinger et al. (1992) reported a higher rate of retention in grade and other results reflecting learning difficulties among higher blood lead children. Several more recent studies, as summarized in U.S. EPA (2006), have since confirmed analogous low-level lead exposure impacts on academic achievement. In one study highlighted by U.S. EPA, Lanphear et al. (2000) used multiple linear regression analyses of standardized academic achievement measures for the 4853 NHANES III children noted above as aged 6–16 years (mean blood lead, 1.9 mg/dL) and found blood lead to be significantly related to decrements in both reading and arithmetic achievement scores. Such decrements were also found in analyses stratified by blood lead concentrations to be inversely related to blood lead for those children with concurrent blood lead values 55 mg/dL. In yet another study noted by U.S. EPA (2006), from among 533 girls aged 6–12 years in Riyadh, Saudia Arabia (having a mean blood lead of 8.1 mg/dL), percentile of class rank was significantly associated with blood lead

HEALTH EFFECTS

787

level in a subset of those with blood lead levels 510 mg/dL (Al-Saleh et al., 2001). Also, significant associations were noted as being found between blood lead values (mean of 11.4 mg/dL) and poorer math and vocabulary scores achieved by 594 second graders in Mexico, with segmented regression analyses showing the slope for the lead effect to be significantly steeper for blood lead values below 10 mg/dL (T
ellez-Rojo et al., 2006). The effects on academic achievement seen in the above studies were statistically significant even after adjustment for IQ, thus raising the possibility (as posed by U.S. EPA, 2006) that the impairment of neurocognitive functions besides those indexed by global intelligence measures may contribute to lead-induced impacts on learning and academic achievement. It should also be noted, as per U.S. EPA (2006), that academic achievement decrements seen in the above studies may possibly be attributable to earlier (but unmeasured) higher pediatric blood lead levels (that usually peak before 3 years of age, then decline). As for lead impacts on neurobehavioral end points besides the above global measures of intelligence or academic achievement, U.S. EPA (2006) also noted that epidemiological studies have evaluated lead effects on more specific cognitive abilities, for example, attention, memory, visuo-spatial processing, and executive functions (impulse control, planning, other integration of higher order cognitive processes, etc.). Recent such studies were further noted as having shown relationships between blood lead and impacts on attentional behaviors and executive function among cohorts of children (varying in age range from 4–5 years to 19–20 years), even in those cohorts with more than 80% of subjects having concurrent blood lead values 510 mg/dL. Epidemiological studies were further noted as having demonstrated childhood lead exposure to be associated with disruptive/antisocial behavior, with such effects apparently persisting into adolescence and early adulthood. As an example, in following from ages 7–11 years, the same cohort of children shown by Bellinger et al. (1992) to have lead-related increased retention in grade, Needleman et al. (1996) found that lead exposure also increased risk for antisocial and delinquent behavior at 11 years of age. Dietrich et al. (2001) also reported behavioral disturbance and/or delinquency among young adults to be significantly related to blood lead measures obtained for them at various earlier time points (prenatally, at intervals during infancy and childhood, etc.) during participation in the Cincinnati prospective cohort lead study. Analogous long-term effects were reported by Burns et al. (1999), who observed that increasing blood lead across the range of 10–30 mg/dL adversely affected the behavioral and emotional development of children in the Port Pirie, Australia, cohort when evaluated at ages 11–13 years. In the same children, increasing blood lead concentration across the range of 10–20 mg/dL was also found to be associated at such ages with a three-point decline in mean IQ (Tong et al., 1996). Lastly, Needleman et al. (2002) found bone lead to be one of the strongest predictors of adjudicated delinquency among high-school-aged White and African-American subjects living in the Pittsburgh, PA, area. Specific neurological substrates and biochemical mechanisms that may be perturbed by lead in contributing to increased behavioral disturbance/ antisocial behavior remain to be more definitively characterized. However, U.S. EPA (2006) indicated that Lidsky and Schneider (2003) have noted that lead affects numerous brain sites and processes involved in impulse control, and Needleman et al. (2002) proposed that lead impacts on cognitive function and academic performance may indirectly contribute to antisocial behavior/delinquency. In addition to the above types of effects on intelligence and other higher order integrative functions, lead has also been shown to exert notable effects on more basic levels of sensory, motor, and sensory-motor integrative functions in children. For example, as noted by U.S. EPA (2006), epidemiological studies have shown lead effects on hearing thresholds and

788

LEAD AND COMPOUNDS

other features of auditory processing in children that appear to persist into young adulthood, including(a) observation by Schwartz and Otto (1987) of significant elevations in pure-tone hearing thresholds (at frequencies within the range of human speech) among more than 4500 NHANES II participants (aged 14–19 years); (b) replication by Schwartz and Otto (1991) of such findings in a sample of approximately 3000 subjects (6–19 years old) in the Hispanic Health and Nutrition Examination Survey (HHANES), even at blood lead levels below 10 mg/dL; (c) observation by Dietrich et al. (1992) of associations between higher prenatal, neonatal, and later postnatal blood lead concentrations in 215 children from the Cincinnati Lead Study and poorer scores on a test of central auditory processing (SCAN) at age 5 years; and (d) demonstration by Osman et al. (1999) of significant associations between concurrent blood lead levels and increased hearing thresholds among 155 children (aged 4–14 years) in Poland, which remained significant in analyses restricted to data for blood lead levels 510 mg/dL. As also noted by U.S. EPA (2006), Bellinger (1995) has suggested that such lead-induced impacts on hearing and auditory processing may be a mechanism contributing to learning impairment by lead. The above studies indicate that such sensory effects occur at rather low exposure levels (e.g., blood lead concentrations 510 mg/dL), although it should be noted again that some of the observed auditory effects could, potentially, derive from somewhat higher earlier peak blood lead concentrations. Several epidemiological studies have also characterized lead-related neuromotor deficits at relatively low levels of exposure, as discussed in U.S. EPA (2006). Dietrich et al. (1993a , 1993b), for example, reported that both pre-and postnatal blood lead concentrations were significantly related to poorer scores on tests of bilateral coordination, visual-motor control, upper limb speed and dexterity, fine motor control, and postural stability among Cincinnati Lead Study children at 6 years of age, with strongest associations being seen with concurrent lead levels (mean of 10.1 mg/dL). Later, 78-month postnatal blood lead levels were associated with poorer fine motor skills at the age of 16 years. Other studies were also noted as finding(a) associations between lifetime average blood lead levels through 54 months of age and poorer fine motor and visual motor function among Yugoslav children (Wasserman et al., 2000a), and (b) significant associations between blood lead values (mean 5.0 mg/dL) and increased reaction time, postural sway oscillations, and action tremor among 110 preschool Inuit children in Canada, even when data from the 10% of the children having blood lead levels 410 mg/dL were excluded from analyses. In addition to the above types of impacts, lead has also been shown to affect various electrophysiological measures of sensory and motor neurological responses. The prior edition of this chapter noted that peripheral nerve conduction (Seppalainen et al., 1979), as well as visual (Araki et al., 1987) and auditory (Schwartz and Otto, 1987) brainstem responses, was altered at relatively low lead exposure levels. Also highlighted were other epidemiological findings of significant lead-related impacts on visual-evoked potential interpeak latencies being seen (Altmann et al., 1998) in an environmentally exposed population of 6-year-old children, with mean blood lead of 4.2 mg/dL and 95th percentile value of 8.9 mg/dL (Walkowiak et al., 1998). During the past 10 years or so, certain innovative new approaches have begun to be used to evaluate neural substrates and neurochemical processes potentially perturbed by lead and possibly contributing to neurobehavioral impacts of lead. For example, U.S. EPA (2006) noted that MRI and MRS have come to be used to evaluate lead-exposed children, with several studies comparing subjects with elevated blood lead levels (420 mg/dL) to control subjects (blood lead 510 mg/dL). It was highlighted that although all subjects had normal MRI, the elevated lead subjects showed significant reductions in the ratios of

HEALTH EFFECTS

789

N-acetylaspartate to creatine and phosphocreatine in frontal gray matter relative to the lower lead control subjects (Trope et al., 2001) and reduced peak values of choline, creatine, and N-acetylaspartate were seen in all the four brain regions of the lead-exposed children (Meng et al., 2005). One or the other such lead effects, it was suggested, could be related to lower neuronal density/neuronal loss, decreased cell membrane turnover, myelin alterations (possibly leading to CNS hypertrophy), or less neuronal cell viability. Also, in another study using functional MRI methods to follow a subsample of 48 young adults (aged 20–23 years) from the Cincinnati Lead Study, higher childhood average blood lead levels were found to be associated with reduced activation in Broca’s area (a brain area involved in speech production) while performing an integrated verb-generating/finger tapping task (Cecil et al., 2005; Yuan et al., 2006). The prior version of this chapter also noted that (a) the acute neurological damage in adults due to high levels of inorganic lead exposure and neurological consequences of exposure to organic lead compounds have been recognized for decades, and (b) additional findings on adverse neurologic effects of lead, including at exposure levels not previously recognized as harmful to the adult, have been elucidated within the past 10 years. For example, Hanninen et al. (1998) reported that in the studies of workers whose blood lead levels had never exceeded 2.4 mmol/L (50 mg/dL), lead was found to be associated with decrements in visual-spatial and visual-motor functions, verbal comprehension, and attention, as well as increased symptoms of impaired well-being as rated by psychological assessments of mood, whereas blood lead increases approximately from 2.4 mmol/L (50 mg/ dL) to 4.9 mmol/L (100 mg/dL) caused persisting, possibly permanent impairment of CNS function (Hanninen et al., 1998). 20.8.2

Other Effects

The previous edition of this chapter noted that adverse renal system effects typically have been described in many earlier reviews of high lead exposure effects. It further noted that acute effects of exposure to high concentrations of lead result in proximal tubule damage manifested by glycosuria and amino acid uria and that overt nephropathy appears to develop when blood lead levels exceed a threshold of 60 mg/dL (about 2.9 mmol/L), as reviewed by Loghman-Adham (1997). However, early renal tubule dysfunction secondary to far lower levels of lead exposure was also indicated as having been detected by measuring of urinary excretion of low molecular weight proteins (81 and 92 microglobulins, retinol binding protein) or the lysosomal enzyme NAG, as well as other brush border proteins (Loghman-Adham, 1997). A cross-sectional study by Fels et al. (1998) was further highlighted as comparing changes in urinary or serum markers of function or integrity of specific nephron segments in children, with those having a mean blood lead concentration of 13 mg/dL showing increased excretion rates for prostaglandins and thromboxane B2, epidermal growth factor, 92-microglobulin, and Clara cell protein compared to children with mean blood lead levels averaging less than 4 mg/dL. This overall pattern of glomerular, proximal, and distal tubular, and interstitial markers was noted as being similar to that earlier found among adults, but occurring at lower blood lead concentrations than in adults (Fels et al., 1998). The clinical significance of these changes in excretion of low molecular weight proteins and/or lysosomal enzymes was noted, however, as not yet being fully clear. Additional evidence derived from several U.S. and European general population studies during the past 2 decades substantiates further that low-level lead exposures affect the above

790

LEAD AND COMPOUNDS

or other indicators of altered renal function, as discussed in recent reviews/assessments (Gonick, 2002; U.S. EPA, 2006). For example, U.S. EPA (2006) noted that the Belgian Cadmibel study(a) was the first large environmental study to adjust for multiple renal risk factors; (b) evaluated several renal outcome measures among a general adult European population; and (c) found decreased creatinine clearance to be related to blood lead and zinc protoporphyrin concentrations in both men (blood lead mean of 11.4 mg/dL and range of 2.3–72.5) and women (blood lead mean of 7.5 mg/dL and range of 1.7–60.3 mg/dL), thus raising concerns that the exposure/dose threshold for adverse lead effects on renal function among the general population might be much lowered than earlier thought based on studies of occupationally exposed workers. U.S. EPA (2006) also highlighted several other published analyses (Payton et al., 1994; Kim et al., 1996; Wu et al., 2003; Tsaih et al., 2004) of data from the Normative Aging Study (a long-term study of Boston area adult men, aged 21–80 years, with participants initially having been recruited in the 1960s and undergoing periodic follow-up evaluations), which demonstrated relationships between blood and/or bone lead concentrations and various indicators of reductions in measured or estimated creatinine clearance, including among men with peak blood lead levels below 10 mg/dL in some analyses. Also highlighted by U.S. EPA (2006) were the results of analyses by Muntner et al. (2003) of associations between blood lead and renal outcomes among 15,000 NHANES III adult participants during 1988–1994. The analyses were stratified because of an interaction between blood lead and the hypertension (HTN) (as per leadrelated cardiovascular effects noted later), with (a) mean blood lead concentrations being 4.2 mg/dL in hypertensive and 3.3 mg/dL in normotensive subjects; (b) the prevalence of elevated serum creatinine levels (indicative of reduced renal clearance of creatinine) being about 10 times higher in hypertensive (11.8%) than in normotensive (1.8%) subjects; and (c) higher blood lead levels being associated with a higher prevalence of chronic kidney disease in diabetics among the nonhypertensive subjects. Lastly, U.S. EPA (2006) also noted ˚´ kesson, 2006) to be significantly related ˚´ kesson et al., 2005; A that blood lead was reported (A to indications of altered renal function (e.g., reductions in estimated creatinine clearance) among 820 women, aged 53–64 years, in the Lund area of Sweden, with the association being apparent over the entire blood lead range (mean blood lead of 2.2 mg/dL). U.S. EPA (2006) also discussed an interesting quantitative comparison (see Chapter 6, Figure 6.8 in the U.S. EPA document) of estimated impacts of lead exposure on renal creatinine clearance derived by the various analyses discussed above and noted that the slopes from those studies fell in the range of 0.2 to minus 1.8 mL/min change in creatinine clearance per 1 mg/dL blood lead increase. Overall, the above analyses appear to substantiate well that the impairment of renal function by lead occurs at much lower exposures/doses than those previously thought to be harmful for nonoccupationally exposed adults in the general population. This appears to include lead exposures resulting in blood lead levels extending to well below 10 mg/dL, but one cannot completely rule out the possible attribution of effects observed late in adulthood in some studies to somewhat higher (but still likely rather low) unmeasured peak exposure levels that may have occurred earlier in the study participants’ lives. As noted at the outset of this section, lead has now come to be recognized as exerting notable effects, including at relatively low exposure levels, on a number of organ systems besides those (neurologic, renal, etc.) classically recognized a key target organs for the metal. Among the most important classes of newly demonstrated types of effects are cardiovascular impacts of lead exposure. U.S. EPA (2006) assesses key information on leadrelated cardiovascular effects that has emerged during the last several decades. U.S. EPA

HEALTH EFFECTS

791

(2006) review noted that epidemiological studies that have examined relationships between human blood lead levels and blood pressure have generally found positive associations, even after controlling for confounding factors such as tobacco smoking, exercise, body weight, achohol consumption, and socioeconomic status. It further highlights recent meta-analyses of such studies that found robust, statistically significant, though small effect size, associations between blood lead concentrations and blood pressure. The Nawrot et al. (2002) metaanalysis was cited as an example in finding that a doubling of blood lead corresponded to a 1 mmHg increase in systolic blood pressure, and it was noted that while not necessarily being clinically meaningful for a given individual, a population shift in blood pressure of 1 mmHg is likely of concern in terms of associated increased risk for more serious cardiovascular outcomes (heart attack, cerebrovascular events, etc.). U.S. EPA (2006) further noted that the majority of recent studies evaluating relationships between the bone lead and the cardiovascular effects found strong associations between the long-term lead exposure (indexed by bone lead concentrations) and the arterial blood pressure. Also noted, too, was highly supportive evidence from numerous animal studies showing that low-level lead exposures for extended periods of time result in the eventual onset of arterial hypertension, which persists long after exposure cessation, with both in vivo and in vitro toxicology studies (Gonick et al., 1997; Vaziri et al., 1997) providing strong evidence that oxidative stress, at least in part, plays a key role in mediating lead-related HTN. An excellent review of pertinent literature on this subject has been provided by Vaziri (2002). Impacts on the immune system have also emerged during the past 20–30 years as yet another important class of lead-related toxic effects subsumed as part of a much broader array beyond those earlier classically defined as typifying lead toxicity, again including notable lead impacts seen with rather low-level exposures. Such lead-related immune system effects have recently been assessed by U.S. EPA (2006) and in a review by Dietert and Piepenbrink (2006). Very notably, as summarized well in U.S. EPA (2006), lead effects on nonhuman animal immune systems appear to include the targeting of T cells and macrophages, with lead-induced alterations being typified by (a) an increased inflammatory profile for macrophages (e.g., elevated tumor necrosis factor-alpha, oxygen radical and prostaglandin production) and (b) skewing of the T cell response away from T helper 1(Th1)-dependent functions toward T helper 2 (Th2)-dependent functions. Resulting impacts on immune system function, as noted by U.S. EPA (2006), include increased production of Th2 cytokines (e.g., IL-4, IL-10) and certain immunoglobulins (e.g., IgE), decreases in Th1-associated cytokines (interferon gamma and IL-12) and Th1 functions (e.g., the delayed-type hypersensitivity, or DTH, response); but not major immune cell population changes (which complicates interpretation of human epidemiological study results). Also noted was an approximate order-of-magnitude age-related difference in immune system sensitivity to lead between the perinatal period and adulthood, with lead effects on immune function being seen at blood lead levels distinctly below 10 mg/dL following gestational or perinatal exposures. Also emphasized was a key point noted by Dietert and Piepenbrink (2006), that is, given that the most informative sources of functionally reactive immune cells (e.g., antigen-reactive ones in lymphoid organs and lymph nodes) are not readily accessible in humans, the circulating lymphocytes and serum or plasma immunoglobulin (e.g., IgE) levels must serve as readily accessible surrogate indices of immune status in human studies. Most importantly, it was highlighted that similar immune system effects have been observed in both humans and laboratory animals in terms of positive associations being seen between the circulating IgE levels and the bood lead concentrations following early life lead exposures, even at blood lead levels below 10 mg/dL.

792

LEAD AND COMPOUNDS

As for the other types of lead health effects, the prior edition of this chapter included brief discussion of reproductive impacts, noting that comparatively recent investigations of the reproductive effects of lead had most extensively focused on male reproductive toxicity (as reviewed by Apostoli et al., 1998). Perhaps of most note is the fact that based on review of 32 experimental animal studies, 22 human epidemiological studies, and one case report in humans, Apostoli et al. (1998) concluded that when blood lead concentrations exceed 40 mg/dL, there appears to be associated decreases in sperm count, volume, motility, and morphological alternations (with a possible effect also on endocrine profile), but dose– response relationships, particularly possible thresholds for effects, remained poorly understood. More recently, Bellinger (2003) and U.S. EPA (2006) have provided additional overviews of lead-related reproductive effects. Also of note is the fact that during the past 20–30 years or so, lead and lead compounds have been recognized as being carcinogenic to animals (IARC, 1987). The prior edition of this chapter also noted that better documentation of possible lead carcinogenicity among humans has begun to emerge, and cited a review of the carcinogenicity of lead by Vainio (1997) following publication of two cohort studies among smelter workers (by Lundstrom et al., 1997 and by Cocco et al., 1997). Overall, Vainio (1997) concluded that a long-term, highlevel exposure to lead compounds is associated with an increased risk of cancer and that the “weight of evidence is beginning to be convincing enough concerning kidney and even lung cancer” for humans. In addition, more recent extensive reviews have been published by both Silbergeld (2003) and U.S. EPA (2006) which evaluate available human and/or animal evidence concerning carcinogenicity effects of lead. The broad lead assessment by U.S. EPA (2006) covers the latest available information not only on all of the types of lead health effects discussed above, but also other classes of lead effects not addressed to any great extent here due to space limitations. It is recommended that the reader must also consult the discussion by Hu et al. (1998) of lead effects on bone and teeth.

20.9 MECHANISMS UNDERLYING LEAD TOXICITY The prior edition of this chapter noted that early investigations into lead toxicity focused on mechanisms underlying the relatively gross effects of lead observed at very high levels of exposure. For instance, the study of kidney pathology in lead-exposed animals and humans was noted as demonstrating deposition of lead in the cells of the proximal tubule, especially in nuclear material (Goyer and Rhyne, 1973), and such studies were also noted as documenting mitochondrial degeneration. Although these studies are consistent with the observed proximal tubular dysfunction seen in lead toxicity, it was also highlighted that they do not fully clarify the mechanism of this toxicity. Similarly, it was noted that studies of the heme biosynthetic pathway, though interesting, reveal certain inconsistencies that remain to be explained (Scott et al., 1971). Lead certainly inhibits several enzymes in this pathway, for example, porphobilinogen synthetase and heme synthetase in particular. However, of these, heme synthetase is of special interest, because it is a mitochondrial-bound enzyme, suggesting the possibility that lead inhibits this enzyme simply by altering mitochondrial function rather than by specifically affecting the enzyme per se. It was further noted that the mechanisms for nervous system toxicity as a major focus of attention remained unclear, but a number of important insights were highlighted. For example, the importance of lead effects on calcium-dependent systems appears to come

MECHANISMS UNDERLYING LEAD TOXICITY

793

as close as to an overarching explanation currently available. That is, ultimately, all lead toxicity may be related to calcium transport or to other mechanisms underlying alteration of functioning of calcium-dependent systems. Also, as noted elsewhere in this chapter, lead accumulates in the nervous system preferentially in areas very metabolically active and rich in mitochondria, and lead has been shown in vitro to be a potent inhibitor of mitochondrial function, possibly by competing with mitochondrial uptake of calcium. Lead compounds, it was noted, may have a variety of targets within the nervous system (Bondy, 1988), and before discussing what some of these targets might be, it may first be useful to classify broadly the types of lead compounds that are of environmental concern. As already pointed out, the inorganic lead compounds are generally of greater environmental concern. A brief discussion of properties of organolead compounds may help to clarify our knowledge about biochemical mechanisms of neurotoxicity for all classes of lead compounds. Consideration of the chemistry of organolead compounds suggests that they differ fundamentally in both chemical and biological properties from ionic compounds of the same metal (Grandjean and Grandjean, 1984). For example, it is well documented that the toxicity of tetraethyl lead results from the breakdown of the compound in the organism to a salt of triethyl lead. The triorganolead compounds form a very distinctive neurotoxic class. Their actions are probably brought about by two distinctive chemical properties: one depends on the lipophilic (rather amphoteric) nature of their chlorides and hydroxides and their affinity constants that allow dissociation at biological concentrations of hydroxyl and chloride ions; the other derives from the potentiality for five coordinate binding. Therefore, it is not surprising that the behavioral toxicity effects of alkyl leads do not closely resemble those of inorganic lead. Some similarities do exist that may be associated with the degradation of the alkyl lead to the stage of divalent lead in the organism. Even a small amount of metabolism of this type could become significant because of the much altered tissue distribution of lead as a result of the lipophilic properties of organolead compounds. This suggests that more attention should be paid to the possible complexes of inorganic lead with hydrophobic ligands such as hemic acid. The general lack of involvement/importance of the divalent lead ion in organolead compound toxicity is also supported by the observation that the usual chelators have little effect on intoxication from organolead compounds. The study of chemical mechanisms of inorganic lead compound toxicity is complicated by a number of factors. As already mentioned, the amount of lead absorbed after its oral administration can vary significantly depending on animal species, age, diet, and on both the chemical and physical form of the inorganic lead compound ingested. For example, in the absence of food in the gut (as during fasting), the lead compound can apparently be more readily acidified and solubilized for tissue uptake. Because of the rather ubiquitous occurrence of lead in the environment and its relatively high toxicity, one must also be concerned about lead contamination of the diet, and even of laboratory reagents (Simons, 1989) used in studies in vitro. Separation of the influence of nutritional status on biokinetics of lead from specific neurotoxic events is another problem that needs more attention. There are many ostensibly different toxic effects of lead described in the literature. It seems likely that there are only a few triggering events in biological systems that would account for the rather potent neurotoxic effects of lead. These in turn could result in a range of secondary effects. It follows also that the important initiating events will depend in some way on rather distinctive chemical properties of divalent lead, since all other divalent metals produce a toxic syndrome that is qualitatively or quantitatively different from that of lead. Considerations of lead chemistry should also permit one to rule out certain possible initiating

794

LEAD AND COMPOUNDS

biochemical pathways. For example, the relatively low affinity of lead for sulfur suggests that interaction with specific sulfur-containing proteins such as metallothionein would not be critical. Carboxyl-oxygen-containing amino acids are more likely points of interactions with proteins, and this possibility is supported by the findings of lead-containing nuclear inclusion bodies in tissues that are rich in acidic amino acids (Choie and Richter, 1972). Since divalent lead is a reasonably stable oxidation state, it seems likely that the divalent lead ion in some way mediates the range of effects seen and that metabolic redox pathways involving tetravalent lead are not important. The toxic effects of divalent lead may be the result of either physical or chemical change in the biochemical systems with which it interacts. Lead may damage membranes by bringing about change in the ultrastructure of cellular components or by initiating oxidative damage. Membrane damage does appear to be a significant factor in lead toxicity, and myelin-containing membranes seem to be especially sensitive. Membrane damage by peroxidative processes may involve change in calcium homeostasis. Effects of lead on energy production could be related to direct interaction with mitochondrial membranes, altering ion transport, or changes in calcium homeostasis within the cell. Various studies have shown that lead accumulates in mitochondria, and this is associated with the inhibition of oxidative phosphorylation. Endothelial cells in brain capillaries contain three–five times more mitochondria than do endothelial cells in the vessels of other organs (Oldendorf and Brown, 1975). This difference may reflect the high rate of metabolic activity necessary to maintain the active transport of ions across the blood– brain barrier and may explain the susceptibility of these cells to a wide variety of toxic compounds that cause brain edema. Mitochondria may be a critical subcellular target for the toxic effects of lead. Inorganic lead has also been shown to compete with some essential divalent metals at several different levels (Chisolm, 1980). These include, in particular, calcium, zinc, copper, and iron. These interactions can occur at several levels, including absorption from the gut, transport across the blood–brain barrier, and at the synapse. For example, active calcium uptake by mitochondria is a critical process required to maintain calcium at very low concentration in the cytosol of cells. Inhibition of calcium accumulation by mitochondria may involve a direct blockage of the calcium pumps, but this may also be attributed to depletion of ATP or key intermediates involved in its synthesis, such as inorganic phosphate. No single metal deficiency shows symptoms identical to those seen in lead exposure. The chemistry of lead suggests that a variety of metals are likely to show altered distribution in biological systems in the presence of lead, with some associated biochemical changes. Dietary supplements of various minerals and vitamins do not completely protect against the toxic effects of lead, suggesting also that there are more critical biochemical changes associated with the potent neurotoxic effects of lead. Other biochemical interactions that may be important in explaining the toxic responses of lead include the possible inhibitory effects of lead on the cholinergic system and activation of catecholaminergic function. This may be related to the calcium agonist property of lead. Lead may be equipotent with calcium in binding to calmodulin. Lead may also be involved in direct reactions carried out by mixed-function oxidases. Complex secondary interactions between organ systems are also a possible factor determining the overall pattern of toxicity. A possible interrelated sequence of events by which lead compounds could cause neurotoxicity, as expressed by behavioral change, has previously been suggested (Bondy, 1988). What aspects of the chemistry of lead may throw light on the important biochemical/molecular mechanisms of toxicity? The importance of the ligand exchange chemistry of divalent lead in the

MECHANISMS UNDERLYING LEAD TOXICITY

795

overall expression of toxicity has already been pointed out. The relative binding strengths of the various endogenous ligands for lead will determine the fate and distribution of lead in vivo and its competition with other metals such as calcium. Once lead is inside the cells, its eventual fate will be to bind to sites that are stronger than cytosolic pool chelators such as citrate. At slightly acidic pH, at which divalent lead ionic species are likely to be present, phosphate ligands are prime candidates because of the very large association binding constants involved here. The extreme tenacity for phosphate may be the most distinctive feature of lead chemistry relative to other divalent metal ions of the same basic types. Several types of phosphate ligands must be considered, including inorganic orthophosphates, ATP, and the phosphate groups of membrane lipophosphates and phosphorylated proteins. This primary mechanism underlying lead toxicity is compatible with the major reproducible biochemical findings described above, particularly the suggestion that mitochondria (where oxidative phosphorylation takes place) may be a target site in cells. It is of course also compatible with the ultimate localization of the lead in bone as insoluble phosphate salts. It was also noted that lead may function as a phosphate scavenger and siphon off minority phosphate species crucial to developing/proliferating cells, especially in neural tissues. If one of the functions of calcium is to store phosphate in the form of calcium phosphate, then lead would be an efficient antagonist for this process. In fact it is, in general, not clear if the effects of lead on calcium homeostasis are the result of direct competition between lead and calcium for binding sites and/or of their differential affinity for phosphate ligands, particularly inorganic phosphate. For example, the report of Markovac and Goldstein (1988) indicating that lead is a potent activator of protein kinase C could be interpreted as a direct effect of lead on sequences of protein–phosphorylating–dephosphorylating that do not involve the enzyme at all (a control experiment in the absence of enzyme was apparently not run). Note that phosphorylating–dephosphorylating sequences are of critical importance to energy transformation in cells and tissues, and this is particularly true of those in the nervous system during rapid growth and maturation. It was noted that the possible direct phosphate-deleting effects of lead resemble in many ways those seen in poisoning by nitrophenols (Clayton and Clayton, 1981), which are known to uncouple oxidative phosphorylation (which presumably also reduces the body’s reservoirs of high-energy phosphate compounds). Such uncoupling apparently stimulates oxidative metabolism and, in turn, heat production of the body. Oxygen consumption, body temperature, respiration, and heart rate are all increased. Some similar effects are also associated with the hyperthyroid state (Dratman, 1978). With regard to mechanisms of lead neurotoxicity, it is interesting to note that there is some evidence that hyperthyroidism is associated with reduced catecholamine production rates in both the peripheral nerve tissue and the brain. In studies on rat, McIntosh et al. (1989) reported that the effects of lead on catecholaminergic and cholinergic transmission are regionally specific within the brain, the midbrain, and the diencephalon, showing the greatest degree of change in concentrations of neurotransmitters (dopamine concentrations usually decreased) and activities of ratelimiting enzymes. See also Cory-Schlecta (1997) for information on relationships between lead effects on neurotransmitter systems and behavioral toxicity. Although lead neurotoxicity, poisoning by nitrophenols, and hyperthyroidism all involve different triggering mechanisms, these responses may have in common certain secondary effects related to the maintenance of cellular homeostasis and metabolism. The remarkable affinity of lead for phosphate may be the most sensitive primary event to explain the fact that the neurotoxic effects of lead are evident at low concentrations of lead exposure and after only a brief exposure period. This suggests that thyroid status and metabolic state might be

796

LEAD AND COMPOUNDS

important factors to consider in examining correlations between the measures of lead exposure and could be causally related to increased oxidative metabolism. It should be possible to test this hypothesis at the biochemical levels, perhaps in some cases using the tools now available in molecular biology. Since the above overview of hypotheses/information related to lead toxicity (especially lead neurotoxicity) appeared, reviews by U.S. EPA (2006) and White et al. (2007) have been published that include evaluation of further advances regarding mechanisms underlying lead neurotoxicity. The White et al. (2007) review, for example, highlighted four major areas of key advances in lead neurotoxicity during recent years: (1) experimental studies showing that stress markedly influences lead effects, possibly mediated by interactions of corticosteroid hormones with components of the mesocorticolimbic dopamine system of the brain (with heightened stress causing consequent elevations in circulating corticosteroid levels hypothesized to contribute to increased vulnerability to many diseases and other dysfuctions among lower socioeconomic status populations); (2) cellular models of learning and memory used to evaluate possible mechanisms of lead-related cognitive deficits (with studies of long-term potentiation in the rodent hippocampus having shown lead-related increased thresholds, decreases of magnitude, and shorter retention times for neural plasticity, and alterations in the form of adult neurogenesis in the hippocampus, which may contribute to learning impairment); (3) in vitro evidence for strong binding of lead to glucose-related protein (GRP-78), induction of GRP aggregation, and blocking of secretion of interleukin-6 (IL-6) by astroglial cells (findings that implicate lead in “chaperone deficiency” processes, which in the long-term could underlie protein confrontational diseases, e.g., Alzheimer’s disease); and (4) implication of lead exposure in the early development and in the later progression of amyloidogenesis in rodent brains during old age (thereby contributing to increased proteins associated with Alzheimer’s disease pathology). Overall, as noted by White et al. (2007), such findings provide compelling evidence for lead exposures having adverse effects on the nervous system, that environmental factors increase nervous system susceptibility to lead, and that lead exposures early in life may contribute to neurodegeneration later in old age.

20.10

TREATMENT OF LEAD TOXICITY

The prior edition of this chapter noted that metal chelation therapy has been used with some success to treat lead poisoning (Bondy, 1988). It further noted, however, that chelators used for this purpose can also remove essential elements resulting in kidney damage, and most of these drugs tend to have unpleasant side effects, that they are also most generally useful for acute rather than sustained therapy, and that the benefits derived are usually only transitory, since blood lead can be rapidly replaced from bone stores. Several other important points were made as well. For example, in view of the nature of lead chemistry and the possibility that lead is basically functioning as a phosphate scavenger, it is unlikely that a complexing/ chelating agent of the usual variety could be found that could compete effectively with phosphate in a ligand exchange reaction, but it may be possible to develop a derivative of phosphate that is sufficiently reactive and excretable to be useful in therapy. On the other hand, soluble forms of phosphate itself might offer some protection against toxic effects, perhaps until lead is deposited in bone. This approach does not reduce the body burden of lead, but it may buy time during the process of deposition of lead in bone, which at least offers transient sequestering of lead.

TREATMENT OF LEAD TOXICITY

797

It was further noted that the prospects for successful chemical treatment of long-term, low-dose lead toxicity are not promising and a number of pertinent points were articulated, as reiterated with only very minor change here. Chelating agents have been used mainly to treat the short-term, high-dose exposures. For children, CDC (1991) has recommended the use of chelation therapy only when blood lead concentrations exceed 45 mg/dL (42.17 mmol/L). It remains unclear whether chelation has sustained benefits for children with blood lead of 25–44 mg/dL. Criteria such as the persistence of elevated blood lead levels despite environmental intervention have also been offered as justifications for chelation (American Academy of Pediatrics, 1993). Deciding which children may respond to chelation therapy is complex. It was found that children with changes in erythrocyte protoporphyrin and hematological index are more likely to respond to chelators with markedly enhanced urinary excretion of lead. Concern has also been raised suggesting that at least some chelators (i.e., succimer) are more effective in removing lead from blood than lead from brain (Pappas et al., 1995; Cory-Slechta, 1988). Smith et al. (1998) cautioned with regard to basing judgments on changes in blood lead concentration on predictions of the impact of chelators on brain lead concentrations. The ratios of change in brain lead and change in blood lead differed over the duration of chelation therapy in rodents. The usefulness of nutritional therapy depends on the timing of its introduction, the severity of lead exposure, and underlying nutritional status. It is clear that marginal nutritional status is associated with increased prevalence of elevated blood lead concentrations. Data from national epidemiological surveys such as the National Health and Nutrition Examination Survey, conducted in the United States during the 1970s through the l990s, demonstrated that young children from socially disadvantaged, low-income, minority families are more likely to have a greater prevalence of elevated blood lead levels (Mahaffey et al., 1982; Pirkle et al., 1994) and of marginal nutritional status (Life Science Research Office, 1996; Mahaffey et al., 1986). The consumption of a higher calcium diet has also been shown to be inversely related to bone lead among women living in Mexico City (HernandezAvila et al., 1996). The physiological changes that accompany poor nutritional status for calcium and iron result in enhanced absorption of these required elements from the GI tract (Fullmer, 1997; Mahaffey, 1995). Because lead absorption also increases with these physiological changes, improving nutritional status results in reduced future absorption of lead. Some reports indicate reduced blood lead following treatment with iron (Granado et al., 1994). Short-term correction of low calcium intake has not been shown to alter blood lead, but it is clear that skeletal mineral can be mobilized for calcium under conditions of physiological stress and that lead will be released along with calcium during this mineral mobilization (Gulson et al., 1997a, 1998a). Whether increased calcium intake reduces this mobilization is a question still to be addressed. Among recommended “treatment” activities is identification through screening of cases for environmental/nutritional/pharmaceutical intervention. In 1991, the “case” definition was lowered to define childhood lead poisoning as a blood lead of greater than or equal to 10 mg/dL (0.48 mmol/L) (CDC, 1991; American Academy of Pediatrics, 1993). Although the 10 mg/dL action level still remain in effect, in a move away from universal screening (CDC, 1997), emphasis is now on geographic areas with higher prevalence of older housing (defined as where 27% or more housing was built before 1950) or presence of other risk factors (e.g., the child receives services from public assistance programs for the poor, or the child has a sibling or playmate who has had lead poisoning). Part of this change reflects the overall decline in blood lead concentrations in the United States.

798

LEAD AND COMPOUNDS

REFERENCES ´˚ ´˚ kesson to Virginia Weaver with attached kesson A (2006) Personal communication [from Agneta A A GFR plots], April 12. ´˚ kesson AT, Lundh T, Vahter M, Bjellerup P, Lidfeldt J, Nerbrand C, Goran S, Str€omberg U, Skerfving A S (2005) Tubular and glomerular kidney effects in Swedish women with low environmental cadmium exposure. Environ. Health Perspect. 113:1627–1631. Alexander FW, Clayton BE, Delves HT (1974) Quart. J Med. New Series 43 (169):89–111. Al-Saleh I, Nester M, DeVol E, Shinwari N, Munchari L, Al-Shahria S (2001) Relationships between blood lead concentrations, intelligence, and academic achievement of Saudi Arabian schoolgirls. Int. J. Hyg. Environ. Health 204:165–174. Altmann L, Sveinsson K, Kramer U, Weishoff-Houben M, Turfeld M, Winneke G, Wiegand H (1998) Visual functions in 6-year-old children in relation to lead and mercury levels. Neuroroxicol. Teratol. 20:9–17. American Academy of Pediatrics/Committee on Environmental Health(1993) Lead poisoning from screening to prevention. Pediatrics 92:176–183. Angle CR, Manton W, Stanek YL (1995) Stable isotope identification of lead sources in preschool children — the Omaha study. J. Toxicol. Clin. Toxicol. 33:657–662. Apostoli P, Kiss P, Porru S, Bonde JP, Vanhoome M and the ASCLEPIOS Study Group.(1998) Male reproductive toxicity of lead in animals and humans. Occup. Environ. Med. 55:364–373. Araki S, Murata K, Aono H (1987) Central and peripheral nervous system dysfunction in workers exposed to lead, zinc, and copper. A follow-up study of visual and somato sensory evoked potentials. In. Arch. Occup. Environ. Health 59:177–187. Aub JC, Fairhall LT, Minot AS, Reznilcuff R (1925)Lead poisoning. In: Edsall DL, Howland J, Chesney AM, editors. Medicine, Vol. 4. Baltimore:Williams and Wilkins. Aufderheide AC, Wittmers LE Jr (1992) Selected aspects of the spatial distribution of lead in bone. Neurotoxicology 13:809–819. Aungst BJ, Fung HL (1981) Kinetic characterization of in vitro lead transport across the rat small intestine. Toxicol. Appl. Pharmacol. 61:39–47. Aungst BJ, Fung HL (1985) Kinetic characterization of in vitro lead absorption, distribution and elimination kinetics in rats. J. Toxicol. Environ. Health 16:147–159. Baghurst PA, McMichael AJ, Wigg NR, Vimpani GV, Robertson EF, Roberts RJ, Tong SO (1992) Environmental exposure to lead and children’s intelligence at the age of seven years: the Port Pirie Cohort study. N. Engl. J. Med. 327:1279–1284. Baghurst PA, McMichael AJ, Tong S, Wigg NR, Vimpani GV, Robertson EF (1995) Exposure to environmental lead and visual-motor integration at age 7 years: the Port Pirie cohort study. Epidemiology 6:104–109. Barltrop D, Khoo HE (1975) The influence of nutritional factors on lead absorption. Postgrad. Med. J. 51:795–800. Barltrop D, Meek F (1979) Effect of particle size on lead absorption from the gut. Environ. Health 34:280–285. Barry PS (1975) A comparison of concentrations of lead in human tissue. Br J. Ind. Med. 32:119–139. Barry PS (1981) Concentrations of lead in tissues of children. Br J. Ind. Med. 38:61–71. Barry PS (1985) Multicompartment kinetic models for lead. I. Bone diffusion mode for long time retention. Environ. Res. 36:441–458. Barton JC, Conrad ME, Harrison L, Nuby S (1978a) Effects of calcium on the absorption and retention of lead. J. Lab. Clin. Med. 92:536–543. Barton JC, Conrad ME, Harrison L, Nuby S (1978b) Effects of iron on the absorption and retention of lead. J. Lab. Clin. Med. 91:366–376.

REFERENCES

799

Beechx RL (1986) Lead poisoning in children. Anal. Chem. 58:274–287. Bellinger DC (1995) Interpreting the literature on lead and child development: the neglected role of the “experimental system.” Neurotoxicol. Teratol. 17:201–212. Bellinger DC (2003) Teratogen update: lead and pregnancy. Birth Defects Res. Part A 73:409–420. Bellinger DC (2008) Neurological and behavioral consequences of Childhood lead exposure. PLoS Med 5(5):e115 doi:10.1371/journal.pmed.0050115. Bellinger D, Leviton A, Needleman KL, Waternaux C, Rabinowitz M (1986) Low-level lead exposure and infant development in the first year. Neurobehav. Toxicol. Teratol. 8:151–161. Bellinger D, Leviton A, Waterman C, Needleman L, Rabinowitz M (1987) Longitudinal analysis of prenatal and postnatal lead exposure and early cognitive development. N. EngI. J. Med. 316:1037– 1043. Bellinger DC, Stiles KM, Needleman HL (1992) Low-level lead exposure, intelligence and academic achievement: a long-term follow-up study. Pediatrics 90:855–871. Bernard SR (1977) Dosimetric data and metabolic models for lead. Health Phys 32:44–46. Bert JL, van Dsen LJ, Grace JR (1989) A generalized model for the prediction of lead body burdens. Environ. Res. 48:117–127. Blake KCH, Mann M (1983) Effect of calcium and phosphorus on the gastrointestinal absorption of 203 Pb in man. Environ. Res. 30:188–194. Bogden JD, Oleske JM, Lourie DB (1997) Lead poisoning — one approach to a problem that won’t go away. Environ. Health Perspect. 105:1284–1287. Bolger JM, Yess NJ, Gunderson EL, Troxell TC, Carrington CD (1996) Identification and reduction of sources of dietary lead in the United States. Food Addit. Contamin. 13:53–60. Boline DR (1981)Some speciation and mechanistic aspects of trace metals in biological systems. In: McKinney JD, editor. Environmental Health Chemistry. Ann Arbor:Science Press. Bondy SC (1988) The neurotoxicity of organic and inorganic lead. In: Bondy SC, Prasad KR, editors. Metal Neurotoxicity. Boca Raton, FL: CRC Press. Boutron CF, Candelone J-P, Hong SM (1994) Past and recent changes in the large-scale tropospheric cycles of lead and other heavy metals as documented in Antarctic and Greenland snow and ice: a review. Geochim. Cosmochim. Acta 58:3217–3225. Bowers T, Beck B (2006) What is the meaning of non-linear dose–response relationship between blood lead concentration and IQ? Neurotoxicology 27:520–524. Brezinski DR (1976) Review of factors determining solubility and absorption of lead and other trace metals in the gastrointestinal tract. J. Coatings Technol. 48:48–57. Burns JM, Baghurst PA, Sawyer MG, McMichael AE, Tong S-L (1999) Lifetime low-level exposure to environmental lead and children’s emotional and behavioral development at ages 11–13 years. Am. J. Epidemiol. 149:740–749. Casteel SW, Cowart RP, Weiss CP, Henningsen GM, Huffman E, Brattin WJ, Guzman RE, Starost MF, Payne PT, Stockham SL, Becker SW, Drexler JW, Turk JR (1997) Bioavailability of lead to juvenile swine dosed with soil from the Smuggler Mountain NPL site of Aspen, Colorado. Fundam. Appl. Toxicol. 36:177–187. Cecil KM, Yuan W, Holland S, Wessel S, Dietrich K, Ris D, Lanphear B (2005) The influence of childhood lead exposure on language function in young adults: an fMRI study [abstract]. Presented at International Society for Magnetic Resonance Imaging: 12th Scientific Meeting and Exhibition, May, Miami, FL. pp.A1443. Centers for Disease Control (1991) Preventing Lead Poisoning in Young Children. A Statement by the Centers for Disease Control. United States Department of Health and Human Service.Atlanta: United States Public Health Service. Centers for Disease Control (1997) Blood lead levels–United States, 1991–1994. Morb. Mortal Wkly. Rep. 46:141–147.

800

LEAD AND COMPOUNDS

Chamberlain AC (1983) Effect of airborne lead on blood lead. Atmos. Environ. 17:677–692. Chamberlain AC (1985) Prediction of response of blood lead to airborne and dietary lead from volunteer experiments with lead isotopes. Proc. Royal Soc. London B224:149–182. Chisolm JJ (1980) Lead and other metals — a hypothesis of interactions. In:Singhal RL, Thomas JA, editors. Lead Toxicity.Baltimore: Urban and Schwartzenberg. Choie DD, Richter GW (1972) Lead poisoning: rapid information of intranuclear inclusion. Science 177:1194–1195. Christensen JJ, Lamb JD, Izatt SR, Starr SE, Weed GC, Astin MS, Stitt BD, Izatt RM (1978) Effect of anion type on rate of facilitated transport of cations across liquid membranes via neutral macrocyclic carriers. J. Am. Chem. Soc. 100:3219–3220. Claudio ES, Godwin HA, Magyar JS (2003) Fundamental coordination chemistry, environmental chemistry, and biochemistry of lead (II). In: Karlin KD, editor. Progress in Inorganic Chemistry, Vol. 51. Hoboken, NY: John Wiley & Sons. pp.1–144. Clayton GG, Clayton LE (1981–1982) Patty’s Industrial Hygiene and Toxicology,3rd edn, Vols 2A, 2B, 2C. New York: Wiley. Cocco P, Hua F, Boffetta P, Carta P, Flore C, Flore V, Onnis A, Picchiri GF, Cohn D (1997) Mortality of Italian lead smelter workers. Scand. J. Work Environ. Health 23:15–23. Committee on Measuring Lead in Critical Populations, National Research Council(1993) Measuring Lead Exposure in Infants, Children and Other Sensitive Populations.Washington, DC: National Academy of Sciences Press. Conrad ME, Umbriet JN, Moore EG, Rodning CR (1992) Newly identified iron-binding protein in human duodenal mucosa. Blood 79:244–247. Cory-Slechta D (1988) Mobilization of lead over the course of DMSA chelation therapy and long-term efficacy. J. Pharmacol. Exp. Ther. 246:84–91. Cory-Slechta DA (1997) Relationships between Pb-induced changes in neurotransmitter system function and behavioral toxicity. Neurotoxicology 18:673–688. Cory-Slechta DA, Weiss B, Cox C (1987) Mobilization and redistribution of lead over the course of calcium disodium ethylene diamine tetracetate chelation therapy. J. Pharmacol. Exp. Ther. 243:804–813. Davis JM, Svendzgaard DJ (1987) Lead and child development. Nature 329:297–300. DeSilva PE (1981) Determination of lead in plasma and studies on its relationship to lead in erythrocytes. Br J. Ind. Med. 38:209–217. Dietrich K, Succop PA, Bomschein RL, Krafft K, Berger O, Hammond PB, Buncher CR (1990) Lead exposure and neurobehavioral development in later infancy. Environ. Health Perspect. 89:13–19. Dietrich KN, Succop PA, Berger OG, Keith RW (1992) Lead exposure and the central auditory processing abilities and cognitive development of urban children: the Cincinnati lead study cohort at age 5 years. Neurotoxicol. Teratol. 14:51–56. Dietrich KN, Berger OG, Succop PA, Hammond PB, Bomschein RL (1993a) The developmental consequences of low to moderate prenatal and postnatal lead exposure: intellectual attainment in the Cincinnati lead study cohort following school entry. Neurotoxicol. Teratol. 15:37–44. Dietrich KN, Berger OG, Succop PA (1993b) Lead exposure and the motor developmental status of urban six-year-old children in the Cincinnati prospective study. Pediatrics 91:301–307. Dietrich KN, Ris MD, Succop PA, Berger OG, Bornschein RL (2001) Early exposure to Pb and juvenile delinquency. Neurotoxicol. Teratol. 23:511–518. Dietert RR, Piepenbrink MS (2006) Perinatal immunotoxicity: why adult exposure assessment fails to predict risk. Environ. Health Perspect. 114:477–483.

REFERENCES

801

Doss M, Becker U, Sixel E, Geisse S, Solcher H, Kufner G, Schiegel H, Stoeppler M (1982) Persistent protoporphyrinemia in hereditary porphobilinogen under low lead exposure: a new molecular basis for the pathogenesis of lead intoxication. Klin. Wochenschr. 60:599–606. Drasch G (1974) Untersuchungen zum Bleigehalt menschiecher Knochen. Arbeitmed. Sozialmed. Praventimed 14:32–34. Drasch G (1997) Are blood, urine, hair, and muscle valid biomonitors for the internal burden of men with heavy metal mercury, lead and cadmium? Trace Elem. Electrol. 14:116–123. Drasch GA, Ott J (1988) Lead in human bones: investigations on an occupationally non-exposed population in southern Bavaria (FGR). II. Children. Sci. Total Environ. 68:61–69. Drasch GA, Bohm J, Baur C (1987) Lead in human bones: investigations on an occupationally non-exposed population in southern Bavaria (FGR). 1. Adults. Sci. Total Environ. 64:303–315. Dratman JB (1978)The mechanisms of thyrozine action. In: Eckerman DA, Glowa JR, Anger WK, editors. Hormonal Protein and Peptides: Thyroid Hormones. Eckerman DA, Glowa JR, Anger WK (1999)Interindividual variability in neurotoxicity. In: Neumann DA, Kimmel CA, editors. Human Variability in Response to Chemical Exposures.Boca Raton, FL:CRC Press. pp. 59–86. Edwards M, Dudi A (2004) Role of chlorine and chlorimine in corrosion of lead-bearing plumbing materials. J. Am. Water Works Assoc. 96:69–78. Fairhall LT (1924) The solubility of various lead compounds in blood serum. J. Biol. Chem. 60:481–488. pp. 59–86. Fels LM, Wunsch M, Baranowski J, Norska–Borowka I, Price RG, Taylor SA, Patel S, DeBroe M, Elsevier MM, Lauwery R, Roels H, Bernard A, Mutti A, Gelpi E, Rosello J, Stolte H (1998) Adverse effects of chronic low level lead exposure on kidney function — a risk group study in children. Nephrol. Dial. Transplant 13:2248–2256. Flannagan PR, Hamilton DL, Haist J, Valberg LS (1979) Interrelationships between iron and lead absorption in iron-deficient mice. Gastroenterology 77:1041–1081. Florence TM, Lilley SG, Stauber JL (1988) Skin absorption of lead. Lancet 2:157–158. Franklin CA, Inskip MJ, Baccanale CL, Edwards CM, Manton WI, Edwards E, O’Flaherty EJ (1997) Use of sequentially administered stable lead isotopes to investigate changes in blood lead during pregnancy in a nonhuman primate (Macaca fascicularis). Fundam. Appl. Toxicol. 39:109–119. French P, Hunt DTE (1988) Thermodynamic calculations of dissolved trace metal speciation in river, estuary, and sea waters. Water Res. Centr. Tech. Rep. 249:38. Fullmer CS (1997) Lead–calcium interactions: involvement of 1,25-dihydroxyvitamin D. Environ. Res. 72:45–55. Fullmer CW, Edelstein S, Wasserman RH (1985) Lead-binding properties of intestinal calciumbinding proteins. J. Biol. Chem. 260:6816–6819. Gallon C, Tessier A, Gobeil C, Alfaro-De La Torre MC (2004) Modeling diagenesis of lead in sediments of a Canadian Shield lake. Geochim. Cosmochim. Acta 68:3531–3545. Garty J (2001) Biomonitoring atmospheric heavy metals with lichens: theory and application. Crit. Rev. Plant Sci. 20:309–371. Gonick HC (2002) Lead, renal disease, and hypertension. Am. J. Kidney Dis. 40:202–204. Gonick HC, Ding Y, Bondy SC, Ni Z, Vaziri ND (1997) Lead-induced hypertension: interplay of nitric oxide and reactive oxygen species. Hypertension 30:1487–1492. Goyer RA, Rhyne BC (1973) Pathological effects of lead. Int. Rev. Exp. Pathol. 12:1–77. Granado MJR, Guisasola FJA, Quirs AB (1994) Estuido de Ia plumbemia en Ia poblani infantil conferropenia. Med. Clin. (Barc.) 102:210–204. Grandjean P, Grandjean EC (1984) Biological Effects of Organolead Compounds. Boca Raton, FL: CRC Press.

802

LEAD AND COMPOUNDS

Grandjean P, Lintrup J (1978) Erythrocyte Zn-protoporphyrin as an indicator of lead exposure. Scand. J. Clin. Lab. Invest. 38:669–675. Greninger D, Kolionitsch V, Kline CK (1978) Lead Chemicals.New York: International Lead and Zinc Organization. Gross SB, Pfitzer EA, Yeager DW, Kehoe RA (1975) Lead in human tissues. Toxicol. Appl. Pharmacol. 132:628–681. Gulson BL, Law AJ, Korsch MJ, Mizon KJ (1994) Effects of plumbing systems on lead content of drinking water and contribution to lead body burden. Sci. Total Environ. 144:279–284. Gulson BL, Mahaffey KR, Mizon KJ, Korsch MJ, Cameron MA, Vimpani G (1995) Contribution of tissue lead to blood lead in adult female subjects based on stable lead isotope methods. J. Lab. Clin. Med. 125:703–712. Gulson BL, Pisamello D, McMichael AJ, Mahaffey KR, Luke C, Mizon KJ, Korsch MJ, Ashbolt R, Vimpani G, Pederson DC (1996) Stable lead isotope profiles in smelter and general urban communities: comparison of biological and environmental measures. Environ. Geochem. Health 18:147–163. Gulson BL, Mahaffey KR, Jameson CW, Vicos M, Law AJ, Mizon KJ, Smith AJM, Korsch MJ (1997) Dietary lead intakes for mother–child pairs and relevance to pharmacokinetic models. Environ. Health Perspect. 105:1334–1342. Gulson BL, Mahaffey KR, Jameson CW, Mizon KJ, Korsch MJ, Cameron MA, Eisman EA (1998a) Mobilization of lead from the skeleton during the postnatal period is even larger than during pregnancy. J. Lab. Clin. Med. 131:324–329. Gulson BL, Cameron MA, Amith AJ, Mizon KJ, Korsch MJ, Vimpani G, McMichael AJ, Pisaniello D, Jameson CW, Mahaffey KR (1998b) Blood lead–urine lead relationships in adults and children. Environ. Res. 78:152–160. Gulson BL, Jameson CW, Mahaffey KR, Mizon KJ, Patison N, Law AJ, Korsch MJ, Salter MA (1998c) Relationships of lead in breast milk to lead in blood, urine, and diet of the infant and mother. Environ. Health Perspect. 106:667–674. Hamilton DL (1978) Interrelationships of lead and iron retention in iron-deficient mice. Toxicol. Appl. Pharmacol. 46:651–661. Hammond PB, Bornschein RL, Succop P (1985) Dose–effect and dose–response relationships of blood lead to erythrocyte protoporphyrin in young children. Environ. Res. 38:187–196. Hanninen H, Aitio A, Kovala T, Luukkonen R, Matikainen E, Mannelin T, Erkkila J, Riihimaki V (1998) Occupational exposure to lead and neuropsychological dysfunction. Occup. Environ. Med. 55:202–209. Hanzlik RP (1981)Toxicity and metabolism of metal compounds: some structure–activity relationships. In: Mckinney JD,editor. Environmental Health Chemistry.Ann Arbor: Science Press. Healy MA, Harrison PG, Aslam M, Davis S, Wilson CG (1982) Lead sulphide and traditional preparations: routes for ingestion and solubility and reactions in gastric fluids. J. Clin. Hosp. Pharmacol. 7:169–173. Heard MJ, Chamberlain AC (1982) Effects of minerals and food on uptake of lead from the gastrointestinal tract in humans. Human Toxicol. 1:411–415. Henning SJ, Cooper L (1988) Intestinal accumulation of lead salts and milk lead by suckling rats (42645). Proc. Soc. Exp. Biol. Med. 187:110–116. Hernandez-Avila M, Gonzalez-Cosio T, Palazuelos E, Romieu I, Aro A, Fishbein E, Peterson KE, Hu H (1996) Dietary and environmental determinants of blood and bone lead levels in lactating postpartum women living in Mexico City. Environ. Health Perspect. 104:1076– 1082. Hill CH, Matrone G (1970) Chemical parameters in the study of in vivo and in vitro interactions and transition elements. Fed. Proc. 29:1474–1481.

REFERENCES

803

Hlavay J, Poly
ak K, Weisz M (2001) Monitoring of the natural environment by chemical speciation of elements in aerosol and sediment samples. J. Environ. Monit. 3:74–80. Hogan K, Marcus A, Smith R, White P (1998) Integrated exposure uptake biokinetic model for lead in children: empirical comparisons with epidemiologic data. Environ. Health Perspect. 106 (Suppl. 6):1557–1567. Hu H (1998) Bone lead as a new biologic marker of lead dose: recent findings and implications for public health. Environ. Health Perspect. 106 (Suppl. 4):961–967. Hu H, Rabinowitz M, Smith D (1998) Bone lead as a biological marker in epidemiologic studies of chronic toxicity: conceptual paradigms. Environ. Health Perspect. 106:1–8. Huisingh D, Huisingh J (1974) Factors influencing the toxicity of heavy metals food. Ecol. Food Nutr. 3:2263–2272. Hunt DTE, Creasey JD (1980) The calculation of equilibrium trace metal speciation and solubility on aqueous systems by a computer method with particular reference to lead. Water Res. Cent. Tech. Rep. 151:32. Hursch JB, Suomela J (1968) Absorption of 212Pb from the gastrointestinal tract of man. Acta Radiol. 7:108–120. International Agency of Research on Cancer (IARC) (1987) Overall Evaluations of Carcinogenicity: An Updating of IARC Monographs, Vols 1–42. IARC Monographs on the Evaluation of Carcinogenic Risk to Humans, Suppl. 7. Lyon: IARC. pp. 1–440. Issac RA, Gil L, Cooperman AN, Hulme K, Eddy B, Ruiz M, Jacobson K, Larson C, Pancorbo OC (1997) Corrosion in drinking water distribution systems: a major contributor of copper and lead to wastewaters and effluents. Environ. Sci. Technol. 31:3198–3203. Izatt RM, Terry RE, Haymore BL, Haven LD, Dailey NK, Avondet AG, Christensen JJ (1976) Calorimetric titration study of the interaction of several uni- and bivalent cations with 15-crown5,18-crown-6, two isomers of dicyclohexo-18-crown-6 in aqueous solutions at 25(C and u ¼ 0.1. J. Am. Chem. Soc. 98:7620–7626. Jackson PJ, Sheiham I (1986) Calculations of lead solubility in water. Water Res. Cent. Tech. Rep. 152:43. Jacobs DE (1995)Lead-based paint as a major source of childhood lead poisoning: a review of the evidence. In: Beard ME, Iska SDA,editors. Lead in Paint, Soil, and Dust: Health Risks, Exposure Studies, Control Measures, Measurement Methods, and Quality Assurance.Philadelphia: American Society for Testing and Materials. pp. 175–187. James M, Hilburn ME, Blair JA (1985) Effects of meals and meal times on uptake of lead from the gastrointestinal tract in humans. Human Toxicol. 40:401–407. Jawrerski JF, Nriagu J, Denny P, Hart BT, Laskeenn MR, Subramanan V, Wong MH (1987)Group report: lead. In: Hutchinson TC, Meemav KM,editors. Scope 31/Lead, Mercury, Cadmium and Arsenic in the Environment. New York: Wiley. Kazantzis G (1988) The use of blood in the biological monitoring of toxic metals. In: Clarkson TW, Friberg L, Nordberg GF, Sager RP,editors. Biological Monitoring of Toxic Metals.New York: Plenum Press. Kehoe RA (1961) The metabolism of lead in man in health and disease. J Royal Inst. Public Health Hyg. 24: 81–97, 129–143, and 177–203. Keller CA, Doherty RA (1980a) Bone lead mobilization in lactating mice and lead transfer to suckling offspring. Toxicol. Appl. Pharmacol. 55:220–228. Keller CA, Doherty RA (1980b) Lead and calcium distributions in blood, plasma, and milk of the lactating mouse. J. Lab. Clin. Med. 95:81–89. Kim R, Rotnitsky A, Sparrow D, Weiss ST, Wager C, Hu H (1996) A longitudinal study of low-level lead exposure and impairment of renal function. The normative aging study. J. Am. Med. Assoc. 275:1177–1181.

804

LEAD AND COMPOUNDS

Kiratli N, Ergin M (1996) Partitioning of heavy metals in surface Black Sea sediments. Appl. Geochem. 11:775–788. Kurkjian R, Dunlap C, Flegal AR (2004) Long-range downstream effects of urban runoff and acid mine drainage in the Debed River. Armenia: insights from lead isotope modeling. Appl. Geochem. 19:1567–1580. Landner L (1987)Speciation of metals in water, sediment, and soil systems. In:Bhattacharji S, Friedman GM, Neugebauer HJ, Seilaeher A,editors. Lecture Notes in Earth Sciences. New York: Springer-Verlag. pp.179–185. Lanphear BP, Matte TD, Rogers J, Clickner RP, Dietz B, Bornschein RL, Succop P, Mahaffey KR, Dixon S, Galke W, Rabinowitz M, Farfel M, Rohde C, Schwartz J, Ashley P, Jacobs DE (1998) The contribution of lead-contaminated house dust and residential soil to children’s blood lead levels. Environ. Res. 79:51–68. Lanphear BP, Dietrich K, Auinger P, Cox C (2000) Cognitive deficits associated with blood lead concentrations510 mg/dL in U.S. children and adolescents. Public Health Rep. 115:521–529. Lanphear BP, Hornung R, Ho M, Howard CR, Eberly S, Knauf K (2002) Environmental lead exposure during early childhood. J. Pediatr.140:40–47. Lanphear BP, Hornung R, Khoury J, Yolton K, Baghurst P, Bellinger DC, Canfield RL, Dietrich KN, Bornschein R, Greene T, Rothenberg SJ, Needleman HL, Schnaas L, Wasserman G, Graziano J, Roberts R (2005) Low-level environmental lead exposure and children’s intellectual function: an international pooled analysis. Environ. Health Perspect. 113:894–899. Lee RG, Becker WC, Collins DW (1989) Lead at the tap: sources and control. J. Am. Water Works Assoc. 81:52–62. Leggett RW (1993) An age-specific kinetic model of lead metabolism in humans. Environ. Health Perspect. 101:598–616. Lester BM, LaGasse LL, Seifer R (1998) Cocaine exposure and children: the meaning of subtle effects. Science 282:633–634. Lidsky TI, Schneider JS (2003) Lead neurotoxicity in children: basic mechanisms and clinical correlates. Brain 37(126):5–19. Life Science Research Office(1996) Executive summary from the third report on nutrition monitoring in the United States. J. Nutr. 126: iii–x and 1907S–1936S. Lilis R, Eisinger J, Blumberg W, Fischbein A, Selikoff IJ (1978) Hemoglobin, serum iron, and zinc protoporphyrin in lead-exposed workers. Environ. Health Perspect. 25:97–102. Lippmann M (1990) Review: 1989 Alice Hamilton Lecture. Lead and human health: background and recent findings. Environ. Res. 51:1–24. Loghman-Adham M (1997) Renal effects of environmental and occupational lead exposure. Environ. Health Perspect. 105:928–939. Lundstrom N-G, Nordberg G, Englyst V, Gehardssen L, Hagmar L, Jin T, Rylander L, Walls S (1997) Cumulative lead exposure in relation to mortality and lung cancer morbidity in a cohort of primary smelter workers. Scand. J. Work Environ. Health 23:24–30. Mahaffey KR (1977) Relation between quantities of lead ingested and health effects of lead in humans. Pediatrics. 59:448–455. Mahaffey KR (1995) Nutrition and lead: strategies for public health. Environ. Health Perspect. 103 (Suppl. 6):191–196. Mahaffey KR, Annest JL (1986) Association of erythrocyte protoporphyrin with blood lead level and iron status in the Second National Health and Nutrition Examination Survey, 1976–1980. Environ. Res. 41:327–338. Mahaffey KR, Annest JL, Roberts J, Murphy RS (1982) National estimates of blood lead levels: United States, 1976–1980. Association with selected demographic and socioeconomic factors. N. Engl. J. Med. 307:573–579.

REFERENCES

805

Mahaffey KR, Garthside PR, Glueck CL (1986) Blood lead and dietary calcium in 2926 1-through 11year-old black and white children: The second National Health and Nutrition Examination Survey, 1976–1980. Pediatrics 78:257–262. Mahaffey-Six K, Goyer RA (1970) Experimental enhancement of lead toxicity by low dietary calcium. J. Lab. Clin. Med. 76:933–942. Mahaffey-Six K, Goyer RA (1972) The influence of iron deficiency on tissue content and toxicity of ingested lead in the rat. J. Lab. Clin. Med. 79:128–136. Manton WI (1977) Sources of lead in blood: identification by stable isotopes. Arch. Environ. Health 32:149–157. Manton WI (1985) Total contribution of airborne lead to blood lead. Br J. Ind. Med. 42:169–172. Marcus AH (1985a) Multicompartment kinetic model for lead. I. Bone kinetics and variable absorption in humans without excessive lead exposures. Environ. Res. 36:441–458. Marcus AH (1985b) Multicompartment kinetic model for lead. II. Linear kinetics and variable absorption in humans without excessive lead exposure. Environ. Res. 36:459–472. Marcus AH (1985c) Multicompartment kinetic models for lead. III. Lead in blood plasma and erythrocytes. Environ. Res. 36:473–489. Markovac J, Goldstein GW (1988) Picomolar concentrations of lead stimulate brain protein kinase C. Nature 334:71–73. McIntosh MJ, Meredith PA, Moore MR, Goldberg A (1989) Action of lead on neurotransmission in rats. Xenobiotica 19:101–113. Meng X-M, Zhu DM, Ruan D-Y, Sheand J-Q, Luo L (2005) Effects of chronic lead exposure on 1H MRS of hippocampus and frontal lobes in children. Neurology 64:1644–1647. Meredith PA, Moore MR, Goldberg A (1977) The effects of calcium on lead absorption in rats. Biochem. J. 166:531–537. Minot AS (1929) The distribution of lead in the organism after absorption by the lungs and subcutaneous tissue. J. Ind. Hyg. 6:137–148. Mohnen VA (1988) The challenge of acid rain. Sci. Am. 259:30–38. Morrow PE, Beiter H, Amato F, Gibb FR (1980) Pulmonary retention of lead: an experimental study in man. Environ. Res. 21:373–384. Muntner P, He J, Vupputuri S, Coresh J, Batuman V (2003) Blood lead and chronic kidney disease in the general United States population: results from NHANES III. Kidney Int. 63:1044–1050. Mushak P (1989) Biological monitoring of lead exposure in children: overview of selected biokinetic and toxicological issues. In:Smith MA, Grant LD, Sors AI,editors.Lead Exposure and Child Development: An International Assessment. Workshop Organized by the Commission of the European Communities and the U.S. Environmental Protection Agency, September 1986, Edinburgh, UK. Mushak P (1991) Gastro-intestinal absorption of lead in children and adults: overview of biological and biophysico-chemical aspects. Chem. Spec. Bioavail. 3 (3/4):87–104. Mushak P (1993) New directions in the toxicokinetics of human lead exposure. Neurotoxicology 14:29–42. Mushak P (1998) Uses and limits of empirical data in measuring and modeling human lead exposure. Environ. Health Perspect. Suppl. 106 (6):1467–1484. Mykkanen H, Hannu M, Wasserman RH (1982) Effect of vitamin D on the intestinal absorption of lead-203 and calcium-47 in chicks. J. Nutr. 112:520–527. National Institute for Occupational Safety and Health(1998) Adult blood lead epidemiology and surveillance—United States, third quarter, 1997. Morb. Mortal Wkly. Rep. 47:77–80. Nawrot TS, Thijs L, Den Hond, EM; Roels HA, Staessen JA (2002) An epidemiological re-appraisal of the association between blood pressure and blood lead: a meta-analysis. J. Hum. Hypertens. 16:123–131.

806

LEAD AND COMPOUNDS

Needleman HL, Gunnoe C, Leviton A, Reed R, Peresie H, Maher C, Barrett P (1979) Deficits in psychological and classroom performance of children with elevated dentine lead levels. N. Engl. J Med. 300:689–695. Needleman HL, Riess JA, Tobin MJ, Biesecker GE, Greenhouse JB (1996) Bone lead levels and delinquent behavior. J. Am. Med. Assoc. 275:363–369. Needleman HL, McFarland C, Ness RB, Fienberg SE, Tobin MJ (2002) Bone lead levels in adjudicated delinquents. A case control study. Neurotoxicol. Teratol. 24:711–717. Nordberg GF, Mahaffey KR, Fowler BA (1991) Introduction and summary. International workshop on lead in bone: implications for dosimetry and toxicology. Environ. Health Perspect. 91:3–7. Nriagu JO (1998) Tales told in lead. Science 281:1622–1623. Nriagu JO, Pacyna JM (1988) Quantitative assessment of worldwide contamination of air, water and soils by trace metals. Nature 333:134–139. O’Flaherty EJ (1993) Physiologically based models for bone-seeking elements. IV. Kinetics of lead disposition in humans. Toxicol. Appl. Pharmacol. 118:16–29. O’Flaherty EJ (1995) Physiologically based models for bone-seeking elements. V. Lead absorption and deposition in childhood. Toxicol. Appl. Pharmacol. 131:297–308. O’Flaherty KE (1998) Physiologically based models of metal kinetics. Crit. Rev. Toxicol. 28:271–317. Oldendorf WH, Brown WJ (1975) Greater numbers of capillary endothelial cell mitochondria in brain than in muscle. Proc. Soc. Exp. Biol. Med. 149:736–738. Osman K, Pawlas K, Schutz A, Gazdzik M, Sokal JA, Vahter M (1999) Lead exposure and hearing effects in children in Katowice, Poland. Environ. Res. 80:1–8. Pappas J, Ahlquist J, Allen E, Banner W (1995) Oral dimercaptosuccinic acid and ongoing exposure to lead: effects on heme synthesis and lead distribution in a rat model. Toxicol. Appl. Pharmacol. 133:121–129. Peltier EF, Webb SM, Gaillard JF (2003) Zinc and lead sequestration in an impacted wetland system. Adv. Environ. Res. 8:103–112. Perkins SM, Filippelli GM, Souch CJ (2000) Airborne trace metal contamination of wetland sediments at Indiana Dunes National Lakeshore. Water Air Soil Pollut. 122:231–260. Piomelli S (1973) A micro method for free erythrocyte porphyrins: the FEP test. J. Lab. Clin. Med. 81:932–940. Piomelli S, Davidow B, Guinee V, Young P, Gay T (1973) The FEP [free erythrocyte porphyrin] test: a screening micro-method for lead poisoning. Pediatrics 51:254–259. Piomelli S, Seaman C, Zullow D, Curran A, Davidson B , Threshold for lead damage of heme synthesis in urban children. (1982)Proc. Natl. Acad. Sci. USA 79:3335–3339. Pirkle JL, Brody JD, Gunter EW, Kramer RQ, Paschal DC, Flegal KK, Matte TD (1994) The decline in blood lead levels in the United States: the National Health and Nutrition Examination Surveys (NHANES). J. Am. Med. Assoc. 272:284–291. Rabinowitz MB, Wetherill G, Kopple J (1975) Absorption, storage and excretion of lead by normal humans. In:Proceedings of 9th Annual Conference on Trace Substances in Environmental Health. Columbia:University of Missouri. pp.361–368. Rabinowitz MB, Wetherill G, Kopple JD (1976) Kinetic analysis of lead metabolism in healthy humans. J. Lab. Invest. 58:260–270. Ragan HA (1983) The bioavailability of iron, lead, and cadmium via gastrointestinal absorption: a review. Sci. Total Environ. 28:317–326. Rhoads BL, Cahill RA (1999) Geomorphological assessment of sediment contamination in an urban stream system. Appl. Geochem. 14:459–483.

REFERENCES

807

Roels HA, Lauwerys RR, Buchet JP, Vrelust MT (1976) Response of free erythrocyte porphyrin and urinary delta-aminolevulinic acid in men and women moderately exposed to lead. Int. Arch. Arbeitsmed. 34:97–108. Rosen JF, Pounds JG (1998) Severe chronic lead insult that maintains body burdens of lead related to those in the skeleton: observations by Dr. Clair Patterson conclusively demonstrated. Environ. Res. 78:140–151. Rosen JF, Markowitz ME, Bijur PE, Jenks ST, Wielopolski L, Kalef-Ezra JA, Slatkin DN (1989) L-line X-ray fluorescence of cortical bone lead compared with the CaNa2EDTA test in lead-toxic children. Public health implications. Proc. Natl. Acad. Sci. USA 86:685–689. Correction 86:7595. Sartorelli E, Loi F, Gori R (1985) Lead silicate toxicity: a comparison among different compounds. Environ. Res. 36:420–425. Sassa S, Granick JL, Granick S, Kappas A, Levere RD (1973) Studies in lead poisoning. I. Microanalysis of erythrocyte protoporphyrin levels by spectrophotometry in the detection of chronic lead intoxication in the subclinical range. Biochem. Med. 8:135–148. Schintu M, Kudo A, Sarritzu G, Contu A (1991) Heavy metal distribution and mobilization in sediments from a drinking water reservoir near a mining area. Water Air Soil Pollution 57:329– 338. Schlag RD (1987) Lead. In:Fishbein L, Furst A, Mehlman MA,editors. Genotoxic and Carcinogenic Metals: Environmental and Occupational Occurrence and Exposure. Princeton: Princeton Scientific Publishing. Schutz A, Skerfving S, Ranstam J, Christoffersson JO (1987a) Kinetics of lead in blood after the end of occupational exposure. J. Work Environ. Health 13:221–231. Schutz A, Skerfving S, Christoffersson J-O (1987b) Chelatable level versus lead in human trabecular and compact bone. Sci. Total Environ. 68:45–59. Schwartz J, Otto D (1987) Blood lead, hearing threshold, and neurobehavioral development in children and youth. Arch. Environ. Health 42:153–160. Schwartz J, Otto D (1991) Lead and minor hearing impairment. Arch. Environ. Health 46:300–305. Scott KM, Hwang KM, Jurikowitz M, Brierley GP (1971) Ion transfer by heart mitochondria XXIII. The effects of lead on mitochondrial reactions. Arch. Biochem. Biophys. 147:557–567. Seppalainen AM, Hernberg S, Koch R (1979) Relationship between blood lead levels and nerve conduction velocities. Neurotoxicology 1:313–332. Skerfving S (1998)Biological monitoring of exposure to inorganic lead. In:Clarkson TW, Friberg L, Nordberg GF, Sager PR,editors. Biological Monitoring of Toxic Metals.New York: Plenum Press. pp.169–197. Shotyk W, Weiss D, Appleby PG, Cheburkin AK, Frei R, Gloor M, Kramers JD, Reese S, Van Der Knapp WO (1998) History of atmospheric lead deposition since 12,370 14C yr BP from a peat bog, Jura Mountains, Switzerland. Science 281:1635–1640. Silbergeld EK (2003) Facilitative mechanisms of lead as a carcinogen. Mutation Res. 533:121–133. Simons TJB (1986) Passive transport and binding of lead by human red blood cells. J. Physiol. 378:267–286. Simons TJB (1989) Lead contamination. Nature 337:514. Singley JE (1994) Electrochemical nature of lead contamination. J. Am. Water Works Assoc. 86:91–96. Smith CM, DeLuca HF, Tanaka Y, Mahaffey KR (1980) Stimulation of lead absorption by vitamin D administration. J. Nutr. 108:843–847. Smith GL, Rees SB, Williams DR (1985) Speciation of strontium on human and bovine milk. Polyhedron 4:713–716.

808

LEAD AND COMPOUNDS

Smith DR, Osterloh JD, Flegal AR (1996) Use of endogenous, stable lead isotopes to determine release of lead from the skeleton. Environ. Health Perspect. 104:60–66. Smith D, Bayer L, Strupp BL (1998) Efficacy of succimer in chelation for reducing brain Pb levels in a rodent model. Environ. Res. 78:168–176. Spear TM, Svee W, Vincent JH, Stanisich N (1998) Chemical speciation of lead dust associated with primary lead smelting. Environ. Health Perspect. 106:565–571. Steenhout A (1982) Kinetics of lead storage in teeth and bones: an epidemiologist’s approach. Arch. Environ. Health. 37:224–231. Steding DJ, Dunlap CE, Flegal AR (2000) New isotopic evidence for chronic lead contamination in the San Francisco Bay estuary system: implications for the persistence of past industrial lead emissions in the biosphere. Proc. Natl. Acad. Sci. USA 97:11181–11186. T
ellez-Rojo MM, Bellinger DC, Arroyo-Quiroz C, Lamadrid-Figueroa H, Mercado-Garc
ıa A, Schnaas-Arrieta L, Wright RO, Hern
andez-Avila M, Hu H (2006) Longitudinal associations between blood lead concentrations 510 mg/dL and neurobehavioral development in environmentally exposed children in Mexico City. Pediatrics 118:323–330. Tsaih SW, Korrick S, Schwartz J, Amarasiriwardena C, Aro A, Sparrow D, Hu H (2004) Lead diabetes, hypertension, and renal function: the Normative Aging Study. Environ. Health Perspect. 112:1178–1182. Tong S, Baghurst P, McMichael A, Sawyer M, Mudge J (1996) Lifetime exposure to environmental lead and children’s intelligence at 11–13 years: the Port Pirie cohort study. Br. Med. J. 312:1569– 1575. Trope I, Lopez-Villegas D, Cecil KM, Lenkinski RE (2001) Exposure to lead appears to selectively alter metabolism of cortical gray matter. Pediatrics. 107:1437–1442. U.S. Environmental Protection Agency (1986a) Air Quality Criteria for Lead. EPA Report No. EPA/ 600/8-83/028aF-dF. 4v. Available from NTIS, Springfield, VA; PB87-142378.Research Triangle Park, NC:Office of Health and Environmental Assessment, Environmental Criteria and Assessment Office. U.S. Environmental Protection Agency (1986b) Lead Effects on Cardiovascular Function, Early Development, and Stature: An Addendum to U.S. EPA Air Quality Criteria for Lead. In: Air Quality Criteria for Lead, Vol. 2. EPA Report No. EPA/600/8-83/028aF. Available from NTIS, Springfield, VA; PB87-142378.Research Triangle Park, NC:Office of Health and Environmental Assessment, Environmental Criteria and Assessment Office. pp.A1–A67. U.S. Environmental Protection Agency (1990a) Summary of Selected New Information on Effects of Lead on Health and Supplement to 1986 Air Quality Criteria for Lead. EPA Report No. EPA/600/8-89. Available from NTIS, Springfield, VA; PB92-235670. Research Triangle Park, NC:Office of Health and Environmental Assessment, Environmental Criteria and Assessment Office. U.S. Environmental Protection Agency (1977) Air Quality Criteria for Lead. EPA Report No. EPA/ 600/8-77-017. Available from NTIS, Springfield, VA; PB-280411. Research Triangle Park, NC: Health Effects Research Laboratory, Criteria and Special Studies Office. U.S. Environmental Protection Agency. (1994a) Guidance Manual for the Integrated Exposure Uptake Biokinetic Model for Lead in Children. Report No. EPA/540/R-93/081. Available from NTIS, Springfield, VA; PB93-963510.Washington, DC: Office of Emergency and Remedial Response. U.S. Environmental Protection Agency (1994b) Technical Support Document: Parameters and Equations Used in Integrated Exposure Uptake Biokinetic Model for Lead in Children (v 0.99d). Report No. EPA/540/R-94/040. Available from NTIS, Springfield, VA; PB94-963505.Washington, DC: Office of Solid Waste and Emergency Response. U.S. EPA (2006) Air Quality Criteria for Lead (Final). EPA/600/R-05/144aF-bF. Washington, DC: U. S. Environmental Protection Agency.

REFERENCES

809

Vainio H (1997) Lead and cancer — association or causation? Scand. J. Work Environ. Health 23:1–3. Vaziri ND (2002) Pathogenesis of lead-induced hypertension: role of oxidative stress. J. Hypertens 20 (Suppl.):S15–S20. Vaziri ND, Ding Y, Ni Z, Gonick HC (1997) Altered nitric oxide metabolism and increased oxygen free radical activity in lead-induced hypertension: effect of lazaroid therapy. Kidney Int. 52:1042–1046. Walkowiak J, Altmann L, Kramer U, Sveinsson K, Turfeld M, Weishoff-Houben M, Winneke G (1998) Cognitive and sensorimotor functions in 6-year-old children in relation to lead and mercury levels: adjustment for intelligence and contrast sensitivity in computerized testing. Neurotoxicol. Teratol. 20:511–521. Wasserman GA, Liu X, Lolocono NJ, Factor-Litvak P, Kline JK, Popovac D, Morina N, Musabergovic A, Vrenezi N, Capum-Paracka S, Lekic V, Preteni-Redjepi E, Hadzialjevic S, Salkovich V, Graziano JH (1997) Lead exposure and intelligence in 7-year-old children: the Yugoslavian prospective study. Environ. Health Perspect. 105:956–962. Wasserman GA, Liu X, Popovac D, Factor–Litvak P, Kline J, Waternaux C, LoIacono N, Graziano JH (2000b) The Yugoslavia prospective lead industry study: contribution of prenatal and postnatal lead exposure to early intelligence. Neurotoxicol. Teratol. 22:811–818. Watson WS, Hume R, Moore MR (1980) Oral absorption of lead and iron. Lancet 2:236–237. Watt GCM, Britton A, Gilmore WH, Moore MR, Murray GD, Robertson SJ, Womersley J (1996) Is lead in tap water still a public health problem? An observational study in Glascow. Br. Med. J. 31:979–981. Weiss D, Shotyk W, Kempf O (1999) Archives of atmospheric lead pollution. Naturwissenschaften 86:262–275. White PD, Van Leeuwan P, Davis BD, Maddaloni M, Hogan KA, Marcus AH, Elias RW (1998) The conceptual structure of the integrated exposure uptake biokinetic model for lead in children. Environ. Health Perspect. 106 (Suppl. 6):1513–1530. White LD, Cory-Slechta DA, Glibert ME, Tiffiny-Castigliani E, Zawia NH, Virgolini M, RossiGeorge A, Lashley SM, Qian YC, Riyaz Basha MD (2007) New and evolving concepts in the neurotoxicity of lead. Toxicol. Appl. Pharmacol. 225:1–27. Willes RF, Lok E, Truelove JF, Sandarum A (1977) Retention and tissue distribution of 210Pb(NO3)2 administered orally to infant and adult monkeys. J. Toxicol. Environ. Health 3:395–406. Wu MT, Kelsey K, Schwartz J, Sparrow D, Weiss S, Hu H. (2003) A “delta”-aminolevulinic acid dehydratase (ALAD) polymorphism may modify the relationship of low-level lead exposure to uricemia and renal function: the Normative Aging Study. Environ. Health Perspect. 111:335–341. Yuan W, Holland SK, Cecil KM, Dietrich KN, Wessel SD, Altaye M, Hornung RW, Ris MD, Egelhoff JC, Lanphear BP (2006) The impact of early childhood lead exposure on brain organization: a functional magnetic resonance imaging study of language function. Pediatrics 118:971–977. Ziegler EE, Edwards BB, Jensen RL, Mahaffey KR, Fomon SJ (1978) Absorption and retention of lead by infants. Pediatr. Res. 12:29–34.

21 MERCURY Philippe Grandjean and Jesper B. Nielsen

21.1 INTRODUCTION Although the toxicity of mercury has been known since ancient times, its therapeutic effects were also utilized in a variety of drugs. In particular, mercurous chloride (calomel) was an important drug for syphilis treatment, although some patients inevitably became mercury poisoned. The occupational health risks were described by Bernardino Ramazzini 300 years ago. Risks due to environmental contamination came to the forefront in around 1960 when Minamata disease in Japan was found to be caused by mercury pollution from a local factory. Recent risk assessments include the Agency for Toxic Substances and Disease Registry (ATSDR, 1994) U.N. Environment Programme (UNEP, 2002) and the U.S. Environmental Protection Agency (2001); these sources may be consulted for further information and additional references.

21.2 CHEMISTRY Mercury exists in three oxidation states: Hg0 (metallic), Hgþ (mercurous), and Hg2þ (mercuric) mercury. In organometallic derivatives, mercuric mercury is covalently bound to one or two carbon atoms, and the organic part of the molecule is often an alkyl group or an alkoxialkyl group. The former compounds are more toxic because they are more easily absorbed and more slowly metabolized. In its elemental form, mercury is a dense, silvery-white, shiny metal, which is liquid at room temperature and boils at 357 C. At 20 C, the vapor pressure of the metal is 0.17 Pa (0.0013 mm Hg), and a saturated atmosphere at this temperature contains 14 mg Hg/m3, which is more than 100 times the occupational exposure limit. Mercury compounds differ greatly in their solubility. Thus, at 25 C, the solubility of metallic mercury, mercurous chloride, and mercuric chloride in water are 60 mg/L, 2 mg/L, and 69 g/L, respectively (IARC, 1994).

Environmental Toxicants, Third Edition Edited by Morton Lippmann Copyright
2009 John Wiley & Sons, Inc.

811

812

MERCURY

Exact data for the solubility of methyl mercuric chloride in water are lacking, but are known to be slightly higher than for mercurous chloride. Certain species of mercury are soluble in nonpolar solvents. These include elemental mercury and the halide compounds of alkylmercurials.

21.3 SOURCES Mercury is emitted to the atmosphere by “degassing” of the earth’s surface and by resuspension of mercury-containing particles previously deposited. Emissions from volcanoes and other natural sources are estimated to constitute about 1000 tons per year (UNEP, 2002). An additional annual emission of about 2500 tons comes from anthropogenic sources, the major source being energy production from fossil fuels, especially coals with high mercury contents (UNEP, 2002). Mercury is produced by the mining and smelting of cinnabar ore. It has been extensively used in chloralkali plants (producing chlorine and sodium hydroxide), but modern plant designs have now made large stores of mercury unnecessary. A myriad of mercurycontaining products have been in use, including compounds in paints as preservatives or pigments, in electrical switching equipment and batteries, in measuring and control equipment (thermometers and other medical equipment), in mercury vacuum instruments, as a catalyst in chemical processes, in mercury quartz and luminescent lamps, in the production and use of high explosives using mercury fulminate, in copper/silver amalgams in dental restoration materials, and as fungicides in agriculture (especially as seed dressings). Many of these uses are now being banned or phased out. According to data from the U.S. EPA, these efforts have been beneficial, as overall mercury emissions from industrial use in the United States have dropped 45% since 1990, and are still decreasing. One of the uses of liquid metallic mercury that has escalated during the last few decades is artisanal gold mining. Alluvial deposits of fine gold particles are often extracted using mercury. The gold particles are dissolved in the mercury as an amalgam, and the mercury is subsequently removed by heating with a gas torch. This use therefore exposes the gold miner to a substantial amount of mercury vapor, and also leads to extensive release of mercury into confined and sometimes ecologically sensitive areas. The annual consumption of mercury in such mining operations is about 650 tons, mainly in Asia, Central Africa, and Latin America. Another consequence of this practice is the contamination of soil, which can remain polluted for many decades. Some previous gold mining sites in the United States (e.g., Carson River, Nevada) are now recognized as being heavily contaminated with mercury, with estimated amounts of mercury residues exceeding 6000 tons. Deposition of sewage sludge and contamination from other industrial activities often involve mercuric salts with low solubility (i.e., sulfides). Ecological and human health implications inorganic mercury in the environment depend on the entent of mercury methylation. Recent studies using stable mercury isotopes (Harris et al., 2007) have documented that sedimentation of airborne mercury onto freshwater ecosystems within several months gets accumulated in fish in the form of methylmercury. Organomercury compounds now only find limited use as fungicides, but methylmercury was extensively used for this purpose in the past, until environmental effects were discovered. The less toxic methoxymethylmercury is still sometimes used for wood treatment or in the paper and pulp industry as an antislime agent. Thimerosal (ethylmercury salicylate) has been widely used as a preservative in the pharmaceutical industry, for example, in vaccines, but is now being phased out.

ENVIRONMENTAL EXPOSURES

813

The various uses of mercury and mercury compounds result in occupational exposures in a range of occupations, in most cases only involving mercury vapor and inorganic compounds. The industrial use of mercury may also lead to releases to the environment due to evaporation of releases in sewage water (IPCS, 1991). Localized problems relating to contamination of river systems and bays have been caused by contamination from chloralkali plants, paper and pulp industries, and pesticide factories. In Japan, the Minamata Bay became severely contaminated from a factory that used methylmercury as a catalyst in the production of acetaldehyde. In addition, airborne emissions from coalfired power plants and incinerators can cause contamination of lakes and rivers. Coal burning contributes 40% of the mercury emissions in the United States, and is associated with increased mercury deposition in Eastern Canada and New England (Rice and Hammitt, 2005). Elemental mercury may be oxidized to Hg2þ, which can then become methylated into methylmercury compounds by chemical or microbiological reactions in the aquatic environment. The intestinal bacterial flora of various animal species, including fish, is also, though to a much lower degree, able to convert ionic mercury into methyl mercuric compounds (Nielsen, 1992). Methylmercury is accumulated by fish and marine mammals and it attains its highest concentrations in large predatory species at the top of the aquatic food chain. By this means, it enters the human diet. Certain microorganisms can demethylate methylmercury, for example, in the gut, while others can reduce Hg2þ to Hg0. Thus, microorganisms are believed to play an important role in the fate of mercury in the environment and in affecting human exposure.

21.4 ENVIRONMENTAL EXPOSURES 21.4.1

Air

In the areas of Europe remote from industrial activity, mean concentrations of total mercury in the atmosphere are reported to be in the range of 2–3 ng/m3 in summer and 3–4 ng/m3 in winter (UNEP, 2002). Mean mercury concentrations in urban air are usually three–fourfold higher. “Hot spots” of mercury concentration exceeding 10,000 ng/m3 have been reported close to industrial emissions or above areas where mercury fungicides have been used extensively. The chemistry of atmospheric mercury is complex and has attracted much research. Except at or near pollution sources, airborne mercury is of limited relevance in regard to human respiratory exposures. The chemical reactions and the partitioning of mercury in gas and aqueous phases, however, are crucial in regard to mercury residence times in the atmosphere and deposition at various latitudes. This is exemplified during the Arctic spring, when the ultraviolet light catalyzes reactions that lead to short-lasting, yet excessive, depositions of mercury (Lindberg et al., 2002). Few data are available on average indoor air pollution due to mercury vapor. Fatalities and severe poisonings have resulted from heating metallic mercury and mercury-containing objects at home. Elemental mercury is sometimes used for certain cultural and religious practices that may involve sprinkling mercury inside, burning it in a candle, or mixing it with perfume; such practices can create exposures that may greatly exceed currently permitted occupational exposure level (Riley et al., 2001). Release of mercury from dental amalgam fillings is otherwise the predominant source of human exposure to inorganic mercury (IPCS, 1991). Energy-saving light bulbs contain mercury, and their increased popularity will no

814

MERCURY

doubt make broken bulbs an important indoor and environmental source of mercury vapor (Appell, 2007). The daily amount of mercury absorbed from the atmosphere into the bloodstream as a result of respiratory exposure in adults is about 32 ng in rural areas and about 160 ng in urban areas. These calculations are based on rural concentrations of 2 ng/m3 and urban concentrations of 10 ng/m3 (absorption rate 80%). Depending upon the number of amalgam fillings, mercury concentrations in inhaled air have ranged up to several thousand ng/m3, and the estimated average daily absorption is thought to vary between 3000 and 17,000 ng (IPCS, 1991). 21.4.2

Diet (Drinking Water and Food)

Mercury in drinking water is usually in the range of 5–100 ng/L, the average value being about 25 ng/L. Although the forms of mercury in drinking water are poorly known, Hg2þ present as complexes and chelates with various ligands is likely the predominant species. The bioaccessibility (i.e., the extent to which a certain mercury complex is available for absorption at the gastrointestinal mucosal surface) may increase or decrease depending on the ligand and the binding strength between the metal and the ligand. Concentrations of mercury in most foodstuffs (EFSA, 2004) are often below the detection limit, and likely to be inconsequential. Freshwater fish, seafood, in general, and marine mammals constitute the dominant sources, mainly in the form of methylmercury compounds (70–90% of the total). The amount of mercury in fish depends on factors such as pH and redox potential of the water, species, age, and size of the fish. The normal concentrations in edible tissues of various species of fish cover a wide range, mostly between 50 and 1400 ng mercury per gram of fresh weight (IPCS, 1990; EFSA, 2004). Large predatory fish, such as pike, swordfish, and tuna, as well as shark, seals, and toothed whales contain the highest average concentrations. Furthermore, exposure might occur from the use of pharmaceuticals, in particular thimerosal, widely applied as a preservative of vaccines and immunogloblins. With up to 100 mg mercury per injection, this preservative caused substantial bolus doses of mercury, especially on a body weight basis, in connection with childhood immunizations. Skin-lightening lotions and soaps used, in Arabian and African countries often contain mercury concentrations of about 1000 mg/kg. Some of these products may even reach concentrations in the percent range. Although mercury may be absorbed through the skin, consumers are usually not warned about the toxic contents. 21.4.3

Relative Significance of Different Routes of Environmental Exposure

Human exposure to the three major forms of mercury present in the environment is summarized in Table 21.1 (based on IPCS, 1991). Although the choice of values given is associated with some uncertainty, the numbers provide a perspective on the relative magnitude of the contributions from various media. Humans may be exposed to additional quantities of mercury occupationally, from living in heavily polluted areas or through the use of skin-lightening creams. The intake from drinking water is about 50 ng mercury per day, mainly as Hg2þ; of which only a small fraction is absorbed. The main pathway of exposure is through the intake of fish and seafood products, mainly in the form of methylmercury. Very high exposures occur in arctic populations, whose diets include marine mammals. Increased levels also occur in Japanese and Mediterranean populations, who frequently eat fish high in the food chain. Exposures are lower in countries, such as the United States, where NHANES III data suggest that 85% of

OCCUPATIONAL EXPOSURES

TABLE 21.1

815

Estimated Average Daily Intake (Retention) of Mercury Compounds Estimated Average Daily Intake (Retention)a in ng of Mercury Per Day Mercury Vapor

Media

b

Air Food Marine Nonmarine Drinking water Dental amalgam Total

Inorganic Mercury Compounds c

Methylmercury

40–200 (30–160)

0

0c

0 0 0

600d (60) 3600 (360) 50 (5)

2400d (2300) — 0

3800–21,000 (3000–17,000) 3900–21,000 (3100–17,000)

0

0

4200 (420)

2400 (2300)

a

Figures in parentheses are the amounts retained that were estimated from the pharmacokinetic parameters (i.e., 80% of inhaled vapor, 95% of ingested methylmercury, and 10% of inorganic mercury are retained). b Assumes an air concentration of 2–10 ng/m3 and a daily respiratory volume of 20 m3. c For the purposes of comparison, it is assumed that in the atmospheric concentrations of species of mercury other than mercury vapor are negligible. d It is assumed that 80% of the total mercury in edible fish tissues is methylmercury and 20% in the form of inorganic mercury compounds. Marine food intake may vary considerably between individuals and across populations.

Americans consume fish at least once a month, 40% once a week, while only 1–2% consume fish or shellfish almost daily. Because fish and seafood is recommended as an essential part of a varied diet, advisories need to identify the types of nutrient-rich fish that are low in mercury to ensure that the benefits exceed the risks. Total dietary mercury intake has usually been measured as part of market basket surveys or as part of specific monitoring. Probabilistic analyses based on dietary questionnaire data and fish analyses suggest that small children, on a body weight basis, may receive a higher exposure than adults (EFSA, 2004). Incomplete information is available on the distribution of high-end intakes from seafood diets, especially among vulnerable population groups, such as pregnant women and children.

21.5 OCCUPATIONAL EXPOSURES Occupational exposure is almost exclusively to inorganic mercury and occurs at chloralkali plants, mercury mines, thermometer factories, fluorescent light tube production plants, refineries, and in dental clinics. High mercury concentrations have been described for all these situations, with considerable variations depending on the working conditions. Some 70,000 workers in the United States were considered exposed to mercury, primarily elemental mercury, but the number is decreasing. Serious mercury exposures may occur in connection with gold mining, especially when gold amalgam is heated. In developing countries, this process is often carried out under field conditions or in small gold vending shops without or with insufficient ventilation. An estimated 10 million workers in Africa, Latin America, and Asia are exposed to high concentrations of elemental mercury through such activities. The impacts on health from these exposures are generally not monitored.

816

MERCURY

21.6 KINETICS AND METABOLISM The bioavailability, kinetics, and biotransformation of mercury depend upon its chemical and physical form. 21.6.1

Absorption

21.6.1.1 Elemental Mercury (Hg0) Approximately 80% of inhaled mercury vapor is absorbed via the lungs and retained in the body. Elemental mercury is poorly absorbed in the gastrointestinal tract (less than 0.01% in rats). Increased blood mercury concentrations have been measured in humans, however, after accidental ingestion of several grams of metallic mercury. 21.6.1.2 Inorganic Mercurous (Hgþ) and Mercuric (Hg2þ) Mercury The absorption of inhaled aerosols of inorganic mercury depends on particle size, solubility, and so on (IPCS, 1991). No data have been reported for humans. In dogs, 45% of deposited mercuric oxide aerosols were cleared in less than 24 h, and the remainder cleared with a half-time of 33 days. Ten–fifteen percent of an oral, nontoxic dose of mercuric mercury is absorbed from the gastrointestinal tract in adults and retained in body tissues, but considerable individual variations may exist. In children, the gastrointestinal absorption is likely to be greater. 21.6.1.3 Organic Mercury Human poisoning cases caused by inhalation indicate that a large fraction of these lipophilic compounds are absorbed into the blood. Alkylmercury compounds are absorbed almost completely in the gastrointestinal tract. Certain methylmercury compounds are probably absorbed through the skin. Parenteral uptake of ethylmercury occurs in connection with vaccines preserved with thimerosal. 21.6.2

Distribution

21.6.2.1 Elemental Mercury (Hg0) After exposure to mercury vapor, the element is found in blood as physically dissolved elemental mercury. Within a few minutes, the enzyme catalase oxidizes mercury into mercuric mercury in the erythrocytes. Through this mechanism, the maximum Hg concentration in the erythrocytes can be observed within an hour after a brief exposure to mercury vapor. In contrast, it takes about 10 h for plasma concentrations to peak. Before oxidation, Hg0 readily crosses cell membranes, including the blood/brain barrier and the placental barrier. After oxidation, the Hg2þ ions (or complexes) are distributed in the body via the blood. The kidneys and the brain are the main retention sites for Hg after exposure to mercury vapor, whereas absorbed inorganic mercury salts are mainly deposited in the kidneys. The uptake and/or elimination of mercury after exposure to mercury vapor can be altered by a moderate intake of alcohol, possibly due to inhibition of catalase. In humans, ingestion of alcohol prior to mercury vapor exposure has resulted in significantly reduced blood mercury levels (Hursh et al., 1980). 21.6.2.2 Inorganic Mercurous (Hgþ) and Mercuric (Hg2þ) Mercury The kidneys are the predominant site of inorganic mercury accumulation. After oral exposure, accumulation also occurs in the cells of the mucous membranes of the gastrointestinal tract. A significant

KINETICS AND METABOLISM

817

part of this accumulation, however, never reaches the general circulation as it is eliminated through cell shedding. Mercuric mercury in blood is divided between erythrocytes and plasma in about equal amounts. In erythrocytes, mercury is probably to a large extent bound to sulfhydryl groups of the hemoglobin molecule and possibly also to glutathione. The distribution between different plasma–protein fractions varies with dose and time after exposure. To a limited extent, mercuric mercury crosses the blood–brain and placental barriers. However, mercuric mercury does accumulate in the placenta, fetal membranes, and amniotic fluid. The rate of uptake from blood and different organs varies widely, as does the rate of elimination from different organs. Thus, the distribution of mercury within the body and within organs varies widely with dose and time lapse after absorption. Yet, under all conditions, the dominating mercury pool in the body after exposure to mercuric mercury is the kidney. Inorganic divalent mercury can induce metallothionein, and a large proportion of the mercury in the kidneys is soluble and bound to this molecule. 21.6.2.3 Organic Mercury Methylmercury is distributed via the bloodstream to all tissues in the body. The pattern of tissue distribution is much more uniform than after inorganic mercury exposure. Red cells are an exception; there the concentration is 10–20 times greater than the plasma concentration. Methylmercury readily crosses the blood–brain and placental barriers. In the fetus, methylmercury is accumulated and concentrated, especially in the brain. As with other forms of mercury, the kidneys retain the highest tissue concentration, but the brain still contains about fivefold higher concentrations than blood. Methylmercury accumulates in hair in the process of formation of hair strands, with average concentrations being about 250-fold higher than in the blood. Methylmercury undergoes biotransformation to inorganic mercury by demethylation, particularly in the gut. Ethylmercury is less stable than methylmercury. 21.6.3

Elimination

21.6.3.1 Elemental Mercury (Hg0) After short-term exposure to mercury vapor, about one third of the absorbed mercury will be eliminated in unchanged form through exhalation. The remaining mercury will be eliminated as mercuric mercury mainly through feces. Assuming first-order kinetics for the clearance of urinary mercury after exposure to mercury vapor, the median half-time was found to be 41 days. Blood concentrations can serve as indicators of recent mercury vapor exposure. Speciation, in this case, should be carried out to eliminate possible influence of dietary intake of mercury from contaminated marine food. 21.6.3.2 Inorganic Mercurous (Hgþ) and Mercuric (Hg2þ) Mercury Excretion of absorbed inorganic mercury is mainly via urine and feces, the rates by each pathway being roughly equal. The whole-body half-time in adults is also about 40 days. The elimination of inorganic mercury follows a complicated pattern, with biological half-times differing according to the tissue and the time after exposure. Thus, mercury concentrations in critical organs may remain high even after having dwindled in urine and blood. Hence, at present, there are no general and suitable indicator media that will reflect concentrations of inorganic mercury in the critical organs, the brain or kidney, under different exposure conditions (IPCS, 1991).

818

MERCURY

21.6.3.3 Organic Mercury Mercury excretion after methylmercury exposure is predominantly via the feces. Methylmercury is slowly demethylated in the gut, and the enterohepatic recirculation of methylmercury explains that most, if not all, of the mercury excreted is in the demethylated inorganic form. Some elimination also occurs via urine. The whole-body half-time of methylmercury is generally about 45 days, although higher estimates have also been published. Laboratory animal studies have shown that following acute dosage with methylmercury, blood mercury concentrations will initially reflect organ concentrations reasonably well. Henceforth, an increasing fraction of the body burden will be in the brain, muscles, and kidney. The blood concentration might be a useful indicator of the body burden of mercury while the erythrocyte mercury concentration is more specific for methylmercury exposure. Accordingly, if exposure to mercury vapor or other inorganic mercury compounds is suspected, mercury should be speciated or a serum sample analyzed. Mercury in hair, when measured along the length of a hair strand, has also been used as an indicator of past blood levels. When using hair as an indicator, it is important to note the history and condition of the hair. Permanent waving may leach mercury from the hair while exogenous mercury can increase its concentration. Cord blood (or cord tissue) measurement is the best method to establish prenatal exposure levels.

21.7 HEALTH EFFECTS 21.7.1

Acute and Local Effects

Acute poisoning with mercury vapor may cause a severe airway irritation, chemical pneumonitis, and, in severe cases, pulmonary edema. Ingestion of inorganic compounds may cause gastrointestinal corrosion and irritation, such as vomiting, bloody diarrhea, and stomach pains. Subsequently, shock and acute kidney dysfunction with uremia may ensue. Local irritation may develop following cutaneous exposure to mercury compounds, which are among the most common allergens in patients with contact dermatitis. 21.7.2

Chronic and Systemic Effects

Chronic intoxication may develop as early as a few weeks after the onset of a mercury exposure. More commonly, however, the exposure has lasted for several months or years, yet early diagnosis is thwarted by the lack of recognition of subtle effects. The symptoms depend on the degree of exposure and the kind of mercury in question. They mainly involve the oral cavity, the peripheral and central nervous system, and the kidneys. As the elemental mercury present in vapor is oxidized to mercuric mercury in the blood, the non-neurotoxic effects of absorbed mercury vapor and other inorganic mercury compounds will be similar. 21.7.2.1 Elemental Mercury (Hg0) Severe exposure to inorganic mercury causes an inflammation of gingiva and oral mucosa, which become tender and bleed easily. Salivation is increased, most obviously so in subacute cases. Often the patient complains of a metallic taste in the mouth. Especially when oral hygiene is bad, a gray border is formed on the gingival edges. In exposures to mercury vapor, the central nervous system is the critical organ, and the classic triad of symptoms includes erethism, intention tremor, and the gingivitis described

HEALTH EFFECTS

819

above. The fine intention tremor of fingers, eyelids, lips, and tongue may progress to spasms of arms and legs. A jerky micrographia is typical as well. The changes in the central nervous system result in psychological effects known as erethism: restlessness, irritability, insomnia, concentration difficulties, decreased memory and depression, sometimes in combination with shyness, unusual psychological vulnerability, and anxiety. “Micromercurialism” is a term used to denote an early stage of erethism in which decreased memory, dizziness, and irritability are most prominent. While recent data do not support an association between dental fillings and deleterious effects some patients believe that their symptoms are linked to amalgam restorations (Bellinger et al., 2006). Induction of minimal tremor by mercury vapor has been reported at urinary excretion levels of 50 mg/L (0.25 mmol/L) and above. Data concerning the effects of mercury vapor on early stages of the human life cycle are limited. While some information is available with regard to effects on pregnancy and birth in women occupationally exposed to mercury vapor, a dose–response relationship has not been established. In children, “pink disease” may occur, as described below. 21.7.2.2 Inorganic Mercurous (Hgþ) and Mercuric (Hg2þ) Mercury The target organ following long-term exposure causing no acute toxicity are the kidneys. In general, the early renal effects of mercury appear to be reversible after cessation of exposure. Nephrotoxic effects include proximal tubular damage, as indicated by an increased excretion of small proteins in the urine (e.g., beta2-micro globulin). Experimental studies suggest that glomerular damage is caused by an autoimmune reaction to mercury complexes in the basal membrane. This mechanism of action, however, has not been confirmed for humans. In children, a different syndrome is seen, the so-called pink disease or acrodynia, diagnosed most frequently in children treated with teething powders, which contained calomel. It has also been occasionally seen in children who had inhaled mercury vapor (e.g., from broken thermometers) (Agocs et al., 1990). A generalized eruption develops and the hands and feet show a characteristic, scaly, reddish appearance. In addition, the children are irritable, sleep badly, fail to thrive, sweat profusely, and have photophobia. This condition was extremely common until 30 years ago, when the etiology was finally found and teething powders were phased out. 21.7.2.3 Organic Mercury Intoxications with alkoxialkyl or aryl compounds are similar to intoxications with inorganic mercury compounds due to their relatively unstable state. Alkyl mercury compounds, such as methylmercury, result in a different syndrome. The earliest symptoms in adults are paresthesias in the fingers, the tongue, and the face, particularly around the mouth. Later on, disturbances occur in the motor functions, resulting in ataxia and dysphasia. The visual field is decreased and, in severe cases, the result may be total blindness. Similarly, impaired hearing may progress to complete deafness. This syndrome appeared as Minamata disease in Japan as a result of methylmercury contamination from a local factory. Epidemics also occurred when methylmercurytreated seed grain was used for baking or animal feed in Iraq and elsewhere. Children and the fetus are more susceptible to the toxic effects of methylmercury than are adults, and congenital methylmercury poisoning may result in a cerebral palsy syndrome, even though the mother remains healthy or suffers only minor symptoms due to the exposure. In populations with a high consumption of fish or marine mammals, methylmercury intakes may approach the levels that resulted in such serious disease in Japan and Iraq. Recent evidence from long-term follow-up of a Faroese birth cohort has shown that prenatal

820

MERCURY

exposure to methylmercury may result in neuropsychological and neurophysiological deficits that are detectable through adolescence (Grandjean et al., 1997; Murata et al., 2004; Debes et al., 2006; Budtz-Jørgensen et al., 2007). In adults, the earliest effects, such as paresthesias, appear to occur when blood concentrations are above 200 mg/L (1 mmol/L). Recent epidemiological studies suggest that adverse cardiovascular effects may occur at much lower exposures than are prevalent among people regularly eating seafood (Virtanen et al., 2005). Although the implications of these findings are not yet clear, they may suggest that methylmercury can be a toxic risk to the population at large, and that benefits of eating seafood must take into account that mercury exposures need to be minimized. Developmental delays appear to be related to maternal hair mercury concentrations of 1–3 mg/g (Grandjean et al., 1997) (i.e., cord blood concentrations of 4–12 mg/L). Despite several years of research, the evidence on the possible adverse health effects of thimerosal in vaccines remains unclear (Institute of Medicine, 2004; Geier and Geier, 2006). Sufficient evidence supports that methylmercury chloride is carcinogenic to experimental animals. In the absence of comprehensive epidemiological data, methylmercury is considered a possible human carcinogen (class 2B) (IARC, 1994). The U.S. EPA has classified both inorganic mercury compounds and methylmercury as possible human carcinogens.

21.8 PREVENTION Prevention should start at the source. For example, the European Union has recently enacted a ban on mercury exports and a variety of mercury uses are being phased out, such as mercury in household thermometers. Of important nonindustrial sources, batteries and fluorescent light bulbs are recycled in many countries. Mercury exposures from dental amalgam fillings should be minimized, and suitable alternative restorative materials are now available for most purposes. Pollution abatement should also focus on point sources, such as coal-fired power plants and incinerators. When mercury emissions from such sources were controlled in Florida and Massachusetts, methylmercury contamination of local freshwater fish significantly decreased after a few years. However, on a global scale, much still remains to be done. This is particularly the case with regard to pollution from burning of mercurycontaining coals in East Asia. In regard to occupational exposures, WHO has recommended that long-term mercury vapor exposures should be limited to a time-weighted average (TWA) limit of 25 mg/m3, a value that has also been adopted by the ACGIH as a threshold limit value. The corresponding TWA for inorganic mercury is 50 mg/m3. Biological monitoring is crucial in the diagnosis of mercury exposure and in the control of occupational exposure levels. Although blood concentrations are highly useful, they do not reflect mercury retained in the brain, where mercury from vapor inhalation has a half-life of several years. Urine levels are usually preferred as an indicator of occupational exposures to inorganic mercury species, and a limit of 50 mg Hg/g creatinine (28 mmol/mol creatinine) has been recommended (IPCS, 1991). For consumers, recent exposure can be ascertained by analysis of hair or blood. In the United States, increased mercury concentrations in freshwater fish have prompted fish advisories in almost every single state. However, current advisories may not be sufficient to help consumers obtaining the optimal nutritional benefits from fish and seafood, while minimizing methylmercury exposure. Current limits of 0.5 mg/g for fish in general and

REFERENCES

821

1.0 mg/g for large fish, such as swordfish and tuna, are not based on a detailed risk assessment, but reflect methylmercury concentrations prevalent in seafood. A healthy diet that includes two fish dinners per week must be based on seafood that averages a mercury concentration of no more than 0.1 mg/g to avoid exceeding the reference dose (RfD) of 0.1 mg/kg body weight per day defined by the U.S. Environmental Protection Agency (2001). Popular and nutritious fish with low mercury concentrations include salmon, sardine, and flounder. Exposure limits for methylmercury have been revised downward to protect sensitive life stages (National Research Council, 2002; JECFA, 2003). The major limits are the reference dose (RfD) of 0.1 mg/kg body weight per day (National Research Council, 2002) and the provisional tolerable weekly intake (PTWI) of 1.6 mg/kg body weight per week (JECFA, 2003) correspond to hair mercury concentrations of approximately 1–2 mg/g. Data on blood mercury concentrations in the U.S. general population suggest that between 5 and 10% of different population sections exceed the RfD.

REFERENCES Agocs MM, Etzel RA, Parrish RG, Paschal DC, Campagna PR, Cohen DS, Kilbourne EM, Hesse JL (1990) Mercury exposure from interior paint. N. Engl. J. Med. 323:1096–1101. Appell D (2007) Toxic bulbs. Sci. Am. 297 (4):30–31. ATSDR (Agency for Toxic Substances and Disease Registry)(1994) Toxicological Profile for Mercury (update), TP-93/10, Atlanta, GA. Available at http://www.atsdr.cdc.gov/toxprofiles/tp46.html (accessed March 12, 2006). Bellinger DC, Trachtenberg F, Barregard L, Tavares M, Cernichiari E, Daniel D, McKinlay S (2006) Neuropsychological and renal effects of dental amalgam in children: a randomized clinical trial. JAMA 295:1775–1783. Budtz-Jørgensen E, Grandjean P, Weihe P (2007) Separation of risks and benefits of seafood intake. Environ. Health Perspect. 115:323–327. Debes F, Budtz-Jørgensen E, Weihe P, White RF, Grandjean P (2006) Impact of prenatal methylmercury toxicity on neurobehavioral function at age 14 years. Neurotoxicol. Teratol. 28:363–375. EFSA (European Food Safety Authority) (2004) Opinion of the Scientific Panel on Contaminants in the Food Chain on a Request from the Commission Related to Mercury and Methylmercury in Food (EFSA-Q-2003-030), Brussels. Available at http://www.efsa.eu.int/science/contam/contam_opinions/259_en.html (accessed March 12, 2006). Geier DA, Geier MR (2006) An evaluation of the effects of thimerosal on neurodevelopmental disorders reported following DTP and Hib vaccines in comparison to DTPH vaccine in the United States. J. Toxicol. Environ. Health A 69:1481–1495. Grandjean P, Weihe P, White RF, Debes F, Araki S, Yokoyama K, Murata K, Sørensen N, Dahl R, Jørgensen PJ (1997) Cognitive deficit in 7-year-old children with prenatal exposure to methylmercury. Neurotox. Teratol. 19:417–428. Harris RC, Rudd JW, Amyot M, Babiarz CL, Beaty KG, Blanchfield PJ, Bodaly RA, Branfireun BA, Gilmour CC, Graydon JA, Heyes A, Hintelmann H, Hurley JP, Kelly CA, Krabbenhoft DP, Lindberg SE, Mason RP, Paterson MJ, Podemski CL, Robinson A, Sandilands KA, Southworth GR, St Louis VL, Tate MG (2007) Whole-ecosystem study shows rapid fish-mercury response to changes in mercury deposition. Proc Natl Acad Sci USA 104:16586–16591. Hursh JB, Greenwood MR, Clarkson TW, Allen J, Demuth S (1980) The effect of ethanol on the fate of mercury vapor inhaled by man. J. Pharmacol. Exper. Therap. 214:520–527.

822

MERCURY

IARC (1994) Monographs on the Evaluation of Carcinogenic Risk to Humans, Vol. 58. Mercury and Mercury Compounds, Lyon. Institute of Medicine (2004) Immunization Safety Review: Vaccines and Autism. Washington: National Academy Press. IPCS (International Programme on Chemical Safety) (1990) WHO Task Group on Environmental Health Criteria for Methylmercury (EHC 101), Geneva: WHO. IPCS (International Programme on Chemical Safety) (1991) WHO Task Group on Environmental Health Criteria for Inorganic Mercury (EHC 118), Geneva: WHO. JECFA (Joint FAO/WHO Expert Committee on Food Additives) (2003) Sixty-first meeting, Rome, 10–19 June 2003. Summary and Conclusions. Available at ftp://ftp.fao.org/es/esn/jecfa/jecfa61sc. pdf (accessed March 19, 2006). Lindberg SE, Brooks S, Lin CJ, Scott KJ, Landis MS, Stevens RK, Goodsite M, Richter A (2002) Dynamic oxidation of gaseous mercury in the Arctic troposphere at polar sunrise. Environ. Sci. Technol. 36:1245–1256. Murata K, Weihe P, Budtz-Jørgensen E, Jørgensen PJ, Grandjean P (2004) Delayed brainstem auditory evoked potential latencies in 14-year-old children exposed to methylmercury. J. Pediatr. 144:177–183. National Research Council (2002) Toxicological Effects of Methylmercury. Washington: National Academy Press. Nielsen JB (1992) Toxicokinetics of mercuric chloride and methylmercuric chloride in mice. J. Toxicol. Environ. Health 37:85–122. Rice G, Hammitt JK (2005) Economic valuation of human health benefits of controlling mercury emissions from U.S. coal-fired power plants, NESCAUM. Available at http://www.nescaum.org/ topics/mercury-control-technology (accessed March 12, 2006). Riley DM, Newby CA, Leal-Almeraz TO, Thomas VM (2001) Assessing elemental mercury vapor exposure from cultural and religious practices. Environ. Health Perspect. 109:779–784. UNEP (United Nations Environment Programme) (2002) Global Mercury Assessment Report, Geneva. Available at http://www.chem.unep.ch/mercury/Report/Final%20Assessment%20report.htm (accessed March 12, 2006). U.S. EPA (Environmental Protection Agency) (2001) Water Quality Criterion for the Protection of Human Health: Methylmercury. Publication EPA-823-R-01-001, Washington, D.C. Available at http://www.epa.gov/waterscience/criteria/methylmercury/document.html (accessed March 12, 2006). Virtanen JK, Voutilainen S, Rissanen TH, Mursu J, Tuomainen TP, Korhonen MJ, Valkonen VP, Seppanen K, Laukkanen JA, Salonen JT (2005) Mercury, fish oils, and risk of acute coronary events and cardiovascular disease, coronary heart disease, and all-cause mortality in men in eastern Finland. Arterioscler. Thromb. Vasc. Biol. 25:228–233.

22 NITROGEN OXIDES Richard B. Schlesinger

22.1 INTRODUCTION Oxides of nitrogen (NOx), so called because they consist of various chemical species, many of which are interconvertible, can exist in the atmosphere as either gases/vapors or particles. The former includes nitric oxide (NO), nitrogen dioxide (NO2), nitrous oxide (N2O), and occasionally, nitrogen trioxide (NO3), dinitrogen trioxide (N2O3), dinitrogen tetroxide (N2O4), and dinitrogen pentoxide (N2O5), while the latter includes nitrate (NO3 ) salts. Species that may exist in either a particulate or gaseous state are nitric acid (HNO3) and nitrous acid (HONO). Nitric oxide and nitrogen dioxide are the most important of the NOx in terms of public health concern since they are often present in the atmosphere in significant concentrations and are quite chemically reactive. Although N2O is also ubiquitous, being released due to natural biological processes in soil, it is not involved in chemical reactions in polluted air. Most other NOx, if found at all, are present at very low concentrations. Two possible exceptions from a health standpoint are HNO3 and inorganic nitrates.

22.2 SOURCES Ambient atmospheric NOx derive from both natural sources, such as forest fires, organic decay, and lightning, and from anthropogenic activities that involve high temperature combustion processes in both mobile and stationary sources. The major mobile source is motor vehicles, while the major stationary source is electric power generation using fossil fuels, with industrial combustion processes being a close second. From a global perspective, however, the total mass of emissions released from natural sources is much greater than that from human activities. NOx, primarily NO2, can also be an important indoor pollutant. The

Environmental Toxicants, Third Edition Edited by Morton Lippmann Copyright
2009 John Wiley & Sons, Inc.

823

824

NITROGEN OXIDES

main indoor source is the use of unvented or improperly vented natural gas or other fossil fuel-fired appliances, such as stoves and heaters. During combustion processes, nitrogen derived from the combustion air and/or the fuel being consumed reacts with atmospheric oxygen. Although most of the resulting NOx produced is initially in the form of NO, this is generally rapidly oxidized to NO2, with the conversion rate depending upon a number of factors, including the concentration of NO, temperature of the combustion process, and distance from the emission zone. Several reaction pathways are possible. While simple oxidation involving molecular oxygen (O2) is the primary one for NO2 production in combustion gas effluents, it does not play a major role in the ambient atmosphere since transformations via other pathways occur at faster rates. Thus, in air containing other reactive chemical species, for example, ozone, irradiation by sunlight can catalyze photochemical reactions leading to the very rapid formation of NO2. HNO3 is also a product of the photooxidation cycle of polluted air but, along with HONO, can additionally derive from primary emissions released by mobile sources. The major production pathway involves reaction between the hydroxyl radical (OH), formed within the smog cycle, with NO2. Other routes, which are potentially important at night, involve reactions between N2O5 with water or nitrate radicals with volatile organics, or production in droplets containing both hydrogen ion (Hþ) and nitrate (NO3 ). Because of its high saturation vapor pressure, HNO3 generally exists as a vapor under most ambient conditions, for example, within photochemical smog, where levels generally peak during daytime hours (Ellestad and Knapp, 1988). Within acidic fogs, however, HNO3 may be found in the particulate state (Jacob et al., 1985). Similarly, HONO can be found in ambient air both as a primary product from combustion sources and as a secondary product of photochemical smog reactions. HNO3 may also be produced indoors via reaction of ozone with NO2, water vapor, and volatile organics (Weschler et al., 1992). Water on indoor surfaces can react with NO2 to form HONO, which can then be released into indoor air as gas phase acid (Dubowski et al., 2004). Nitrate salts may be formed in the atmosphere via various pathways, many of which involve gaseous HNO3. For example, ammonium nitrate results from the homogeneous reaction between nitric acid and atmospheric ammonia. Nitrates may also be formed by heterogeneous reactions involving NO2 or NO and water droplets, or HNO3 vapor and dust or sea salt particles.

22.3 NITROGEN DIOXIDE 22.3.1

Exposure

22.3.1.1 Atmospheric Concentration Outdoor levels of NO2 are often directly related to motor vehicle emissions and traffic density around busy roadways and, along with particulate matter (PM) and various organics, NO2 is considered to be a good indicator of the complex particulate–gas mixture that derives from vehicular traffic (Gauderman et al., 2000, 2002; McConnell et al., 2003; Seaton and Dennekamp, 2003). Outdoor concentrations in urban areas are generally characterized by two daily peaks related to traffic patterns in the morning and afternoon. In areas having significant stationary sources, the pattern is characterized by a baseline NO2 level superimposed upon which are higher spikes occurring on an irregular basis. In those areas not impacted by significant local sources,

NITROGEN DIOXIDE

825

NO2 has little variation on an hourly basis throughout the day, unless there is transport of NO2 into the region. The average daily l-h maximum outdoor concentration of NO2 in urban regions across the United States was 0.03 ppm; levels in nonmetropolitan or rural areas would tend to be lower. The 24-h average is about half this value (U.S. EPA, 2008). Data from various rural, suburban, and urban sites in Southern California indicated 8-year (1994–2001) mean concentrations ranging from 0.003 to 0.038 ppm, with annual means at the most rural sites ranging from 0.003 to 0.02 ppm. While, as noted, natural emissions far outweigh those from anthropogenic sources on a total mass basis, the former are distributed over a wider area; this results in very low background levels due to natural sources. Nitrogen dioxide is also an indoor air pollutant, deriving from combustion sources such as gas-fired ranges, kerosene heaters, and improperly or unvented gas space heaters. Nitrogen oxides are also major components of smoke derived from the burning of tobacco products. Cigarette smoke contains high levels of NO, which is oxidized to NO2 as the smoke ages. The indoor/outdoor concentration ratio for NO2 in the absence of significant indoor sources is 0.5–0.6, but is often >1 when such sources are present (Berglund, 1993). Indoor levels vary widely depending upon the strength of the specific sources and the degree of ventilation. Furthermore, because combustion from indoor sources tends to be episodic, fairly high short-term peaks are possible. Daily (24-h average) levels of NO2 in homes using gas-fired ranges or heaters can range between 0.05 and 0.5 ppm, but short-term peaks can exceed 1 ppm (U.S. EPA, 1993; Spengler and Cohen, 1985; Goldstein et al., 1988). 22.3.1.2 Exposure Assessment From a health standpoint, the only relevant route of exposure to NOx is via inhalation and, from the above discussion, it is evident that such exposures can occur in numerous settings, which include residential areas, transportation vehicles, as well as the outdoor atmosphere. The integrated exposure is, therefore, the sum of the individual exposures over all possible time intervals and for all of these different environments. Such exposure can be assessed either by direct methods, which include biomarkers and personal monitoring, and indirect methods, which involve measurement of pollutant levels at monitoring sites and the use of mathematical models for the estimation of actual individual or population exposures. There is currently no accepted biomarker for exposure to NO2. Some suggested ones have included urinary hydroxyproline excretion (Adgate et al., 1992), the NO-heme protein complex in bronchial lavage (Maples et al., 1991) and 3-nitrotyrosine in urine (Oshima et al., 1990). However, because of their lack of sensitivity and/or specificity, these have not been shown to be practical for assessing environmental NO2 exposures. Outdoor measures of NO2 levels, while related to and contributing to total exposure, are poor predictors of total personal exposures for most people. Because indoor concentrations are often greater than those outdoors, indoor exposure is commonly the main contributor to total exposure, and actual personal exposures to NO2 may differ from what would be assumed based upon ambient outdoor air measures (Linaker et al., 1996). Thus, indoor residential levels are generally a much better predictor of personal exposure, explaining over 50% of the variation in such exposure. However, it should be borne in mind that this is a generalization, and there are likely to be selected groups of people for which indoor levels are not a good predictor of total exposure due to greater percentages of time spent in other significant NOx-containing environments, especially those occupationally exposed or those living near heavily traveled roadways.

826

NITROGEN OXIDES

TABLE 22.1

Exposure Limits for Nitrogen Oxides

Nitric oxide TLVa PELb RELc IDLHd

25 ppm 25 ppm 25 ppm 100 ppm

Nitrogen dioxide NAAQSe PELf RELg EEGLh TLVa STELi IDLHd

0.053 ppm 5 ppm 1 ppm 1 ppm 3 ppm 5 ppm 20 ppm

a

Threshold limit value (ACGIH; time weighted average for an 8-h work day and a 40-h work week). Permissible exposure limit (OSHA; time weighted average for an 8-h work day). c Recommended exposure limit (NIOSH; time weighted average for an 8-h work day). d Immediately dangerous to life and health (NIOSH; 30-min average). e National ambient air quality standard (USEPA; annual average). f Permissible exposure limit for general industry (OSHA; ceiling for 15 min). g Recommended exposure limit (NIOSH; ceiling for 15-min exposure). h Emergency exposure guidance level (NAS; 1-h exposure). i Short-term exposure limit (ACGIH; ceiling for 15-min exposure). b

22.3.1.3 Exposure Limits There are various ambient and occupational exposure limits for NOx. Some of the major ones are listed in Table 22.1. 22.3.2

Dosimetry

Up to 90% of the NO2 inspired during normal respiration can be removed within the human respiratory tract (Wagner, 1970). Estimates of regional uptake for the upper respiratory tract (i.e., airways proximal to the trachea) based upon laboratory animal studies range from 28 to 90% of the amount inhaled (Cavanagh and Morris, 1987; Dalhamn and Sjoholm, 1963; Yokoyama, 1968; Vaughan et al., 1969; Kleinman and Mautz, 1989), while that for the lungs range from 36 to 90% of the amount entering the trachea (Postlethwait and Mustafa, 1981; Kleinman and Mautz, 1989). Specific ventilatory factors influence the extent of uptake. Thus, more NO2 will be absorbed in the upper respiratory tract during nasal breathing than during oral breathing (Kleinman and Mautz, 1989), implying that the latter would allow a greater percentage of inhaled NO2 to reach the lungs. Increased ventilation, such as due to exercise, may alter regional gas distribution from that occurring at rest by reducing NO2 uptake in the upper respiratory tract and tracheobronchial tree and, thus, increasing the amount of NO2 delivered to and absorbed in the respiratory (alveolar) region of the lungs (Miller et al., 1982; Overton, 1984; Kleinman and Mautz, 1989; Wagner, 1970; Mohsenin, 1994). Within the lungs, inhaled NO2 can be absorbed throughout the entire tracheobronchial tree and respiratory region, although the major dose to tissue is delivered at the junction between the conducting and respiratory airways (Miller et al., 1982; Overton, 1984). Regardless of the site of initial contact with airway surfaces, a primary determinant of

NITROGEN DIOXIDE

827

NO2 uptake is surface reactivity, that is, direct interaction with airway lining fluid and/or cellular components (Postlethwait and Bidani, 1990). Potential substrates include oxidizable chemical species, such as amino acids, proteins, and unsaturated fatty acids (Hood et al., 1993), resulting in the production of nitrite ion or various radicals (Postlethwait and Mustafa, 1981; Saul and Archer, 1983; Postlethwait and Bidani, 1989, 1990), which can then interact with the epithelium or rapidly pass into the bloodstream and undergo other chemical reactions in extrapulmonary sites, for example, oxidation to nitrate by interaction with hemoglobin in red blood cells (Parks et al., 1981; Oda et al., 1981; Kosaka et al., 1979; Case et al., 1979). Nitric and nitrous acids, as well as their nitrate salts, have been detected in blood and urine following exposure to NO2 (U.S. EPA, 1993; Garn et al., 2003). Antioxidants present within airway lining fluid can react with deposited NO2, potentially modulating its toxicological impact (Kelly et al., 1996). Any NO2 that dissolves in airway fluids could result in the production of nitric and nitrous acids, with any subsequent toxicity likely due to the hydrogen or nitrite ions (Goldstein et al., 1977, 1980). It is, however, likely that both oxidative and nonoxidative mechanisms are involved in toxicity from inhaled NO2. 22.3.3

Health Effects––Epidemiology

Epidemiological studies have attempted to assess the potential role of exposure to NO2 in producing adverse human health effects. Many studies relate health end points to outdoor concentrations, but the current trend is to provide better measures of actual personal exposures, which, as noted, can be reflections of strong indoor sources. Some studies have used NO2 as the only pollutant, while others have used NO2 as a general marker for pollution derived from motor vehicles. A major problem, however, is, as noted above, a close association between NO2 and other pollutants, especially PM, derived from the same combustion sources, making it often difficult to determine any independent effects due solely to NO2. While the robustness of some epidemiological studies are affected by a lack of reliable estimates of actual NOx exposure conditions, inadequate sample size, inadequate compensation for the effects of covariates, and/or misclassification of health end points, they do provide a linkage between controlled exposure (toxicology) studies and “real-world” exposure of humans. These studies have examined the relationship between acute exposure and effects, as well as responses to long-term exposure. Ambient NO2 has been related to increased mortality in some evaluations (e.g., Wietlisbach et al., 1996; Sunyer et al., 1996; Anderson et al., 1996). A meta-analysis examining daily mortality that incorporated studies published between 1982 and 2000 and that used data from 1958 to 1999 (Stieb et al., 2002, 2003) indicated an overall effect estimate for overall mortality from NO2 alone to be 2.8% (with a 95% confidence interval (CI) of 2.1– 3.5%) per 0.024 ppm NO2 (24-h mean); this value was reduced to 0.9% (CI ¼ 0.1–2.0) in a multipollutant model that included PM as well, highlighting the difficulty in evaluating effects due to NO2 alone. A multicity study found a 1.3% increase in daily mortality (95% CI ¼ 0.9–1.8) per 50 mg/m3 (0.028 ppm) 1-h maximum concentration increase in NO2 (Touloumi et al., 1997). A study examining European cities (Katsouyanni et al., 2001) noted a higher daily mortality in those having higher NO2. Analysis of daily air pollution and mortality in 90 cities in the United States (Dominici et al., 2003) showed a significant increase in daily mortality ranging from 0.3 to 0.4% per 0.010 ppm (using a 1-day lag between concentration and response), but this disappeared when adjusted for PM and ozone.

828

NITROGEN OXIDES

Most studies of acute responses to NO2 used indices of respiratory illness and/or changes in pulmonary mechanical function to assess the health consequences of exposure. Some examples of representative results obtained in such studies are provided. In a classic series of surveys conducted in a number of cities in the United States selected to represent a range of outdoor air quality (Harvard Six-Cities Air Pollution Health Study), grade school age children within each community were followed for several years by reporting on questionnaires and by annual measurements of pulmonary function. Outdoor NO2 levels were measured at various sites within each community and indoor levels were also measured in selected households. While results of this study from 1974 to 1977 on over 8000 children aged 6–10 years indicated a significant increase in the rate of respiratory illness before age 2 in homes with gas-fired stoves compared to those with electric stoves (Speizer et al., 1980), a later examination of the same communities over a longer time period did not show any NO2related increase in respiratory illness (Ware et al., 1984). A further analysis of over 5000 children aged 7–11 years during the period 1983–1986 noted marginal significance for physician-diagnosed respiratory illness prior to age 2 in homes using gas-fired stoves compared to those using electric stoves (Dockery et al., 1989). When pulmonary mechanical indices were evaluated in the above-mentioned children (Ware et al., 1984), gas stove use was associated with significant reductions in parameters of expiratory flow (FEV1, FVC) in a first examination, but no such relationship was found in a subsequent evaluation. Certain subpopulations, based upon age or preexisting disease state, or both, may be more susceptible to effects of NO2 than are others. A borderline significant effect was noted between peak flow reduction in healthy children residing in homes having gas stoves, while a much stronger association was noted in asthmatics (Lebowitz et al., 1985). Children with asthmatic symptoms appeared to be more susceptible to reduced lung function when outdoor average NO2 concentrations exceeded a certain level (0.02 ppm), but no such effect was found with children having no asthmatic symptoms (Moseler et al., 1994). A relationship between outdoor levels of NO2 and common respiratory symptoms (e.g., cough, sore throat, etc.) in children up to 5 years of age was noted in one study (Gnehm et al., 1988), while another found an association between NO2 exposure and wheeze in females, but not in males, aged 4 months to 4 years (Pershagen et al., 1995). Braun-Fahrlander et al. (1992) examined symptoms in children in relation to outdoor and indoor levels of NO2. While the incidence of symptoms was not associated with either indoor or outdoor levels, the duration of increased symptoms was associated with outdoor NO2 concentration. In studies of the effect of NO2 on respiratory health in 6–9-year-old children, personal exposures to NO2 were measured, as were indoor levels in the home (Houthuijs et al., 1987; Brunekreef et al., 1987). The prevalence of lung disease was found to be associated with the presence of unvented gas water heaters, with weekly average exposures estimated at 0.021 ppm. On the contrary, Dijkstra et al. (1990) found no association between respiratory symptoms with indoor NO2 measurements in homes. Koo et al. (1990) used personal samplers to monitor NO2 exposure in children aged 7–13 years in Hong Kong. No association was noted between exposure levels (means ranged from 0.013 to 0.023 ppm for a 1-week period) and respiratory symptoms, such as wheeze, running nose, or cough. The effects of both indoor and outdoor air pollution on respiratory illness in a cohort of primary school children indicated a gradient of increased respiratory symptoms with increasing indoor levels of NO2 in homes with gas stoves (Melia et al., 1977). A later assessment also indicated some increase in relative risk in homes with gas stoves, but this was not a consistent finding (Melia et al., 1979). In this case, levels of NO2 measured in bedrooms of homes having gas stoves ranged from 0.003 to 0.017 ppm. In another study of children

NITROGEN DIOXIDE

829

aged 5–6 years, no significant relationship was noted between levels of NO2 and the prevalence of respiratory illness (Melia et al., 1982); levels of NO2 in the bedrooms of homes with gas stoves were 0.005–0.029 ppm. Other attempts to relate gas stove use in homes to acute respiratory illness, respiratory symptoms or indices of reduced lung function for various age populations have had mixed results, with some studies reporting no association and others reporting some relationship (Keller et al., 1979; Comstock et al., 1981; Schenker et al., 1983; Dodge, 1982; Ekwo et al., 1983). A strong association was found between NO2 and asthma admissions to hospital for children under 14 years of age in three European cities (Sunyer et al., 1997), while significant effects of NO2 on doctors visits or hospital admissions for asthma were found in children in various areas of the world (Medina et al., 1997; Anderson et al., 1998; Morgan et al., 1998; Lee et al., 2002; Lin et al., 2002). A stronger association was found to exist between NO2 and asthma symptoms in children in London than for adults in the same area (Hajat et al., 1999). In a study examining respiratory symptoms in adult women and children aged 13 years and younger (Berwick et al., 1984), indoor NO2 levels were measured in homes. Children under the age of 7 years exposed to 0.016 ppm were found to be at an increased risk of upper and lower respiratory tract symptoms compared to those who were not so exposed. No increased risk was found in older children or adults. In a later study (Samet et al., 1993), no association was found between indoor levels of NO2 and either the incidence or duration of respiratory illness in infants examined during their first 18 months of life. A meta-analysis (Hasselblad et al., 1992) of studies using indoor NO2 levels suggested a relationship between incidence of lower respiratory tract symptoms and chronic exposure in children less than 12 years of age, while no effect on lower respiratory tract illness during the first year of life was seen in relation to indoor NO2 in another study (Sunyer et al., 2004). A meta-analysis using data from Australia and New Zealand (Barnett et al., 2005) noted an association between all respiratory-related hospital admissions, as well as admissions specifically for asthma, pneumonia, and acute bronchitis in children grouped into various age ranges. While both PM and NO2 were noted to be associated with total respiratory admissions in children in the 1–4-year age group, the largest effect was noted for NO2. In older children, aged 5–14 years, an association was also found with PM and NO2, but with the latter showing a larger effect. However, the greatest association was noted for asthma admissions increase, related to an increase of about 5 ppm in the 24-h mean concentration of NO2. However, when PM was controlled for, the effect of NO2 in the younger children was attenuated, but that in the older age group was not. Finally, a number of studies have noted associations between NO2 and various symptoms, such as cough, wheeze, and shortness of breath, in children with asthma (e.g., Just et al., 2002; Segala et al., 1998; Quackenboss et al., 1991; Mortimer et al., 2002). In an examination of the relationship between air pollutants and emergency room visits in Atlanta, GA (Peel et al., 2005), the effect estimate for respiratory admissions due to a 1-h exposure to 0.020 ppm was found to be 1.6% (95% CI ¼ 0.6–2.7) for all respiratory admissions, and 1.9% (95%CI ¼ 0.6–3.1) for upper respiratory infections only. While the above-mentioned effects were attenuated in multipollutant models that also considered PM and CO, the effect of NO2 on emergency room visits specifically for asthma was not, and NO2 showed the strongest association with these visits of all pollutants in the model. Asthma may not be the only disease that can predispose to enhanced effects from exposure. Some studies (e.g., D’Ippoliti et al., 2003; Burnett et al., 1997; Wong et al., 1999) noted an association between NO2 and admissions for cardiovascular disease but, as above, effect estimates were often modulated when other pollutants, especially PM, were

830

NITROGEN OXIDES

considered in the model as well. Some studies strongly suggest (Mann et al., 2002; Metzger et al., 2004) that individuals with ischemic heart disease and accompanying congestive heart failure and/or arrythmia may be a group especially sensitive to effects of motor vehicle derived pollution, which includes NO2. In an examination of air pollution and emergency room visits for cardiovascular causes in Atlanta, GA (Metzger et al., 2004), there was found to be a significant effect of NO2 on cardiovascular emergency room visits that remained so even after adjusting for PM. In an examination of air pollutants and congestive heart failure (Wellenius et al., 2005), effects with NO2 were modulated when considering CO and PM, but there was a conclusion that the general mix of pollution from motor vehicles was responsible for the observed effects. Finally, daily outdoor concentrations of NO2 were associated with emergency room admissions due to cerebrovascular disease and short-term ischemic attacks (Ponka and Virtanen, 1996). The discussion above involved acute responses to exposure. A few studies have examined the prevalence and/or incidence of asthma or allergic airway disease related to long-term exposure to NO2. These often show conflicting results, and there is no unequivocal evidence that chronic exposure to NO2 will increase these health outcomes (Studnicka et al., 1997; Dockery et al., 1989; Braun-Fahrlander et al., 1997; McConnell et al., 1999, 2003; Shima and Adachi, 2000; Peters et al., 1999b). There does, however, appear to be an association between long-term exposure to NO2 and decreased lung function growth with age in children, based upon studies in southern California (Peters et al., 1999a, 1999b; Gauderman et al., 2000, 2002), but NO2 was correlated with other motor vehicle related pollutants, again implicating motor vehicle derived emissions in these effects. A study of adults in Europe (Schindler et al., 1998) indicated a relationship between NO2 exposure and changes in lung function, as noted by FVC. Some European studies have provided indication that chronic exposure to NO2 is associated with increased risk of all cause mortality (Hoek et al., 2002; Nafstad et al., 2004; Filleul et al., 2005). These studies have suggested a specific association of NO2 with cardiopulmonary mortality as well. However, studies in the United States (Dockery et al., 1993; Pope et al., 2002) do not seem to provide similar evidence for such an effect of chronic NO2 exposure and all-cause mortality. In summary, given the available epidemiological evidence, it is not possible to provide an unequivocal conclusion regarding adverse health effects of NO2. There have been both positive and negative findings at various levels of NO2 exposure, and with various degrees of precision in measuring actual exposure levels. There does appear to be a relationship between exposure and increased mortality due to all cause, or due to cardiovascular and respiratory effects, although the effect estimate is generally reduced when adjustments are made for other pollutants, specifically PM and ozone. Thus, while short-term variations in NO2 appear to correlate with increased daily mortality, a definitive causal relationship cannot be concluded. Some results are also suggestive that an increase in acute respiratory illness, especially in younger children, may be associated with chronic exposure, although the extent of any such effect or excess risk is small. While there also appears to be an effect from NO2, which is independent of that from other related air pollutants, the exact extent of the contribution of NO2 is not always clear. However, the strongest association does appear to be in asthmatics, and this is either not modulated, or modulated to a lesser degree, by other pollutants. A number of studies indicate there to be a fairly strong association between NO2 and hospital admissions or emergency room visits for asthma in children where NO2 was the only pollutant associated, or where adjusting for other pollutants did not affect the

NITROGEN DIOXIDE

831

association with NO2. Finally, there is fairly strong evidence that chronic exposure to NO2 adversely affects lung function growth in children; this could be reflected in reduced function as adults. Thus, while both acute and chronic exposure to NO2 has been associated with adverse health outcomes, it is often unclear as to whether there is an independent effect; but there is clear evidence for traffic-related air pollution, which contains NO2, having adverse health effects. 22.3.4

Health Effects––Toxicology

Toxicological studies can be helpful in providing biological plausibility for health outcomes noted in epidemiological studies. However, a significant fraction of the NO2 toxicological database involves experimental exposures to concentrations >5 ppm. While such studies may help elucidate mechanisms of toxicity that influence responses to high concentrations, they are often of limited use in attempts to determine the public health significance from actual, much lower concentration, ambient air exposures. Thus, in this chapter, generally only studies using 5 ppm will be discussed. However, when necessary to help elucidate certain mechanisms, effects at higher levels will be presented as well. 22.3.4.1 Studies in Animals The largest database concerning the biological effects of NO2 is that derived from controlled exposures of laboratory animals. Since the mechanisms underlying many responses are similar across species, effects in these animals may have implications for humans. It should be borne in mind, however, that the exposure concentrations needed for comparable response likely differ between species. In any case, nitrogen oxides have been shown to exert a wide range of biological effects, mostly within the respiratory tract. Respiratory Tract Defenses Mucociliary transport provides a first line of defense against prolonged retention of deposited particles in the tracheobronchial tree. Acute (1–2 h) exposures to NO2 at levels 10 ppm did not alter mucociliary transport rate from the tracheobronchial tree of laboratory animals (Schlesinger, 1989). Rats exposed for 6 weeks to 6 ppm NO2 showed a transient depression in mucociliary activity (Giordano and Morrow, 1972), while rabbits exposed for 2 h/day for 14 days to 0.3 or 1 ppm did not show altered tracheobronchial mucociliary transport (Schlesinger et al., 1987). Thus, the available data suggests that with either single or short-term repeated exposures, higher than ambient levels are needed to alter tracheobronchial mucociliary transport. While the mechanisms underlying any such changes are not certain, they may involve NO2-induced effects upon ion transport across the airway epithelium (Robison and Kim, 1995) or upon ciliary beat activity (Ohashi et al., 1993). Particle clearance from the respiratory region of the lungs has also been assessed following exposures to NO2. Rats exposed to 1, 15, and 24 ppm showed a decrease in clearance after 22 daily exposures to 15 and 24 ppm, but accelerated clearance after exposures to 1 ppm (Ferin and Leach, 1977). Rabbits exposed for 2 h to 0.3, 1, 3, or 10 ppm showed accelerated clearance at all concentrations, while repeated 2 h/day exposures for 14 days to 1 or 10 ppm NO2 resulted in clearance patterns similar to those with single exposures at the same concentration (Vollmuth et al., 1986). Ferrets exposed to either 0.5 or 10 ppm NO2 for 4 h/day, 5 days/week for 8 or 15 weeks showed a reduction in clearance measured 12 weeks after the start of either exposure regime (Rasmussen et al., 1994).

832

NITROGEN OXIDES

Alveolar macrophages play a central role in the defense of the lungs, and alterations in numbers and functional properties of these cells may affect susceptibility to disease or injury. Macrophage numbers increased with continuous exposure of rats to 17 ppm, but not with 2 ppm (Stephens et al., 1972), and after 7 days of continuous exposure of rats to 4 ppm (Mochitate et al., 1986). However, no change in cell number was found following exposure of rabbits to 0.3 or 1 ppm NO2 for 2 h/day for 13 days (Schlesinger, 1987). A subpleural accumulation of alveolar macrophages was found in rats exposed for 7 h/day, 5 days/week for 15 weeks to 5 ppm NO2, but not to 1 ppm (Gregory et al., 1983). Rombout et al. (1986) noted some increase, by 2 days, in the number of macrophages in terminal bronchioles and adjacent alveoli in the lungs of rats exposed continuously to 5 ppm NO2; this was not seen with 1 or 2.5 ppm. Others have noted concentration-related increases in macrophage numbers with exposure to 5–40 ppm for 2 days to 15 week (Kleinerman et al., 1982; DeNicola et al., 1981; Busey et al., 1974; Wright et al., 1982; Foster et al., 1985). Various functional properties of macrophages essential to adequate defense, for example, surface attachment, mobility, and phagocytosis, have been assessed following exposure to NO2. Schlesinger (1987) exposed rabbits to 0.3 or 1 ppm for 2 h/day and found no effect on attachment, but a depression of mobility at day 3 in the 0.3 ppm group. Macrophages obtained from baboons exposed to 2 ppm for 8 h/day, 5 days/week for 6 months showed reduced responsiveness to migration inhibitory factor, a lymphokine that mediates cell movement (Greene and Schneider, 1978). Suzuki et al. (1986) found depressed phagocytic activity in macrophages obtained from rats exposed for 10 days to 4 or 8 ppm NO2, while Lefkowitz et al. (1986) noted no change in such activity in macrophages from mice exposed for 7 days to 5 ppm. The phagocytic activity of rabbit macrophages was reduced by in vivo exposure to 0.3 ppm, but was enhanced with exposure to 1 ppm by 3 days, and returned to control values by 7 days and remained there through 13 days of exposure (Schlesinger, 1987). An exposure–concentration dependent difference in the direction of phagocytic response seems to be a characteristic of NO2. Thus, Schlesinger (1989) found a reduction in phagocytic activity of macrophages recovered immediately after a 2-h exposure of rabbits to 1 ppm NO2; with 10 ppm, no change was seen immediately after exposure, but activity was increased 24 h postexposure. Ehrlich et al. (1979) found exposure of mice to 0.5 ppm NO2 for 3 h/day, 5 days/week for 2 months to depress phagocytosis, while Sone et al. (1983) showed enhanced phagocytosis in macrophages obtained from rats exposed to 40 ppm for 4 h/day for 7 days. The reasons for such differences in direction of response are unknown. Macrophages are a source of various biological mediators, and their ability to produce these may be compromised by pollutant exposure. The eicosanoids are a class of mediators produced in response to a wide variety of cellular perturbation and have various effects on airway physiology and the immune system. Alveolar macrophages obtained from rats exposed to 0.5 ppm NO2 for 0.5–10 days exhibited complex responses related to the production of eicosanoids (Robison et al., 1993). An initial depression of production was followed by recovery for some of these mediators, but not others, with increasing exposure duration. The complexity of response was also noted in a study of rat alveolar macrophages acutely exposed in vitro to 0.1–20 ppm NO2 (Robison and Forman, 1993). Low concentrations (up to 5 ppm) had small effects on basal synthesis of eicosanoids but amplified response to stimulated production of eicosanoids, while high concentrations (20 ppm) showed the reverse pattern of response. Finally, rats continually exposed to 10 pm for 1, 3, or 20 days showed reduced levels of tumor necrosis factor in lavage and suppression of cytokine

NITROGEN DIOXIDE

833

signaling-3 mRNA, both of which were likely due to NO2-induced changes in activation state of the macrophages (Garn et al., 2003). NO2 may impair the ability to resist infectious agents; this is suggested by some epidemiological studies noted above. Mice exposed continuously to 0.5 ppm NO2 showed increased mortality to K. pneumonia after 3 months of exposure (Ehrlich and Henry, 1968), while 0.05 ppm for 24 h/day for 15 days did not change bacterial resistance (Gardner et al., 1982). The finding of increased susceptibility does, however, depend upon the specific organism being used. Thus, while exposure of mice for 3 h/day for 3 months to 0.5 ppm increased mortality to Streptococcus sp. (Ehrlich et al., 1979), exposure to 0.5 to 1.5 ppm NO2 continuously for 3 months produced no effect on mortality due to K. pneumoniae (McGrath and Oyervides, 1985); on the contrary, exposure to 5 ppm for 3 days did result in enhanced mortality. It may be that peak exposure and exposure pattern are important modulators of response to NO2. A number of infectivity studies involved exposure to a baseline NO2 concentrations upon which spikes to a higher level were superimposed to mimic ambient exposures. The relative effect of such spikes is not always clear, but seems to depend upon both spike duration and time between spikes. Miller et al. (1987) noted that mortality due to infection was greater in a spike regimen (to 0.8 ppm) than in the baseline-exposed group (0.2 ppm). Others have found that both the number and amplitude of spikes are of importance in increasing mortality (Gardner et al., 1979; Graham et al., 1987). In fact, effects from such exposure excursions may approach those due to more continuous exposure to a lower concentration. This is consistent with the notion that, in general, brief exposures to high NO2 levels are more hazardous than are longer duration exposures to lower concentrations (Lehnert et al., 1994). The effect of NO2 on mortality due to bacterial infection appears to increase with both exposure duration (T) and peak concentration (C), although the latter seems to have more influence than the former for fixed C  T values (Gardner et al., 1979). Any differences between intermittent and continuous exposure also seem to disappear as the number of days of exposure increases (Gardner et al., 1979). Other studies suggest that, as concentration increases, a shorter exposure time is needed for intermittent and continuous exposure regimes to produce similar degrees of effect (Ehrlich and Henry, 1968; Ehrlich, 1979). Mortality is also proportional to exposure duration if the bacterial challenge is given immediately after exposure, but may not be when the challenge is given much later (Gardner et al., 1982). For example, no effect of 3.5 ppm NO2 for 2 h was seen in mice when bacteria were administered 27 h after exposure, while increased mortality was evident when administration was immediately after NO2 inhalation (Ehrlich, 1980). Effects of 25 ppm for 2 h on mice were seen only when the microbial challenge was given within 72 h after NO2 exposure (Purvis and Ehrlich, 1963). These results suggest that a critical time frame exists between exposure and bacterial challenge after which NO2 will not affect resistance. The mechanism(s) underlying any NO2-induced change in host resistance to bacteria are not known. However, since exposure levels that alter resistance do not affect physical clearance processes, the response to NO2 may be due to impaired intracellular killing of microbes, perhaps reflecting macrophage dysfunction. For example, macrophages are a source of numerous biochemical mediators that are directly involved in antibacterial action, for example, superoxide anion, and a depression in superoxide production has been noted following NO2 exposure in some studies (Amoruso et al., 1981; Suzuki et al., 1986; Robison et al., 1993), although at higher than ambient levels. However, human alveolar macrophages exposed in vitro to 0.1–0.5 ppm NO2 for 30–120 min showed increased reactive oxygen

834

NITROGEN OXIDES

intermediate production in a dose-dependent fashion (Kienast et al., 1994). The effects of NO2 on viral infectivity have also been examined. Exposures to 0.5 ppm or greater on a continuous basis likely increase susceptibility, while at higher concentrations the exposure duration needed for any effect is lowered (Ito, 1971; Rose et al., 1988). Furthermore, environmental stresses may enhance the lethality of infectious agents over and above that due solely to NO2 exposure. These may include exercise (Illing et al., 1980), elevated temperature (Gardner et al., 1982), and the presence of other pollutants. Exposure to NO2 may affect allergic response. Rats exposed to 5 ppm for 3 h after sensitization with house dust mite antigen had higher levels of serum IgE and local respiratory tract IgA, IgG, and IgE antibodies than did controls (Gilmour, 1995). The exposed animals also had increased lymphocyte activity in the spleen and local lymph nodes and showed an increase in respiratory tract inflammatory cells. This suggests that NO2 may enhance immune responsiveness and increase the severity of pulmonary inflammation in sensitized lungs and may, thus, play some role in the exacerbation of immune-mediated respiratory disease. A number of studies have examined the effects of NO2 on specific parameters of respiratory and/or systemic humoral and cellular immunity. While immune suppression and/or enhancement of factors involved in airway hyperreactivity has clearly been shown to follow exposure to levels of NO2 above 5 ppm, as evidenced by various end points including the response of T-cells, antibodies, or production of interferon or other inflammatory mediators (e.g., Holt et al., 1979; Valand et al., 1970; Fujimaki and Shimizu, 1981; Campbell and Hilsenroth, 1976; Ayyagari et al., 2004), there are only a few reports of response to lower levels. These studies suggest that short-term repeated exposures may result in reductions in counts of lymphocytes in the lungs or spleen, or a depression in antibody responsivity to particular antigens. Mice exposed for 7 h/day, 5 days/week for 7 weeks to 0.25 ppm NO2 showed reduced total T-lymphocyte numbers in the spleen, with concomitant reductions in certain subpopulations of these cells, for example, helper cells (Richters and Damji, 1988). Exposure of mice for 3 months to 0.5 ppm NO2 resulted in a depressed responsiveness of both T- and B-lymphocytes in spleen (Maigetter et al., 1978). No effect on the cell-mediated immune system was found either in mice exposed for 24 h to 5 ppm NO2 (Lefkowitz et al., 1986), or in those exposed to 1.6 ppm NO2 for 4 weeks (Fujimaki et al., 1982). Mice exposed to 0.4 or 1.6 ppm NO2 for 4 weeks showed depressed primary antibody responsivity to sheep red blood cells in vitro (Fujimaki et al., 1982) while mice exposed to 4 ppm NO2 continuously for up to 56 days showed no change in the antibody response to Tcell-dependent and independent antigens in spleen (Fujimaki, 1989). In another study, mice were vaccinated with influenza virus after they had undergone 3 months of continuous exposure to 0.5 or 2 ppm NO2 with daily spikes (1 h) of 2 ppm for 5 days/week. Both concentrations resulted in a reduction in mean serum neutralizing antibody titers (Ehrlich et al., 1975). Guinea pigs exposed to 1 ppm for 6 months showed a reduction in all immunoglobulin fractions (Kosmider et al., 1973). On the contrary, Balchum et al. (1965) noted an increase in serum antibody titers against lung tissue in guinea pigs exposed to 5 ppm NO2 for 4 h/day after 160 h of exposure, and further increases as exposure duration increased. Enhanced immune function may be just as detrimental as suppressed function, through overstimulation of response and hypersensitivity. As with other end points, the effects of NO2 upon the immune system appear to be related to various exposure parameters. While some studies show no effects, others show enhancement or depression of immune parameters, depending upon the exposure concentration, the

NITROGEN DIOXIDE

835

length of exposure, and the animal species used. In addition, the direction of change appears to depend upon exposure concentration. For example, humoral response in monkeys chronically exposed to NO2 was enhanced at a low concentration (1 ppm), but suppressed at a higher level (5 ppm) (Fenters et al., 1971, 1973). Respiratory Tract Structure Exposure to NO2 may produce structural alterations in the respiratory tract. As noted, the anatomic region most sensitive to NO2 is the area encompassing the terminal and respiratory bronchioles and adjacent alveolar ducts and alveoli. The primary cellular targets within this region are ciliated cells of the bronchiolar epithelium and Type 1 cells of the alveolar epithelium. Acute exposure to NO2 can result in hypertrophy and hyperplasia of alveolar Type 1 cells, followed by cell death and desquamation and proliferation of and replacement by Type 2 cells. The end result can be a thickened air–blood barrier. The bronchiolar response is characterized by hypertrophy and hyperplasia of epithelial cells, loss of secretory granules and surface protrusions of Clara cells, and loss of ciliated cells, or of cilia. With chronic exposure, many of these same changes are seen, but there is increased cilia loss over larger areas of epithelium and in more proximal airways, and the structure of the remaining cilia may be altered. The temporal progression of NO2-induced lesions has best been described for the rat. The earliest alterations resulting from concentrations 2 ppm occur within 24–72 h of continuous exposure, with repair of injured tissue and replacement of destroyed cells beginning within 24–48 h of continuous exposure. Division of Type 2 cells is observed within 12 h after initial NO2 exposure, the rate becoming maximal by about 48 h, and then decreasing to preexposure levels by about 6 days, even with continued exposure. In some cases, the resolution of NO2-induced morphologic changes may be complete after exposures end; on the contrary, some lesions may resolve while others remain, even when exposure continues (Rombout et al., 1986; DeNicola et al., 1981; Kubota et al., 1987). Chronic exposure to NO2 may result in alterations in lung architecture resembling emphysema-like disease, for example, enlargement of airspaces, increase in mean linear intercept (a measure of the distance between alveolar walls), and reduction in the internal surface area of the alveolar region. However, the relationship between exposure and the development of emphysema remains unclear. A problem in evaluating reported emphysematic changes in animal models is the definition of the disease, which has changed over the years and which has been defined differently by various professional groups (NIH, 1985). While long-term exposure to high NO2 concentrations (>10 ppm) are required to produce clearly definable emphysema-like changes (e.g., Barth et al., 1995), there is evidence that lower NO2 levels may result in emphysema, emphysema-like changes, or altered alveolar dimensions if present in complex mixtures of NOx (Hyde et al., 1978) or when administered during lung development (Rasmussen and McClure, 1992). However, clear evidence of changes characteristic of human emphysema, that is, alveolar septal degeneration, enlarged airspaces, and associated functional changes, is absent with exposure at low levels. There is, however, some evidence for changes similar to those seen in human emphysema with exposure to high concentrations. These involved exposures to levels ranging from 8 to 20 ppm for up to 33 months (Haydon et al., 1967; Freeman et al., 1972). Another study that involved exposure of dogs for 5.5 years to a mixture of NO2 at 0.64 ppm and NO at 0.25 ppm followed by a postexposure period of 2.5 years noted structural changes similar to human centrilobular emphysema that were noted after the postexposure period had ended (Hyde et al., 1978).

836

NITROGEN OXIDES

While the extent and degree of structural alterations induced by NO2 appear to be related to exposure concentration, little is known about effects of other modifying factors, for example, exposure duration or the temporal pattern of exposure. The contribution of exposure time in the histopathologic response to acute inhalation was examined in rats (Stavert and Lehnert, 1988). The most pronounced effects were found with the highest NO2 concentration in any particular set of exposures where the product of concentration and time was equivalent, indicating that concentration played a more important role than did exposure time in tissue injury. This is consistent with the relative roles of C and T in infectivity, discussed previously. Another study (Rombout et al., 1986) assessed the concentration–time response relation for intermittent and continuous exposures, and likewise concluded that concentration played a more important role in inducing morphologic lesions than did exposure duration, as long as the product of C  T was constant. The effect of concentration was found to be greater with intermittent than with continuous exposure, and the onset of response was also delayed with intermittent compared to continuous exposure. The morphological effects of exposure patterns involving transient spikes were examined in a number of studies (Gregory et al., 1983; Miller et al., 1987; Crapo et al., 1984; Chang et al., 1986). Results are equivocal, and it is not clear whether these peaks significantly contributed to morphological damage in excess of that due to integrated exposure. In spite of the fact that there is a fairly extensive database concerning morphologic effects of NO2 in animal models, it is still quite difficult to establish a threshold exposure condition for these end points. This is due to the great complexity of changes occurring with exposure, as well as to large interspecies differences in response. For example, the rat appears to be less sensitive to NO2 compared to other species, such as the guinea pig or monkey. Furthermore, different cell types show differential sensitivity to NO2. In general, morphological alterations, some of which may be persistent, are found with chronic exposure to concentrations <1 ppm. However, long-term exposures to levels 2 ppm are generally required to produce more extensive or permanent changes. An added complication in evaluating morphologic effects is that they may depend upon the age of the animals at the time of exposure (Stephens et al., 1982; Azoulay-Dupuis et al., 1983; Chang et al., 1986; Kyono and Kawai, 1982). Neonates, specifically prior to weaning, seem to be relatively resistant to NO2, with sensitivity increasing with age until adulthood. However, the response in animals of different ages is similar in terms of the cell types affected, the nature of the damage incurred, and repair capacity. Age-related differences occur in the extent of damage and in the time required for repair, this latter taking longer in older animals. The reasons for age differences in sensitivity are not known, but may reflect diet and variable sensitivity of target cells during different growth phases (Stephens et al., 1982; Hahn, 1979). In any case, age-related differences in response are also observed in epidemiological studies. Of importance in assessing the morphological effects of NO2 is consideration of individuals with compromised lung function, for example, those with respiratory disease. There is, however, a very limited database in laboratory animals with preexisting chronic disease. No effect of NO2 exposure upon pneumoconiosis development in guinea pigs was found in one study (Gross et al., 1968). Two studies assessed whether prolonged exposure to NO2 would alter the progression or severity of preexisting emphysema (produced by elastase instillation). Nitrogen dioxide did not potentiate preexisting disease in rats (Stavert et al., 1986), but may have done so in hamsters (Lafuma et al., 1987). It is also possible that acute lung disease may affect NO2 toxicity. Fenters et al. (1973) challenged squirrel monkeys with an influenza virus at various times during continuous

NITROGEN DIOXIDE

837

exposure to 1 ppm NO2 for 16 months, and compared the response to that seen in animals not challenged but similarly exposed. Only the virus-challenged animals showed effects of NO2, namely slight emphysema-like changes and thickening of the bronchial and bronchiolar epithelium. This suggests that the presence of acute lung disease may have affected NO2 toxicity. Respiratory Tract Biochemistry Nitrogen dioxide is, as noted, quite reactive. Exposure can result in damage to the cell membrane, and fairly low exposure levels have been associated with alterations to specific membrane components. For example, lipid peroxidation was noted in rats exposed to 0.04 ppm for 9 months (Sagai et al., 1984). However, extended exposures at low levels may be needed for such effects, since rats and guinea pigs exposed to 0.4 ppm NO2 for 24 h/day for only 2 weeks showed no change in the level of lipid peroxides in the lungs (Ichinose and Sagai, 1989). With the exception of effects on lipids, most studies of biochemical pulmonary alterations show significant effects with acute or short-term repeated exposures generally only at levels above about 2 ppm. Such effects include oxidation of protein or protein components, such as elastin and collagen. Thickened collagen fibrils were noted in the lungs of monkeys exposed to 3 ppm for 4 h/day for 4 days (Bils, 1976), while increased rates of lung collagen synthesis, a possible marker for development of fibrosis, has been noted in NO2-exposed rats (Last et al., 1983; Last and Warren, 1987). Exposures to high concentrations (Kleinerman et al., 1985; Kleinerman and Ip, 1979) suggest that NO2 may reduce elastin content via an increase in the activity of neutrophil elastase, the enzyme responsible for elastin breakdown. Other NO2-induced biochemical effects related to proteins involve changes in activity of various pulmonary enzymes. For example, glutathione (GSH) is a reducing compound found in the lungs, and NO2 exposure has been reported to alter the activity of enzymes that regulate its levels, or to affect the lung content of glutathione itself. Suppressed GSH-peroxidase activity has been noted, for example, in mice exposed continuously for 17 months to 1 ppm, but not to 0.5 ppm (Ayaz and Csallany, 1978), while an increase in GSH-reductase activity was noted in mice exposed to 6 ppm for 4 h/day for 30 days (Csallany, 1975) and in rats exposed to 6.2 (but not 1 or 2.3 ppm) for 4 days (Chow et al., 1974). High NO2 exposure levels are, thus, apparently needed for changes in GSH metabolism. The biochemical response to NO2 may be modulated by dietary factors, in particular, levels of certain antioxidant vitamins. Increases in the lavage content of proteins and lipids were noted in vitamin C depleted guinea pigs exposed to 1 ppm NO2 for 72 h, but not in normal controls (Selgrade et al., 1981); similarly depleted guinea pigs exposed for 1 week to 0.4 ppm showed an increase in lavage protein content, an indicator of serum transudation and possible membrane damage (Sherwin and Carlson, 1973). Another possible modulator of effect is vitamin E. Changes in protein content and enzyme activity in lung homogenates from rats exposed for 7 days to 3 ppm were found to be more severe in those animals which were deficient in this vitamin (Elsayed and Mustafa, 1982). Respiratory tract tissue can metabolize various xenobiotics, and it is possible that inhaled pollutants may alter the ability to handle these materials. However, if NO2 does alter xenobiotic metabolism, it is only at high exposure levels. Thus, exposure of rats continuously to 4 ppm NO2 for up to 2 months produced no change in the cytochrome P-450 content of the lungs, although some other xenobiotic metabolizing enzymes were decreased (Takahashi and Miura, 1989).

838

NITROGEN OXIDES

Respiratory Mechanics The effects of NO2 on respiratory mechanics have been studied in laboratory animals using standard indices of function, with mixed results. Hamsters exposed to 2 ppm for 8 h/day, 5 days/week for 8 weeks exhibited an increase in tidal volume, but no change in compliance or vital capacity (Lafuma et al., 1987). Exposure of mice to 0.2 ppm NO2 for 23 h/day, 7 days/week for up to 52 weeks resulted in no change in pulmonary mechanics (Miller et al., 1987); however, when 1 h spikes to 0.2 ppm twice daily for 5 days/week were superimposed upon this baseline, a significant decrease in endexpiratory volume and vital capacity, as well as a trend toward increased residual volume, were found. Stevens et al. (1988) exposed neonate and 7-week-old mice continuously for 1, 3, or 6 weeks to baseline levels of 0.5, 1, or 2 ppm NO2 upon which were superimposed twice daily 1 h spikes (5 days/week) to 1.5, 3, or 6 ppm, respectively. The two higher levels produced increased vital capacity and compliance by 3 weeks only in the neonates, but the effect resolved by 6 weeks. Furthermore, adult animals showed a reduction in compliance after 6 weeks of exposure to 2 ppm. Suzuki et al. (1982) noted a concentrationrelated increase in respiratory rate in mice exposed to 5–20 ppm NO2 for 24 h; exposure to 5 ppm also resulted in a decrease in arterial CO2 tension (PaCO2), suggesting hyperventilation. Bronchoprovocation challenge testing is often used to assess nonspecific airway hyperresponsivity. Silbaugh et al. (1981) examined the effects of histamine aerosol on guinea pigs exposed to NO2 for 1 h at 7–146 ppm. While a concentration-related increase in sensitivity to histamine was noted when the latter was inhaled 10 min after NO2 exposure (but not 2 or 19 h after exposure), the response became significant only when NO2 levels were >25 ppm. A study involving long-term exposures (Kobayashi and Miura, 1995) involved exposure of guinea pigs to 0.06, 0.5, 1, 2, or 4 ppm NO2 for 24 h/day for 6 or 12 weeks. Airway responsiveness to histamine and specific airway resistance were assessed on the last day of each exposure. Exposure to 2 and 4 ppm by 6 weeks of exposure resulted in increased airway responsiveness, while exposure to the same concentration resulted in increased resistance by 12 weeks of exposure. Tepper et al. (1993) performed a long-term exposure of rats to NO2. Animals were exposed to NO2 having a 0.5 ppm background with 1.5 ppm peaks (2 h) for up to 78 weeks. No exposure-related changes in nitrogen washout, compliance, lung volume, or CO diffusion capacity were noted, but at 78 weeks there was some reduction in a measure of forced expiratory flowrate. However, the authors indicated that the change was borderline, and suggested that long-term exposure to high ambient urban levels did not lead to any dysfunction suggestive of degenerative lung disease. The overall database suggests that NO2 at realistic levels in terms of ambient exposure has not been shown to significantly alter pulmonary mechanics or bronchial responsivity in animal models, consistent with results of controlled clinical studies in humans. Extrapulmonary Effects Exposure to NO2 may affect target sites beyond the respiratory tract. End points that have been shown to be altered include body weight, blood cell counts, blood cell membrane and serum chemistry, liver and kidney function, brain protein enzymes, and neuromotor function (e.g., Graham et al., 1982; Tabacova et al., 1985; Freeman et al., 1966; Wagner et al., 1965; Case et al., 1979; Kaya et al., 1980; Kaya and Miura, 1982; Mochitate et al., 1984; Kosmider et al., 1973; Miller et al., 1980; Takahashi et al., 1986; Sherwin and Layfield, 1974). However, the data are conflicting and, because of this, the ability to relate reported changes to human health effects is severely limited.

NITROGEN DIOXIDE

839

Carcinogenicity/Reproductive Toxicity Exposure to NO2 even at high levels does not seem to be genotoxic or teratogenic in appropriate assay systems (Kripke and Sherwin, 1984; Gooch et al., 1977). While one study did note an increase in the rate of DNA strand breaks in hamster cells exposed in vitro to 10 ppm for 20 min, exposure to 5 ppm for up to 30 min had no such effect (G€ orsdorf et al., 1990), and in vivo exposure of mice to 20 ppm for up to 23 h did not result in any genotoxicity (Victorin et al., 1990). This apparent conflict in response may be due to repair mechanisms operating in vivo that are not operative in in vitro assays. The ability of NO2 to act as a carcinogen, or cocarcinogen, is unclear, but there is no direct evidence that NO2 exposure results in the development of tumors. Some concern is, however, based upon the fact that exposure can result in nitrite in blood and this, in turn, may produce carcinogens, such as nitrosamines, after further reaction in the body. Although there have been no long-term carcinogenesis bioassays performed with NO2, one chronic inhalation study, in which mice were exposed to 1, 5, and 10 ppm NO2 for 6 h/day, 5 days/week for 6 months, suggested a small increase in tumor (pulmonary adenoma) frequency and incidence in the highest dose group (Adkins et al., 1986). However, such data must be interpreted with caution, and the relationship between cancer development in mice and that in humans is not clear. Although not likely a carcinogen itself, NO2 may modulate tumorogenic processes in the lungs (Witschi, 1988). For example, in conjunction with a specific carcinogen, NO2 exposure may be involved in the pathogenesis of small cell carcinoma (Witschi, 1988), especially since it has been shown to modulate the number of neuroendocrine cells, the precursor cells for this disease (Kleinerman et al., 1981; Palisano and Kleinerman, 1980). As another example, an enhancement of tumor colonization in the lungs of mice injected (IV) with melanoma cells was noted after exposure to NO2 at 0.4 or 0.8 ppm for 8 h/day, 5 days/week for 10–12 weeks (Richters and Kuraitis, 1981). This could be due to injury of lung capillary endothelium by NO2, facilitating metastases of blood-borne cancer cells to the lungs (Richters and Richters, 1989), or to the suppression of immune system components. However, as with other end points, the database regarding the role of NO2 in carcinogenic processes is conflicting. For example, NO2 has been shown to actually enhance the cytotoxic response of macrophages (Sone et al., 1983), which implies greater antitumor defense capabilities. Thus, any role for NO2 in cancer etiology requires further evaluation. Is there any epidemiological evidence that NO2 is involved in the etiology of cancer? Some studies do suggest a relationship between NO2 and lung cancer (e.g., Hoek et al., 2002; Nafstad et al., 2004). However, NO2 is generally associated with other pollutants from the same source, many of which are known carcinogens, so all results related directly to NO2 must be interpreted with great caution. There have been some epidemiological studies suggesting that exposures to mixtures containing NO2 during pregnancy may be associated with fetal/reproductive effects, such as low birth weight and perinatal mortality but, again, any independent effect from NO2 is unclear (Liu et al., 2003; Wilhelm and Ritz, 2003). 22.3.4.2 Controlled Studies with Humans By their nature, studies with human volunteers can only be used to evaluate transient effects of acute exposure. They have generally used changes in standard respiratory mechanical indices as markers of response; a few studies, however, have employed other end points, which include bronchoprovocation challenge testing, clearance of inhaled aerosols, and analysis of biochemical and cellular

840

NITROGEN OXIDES

components of bronchopulmonary lavage. Various subject groups have been examined. These include healthy individuals with no history of respiratory disease, allergy, etc., as well as people with allergies or a history of asthma or chronic obstructive pulmonary disease (COPD). Acute exposures (up to about 2 h) to NO2 at levels 1 ppm have not resulted in any consistent, significant changes in respiratory mechanics in normal, healthy adult subjects at rest (Beil and Ulmer, 1976; Bylin et al., 1985; Hazuch et al., 1982; Koenig et al., 1985; Bascom et al., 1996). Regarding higher levels, the study of Beil and Ulmer (1976) involving 2-h exposures to 1, 2.5, 5, and 7.5 ppm, indicated a change in total respiratory resistance occurring at 2.5 ppm, although the effects were quite small. von Nieding et al. (1973) noted a decrease in diffusing capacity (DLco) with a 15-min exposure to 5 ppm. Finally, no changes in lung mechanical function were found with exposure to 2 ppm for 2 h, or for 2 h/day for 3 days (Mohsenin, 1988; Goings et al., 1989). Exposure to 1 ppm NO2 in conjunction with various degrees of exercise have also resulted in inconsistent effects on respiratory mechanics in healthy people; most of the studies showed no effects which could be unequivocally attributed to NO2 (Folinsbee et al., 1978; Frampton et al., 1989a; Hackney et al., 1978; Kerr et al., 1979; Morrow and Utell, 1989). Reduced compliance was noted following exposure for 2 h at 0.5 ppm (Kulle, 1982), but the meaning of this with a lack of other lung mechanical changes was not clear. Linn et al. (1985a) found no change in resistance or spirometry with exposure at 4 ppm for 75 minutes. Increased airway responsivity in healthy subjects has been noted following a 2-h exposure to 7.5 ppm (Beil and Ulmer, 1976), with 1 h to 2 ppm (Mohsenin, 1988), and with 3 h (with intermittent exercise) to 1.5 ppm (Frampton et al., 1989a). Again, however, the results are not consistent, with other studies at similar concentrations and exposure durations finding no change (e.g., Kulle and Clements, 1987). Exposures at 0.6 ppm have not produced any change in responsivity at all (Morrow and Utell, 1989; Frampton et al., 1989a; Bylin et al., 1985; Hazuch et al., 1983). Particular subsegments of the population may be especially susceptible to the effects of NO2. As noted in epidemiological studies, one such group is asthmatics. A number of studies have been performed with exposure levels ranging from 0.1 to 4 ppm for durations ranging up to 4 h, usually with exercise; effects on various aspects of mechanical function, such as spirometry or airway resistance, have ranged from none to slight, and all with much inconsistency (e.g., Avol et al., 1988; Bauer et al., 1986; Ahmed et al., 1982; Hazuch et al., 1982, 1983; Bylin et al., 1985; Koenig et al., 1985, 1987; Kerr et al., 1979; Rubinstein et al., 1990; Morrow and Utell, 1989; Roger et al., 1990; Kleinman et al., 1983; Linn et al., 1985a; Mohsenin, 1987; Salome et al., 1996; Morrow et al., 1992). A study that measured pulmonary function in adult asthmatics in their home and also monitored indoor NO2 levels noted that average exposures to >0.3 ppm produced a decline in certain pulmonary function measures, but inconsistent effects were seen at lower exposure levels (Goldstein et al., 1988). Finally, when there is any response, it may only occur with exercise. Exposure to 0.3 ppm for 30 min produced no change in pulmonary mechanics indices in resting asthmatics, but effects were noted when exercise was incorporated into the exposure protocol (Bauer et al., 1986). The most sensitive pulmonary mechanical response to NO2 in people with airway disease appears to involve changes in airway responsiveness. However, there is much variabilty in results from different studies, and also an apparent lack of a dose–response relationship. While some studies have indicated increased responsiveness due to NO2 exposures at 0.14– 0.5 ppm (Bauer et al., 1986; Bylin et al., 1988; Kleinman et al., 1983; Mohsenin, 1987;

NITROGEN DIOXIDE

841

Salome et al., 1996; Strand et al., 1996), others have indicated no such effects at similar levels (Hazuch et al., 1983; Roger et al., 1990; Linn et al., 1986; Orehek et al., 1981; Avol et al., 1988; Bylin et al., 1988), and exposure to a much higher level (3 ppm) has also produced no effect (Linn et al., 1986). Any NO2-induced increased responsiveness may occur with a several-hour delay following exposure in asthmatics (Strand et al., 1996). Potentiation of cold-induced airway constriction and airway responsiveness to histamine in asthmatics was enhanced by exposure to 0.3 or 0.26 ppm, respectively (Bauer et al., 1986; Strand et al., 1996). A meta-analysis of a number of studies involving both asthmatics and normals indicated that acute exposure to NO2 would enhance responsiveness to various stimuli with exposure to at least 0.11 ppm in asthmatics, but at least 1 ppm in normals (Folinsbee, 1992). While the mechanism of any NO2-induced hyperresponsiveness is not known, it may involve alterations in the metabolism of endogenous bronchoconstrictors (Hoshi et al., 1996) or activation of specific cells within the airways (Ohashi et al., 1993). Mild asthmatics and normals were exposed to 1 ppm NO2 (with intermittent exercise) for 3 h, followed by bronchopulmonary lavage 1 h postexposure. While no change in differential cell counts was noted in either group, the asthmatics showed changes in lung eicosanoids not seen in normals, suggesting that NO2 could activate cells compatible with airway inflammation (Jorres et al., 1995). Even if asthmatics or allergic individuals may not show any enhanced response directly to NO2, exposure may alter their response to antigens. A number of studies have examined response to NO2 in terms of enhancement of the response to inhaled allergens in sensitized individuals. Humans having a history of allergic rhinitis were exposed to 0.4 ppm NO2 for 6 h, followed by challenge with an allergen (Wang et al., 1995). There was some evidence that NO2 primed eosinophils for subsequent activation by the allergen. Acute exposure to 0.43 ppm NO2 enhanced airway constriction in mild asthmatics in response to inhaled house dust mite antigen (Tunnicliffe et al., 1994). Similarly, airway constriction to pollen was enhanced in allergic asthmatics acutely exposed to 0.27 ppm NO2 (Strand et al., 1997) and following repeated exposure at 0.27 ppm (Strand et al., 1998). Finally, allergic asthmatics were exposed in a roadway tunnel to NO2 at a median level of 0.17 ppm, but ranging from 0.11 to 0.25 ppm, for 30 min; subsequent inhalation of an allergen resulted in greater early asthmatic reaction and more symptoms during the later phase asthmatic response when compared to air control exposed individuals (Svartengren et al., 2000). However, one must realize that roadway exposure involves more than just NO2, so effects may not have been due to NO2 alone, or at all. Nitrogen dioxide induced modulation of response to antigens may be due to recruitment of eosinophils (Barck et al., 2002). Thus, for example, subjects with allergic asthma were exposed to 0.27 ppm for 15 min on one day and for two 15-min intervals the next day; they were noted to have increased levels of eosinophil cationic protein, a component found in eosinophil granules, in both systemic blood and sputum when subsequently exposed to allergen. Another possibly sensitive subsegment of the population is people with COPD, that is, chronic bronchitis and emphysema. Increased airway resistance has been found in individuals with COPD after exposure to 1.6 ppm in conjunction with exercise (von Nieding and Wagner, 1979), while a decrease in FVC was noted following exposure to 0.3 ppm for 4 h with intermittent exercise (Morrow and Utell, 1989), and a decrease in FEV1 was noted following exposure for 1 h to 0.3 ppm (Vagaggini et al., 1996). On the contrary, no changes in airway resistance in chronic bronchitics exposed to 0.5 ppm for 2 h with exercise, or in

842

NITROGEN OXIDES

spirometry of COPD patients exposed to 0.5–2 ppm for 1 h also with exercise, have been noted (Kerr et al., 1979; Linn et al., 1985b). Thus, the database is currently not sufficiently robust to allow determination of the specific exposure conditions, that is, concentration, duration, and ventilation, for threshold effects on lung function in healthy humans with acute exposure. Lung mechanics may, in fact, not provide very sensitive indices of response in such people. On the contrary, functional changes may occur in individuals with asthma and/or COPD following exposure to lower levels of NO2 than those affecting normals. Again, however, the results are inconsistent. Results of one study examining pulmonary functional indices with asthmatics have not been confirmed by a subsequent one, or responses of a particular subject group are not always reproducible (Orehek et al., 1976; Hazuch et al., 1983; Bauer et al., 1985; Bromberg, 1988). There is, however, some evidence that especially sensitive subgroup(s) may exist within the asthmatic population (Bauer et al., 1986; Morrow and Utell, 1989). That is, the variability in responses noted above may be the result of differences in the severity or type of asthma in the subjects examined within one study, or between different studies. Asthmatics also exhibit a wide range of response to external stimuli, so some variability may merely be due to an interindividual variation in response to NO2. The lowest concentration that does result in observed effects on airway responsivity in exercising asthmatics is in the 0.2–0.5 ppm range; in normals, levels of 5 ppm may cause bronchoconstriction, but minimum levels of at least 1– 2 ppm are generally needed for changes in pulmonary functional parameters. Most mild asthmatics are not sensitive to NO2 at less than or equal to 0.6 ppm, at least in terms of changes in respiratory mechanics, while nonspecific airway responsiveness in mild asthmatics may be increased at levels >0.1 ppm. Controlled clinical studies have examined other aspects of pulmonary biology after exposure to NO2. Humans exposed for 20 min to 1.5–3.5 ppm NO2 did show a reduction of mucociliary activity measured 45 min following exposure (Helleday et al., 1995). The effects upon infectivity of an attenuated influenza virus in healthy humans was assessed by Kulle and Clements (1987); NO2 exposure levels were 1 to 3 ppm. There were no overall statistically significant changes in infectivity rates, although they were elevated in some of the NO2 exposed groups. In another study (Goings et al., 1989), there was suggestive evidence that exposure for 2 h/day for 3 days to 1 or 2 ppm NO2 increased susceptibility to respiratory viruses in healthy adults. Frampton et al. (1989b) examined the effect of NO2 exposure in vivo on the ability of alveolar macrophages to inactivate influenza virus in vitro. Healthy humans were exposed either to 0.6 ppm for 3 h or to 0.05 ppm for 3 h with three 15-min spikes to 2 ppm. There appeared to be less effective inactivation of the virus by macrophages harvested from the humans exposed to 0.6 ppm, but the results just missed statistical significance. No effects were noted in the individuals exposed to the lower concentration with the 2 ppm spikes. There also seemed to be a trend of increased production of interleukin-1 (IL-1) by macrophages from some individuals, namely those whose cells tended to have reduced viral inactivation activity. Effects on IL-1 were also examined by Pinkston et al. (1988), with exposure of macrophages harvested by lavage to 5–15 ppm NO2 for 3 h. No change in cell viability nor in release of IL-1 was noted. In any case, increased infectivity in NO2-exposed laboratory animals together with the above suggestive findings in humans indicates that NO2 may indeed alter host defense in humans. Healthy subjects exposed to 1–3 ppm NO2 for 2hr/day for 3 days and then exposed to attenuated influenza virus showed a slight trend toward increased infectivity (Goings et al., 1989). However, exposure for 3.5 h to 0.6 ppm NO2 resulted in a decreased inactivation of the influenza virus by alveolar macrophages (Frampton et al., 1989b).

NITROGEN DIOXIDE

843

Healthy subjects exposed to 0.6 ppm for 2 h on 4 days showed a small increase in the percentage of NK-lymphocytes (Rubinstein et al., 1991), but repeated exposures to 1.5 or 4 ppm for 20 min every other day for a total of 6 days reduced numbers of both B- and NKlymphocytes and altered the ratio of CD4 þ /CD8 þ cells (Sandstr€ om et al., 1992a, 1992b). Healthy subjects exposed to 1.92 ppm NO2 for 4 h on 4 days showed upregulation of the expression of IL-5, IL-10, IL-13, and ICAM-1 (Pathmanathan et al., 2003); the effects on the interleukins suggest that repeated exposure may exert a proallergic effect on the airway epithelium, while the effect on ICAM suggests a mechanism for neutrophil influx into the epithelium during an inflammatory response. Some other biochemical effects of inhaled NO2 have been examined in controlled clinical studies. In vitro exposure of human blood to high levels (>6 ppm) of NO2 has been shown to result in production of methemoglobin (metHb) (Chiodi et al., 1983), but Chaney et al. (1981) found no such change in normal humans exposed for 2 h to 0.2 ppm. A reported elevation of glutathione in these exposed subjects was not supported by the results of Posin et al. (1978), who found no effect following exposure to 1 ppm. The results of some of these studies are clouded by the lack of any consistent dose–response relationship. Exposure to 4 ppm NO2 for 20 min resulted in an inflammatory response in healthy individuals, as evidenced by changes in lymphocyte counts in lavage fluid obtained 4–24 h after exposure (Sandstr€ om et al., 1990). This is, however, not a consistent finding in humans (e.g., Mohsenin and Gee, 1987), possibly due to differences in experimental protocols, such as the times at which lavage was performed after exposure. Thus, exposure to 0.3 ppm for 1 h with exercise produced no acute inflammation in the proximal airways of normals, asthmatics or people with COPD (Vagaggini et al., 1996). It should be noted that exposures of laboratory animals to NO2 at levels up to 8 ppm for up to 10 days did not produce evidence of acute inflammation (Schlesinger et al., 1987; Gregory et al., 1983; Mochitate et al., 1986; Suzuki et al., 1986). Perhaps NO2 is not very effective in eliciting an inflammatory response at ambient levels with short-term exposure. Nitrogen dioxide exposure has been associated with development of emphysema in animal models. A component of the lungs’ defense against proteolysis is a-1-protease inhibitor. Mohsenin and Gee (1987) noted a decrease in levels of this enzyme in the lavage fluid of subjects exposed to 3–4 ppm for 3 h. However, the investigators noted that the extent of the decrease was not associated with any increased risk of emphysema. On the contrary, exposure of normal humans for 3 h (with intermittent exercise) to 1.5 ppm, or for 3 h to 0.05 ppm with three 2 ppm peaks, did not result in any change in activity of a-1-protease inhibitor in lavage fluid (Johnson et al., 1990). A 3-h exposure to 0.6 ppm resulted in an increase in levels of another antiprotease (a-2-macroglobulin) in lung lavage (Frampton et al., 1989c). In another study of potential lung damage, normal humans exposed to 0.6 ppm NO2 for 4 h/day for 3 days showed no effect on the excretion of hydroxyproline, a marker for connective tissue injury (Muelenaer et al., 1987). Effects of repeated exposures, which would more likely be involved in disease development, on these end points are unknown. 22.3.5

Health Effects––Summary and Conclusions

A large database exists concerning biological responses resulting from the inhalation of NO2. While there have been a significant number of epidemiology studies conducted over the past 10 years, there are very few new toxicology studies aimed at assessing mechanisms of response to NO2. In any case, comparisons between animal studies, controlled human exposures or epidemiologic studies is difficult, since the assays used in these different types

844

NITROGEN OXIDES

of evaluations are not always directly comparable. One type of response index that has been examined in all of these studies is respiratory mechanics. However, changes in pulmonary function may not be very sensitive to NO2 due to the tendency of such tests to reflect changes in the large airways while the major targets for NO2 are the smaller conducting airways and respiratory region. In any case, there is little evidence that exposure of normal humans or laboratory animals to 1 ppm NO2 affects standard pulmonary mechanics responses. Even exposure to higher levels has resulted in inconsistent results. Airway responsiveness may be increased in normal human subjects, but generally only with exposures at >1 ppm NO2. Epidemiological data suggest that there may be long-term effects of NO2 on pulmonary function in children. Asthmatics may represent a population subgroup showing susceptibility to NO2. However, even among asthmatics, responses were not always consistent or reproducible, and those that have occurred involved increased airway responsiveness rather than changes in standard respiratory function indices. Surprisingly, effects noted in some studies at 1 ppm NO2 have not always been found with higher levels (up to 3 ppm), and this apparent lack of dose–response complicates any evaluation of the health significance of NO2 exposure. While it is possible that differences in the degree of asthma severity in the subjects used in the various studies may have accounted for some of this discrepancy, it does seem that mild asthmatics are not generally sensitive to NO2 concentrations <0.6 ppm when respiratory function and airway responsiveness are examined. The database for pulmonary function effects in COPD patients is similarly conflicting, with some studies showing effects and others none at NO2 exposure levels <2 ppm. Nevertheless, while controlled clinical studies do not unequivocally indicate any enhanced susceptibility to NO2 among mild asthmatics or people with COPD, there is indication that some asthmatics may respond at lower exposure concentrations than do healthy individuals. Various biological responses not generally examined in humans have been assessed in animal models, and these indicate effects due to NO2 that may have potential health significance. This includes NO2-related alterations in various host defense parameters, such as mucociliary clearance, pulmonary macrophage and immunologic function, and susceptibility to respiratory infection. Tracheobronchial mucus transport rates remain unaltered by single exposures up to 10 ppm or short-term repeated exposures to 1 ppm, while respiratory region clearance may be affected by short-term repeated at <1 ppm. Morphological changes in macrophages begin to occur at 0.5 ppm, while functional activity has been affected by short-term repeated exposures to 0.3 ppm. As a likely consequence of altered defenses, animals may be less able to cope with respiratory infection. Nitrogen dioxide levels as low as 0.5 ppm will increase bacterial infectivity if exposure is prolonged. Clear, direct suppressive effects on humoral or cellular immunity have been noted only with exposure to >5 ppm NO2, with lower levels possibly resulting in some depression or activation of immune system components. Limited results from controlled clinical studies suggest that some similar responses may be occurring in humans. Additional evidence for human health effects resulting from NO2 exposure comes from epidemiological examinations of acute respiratory illness, especially since this is supported by toxicological studies on host defense mechanisms and a limited number of controlled clinical studies. Those which provide relatively reliable estimates of NO2, either by direct measure or suitable surrogate, are somewhat suggestive of increased risk or susceptibility to lower and/or upper respiratory tract infection in young children associated with long-term NO2 exposure.

NITRIC OXIDE

845

Animal models have been extensively used in studying effects of NO2 on pulmonary morphology. The target site is consistently the area around the terminal/respiratory bronchiolar junction and associated alveoli, and the cells that are most sensitive are the ciliated cells of the bronchiolar epithelium and the Type 1 cells of the alveolar epithelium. Neonates seem to be more resistant than adults to these morphological effects. Acute exposure to <5 ppm can produce hyperplasia and hypertrophy of bronchiolar and alveolar cells, and proliferation of Type 2 cells. Long-term exposure to 0.3–0.5 ppm can result in similar lesions, although chronic exposures to 2 ppm are generally needed to produce extensive and permanent pulmonary structural changes. Some changes may resolve even with continued exposure. Although the data base does not currently allow for determination of the lowest NO2 level and shortest exposure duration that will produce clear and permanent morphological effects, the concentrations that seem to result in such changes are well above those currently found in most outdoor or indoor environments. The primary target organ for NO2 is the respiratory tract, but there may be some extrapulmonary effects of exposure as well. However, conclusions as to the possible health significance of these cannot, as yet, be made. Furthermore, there is no support for any teratogenic or genotoxic potential for NO2, nor for any direct carcinogenic action. Although NO2 may modulate pulmonary cancer originating elsewhere, the database is weak in this regard as well. Quite a few questions regarding health effects from exposure to NO2 remain unanswered. For example, there is no complete picture of the transition between acute and chronic effects, nor is the extent of reversibility of effects resolved, especially with short-term peak exposures. High concentrations of NO2 (5 ppm) are associated with structural and physiological changes in the respiratory tract. However, the extent to which these may occur at levels more relevant to either outdoor or indoor exposures is not clear. Furthermore, the relationship between biological responses and specific exposure patterns, that is, constant low level versus low baseline plus higher spikes, is also not clear; the latter scenario may be more relevant to indoor exposure, the former to most outdoor exposures. The contribution of differing biochemical mechanisms, that is, acid versus oxidative, in the expression of NO2 toxicity is not fully understood. Observations that the direction of change in various biological end points seems to depend upon exposure concentration may reflect differences in underlying mechanisms of action. Thus, while toxicologic studies may provide indications of possible mechanisms of action leading to adverse health effects, controlled clinical and epidemiologic studies have not as yet resulted in a consistent pattern of responses. The epidemiology studies often cannot separate effects of NO2 from those due to copollutants released by the same source, and effects of NO2 could very well be due to or modulated by one of more of these copollutants. Thus, determination of independent effects of NO2 is difficult. However, what is emerging seems to be an independent effect from NO2 primarily on hospital admissions or emergency room visits for cardiopulmonary disease, or on lung functional development in children.

22.4 NITRIC OXIDE 22.4.1

Exposure

There are much less data on ambient levels of NO than for NO2. Maximum hourly average NO concentrations can range from 0.17 to 1 ppm in metropolitan areas, while annual

846

NITROGEN OXIDES

averages are in the range of 0.01–0.06 ppm (U.S. EPA, 2008). Rural areas show maximum hourly averages of 0.01–0.4 ppm, and annual averages of 0.005–0.009 ppm. Monitoring of various regions in Southern California indicated an 8-year (1994–2001) mean concentration range of 0.001–0.039 ppm (Peters, 2004). However, there is a wide variability in regional NO concentrations. For example, hourly average concentrations in California were found to range from 0 to 0.87 ppm during 1 year (1999), while maximum hourly average concentrations during the same time period were noted to range from 0.50–0.79 ppm in various areas in the Midwest to Northeast (Lazarus, 2001). Indoor concentrations are not commonly measured and, thus, data are limited. While it seems that most of the NO found indoors derives from penetration from outdoor air (Lazarus, 2001), direct indoor emission can result from combustion appliances, such as natural gas fired cooking stoves and gas or kerosene heaters. Indoor concentrations, thus, are strongly influenced by utilization of such devices. Indoor levels that ranged from 0.001 to 0.386 ppm were noted in one home in Southern California during a 1-year period (Weschler et al., 1994). 22.4.2

Dosimetry and Toxicology of NO

The lower aqueous solubility of NO compared to NO2 may result in greater amounts of the former reaching the respiratory region (Yoshida and Kasama, 1987). This NO will then diffuse rapidly, at a somewhat faster rate than would NO2 (Chiodi and Mohler, 1985), through pulmonary tissue with little reaction, but it is not transported to any great extent through the vasculature due to its rapid interaction with oxyhemoglobin, as discussed further below. While the direct toxicity of NO is low, indirect toxicity can result from its reaction with superoxide to produce peroxynitrite, a potent oxidant. NO entering the bloodstream binds to hemoglobin (Hb), producing nitrosylhemoglobin (NOHb). The affinity of Hb for NO is very high, much higher even than that for O2. The NOHb formed is rapidly oxidized to methemoglobin (MetHb) in the presence of O2. The MetHb is subsequently reduced into ferrous Hb by MetHb-reductase, an enzyme present in red blood cells. In spite of the affinity for hemoglobin, in vivo exposures to NO at levels ranging from 2 to 10 ppm have shown that the amount of NOHb in blood was such that any reduction of oxygen transport was not lethal nor damaging to organs sensitive to O2 depletion (e.g., Oda et al., 1980a, 1976; Azoulay et al., 1977). Exposure of humans to 40 ppm for 2 h resulted in a small increase in MetHb in peripheral blood that was considered as not clinically significant (Luhr et al., 1998). Thus, it appears that as long as the activity of MetHb-reductase is maintained, the conversion of NOHb to MetHb should mitigate any potential toxicity due to NO-related oxygen transport effects. However, it should be borne in mind that some groups, especially neonates, have less capacity than do adults to remove MetHb from their circulation and may, thus, show greater effects at lower exposure concentrations. NO is synthesized endogenously in the cells of many tissues from arginine and molecular oxygen via various nitric oxide synthase (NOS) enzymes. Human tissues contain three such enzymes, nNOS in neurons, iNOS in macrophages and eNOS in endothelial cells. Because NO diffuses freely across cell membranes and there are many molecules with which it can react, it is consumed very rapidly near the site of synthesis. Since endogeneous NO is involved in numerous physiological processes, such as nervous system signaling, regulation of pulmonary and systemic vascular resistance, and mediation of immune defenses, the impact of inhaled, exogenous NO, especially at low concentrations, is often difficult to evaluate.

NITRIC OXIDE

847

Nitric oxide can react with thiol-associated iron in enzymes, which is a mechanism for cytotoxicity. It can also react with superoxide, producing peroxynitrite which can then react with proteins (Ischiropoulos et al., 1992). Many of these effects have been noted in vitro and offer potential explanations for effects of NO on host defenses. Whether they can explain any effects of NO inhalation exposure is not clear. There is, however, indication that, at least for some end points, effects of endogeneous NO can be mimicked by exposure to exogeneous NO (Gustafsson, 1993). It has been suggested, for example, that the vasodilatory response of the bronchial and pulmonary vascular systems to cigarette smoke is due to NO in the smoke (Alving et al., 1992). Furthermore, individuals with depressed endogenous NO may be more sensitive to inhaled NO. The specific substrates and reactions that mediate NO toxicity are not clear. Some studies indicate that the toxic effects of NO are different from the membrane damage due to NO2. For example, NO may target fibroblasts that are responsible for the maintenance and repair of the alveolar interstitium (Mercer et al., 1995). Any respiratory tract morphologic responses to NO are similar to those found with NO2, except that NO levels needed to produce them in most studies were higher, that is, 2 ppm with continuous exposure (Azoulay et al., 1977; Oda et al., 1980; Hugod, 1979; Holt et al., 1979; Oda et al., 1976). While little NO appears to react with lung tissue at exposure concentrations found in ambient outdoor or indoor air, with most diffusing into the blood, chronic exposure of rats to 0.5 ppm (with spikes to 1.5 ppm) produced interstitial lung damage (Mercer et al., 1995). While this may suggest that NO is more potent than NO2 for certain types of morphological injury, there was found to be no structural change in the alveoli in rats exposed continuously to 2 or 6 ppm NO for 6 weeks (Mercer, 1999). Studies of physiological effects of inhaled NO are sparse, and exposure levels used were quite high. Murphy (1964) found no change in pulmonary mechanical function of guinea pigs exposed for 4 h to NO at 16 or 50 ppm. Holt et al. (1979) examined immunological end points in mice exposed to 10 ppm NO for 2 h/day, 5 days/week up to 30 weeks. Leukocytosis was evident by 5 weeks of exposure, while a decrease in mean hemoglobin content of red blood cells was found by 30 weeks. The ability of spleen cells to mount a graft versus host reaction was stimulated by 20 weeks of exposure, but suppressed by 26 weeks. When the ability of mice to reject virus-induced tumors was assessed, less of the NO-exposed animals survived tumor challenge compared to control; this suggests that NO, at high levels, may have affected immunologic competence. In this regard, mice were exposed continuously to 2 ppm NO for 6 h up to 4 weeks, to assess the effect on resistance to bacterial infection (Azoulay et al., 1981). There was some indication that NO-exposed females, but not males, showed a significant increase in mortality and a significant decrease in survival time. Exposure to 20 ppm NO for 2 h was noted to reduce the viability and production of superoxide by neutrophils (Daher et al., 1997). Finally, NO does not appear to be genotoxic (G€ orsdorf et al., 1990). There are very few data on controlled exposures of NO in humans. Inhalation of 30 ppm for 0.5 h in normal subjects did not affect platelet function (Albert et al., 1999). On the contrary, exposure for 40 min at up to 40 ppm resulted in increased bleeding time at the highest concentration (Gries et al., 1997). Healthy males inhaling 50 ppm NO for 0.5 h showed no effect on systemic blood pressure or heart rate (Krejcy et al., 1995). Following a 2-h exposure to 1 ppm NO (Kagawa, 1982), there appeared to be some variability in pulmonary function response among subjects, but only one of a large battery of tests showed statistical significance; it is likely that this effect, if not due to chance, has little biological significance. On the contrary, vasomotor tone is a sensitive target for NO, and pulmonary

848

NITROGEN OXIDES

vasodilation has been noted with acute exposure to 5–10 ppm in normal animals and humans (Gustafsson, 1993). Normal adults and those with airway hyperresponsivity, asthma, or COPD were exposed to 80 ppm NO for 10 min with measurements of specific airway conductance and functional residual capacity (Hogman et al., 1993). Normals and those with hyperactive airways showed no effect, while asthmatics actually showed some improvement in conductance with exposure. The interpretation of responses to exogenous NO is complicated by the presence, as noted, of endogenous NO, which can act as a bronchodilator. In summary, a large fraction of inhaled NO reaches the respiratory region of the lungs, where it rapidly diffuses into blood and reacts with hemoglobin; little NO directly interacts with lung tissue, especially at ambient concentrations. In spite of any binding with hemoglobin, anoxia of O2-sensitive organs does not seem to occur, at least with NO exposure levels 10 ppm.

22.5 NITRIC/NITROUS ACID 22.5.1

Exposure

There are few data for ambient HNO3 levels, and those that are available suggest much variability. Levels in various cities in mid- to southern California were found to range from 0.075–5.6 mg/m3 (0.003–0.22 ppm), averaging about 0.07 ppm (Munger et al., 1990; Fischer et al., 2003); concentrations in a relatively urban area in Southern California were noted to range as high as 0.5 ppm. Indoor levels of nitric acid have been reported to range up to 0.001 ppm (Brauer et al., 1991). Indoor levels of nitrous acid can reach 0.1 ppm when gas stoves and unvented kerosene heaters are used (Beckett et al., 1995), while outdoor (short term) levels of 0.007–0.016 ppm have been reported in Southern California (Winer and Biermann, 1991). 22.5.2

Dosimetry and Toxicology

The dosimetry of inhaled HNO3 is unknown. However, because of its very high water solubility and vapor state, inhaled HNO3 should undergo significant removal in the upper respiratory tract. It has also been suggested that inhaled vapor phase HNO3 may be converted into or deposited onto small particles within the humid atmosphere of the respiratory tract, thus facilitating its transport to and deposition within the deep lung (Chen and Schlesinger, 1996). By contrast, HNO3 inhaled in fog droplets would deposit in large airways. The database concerning potential health effects from exposure to HNO3 is limited. In one study, both normal and allergic sheep were exposed for 4 h to 1.6 ppm (42.7 mg/m3) HNO3 vapor (Abraham et al., 1982). A decrease in specific pulmonary flow resistance, compared to preexposure control values, in both groups of sheep was noted. However, allergic sheep showed increased airway responsiveness, both immediately and 24 h after HNO3 exposure. Although there was no significant change in responsiveness in the normal group as a whole, two of the animals showed an increase in responsiveness to bronchoconstrictor challenge (carbachol) after HNO3 exposure. According to the investigators, this suggested that some individuals in a normal population may be more sensitive than are others. Allergic adolescent asthmatic human subjects were exposed for 40 min during rest and moderate exercise to 0.05 ppm (1.3 mg/m3) HNO3. An increase in total respiratory resistance and a decrease in forced expiratory volume were noted (Koenig et al., 1989). In

INORGANIC NITRATES

849

another report, Koenig (1989) examined exercising adolescent asthmatics exposed to 0.057 (1.5 mg/m3) ppm HNO3 vapor for 45 min. Small, but not statistically significant, decreases in forced expiratory volume and expiratory flow rates were found. Particulate HNO3 has also been found to enhance bronchoconstriction in humans produced by exposure to hypoosmolar aerosols (Balmes et al., 1988). Aris et al. (1991) measured pulmonary functional parameters (specific airway resistance, SRaw, FEV1, FVC) and lavage indices (total and differential cell counts, LDH, fibronectin, total protein) in healthy subjects exposed for 4 h (including moderated exercise) to 500 mg/m3 HNO3 vapor. Lavage was performed 18 h postexpsoure. No HNO3-related effects on any of the measured end points were found. Healthy adults exposed to 500 mg/m3 HNO3 for 4 h with exercise showed no change in measures of pulmonary mechanics (CARB, 1996). Heat shock proteins (HSP) have been correlated with environmental stress and pathophysiological conditions. Stress-induced HSP 70 in rat lungs was examined following inhalation exposure 4 h/day, 3 days/week for 40 weeks to 50 mg/m3 HNO3 (Wong et al., 1996). HNO3 was found to elevate lung stress inducible HSP above baseline control levels. Schlesinger et al. (1994) exposed rabbits for 4 h/day, 3 days/week for 4 weeks to HNO3 vapor at 0, 50, 150, and 450 mg/m3. Exposure was followed by assays of biochemical markers in lavage fluid, pulmonary macrophage function, and in vitro bronchial responsivity to smooth muscle constrictor challenge. Nitric acid had no effect either on viability or numbers of cells recovered, or on lactate dehydrogenase or total protein in lavage. All acid concentrations reduced both basal levels and stimulated production of superoxide anion by macrophages, while the release/activity of tumor necrosis factor by stimulated macrophages was reduced following exposure to 150 mg/m3 HNO3. Bronchi from rabbits exposed to 150 mg/m3 HNO3 exhibited reduced smooth muscle responsivity in vitro compared to control. HNO3-induced alterations in both conducting and respiratory airways were also noted by Mautz et al. (1993), who observed changes in breathing pattern, alveolar macrophage receptor binding capacity, and alveolar morphometry in rats exposed to 50, 170, and 470 mg/m3 HNO3 for 4 h/day, 3 days/week for 4 weeks. Further evidence for penetration of HNO3 into the deep lung was provided by Nadziejko et al. (1992), who noted reduced production of superoxide anion by macrophages harvested from rats exposed to 250 mg/m3 for 4 h/day for 4 days. Similar to results of Schlesinger et al. (1994), Nadziejko et al. (1992) found no effects of acid exposure either on total numbers of cells recovered by lavage, differential counts, or in total soluble protein in lavage fluid. Healthy adults exposed, with some exercise, to HNO2 for 3.5 h at 77 and 395 ppb showed a decrease in specific airway conductance compared to air exposure (Rasmussen et al., 1995). Mildly asthmatic adults exposed to 650 ppb HNO2 for 3 h with exercise periods showed a decrease in FVC, which began during the exposure period. Respiratory symptoms indicative of irritation were also associated with the acid exposure (Beckett et al., 1995).

22.6 INORGANIC NITRATES There are limited data on ambient levels of particulate inorganic nitrates (NO3 ). Maximum (24-h average) ambient concentrations are generally well below 10 mg/m3, although certain regions having persistent smog, for example, southern California, may show peaks between 20 and 35 mg/m3 (Pierson and Brachaczek, 1988; Ellestad and Knapp, 1988; Shaw et al., 1982). The annual averages for fine particulate nitrate over an 8-year period (1994–2001) in a

850

NITROGEN OXIDES

number of rural areas in California were noted to range from around 1 to 5 mg/m3 while 8 year averages over a number of rural, suburban, and urban areas ranged from 0.76–11.5 mg/m3 (Peters, 2004). In the United States, nitrates generally account for 1–5% or 8–15% of the total PM2.5 (fine particle) mass in the eastern or western parts of the country, respectively (U.S. EPA, 1993), although in parts of California it may account for up to 40% (Peters, 2004). The toxicologic database supporting any health effects from inhaled nitrates is sparse. Anesthetized dogs exposed to sodium nitrate (NaNO3) at up to 10,000 mg/m3 for 7.5 min showed no effect on pulmonary function end points, while exposure for 4 h to 5000 mg/m3 produced no alterations in pulmonary function, pulmonary or systemic arterial blood pressures, cardiac output or heart rate (Sackner et al., 1979). Conscious sheet similarly exposed for 4 h to 5000 mg/m3 demonstrated no alteration in tracheal mucous velocity (Sackner et al., 1979). Both normal rats and guinea pigs or those with elastase-induced emphysema were exposed to 1000 mg/m3 ammonium nitrate for 6 h/day, 5 days/week for 5 or 20 days; the guinea pigs showed no exposure-related effect on lung volumes or lung compliance while rats showed only minor changes in pulmonary function and there was no additional effect related to the emphysema state (Loscutoff et al., 1985). Furthermore, morphological analysis of those animals exposed for 20 days showed no effect of exposure. Normal mice and those sensitized to ovalbumin were exposed to ammonium nitrate at either 140 mg/m3 (0.58 mm) or 250 mg/m3 (0.22 mm) for 4 h/day for 3 days (Cassee et al., 1998). There were no effects on protein or lactate dehydrogenase levels in lung lavage fluid, but the level of N-acetyl glucosaminidase was increased in the sensitized mice exposed to the smaller particles. Mice in both groups exposed to the larger particles showed increased airway responsiveness, suggesting that the ammonium nitrate did not exacerbate airway responsiveness differentially in atopic animals. In a similar study (Cassee et al., 1999), normal rats and those treated with monocrotaline exposed to ammonium nitrate for 4 h/day for 3 days at 418 mg/m3 (0.087 mm) or 361 mg/m3 (0.643 mm) showed no effect on various enzymes in lavage, but did show an increased number and severity of lesions in the lungs which were determined to be due to a background bacterial infection, suggesting some effect on resistance to infection. Ehrlich (1979) examined the effect of 3-h exposures to various nitrate salts (1290–4500 mg/m3) on resistance to respiratory bacterial infection in mice. Only zinc nitrate (Zn(NO3)2) resulted in any significant mortality increase, the extent of which seemed to be exposure concentration related. However, since the response was similar to that seen with zinc sulfate (ZnSO4), the effect was likely due to the zinc ion (Znþ2) rather than to the NO3 . Charles and Menzel (1975) examined the effects of nitrate upon the release of histamine by guinea pig lung fragments; response to some pollutants may be a function of their ability to elicit biologic mediators. Histamine was released in proportion to the concentration of salt present. However, the response was not totally due to NO3 ; ammonium (NH4þ) ion was also a possible contributor. The relation of this to actual in vivo exposures is, however, not clear. Other in vitro studies suggest that NO3 may affect red blood cells by altering the transport of calcium across the cell membrane (Kunimoto et al., 1984). Some controlled clinical studies have been conducted with NO3 aerosols (Kleinman et al., 1980; Sackner et al., 1979; Stacy et al., 1983; Utell et al., 1979, 1980). Concentrations ranged from 200 to 7000 mg/m3, and pulmonary function was the end point. The only effects noted were decreases in airway conductance and partial-expiratory flow–volume curves in subjects with influenza exposed for 16 min to 7000 mg/m3 of NaNO3 aerosol (Utell et al., 1980). This was not seen in normals or asthmatics (Utell et al., 1979).

REFERENCES

851

The concentrations used in most studies with nitrates are well above ambient levels. The results suggest that there are likely to be no adverse effects, as far as cardiopulmonary function is concerned, from current levels of NO3 aerosols, even in presumably more sensitive asthmatics. However, some potentially sensitive measures of cardiopulmonary function, such as heart rate variability, have not been assessed in controlled studies with nitrate particles, and these measures have been shown to be altered by exposure to various types of ambient particulate matter (Schlesinger et al., 2006).

REFERENCES Abraham WM, Kim CS, King MM, Oliver Jr, W, Yerger L (1982) Effects of nitric acid on carbachol reactivity of the airways in normal and allergic sheep. Arch. Environ. Health 37:36–40. Adgate JL, Reid HF, Morris R, Helms RW, Berg RA, Hu P-C, et al.(1992) Nitrogen dioxide exposure and urinary excretion of hydroxyproline and desmosine. Arch. Environ. Health 47:376–384. Adkins Jr, B, Van Stee EW, Simmons JE, Eustis SL (1986) Oncogenic response of strain A/J mice to inhaled chemicals. J. Toxicol. Environ. Health 17:311–322. Ahmed T, Marchette B, Danta I, Birch S, Dougherty RL, Schreck R, Sackner MA (1982) Effect of 0.1 ppm NO2 on bronchial reactivity in normals and subjects with bronchial asthma. Am. Rev. Respir. Dis. 125 (Suppl.):152S. Albert J, Wallen NH, Li N, Frostell C (1999) Neither endogenous nor inhaled nitric oxide influences the function of circulating platelets in healthy volunteers. Clin. Sci. 97:345–353. Alving K, Fornhem C, Weitzbergm E, Lundberg JM (1992) Nitric oxide mediates cigarette smokeinduced vasocilatory responses in the lung. Acta Physiol. Scand. 146:407–408. Amoruso MA, Witz G, Goldstein BD (1981) Decreased superoxide anion radical production by rat alveolar macrophages following inhalation of ozone or nitrogen dioxide. Life Sci. 28: 2215–2221. Anderson HR, Ponce de Leon A, Bland JM, Bower JS, Strachan DP (1996) Air pollution and daily mortality in London: 1987-92. BMJ 312:665–669. Anderson HR, Ponce de Leon A, Bland JM, Bower JS, Emberlin J, Strachan DP (1998) Air pollution, pollens, and daily admissions for asthma in London 1987-92. Thorax 53:842–848. Aris RD, Christian D, Sheppard D, Balmes JR (1991) The effects of sequential exposure to acidic fog and ozone on pulmonary function in exercising subjects. Am. Rev. Resp. Dis. 143:85–91. Avol EL, Linn WS, Peng RC, Valencia G, Little D, Hackney JD (1988) Laboratory study of asthmatic volunteers exposed to nitrogen dioxide and to ambient air pollution. Am. Ind. Hyg. Assoc. J. 49:143–149. Ayaz KL, Csallany AS (1978) Long-term NO2 exposure of mice in the presence and absence of vitamin E: II. Effect of glutathione peroxidase. Arch. Environ. Health 33:292–296. Ayyagari VN, Januszkiewicz A, Nath J (2004) Pro-inflammatory responses of human bronchial epithelial cells to acute nitrogen dioxide exposure. Toxicology 197:149–164. Azoulay E, Bouley G, Blayo MC (1981) Effects of nitric oxide on resistance to bacterial infection in mice. J. Toxicol. Environ. Health 7:873–882. Azoulay E, Soler P, Blayo MC, Basset F (1977) Nitric oxide effects on lung structure and blood oxygen affinity in rats. Bull. Eur. Physiopathol. Respir. 13:629–644. Azoulay-Dupuis E, Torres M, Soler P, Moreau J (1983) Pulmonary NO2 toxicity in neonate and adult guinea pigs and rats. Environ. Res. 30:322–339. Balchum OJ, Buckley RD, Sherwin R, Gardner M (1965) Nitrogen dioxide inhalation and lung antibodies Arch. Environ. Health 10:274–277.

852

NITROGEN OXIDES

Balmes JR, Fine JM, Christian D, Gordon T, Sheppard D (1988) Acidity potentiates bronchoconstriction induced by hypoosmolar aerosols Am. Rev. Respir. Dis. 138:35–39. Barck C, Sandstrom T, Lundahl J, Hallden G, Svartengren M, Strand V, Rak S, Bylin G (2002) Ambient level of NO2 augments the inflammatory response to inhaled allergen in asthmatics. Respir. Med. 96:907–917. Barnett AG, Williams GM, Schwartz J, Neller AH, Best TL, Petroeschevsky AL, Simpson RW (2005) Air pollution and child respiratory health: A case-crossover study in Australia and New Zealand. Am. J. Respir. Crit. Care Med. 171:1272–1278. Barth PJ, Muller B, Wagner U, Bittinger A (1995) Quantitative analysis of parenchymal and vascular alterations in NO2-induced lung injury in rats Eur. Respir. J. 8:1115–1121. Bascom R, Bromberg PA, Costa DL, et al.(1996) Health effects of outdoor air pollution. Am. J. Respir. Crit. Care Med. 153:477–498. Bauer MA, Utell MJ, Morrow PE, Speers DM, Gibb FR (1985) Route of inhalation influences airway responses to 0.3 ppm nitrogen dioxide in asthmatic subjects. Am. Rev. Respir. Dis. 131: A171. Bauer MA, Utell MJ, Morrow PE, Speers DM, Gibb FR (1986) Inhalation of 0.30 ppm nitrogen dioxide potentiates exercise-induced bronchospasm in asthmatics. Am. Rev. Respir. Dis. 134:1203–1208. Beckett WS, Russi MB, Haber AD, Rivkin RM, Sullivan JR, Tameroglu Z, Mohsenin V, Leaderer BR (1995) Effect of nitrous acid on lung function in asthmatics: A chamber study. Environ. Health Perspect. 103:372–375. Beil M, Ulmer WT (1976) Wirkung von NO2 in MAK-Bereich auf Atemmechanik und bronchiale Acetylcholinempfindlichkeit bei Normalpersonen [Effect of NO2 in workroom concentrations on respiratory mechanics and bronchial susceptibility to acetylcholine in normal persons]. Int. Arch. Occup. Environ. Health 38:31–44. Berglund M (1993) Exposure. Scand. J. Work Environ. Health 19:14–20. Berwick M, Zagraniski RT, Leaderer BP, Stolwijk JA (1984) Respiratory illness in children exposed to unvented combustion sources. In:Berglund B, Lindvall T, Sundell J, editors. Indoor Air. Volume 2. Stockholm:Swedish Council for Building Research. pp. 255–260. Bils RF (1976) The connective tissues and alveolar walls in the lungs of normal and oxidant-exposed squirrel monkeys. J. Cell Biol. 70:318. Brauer M, Koutrakis P, Keeler GJ, Spengler JD (1991) Indoor and outdoor concentrations of inorganic acidic aerosols and gases. J. Air Waste Manage. Assoc. 41:171–181. Braun-Fahrlander C, Ackermann-Liebrich U, Schwartz J, Gnehm HP, Rutishauser M, Wanner HU (1992) Air pollution and respiratory symptoms in preschool children. Am. Rev. Respir. Dis. 145:42–47. Braun-Fahrlander C, Vuille JC, Sennhause FH , et al. (1997) Respiratory health and long-term exposure to air pollutants in Swiss schoolchildren. SCARPOL Team. Swiss Study on Childhood Allergy and Respiratory Symptoms with Respect to Air Pollution, Climate and Pollen. Am. J. Respir. Crit. Care Med. 155:1042–1049. Bromberg PA, (1988) Asthma and automotive emissions. In:Watson AY, Bates RR, Kennedy D, editors. Air Pollution, the Automobile and Public Health. Washington, DC:National Academy Press. pp. 465–498. Brunekreef B, Fischer P, Houthuijs D, Remijn B, Boleij J (1987) Health effects of indoor NO2 pollution. In:Seifert B, Esdorn H, Fischer M, Rueden H, Wegner J, editors. Indoor Air ’87. Volume 1. Berlin, FRG:Institute for Water, Soil and Air Hygiene. pp. 304–308. Burnett RT, Cakmak S, Brook JR, Krewski D (1997) The role of particulate size and chemistry in the association between summertime ambient air pollution and hospitalization for cardiorespiratory diseases. Environ. Health Perspect. 105:614.

REFERENCES

853

Busey WM, Coate WB, Badger DW (1974) Histopatholic effects of nitrogen dioxide exposure and heat stress in cynomolgus monkeys. Toxicol. Appl. Pharmacol. 29:130. Bylin G, Hedenstierna G, Lindvall T, Sundin B (1988) Ambient nitrogen dioxide concentrations increase bronchial responsiveness in subjects with mild asthma. Eur. Respir. J. 1:606–612. Bylin G, Lindvall T, Rehn T, Sundin B (1985) Effects of short-term exposure to ambient nitrogen dioxide concentrations on human bronchial reactivity and lung function. Eur. J. Respir. Dis. 66:205–217. Campbell KI, Hilsenroth RH (1976) Impaired resistance to toxin in toxoid-immunized mice exposed to ozone or nitrogen dioxide. Clin. Toxicol. 9:943–954. CARB (1996) California Air Resources Board. Does ambient nitric acid compromise lung function. Research Notes No. 96-6. Sacramento, CA:CARB. CARB (2004) California Air Resources Board. The 2004 California Almanac of Emissions and Air Quality. Appendix B. Air Quality Trend Data by Pollutant. Sacramento CA:CARB. Case GD, Dixon JS, Schooley JC (1979) Interactions of blood metalloproteins with nitrogen oxides and oxidant air pollutants. Environ. Res. 20:43–65. Cassee FR, Boere AJF, Fokkens PHB, Dormans JAMA, Van Bree L, Rombout PJA (1999) Toxicity of ambient particulate matter. IV. Acute toxicity in pulmonary hypertensive rats following 3-day exposure to ultrafine and fine model components for the secondary aerosol fraction of PM10. Report No. 650010016. Bilthoven, The Netherlands:RIVM. Cassee FR, Dormans JAMA, Van Loveren H, Van Bree L, Rombout PJA (1998) Toxicity of ambient particulate matter. I. Ultrafine and fine ammonium bisulfate. Report No. 650010010. Bilthoven, The Netherlands:RIVM. Cavanagh DG, Morris JB (1987) Mucus protection and airway peroxidation following nitrogen dioxide exposure in the rat. J. Toxicol. Environ. Health 22:313–328. Chaney S, Blomquist W, DeWitt P, Muller K (1981) Biochemical changes in humans upon exposure to nitrogen dioxide while at rest. Arch. Environ. Health 36:53–58. Chang L-Y, Graham JA, Miller FJ, Ospital JJ, Crapo JD (1986) Effects of subchronic inhalation of low concentrations of nitrogen dioxide I. The proximal alveolar region of juvenile and adult rats. Toxicol. Appl. Pharmacol. 83:46–61. Charles JM, Menzel DB (1975) Ammonium and sulfate on release of histamine from lung fragments. Arch. Environ. Health 30:314–316. Chen LC, Schlesinger RB (1996) Considerations for the respiratory tract dosimetry of inhaled nitric acid vapor. Inhal. Toxicol. 8:639–654. Chiodi H, Mohler JG (1985) Effects of exposure of blood hemoblogin to nitric oxide. Environ. Res. 37:355–363. Chiodi H, Collier CR, Mohler JG (1983) In vitro methemoglobin formation in human blood exposed to NO2 Environ. Res. 30:9–15. Chow CK, Dillard CJ, Tappel AL (1974) Glutathione peroxidase system ad lysozyme in rats exposed to ozone or nitrogen dioxide. Environ. Res. 7:311–319. Comstock GW, Meger MB, Helsing KJ, Tockman MJ (1981) Respiratory effects of household exposures to tobacco smoke and gas cooking. Am. Rev. Respir. Dis. 124:143–148. Crapo JD, Barry BE, Chang L-Y, Mercer RR (1984) Alterations in lung structure caused by inhalation of oxidants. J. Toxicol. Environ. Health 13:301–321. Csallany AS (1975) The effect of nitrogen dioxide on the growth of vitamin E deficient, vitamin E supplemented and DPPD supplemented mice. Fed. Proc. Fed. Am. Soc. Exp. Biol. 34:913. Daher AH, Fortenberry JD, Owens ML, Brown LA (1997) Effects of exogenous nitric oxide on neutrophil oxidative function and viability. Am. J. Respir. Cell Mol. Biol. 16:407–412.

854

NITROGEN OXIDES

Dalhamn T, Sjoholm J (1963) Studies on SO2 NO2 and NH3: Effect on ciliary activity in rabbit trachea of single in vitro exposure and resorption in rabbit nasal cavity. Acta Physiol. Scand. 58:287–291. DeNicola DB, Rebar AH, Henderson RF (1981) Early damage indicators in the lung V. Biochemical and cytological response to NO2 inhalation. Toxicol. Appl. Pharmacol. 60:301–312. Dijkstra L, Houthuijs D, Brunekreef B, Akkerman I, Boleij JSM (1990) Respiratory health effects of the indoor environment in a population of Dutch children. Am. Rev. Respir. Dis. 142: 1172–1178. D’Ippoliti D, Forastiere F, Ancona C, Agabiti N, Fusco D, Michelozzi P, Perucci CA (2003) Air pollution and myocardial infarction in Rome: a case-crossover analysis. Epidemiology 14:528–535. Dockery DW, Speizer FE, Stram DO, Ware JH, Spengler JD, Ferris Jr, BG (1989) Effects of inhalable particles on respiratory health of children. Am. Rev. Respir. Dis. 139:587–594. Dockery DW, Pope AC, Xu X , et al.(1993) An association between air pollution and mortality in six U. S. Cities. N. Engl. J. Med. 329:1753–1759. Dodge R (1982) The effects of indoor pollution on Arizona children. Arch. Environ. Health 37:151– 155. Dominici F, McDermott A, Daniels M, Zeger SL, Samet JM (2003) Mortality among residents of 90 cities. Health Effects Institute. Revised Analysis of Time-Series Studies of Air Pollution and Health. Special Report. Montpelier, VT:Capital Cities Press. Dubowski Y, Sumner AL, Menke EJ (2004) Interactions of gaseous nitric nitric acid with surfaces of environmental interest. Phys. Chem. Chem. Phys. 6:3879–3888. Ehrlich R (1979) Interaction between environmental pollutants and respiratory infections. In:Gardner DE, Hu EPC, Graham JA, editors. Proceedings of the Symposium on Experimental Models for Pulmonary Research. EPA-600/9-79-022, USEPA, Research Triangle Park NC. pp. 145– 163. Ehrlich R (1980) Interaction between environmental pollutants and respiratory infections. Environ. Health Perspect. 35:89–100. Ehrlich R (1983) Changes in susceptibility to respiratory infection caused by exposures to photochemical oxidant pollutants. In:Lee SD, Mustafa MG, Mehlman MA, editors. International Symposium on the Biomedical Effects of Ozone and Related Photochemical Oxidants. Princeton, NJ: Princeton Scientific. pp. 273–285. Ehrlich R, Henry MC (1968) Chronic toxicity of nitrogen dioxide: I. Effect on resistance to bacterial pneumonia. Arch. Environ. Health 17:860–865. Ehrlich R, Silverstein E, Maigetter R, Fenters JD, Gardner D (1975) Immunologic response in vaccinated mice during long-term exposure to nitrogen dioxide. Environ. Res. 10:217–223. Ehrlich R, Findlay J, Gardner DE (1979) Effects of repeated exposures to peak concentrations of nitrogen dioxide and ozone on resistance to streptococcal pneumonia. J. Toxicol. Environ. Health 5:631–642. Ekwo EE, Weinberger MW, Lachenbruch PA, Huntley WH (1983) Relationship of parental smoking and gas cooking to respiratory disease in children. Chest 84:662–668. Ellestad TG, Knapp KT (1988) The sampling of reactive atmospheric species by transition-flow reactor. Application to nitrogen species. Atmos. Environ. 22:1595–1600. Elsayed NM, Mustafa MG (1982) Dietary antioxidants and the biochemical response to oxidant inhalation I. Influence of dietary vitamin E on the biochemical effects of nitrogen dioxide exposure in rat lung. Toxicol. Appl. Pharmacol. 66:319–328. Fenters JD, Findlay JD, Port CD, Ehrlich R, Coffin DL (1973) Chronic exposure to nitrogen dioxide: Immunologic, physiologic, and pathologic effects in virus-challenged squirrel monkeys. Arch. Environ. Health 27:85–89.

REFERENCES

855

Fenters JD, Ehrlich R, Findlay J, Spangler J, Tolkacz V (1971) Serologic response in squirrel monkeys exposed to nitrogen dioxide and influenza virus. Am. Rev. Respir. Dis. 104:448– 451. Ferin J, Leach LJ (1977) The effects of selected air pollutants on clearance of titanic oxide particles from the lungs of rats. In:Walton WH, editor. Inhaled Particles IV. New York:Pergamon Press. pp. 333–341. Filleul L, Rondeau V, Vandentorren S, LeMoual N, Cantagrel A. Annesi-Maesano I, Charpin D, Declercq C, Neudirch F, Paris C, Vervloet D, Brochard P, Tessier JF, Kauffmann F, Baldi I (2005) Twenty five year mortality and air pollution: results from the French PAARC survey. Occup. Environ. Med. 62:453–460. Fischer ML, Littlejohn D Lunden MM, Brown NJ (2003) Automated measurements of ammonia and nitric acid in indoor and outdoor air. Environ. Sci. Technol. 37:2114–2119. Folinsbee LJ (1992) Does nitrogen dioxide exposure increase airways responsiveness? Toxicol Ind. Health 8:189–194. Folinsbee LJ, Horvath SM, Bedi JF, Delehunt JC (1978) Effect of 0.62 pm NO2 on cardiopulmonary function in young male nonsmokers. Environ. Res. 15:199–205. Foster JR, Cottrell RC, Herod IA, Atkinson HAC, Miller K (1985) A comparative study of the pulmonary effects of NO2 in the rat and hamster. Br. J. Exp. Pathol. 66:193–204. Frampton MW, Gibb FR, Speers DM, Morrow PE, Utell MJ (1989a) Effects of NO2 exposure on pulmonary function and airway reactivity. Am. Rev. Respir. Dis. 139:A124. Frampton MW, Smeglin AM, Roberts Jr, NJ, Finkelstein JN, Morrow PE, Utell MJ (1989b) Nitrogen dioxide exposure in vivo and human alveolar macrophage inactivation of influenza virus in vitro. Environ. Res. 48:179–192. Frampton MW, Finkelstein JN, Roberts NJ, Smeglin AM, Morrow PE, Utell M (1989c) Effects of nitrogen dioxide exposure on bronchoalveolar lavage proteins in humans. Am. J. Respir. Cell Mol. Biol. 1:499–505. Freeman G, Furiosi NJ, Haydon GB (1966) Effects of continuous exposure of 0.8 ppm NO2 on respiration of rats. Arch. Environ. Health 13:454–456. Freeman G, et al. (1972) Covert reduction in ventilatory surface in rats during prolonged exposure to subacute nitrogen dioxide. Am. Rev. Respir. Dis. 106:563–579. Fujimaki H, (1989) Impairment of humoral immune responses in mice exposed to nitrogen dioxide and ozone mixtures. Environ. Res. 48:211–217. Fujimaki H, Shimizu F (1981) Effects of acute exposure to nitrogen dioxide on primary antibody response Arch. Environ. Health 36:114–119. Fujimaki H, Shimizu F, Kubota K (1982) Effect of subacute exposure to NO2 on lymphocytes required for antibody response. Environ. Res. 29:280–286. Gardner DE, Miller FJ, Blommer EJ, Coffin DL (1979) Influence of exposure mode on the toxicity of NO2. Environ. Health Perspect. 30:23–29. Gardner DE, Miller FJ, Illing JW, Graham JA (1982) Non-respiratory function of the lungs: Host defenses against infection. In:Schneider T, Grant L, editors. Air Pollution by Nitrogen Oxides, Proceedings of the US-Dutch International Symposium. 24–28 May. Maastricht, The Netherlands: Elsevier. pp. 401-415, Garn H, Siese A, Stumpf S, Barth PJ, Muller B, Gemsa D (2003) Shift toward an alternatively activated macrophage response in lungs of NO2-exposed rats. Am. J. Respir. Cell Mol. Biol. 28:386–396. Gauderman WJ, McConnell R, Gilliland F, London S, Thomas D, Avol E, Vora H, Berhane K, Rappaport EB, Lurmann F, Margolis HG, Peters J (2000) Association between air pollution and lung function growth in southern California children. Am. J. Respir. Crit. Care Med. 162:1383–1390.

856

NITROGEN OXIDES

Gauderman WJ, Gilliland F, Vora H, Avol E, Stram D, McConnell R, Thomas D, Lurmann F, Margolis HG, Rappaport EB, Berhane K, Peters J (2002) Association between air pollution and lung function growth in southern California children Results from a second cohort. Am. J. Respir. Crit. Care Med. 166:76–84. Gilmour MI (1995) Interaction of air pollutants and pulmonary allergic responses in experimental animals. Toxicology 105:335–342. Giordano Jr, AM, Morrow PE (1972) Chronic low-level nitrogen dioxide exposure and mucociliary clearance. Arch. Environ. Health 25:443–449. Gnehm HE, Ackerman U, Braun C, Rutishauser M, Wanner HU (1988) Significant association of respiratory symptoms in small children with outdoor NO2 air pollution. Pediatr. Res. 23:291A. Goings SAJ, Kulle TJ, Bascom R, Sauder LR, Green DJ, Hebel JR, Clements ML (1989) Effect of nitrogen dioxide exposure on susceptibility to influenza A virus infection in healthy adults. Am. Rev. Respir. Dis. 139:1075–1081. Goldstein E, Goldstein F, Peek NF, Parks NJ (1980) Absorption and transport of nitrogen oxides. In: Lee SD, editor. Nitrogen Oxides and Their Effects on Health, Ann Arbor:Ann Arbor Sciences. pp. 143–160. Goldstein E, Peek NF, Parks NJ, Hines HH, Steffey EP, Tarkington B (1977) Fate and distribution of inhaled nitrogen dioxide in rhesus monkeys. Am. Rev. Respir. Dis. 115:403–412. Goldstein IF, Lieber K, Andrews LR, Kazembe F, Foutrakis G, Huang P, Hayes C (1988) Acute respiratory effects of short-term exposures to nitrogen dioxide. Arch. Environ. Health 43:138–142. Gooch PC, Luippold HE, Creasia DA, Brewen JG (1977) Observations on mouse chromosomes following nitrogen dioxide inhalation. Mutat. Res. 48:117–120. G€ orsdorf S, Appel KE, Engeholm C, Obe G (1990) Nitrogen dioxide induces DNA single-strand breaks in cultured Chinese hamster cells. Carcinogenesis 11:37–41. Graham JA, Menzel DB, Miller FJ, Illing JW, Gardner DE (1982) Effect of ozone on drug-induced sleeping time in mice pretreated with mixed-function oxidase inducers and inhibitors. Toxicol. Appl. Pharmacol. 62:489–497. Graham JA, Gardner DE, Blommer EJ, House DE, Menache MG, Miller FJ (1987) Influence of exposure patterns of nitrogen dioxide and modifications by ozone on susceptibility to bacterial infectious disease in mice. J. Toxicol. Environ. Health 21:113–125. Greene ND, Schneider SL (1978) Effects of NO2 on the response of baboon alveolar macrophages to migration inhibitory factor. J. Toxicol. Environ. Health 4:869–880. Gregory RE, Pickrell JA, Hahn FF, Hobbs CH (1983) Pulmonary effects of intermittent subacute exposure to low-level nitrogen dioxide. J. Toxicol. Environ. Health 11:405–414. Gries A, Bottiger BW, Dorsam J, Bauer H, Weimann J, Bode C, Martin E, Motsch J (1997) Inhaled nitric oxide inhibits platelet aggregation after pulmonary embolism in pigs. Anaesthesia 86:387– 393. Gross P, DeTrevill RTP, Babyak MA, Kaschak M, Tolker EB (1968) Experimental emphysema: effect of chronic nitrogen dioxide exposure and papain on normal and pneumoconiotic lungs. Arch. Environ. Health 16:51–58. Gustafsson LE (1993) Experimental studies on nitric oxide. Scand. J. Work Environ. Health 19: 44–49. Hackney JD, Thiede FC, Linn WS, Pedersen EE, Spier CE, Law DC, Fischer DA (1978) Experimental studies on human health effects of air pollutants IV. Short-term physiological and clinical effects of nitrogen dioxide exposure. Arch. Environ. Health 33:176–181. Hahn P (1979) Nutrition and metabolic development in mammals. In:Winick M, editor. Nutrition: Pre- and Postnatal Development. New York:Plenum Press. pp. 1–40.

REFERENCES

857

Hajat S, Haines A, Goubet SA, Atkinson RW, Anderson HR (1999) Association of air pollution with daily GP consultations for asthma and other lower respiratory conditions in London. Thorax 54:597–605. Hasselblad V, Eddy DM, Kotchmar DJ (1992) Synthesis of environmental evidence: nitrogen dioxide epidemiology studies. J. Air Waste Manage. Assoc. 42:662–671. Haydon GB, Davidson JT, Lillington GA, Wasserman K (1967) Nitrogen dioxide-induced emphysema in rabbits. Am. Rev. Respir. Dis. 95:797–805. Hazuch MJ, Ginsberg JF, McDonnell WF, Haak Jr, ED, Pimmel RL, House DE, Bromberg PA (1982) Changes in bronchial reactivity of asthmatics and normals following exposures to 0.1 ppm NO2. In:Schneider T, Grant L, editors. Air Pollution by Nitrogen Oxides. Amsterdam:Elsevier. pp. 387–400. Hazucha MJ, Ginsberg JF, McDonnell WF, Haak Jr, ED, Pimmel RL, Salaam SA, House DE, Bromberg PA (1983) Effects of 0.1 ppm nitrogen dioxide on airways of normal and asthmatic subjects. J. Appl. Physiol. 54:730–739. Helleday R, Huberman D, Blomberg A, Stjernberg N, Sandstron T (1995) Nitrogen dioxide exposure impairs the frequency of the mucociliary activity in healthy subjects. Eur. Respir. J. 8:1664–1668. Hoek G, Brunekreff B, Goldbohm S, Fischer P, Van den Brandt PA (2002) Association between mortality and indicators of traffic-related air pollution in the Netherlands: a cohort study. Lancet 360:1203–1209. Hogman M, Frostell CG, Hedenstrom H, Hedenstierna G (1993) Inhalation of nitric oxide modulates adult human bronchial tone. Am. Rev. Respir. Dis. 148:1474–1478. Holt PG, Finlay-Jones LM, Keast D, Papadimitrou JJ (1979) Immunological function in mice chronically exposed to nitrogen oxides (NOx). Environ. Res. 19:154–162. Hood DB, Gettins P, Johnson DA (1993) Nitrogen dioxide reactivity with proteins: Effects on activity and immunoreactivity with alpha-1-proteinase inhibitor and implications for NO2-mediated peptide degradation. Arch. Biochem. Biophys. 304:17–26. Hoshi H, Yamauchi K, Sekizawa K, Ohkawara Y, Iijima H, Sakurai E, Maeda K, Okinaga S, Ohno I, Honma M, Tamura G, Tanno Y, Watanabe T, Sasaki H, Shirato KK (1996) Nitrogen dioxide exposure increases airway contractile response to histamine by decreasing histamine N-methyltransferase activity in guinea pigs. Am. J. Respir. Cell Mol. Biol. 14:76–83. Houthuijs D, Remijn B, Brunekreef B, de Koning R (1987) Exposure to nitrogen dioxide and tobacco smoke and respiratory health of children. In:Seifert B, Esdorn H, Fischer M, Rueden H, Wegner J, editors. Indoor Air ’87. Volume 1. Berlin, Germany:Institute for Water, Soil and Air Hygiene. pp. 463–467. Hugod C (1979) Ultrastructural changes of the rabbit lung after a 5 ppm nitric oxide exposure. Arch. Environ. Health 34:12–17. Hyde D, Orthoefer J, Dungworth D, Tyler W, Carter R, Lum H (1978) Morphometric and morphologic evaluation of pulmonary lesions in beagle dogs chronically exposed to high ambient levels of air pollutants Lab. Invest. 38:455–469. Ichinose T, Sagai M (1982) Studies on biochemical effects of nitrogen dioxide III. Changes of the antioxidative protective systems in rat lungs and of lipid peroxidation by chronic exposure. Toxicol. Appl. Pharmacol. 66:1–8. Illing JW, Miller FJ, Gardner DE (1980) Decreased resistance to infection in exercised mice exposed to NO2 and O3. J. Toxicol. Environ. Health 6:843–851. Ischiropoulos H, Zhu L, Beckman JS (1992) Peroxy-nitrite formation from macrophage-derived nitric oxide. Arch. Biochem. Biophys. 298:446–451. Ito K (1971) Effect of nitrogen dioxide inhalation on influenza virus infection in mice. Nippon Eiseigaku Zasshi 26:304–314.

858

NITROGEN OXIDES

Jacob DS, Waldman JW, Munger JW, Hoffman MR (1985) Chemical composition of fogwater collected along the California coast. Environ. Sci. Technol. 19:730–736. Johnson DA, Frampton MW, Winters RS, Morrow PE, Utell MJ (1990) Inhalation of nitrogen dioxide fails to reduce the activity of human lung alpha-1-proteinase inhibitor. Am. Rev. Respir. Dis. 142:758–762. Jorres R, Nowak D, Grimminger F, Seeger W, Oldigs M, Magnussen H (1995) The effect of 1 ppm nitrogen dioxide on bronchoalveolar lavage cells and inflammatory mediators in normal and asthmatic subjects Eur. Respir. J. 8:416–424. Just J, Segala C, Sahraoui F, Priol G, Grimfeld A, Neukirch F (2002) Short-term health effects of particulate and photochemical air pollution in asthmatic children. Eur. Respir. J. 20:899– 906. Kagawa J (1982) Respiratory effects of 2-hr exposure to 1.0 ppm nitric oxide in normal subjects. Environ. Res. 27:485–490. Katsouyanni K, Touloumi G, Samoli E, Gryparis A, LeTerte A, Monopolis Y, Rossi G, Zmirou D, Ballester F, Boumghar A, Anderson HR, Wojtyniak B, Paldy A, Baunstein R, Pekkanen J, Schindler C, Schwartz J (2001) Confounding and effect modification in the short-term effects of ambient particles on total mortality: results from 29 European cities within the APHEA2 project. Epidemiology 12:521–531. Kaya K, Miura T (1982) Effects of nitrogen dioxide on fatty acid compositions of red cell membranes, sera, and livers in rats. Environ. Res. 27:24–35. Kaya K, Miura T, Kubota K (1980) Effects of nitrogen dioxide on red blood cells of rats: Changes in components of red cell membranes during in vivo exposure to NO2. Environ. Res. 23:397–409. Keller MD, Lanese RR, Mitchell RI, Cote RW (1979) Respiratory illness in households using gas and electricity for cooking. I. Survey of incidence. Environ. Res. 19:495–503. Kelly FJ, Blomberg A, Frew A, Holgate ST, Sandstron T (1996) Antioxidant kinetics in lung lavage fluid following exposure of humans to nitrogen dioxide. Am. J. Respir. Crit. Care Med. 154 (6 Pt 1):1700–1705. Kerr HD, Kulle TJ, McIlhany ML, Swidersky P (1979) Effects of nitrogen dioxide on pulmonary function in human subjects: An environmental chamber study. Environ. Res. 19:392–404. Kienast K, Knorst M, Lubjuhn S, Muller-Quernheim J, Ferlinz R (1994) Nitrogen dioxode-induced reactive oxygen intermediates production by human alveolar macrophages and peripheral blood mononuclear cells. Arch. Environ. Health 49:246–250. Kleinerman J, Marchevsky AM, Thorton J (1981) Quantitative studies of APUD cells in airways of rats: The effects of diethylnitrosamine and NO2. Am. Rev. Respir. Dis. 124:458–462. Kleinerman J, Ip MPC (1979) Effects of nitrogen dioxide on elastin and collagen contents of lung Arch. Environ. Health 34:228–232. Kleinerman J, Ip MPC, Sorensen J (1982) Nitrogen dioxide exposure and alveolar macrophage elastase in hamsters. Am. Rev. Respir. Dis. 125:203–207. Kleinerman J, Gordon RE, Ip MPC, Collins A (1985) Structure and function of airways in experimental chronic nitrogen dioxide exposure. In:Gammage RB, Kaye SV, editors. Indoor Air and Human Health. Chelsea, MI:Lewis Publishers Inc. pp. 297–301. Kleinman MT, Phalen RF, Mautz WJ, Mannix RC, McClure TR, Crocker TT (1989) Health effects of acid aerosols formed by atmospheric mixtures. Environ. Health Perspect. 79:137– 145. Kleinman MT, Bailey RM, Linn WS, Anderson KR, Whynot JD, Shamoo DA, Hackney JD (1983) Effects of 0.2 ppm nitrogen dioxide on pulmonary function and response to bronchoprovocation in asthmatics. J. Toxicol. Environ. Health 12:815–826.

REFERENCES

859

Kleinman MT, Linn WS, Bailey RM, Jones MP, Hackney JD (1980) Effect of ammonium nitrate aerosol on human respiratory function and symptoms. Environ. Res. 21:317–326. Kobayashi T, Miura T (1995) Concentration- and time-dependent increase in specific airway resistance after induction of airway hyperresponsiveness by subchronic exposure of guinea pigs to nitrogen dioxide. Fundam. Appl. Toxicol. 25:154–158. Koenig JQ (1989) Paper No. 89-92.4. Presented at 82nd Annual Meeting of Air and Waste Management Association. Anaheim, CA, June. Koenig JQ, Covert DS, Pierson WE (1989) Effects of inhalation of acidic compounds on pulmonary function in allergic adolescent subjects. Environ. Health Perspect. 79:173–178. Koenig JQ, Covert DS, Morgan MS, Horike M, Horike N, Marshall SG, Pierson WE (1985) Acute effects of 0.12 ppm ozone or 0.12 ppm nitrogen dioxide on pulmonary function in healthy and asthmatic adolescents. Am. Rev. Respir. Dis. 132:648–651. Koenig JQ, Covert DS, Marshall SG, Van Belle G, Pierson WE (1987) The effects of ozone and nitrogen dioxide on pulmonary function in healthy and asthmatic adolescents. Am. Rev. Respir. Dis. 136:1152–1157. Koo LC, Ho JHC, Ho C-Y, Matsuki H, Shimizu H, Mori T, Tominaga S (1990) Personal exposure to nitrogen dioxide and its association with respiratory illness in Hong Kong. Am. Rev. Respir. Dis. 141:1119–1126. Kosaka H, Imaizumi K, Imai K, Tyuma I (1979) Stoichiometry of the reaction of oxyhemoglobin with nitrate. Biochim. Biophys. Acta 581:184–188. Kosmider S, Misiewicz A, Felus E, Drozdz M, Ludyga K (1973) Experimentelle und klinische Untersuchungen ueber den Einfluss der Stickstoffoxyde auf die Immunitaet [Experimental and clinical studies on the effects of nitrogen oxides on immunity]. Int. Arch. Arbeitsmed. 31:9–23. Krejcy K, Schmetterer L, Kastner J, Nieszpaur-Los M, Monitzer B, Schutz W, Eichler H-G, Kyrle PA (1995) Role of nitric oxide in hemostatic system activation in vivo in humans. Artriosclerosis Thrombosis Vasc. Biol. 15:2063–2067. Kripke BJ, Sherwin RP (1984) Nitrogen dioxide exposure–influence on rat testes. Anesth. Analg. 63:526–528. Kubota K, Murakami M, Takenaka S, Kawai K, Kyono H (1987) Effects of long-term nitrogen dioxide exposure on rat lung: Morphological observations. Environ. Health Perspect. 73:157– 169. Kulle TJ (1982) Effects of nitrogen dioxide on pulmonary function in normal healthy humans and subjects with asthma and bronchitis. In:Schneider T, Grant L, Air Pollution by Nitrogen Oxides. Amsterdam:Elsevier. pp. 477–486. Kulle TJ, Clements ML (1987) Susceptibility to virus infection with exposure to nitrogen dioxide. Cambridge:Health Effects Institute. Research Report No. 15. Kunimoto M, Tsubone H, Tsujii N, Mochitate K, Kaya K, Shimojo N, Miura T (1984) Effects of nitrate and nitrite, chemical intermediates of inhaled nitrogen dioxide, on membrane components of red blood cells of rats. Toxicol. Appl. Pharmacol. 74:10–16. Kyono H, Kawai K (1982) Morphometric study on age-dependent pulmonary lesions in rats exposed to nitrogen dioxide. Ind. Health 20:73–99. Lafuma C, Harf A, Lange F, Bozzi L, Poncy JL, Bignon J (1987) Effect of low-level NO2 chronic exposure on elastase-induced emphysema. Environ. Res. 43:75–84. Last JA, Warren DL (1987) Synergistic interaction between nitrogen dioxide and respirable aerosols of sulfuric acid or sodium chloride on rat lungs. Toxicol. Appl. Pharmacol. 90:34–42. Last JA, Gelzleichter TR, Pinkerton KE, Walker RM, Witschi H (1993) A new model of progressive pulmonary fibrosis in rats. Am. Rev. Respir. Dis. 148:487–494.

860

NITROGEN OXIDES

Lazarus SC (2001) The Health Impact of Nitric Oxide: Effects on Lung Function and Cellular and Biochemical Processes in Healthy Humans. Final Report No. 97-329. Sacramento, CA:California Air Resources Board. Lebowitz MD, Holberg CJ, Boyer B, Hayes C (1985) Respiratory symptoms and peak flow associated with indoor and outdoor air pollutants in the southwest. J. Air Pollut. Control Assoc. 35:1154–1158. Lee LT, Kim H, Song H, Hong YC, Cho YS, Shin SY, Hyun YJ, Kim YS (2002) Air pollution and asthma among children in Seoul Korea. Epidemiology 13:481–484. Lefkowitz SS, McGrath JJ, Lefkowitz DL (1986) Effects of NO2 on immune responses. J. Toxicol. Environ. Health 17:241–248. Lehnert BE, Archuleta DC, Ellis T, Session WS, Lehnert NM, Gurley LR, Stavert DM (1994) Lung injury following exposure of rats to relatively high mass concentrations of nitrogen dioxide. Toxicology 89:239–277. Linaker CH, Chauhan AJ, Inskip H, Frew AJ, Sillence A, Coggon D, Holgate ST (1996) Distribution and determinants of personal exposure to nitrogen dioxide in school children. Occup. Environ. Med. 53:200–203. Lin M, Chen Y, Burnett RT, Villeneuve PJ, Krewski D (2002) The influence of ambient coarse particulate matter on asthma hospitalization in children: case-crossover and time-series analyses. Environ. Health Perspect. 110:575–581. Linn WS, Solomon JC, Trim SC, Spier CE, Shamoo DA, Venet TG, Avol EL, Hackney JD (1985a) Effects of exposure to 4 ppm nitrogen dioxide in healthy and asthmatic volunteers. Arch. Environ. Health 40:234–239. Linn WS, Shamoo DA, Spier CE, Valencia LM, Anzar UT, Venet TG, Avol EL, Hackney JD (1985b) Controlled exposure of volunteers with chronic obstructive pulmonary disease to nitrogen dioxide. Arch. Environ. Health 40:313–317. Linn WS, Shamoo DA, Avol EL, Whynot JD, Anderson KR, Venet TG, Hackney JD (1986) Doseresponse study of asthmatic volunteers exposed to nitrogen dioxide during intermittent exercise. Arch. Environ. Health 41:292–296. Liu S, Krewski D, Shi Y, Chen Y, Burnett RT (2003) Association between gaseous ambient air pollutants and adverse pregancy outcomes in Vancouver, Canada. Environ. Health Perspect. 111:1773–1778. Loscutoff SM, Cannon WC, Buschbom RL, Busch RH, Killand BW (1985) Pulmonary function in elastase-treated guinea pigs exposed to ammonium sulfate or ammonium nitrate aerosols. Environ. Res. 36:170–180. Luhr OR, Frostell CG, Heywood R, Riley S, Lonnqvist S (1998) Induction of chromosome aberrations in peripheral blood lymphocytes after short time inhalation of nitric oxide. Mutation Res . 414:107–115. Maigetter RZ, Fenters JD, Findlay JC, Ehrlich R, Gardner DE (1978) Effect of exposure to nitrogen dioxide on T and B cells in mouse spleens. Toxicol. Lett. 2:157–161. Mann JK, Tager IB, Lurmann F, Segal M, Quesenberry Jr, CP, Lugg MM, Shan J, Van den Eeden SK (2002) Air pollution and hospital admissions for ischemic heart disease in persons with congestive heart failure or arrhythmia. Environ. Health Perspect. 110:1247–1252. Maples KR, Sandstr€om T, Su Y-F, Henderson RF (1991) The nitric oxide/heme protein complex as a biological marker of exposure to nitrogen dioxide in humans, rats and in vitro models. Am. J. Respir. Cell Mol. Biol. 4:538–543. Mautz WJ, Kleinman MT, Bufalino C, Cheng W (1993) Effects of episodic exposure to nitric acid on pulmonary structure and function in the rat. Toxicologist 13:150. McConnell R, Berhane K, Gilliland F, London SJ, Vora H, Avol E, Gauderman WJ, Margolis HG, Lurmann F, Thomas DC, Peters JM (1999) Air pollution and bronchitic symptoms in Southern California children with asthma. Environ. Health Perspect. 107:757–760.

REFERENCES

861

McConnell R, Berhane K, Gilliland F, et al. (2003) Prospective study of air pollution and bronchitic symptoms in children with asthma. Am. J. Respir. Crit. Care Med. 168:790–797. McGrath JJ, Oyervides J (1985) Effects of nitrogen dioxide on resistance of Klebsiella pneumoniae in mice. J. Am. Coll. Toxicol. 4:227–231. Medina S, LeTerte A, Quenel P, Lemoullec Y, Lameloise P, Guzzo JC, Festy B, Ferry R, Dab W (1997) Air pollution and doctors’ house calls: results from the ERPURS system for monitoring the effects of air pollution on public health in Greater Paris. France Environ. Res. 75:73–84. Melia RJW, Florey CDV, Chinn S (1979) The relation between respiratory illness in primary schoolchildren and the use of gas for cooking. I. Results from a national survey. Int. J. Epidemiol. 8:333–339. Melia RJW, Florey CDV, Morris RW, Goldstein BD, John HH, Clark D, Craighead IB, Mackinlay JC (1982) Childhood respiratory illness and the home environment: II. Association between respiratory illness and nitrogen dioxide, temperature and relative humidity. Int. J. Epidemiol. 11:164–169. Melia RJW, Florey CV, Altman DG, Swan AV (1977) Association between gas cooking and respiratory disease in children. BMJ. 2:149–152 Mercer RR, Costa DL, Crapo JD (1995) Effects of prolonged exposure to low doses of nitric oxide or nitrogen dioxide on the alveolar septa of the adult rat lung. Lab. Invest. 73:20–28. Mercer RR (1999) Morphometric analysis of alveolar responses of F344 rats to subchronic inhalation of nitric oxide. Research Report No. 88. Cambridge, MA:Health Effects Institute. Metzger KB, Tolbert PE, Klein M, Peel JL, Flanders WD, Todd K, Mhlholland JA, Ryan PB, Frumkin H (2004) Ambient air pollution and cardiovascular emergency department visits. Epidemiology 15:46–56. Miller FJ, Graham JA, Raub JA, Illing JW, Menache MG, House DE, Gardner DE (1987) Evaluating the toxicity of urban patterns of oxidant gases II. Effects in mice from chronic exposure to nitrogen dioxide. J. Toxicol. Environ. Health. 21:99–112. Miller FJ, Graham JA, Illing JW, Gardner DE (1980) Extrapulmonary effects of NO2 as reflected by pentobarbital-induced sleeping time in mice. Toxicol. Lett. 6:267–274. Miller FJ, Overton JH, Myers ET, Graham JA (1982) Pulmonary dosimetry of nitrogen dioxide in animals and man. In:T. Schneider, L Grant, editors. Air Pollution by Nitrogen Oxides. Amsterdam:Elsevier. pp. 377–386 Mochitate K, Kaya K, Miura T, Kubota K (1984) In vivo effects of nitrogen dioxide on membrane constituents in lung and liver of rats. Environ. Res. 33:17–28. Mochitate K, Takahashi Y, Ohsumi T, Miura T (1986) Activation and increment of alveolar macrophages induced by nitrogen dioxide. J. Toxicol. Environ. Health 17:229–239. Mohsenin V (1994) Human exposure to oxides of nitrogen at ambient and supra-ambient concentrations. Toxicology 89:301–312. Mohsenin V (1987) Airway responses to nitrogen dioxide in asthmatic subjects. J. Toxicol. Environ. Health 22:371–380. Mohsenin V (1988) Airway responses to 2.0 ppm nitrogen dioxide in normal subjects. Arch. Environ. Health. 43:242–246. Mohsenin V, Gee JBL (1987) Acute effect of nitrogen dioxide exposure on the functional activity of a1-protease inhibitor in bronchoalveolar lavage fluid of normal subjects. Am. Rev. Respir. Dis. 136:646–650. Morgan G, Corbett S, Wlodarczyk J (1998) Air pollution and hospital admissions in Sydney, Australia. 1990–1994. Am. J. Public Health 88:1761–1766. Morrow PE, Utell MJ, Bauer MA, et al.(1992) Pulmonary performance of elderly normal subjects and subjects with chronic obstructive pulmonary disease exposed to 0.3 ppm nitrogen dioxide. Am. Rev. Respir. Dis. 145:291–300.

862

NITROGEN OXIDES

Morrow PE, Utell MJ (1989) Responses of susceptible subpopulations to nitrogen dioxide. Cambridge, MA:Health Effects Institute. Research Report No. 23. Mortimer KM, Neas LM, Dockery DW, Redline S, Tager IB (2002) The effect of air pollution on inner-city children with asthma. Eur. Respir. J. 19:699–705. Moseler M, Hendel-Kramer A, Karmaus W, Forster J, Weiss K, Urbanek R, Kuehr J (1994) Effect of moderate NO2 air pollution on the lung function of children with asthmatic symptoms. Environ. Res. 67:109–124. Muelenaer P, Reid H, Morris R, Saltzman L, Horstman D, Collier A, Henderson F (1987) Urinary hydroxyproline excretion in young male exposed experimentally to nitrogen dioxide. In:Seifert B, Esdorn H, Fischer M, Rieden H, Wegner J, editors. Indoor Air ’87.Volume 2. Berlin:Institute for Water, Soil and Air Hygiene. pp. 97–103. Munger JW, Collett Jr, J, Danbe Jr, B, Hoffman MR (1990) Fogwater chemistry at Riverside California. Atmos. Environ. 24B:185–205. Murphy SD (1964) A review of effects on animals of exposure to auto exhaust and some of its components. J. Air Pollut. Contr. Assoc. 14:303–308. Nadziejko CE, Nansen L, Mannix RC, Kleinman MT, Phalen RF (1992) Effect of nitric acid vapor on the response to inhaled ozone. Inhal. Toxicol. 4:343–358. Nafstad P, Haheim LL, Wisloff T, et al. (2004) Urban air pollution and mortality in a cohort of Norwegian men. Environ. Health Perspect. 112:610–615. National Institutes of Health (NIH). (1985) The definition of emphysema: Report of a National Heart, Lung, and Blood Institute Division of Lung Diseases Workshop. Am. Rev. Respir. Dis. 132:182–185. Oda H, Nogami H, Kusumoto S, Mukajima T, Kurata A (1980) Lifetime exposure to 2.4 ppm nitric oxide in mice. Environ. Res. 22:254–263. Oda H, Nogami H, Kusumoto S, Nakajima T, Kurata A, Imai K (1976) Long-term exposure to nitric oxide in mice. J. Jap. Soc. Air Pollut. 11:150–160. Oda H, Tsubone H, Suzuki A, Ichinose T, Kubota K (1981) Alterations of nitrite and nitrate concentrations in the blood of mice exposed to nitrogen dioxide. Environ. Res. 25:294–301. Ohashi Y, Nakai Y, Sugiura Y, Ohno Y, Okamoto H (1993) Nitrogen dioxide induced eosinophilia and mucosal injury in the trachea of the guinea pig. J. Oto-Rhino-Laryngol. 55:36–40. Orehek J, Grimaldi F, Muls E, Durand JP, Viala A, Charpin J (1981) Response bronchique aux allergenes apres exposition controlee au dioxyde d’azote [Bronchial response to allergens after controlled NO2 exposure]. Bull. Eur. Physiopathol. Respir. Clin. Respir. Physiol. 17:911–915. Orehek J, Massari JP, Gayrard P, Grimaud C, Charpin J (1976) Effect of short-term, low-level nitrogen dioxide exposure on bronchial sensitivity of asthmatic patients. J. Clin. Invest. 57:301– 307. Oshima H, Friesen M, Brouet I, Bartsch H (1990) Nitrotyrosine as a new marker for endogenous nitrosation and nitration of products. Food Chem. Toxicol. 28:647–652. Overton Jr, JH (1984) Physicochemical processes and the formulation of dosimetry models. In:Miller FJ, Menzel DB, editors. Fundamentals of Extrapolation Modeling of Inhaled Toxicants: Ozone and Nitrogen Dioxide. Washington, DC:Hemisphere Publishing Company. pp. 93–114. Palisano JR, Kleinerman J (1980) APUD cells and NEB in hamster lung. Methods, quantitation and response to injury. Thorax 35:5–11. Parks NJ, Krohn KA, Mathis CA, Chasko JH, Greiger KR, Gregor ME, Peek NF (1981) Nitrogen13-labeled nitrate: Distribution and metabolism after intratracheal administration. Science. 212: 58–61. Pathmanathan S, Krishna MT, Blomberg A, Helleday R, Kelly FJ, Sandstrom T, Holgate ST, Wilson SJ, Frew AJ (2003) Repeated daily exposure to 2 ppm nitrogen dioxide upregulates the expression

REFERENCES

863

of IL-5, IL-10, IL-13, and ICAM-1 in the bronchial epithelium of healthy human airways. Occup. Environ. Med. 60:892–896. Peel JL, Tolbert PE, Klein M, Metzger KB, Flanders WB, Todd K, Mhlholland JA, Ryan PB, Frumkin H (2005) Ambient air pollution and respiratory emergency department visits. Epidemiology 16:164–174. Pershagen G, Rylander E, Norberg S, Eriksson M, Nordvall SI (1995) Air pollution involving nitrogen dioxide exposure and wheezing bronchitis in children. Int. J. Epidemiol. 24:1147–1153. Peters JM, Avol E, Gauderman WJ, et al. (1999a) A study of twelve Southern California communities with differing levels and types of air pollution II. Effects on pulmonary function. Am. J. Respir. Crit. Care Med. 159:768–775. Peters JM, Avol E, Navidi W , et al.(1999b) A study of twelve Southern California communities with differing levels and types of air pollution I. Prevalence of respiratory morbidity. Am. J. Respir. Crit. Care Med. 159:760–767. Pierson WR, Brachaczek WW (1988) Coarse- and fine-particle atmospheric nitrate and HNO3(g) in Claremont, California, during the 1985 nitrogen species methods comparison. Atmos. Environ. 22:1665–1668. Pinkston P, Smeglin A, Roberts Jr, NJ, Gibb FR, Morrow PE, Utell MJ (1988) Effects of in vitro exposure to nitrogen dioxide on human alveolar macrophage release of neutrophil chemotactic factor and interleukin-1. Environ. Res. 47:48–58. Ponka A, Virtanen M (1996) Low-level air pollution and hospital admissions for cardiac and cerebrovascular diseases in Helsinki. Am. J. Public Health 86:1273–1280. Pope CA, Burnett RT. Thun MJ, et al. (2002) Lung cancer, cardiopulmonary mortality, and long-term exposure to fine particulate air pollution. JAMA 287:1132–1141. Posin C, Clark K, Jones MP, Patterson JV, Buckley RD, Hackney JD (1978) Nitrogen dioxide inhalation and human blood biochemistry. Arch. Environ. Health 33:318–324. Postlethwait EM, Bidani A (1989) Pulmonary disposition of inhaled NO2-nitrogen in isolated rat lungs. Toxicol. Appl. Pharmacol. 98:303–312. Postlethwait EM, Bidani A (1990) Reactive uptake governs the pulmonary air space removal of inhaled nitrogen dioxide. J. Appl. Physiol. 68:594–603. Postlethwait EM, Mustafa MG (1981) Fate of inhaled nitrogen dioxide in isolated perfused rat lung. J. Toxicol. Environ. Health 7:861–872. Purvis MR, Ehrlich R (1963) Effect of atmospheric pollutants on susceptibility to respiratory infection: II Effect of nitrogen dioxide. J. Infect. Dis. 113:72–76. Quackenboss JJ, Krzyzanowski M, Lebowitz MD (1991) Exposure assessmente approaches to evaluate respiratory health effects of particulate matter and nitrogen dioxide. J. Expo. Analyt. Environ. Epidemiol. 1:83–107. Rasmussen RE, McClure TR (1992) Effect of chronic exposure to NO2 in the developing ferret lung. Toxicol. Lett. 63:253–260. Rasmussen RE, Mannix RC, Oldham MJ, Phalen RF (1994) Effects of nitrogen dioxide on respiratory tract clearance in the ferret. J. Toxicol. Environ. Health 41:109–120. Rasmussen TR, Brauer M, Kjaergaard S (1995) Effects of nitrous acid exposure on human mucous membranes. Am. J. Respir. Crit. Care Med. 151:1504–1511. Richters A, Kuraitis K (1981) Inhalation of NO2 and blood borne cancer cell spread to the lungs. Arch. Environ. Health 36:36–39. Richters A, Damji KS (1988) Changes in T-lymphocyte subpopulations and natural killer cells following exposure to ambient levels of nitrogen dioxide. J. Toxicol. Environ. Health 25:247–256. Richters A, Richters V (1989) Nitrogen dioxide (NO2) inhalation, formation of microthrombi in lungs and cancer metastasis. J. Environ. Pathol. Toxicol. Oncol. 9:45–51.

864

NITROGEN OXIDES

Robison TW, Forman HJ (1993) Dual effect of nitrogen dioxide on rat alveolar macrophage arachidonate metabolism. Exp. Lung Res. 19:21–36. Robison TW, Kim KJ (1995) Dual effect of nitrogen dioxide on barrier properties of guinea pig tracheobronchial epithelial monolayers cultured in an air interface. J. Toxicol. Environ. Health 44:57–71. Robison TW, Murphy JK, Beyer LL, Richters A, Forman HJ (1993) Depression of stimulated arachidonate metabolism and superoxide production in rat alveolar macrophages following in vivo exposure to 0.5 ppm NO2. J. Toxicol. Environ. Health 38:273–292. Roger LJ, Horstman DH, McDonnell WF, Kehrl H, Ives PJ, Seal E, Chapman R, Massoro E (1990) Pulmonary function, airway responsiveness, and respiratory symptoms in asthmatics following exercise in NO2. Toxicol. Ind. Health 6:155–171. Rombout PJA, Dormans JAMA, Marra M, Van Esch GJ (1986) Influence of exposure regimen on nitrogen dioxide-induced morphological changes in the rat lung. Environ. Res. 41:466–480. Rose RM, Fuglestad JM, Skornik WA, Hammer SM, Wolfthal SF, Beck BD, Brain JD (1988) The pathophysiology of enhanced susceptibility to murine cytomegalovirus respiratory infection during short-term exposure to 5 ppm nitrogen dioxide. Am. Rev. Respir. Dis. 137:912–917. Rubinstein I, Bigby BG, Reiss TF, Bousley HA (1990) Short-term exposure to 0.3 ppm nitrogen dioxide does not potentiate airway responsiveness to sulfur dioxide in asthmatic subjects. Am. Rev. Respir. Dis. 141:381–385. Rubenstein I, Reiss TF, Bigby BG, Stites DP, Boushey HA Jr (1991) Effects of 0.6 ppm nitrogen dioxide on circulating and bronchoalveolar lavage lymohocyte phenotypes in healthy subjects. Environ. Res. 55:18–30. Sackner MA, Dougherty RD, Chapman GA, Zarzecki S, Zarzemske L, Schreck R (1979) Effects of sodium nitrate aerosol on cardiopulmonary function of dogs, sheep, and man. Environ. Res. 18:421–436. Sagai M, Ichinose T, Kubota K (1984) Studies on the biochemical effects of nitrogen dioxide IV. Relation between the change of lipid peroxidation and the antioxidative protective system in rat lungs upon life span exposure to low levels of NO2. Toxicol. Appl. Pharmacol. 73:444–456. Salome CM, Brown NJ, Marks GB, Woolcock AJ, Johnson GM, Nancarrow PC, Quigley S, Tiong J (1996) Effect of nitrogen dioxide and other combustion products on asthmatic subjects in a homelike environment. Eur. Respir. J. 9:910–918. Samet JM, Lambert WE, Skipper BJ, Cushing AH, Hunt WC, Young SA, McLaren LC, Schwab M, Spengler JD (1993) Nitrogen dioxide and respiratory illness in infants. Am. Rev. Respir. Dis. 148:1258–1265. Sandstr€om T, Anderson MC, Kolmodin-Hedman B, Stjernberg N, Angstr€om T (1990) Bronchoalveolar mastocytosis and lymphocytosis after nitrogen dioxide exposure in man: A time-kinetic study. Eur. Respir. J. 3:138–143. Sandstr€om T, et al. (1992a) Effects of repeated exposure to 4 ppm nitrogen dioxide on bronchoalveolar lymphocyte subsets and macrophages in healthy me. Eur. Respir J. 5:1092–1096. Sandstr€om T, Ledin MC, Thomasson L, Helleday R, Stjernberg N (1992b) Reductions in lymphocyte subpopulations after repeated exposure to 1.5 ppm nitrogen dioxide. Br. J. Ind. Med. 49:850– 854. Saul RL, Archer MC (1983) Nitrate formation in rats exposed to nitrogen dioxide. Toxicol. Appl. Pharmacol. 67:284–291. Schenker MB, Samet JM, Speizer FE (1983) Risk factors for childhood respiratory disease: The effect of host factors and home environmental exposures. Am. Rev. Respir. Dis. 28:1038–1043. Schindler C. Ackerman Un-Liebrich Leuenberger P, et al. (1998) Associations between lung function and estimated average exposure to NO2 in eight areas of Switzerland The SAPALDIATeam. Swiss study of air pollution and lung diseases in adults. Epidemiology 9:405–411.

REFERENCES

865

Schlesinger RB (1987) Intermittent inhalation of nitrogen dioxide: Effects on rabbit alveolar macrophages. J. Toxicol. Environ. Health 21:127–139. Schlesinger RB (1989) Comparative toxicity of ambient air pollutants: Some aspects related to lung defense. Environ. Health Perspect. 81:123–128. Schlesinger RB, El-Fawal HAN, Zelikoff JT, Gorczynski JE, McGovern T, Nadziejko CE, Chen LC (1994) Pulmonary effects of repeated episodic exposures to nitric acid vapor alone and in combination with ozone. Inhal. Toxicol. 6:21–41. Schlesinger RB, Dirscoll KE, Vollmuth TA (1987) Effect of repeated exposures to nitrogen dioxide and sulfuric acid mist alone or in combination on mucociliary clearance from the lungs of rabbits. Environ. Res. 44:294–301. Schlesinger RB, Kunzli N, Hidy GM, Gotschi T, Jerrett M (2006) The health relevance of ambient particulate matter characteristics: coherence of toxicological and epidemiological inferences. Inhal. Toxicol. 18:95–125. Seaton A, Dennekamp M (2003) Hypothesis: Ill health associated with low concentrations of nitrogen dioxide-an effect of ultrafine particles? Thorax 58:1012–1015. Segala CB, Fauroux, Just J, Pascual L, Grimfeld A, Neukirch F (1998) Short-term effect of winter air pollution on respiratory health of asthmatic children in Paris. Eur. Respir. J. 11:677–685. Selgrade MJK, Mole ML, Miller FJ, Hatch GE, Gardner DE, Hu PC (1981) Effect of NO2 inhalation and vitamin C deficiency on protein and lipid accumulation in the lung. Environ. Res. 26:422– 437. Shaw Jr, RW, Stevens RK, Bowermaster J, Tesch JW, Tew E (1982) Measurements of atmospheric nitrate and nitric acid: The denuder difference experiment. Atmos. Environ. 16:845–853. Sherwin RP, Carlson DA (1973) Protein content of lung lavage fluid of guinea pigs exposed to 0.4 ppm nitrogen dioxide: Disc-gel electrophoresis for amount and types. Arch. Environ. Health 27:90–93. Sherwin RP, Layfield LJ (1974) Proteinuria in guinea pigs exposed to 0.5 ppm nitrogen dioxide. Arch. Environ. Health 28:336–341. Shima M, Adachi M (2000) Effect of outdoor and indoor nitrogen dioxide on respiratory symptoms in schoolchildren. Int. J. Epidemiol. 29:862–870. Silbaugh SA, Mauderly JL, Macken CA (1981) Effects of sulfuric acid and nitrogen dioxide on airway responsiveness of the guinea pig. J. Toxicol. Environ. Health 8:31–45. Sone S, BRennan LM, Creasia DA (1983) In vivo and in vitro NO2 exposures enhance phagocytic and tumoricidal activities of rat alveolar macrophages. J. Toxicol. Environ. Health 11:151–163. Speizer FE, Ferris Jr, B, Bishop YMM, Spengler J (1980) Respiratory disease rates and pulmonary function in children associated with NO2 exposure. Am. Rev. Respir. Dis. 121:3–10. Spengler JD, Cohen MA (1985) Emissions from indoor sources. In:Gammage RB, Kaye SV, editors. Indoor Air and Human Health. Chelsea, MI:Lewis. pp. 261–278. Stacy RW, Seal Jr, E, House DE, Green J, Roger LJ, Raggio L (1983) A survey of effects of gaseous and aerosol pollutants on pulmonary function of normal males. Arch. Environ. Health 38:104– 115. Stavert DM, Lehnert BE (1988) Concentration versus time is the more important exposure variable in nitrogen dioxide-induced acute lung injury. Toxicologist 8:140. Stavert DM, Archuleta DC, Holland LM, Lehnert BE (1986) Nitrogen dioxide exposure and development of pulmonary emphysema. J. Toxicol. Environ. Health 17:249–267. Stephens RJ, Tallent C, Hart C, Negi DS (1982) Postnatal tolerance to NO2 toxicity. Exp. Mol. Pathol. 37:1–14. Stephens RJ, Freeman G, Evans MJ (1972) Early response of lungs to low levels of nitrogen dioxide: Light and electron microscopy. Arch. Environ. Health 24:160–179.

866

NITROGEN OXIDES

Stevens MA, Menache MG, Crapo JD, Miller FJ, Grahan JA (1988) Pulmonary function in juvenile and young adult rats exposed to low-level NO2 with diurnal spikes. J. Toxicol. Environ. Health 23:229–240. Stieb DM, Judek S, Burnett RT (2003) Meta-analysis of time-series studies of air pollution and mortality: update in relation to the use of generalized additive models. J. Air Waste Manage. Assoc. 53:258–261. Stieb DM, Judek S, Burnett RT (2002) Meta-analysis of time-series studies of air pollution and mortality: effects of gases and particles and the influence of cause of death, age and season. J. Air Waste Manage. Assoc. 52:470–484. Strand V, Salomonsson P, Lundahl J, Bylin G (1996) Immediate and delayed effects of nitrogen dioxide exposure at an ambient level on bronchial responsiveness to histamine in subjects with asthma. Eur. Respir. J. 9:733–40. Strand V, Rak S, Svartengren M, Bylin M (1997) Nitrogen dioxide exposure enhances asthmatic reaction to inhaled allergen in subjects with asthma. Am. J. Respir. Crit. Care Med. 155:881–887. Strand V, Svartengren M, Rak S, Barck C, Bylin G (1998) Repeated exposure to an ambient level of NO2 enhances asthmatic response to a nonsymptomatic allergen dose. Eur. Respir. J. 12:6–12. Studnicka M, Hackl E, Pischinger J , et al. (1997) Traffic-related NO2 and the prevalence of asthma and respiratory symptoms in seven year olds. Eur. Respir. J. 10:2275–2278. Sunyer JJ, Castellsague, Saez M, Tobias A, Anto JM (1996) Air pollution and mortality in Barcelona. J. Epidemiol. Commun. Health 50 (Suppl. 1):76–80. Sunyer J, Spic C, Quenel P, Ponce-de-Leon A, Ponka A, Barumandzadeh T, Touloumi G, Bacharova L, Wojtyniak L, Vonk J, Bisanti L, Schwartz J, Katsouyani K (1997) Urban air pollution and emergency admissions for asthma in four European cities: the APHEA project. Thorax. 52:760– 765. Sunyer J, Puig C, Torrent M, Garcia-Algar O, Calico I, Munoz-Ortiz L, Barnes M, Cullican P (2004) Nitrogen dioxide is not associated with respiratory infection during the first year of life. Int. J. Epidemiol. 33:116–120. Suzuki AK, Tsubone H, Kubota K (1982) Changes of gaseous exchange in the lung of mice acutely exposed to nitrogen dioxide. Toxicol. Lett. 10:327–335. Suzuki T, Ikeda S, Kanoh T, Mizoguchi I (1986) Decreased phagocytosis and superoxide anion production in alveolar macrophages of rats exposed to nitrogen dioxide. Arch. Environ. Contam. Toxicol. 15:733–739. Svartengren M, Strand V, Bylin G, Jarup L, Pershagen G (2000) Short-term exposure to air pollution in a road tunnel enhances the asthmatic response to allergen. Eur. Respir. J. 15:716–724. Tabacova S, Nikiforov B, Balabaeva L (1985) Postnatal effects of maternal exposure to nitrogen dioxide. Neurobehav. Toxicol. Teratol. 7:785–789. Takahashi Y, Miura T (1989) Effects of nitrogen dioxide and ozone in combination on xenobiotic metabolizing activities of rat lungs. Toxicology 56:253–262. Takahashi Y, Mochitate K, Miura T (1986) Subacute effects of nitrogen dioxide on membrane constituents of lung, liver, and kidney of rats. Environ. Res. 41:184–194. Tepper JS, Costa DL, Winsett DW, Stevens MA, Doerfler DL, Watkinson SP (1993) Near-lifetime exposure of the rat to a simulated urban profile of nitrogen dioxide: Pulmonary function evaluation. Fund. Appl. Toxicol. 20:88–96. Touloumi G, Katsouyanni K, Zmirou D, Schwartz J, Spix C, Ponka A , et al.(1997) Shortterm effects of ambient oxidant exposure on mortality: A combined analysis within the APHEA project Air Pollution and Health: A European Approach. Am. J. Epidemiol. 146:177–185. Tunnicliffe WS, Burge PS, Ayres JG (1994) Effect of domestic concentrations of nitrogen dioxide on airway responses to inhaled allergen in asthmatic patients. Lancet 344:1733–1736.

REFERENCES

867

U.S. EPA (1993) United States Environmental Protection Agency Air Quality Criteria for Oxides of Nitrogen. EPA/600/8-91/049aF. Washington, DC:USEPA. U.S. EPA (2008) Integrated Science Assessment for Oxides of Nitrogen and Sulfur: Environmental Criteria. NCEA, ORD, Research Triangle Park, NC. EPA/600/R-08/082. Utell MJ, Swinburne AJ, Hyde RW, Speers DM, Gibb FR, Morrow PE (1979) Airway reactivity to nitrates in normal and mild asthmatic subjects. J. Appl. Physiol. 46:189–196. Utell MJ, Aquilina AT, Hall WJ, Speers DM, Douglas Jr, RG, Gibb FR, Morrow PE, Hyde RW (1980) Development of airway reactivity to nitrates in subjects with influenza. Am. Rev. Respir. Dis. 121:233–241. Vagaggini B, Paggiaro PS, Giannini D, Franco AD, Cianchetti S, Carnevali S, Taccola M, Bacci E, Bancalari L, Dente FL, Giuntini C (1996) Effects of short term NO2 exposure on induced sputum in normal, asthmatic and COPD subjects. Eur. Med. J. 9:1852–1857. Valand SB, Acton JD, Myrvik QN (1970) Nitrogen dioxide inhibition of viral-induced resistance in alveolar monocytes. Arch. Environ. Health 20:303–309. Vaughan Jr, TR, Jennelle LF, Lewis TR (1969) Long-term exposure to low levels of air pollutants Effects on pulmonary function in the beagle. Arch. Environ. Health 19:45–50. Victorin K, Busk L, Cederberg H, Magnusson J (1990) Genotoxic activity of 1.3 butadiene and nitrogen dioxide and their photochemical reaction products in Drosophila and in the mouse bone marrow micronucleus assay. Mutat. Res. 228:203–209. Vollmuth TA, Dirscoll KE, Schlesinger RB (1986) Changes in early alveolar particle clearance due to single and repeated nitrogen dioxide exposures in the rabbit J. Toxicol. Environ. Health 19:255–266. von Nieding G, Wagner HM (1979) Effects of NO2 on chronic bronchitics Environ. Health Perspect. 29:137–142. von Nieding G, Krekeler H, Fuchs R, Wagner M, Koppenhagen K (1973) Studies of the acute effects of NO2 on lung function: Influence on diffusion, perfusion and ventilation in the lungs. Int. Arch. Arbeitsmed. 31:61–72. Wagner H-M (1970) Absorption of NO and NO2 in MIK and MAK concentrations during inhalation. Staub. Reinhalt. Luft (Engl.) 30:25–26. Wagner WD, Duncan BR, Wright PG, Stokinger HE (1965) Experimental study of threshold limit of NO2. Arch. Environ. Health 10:455–466. Wang JH, Devalia JL, Duddle JM, Hamilton SA, Davies RJ (1995) Effect of six-hour exposure to nitrogen dioxide on early-phase nasal response to allergen challenge in patients with a history of seasonal allergic rhinitis. J. Allergy Clin. Immunol. 96 (5 Pt 1):669–676. Ware JH, Dockery DW, Spiro III, A, Speizer FE, Ferris Jr, BG (1984) Passive smoking, gas cooking, and respiratory health of children living in six cities. Am. Rev. Respir. Dis. 129:366–374. Watanabe H, Fukase O, Isomura K (1980) Combined effects of nitrogen oxides and ozone on mice. In:Lee SD, editor. Nitrogen Oxides and Their Effects on Health. Ann Arbor, MI:Ann Arbor Science. pp. 181–189. Wellenius GA, Bateson TG, Mittlemen MA, Schwartz J (2005) Particulate air pollution and the rate of hospitalization for congestive heart failure among Medicare beneficiaries in Pittsburgh, Pennsylvania. Am. J. Epidemiol. 161:1030–1036. Weschler CJ, Brauer M, Koutrakis P (1992) Indoor ozone and nitrogen dioxide: A potential pathway to the generation of nitrate radicals, dinitrogen pentaoxide, and nitric acid indoors. Environ. Sci. Technol. 30:3250–3258. Weschler CJ, Shields HC, Naik DV (1994) Indoor chemistry involving O3, NO, and NO2 as evidenced by 14 months of measurements at a site in Southern California. Environ. Sci. Technol. 28:2120–2132. Wietlisbach V, PopeIII, CA, Ackerman Un-Liebrich (1996) Air pollution and daily mortality in three Swiss urban areas. Soz. Praventivmed. 41:107–115.

868

NITROGEN OXIDES

Wilhelm M, Ritz B (2003) Residential proximity to traffic and adverse birth outcomes in Los Angeles county, California, 1994–1996. Environ. Health Perspect. 111:207–216. Winer AM, Biermann HW (1991) Measurements of nitrous acid, nitrate radicals, formaldehyde and nitrogen dioxide for the Southern California Air Quality Study by differential optical absorption spectroscopy. In: Conference on Chemical Sensing of the Environment: Measurement of Atmospheric Gases, January. Los Angeles, CA.Proc. SPIE-Int. Soc. Opt. Eng. 1433: pp. 44–55. Witschi H (1988) Ozone, nitrogen dioxide and lung cancer: A review of some recent issues and problems. Toxicology 48:1–20. Wong CG, Bonakdar M, Mautz WJ, Kleinman MT (1996) Chronic inhalation exposure to ozone and nitric acid elevates stress-inducible heat shock protein 70 in the rat lung. Toxicology 107:111–119. Wong TW, Lau TS, Yu TS, Neller A, Wong SL, Tam W, Pang SW (1999) Air pollution and hospital admissions for respiratory and cardiovascular diseases in Hong Kong. Occup. Environ. Med. 56:679–683. Wright ES, Vang MJ, Finkelstein JN, Mavis RD (1982) Changes in phospholipid biosynthetic enzymes in type II cells and alveolar macrophages isolated from rat lungs after NO2 exposure. Toxicol. Appl. Pharmacol. 66:305–311. Yanagisawa Y, Nishimura H (1982) A badget-type personal sampler for measurement of personal exposure to NO2 and NO in ambient air. Environ. Int. 8:235–242. Yokoyama E (1968) Effects of acute controlled exposure to NO2 on mechanics of breathing in health subjects. Koshu Eiseiin Kenkyu Hokoku 17:337–346. Yoshida K, Kasama K (1987) Biotransformation of nitric oxide. Environ. Health Perspect. 73: 201–206.

23 OZONE Morton Lippmann

23.1 INTRODUCTION In 1851, soon after its initial laboratory synthesis, Schonbein recognized ozone (O3) as a powerful lung irritant (Bates, 1989). Health effects among the general community were first reported among high school athletes in California, in terms of lesser performance on high O3 exposure days (Wayne et al., 1967). The effects of concern with respect to acute response in the population at large are reductions in lung function and increases in respiratory symptoms, airway reactivity, airway permeability, and airway inflammation. For persons with asthma, there are also increased rates of medication usage, as well as restriction in activities. Margin-of-safety considerations include: (1) the influence of repetitive elicitation of such responses in the progression of chronic damage to the lung of the kinds seen in chronic exposure studies in rats and monkeys; and (2) evidence from laboratory and field studies that ambient air copollutants potentiate the responses to O3. The bases for these concerns are discussed later in this chapter. Ozone is almost entirely a secondary air pollutant, formed in the atmosphere through a complex photochemical reaction sequence requiring reactive hydrocarbons (HCs), nitrogen dioxide (NO2) and sunlight. Ambient O3 can be controlled only by reducing the concentrations of HCs and NO2, or both. Both NO and NO2 are primary pollutants emitted by fossil fuel combustors and are known collectively as NOx. In the atmosphere, NO is gradually converted to NO2. Motor vehicles, one of the major categories of sources of HCs and NOx, have been the target of control efforts, and major reductions (>90%) have been achieved in HC emissions per vehicle. However, there have been major increases in vehicle size and miles driven. Thus, reductions in overall NOx from motor vehicles and stationary-source combustion have been modest. Some net reductions in exposure have occurred in areas with more stringent controls, such as California, while some increases in exposure have occurred in other parts of the United States. In 1988, record high levels of ambient O3 that exceeded the Environmental Toxicants, Third Edition Edited by Morton Lippmann Copyright
2009 John Wiley & Sons, Inc.

869

870

OZONE

FIGURE 23.1 Trends for 1983–2002 period for the annual second maximum 1 h average concentration. Source: U.S. EPA (2003).

1 h 120 ppb limit were reported in 96 communities having over 150 million people. Since then, there has been a gradual decrease in ambient O3 in most of the United States, as indicated in Fig. 23.1. Over the same period, there has been a smaller decrease in the maximum 8 h average concentrations (Fig. 23.2). We know a great deal about O3 chemistry and have developed highly sophisticated O3 air quality models (U.S. EPA, 2006). Unfortunately, the models, and their applications in control strategies have clearly been inadequate in terms of community compliance with the National Ambient Air Quality Standard (NAAQS). We also know a great deal about some of the health effects of O3. However, much of what we know relates to transient, apparently reversible, effects that follow acute exposures lasting from 5 min to 6.6 h. These effects include changes in lung capacity, flow resistance, epithelial permeability, and reactivity to bronchoactive challenges; such effects can be observed within the first few hours after the start of the exposure and may persist for many hours or days after the exposure ceases. Repetitive daily exposures over several days can exacerbate and prolong these transient

FIGURE 23.2 Trends for 1983–2002 period for the annual fourth maximum 8 h average concentration. Source: E.P.A. (2003).

INTRODUCTION

871

effects. There has been a great deal of controversy about the health significance of such effects and whether such effects are sufficiently adverse to serve as a basis for the O3 NAAQS (Lippmann, 1988, 1991, 1993). Decrements in respiratory function such as forced vital capacity (FVC) and forced expiratory volume in the first second of a vital capacity maneuver (FEV1) fall into the category where adversity begins at some specific level of pollutant-associated change. However, there are clear differences of opinion on what the threshold of adversity ought to be. The 1996 O3 Staff Paper (U.S. EPA, 1996b) focused the discussion of thresholds for adversity on persons with impaired respiratory symptoms, as well as on healthy people, because NAAQS are generally set to protect sensitive subgroups of the population. The gradations are presented in Tables 23.1 and 23.2 for healthy persons and persons with impaired respiratory symptoms, respectively. TABLE 23.1 (a–c) Gradation of Individual Responses to Short-Term Ozone Exposure in Healthy Persons Functional Response (a) FEV1

None Within normal range (3%) Within normal range

Nonspecific bronchial responsiveness Duration of None response (b) Symptomatic Response Cough

Moderate

Large

Decrements of Decrements of Decrements of 3% to 10% >10% but <20% 20% Increases of <0% Increases of 300% Increases of >300% <4 h

>4 but 24 h

>24 h

Normal

Mild

Moderate

Severe

Infrequent cough

Persistent Cough with deep Frequent uncontrollable breath spontaneous cough cough Discomfort just Marked discomfort Severe discomfort on exercise on exercise or noticeable on deep breath or deep exercise or deep breath <4 h >4 but 24 h >24 h

Chest pain

None

Duration of response

None

(c) Impact of various functional and/or symptomatic responses Interference with normal activity

Small

Small functional Normal and/or mild functionaland/or symptomatic symptomatic responses responses

Large functional Moderate and/or severe functional and/or symptomatic symptomatic responses responses

None

Many sensitive A few sensitive individuals individuals likely to limit activity likely to limit activity

Source: U.S. EPA (1996b).

None

872

OZONE

TABLE 23.2 (a–c) Gradation of Individual Responses to Short-Term Ozone Exposure in Persons with Impaired Respiratory Systems Functional Response

None

(a) FEV1 change

Decrements of <3% Nonspecific Within bronchial normal responsiveness range Within Airway normal resistance range (SRaw) (20%) Duration of None response

Small

Moderate

Large

Decrements of Decrements of Decrements of 3%–10% >10% but <20% 20% Increases of <100% Increases of 300% Increases of >300%

SRaw increased <100%

SRaw increased up SRaw increased to 200% or up to >200% or more 15 cm H2O/s than 15 cm H2O/s

<4 h

>4 but 24 h

>24 h

Normal

Mild

Moderate

Severe

None

With otherwise With shortness of normal breathing breath

Cough

Infrequent cough

Cough with deep breath

Chest pain

None

Duration of response

None

Discomfort just noticeable on exercise or deep breath <4 h

(b) Symptomatic response Wheeze

Persistent with shortness of breath Persistent Frequent uncontrollable spontaneous cough cough Marked discomfort Severe discomfort on exercise or on exercise or deep breath deep breath

>4 but 24 h

>24 h

(c) Large functional Moderate Small functional Impact of various Normal and/or severe functional and/or and/or mild functional functional and/ symptomatic symptomatic symptomatic or symptomatic and/or responses responses responses symptomatic responses sresponses Interference with None Few individuals Many individuals Most individuals normal activity likely to limit likely to limit likely to limit activity activity activity Medical No change Normal medication Increased frequency Increased treatment/self as needed or additional likelihood of medication medication use physician or ER visit Source: U.S. EPA (1996b).

BACKGROUND ON EXPOSURES AND HEALTH-RELATED EFFECTS

873

With respect to adversity, the 1996 staff paper concluded that responses listed as large or severe were clearly adverse. For responses listed as moderate, it was concluded that they could be considered adverse if there were repetitive exposures. Although we know a great deal about the transient effects following single exposure to O3 resulting from controlled laboratory exposures, and short-term responses in populations associated with peak ambient air concentrations, our current knowledge about the chronic health effects of O3 is much less complete. As discussed in the latter part of this chapter, the known chronic effects include alterations in lung function or structure. Such effects may result from cumulative damage and/or from the side effects of adaptive responses to repetitive daily or intermittent exposures. This review does not discuss the effects of O3 or its metabolites on nonpulmonary tissues or organs. It also does not discuss the health effects of increased ultraviolet radiation resulting from the depletion of stratospheric O3, which are viewed as minor in the 2006 O3 criteria document (U.S. EPA, 2006). This chapter provides a critical review of the health effects data and their significance to public health in relation to the populations exposed. The judgments made have been influenced by my participation in public CASAC reviews of U.S. Environmental Protection Agency (EPA) documents, but they differ, in some cases, from those of the EPA and of others on the CASAC panels.

23.2 BACKGROUND ON EXPOSURES AND HEALTH-RELATED EFFECTS 23.2.1

Sources and Distribution of O3 in Ambient Air

O3 in ambient air is attributable to several different sources. One is the intrusion of stratospheric O3, especially in the spring when the stratospheric–tropospheric air exchange is greatest. The other sources are driven by complex photochemical reaction sequences requiring input of HCs, NOx, and actinic radiation. Reactive organic vapors such as olefinic hydrocarbons, formaldehyde, and m-xylene, which are largely products of anthropogenic activities, are highly efficient contributors to O3 formation. On the other hand, methane (CH4), a major product of natural biogenic decay, and a relatively nonreactive hydrocarbon, can also contribute to O3 formation. Actually, the background concentration of CH4 has been rising over the last 100 years as a result of increasingly intensive agriculture and animal husbandry. The coincident increase in continental background O3 over the past century, from 10 to 20 ppb (Altshuller, 1987) to the current level near 40 ppb, may be due to the rising background of both CH4 and NOx. The NOx concentrations have grown continuously as fossil fuel usage has increased. The increase in NOx may also account for a greater rate of O3 formation by photochemical reactions with isoprene and terpenes emitted by trees. As also noted by Altshuller (1987), the role of NOx in tropospheric O3 formation is especially critical. Unless NOx concentrations exceed about 0.02 to 0.03 ppb, photochemical O3 loss exceeds photochemical O3 production. There are remote regions of the troposphere where the NOx concentrations may be below such values. On the other hand, NOx concentrations in the rural planetary boundary layer over the United States usually exceed 1 ppb. NOx concentrations of 5 to 10 ppb are typical of rural areas within more heavily populated areas in the United States and Europe. Empirical estimates based on O3 and NOx measurements at a site at 3 km elevation, that is, Niwot Ridge in Colorado, indicate that in summer 17  3 ppb of O3 is formed per 1 ppb of NOx when NOx concentrations are below 1 ppb. At lower elevation rural sites elsewhere in the United States, where NOx

874

OZONE

FIGURE 23.3 Frequency functions of duration of O3 concentrations in excess of 80 ppb over New Jersey and Connecticut for 2 years. From Rao (1988).

concentrations were within the 1 to 10 ppb range, 5 to 7 ppb of O3 was estimated to be formed per 1 ppb of NOx reacted. Since O3 is highly reactive with ground-level surfaces, it drops markedly in the evening. However, it can remain at elevated concentrations in the ambient air above the mixing layer. This elevated reservoir of O3 can then contribute to elevated ground level O3 on the following day as air mixing increases. This contributes to multiday summer episode exposures. Rao (1988) has shown that the likelihood of O3 >80 ppb continuing for 3 days or longer, once it has been in existence for 1 day, is high in the northeastern United States. This is illustrated in Fig. 23.3 for a typical year (1981) and a relatively high O3 exposure year (1983). The peak concentrations of O3 during a specific day at a specific location are determined largely by the baseline level in the air aloft, the photochemical production rate during the day, and the concentration of O3 scavenging chemicals such as nitric oxide (NO) and ethylene, and depends on the ambient ratio of reactive organic gases (ROGs) to NOx concentrations. When [ROG]/[NOx] is approximately 5–6, the two species have about an equal chance of reacting with hydroxy (OH) radical. If this ratio is much larger than 5 to 6, there is a shortage of NO that can be oxidized to NO2, and O3 production is controlled by the amount of NOx available. In this region, decreasing NOx leads to a decrease in the peak O3. However, when [ROG]/[NOx] is on the order of 5 or less, the ready availability of NOx makes O3 formation dependent on ROG. NO will scavenge O3 faster than it reacts with RO2, and NO2 will react with OH to form nitric acid. Decreasing NOx can lead to an increase in peak O3 as the efficiency of O3 formation increases. The daily formation of O3, in the absence of a substantial baseline level from the air aloft or upwind, leads to a relatively sharp daily peak in concentration, with a major part of the effective exposure taking place over a relatively few hours. However, in recent years, this type of exposure pattern has become relatively rare. In heavily populated regions, such as the eastern United States and western Europe, a typical daily plateau of exposure occurs after 10 a.m. and the maximum 8 h exposure is approximately 90% of the maximum 1 h exposure (Rombout et al., 1986). This type of exposure pattern is consistent with the hypothesis that relatively little of the exposure on a typical high exposure summer day is attributable to local

BACKGROUND ON EXPOSURES AND HEALTH-RELATED EFFECTS

875

FIGURE 23.4 Three-day sequence of hourly O3 concentrations at Montague, MA SURE station showing locally generated midday peaks and long-range transport late peaks. From U.S. EPA (1986).

sources or amenable to local source control. Rather, the local generation of O3 represents a bump on a broad daily hump arising from a series of upwind sources and photochemistry. The size of the bump depends on the concentration of precursor reactants in the incoming air and the local increments of reactants. The broad humps can be attributed to the sum of the contributions of stratospheric O3 injections and O3 formed upwind and retained aloft for one or many days. The nature of contemporary O3 exposure is illustrated for a rural area of western Massachusetts in Fig. 23.4, showing both locally generated peaks and late-afternoon peaks from upwind population centers superimposed on a broad daily plateau (Lioy and Dyba, 1989). It clearly illustrates that the O3 exposure problem affects broad areas of the country and is not an urban problem only. 23.2.2

Ozone Exposures and Dosimetry

For O3, the only significant exposure route is inhalation, and exposure can be defined as the concentration at the nose and mouth. There have been few personal measurements of O3, and it is generally assumed that the concentrations that we breathe are the same as those measured at central monitoring sites. However, this assumption has limited validity. For time spent outdoors, local concentrations are reduced in the vicinity of heavy vehicular traffic due to scavenging by NO. However, less traffic areas downwind of the monitor may have a higher O3 concentration because of the enrichment of the air mass with motor vehicle exhaust precursor chemicals and active photochemistry. Thus, outdoor O3 concentrations can be either higher or lower than those measured at monitoring sites. Indoor concentrations of O3 are almost always substantially lower than those outdoors because of efficient scavenging by indoor surfaces and the lack of indoor sources. The only common indoor sources are copying machines and electrostatic air cleaners. Since most people spend more than 80% of their time indoors, their O3 exposure is much lower than estimates based on outdoor concentrations. Ozone exposure is only one determinant of O3 dose. Dose is also determined by the volumes of air inhaled and by the pattern of uptake of O3 molecules along the respiratory tract. When people work or exercise outdoors and increase their rate of ventilation, the

876

OZONE

contribution of outdoor exposure to total dose of O3 becomes the major determinant of total O3 dose. The dose to target tissues in the respiratory acini (the region from the terminal bronchioles through the alveolar ducts) increases even more with exercise than does total respiratory tract dose, since O3 penetration to distal lung airways increases with tidal volume and flow rate. Gerrity et al. (1988) measured the efficiency of O3 removal from inspired air by the extrathoracic and intrathoracic airways in healthy, nonsmoking young male volunteers for O3 concentrations of 100, 200, and 400 ppb for nose only, mouth only, and oronasal breathing, respectively, at 12 and 24 breaths/min. The mean extrathoracic removal efficiency for all measurements was 39.6  0.7%, and the mean intrathoracic removal efficiency was 91.0  0.5%. The effects of concentration, breathing frequency, and mode of breathing on removal efficiency while significant were relatively small. Gerrity et al. (1995) used a bronchoscope to sample air at various lung depths in healthy nonsmokers, with the distal end sequentially positioned at the bronchus intermedius (BI), main carina (CAR), upper trachea, and above the vocal cords. O3 concentration was measured continuously at each site using a rapid-responding O3 analyzer. The subjects breathed through a mouthpiece at 12 breaths/min. Integration of the product of the flow and O3 concentrations during inspiration and expiration provided the O3 mass passing each anatomic location during each phase of respiration. On inspiration, the fractional uptake of O3 by structures between the mouth and each location were 0.18  0.04 (SE), 0.27  0.02, 0.36  0.03, and 0.33  0.03 for above the vocal cords, upper trachea, CAR, and BI, respectively. A significant effect of location on uptake was found by an analysis of variance. Studies of the O3 uptake within the human respiratory tract have been conducted by Rigas et al. (2000) in tidal breathing. Male and female adults inhaled 200 or 400 ppb O3 while exercising at 20 L/min for 60 min or 40 L/min for 30 min. Fractional absorption ranged from 0.56 to 0.98, with an intersubject variability of approximately 10%. In the same laboratory, Asplund et al. (1996) used a continuous O3 exposure followed by an O3 bolus, and Rigas et al. (1997) used O3 boli following continuous NO2 and SO2 exposures. With continuous O3 exposure, the absorbed fraction of the bolus decreased, suggesting that biochemical substances on the airways were being depleted, whereas with continuous NO2 and SO2 exposures, the absorbed fraction of the O3 bolus increased, suggesting that the NO2 and SO2 exposures were increasing the availability of the biochemical substances that absorb O3. The tissues within the respiratory acini of humans, rabbits, guinea pigs, and rats receive the highest local dose from inhaled O3 according to the models developed by Miller and colleagues (Hatch et al., 1989; Miller et al., 1978b; Overton and Miller, 1987), with the dose in humans being about twice that in rats for the same exposure (Gerrity and Wiester, 1987), and with children having somewhat greater doses than adult humans (Overton and Graham, 1989). This comparative dosimetry is consistent with the greater effects of O3 on lung function seen for comparable exposures in humans than in rats (Costa et al., 1989). 23.2.3

Populations of Concern for Health Effects

In general, the NAAQSs have been established to protect against adverse health effects in the most sensitive subpopulation that is identifiable (Lippmann, 1987). For example, cardiovascular patients were of paramount concern in establishing the NAAQS for carbon monoxide (CO), which binds to hemoglobin and further reduces their already limited capacity to oxygenate the blood. Asthmatics were of special concern in establishing the sulfur dioxide (SO2) NAAQS because the concentrations required to produce comparable

EFFECTS OF SHORT-TERM EXPOSURES TO OZONE IN HUMANS

877

levels of bronchoconstriction were about an order of magnitude lower than those for normal people and because of the potential for a disabling or fatal bronchospasm being initiated by a transient high concentration of SO2. In the case of O3, no special functional responsiveness has yet been clearly demonstrated among the potentially more sensitive groups with preexisting disease (Lippmann, 1989a, 1989b, 1993). Thus, consideration is being given to healthy people who exercise regularly outdoors as a primary population of concern on the basis of their higher O3 exposures and doses. The EPA ozone staff papers have also identified people with asthma as a population of concern on the basis of reports of symptomatic responses, increased rates of visits to clinics, and hospital admissions at very low ambient O3 concentrations. 23.2.4

Health-Related Responses of Concern

As has been previously noted, O3 in ambient air has been associated with a variety of transient effects on the respiratory airways. Among the best documented of these changes are dose-related decrements in indices of forced expiratory flow capacity, which is reproducible in individuals and highly variable among the population. Increased rates of symptoms, clinic visits, and hospital admissions are other responses of concern with respect to peak exposures. More persistent physiological decrements associated with structural alterations of lung airways could also be considered adverse effects if they occurred in humans as a result of repetitive O3 exposures. Although human evidence is currently lacking, such effects have been produced in laboratory animals following chronic exposures. Thus, these effects are also of concern for human populations with high levels of chronic exposure.

23.3 EFFECTS OF SHORT-TERM EXPOSURES TO OZONE IN HUMANS 23.3.1

Respiratory Mechanical Function Responses

23.3.1.1 One- and Two-Hour Chamber Exposure Studies There are more data on respiratory function responses than on any other coincident responses to short-term O3 inhalation. Such functional responses can be obtained with noninvasive, readily performed protocols and can be detected at levels of exposure as low as or lower than any of the other well-established assays. The major debate about very small but statistically significant decrements in function from such studies is how to interpret their health significance (Lippmann, 1988). It is well established that the inhalation of O3 causes concentration-dependent mean decrements in exhaled volumes and flow rates during forced expiratory maneuvers, and that the mean decrements increase with increasing depth of breathing (Hazucha, 1987). There is a wide range of reproducible responsiveness among healthy subjects (Frampton et al., 1997; McDonnell et al., 1985a; Weinmann et al.,1995). Functional responsiveness to O3 is not greater, and usually lower, among cigarette smokers (Frampton et al., 1997; Kagawa, 1984; Shephard et al., 1983), older adults (Drechsler-Parks et al., 1987; McDonnell et al., 1993, 1995; Reisenauer et al., 1988), asthmatics (Koenig et al., 1987; Linn et al., 1980), and patients with chronic obstructive pulmonary disease (COPD) (Linn et al., 1983; Solic et al., 1982). An exception is that patients with allergic rhinitis had greater changes in airway resistance (McDonnell et al., 1987).

878

OZONE

A prospective confirmation of reduced responsiveness to O3 among asymptomatic cigarette smokers was produced by Emmons and Foster (1991). They measured respiratory function before and after 2 h of O3 at 400 ppb with light exercise in smokers before they stopped smoking and again after 6 months of not smoking. None was responsive to O3 exposure before smoking cessation. During smoking cessation, their mean baseline FEF25–75 was raised from 3.0 to 4.1 L/sec. For the subjects re-exposed to O3 6 months later, the exposure reduced their mean FEF25–75 from 3.9 to 3.0 L/sec. The subjects with the greatest improvement in FEF25–75 after withdrawal had the largest acute decrements after O3 exposure. Smoking cessation did not significantly increase FVC or forced expiratory volume in 1 s (FEV1), and O3 exposure after smoking cessation did not produce significant decrements in these respiratory function parameters. Weinmann et al. (1995) showed that O3-induced changes in FEF25–75 were unexplained and followed a different time-course than O3-induced changes in FVC. Their analysis indicated that intrinsic narrowing of the small airways might be a significant indicator of the functional response. While the results of some laboratory studies have indicated that responses in young females was greater than those in young males (Messineo and Adams, 1990), the largest study of both males and females did not find gender-related differences in responsiveness to O3 among either black or white adults (Seal et al., 1993). The first indications that the effects of O3 on respiratory function accumulate over more than 1 h were the observations of McDonnell et al. (1983) and Kulle et al. (1985) in chamber exposures to O3 in purified air for 2 h with the volunteers engaged in vigorous intermittent exercise. Significant function decrements observed after 2 h of exposure were not present at measurements made after 1 h. 23.3.1.2 Field Studies Spektor et al. (1998a) noted that children at summer camps with active outdoor recreation programs had greater decrements in lung function than children exposed to O3 at comparable concentrations in chambers for 1 or 2 h. Furthermore, their activity levels, although not measured, were known to be considerably lower than those of the children exposed in the chamber studies while performing very vigorous exercise. Since it is well established that functional responses to O3 increase with levels of physical activity and ventilation (Hazucha, 1987), the greater responses in the camp children had to be caused by other factors, such as greater cumulative exposure, or by the potentiation of the response to O3 by other pollutants in the ambient air. Cumulative daily exposures to O3 were generally greater for the camp children, since they were exposed all day long rather than for a 1 or 2 h period preceded and followed by exposure to clean air. Similar considerations apply to the studies of Kinney et al. (1988) and Hoek et al. (1993) of school children. In the Kinney et al. (1988) study in Kingston and Harriman, TN, lung function was measured in school on six occasions during a 2-month period in the late winter and early spring. Child-specific regressions of function versus maximum 1 h O3 during the previous day indicated significant associations between O3 and function, with coefficients similar to those seen in the summer camp studies of Lippmann et al. (1983), Spektor et al. (1988a, 1991, Higgins et al. (1990), and Hoek et al. (1993). Since children in school may be expected to have relatively low activity levels, the relatively high response coefficients may be related to potentiation by other pollutants or to a low level of seasonal adaptation. Kingston–Harriman is notable for its relatively high levels of aerosol acidity. As shown by Spengler et al. (1989), Kingston–Harriman has higher annual average and higher peak acid aerosol concentrations than other cities studied, that is, Steubenville, OH; St. Louis,

EFFECTS OF SHORT-TERM EXPOSURES TO OZONE IN HUMANS

879

MO; and Portage, WI. Alternatively, the relatively high response coefficients could have been caused by the fact that the measurements were made in the late winter and early spring. Linn et al. (1988) have shown evidence for a seasonal adaptation, and children studied during the summer may not be as responsive as children measured earlier in the year. This study will be discussed further Section 23.8. In a study of children with moderate to severe asthma at a summer camp in the Connecticut River Valley (Thurston et al., (2001)), the association between decrements in peak expiratory flow rates associated with ambient O3 concentrations were similar in magnitude to those reported by the same group of investigators for healthy children at other summer camps in northeastern United States (Spektor et al., (1988a, 1991). However, the level of physical activity of the asthmatic children, and hence their O3 intake, was much lower. Also, the asthmatic children have less reserve functional capacity. Thus, the level of health concern for such comparable functional decrements is much greater. Other recent studies of the effects of O3 on lung function in children in natural settings have also demonstrated O3-related functional decrements. Braun-Fahrlander et al. (1994) showed O3-related reductions in peak expiratory flow rate (PEFR) among 9–11-year-old Swiss children following 10 min of heavy exercise at peak O3 concentrations below 80 ppb. Neas et al. (1995) demonstrated O3-related reductions in PEFR between morning and evening in fourth- and fifth-grade children in Uniontown, PA, in relation to 12 h av. O3 below 88 ppb. Castillejos et al. (1995) studied the change in lung function following exercise out of doors for 7 1/2–11-year-old children in Mexico City who were repeatedly exposed to high ambient levels of O3 and particulate matter (PM). They had O3-related decrements in FVC, FEV1.0, FEF25–75, and FEV1.0/FVC when peak 1 h O3 exceeded 150 ppb. Field studies of functional responses of adults engaged in recreational activities outdoors in the presence of varying levels of O3 have also been performed. Spektor et al. (1988b) made pre- and postexercise respiratory function measurements on young adults who were engaged in daily outdoor exercise for about one-half hour per day in an area with regional summer haze but no local point sources. The magnitudes of the functional decrements per unit of ambient O3 concentration were similar to those observed in volunteers exposed while exercising vigorously for 1 or 2 h in controlled chamber exposure studies. Functional decrements in proportion to relatively low ambient O3 concentrations have also been reported for joggers in Houston, TX (Selwyn et al., 1985), competitive cyclists in The Netherlands (Brunekreef et al., 1994), hikers on Mount Washington in NH (Korrick et al., 1998), and agricultural workers in British Columbia (Brauer et al., 1996). 23.3.1.3 Prolonged Daily Exposures in Chambers The observations from the field studies in children’s camps stimulated Folinsbee et al. (1988) at the EPA Clinical Studies Laboratory in Chapel Hill, NC, to undertake a chamber exposure study of 10 adult male volunteers involving 6.6 h of O3 exposure at 120 ppb. Moderate exercise was performed for 50 min/h for 3 h in the morning and again in the afternoon. They found that the functional decrements become progressively greater after each hour of exposure, reaching average values of approximately 400 mL for forced vital capacity and approximately 540 mL for forced expiratory volume in 1 s by the end of the day. The effects were transient in the sense that there were no residual functional decrements on the following day. The decrements in FEV1 after 6.6 h of exposure at 120 ppb averaged 13.6% and were comparable to those seen previously in the same laboratory on similar subjects following 2 h of intermittent heavier exercise (68 L inhaled per minute for a total exercise time of 60 min) at an interpolated concentration of approximately 120 ppb. Assuming that the rate of ventilation was 10 L/min

880

OZONE

FIGURE 23.5 Mean FEV1 after each 50 min of exercise during exposures to O3 at 0 (open circles), 80 ppb (squares), 100 ppb (triangles), and 120 ppb (solid circles). Asterisks indicate significant reduction in FEV1 from corresponding values at 0 ppb. From Horstman et al. (1990).

between exercise periods, the total amount of O3 inhaled during 2 h of intermittent heavy exercise at 220 ppb (430 mg/m3) would be (60 min  0.068 m3/min þ 60 min  0.010 m3/ min)  430 mg/m3 ¼ 2.01 mg O3. The corresponding amount of O3 inhaled during 6.6 h of intermittent moderate exercise at 120 ppb would be (300 min  0.040 m3/min þ 100 min  0.010 m3/min)  235 mg/m3 ¼ 3.06 mg O3. Thus, the effect accumulated with time, but there was a temporal decay of the effect going on at the same time. Follow-up studies (in the same laboratory) by Horstman et al. (1990) were done on 21 adult males with 6.6 h exposures at 80, 100, and 120 ppb. The exposures at 120 ppb produced very similar responses, for example, a mean FEV1 decline of 12.3%, whereas those at 80 and 100 ppb showed lesser changes that also became progressively greater after each hour of exposure (Fig. 23.5). A further follow-up study using the same exposure protocol on 38 additional healthy young men was done by McDonnell et al. (1991) at 80 ppb. There was a mean FEV1 decline of 8.4%, which was similar to that seen by Horstman et al. (1990) at that concentration. The timescale for an effective O3 dose in relation to functional response was explored further by Hazucha et al. (1992) in exposures of healthy young adults lasting 8 h, with 30 min of exercise (@ 40 L/min) at the beginning of each hour. The O3 concentration rose from 0 to 240 ppb over the first 4 h and dropped back to zero over the second 4 h. The functional responses were compared with both sham exposures and constant 120 ppb exposures in the same subjects. By 4 h, the FEV1 changes from both O3 exposures were similar, and the largest decrement in FEV1, which occurred after 6 h of exposure, was about twice as large as that after 5 to 8 h of constant exposure at 120 ppb. The peak response faded by the end of 8 h and was not significantly greater than that produced by the constant 120 ppb exposure at the eighth hour. Another study looking into the integral effects of temporally varying exposures with the same integral exposure was performed by McKittrick and Adams (1995). Aerobically trained young adult men were exposed while exercising at 60 L/min to either 1 h @ 300 ppb O3 followed by 1 h of clean air; intermittent 1/2 h at 300 ppb and 0 ppb, or intermittent quarter hours at 300 and 0 ppb. The FEV1 decrements at the end of exposure to O3 were essentially the same, that is, 17.6, 17.0 and 17.9%. Larsen et al. (1991) modeled the data of Horstman et al. (1990) using multiple linear regressions on the mean responses at each hour for all three concentrations, but

EFFECTS OF SHORT-TERM EXPOSURES TO OZONE IN HUMANS

881

FIGURE 23.6 Comparison of mean O3-induced FEV1 decrements due to 6.6 hr exposures with mild exercise in the various studies cited in the inset.

excluding those with FEV1 decreases of less than 0.5%. With O3 concentration and duration of exposure as the only two independent variables, the model explained 95% of the variance of the dependent variable Z, a Gaussian transform of the percentage decrease in FEV1. In this model, the exponent of the exposure duration is 0.754. This further demonstrates that exposure time is almost equally important to exposure concentration in cumulative response when concentrations are in the range of normal peak ambient levels. Further evidence of the time scale for the biological integration of O3 exposure can be deduced from the rate at which the effects dissipate. In a study by Folinsbee and Hazuch (), young adult females were exposed to 350 ppb O3 for 70 min, including two 30 min periods of treadmill exercise at 40 L/min. Their mean decrement in FEV1 at the end of the exposure was 21%. After 18 h, their mean decrement was 4%, whereas at 42 h it was 2%. The large interindividual variability of O3-induced functional responses that is illustrated in Fig. 23.6 is not yet understood, and functional responses in individuals do not correlate well with the other responses that will be discussed below. Using the large EPA database, McDonnell et al. (1993) found that O3 concentration explained 31% of the variance in FEV1 responses, and subject age explained another 4%. The modeled influence of age is illustrated in Fig. 23.7. Upon further modeling of this large data set, McDonnell et al. (1997) reported that a sigmoid-shaped model was consistent with previous observations of O3 exposure–response (ER) characteristics and accurately predicted the mean response with independent data. Neither did they find that response was more sensitive to changes in C than in VE nor did they find convincing evidence of an effect of body size upon response, but response to O3 decreased with age. Using the data collected for 68 individuals exposed two or more times for 6.6 h, McDonnell et al. (1995) found that 47% of those exposed to 120 ppb had an FEV1 decrement of 10% or more. These analyses helped demonstrate that the respiratory function effects can accumulate over many hours and that an appropriate averaging time for transient functional decrements caused by O3 is 6 h. This was a major factor for the change in the

882

OZONE

FIGURE 23.7 Predicted mean decrements in forced expiratory volume in L(DFEV1) following 2 h esposures to ozone while undergoing heavy intermittent exercise for three ages. (Note: to convert DFEV1 to % DFEV1, multiply by 22.2%.). From McDonnell et al. (1993).

averaging time for the primary O3 NAAQS from 1 h to 8 h. Another factor was the recognition that O3 exposures in ambient air can have broad peaks with 8 h averages equal to approximately 90% of the peak 1 h averages (Rombout et al., 1986). 23.3.2

Effects on Athletic Performance

It has been four decades since epidemiological evidence suggested that the percentage of high school track team members failing to improve performance increased with increasing oxidant concentrations the hour before a race (Wayne et al., 1967). The effects may have been related to increased airway resistance or to associated discomfort, which may have limited motivation to run at maximal levels. Controlled exposure studies of heavily exercising competitive runners have demonstrated decreased function at 200 to 300 ppb (Adams and Schelegle, 1983; Savin and Adams, 1979). At 210 ppb O3, Folinsbee et al. (1984) reported symptoms as well in seven distance cyclists exercising heavily (VE ¼ 81 L/min). Some studies have shown reduced performance at lower O3 concentrations. Schelegle and Adams (1986) exposed young male adult endurance athletes to 120, 180, and 240 ppb O3 whereas the exercised at a mean VE of 54 L/min for 30 min, followed by a mean VE of 120 L/min for an additional 30 min. Although they all completed the protocol for filtered air (FA) exposure, some of them could not complete it for the 120, 180, and 240 ppb exposures. Linder et al. (1988) also found that maximum performance time was reduced for their 16–28 min progressive maximum exercise for VE of 30–120 L/min in young adults when O3 was present. For example, performance was reduced 11% in females exposed to 130 ppb O3. 23.3.3

Symptomatic Responses

Respiratory symptoms have been closely associated with group mean pulmonary function changes in adults acutely exposed in controlled exposures to O3 and in ambient air containing O3 as the predominant pollutant. However, Hayes et al. (1987) found only a weak-tomoderate correlation between FEV1 changes and symptoms severity when the analysis was conducted using individual data. In controlled 2 hr O3 exposures, McDonnell et al. (1983) reported that some heavily exercising adult subjects experienced cough, shortness of breath, and pain on deep

EFFECTS OF SHORT-TERM EXPOSURES TO OZONE IN HUMANS

883

inspiration at 120 ppb O3, although the group mean response was statistically significant for cough only. Above 120 ppb O3, respiratory and nonrespiratory symptoms include throat dryness, chest tightness, substernal pain, cough, wheeze, pain on deep inspiration, shortness of breath, dyspnea, lassitude, malaise, headache, and nausea. The prolonged exposure studies involving 6.6 h of exposure at concentrations between 80 and 120 ppb also produced significant increases in respiratory symptoms including cough and pain on deep inspiration (Koenig et al., 1987; Linn et al., 1980). Linder et al. (1988) reported that brief exposures (16–28 min) to 120 to 130 ppb O3 at high ventilatory rates (30–120 L/min) produced symptoms of irritation and cough in young adults. Although O3 causes symptomatic responses in adults at current peak levels, such responses do not occur in healthy children (Avol et al., 1985, 1987). Children (ages 8– 11) exposed for 2.5 h at 120 ppb O3 while intermittently exercising (VE ¼ 39 L/min) showed small but statistically significant decreases in FEV1 but showed no changes in frequency or severity of cough compared to controls (McDonnell et al., 1985a, 1985b). Similarly, adolescents (age 12–15) continuously exercising (VE ¼ 31–33 L/min) during exposure to 144 ppb mean O3 in ambient air showed no changes in symptoms despite statistically significant decrements in group mean FEV1 (4%), which persisted at least 1 h during postexposure resting (Avol et al., 1985). These laboratory results are consistent with the results obtained in a series of field studies of healthy children at summer camps, which failed to find any symptomatic responses despite the occurrence of relatively large decrements in function that were proportional to the ambient O3 concentrations (Spektor et al., 1988a). In a study by Hoek and Brunekreef (1995) of a general population sample of 300 children aged 7–11 years who had shown functional responses to O3 in ambient air, there were no responses in terms of symptoms based on diaries maintained by their parents. In panels of 300 healthy children in the Harvard six-cities study, diaries of respiratory symptoms were kept over a 1-year period. In single pollutant models for the April–August period, there was a significant association between O3 and the incidence of cough that was independent of other measured pollutants (Schwartz et al., 1994). For a group of 7–9-year-old children in Mexico City who were repeatedly exposed to high concentration of O3 and PM, Castillejos et al. (1995) reported that mean O3 in the previous 48 h was associated with a child’s report of cough or phlegm, while mean O3 in the previous day or week was not. For a panel of 71 asthmatic, 5–7-year-old children in Mexico City. respiratory symptoms (coughing, phlegm production, wheezing, and difficult breathing) and the frequency of lower respiratory illness on the same day were associated with both O3 and PM10 (Romieu et al., 1996). In a study of children with moderate-to-severe asthma in the Connecticut River Valley, where O3 exposures were much lower than those in Mexico city, Thurston et al. (1997) fount that respiratory symptoms were significantly associated with O3. Other epidemiology studies have provided evidence of qualitative associations between ambient oxidant levels >0.10 ppm and symptoms in children and young adults, such as throat irritation, chest discomfort, cough, and headache (Hammer et al., 1974; Makino and Mizoguchi, 1975). Thus, symptoms reported in individuals exposed to O3 in purified air are similar to those found in individuals exposed to ambient air except for eye irritation, a common symptom associated with exposure to photochemical oxidants, which has not been reported for controlled exposures to O3 alone. Other oxidants, such as aldehydes and peroxyacetyl nitrate (PAN), are primarily responsible for eye irritation and are generally found in atmospheres containing higher ambient O3 levels (Altshuller, 1977; National Research Council, 1977).

884

OZONE

There have been several studies reporting associations between ambient photochemical oxidant pollution and exacerbation of asthma (Holguin et al., 1985; Schoettlin and Landau, 1961; Whittemore and Korn, 1980), but the role of specifically O3 and the nature of the exposure–response relationships remain poorly defined. Respiratory symptoms in healthy young adult females (student nurses) in Los Angeles, in relation to ambient pollution levels, were monitored by Hammer et al. (1974). Schwartz and Zeger (1990) reexamined the original diaries from this study, which contained smoking and allergy histories as well as symptom reports that had never been analyzed. Diaries were compiled daily and collected weekly for as long as 3 years. Air pollution was measured at a monitoring location within 2.5 miles of the school. Incidence and duration of a system were modeled separately. Photochemical oxidants (74 ppb) were associated with increased risk of chest discomfort (odds ratio (OR) ¼ 1.17; p < 0.001) and eye irritation (OR ¼ 1.20; p < 0.001). Ostro et al. (1993) recorded the respiratory symptoms in nonsmoking adults residing in Southern California. Participants recorded the daily incidence of several respiratory symptoms over a 6 month period between 1978 and 1979. Ambient concentrations of O3, SO42 , and other air pollutants were measured. Using a logistic regression model, the authors found a significant association between the incidence of lower respiratory tract symptoms and 7 h O3 (OR ¼ 1.32; 95% confidence interval (CI): 1.14–1.52, for a 100 ppb change), and SO42 (OR ¼ 1.30; 95% CI: 1.09–1.54, for a 10 mg/m3 change), but no association was found with coefficient of haze, a more general measure of PM. The existence of a gas stove in the home was also associated with lower respiratory tract symptoms (OR ¼ 1.23; 95% CI: 1.03–1.47). The effects of O3 were greater in the subpopulation without a residential air conditioner. In addition, O3 had a greater effect on individuals with a preexisting respiratory infection. Desqueyroux et al. (2002) studied symptomatic responses to community air pollutants among patients with COPD. During a 14 month period, Parisian adults with severe COPD were monitored by their physicians. Daily levels of four air pollutants were provided by an urban air quality network. Exacerbation of COPD was associated only with O3 (OR ¼ 1.44 for a 5 ppb increase in O3; 95% CI: 1.14, 1.82), with a lag of 2–3 days. The effect of O3 was greater in patients whose CO2 pressure (PaCO2) was higher than 43 mmHg (OR ¼ 1.83; 95% CI: 1.36, 2.47). 23.3.3.1 Effects on Airway Reactivity Exposure to O3 can also alter the responsiveness of the airways to other bronchoconstrictive challenges as measured by changes in respiratory mechanics. For example, Folinsbee et al. (1988) reported that airway reactivity to the bronchoconstrictive drug methacholine for the group of subjects as a whole was approximately doubled following 6.6 h exposures to 120 ppb O3. Airway hyperresponsiveness (to histamine) had previously been demonstrated but only at O3 concentrations 400 ppb (Holtzman et al., 1979; Seltzer et al., 1986). On an individual basis, Folinsbee et al. (1988) found no apparent relationship between the O3-associated changes in methacholine reactivity and those in FVC or FEV1. On the other hand, Aris et al. (1991) reported a closer relationship, more similar to reported responses to inhaled H2SO4 aerosol, where changes in function correlated closely with changes in reactivity to carbachol aerosol, a bronchonconstrictive drug (Utell et al., 1983). The O3-associated changes in bronchial reactivity may predispose individuals to bronchospasm from other environmental agents such as acid aerosol and naturally occurring aeroallergens.

EFFECTS OF SHORT-TERM EXPOSURES TO OZONE IN HUMANS

885

Tests by Horstman et al. (1990), involving 6.6 h exposures to 80, 100, and 120 ppb, produced 56%, 89%, and 121% increases in methacholine responsiveness, respectively. Increased responsiveness to methacholine was also seen in the Folinsbee and Hazuch (1989) study with 1 h O3 exposure at 350 ppb. An increased responsiveness to histamine was seen by Gong et al. (1988) in one of 17 competitive cyclists exposed at 120 ppb for 1 h at VE of 89 L/ min followed by 3–4 min at 150 L/min. At 200 ppb, responsiveness increased in 9 of 17 subjects. McDonnell et al. (1987) found increased histamine responsiveness in 26 young adult males with allergic rhinitis after O3 at 180 ppb during 2 h of exercise at 64 L/min. Jorres et al. (1996) exposed 24 subjects with mild stable allergic asthma, 12 subjects with allergic rhinitis without asthma, and 10 healthy subjects to 250 ppb O3 or FA for 3 h with intermittent exercise. They determined the concentration of methacholine (PC20FEV1) and the dose of allergen (PD20FEV1) producing a 20% fall in FEV1. In subjects with asthma, FEV1 decreased by 12.5  2.2%, PC20FEV1 of methacholine by 0.91  0.19 doubling concentrations and PD20FEV1 of allergen by 1.74  0.25 doubling doses after O3 compared with sham exposure to FA. The changes in lung function, methacholine, and allergen responsiveness did not correlate with each other. In subjects with rhinitis, mean FEV1 decreased by 7.8% and 1.3% when O3 or FA, respectively, were followed by allergen inhalation. 23.3.3.2 Effects on Airway Permeability Kehrl et al. (1987) studied the effects of inhaled O3 on respiratory epithelial permeability in healthy, nonsmoking young men. They were exposed for 2 h to purified air and 400 ppb ozone while performing intermittent treadmill exercise at 67 L/min. Specific airway resistance (SRaw) and FVC were measured before and at the end of exposures. Seventy-five minutes after the exposures, the pulmonary clearance of a radioisotope-labeled organic molecule, that is, [99m Tc]DTPA, was measured as an index of epithelial permeability. O3 exposure caused respiratory symptoms in all eight subjects and was associated with a 14  2.8% (mean  SE) decrement in FVC (p < 0.001) and a 71  22% increase in SRaw (p ¼ 0.04). Compared to the air exposure day, seven of the eight subjects showed increased [99m Tc]DTPA clearance after the O3 exposure, with the mean value increasing from 0.59  0.08 to 1.75  0.43%/min (p ¼ 0.03). Thus, O3 exposure sufficient to produce decrements in the respiratory function of human subjects also causes an increase in permeability. An increased permeability could facilitate the uptake of other inhaled toxicants and/or the release of inflammatory cells such as neutrophils onto the airway surfaces. Foster and Stetkiewicz (1996) studied the influence of O3 on lung permeability in healthy subjects at 18–20 h after 2 h exposures at 150 and 350 ppb. Permeability was measured in terms of the clearance rate of a water-soluble aerosol containing 99m Tc-labeled DTPA (diethylamine pentaacetic acid). Based on a sequence of g-camera measurements of 99m Tc clearance from the lungs, they concluded that 99m Tc-DTPA clearance from the lung periphery and apexes was significantly increased by O3 but changes in clearance for the base of the lung were not significant. The FEV1 at the late time after O3 was slightly but significantly reduced ( 2.1%) from pre-exposure levels. There was no relationship between the functional changes observed acutely after exposure to O3 and subsequent changes in 99m Tc-DTPA clearance or FEV1 observed at the late period. These results suggest that epithelial permeability of the lung is altered 18–20 h post-O3; this injury is regional, and the lung base appears to have a different time course of response or is in an adapted state with respect to O3 exposure. 23.3.3.3 Effects on Airway Inflammation Seltzer et al. (1986) showed that O3-induced airway reactivity to methacholine is associated with neutrophil influx into the airways and

886

OZONE

with changes in cyclooxygenase metabolites of arachidonic acid. For 2 h exposures to O3 at 400 ppb with intermittent exercise, the bronchoalveolar lavage (BAL) fluid had increased prostaglandins E2 and F2a and thromboxane B2 3 h after the O3 exposure. Reports of Koren et al. (1989) and Devlin et al. (1991) also described inflammatory and biochemical changes in the airways following O3 exposure. In the initial studies, subjects were exposed to 400 ppb for 2 h while performing intermittent exercise at a ventilation of 70 L/min to examine cellular and biochemical responses in the airways. The BAL was performed 18 h after the O3 exposure. An 8.2-fold increase in polymorphonuclear leukocytes (PMNs or neutrophils) was observed after ozone exposure, confirming the observations of Seltzer et al. (1986). Twofold increases in protein, albumin, and IgG were indicative of increased epithelial permeability, as previously suggested by the [99m Tc]DTPA clearance studies of Kehrl et al. (1987). In addition to confirmation of the Kehrl et al. (1987) findings, Koren et al. (1989) provided evidence of stimulation of fibrogenic processes including increases in fibronectin (6.4), tissue factor (2.1), factor VII (1.8), and urokinase plasminogen activator (3.6). There was a twofold increase in the level of prostaglandin E2 and a similar elevation of the complement component C3a. Levels of leukotrienes C4 and B4 were not affected. Devlin et al. (1991) reported that a significant inflammatory response, as indicated by increased levels of PMN, was also observed in BAL fluid from subjects exposed to either 80 or 100 ppb O3 for 6.6 h. As illustrated in Fig. 23.8, the 6.6 h at 100 ppb O3 produced a 3.8-fold increase in PMNs at 18 h after the exposure, whereas the 6.6 h at 80 ppb produced a 2.1-fold increase. The amounts of O3 inhaled in the 80 and 100 ppb protocols were approximately 2.0 and 2.5 mg and approximately 3.6 mg in the 400 ppb protocol. Thus, the effect of concentration was apparently somewhat greater than that of exposure duration. The significant increase in PMNs at a concentration as low as 80 ppb suggests that lung inflammation from inhaled O3 has no threshold down to ambient background O3 levels. The inflammatory process caused by O3 exposure is promptly initiated (Seltzer et al., 1986) and persists for at least 18 h (Koren et al., 1989). The time-course of this inflammatory response, and the O3 exposures necessary to initiate it, however, has not yet been fully

FIGURE 23.8 Range of subject response 18 h after 6.6 h of O3 exposure at 100 ppb (closed circles) or 80 ppb (open circles). Squares indicate the mean changes (SE). From Devlin et al. (1991).

EFFECTS OF SHORT-TERM EXPOSURES TO OZONE IN HUMANS

887

elucidated. Furthermore, these studies demonstrate that cells and enzymes capable of causing damage to pulmonary tissues were increased, and the proteins that play a role in the fibrotic and fibrinolytic processes were elevated as a result of O3 exposure. Scannell et al. (1996) studied a group of asthmatic subjects exposed to O3 using the same exposure protocol previously used by the same investigators for 81 healthy subjects. They reported no significant differences in lung function responses and a trend toward higher airway resistance (p < 0.13). By contrast, the asthmatic subjects had significantly greater (p < 0.05) O3-induced increases in inflammatory end points (% neutrophils and total protein) in bronchoalveolar lavage fluid (BALF) as compared to 20 of the normal subjects who also underwent bronchoscopy. Prolonged inflammatory processes following repetitive exposures to O3 in ambient air were reported by Kinney et al. (1996) in terms of reduced release of reactive oxygen species, increased levels of LDH, IL-8, and PGE2 in the BAL. Interpretation of the nature and significance of the inflammatory responses following short-term O3 exposures is difficult without knowledge of the cumulative effects that may be triggered by repetitive episodes of lung inflammation. The relation of the inflammatory responses, if any, to the well-studied respiratory function responses also remains unknown. We do know that these responses are poorly correlated. Balmes et al. (1996) tested the hypothesis that changes in lung function induced by O3 are correlated with indices of respiratory tract/injury inflammation. They exposed healthy subjects, on separate days, to O3 (0.2 ppm) and filtered air for 4 h during exercise. Symptom questionnaires were administered before and after exposure, and pulmonary function tests (FEV1, FVC, and Sraw) were performed before, during, and immediately after each exposure. Fiberoptic bronchoscopy, with isolated left main bronchus proximal airway lavage (PAL) and bronchoalveolar lavage (bronchial fraction, the first 10 ml of fluid recovered) of the right middle lobe, was performed 18 h after each exposure. The PAL, bronchial fraction, and BAL fluids were analyzed for total and differential cell counts, total protein, fibronectin, interleukin-8 (IL-8), and granulocyte– macrophage colony-stimulating factor (GM-CSF) concentrations. The study population was divided into two groups, least sensitive (n ¼ 12; mean O3-induced change in FEV1 ¼ 7.0%) and most sensitive (n ¼ 8; mean O3-induced change in FEV1 ¼ 36.0%). They found a highly significant O3 effect on Sraw and lower respiratory symptoms for all subjects combined but no significant differences between the least and most sensitive groups. O3 exposure increased significantly percent neutrophils in PAL; percent neutrophils, total protein, and IL-8 in bronchial fraction (p < 0.001, p < 0.001, and p < 0.01, respectively); and percent neutrophils, total protein, fibronectin, and GM-CSF in BAL for all subjects combined; there were no significant differences, however, between least and most sensitive groups. Thus, levels of O3-induced symptoms and respiratory tract injury/inflammation were not correlated with the magnitude of decrements in FEV1 and FVC. A similar conclusion was drawn by Torres et al. (1997), who studied whether individuals who differed in lung function responsiveness to O3, or in smoking status, also differed in susceptibility to airway inflammation. Healthy subjects were selected on the basis of responsiveness to a classifying exposure to 220 ppb O3 for 4 h with exercise (responders, with a decrease in FEV1 > 15%; and nonresponders, with a decrease in FEV1 < 5%). Three groups were studied: nonsmoker-nonresponders (n ¼ 12), nonsmoker-responders (n ¼ 13), and smokers (n ¼ 13; 11 nonresponders and two responders). Each subject underwent two exposures to O3 and one to air, separated by at least 3 weeks; bronchoalveolar and nasal lavages were performed on three occasions: immediately (early) and 18 h (late) after O3 exposure, and either early or late after air exposure. Recovery of PMNs increased progressively in all groups, and by up to 6-fold late after O3 exposure. IL-6 and IL-8 increased early

888

OZONE

(by up to 10-fold and up to 2-fold, respectively), and correlated with the late increase in PMN. Lymphocytes, mast cells, and eosinophils also increased late after exposure. Thus, O3-induced airway inflammation was independent of smoking status or airway responsiveness to O3. Alexis et al. (2000) used indomethacin pretreatment prior to O3 exposure to investigate the role that cyclooxygenase (COX) metabolites of arachidonic acid might play. They reported that COX metabolites contribute to restrictive-type changes in normal and obstructive-type changes in small airways in asthmatic subjects. Grievink et al. (1999) reported that 100 mg of vitamin E and 500 mg of vitamin E provided partial protection against O3-related function decrements in adult Dutch cyclists with O3 concentrations up to 93 ppb during the exercise. For 9-year-old children with moderate-tosevere asthma in Mexico City, with 8 h average O3 concentrations up to 184 ppb, daily supplements of 50 mg of Vitamin E and 250 mg of Vitamin C modulated the pulmonary effects of O3 (Romieu et al., 2002). However, 2 weeks of pretreatment of inhaled budesonide (a corticosteroid) with 800 mg twice a day provided no protection against inhaled O3 in terms of either pulmonary function, methacholine reactivity, or neutrophil recruitment (Nightingale et al., 2000). Holz et al. (1999) reported that respiratory function and O3-induced airway inflammatory changes differed between individuals, both for healthy and asthmatic subjects, were reproducible but were not related to each other. Vagaggini et al. (2002) studied subjects with mild asthma, as indicated by a methacholine challenge. They exposed them to an allergen 24 h before inhaling O3. The O3 exposure increased the percentage of eosinophils but not of PMNs in induced sputum above that associated with the allergen challenge alone. Samet et al. (2001) studied the pulmonary effects of O3 on healthy adults with and without dietary supplementation of antioxidants and found that the antioxidants reduced the O3induced functional decrements but not its effect on increasing PMNs and IL-6 in lavage fluid. Inflammatory reactions occur in the nasal passages as well as in the lungs. Graham et al. (1988) exposed 41 subjects to either filtered air or 500 ppb O3 for 4 h for 2 consecutive days. Nasal lavages (NLs) were taken before and immediately after each exposure and 22 h after the last exposure. Lavage PMN counts increased significantly (p ¼ 0.005) in the O3-exposed group, with 3.5-, 6.5-, and 3.9-fold increases over the air-exposed group at the post-1, pre-2, and post-2 time points, respectively. Graham and Koren (1990) compared the cellular changes detected in NL with those detected in BAL taken from the same individual. Subjects were exposed to either filtered air or 400 ppb O3, with exercise, for 2 h. The NL was done prior to, immediately after, and 18 h postexposure; the BAL was done only at 18 h postexposure. A significant increase in PMNs was detected in the NL immediately postexposure to O3 (7.7-fold increase; p ¼ 0.003) and remained elevated in the 8 h postO3 NL (6.1-fold increase; p < 0.001). A similar increase in PMNs was detected in the BAL 18 h after exposure to O3 (6.0-fold increase; p < 0.001). The albumin levels in the NL and BAL were also similarly increased 18 h after O3 (3.9-fold and 2.2-fold, respectively). Although a qualitative correlation in the mean number of PMNs existed between the upper and lower respiratory tracts after O3, a comparison of the NL and BAL PMNs from each individual showed a significant quantitative correlation for the air data (r ¼ 0.741; p ¼ 0.014) but not for the O3 data (r ¼ 0.408; p ¼ 0.243). The utility of this approach at low ambient levels of O3 was demonstrated by Frischer et al. (1993). They studied nasal airways inflammation after O3 exposure in children by repeated NL from May to October 1991. During this period, five to eight NLs were performed on each child. On day 14 following “high” O3 (>90 ppb), 148 NLs were performed and on day 10 following “low” O3 (<70 ppb), 106 NLs were performed. A significant increase in

EFFECTS OF SHORT-TERM EXPOSURES TO OZONE IN HUMANS

889

PMN counts from low to high O3 was observed. Concomitant with a decrease in O3 in the fall, mean PMN counts showed a downward trend. Humoral markers of inflammation were also measured. A significant increase was observed for eosinophilic cationic protein and myeloperoxidase. Thus, O3 at ambient concentrations initiated a reversible inflammatory response of the upper airways in normal children. Peden et al. (1995) studied the role that O3 may play in the exacerbation of airway disease in asthmatics, either by priming the airway mucosa such that cellular responses to allergen are enhanced or by exerting an intrinsic effect on airway inflammation. The effect of exposure to 400 ppb O3 on nasal inflammation was examined in allergic asthmatics sensitive to Dermatophygoides farinae. This study design emphasized the effect of O3 exposure on the late-phase reaction to allergen using eosinophil influx and changes in eosinophil cationic protein as principal end points. By employing a “split-nose” design, in which allergen was applied to only one side of the nose while saline was applied to the contralateral side, both the effect of O3 on nasal inflammation due to allergen challenge and its direct action on nonallergen-challenged nasal tissues were examined. They found that O3 exposure had both a priming effect on allergen-induced responses and an intrinsic inflammatory action in the nasal airways. 23.3.4

Effect on Particle Clearance

Foster et al. (1987) studied the effect of 2 h exposures to 200 or 400 ppb O3 with intermittent light exercise on the rates of tracheobronchial mucociliary particle clearance in healthy adult males. The 400 ppb O3 exposure produced a marked acceleration in particle clearance from both central and peripheral airways, as well as a 12% drop in FVC. It is interesting to note that the 200 ppb O3 exposure produced a significant acceleration of particle clearance in peripheral airways but failed to produce a significant reduction in FVC, suggesting that significant changes in the ability of the deep lung to clear deposited particles take place before significant changes in respiratory function take place. 23.3.5

Effects on Aerosol Dispersion

In order to study the potential effects of O3 on small airways in humans, Keefe et al. (1991) employed a test of aerosol dispersion. Healthy nonsmoking male volunteers were exposed to 400 ppb O3 for 1 h while exercising at 20 L/min/m2body surface area (BSA). Prior to and immediately after exposure, tests of spirometry (FVC, FEV1, and FEF25-75) and plethysmography (Raw and sRaw) were performed. Subjects also performed an aerosol dispersion test before and after exposure. Each test involved a subject inhaling 5 to 7 breaths of a 300 mL bolus of a 0.5 mm triphenyl phosphate (TPP) aerosol injected into a 2 L tidal volume. The bolus was injected into the tidal breath at three different depths: at depth A after 1.6 L of clean air from functional residual capacity (FRC); at depth B after 1.2 L; and at depth C after 1.2 L but with inhalation beginning from residual volume (RV). The primary measure of bolus dispersion was the expired concentration half-width (HW). Changes in pulmonary function following ozone exposure were consistent with previous findings. When corrected for exercise, FVC, FEV1, and FEF25-75 all significantly declined (p < 0.001, p < 0.002, and p < 0.03, respectively) with nonsignificant increases in Raw and sRaw. The HW significantly increased following O3 exposure relative to air exposure at all depths (17 mL, p < 0.05 at depth A, 56 mL, p < 0.001 at depth B, and 53 mL, p < 0.005 at depth C). The HW was only weakly correlated with spirometric measures, accounting for less than 25% of the variance. Half-width was not correlated with Raw or sRaw. They concluded that the changes in aerosol

890

OZONE

dispersion seen with O3 exposure were related to changes in turbulent mixing and/or regional time constants in the small airways, thus suggesting a possible O3 effect in that region of the lung as well as effects in the larger airways that produce the respiratory function decrements. 23.3.6

Other Responses

Rivas-Arancibia et al. (1998) studied the effect of different O3 exposures on memory and correlated it with pulmonary and brain Cu/Zn superoxide dismutase (SOD) levels. Male Wistar rats were exposed for 4 h to O3 at 0, 100, 200, 500, or 1,000 ppb. Subsequently, they were tested in a passive avoidance conditioning protocol to measure short- and long-term memory. Motor activity was determined 1 and 24 h after O3 exposure. Cu/Zn SOD levels in the brain and pulmonary tissue were also measured. Rats exposed for 4 h to 200, 500, and 1,000 ppb O3 showed long-term memory deterioration and decreased motor activity, which was reversed 24 h later. Brain and pulmonary Cu/Zn SOD levels were increased in animals exposed to 100, 200, and 500 ppb O3 doses but decreased in animals exposed to 1,000 ppb O3. The results suggest that O3 exposure affects long-term memory possibly in association with oxidative stress. Gong et al. (1998) hypothesized that O3 exposure could acutely affect cardiovascular hemodynamics in humans and, in particular, in subjects with essential hypertension. They studied 10 nonmedicated hypertensive and 6 healthy male adults. Each subject, after catheterization of the right heart and a radial artery, was exposed to filtered air for 1 day and to 300 ppb O3 on the following day for 3 h with intermittent exercise. Relative to FA exposure, O3 exposure induced no statistically significant changes in cardiac index, ventricular performance, pulmonary artery pressure, pulmonary and systemic vascular resistances, ECG, serum cardiac enzymes, plasma catecholamines and atrial natriuretic factor, and SaO2. The overall results did not indicate major acute cardiovascular effects of O3 in either the hypertensive or the control subjects. However, mean pre-exposure to postexposure changes were significantly (p < 0.02) larger with O3 than with FA for rate– pressure product (1,353 beats/min/mmHg) and for heart rate (8 beats/min); these responses were not significantly different between the hypertensive and the control subjects. Significant O3 effects were also observed for mean FEV1 ( 6%), and AaPo2 (>10 mmHg increase), which were not significantly different between the two groups. These results suggest that O3 exposure can increase myocardial work and impair pulmonary gas exchange to a degree that might be clinically important in persons with significant pre-existing cardiovascular impairment, with or without concomitant lung disease. 23.4 FACTORS AFFECTING THE VARIABILITY OF RESPONSIVENESS IN HUMANS Although there is a great deal of knowledge about O3 exposure–respiratory function response in humans, as summarized above, we still know very little about the mechanisms responsible for the responses. Other irritants, such as SO2, NO2, and H2SO4, produce greater responses among asthmatics than among healthy human subjects but, as indicated previously, this is not true for O3. For other irritants, functional responses correlate with responsiveness to bronchoconstrictor challenge. For example, Utell et al. (1983) found a high correlation between reactivity to inhaled carbachol and responsiveness to inhaled H2SO4 in asthmatics (r ¼ 0.90, p< 0.001), whereas Horstman et al. (1986) reported that methacholine (MCh) reactivity and SO2 response were significantly but weakly correlated

FACTORS AFFECTING THE VARIABILITY OF RESPONSIVENESS IN HUMANS

891

(r ¼ 0.31). Although both functional decrements and bronchial responsiveness are produced by O3 exposure, Folinsbee et al. (1988) and Horstman et al. (1990) found no apparent relationship between these responses for individual subjects. On the other hand, Aris et al. (1991) screened healthy, nonsmoking volunteers for their functional responsiveness to 3 h of exposure to 200 ppb O3 at a ventilatory rate of 40 L/min and found that the MCh responsiveness of 10 O3-sensitive subjects (PC100 ¼ 3.0  0.8) was significantly greater than that of 10 O3-nonsensitive subjects (PC100 ¼ 18.7  4.5). Beckett et al. (1985) examined the effect of atropine, a muscarinic receptor blocker, on responses to exposure to 400 ppb O3. Atropine pretreatment prevented the significant increase in airway resistance with O3 exposure and partially blocked the decrease in forced expiratory flow rates, but it did not prevent a significant fall in FVC, changes in respiratory frequency, and tidal volume, or the frequency of reported respiratory symptoms. These results suggest that the increase in pulmonary resistance during O3 exposure is mediated by a parasympathetic mechanism and that changes in other measured variables are mediated, at least partially, by mechanisms not dependent on muscarinic cholinergic receptors of the parasympathetic nervous system. Gong et al. (1988) studied the contribution of b-adrenergic mechanisms to the acute airway responses to O3 in a study in which symptoms, pulmonary function, exercise performance, and postexposure histamine bronchoprovocation were studied in nonasthmatic athletes exposed to 210 ppb O3 during heavy continuous exercise, with mean minute ventilation (VE) 80 L/min for 60 min, followed by a maximal sprint (peak VE > 140 L/min) until exhaustion. Each subject was exposed randomly to either 210 ppb O3 or filtered air during the four single-blinded exposure sessions. Albuterol pretreatment resulted in modest but significant bronchodilation compared to placebo. However, albuterol did not prevent O3induced respiratory symptoms, decrements in FVC, FEV1, and maximum midexpiratory flow rate (FEF25–75), and positive histamine challenges compared to that with placebo and O3. There were statistically no significant differences in the metabolic data or ride times across all drugs and exposures, although the peak VE was significantly lower with O3 than FA regardless of drug. The results indicate that acute pretreatment with inhaled albuterol is unable to prevent or ameliorate O3-induced symptoms and alterations in pulmonary function and exercise performance. The contribution of b-adrenergic mechanisms in the acute airway responses to O3 appears to be minimal. In their study on bronchial hyperresponsiveness to O3 exposure, Seltzer et al. (1986) found significant increases in the concentration of prostaglandins E2 and F2a and thromboxane B2 in bronchoalveolar lavage fluid. Prostaglandins E2 and F2a stimulate pulmonary neural afferents that initiate several responses characteristic of acute O3 exposure (Coleridge et al., 1976; Roberts et al., 1985), suggesting that the release of prostaglandins in the lung may be involved in routinely observed pulmonary function decrements and perhaps in altered exercise ventilatory pattern and reported subjective symptomotology. Schelegle et al. (1987) studied whether O3-induced pulmonary function decrements could be inhibited by the prostaglandin synthetase inhibitor, indomethacin, in healthy human subjects. College-age males completed six 1 h exposure protocols with workloads set to elicit a VE of 60 L/min, with no drug, placebo, and indomethacin pretreatments, with filtered air and O3 (350 ppb) exposures within each pretreatment. Exposures consisted of 1 h exercise on a bicycle ergometer. Significant differences were found for comparisons of no drug versus indomethacin and of placebo versus indomethacin, suggesting that cyclooxygenase products of arachidonic acid, which are sensitive to indomethacin inhibition, play a

892

OZONE

prominent role in the development of pulmonary function decrements consequent to acute O3 exposure. In a similar study, Ying et al. (1990) administered indomethacin for 4 days to young adult male nonsmokers prior to 2 h O3 exposures with intermittent exercise at 400 ppb to determine if it would alter their O3 responsiveness as well as their lung function. For subjects who had detectable serum levels of indomethacin and significant responses to methacholine on the sham exposure day, the indomethacin attenuated the O3-induced decrements in lung function but did not attenuate the O3-induced responsiveness to methacholine. They concluded not only that the O3-induced decrements in respiratory function are mediated by cyclooxygenase products but also that the O3-induced increase in airway reactivity occurs by some other mechanism. The mechanism by which the release of cyclooxygenase products in the lung leads to pulmonary function decrements in humans upon O3 exposure remains undefined. Available data indicate that O3-induced pulmonary function decrements and ventilatory pattern changes are neurally mediated (Lee et al., 1979; Hazucha et al., 1989). Hazucha et al. (1989) concluded that O3 inhalation stimulates airway receptors, which leads to an involuntary inhibition of full inspiration, reduction in FVC, and a concomitant decrease in maximal expiratory flow rates in humans. The observation that cyclooxygenase products stimulate neural afferents in the lung (Coleridge et al., 1976; Roberts et al., 1985), combined with the observation of reduced O3-induced pulmonary function decrements after indomethacin pretreatment, suggests that cyclooxygenase products released consequent to O3induced tissue damage stimulate neural afferents in the lung, which results in the observed pulmonary function decrements.

23.5 STUDIES OF POPULATIONS EXPOSED TO OZONE IN AMBIENT AIR Observational studies of the influence of O3 on human health are often difficult to interpret because the population is also exposed to other pollutants in the ambient air that could affect the responses observed or to other environmental challenges that may produce comparable effects, such as environmental tobacco smoke, other pollutants in indoor air, and allergens found in indoor and outdoor air. For time-series studies of daily mortality and admissions to emergency departments, hospital admissions, and other health service providers, appropriate corrections need to be made for ambient temperature, which can covary with both O3 concentrations and health effect indices. 23.5.1

Mortality

Recent studies examining the possible influence of O3 on daily mortality have reported independent effects of O3 in multiple regression analyses. Sartor et al. (1995, 1997) found that O3 affected mortality during the summer of 1994 both for the individuals of all age groups and for the elderly in Belgium and that temperature potentiated the response to O3. Verhoeff et al. (1996) examined daily mortality in Amsterdam, The Netherlands, for the period of 1986–1992 and reported that O3 with a 2-day lag was positively associated with mortality, as were current day black smoke (BS) and PM10. There was no association with SO2 or CO. Anderson et al. (1996) studied air pollution and daily mortality in London, England, during 1987–1992. They reported that both same-day O3 and BS were independently associated with all causes of mortality, which was greater on warm days, and

STUDIES OF POPULATIONS EXPOSED TO OZONE IN AMBIENT AIR

893

independent of the effects of other pollutants. O3 was also significantly associated with mortality caused by cardiovascular and respiratory diseases. Touloumi et al. (1997) performed a combined analysis of daily mortality for six western and central European cities participating in the project Air Pollution and Health: A European Approach (APHEA). They reported that a 25 ppb increase in daily 1 h max. O3 concentration was associated with a 2.9% increase in the number of deaths, and the effect was independent of the BS concentration change and consistent across the cities. Thurston and Ito (2001) re-examined the data from a number of earlier time-series mortality studies that had not adequately corrected for ambient temperature. For all of the total mortality–air pollution time-series studies considered, the combined analysis yielded a relative risk (RR) ¼ 1.036 per 100 ppb increase in daily 1 h maximum O3 (95% CI: 1.023– 1.050). However, the subset of studies that specified the nonlinear nature of the temperature– mortality association yielded a combined estimate of RR ¼ 1.056 per 100 ppb (95% CI: 1.032–1.081) indicating that past time-series studies using linear temperature–mortality specifications underpredicted the mortality effects of O3 air pollution. For Detroit, MI, an illustrative analysis of daily total mortality during 1986–1990 also indicated that the model weather specification choice could influence the O3 health effect estimates. Results were intercompared for alternative weather specifications. Nonlinear specifications of temperature and relative humidity (RH) yielded lower intercorrelations with the O3 coefficient, and larger O3 RR estimates, than a base model employing a simple linear spline of hot and cold temperature. They concluded that, unlike for PM mass, the mortality effect estimates derived by time-series analyses for O3 can be sensitive to the way weather is addressed in the model. Generally, they found that the O3-mortality effect estimate increased in size and statistical significance when the nonlinearity and the humidity interaction of the temperature–health effect association were incorporated into the model weather specification. In recent years, many studies have been focused on the associations between short-term O3 exposures and daily mortality rates in urban centers. The National Mortality and Morbidity Air Pollution Study (NMMAPS) used EPA’s Atmospheric Information Retrieval System (AIRS) data on ambient O3 from 95 United States communities and publicly available daily mortality data in a preselected analytical model. As shown in Fig. 23.9, a positive association was found in all but 2 communities, and a statistically significant association was shown for 7 communities and for 95 communities as a whole (U.S. EPA, 2006). As shown in Fig. 23.10, the 95-community effect was strongest on the same day, and highly significant on 1- and 2-day lags, as well as being even stronger when the distributed lag over 6 days was considered. WHO-EURO commissioned a meta-analysis of time-series studies of the associations between ambient O3 and daily mortality in more than 80 studies published between 1996 and 2001. The results are summarized in Table 23.3. The relationship between acute effects of O3 on mortality was reinforced by the recent publication of four meta-analyses (Bell et al., 2005; Gryparis et al., 2005; Ito, 2005; Levy et al., 2005) that were coherent in showing a significant association between O3 and short-term mortality, which is not confounded by other pollutants (including particulate matter), temperature, weather, season and strategy of modeling. Increases in total mortality have been observed in a concentration as low as 75 mg/m3 (1 h mean) (Gryparis et al., 2005). While O3 does appear to have a significant impact on daily mortality rates, especially in the warmer months of the year, its average long-term concentration has not been found to have a significant influence on annual mortality rate. Pope et al. (2002) found fine particles to be associated with significant increase in cardiovascular and lung cancer mortality.

894

OZONE

FIGURE 23.9 Bayesian city-specific and national average estimates for the percent change (95% CI) in daily mortality per 10 ppb increase in 24 h average. O3 in the previous week using a constrained distributed lag model for 95 U.S. communities (NMMAPS), arranged by size of the effect estimate. Source: EPA O3 PM. Data derived from Bell et al. (2004).

23.5.2

Morbidity

Associations between ambient air pollutants and respiratory morbidity were examined by Ostro and Rothschild (1989) using the Health Interview Survey (HIS), a large cross-sectional database collected by the National Center for Health Statistics. They attempted to determine the separate health consequences of O3 and particulate matter using six separate years of the HIS. The results, using a fixed effects model that controls for intercity differences, indicate an association between fine PM and both minor restrictions in activity and respiratory

FIGURE 23.10 Comparison of single-day lags (0-, 1-, 2, and 3-day) to a cumulative multiday lag (0–6 day) for percent changes in all cause mortality per 20 ppb increase in 24 year average O3 in all ages. Source: EPA O3CD, (2005). Data derived from Bell et al. (2004).

STUDIES OF POPULATIONS EXPOSED TO OZONE IN AMBIENT AIR

895

TABLE 23.3 Summary of Meta-Analysis of Time-Series Studies Published During the Period 1996–2001

RR < 1

Random Effects Summary Coefficientc

No. of Studies Cause Mortality

Season All

Lag 0

Asthma admissions in children

RR > 1 a

Summer

Any

All

Selected

1h 8h 24 h 1h 24 h 1h 8h 24 h 1h

Selected

8h 24 h 1h

10 (6a) 6 (2a) 3 (0)

1 (0) 1 (0) 1 (0)

0.7 (0.3–1.0) 0.6 (0.2–1.0) 0.1 ( 0.4, 0.6)

8h

4 (2a)

3 (2a)

0.1 ( 1.2, 1.3)

Selectedb

Hospital admissions respiratory

Timing

All

13 (8 ) 9 (5a) 8 (5a) 17 (13a) 22 (8a) 6 (5a) 6 (5a) 2 (2a) 4 (2a)

3 3 3 3 3 1 1 0 1

(0) (0) (1a) (0) (0) (0) (0) (0)

0.2 (0.1–0.3) 0.4 (0.2–0.5) 0.4 (0.1–0.6) 0.3 (0.2–0.4) 0.4 (0.3–0.6) 0.4 (0.1–0.6) 0.6 (0.3–0.9) – 0.5 (0.1–1.0)

a

Number of single studies with a p < 0.05. “selected” lag ¼ If results for more than one lag were presented, the lag selected was chosen as lag focused on by the author, most statistically significant, or largest estimate. c Percentage change per 10 mg/m3 increase and (95% CI), preliminary results. b

conditions severe enough to result in work loss and bed disability in adults. Ozone, however, was associated only with the more minor restrictions. Bates and Sizto (1989) examined associations between ambient air pollutants and hospital admissions for respiratory disease in Southern Ontario. They found a consistent association in summer between hospital admissions for respiratory disease and daily levels of SO42 , O3, and temperature but no association for a group of nonrespiratory conditions. Multiple regression analyses showed that all environmental variables together accounted for 5.6% of the variability in respiratory admissions and that when temperature was forced into the analysis first, it accounted for only 0.89% of the variability. It was found that daily SO42 data collected at one monitoring site in the center of the region were not correlated with respiratory admissions, whereas the SO42 values collected every 6th day, on different days of the week, at 17 stations in the region had the highest correlation with respiratory admissions. They concluded that probably neither O3 nor SO42 alone is responsible for the observed associations with acute respiratory admissions but that either some unmeasured species or some pattern of sequential or cumulative exposure was responsible for the observed morbidity. Burnett et al. (1994) also employed the Ontario acute care hospital database to analyze the effects of air pollution on hospital admissions, but their analysis considered all of Ontario and analyzed the data from each individual hospital, rather than aggregating the counts by region. Slow moving temporal cycles, including seasonal and yearly effects, were removed

896

OZONE

and day-of-week effects were controlled prior to the analysis. Poisson regression techniques were employed because of the low daily admission counts at individual hospitals. O3 displayed a positive association with respiratory admissions in 91% of the 168 hospitals, and 5% of summertime (May through August) respiratory admissions (mean ¼ 107/day) were attributed to O3 (mean ¼ 50 ppb). Positive associations were found in all age groups (0–1, 2– 34, 35–64, and 65þ). A parallel analysis of nonrespiratory admissions showed no such associations. Thurston et al. (1994) focused their analysis of respiratory hospital admissions in the Toronto metropolitan area during the summers (July through August) between 1986 and 1988, when they directly monitored for strong particulate acidity (Hþ) pollution on a daily basis at several sites in that city. Long-wave cycles, and their associated autocorrelations, were removed. Strong and significant positive associations with asthma and respiratory admissions were found for both O3 and Hþ, and somewhat weaker significant associations with SO42 , PM2.5, PM10, and TSP, as measured at a central site in downtown Toronto. No such associations were found either for SO2 or NO2 or for any pollutant with nonrespiratory control admissions. Temperature was only weakly correlated with respiratory admissions and became nonsignificant when entered in regressions with air pollution indices. Simultaneous regressions and sensitivity analyses indicated that O3 was the summertime haze constituent of greatest importance to respiratory and asthma admissions, although elevated Hþ was suggested as a possible potentiator of this effect. During multipollutant, simultaneous regressions on admissions, O3 was consistently the most significant. Of the PM metrics, only Hþ remained statistically significant when entered into the admission regressions simultaneously with O3. Sensitivity analyses also showed that dropping all days with 1 h O3 above 120 ppb (2 of a total 117 days) did not significantly change the O3 coefficients. The simultaneous O3, Hþ, and temperature model indicated that 21  8% of all respiratory admissions during the three summers were associated with O3 air pollution, on average, and that admissions rose an estimated 37  15% above that otherwise expected on the highest O3 day (159 ppb). Moreover, despite differing health care systems, the Toronto regression results for the summer of 1988 were remarkably consistent with previously reported results for that same summer in Buffalo, NY (Thurston et al., 1992). Delfino et al. (1994a) studied daily urgent hospital admissions for respiratory and other illnesses at 31 hospitals in Montreal, Canada, during the warm periods of the year between 1984 and 1988. Both 1 h and 8 h maximum O3 concentrations were considered in the analyses, as well as weather variables (temperature and relative humidity) and PM measurements (Delfino et al., (1994b). For the months of July and August, a significant association was found between all respiratory admissions and both 8 h daily maximum O3 (p  0.01) and 1 h daily maximum O3 (p  0.03) 4 days prior to admission, despite the low O3 concentrations (90th percentile ¼ 60 ppb O3). No significant correlations were found between O3 and nonrespiratory, control admissions. Lipfert and Hammerstrom (1992) reanalyzed the Bates and Sizto (1989) hospital admissions data set for 79 acute-care hospitals in Southern Ontario, incorporating more elaborate statistical methods and extending the data set through 1985. O3, SO42 , and SO2 had significant effects on hospital admissions. By contrast, pollution associations with hospital admissions for accidental causes were nonsignificant in these models. The pollutant mean effect accounted for 19–24% of all summer respiratory admissions. Burnett et al. (1997) extended their study of the effects of O3 on hospitalization for respiratory disease to 16 cities across Canada representing 12.6 million people from 1981 to 1991. There were 720,519 admissions for which the principal diagnosis was a respiratory

STUDIES OF POPULATIONS EXPOSED TO OZONE IN AMBIENT AIR

897

disease. After controlling for SO2, NO2, CO, soiling index, and dew point, the daily high hour concentration of O3 recorded 1 day previous to the date of admission was positively associated with respiratory admissions in the April–December period but not in the winter months. The association between O3 and respiratory hospitalizations varied among cities, with relative risks ranging from 1.000 to 1.088 after simultaneous covariate adjustment. PM and CO were also positively associated with respiratory hospitalizations. Thurston et al. (1992) analyzed admissions to acute-care hospitals in three New York State metropolitan areas during the summers of 1988 and 1989. Environmental variables considered included daily 1 h maximum O3 and 24 h average SO42 and Hþ concentrations, as well as daily maximum temperature recorded at central sites in each community. The strongest O3-respiratory admission associations were found during the period of high pollution in the summer of 1988 and in the most urbanized communities considered (i.e., Buffalo and New York City). After controlling for temperature effects via simultaneous regression, the summer haze pollutants (i.e., SO42 , Hþ, O3) remained significantly related to total respiratory and asthma admissions, but high intercorrelation prevented the clear discrimination of a single pollutant as the causal agent. Depending on the index pollutant, the admission category, and the city considered, it was found that summer haze pollutants accounted for approximately 5–20% of June through August total respiratory and asthma admissions, on average, and that these admissions increased approximately by 30% above average on the highest pollution days. White et al. (1994) reported daily emergency room visit records from June through August 1990 at a large inner city hospital in Atlanta, GA. Daily counts of visits for asthma or reactive airways disease by patients 1–16 years of age (mean ¼ 6.6/day) were related to daily levels of O3, SO2, PM10, pollen, and temperature. The model yielded a 1.42 admissions rate ratio (p ¼ 0.057, 95% CI: 0.99 to 2.0) for the number of asthma visits following days with O3 levels equal to or exceeding a 1 h maximum of 0.11 ppm, which is consistent with the relative risk values reported by Thurston et al. (1992, 1994). In a study of Birmingham, AL, data, Schwartz (1994a) separately examined O3 and PM10 influences on hospital admissions of the elderly for pneumonia (mean ¼ 5.9/day) and COPD (mean ¼ 2.2/day) causes from 1986 to 1989. Base model results (excluding winter months) yielded a 2-day lag RR estimate of 1.14 for pneumonia admissions from a 50 ppb increase in 24 h average O3 (95% CI: 0.94–1.38). Excluding days exceeding 120 ppb yielded similar results (RR ¼ 1.12; CI: 0.92–1.37). For COPD, the basic model yielded RR ¼ 1.17 (CI: 0.86–1.60), whereas excluding days above 120 ppb similarly gave RR ¼ 1.18 (CI: 0.86–1.62). Schwartz (1994b) analyzed O3 and PM10 air pollution relationships with daily hospital admissions of 65-year-old or older persons in the Detroit, MI, metropolitan statistical area from 1986 to 1989. Daily counts for pneumonia (mean ¼ 15.7/day), asthma (mean ¼ 0.75/ day), and all other COPDs (mean ¼ 5.8/day) were regressed on the pollution variables. O3 was analyzed with respect to both its daily 24 h average and 1 h maximum. Both O3 and PM10 were significant in simultaneous pollutant models for pneumonia and COPD but not for asthma (which was ascribed to the low daily counts for this category). According to the regression coefficients and data presented, the mean effect for O3 (11.6%) was double that for PM10 (5.7%) in the pneumonia model, but comparable for COPD (12.2% for O3 versus 10.2% for PM10). Schwartz (1994c) evaluated the associations of both PM10 and O3 with respiratory hospital admissions of the elderly in Minneapolis-St. Paul, MN, from 1986 to 1989. Although no association was found for COPD in the elderly, O3 did make a significant

898

OZONE

independent contribution to hospital admissions of the elderly for pneumonia (mean ¼ 6.0/ day), even after controlling for weather and PM10. In summary, the Schwartz studies of the elderly suggest that a large portion of the O3 effects on total respiratory hospital admissions is contributed by COPD and pneumonia cases in the elderly. According to the results presented by Thurston et al. (1992, 1994, the other major contributor is asthma admissions, which are usually more prevalent in younger age groups. The results of a WHO meta-analysis of hospital admission studies are summarized in Table 23.3. They indicate that ambient O3 often has a significant effect on hospital admissions for respiratory causes (WHO, 2003). A variety of recent population studies have analyzed associations between ambient O3 and emergency room (ER) admissions. Cody et al. (1992) analyzed central New Jersey hospital ER visits to the high O3 season (May through August). For simultaneous regression of respiratory visits on both temperature and O3, there was a significant positive coefficient for O3 and a negative coefficient for temperature. Day-of-week influences were considered but found to be unimportant for these ER visit data. Weisel et al. (1995) examined central New Jersey hospital ER visits for asthma (mean ¼ 5.4/day) during the high O3 season (May through August) for 1986 through 1990. Using a stepwise regression analysis, a significant positive coefficient for O3 and a negative coefficient for morning temperature were found. Other environmental factors considered were not found to be correlated with asthma visits. Stieb et al. (1996) examined the relationship of asthma ER visits to daily concentrations of O3 and other air pollutants in Saint John, New Brunswick, Canada. Data on ER visits with a presenting complaint of asthma (n ¼ 1987) were abstracted for the period 1984– 1992 (May–September). Air pollution variables included O3, SO2, NO2, SO42 , and TSP; weather variables included temperature, dewpoint, and relative humidity. The mean daily 1 h O3 max. concentration during the study period was 41.6 ppb. A positive, statistically significant (p < 0.05), association was observed between O3 and asthma ER visits 2 days later, and the strength of the association was greater in nonlinear models. The frequency of asthma ER visits was 33% higher (95% CI: 10–56%) when the daily 1 h O3 max. concentration exceeded 75 ppb (the 95th percentile). The O3 effect was not significantly influenced by the addition of weather or other pollutant variables into the model or by the exclusion of repeat ER visits. Yang et al. (1997) examined the association between air pollution and the ER visits for asthma in Reno, Nevada, for the period 1992–1994. All three hospitals in the region were included, and there were a total of 1593 ER visits for asthma during this period. The air pollution variables were collected from seven monitoring stations, including PM10, O3, and CO. Levels of pollution were moderately elevated (the average concentrations of PM10, CO, and O3 were 38.0 mg/m3, 4.55 ppm, and 51.0 ppb, respectively). Weighted least-square (WLS) regression and autoregressive integrated moving average (ARIMA) time-series analyses were applied and compared. The daily 1 h maximum O3 concentration was a significant predictor of asthma ER visits. Total asthma visits increased by 33.7% (95% CI: range 6.0–61.5%) for each 100 ppb increase in the O3 level. No association of the concentration of other measured pollutants with daily asthma ER visits was found. Another index of respiratory morbidity that has been studied is clinic visits. HernandezGarduno et al. (1997) monitored patient visits for upper respiratory tract infections in Mexico City at five clinics, and collected data on levels of O3, NO2, CO, SO2, and climatological

STUDIES OF POPULATIONS EXPOSED TO OZONE IN AMBIENT AIR

899

variables. Correlations of filtered data revealed an association between NO2 and O3 and an increase in visits to clinics because of respiratory problems. Autoregressive analysis indicated that pollutant levels/respiratory visits associations remained significant even after simultaneous inclusion of temperature, suggesting that air pollution was associated with 10 to 16% of the clinic visits. High levels of O3 and NO2 increased the total number of clinic visits to 19–43% above average. The other pollutants and the control group did not demonstrate significant associations. Overall, these results are consistent with an O3 effect on asthma morbidity. The results of the recent WHO-EURO meta-analysis on associations between O3 and hospital admissions for respiratory diseases are summarized in Table 23.3. A number of epidemiological studies have shown a consistent relationship between daily variations in ambient oxidant exposure and acute respiratory morbidity in the population. Decreased lung function and increased respiratory symptoms, including exacerbation of asthma, occur with increasing ambient O3, especially in children. Table 23.3 summarizes studies of hospital admissions for asthma in children from the WHO-EURO meta-analysis. In an analysis of respiratory hospital admissions in 14 Canadian cities, Burnett et al. (2001) should that the effect was greatest at a 1- or 2-day lag, but greatest of all for a distributed lag over 4 days. Modifying factors, such as ambient temperature, aeroallergens, and other copollutants (e.g., particles) also can contribute to this relationship. Ozone air pollution can account for a portion of summertime hospital admissions and emergency room visits for respiratory causes. It has been estimated from these studies that O3 may account for roughly 1–3 excess summertime respiratory hospital admissions per 100 ppb O3, per million persons. A recent study by Yang et al. (2003) reported significant associations between O3 respiratory hospital admissions for children less than 3 years of age and for the elderly in Vancouver, Canada, where the 24 h average O3 concentration was only 13.4 ppb. In the Children’s Health Study (CHS), O3 was significantly associated with bronchitic symptoms in children with asthma, but the effects were more strongly associated with organic carbon (OC) and NO2 than with O3 (McConnell et al., 2003). One effect of O3 in the CHS that was not influenced by the other measured pollutants was an increase in absence from school due to respiratory illnesses (Gilliland et al., 2001). Similar effects were seen in a study of asthmatic children in seven United States communities, that is, a strong association of school absence associated most strongly with O3, and respiratory symptoms more strongly associated with pollutants more closely associated with motor vehicle exhaust (O’Connor et al., 2006). Another effect observed in the CHS that was more closely associated with O3 than the other measured pollutants was incident cases of asthma, particularly in children who were engaged in three or more team sports (McConnell et al., 2002). Incident asthma in association with chronic O3 exposure has also been reported in adult males in the AHSMOG cohort study of Seventh Day Adventists in California (McDonnell et al., 1999). The overall implications of the CHS studies on public health and cost–benefit considerations for air pollution control have been reviewed by Kunzli et al. (2003). Another index of morbidity of respiratory morbidity in asthmatics is physicianauthorized medication usage. In their study of children with moderate-to-severe asthma at a summer camp in the Connecticut River Valley, Thurston et al. (2001) found that the camp physician authorized supplemental medication to children in the group at a rate proportional to the ambient O3 concentration.

900

OZONE

23.6 EFFECTS OBSERVED IN STUDIES IN LABORATORY ANIMALS 23.6.1

Effects on Athletic Performance

There are animal models for decreased performance during O3 exposure. Tepper et al. (1985) exposed rats and mice for 6 h to O3 at 80, 120, 250, or 500 ppb while housed in running wheels. Running in both species decreased in a concentration-related manner during exposure to O3, with the decrease being greater with increasing time of exposure. The decrease in running activity produced by O3 persisted for several hours after exposure. At comparable concentrations, activity in rats decreased more than in mice. 23.6.2

Effects on Airways Reactivity

The basis for the effect of O3 on airways reactivity was examined by Gordon et al. (1981) in guinea pigs exposed for 1 h to either 100 or 800 ppb O3. Both exposures significantly inhibited lung cholinesterase activity compared to levels in unexposed animals. The O3-induced responsiveness may be centered on the peripheral lung and be retained long after the O3 exposure ceases, according to a study by Beckett et al. (1988). They exposed the peripheral lungs of anesthetized dogs to 1000 ppb O3 for 2 h using a wedged bronchoscope technique. A contralateral sublobar segment was simultaneously exposed to air as a control. In the O3-exposed segments, collateral resistance (Rcs) was increased within 15 min and remained elevated approximately 150% throughout the 2 h exposure period. Fifteen hours later, the baseline Rcs of the O3-exposed sublobar segments was significantly elevated, and these segments demonstrated increased responsiveness to aerosolized acetylcholine (100 and 500 mg/mL). There were no differences in neutrophils, mononuclear cells, or mast cells (numbers or degree of mast cell degranulation) between O3- and air-exposed airways at 15 h. The small airways of the lung periphery thus are capable of remaining hyperresponsive hours after cessation of localized exposure to O3, but this does not appear to depend on the presence of inflammatory cells in the small airway wall. 23.6.3

Effects on Airway Permeability in Laboratory Tests in Rats

Bhalla et al. (1987) reported that exposure to resting rats to 800 ppb O3 increased tracheal and bronchoalveolar permeability to DTPA at 1 h after the exposure. Bronchoalveolar but not tracheal permeability remained elevated 24 h after the exposure. Exercise during exposure to O3 increased permeability to both tracers in the tracheal and the bronchoalveolar zones and prolonged the duration of increased permeability in the tracheal zone from 1 to 24 h and that in the bronchoalveolar zone from 24 to 48 h. Exposure at rest to 600 ppb O3 plus 2500 ppb NO2 significantly increased bronchoalveolar permeability at 1 and 24 h after exposure, although exposure at rest to 600 ppb O3 alone increased bronchoalveolar permeability only at 1 h after exposure. Exposure to O3 and NO2 during exercise led to significantly greater permeability to DTPA than did exercising exposure to O3 alone. Nitric acid vapor was formed in the O3 þ NO2 atmosphere, suggesting that acidic components in the atmospheres produced effects that were additive on the effect of O3 in producing both increase and prolongation of permeability in tracheal and bronchoalveolar zones of the respiratory tract. Guth et al. (1986) examined changes in apparent lung permeability in rats by measuring the recovery of labeled bovine serum albumin in lung lavage fluid after intravenous injection at the end of the O3 exposure. Their permeability index increased in an exposure

DETERMINANTS OF RESPONSIVENESS TO OZONE EXPOSURES IN ANIMAL STUDIES

901

concentration-dependent manner after 6 or 24 h of exposure to O3 at or above levels of 400 ppb. It was also increased after 2 days of exposure to 200 ppb of O3. Abraham et al. (1984) measured changes in airway permeability of tritiated histamine in sheep after exposure to O3. Measurements made 24 h after a 2 h exposure to 500 ppb showed increased permeability. This persistently increased permeability is consistent with the observations of Bhalla et al. (1987) in rats.

23.6.4

Effects on Airway Inflammation

Arsalane et al. (1999) evaluated Clara cell protein (CC16), a 16–17 kDa protein secreted by Clara cells in the bronchial lining fluid of the lung, as a peripheral marker of the integrity of Clara cells and/or of the bronchoalveolar/blood barrier. CC16 was determined in the serum of rats after a single 3 h exposure to 300, 600, or 1,000 ppb of O3. The urinary excretion of the protein was also studied in rats repeatedly exposed to 1,000 ppb O3, 3 h/day, for up to 10 days. The concentrations of CC16 in the lung or trachea homogenates, the lung CC16 mRNA levels, and classical markers of lung injury in BALF were also determined. O3 produced a transient increase of CC16 concentration in serum that reached values that were, on average, 13 times above normal values 2 h after exposure to 1,000 ppb O3. The intravascular leakage of CC16 was dose-dependent and correlated with the extent of lung injury as assessed by the levels of total protein, LDH, and inflammatory cells in BALF. This effect was most likely responsible for the concomitant marked reduction of CC16 concentrations in BALF and lung homogenate, since the CC16 mRNA levels in the lungs were unchanged, and the absolute amounts of CC16 leaking into serum or lost from the respiratory tract were similar. These changes were paralleled by an elevation of the urinary excretion of CC16 resulting from an overloading of the tubular reabsorption process. These results demonstrated the utility of this assay to detect the increased lung epithelial permeability induced by O3. Broeckaert et al. (2000) applied this assay to humans, specifically to cyclists who exercised for 2 h during episodes of photochemical smog, and found that O3 induces an early leakage of lung Clara cell protein. The protein levels increased significantly into the serum from exposure levels as low as 60–84 ppb. These findings confirmed that there is almost no safety margin for the effects of ambient O3 on airway permeability. The assay of CC16 in the serum represents a sensitive noninvasive test allowing the detection of early effects of ambient O3 on the lung epithelial barrier. 23.7 DETERMINANTS OF RESPONSIVENESS TO OZONE EXPOSURES IN ANIMAL STUDIES There is evidently a large genetic component to this responsiveness. Kleeberger (1995) and Kleeberger et al. (1997) have explored the contribution of genetic background to the pathogenesis of airway responses to air pollutants. Using inbred mice strains, Kleeberger (1995) demonstrated a strong genetic influence on responsiveness to O3, and the follow-up study with NO2, another ambient air oxidant, examined the genetic basis for differences in response to the two agents. Kleeberger et al. (1997) determined significant genetic contributions in susceptibility to lung injury and inflammation induced by single and repeated acute exposures to NO2 and whether similar genetic factors control susceptibility to O3. Nine strains of inbred mice (male, 5–6 weeks) were studied. Each was exposed for 3 h to filtered air (controls) or 15 ppm NO2, and cellular inflammation, epithelial injury, and cytotoxicity were

902

OZONE

measured 2, 6, and 24 h thereafter. NO2 exposure caused significant increases in cytotoxicity and lavageable macrophages, epithelial cells, polymorphonuclear leukocytes, and protein in all strains. Interstrain variation in each of these effects indicated that genetic background contributed a significant portion of the variance in responses to this oxidant. Two strains that were differentially susceptible to 3 h exposure to 15 ppm NO2 [C57BL/6J (B6) and C3H/HeJ (C3)] were also exposed for 6 h/day to 10 ppm NO2 on 5 consecutive days. Each of the responses to NO2 was completely adapted after 5 days in resistant C3 mice. Only the lavageable total protein response was adapted in susceptible B6 mice. To determine whether mechanisms of susceptibility to NO2 and O3 were the same, each strain was exposed for 3 h to filtered air or 2 ppm O3 and inflammation was assessed 6 and 24 h thereafter. Strain distribution patterns for responses to each oxidant were not significantly concordant and indicated that susceptibility mechanisms were different. Results of these studies suggest that there is a strong genetic component to NO2 susceptibility that is partially adaptable and significantly different from O3 susceptibility. Studies in laboratory animals have examined the roles of O3 concentration and exposure time in biochemical and cellular responses. Rombout et al. (1989) exposed mice and rats to 380, 750, 1250, and 2000 ppb O3 for 1, 2, 4, and 8 h and measured BAL protein with both daytime and nighttime exposures. Observation times extended from 1 to 54 h. The responses varied with O3 concentration, duration of exposure, time after the start of the exposure, and minute volume, with time of exposure having a greater than proportional influence. For 4and 8 h exposures, the protein content of BAL peaked at 24 h and remained at elevated levels even at 54 h. As indicated previously, Devlin et al. (1991) found increased BAL protein in humans 18 h after an exposure to 100 ppb O3 for 6.6 h. Bhalla and Young (1992) studied the sequence of changes in lung epithelial permeability, free cells in the airways, prostaglandin E2 (PGE2) levels, PMN flux, and alveolar lesions in rats exposed to 800 ppb O3 for 3 h and then studied at 4 h intervals up to 24 h postexposure. Protein content of the BAL increased immediately after O3 exposure and returned to control levels by 16 h postexposure. Albumin concentration in the BAL increased more gradually and the albumin concentrations at 20 and 24 h postexposure were still higher than the control levels. While the total protein in the BAL could be attributed to tissue injury and increased transmucosal transport, the albumin primarily reflected elevated transport from the serum. Total cells in the BAL decreased immediately after the O3 exposure but returned to near normal levels by 4 h. PGE2 levels did not change significantly after O3 exposure. PMNs in the lung sections increased in number with time, peaked at 8 h, and returned to normal levels by 16 h after O3 exposure. The data suggest that the permeability changes may be produced by the direct toxic effects of O3 on the airway epithelia, but the PMNs contribute to the injury process, especially at the later stages. Lung lesions, represented by the thickening of the alveolar septae and increased cellularity, were present at 12 h postexposure and increased with time, thus coinciding with declining permeability at the later stages. The morphological changes lag behind the functional perturbations and appear to represent a phase of functional recovery. The weight of the evidence from these results, showing both functional and biochemical responses in humans and laboratory animals that accumulate over multiple hours and persist for many hours or days after exposure ceases, is clear and compelling. Both functional changes and inflammatory processes were shown to occur in humans following exposures to 100 ppb O3 for 6.6 h, whereas higher concentrations were required to elicit comparable responses in rats. Thus, the rat data, which provide evidence of other effects as well, appear to provide conservative indications of effects on humans.

DETERMINANTS OF RESPONSIVENESS TO OZONE EXPOSURES IN ANIMAL STUDIES

23.7.1

903

Effect of Single and Multiday Exposures on Particle Clearance

The effects of O3 on mucociliary particle clearance have been studied in rats and rabbits. Rats exposed for 4 h to O3 at concentrations in the range of 400–1200 ppb exhibited a slowing of particle clearance at 800 ppb but not at 400 ppb (Frager et al., 1979; Kenoyer et al., 1981). Rabbits exposed for 2 h to 100, 250, and 600 ppb O3 showed a concentration-dependent trend of reduced clearance rate with increasing concentrations, with the change at 600 ppb being approximately 50% and significantly different from control (Schlesinger and Driscoll, 1987). It is not known why the animal tests show only retarded mucociliary clearance in response to O3 exposure whereas the human tests show accelerated clearance. In corresponding tests with other irritants, that is, H2SO4 aerosol and cigarette smoke, both humans and animals have exhibited accelerated clearance at lower exposures and retarded clearance at higher exposures (Lippmann et al., 1987). Phipps et al. (1986) examined the effects of acute exposure to O3 on some of the factors that affect mucociliary transport rates in studies in which sheep were exposed to 500 ppb O3 for 2 h on two consecutive days. The exposures produced increased basal secretion of sulfated glycoproteins but had no effect on ion fluxes. Histological examination indicated a moderate hypertrophy of submucosal glands in the lower trachea, and they concluded that the exposure caused airway mucus hypersecretion. Studies on the effects of O3 on alveolar macrophage-mediated particle clearance during the first few weeks have also been performed in rats and rabbits. Rats exposed for 4 h to 800 ppb O3 had accelerated particle clearance (Frager et al., 1979). Rabbits exposed to 100, 600, or 1200 ppb O3 once for 2 h had accelerated clearance at 100 ppb and retarded clearance at 1200 ppb. Rabbits exposed for 2 h/day to 100 or 600 ppb O3 for 13 consecutive days had accelerated clearance for the first 10 days, with a greater effect at 600 ppb (Driscoll et al., 1986). The responses of the alveolar macrophages to these exposures were examined by Driscoll et al. (1987). A single exposure to 100 ppb resulted in increased macrophage numbers at 7 days, and repeated exposures resulted in an increase in macrophages and neutrophils on days 7 and 14. Macrophage phagocytosis was depressed immediately and 24 h after acute exposure to 100 ppb and at all times after exposure to 1200 ppb. Repeated exposures to 100 ppb produced reductions in the numbers of phagocytically active macrophages on days 3 and 7, with a return to control levels by day 14. Substrate attachment by macrophages was impaired immediately after exposure to 1200 ppb. The results of these studies demonstrated significant alterations in the numbers and functional properties of alveolar macrophages as a result of single or repeated exposure to 100 ppb O3, a level frequently encountered in areas of high photochemical air pollution. The timescale for the biological integration of the effects of a single O3 exposure has also been examined in studies on laboratory animals. Costa et al. (1989) exposed F-344 rats for 2, 4, and 8 h to O3 at 100, 200, 400 and 800 ppb. Lung function was measured immediately after exposure, and BAL was performed immediately and 24 h later. Functional decrements increased with the product ppb h, leveling off at >6,000, whereas BAL proteins increased rapidly for ppb h >4,000. In another test series involving 6.6 h of exposure with 8% CO2 to stimulate respiration, rats exposed to 500 ppb O3 had functional decrements closely matching those seen in humans at 120 ppb. Thus, rats can provide a good test model for the observed human responses to O3, even though they are a less sensitive species than humans. The lesser responses to a given O3 concentration reported here are consistent with the lesser retention of O3 by rats, as discussed previously.

904

OZONE

This issue was further addressed by Highfill et al. (1992) through an examination of relationships between concentration (C) and exposure time (T) and the impact of changes in the C  T product on toxic responses. Using protein concentration of bronchoalveolar lavage fluid as an index of O3-induced lung damage, models were developed from a matrix of C (0.0, 0.1, 0.2, 0.4, and 0.8 ppm) and T (2, 4, and 8 h) values in rat and guinea pig. Equal C  T products with different levels of C and T were incorporated into the protocol. Polynomial and exponential least-square models were developed, and the lognormal linear model (Larsen et al., 1991) was evaluated for the rat and guinea pig data. For equal C  T products, the results showed similar BALP responses at low C  T products. Calculations from the data and the models showed that (1) the models were consistent with reported experiments (Hatch et al., 1986); (2) exercising humans were more responsive to O3 exposure (without adjustments for ventilation rates) than were either rats or guinea pigs as measured by changes in BALP (Koren et al., 1989); and (3) the exponential model provided more generality than Haber’s law by providing estimates of BALP levels for various C  T. Ozone-induced bronchial hyperresponsiveness in dogs is inhibited by neutrophil depletion (O’Bryne et al., 1984a) and indomethacin pretreatment (O’Byrne et al., 1984b), suggesting that neutrophils that infiltrate the airways after acute O3 exposure (O’Byrne et al., 1984b; Holtzman et al., 1979) are the cells that release the cyclooxygenase products responsible for O3-induced bronchial hyperreactivity. However, neutrophil infiltration is a relatively late effect (i.e., occurring more than 6 h after exposure) and is not likely to account for the immediate responses. In a follow-up study (Jones et al., 1990), thromboxane antagonists were given to the dogs to further determine the role of thromboxane in O3-induced airway hyperresponsiveness. The antagonists did not inhibit the response, indicating that thromboxane does not have an important role in causing O3-induced airway hyperresponsiveness. Leikauf et al. (1988) investigated the hypothesis that oxidant damage to the tracheal epithelium may result in elaboration of various eicosanoids. After exposure to O3, epithelial cells derived from bovine trachea were isolated and grown to confluency. Monolayers were alternately exposed to O3 and culture medium for 2 h in a specially designed in vitro chamber using a rotating inclined platform (Valentine, 1985). O3 induced increases in cyclooxygenase and lipoxygenase product formation with significant increases in prostaglandins E2, F2a, 6-keto F1a, and leukotriene B4. Release rates of immunoreactive products were dosedependent, and ozone concentrations as low as 100 ppb produced an increase in prostaglandin F2a. Thus, O3 can augment eicosanoid metabolism in airway epithelial cells. 23.7.2

Effects of Single and Multiday Exposures on Lung Infectivity

Both in vivo and in vitro studies have demonstrated that O3 can affect the ability of the immune system to defend against infection. Increased susceptibility to bacterial infection has been reported in mice at 80–100 ppb O3 for a single 3 h exposure (Coffin et al., 1967; Ehrlich et al., 1977; Miller et al., 1978a). Related alterations of the pulmonary defenses caused by short-term exposures to O3 include impaired ability to inactivate bacteria in rabbits and mice (Coffin et al., 1968; Coffin and Gardner, 1972; Goldstein et al., 1977; Ehrlich et al., 1979) and impaired macrophage phagocytic activity, mobility, fragility and membrane alterations, and reduced lysosomal enzymatic activity (Witz et al., 1983; Dowell et al., 1970; Hurst and Coffin, 1971; Hurst et al., 1970; Goldstein et al., 1971a, 1971b; McAllen et al., 1981; Amoruso et al., 1981). Some of these effects have been shown to occur in a variety of species including mice, rats, rabbits, guinea pigs, dogs, sheep, and monkeys.

DETERMINANTS OF RESPONSIVENESS TO OZONE EXPOSURES IN ANIMAL STUDIES

905

Other studies indicate similar effects for short-term and subchronic exposures of mice to O3 combined with pollutants such as SO2, NO2, H2SO4, and particles (Gardner et al., 1977; Aranyi et al., 1983; Ehrlich, 1980; Grose et al., (1980a, 1980b; Phalen et al., 1980). Similar to the human pulmonary function response to O3, activity levels of mice exposed to O3 have been shown to play a role in determining the lowest effective concentration that alters the immunity (Illing et al., 1980). In addition, the duration of exposure must be considered. In groups of mice exposed to 200 ppb O3 for 1, 3, and 6 h, superoxide anion radical production decreased 8, 18, and 35%, respectively, indicating a progressive decrease in bacteriocidal capacity with increasing duration of exposure (Amoruso and Goldstein, 1988). The major limitation of this large body of data on the influence of inhaled O3 on lung infectivity is that it requires uncertain interspecies extrapolating to estimate the possible effects of O3 on infectivity in humans. Gilmour and Selgrade (1993) compared the pulmonary defenses of rats and mice against streptococcal infection following O3 exposures. In mice, 3 h O3 exposures at 400 ppb resulted in bacterial proliferation and PMN influx in the lungs and excess mortality. By contrast, the rats had only a transient impairment of microbial inactivation. These results indicate that caution is needed in translating the results from either species to predictions of human responses. 23.7.3

Effects of Other Pollutants on Responses to Ozone

A study that addressed the issue of the potentiation of the characteristic functional response to inhaled O3 by other environmental cofactors was performed in Tuxedo, NY (Spektor et al., (1988b). It involved healthy adult nonsmokers engaged in a daily program of outdoor exercise with exposures to an ambient mixture containing low concentrations of acidic aerosols and NO2 as well as O3. Each subject did the same exercise each day, but exercise intensity and duration varied widely between subjects, with minute ventilation ranging from 20 to 153 L, with an average of 79 L, and with duration of daily exercise ranging from 15 to 55 min, with an average of 29 min. Respiratory function was measured immediately before and after each exercise period. The O3 concentrations during exercise ranged from 21 to 124 ppb. All respiratory function indices thus measured showed significant (p < 0.01) O3associated mean decrements. The functional decrements were similar, in proportion to lung volume, to those seen in children engaged in supervised recreational programs in summer camps. They were as large (FEV1) as or much larger (FVC, FEF25–75, PEFR) than those seen in controlled 1- and 2 h exposures in chambers. For the subgroup with the most comparable levels of physical activity, the responses in the field study were even greater. Since the ambient exposures of the adults exercising outdoors were for approximately 1/2 h, compared to the 1- or 2 h exposures in the chamber studies, it was concluded that ambient cofactors potentiate the responses to O3. Thus, the results of the exposures in chambers to O3 in purified air underestimate the O3-associated responses that occur among populations engaged in normal outdoor recreational activity and exposed to O3 in ambient air in the northeastern United States. The apparent potentiation of O3-induced functional decrements observed by Spektor et al. (1988b) in rural New York was not seen by Avol et al. (1984) in a study in southern California in which 42 healthy young men and 8 healthy young women were exposed for 1 h to ambient air containing an average of 153 ppb O3 while exercising heavily in a chamber. The functional decrements were slightly but not significantly smaller than those produced in the same subjects on another day when they were exposed to 160 ppb O3 in purified air. The ambient air in southern California has much higher NO2 concentrations and much lower acid aerosol concentrations than the ambient air in the northeastern United States. Thus, it appears

906

OZONE

that the Hþ content of aerosols is a more likely causal factor for the potentiation seen by Spektor et al. (1988b) than is NO2. However, it must be noted that the Spektor et al. (1988b) study on exercising adults and earlier studies on children at summer camps (Lioy et al., 1985; Spektor et al., (1988a) were not able to demonstrate the specific effect of any of the measured environmental variables, including heat stress and acid aerosol concentration, on the O3-associated responses. The inability to show the individual effects of other environmental cofactors on the response to ambient O3 may result from inadequate knowledge on the appropriate biological averaging time for these other factors. However, in the study of functional responses of children to ambient pollution in Mendham, NJ, a week-long baseline shift in PEFR was associated with both O3 and H2SO4 exposures during a 4-day pollution episode that preceded it (Lioy et al., 1985). A similar response to a brief episode with elevated O3 and a much higher peak 4 h concentration of H2SO4 (46 mg/m3) was seen among girls attending a summer camp in 1986 at Dunnville, on the northeast shore of Lake Erie, Ontario, Canada (Raizenne et al., 1989). Several controlled human exposure studies in chambers have not demonstrated synergy in functional response between O3 and NO2 (Koenig et al., 1988) or between O3 and H2SO4, although Stacy et al. (1983) did report that the mean responses to 400 ppb O3 and 100 mg/m3 H2SO4 after 2 h of exposure at rest were 9.0% for FVC and 11.5% for FEV1, compared to corresponding values of 5.7 and 7.7% for O3 alone, 1.4, and 1.2% for sham exposure, and þ0.9 and þ0.9% for H2SO4 alone. One possible reason why these mean differences, which appear to indicate an enhancement of the O3 response by H2SO4, were not statistically significant was the very high variability of the sham exposure results. By contrast, Koenig et al. (1990) did demonstrate that prior exposure to O3 at 120 ppb with intermittent exercise for 45 min potentiated the subsequent respiratory function response to a 15 min exposure to SO2 at 100 ppb. Frampton et al. (1995) exposed 30 healthy and 30 asthmatic volunteers to either 100 mg/m3 H2SO4 or NaCl for 3 h followed 24 h later by 3 h exposures to O3 at 80, 120, or 180 ppb. For the healthy group, no convincing symptomatic or physiologic effects of exposure to either aerosol or O3 on lung function were found. For the asthmatic group, preexposure to H2SO4 altered the pattern of response to O3 in comparison with NaCl preexposure and appeared to enhance the small mean decrements in FVC that occurred in response to 180 ppb O3. Individual responses among asthmatic subjects were quite variable, some demonstrating reductions in FEV1 of more than 35% following O3 exposure. Analysis of variance of changes in FVC revealed evidence for interactions between aerosol and O3 exposure both immediately after (P ¼ 0.005) and 4 h after (P ¼ 0.030) exposure. Similar effects were seen for FEV1. When normal and asthmatic subjects were combined, four-way analysis of variance revealed an interaction between O3 and aerosol for the entire group (P ¼ 0.0022) and a difference between normal and asthmatic subjects (P ¼ 0.0048). There was no significant effect of exposures on symptoms for either normal or asthmatic subjects. Pollutant interactions that potentiate the characteristic O3 response have also been reported for other effects in controlled exposure studies in animals. Osebold et al. (1980) exposed antigenically sensitized mice to 500 ppb O3 for 3 days, with and without concurrent exposure to 1 mg/m3 of submicrometer H2SO4 droplets. There was an increase in atopic reactivity that was greater than that for each pollutant alone. Lee et al. (1990) exposed 3-month-old male rats to either filtered room air (control) or 1,200 ppb NO2, 300 ppb O3 or a combination of the two oxidants continuously for 3 days. They studied a series of parameters in the lung, including lung weight and enzyme activities related to NADPH generation, sulfhydryl metabolism, and cellular detoxification. The results showed that relative to control, exposure to NO2 caused small (nonsignificant) changes in all the parameters; O3

DETERMINANTS OF RESPONSIVENESS TO OZONE EXPOSURES IN ANIMAL STUDIES

907

caused significant increases in all the parameters except for superoxide dismutase; a combination of NO2 and O3 caused increases in all the parameters, and the increases were greater than those caused by NO2 or O3 alone. Statistical analysis of the data showed that the effects of combined exposure were synergistic for 6-phosphogluconate dehydrogenase, isocitrate dehydrogenase, glutathione reductase, and superoxide dismutase activities, and additive for glutathione peroxidase and disulfide reductase activities but not different from those of O3 exposure for other enzyme activities. Kleinman et al. (1989) reported that lesions in the gas-exhange region of the lung of rats exposed to O3 were greater in size in rats exposed to mixtures containing O3 with either H2SO4 or NO2. Graham et al. (1987) reported a synergistic interaction between O3 and NO2 in terms of mortality in mice challenged with streptococcal infection either immediately or 18 h after pollutant exposure. Last (1989) reported synergistic interaction in rats, in terms of a significant increase in lung protein content, following 9-day exposures at 200 ppb O3 with 20 or 40 mg/m3 H2SO4, and a nonsignificant increase for 9 days at 200 ppb O3 with 5 mg/m3 H2SO4. In summary, single O3 exposures to healthy nonsmoking young adults at concentrations in the range of 80–200 ppb have produced a complex array of pulmonary responses including decreases in respiratory function and athletic performance and increases in symptoms, airway reactivity, neutrophil content in lung lavage, and rate of mucociliary particle clearance. The responses to O3 in purified air in chambers occur at concentrations of 80 or 100 ppb when the exposures involve moderate exercise over 6 h or more and require concentrations of 180 or 200 ppb when the duration of exposure is 2 h or less. On the other hand, mean FEV1 decrements more than 5% have been seen at 100 ppb of O3 in ambient air for children at summer camps and for adults engaged in outdoor exercise for only 1/2 h. The apparently greater responses to O3 in ambient air may result from the presence of, or prior exposures to, acidic aerosol, but this tentative hypothesis need further investigation. Further research is also needed to establish the interrelationships between small transient functional decrements, such as FEV1, PEFR, mucociliary clearance rates, and changes in symptoms, performance, reactivity, permeability, and neutrophil counts. The latter may be adverse or may be more closely related to the accumulation or progression of chronic lung damage. If transient changes in readily measured functions, such as FEV1 or PEFR, are closely correlated with other, more significant health effects, then they could be established as useful surrogates in large-scale laboratory, field, and epidemiologic research as well as further retrospective analyses of published data on human exposure–response. Finally, we need more investigation of the mechanisms underlying the pulmonary responses to inhaled O3. The mechanisms have been summarized by Bates (1989, 1995). The author notes that after O3 exposure, the inspiratory capacity is first reduced as a consequence of a lower maximal negative intrapleural pressure on taking a full inspiration. Maximal inspiratory and expiratory mouth pressures are not affected. He emphasizes that the FEV1/FVC ratio is not initially affected after O3 exposure, which is to say that the FVC and FEV1 initially fall together. He postulates that stimulation of the C-fiber system in the lung must lead to a “braking” effect on the inspiratory muscles as a first consequence of O3 exposure, and this probably occurs as a result of induced inflammation. The increased respiratory rate after O3 exposure, the increased lung permeability, the increased airway reactivity, and the fact that b blockers do not prevent the changes induced by O3 all support this hypothesis. A similar mechanism was suggested by Lee (1990) for the inhibitory effect of the gas phase of cigarette smoke on breathing. He showed that the effect could be largely prevented in rats by the administration of dimethylthiourea, a hydroxyl radical scavenger. The reduced response after repeated

908

OZONE

exposures might result from a thicker lining of mucus over the surface of the airway or from actual cell replacement after exposure.

23.8 EFFECTS OF MULTIPLE DAY AND AMBIENT EPISODE EXPOSURES Since single exposures lasting for 1 h or more at current peak ambient O3 levels produce measurable biological responses in healthy humans and since there is a high probability that one high-O3 day will be followed by several more high-O3 days (Rao, 1988), it is important to know the extent to which the effects accumulate or progress over multiple days. This section reviews the fairly substantial database on functional adaptation to repetitive exposures and the more limited database on biochemical and structural changes that such exposures produce. It should be noted that the data on functional adaptation is largely, but not exclusively, based on studies in human volunteers, whereas the database on biochemical and structural changes caused by O3 is based entirely on studies in laboratory animals. Data on exposures lasting more than 2 weeks are discussed in the Section 23.9. It is well established that repetitive daily exposures, at a level that produces a functional response upon single exposure, result in an enhanced response on day 2, with diminishing responses on days 3 and 4 and virtually no response by day 5 (Farrell et al., 1979; Folinsbee et al., 1980; Hackney et al., 1977). Brookes et al. (1989) found enhanced responses on the second day of successive exposures of exercising young adult males to 350 ppb O3 for 1 h as well as an enhanced response to 250 ppb when the previous day’s exposure was to 350 ppb. In older adults (60–89 years), successive days of 2 h exposures to 450 ppb O3 with light exercise led to small functional decrements on the first 2 days but no changes on successive days (Bedi et al., 1989). For repeated 6.6 h/day exposures to 120 ppb O3, the peak functional response occurs on the first day, with progressively lesser responses after the second, third, and fourth days of exposure. However, for these same subjects, their responsiveness to methocholine challenge peaked on the second day and remained elevated throughout all 5 days of exposure (Folinsbee et al., 1994). The persistent elevation of airway responsiveness is one reason to discount the view of some people that the functional adaptation phenomenon indicates that transient functional decrements are not an important health effect. Additional evidence comes from research in animals showing that persistent damage to lung cells accumulates even as functional adaptation takes place. This kind of functional adaptation to exposure disappears about a week after exposure ceases (Horvath et al., 1981; Kull et al., 1982). The adaptation phenomenon has led some people to conclude that transient functional decrements are not important health effects. However, recent research in animals has shown that persistent damage to lung cells accumulates even as functional adaptation takes place. Tepper et al. (1989) exposed rats to 350, 500, or 1000 ppb O3 for 2.25 h on 5 consecutive days. CO2 (8%) was added to the exposure during alternate 15 min periods to stimulate breathing and thereby increase O3 uptake and distribution. The consequences of exposure on pulmonary function, histology, macrophage phagocytosis, lavageable protein, differential cell counts, and lung tissue antioxidants were assessed. Tidal volume, frequency of breathing, inspiratory time, expiratory time, and maximal tidal flows were affected by O3 during days 1 and 2 at all O3 concentrations. By day 5, these O3 responses were completely adapted at 350 ppb, greatly attenuated at 500 ppb, but showed no signs of adaptation in the group exposed to

EFFECTS OF MULTIPLE DAY AND AMBIENT EPISODE EXPOSURES

909

1000 ppb. Unlike the pulmonary function data, light microscopy indicated a pattern of progressive epithelial damage and inflammatory changes associated with the terminal bronchiole region. Over the 5 day testing period, a sustained 37% increase in lavageable protein and 60% suppression of macrophage phagocytic activity were observed with exposure to 500 ppb. There were no changes in differential cell counts. Lung glutathione was initially increased but was within the control range on days 4 and 5. Lung ascorbate was significantly elevated above control on days 3–5. These data suggest that attenuation of the pulmonary functional response occurs while aspects of the tissue response reveal progressive damage. Van Bree et al. (1989) reported the influence of exposure time per day and number of exposure days on biochemical and cellular changes in the lung. Seven-day exposures at 800 ppb produced a loss of normal cilia, a hypertrophic response of Clara cells, and an increase in P-450 isoenzyme activity, whereas 4 day exposures produced increases in protein, G6PDH, and GSHP4. In rats exposed for 2, 4, 8, and 16 days at 400 ppb O3 for 4, 8, and 24 h/ day, the quantity of antioxidant in whole-lung tissue was influenced about twice as much by the exposure duration per day as by the number of exposure days. Finally, in rats exposed to 400 ppb O3 for 12 h at either daytime or nighttime, the effects at night, when they were active, were much greater, once again showing the influence of physical activity on responses to O3. Further indications that functional adaptation, as measured in the days and weeks following exposure, is not fully protective against the development of pathological changes have been provided by a study reported by Farman et al. (1997). This was a follow-up study of one by Last et al. (1993) that showed that rats exposed to 800 ppb O3 and 14.4 ppm NO2 for 6 h daily developed progressive bronchiolitis and pulmonary fibrosis after about 8–10 weeks of exposure, with a high level of mortality. To begin to understand what processes are occurring during the approximately 2- to 2.5-month long period of lesion development, they studied the time course of evolution of fibrotic lesions in rats exposed to O3 and NO2. Rats were sampled weekly for 9 weeks from the onset of exposure, and biochemical and histopathological evaluations were performed. They also quantified the reparative potential of the airway epithelium after 4 and 8 weeks of exposure by in vivo labeling with bromodeoxyuridine (BrdU). Histopathological evaluation indicated a triphasic response temporally: inflammatory and fibrotic changes increased mildly for the first 3 weeks of exposure, stabilized or apparently decreased during 4–6 weeks, and demonstrated severe increases over 7–9 weeks. Biochemical quantification of lung 4-hydroxyproline (collagen) content showed a pattern consistent with the histopathology: no significant differences from controls for the first 3 weeks of exposure, significant increases in collagen content after 4–5 weeks of exposure, and a stabilization of lung collagen content after 6 weeks of exposure. In vivo determination of cumulative labeling indexes showed normal (or slightly decreased) repair of the small airway and alveolar epithelium after 4 weeks of exposure, with significantly diminished reparative capacity after 8 weeks. The diminished reparative capacity of the bronchiolar and alveolar epithelium may be causally linked to the rapid, progressive fibrosis that occurs in this model after about 7–8 weeks of exposure to O3 plus NO2. However, it should be noted that this progression of responses may not be relevant to lower level of exposures, since the Last et al. (1993) paper reported no long-term effects in the rats exposed to 200 ppb O3 plus 3.6 ppm NO2 for 90 days. The effects of multiday O3 exposures of laboratory animals on particle clearance from the lungs and on lung infectivity were reviewed previously. They also show that O3-induced transient effects often become greater with repetitive exposures.

910

OZONE

Effects of a multiday episode-type exposure to O3 on humans in ambient air were described by Lioy et al. (1985). During a study focused on daily variations in lung function among 39 children attending a summer day camp in Mendham, NJ, a summer haze episode occurred in which the daily 1 h peak O3 concentrations exceeded 120 ppb on four consecutive days, with the highest concentration being 185 ppb. During the week following the episode, there were consistent deviations in function from the concentration versus peak flow regressions for the individual children, indicating a persistent loss of lung function during that time. In a subsequent reanalysis of the data from this study, Lioy and Dyba (1989) suggested that the persistence of the reduced function during the week following the episode was more likely due to the cumulative daily exposure than by the daily peak concentrations. In any case, the exposure episode was apparently responsible for an approximately 1-weeklong shift in the function baseline, suggesting that epithelial cell death and regeneration were involved and not just a reflex airway constriction. In summary, successive days of exposure of adult humans in chambers to O3 at current high ambient levels lead to a functional adaptation in that the responses are attenuated by the third day and are negligible by the fifth day. However, a comparable functional adaptation in rats does not prevent the progressive damage to the lung epithelium. Daily exposures of animals also increase other responses in comparison to single exposures, such as a loss of cilia, a hypertrophic response of Clara cells, alterations in macrophage function, and alterations in the rates of particle clearance from the lungs. For children exposed to O3 in ambient air there was a weeklong baseline shift in peak flow following a summer haze exposure of a 4-day duration with daily peak O3 concentrations ranging from 125 to 185 ppb. Since higher concentrations used in adult adaptation studies in chambers did not produce such effects, it is possible that baseline shifts require the presence of other pollutants in the ambient air. A baseline shift in peak flow in camp children was also reported by Raizenne et al. (1989) following a brief episode characterized by a peak O3 concentration of 143 ppb and a peak acidic aerosol concentration of 559 nmol/m ¼ 3.

23.9 CHRONIC EFFECTS OF AMBIENT OZONE EXPOSURES The chronic effects database includes a quite limited amount of information on human effects and a more substantial volume of data on effects seen in laboratory animals undergoing chronic exposures. 23.9.1

Controlled Laboratory Exposure Studies: Human Responses

A study by Linn et al. (1988) in Southern California provided evidence for a seasonal adaptation of lung function. In this study, a group of subjects selected for their relatively high functional responsiveness to O3 had much greater functional decrements following 2 h of exposure to O3 at 180 ppb with intermittent exercise in a chamber in the spring than they did in the following autumn or winter, although their responses in the following spring were equivalent to those in the preceding spring. These findings suggest that some of the variability in acute response coefficients reported for earlier controlled human exposures to O3 in chambers could have been related to seasonal variations in responsiveness, which, in turn, may be related to a long-term adaptation to chronic O3 exposure.

CHRONIC EFFECTS OF AMBIENT OZONE EXPOSURES

23.9.2

911

Epidemiological Studies

23.9.2.1 Baseline Respiratory Function Some cross-sectional studies also suggest O3related decrements in respiratory function. Stern et al. (1994) examined differences in the respiratory health status of school children, aged 7–11 years, who resided in 10 rural Canadian communities in areas of moderate and low exposure to regional SO42 and O3 pollution. Five of the communities were located in central Saskatchewan, a low-exposure region and five were located in southwestern Ontario, an area with moderately elevated exposures resulting from long-range atmospheric transport of polluted air masses. Summertime 1 h daily O3 maxima means were 69.0 ppb in Ontario and 36.1 ppb in Saskatchewan. Concentrations of SO42 were three times higher in Ontario than in Saskatchewan; there were no significant differences in levels of PM10 or particulate nitrates. Levels of SO2 and NO2 were low in both regions. After controlling for the effects of age, sex, parental smoking, parental education, and gas cooking, no significant regional differences were observed in symptoms. Children living in Ontario had statistically significant (p < 0.01) mean decrements of 1.7% in FVC and 1.3% in FEV1.0 compared to Saskatchewan children, after adjusting for age, sex, weight, standing height, parental smoking, and gas cooking, but therewere no statistically significant regional differences in the pulmonary flow parameters. The differences could have been due to exposures to either O3 or SO42 or their combination. A more definitive study by Kunzli et al. (1997) regressed mid- and end-expiratory flows (FEF25–75%, FEF75%) against effective exposure to O3. A convenience sample of 130 UC Berkeley freshmen, ages 17–21, participated twice in the same tests (residential history, questionnaire, pulmonary function), 5–7 days apart. Students had to be life-long residents of Northern (SF) or Southern (LA) California. Monthly ambient 8 h O3 concentrations were assigned based on the lifetime residential history and nearby monitoring data for O3. For a 20 ppb increase (interquartile range) in 8 h O3, FEF75% decreased, 14% (95% Cl: 1.0–28.3%) of the population mean FEF75%. The effect on FEF25–75% was 7.2% of the mean. Negative confounding factors were region (SF versus LA), gender, and ethnicity. Lifetime 8 h average O3 ranged from 16 to 74 ppb with little overlap between regions. There was no evidence for different O3 effects across regions. Effects were independent of lifetime mean PM10, NO2, temperature, or humidity. Effects on FEV1 tended to be negative whereas those for FVC, although negative in some models, were inconsistent and small. The strong relationship of lifetime effects of ambient O3 on mid- and end-expiratory flows of college freshmen and the lack of association with FEV1 and FVC are consistent with biologic models of chronic effects of O3 in the small airways. Evidence of chronic effects of O3 were reported by Schwartz (1989) based upon an analysis of pulmonary function data in a national population study in 1976–80, that is, the second National Health and Nutrition Examination Survey (NHANES II). Using ambient O3 data from nearby monitoring sites, he reported a highly significant O3-associated reduction in lung function for people living in areas where the annual average O3 concentrations exceeded 40 ppb. Recent studies of lung function growth in cohorts of children in 12 Southern California communities suggest that ambient air pollutants other than O3 are more responsible for the pulmonary function effects (Gauderman et al., 2000, Gauderman et al., 2002, 2005; Avol et al., 2001). 23.9.2.2 Lung Structure An autopsy study of 107 lungs from 14–25-year-old accident victims in Los Angeles County by Sherwin and Richters (1991) reported that 27% had

912

OZONE

what were judged to be severe degrees of structural abnormalities and bronchiolitis not expected for such young subjects, whereas another 48% of them had similar, but less severe, abnormalities. In the absence of corresponding analyses of lungs of comparable subjects from communities having much lower levels of air pollution, the possible association of the observed abnormalities with chronic O3 exposure remains speculative. Some of the abnormalities observed could have been due to smoking and/ or drug abuse, although the authors noted that published work on the association between smoking and small airway effects showed lesser degrees of abnormality (Cosio et al., 1980). 23.9.2.3 Development of Chronic Disease The effects of chronic exposure to O3 and PM was followed for 10 years in a prospective cohort study by Abbey et al. (1995) in 6340 nonsmoking 7th-Day Adventists living in California. Ambient air monitoring data were available for O3, TSP, SO42 , NO2, and SO2. No significant associations were found for NO2 or SO2. O3 was found significantly associated with increasing severity of asthma and with the development of asthma in males. Measured TSP and SO42 and estimated PM2.5 and PM10 were associated with the development of airway obstructive disease, chronic bronchitis, and asthma, and these were not confounded by the presence of the gaseous pollutants. 23.9.2.4 Effect on Longevity The limited evidence available to date is largely negative. Mendelsohn and Orcutt (1979), in a study utilizing the Public Use Sample containing data on 2 million individuals in the United States obtained both death certificate data and air pollution network data in eight regions of the United States. Highly significant and consistent associations with mortality were found for SO42 . Significant, but weaker and less consistent, associations were seen for SO2 and CO. No significant associations were seen for O3 or NO2. The only other multipollutant studies of annual mortality rates were the six-city study of Dockery et al. (1993) and the American Cancer Society study of Pope et al. (2002). In the six-city prospective cohort study of 8,111 adults over 14–16 years of age, highly significant and consistent mortality effects were seen for PM2.5 and SO42 , with smaller effects indicated for TSP, SO2, and NO2. The variations in O3 across the six cities were too small for effects to be detected. In the Pope et al. (2002) study, there was no significant association of premature mortality with PM2.5, SO42 , and SO2, but not with O3. 23.9.3 Controlled Subchronic and Chronic Laboratory Exposure Studies: Animal Responses Most of the inhaled O3 penetrates beyond the sites in the airways that trigger the functional responses. In this deeper region of the lung, at and just beyond the terminal bronchioles, the effects produced by O3 include changes in biochemical indices, lung inflammation, and airway structure. Furthermore, the effects of O3 exposure in this region appear to be cumulative and persistent, even in animals that have adapted to the exposure in terms of respiratory mechanics (Frank et al., 2001). In a series of inhalation studies, rats were exposed to O3 at constant concentrations of either 120 or 250 ppb for 12 h per day for 6 and 12 weeks, or to a daily cycle with a baseline of 60 ppb for 15 h with a broad peak for 8 h averaging 180 ppb for a period of 3 to 12 weeks.

CHRONIC EFFECTS OF AMBIENT OZONE EXPOSURES

913

Hyperplasia of type I alveolar cells in the proximal alveoli was linearly related to the cumulative O3 dose (Huang et al., 1988). The highest O3 dose is received at the acinus, where the terminal bronchioles lead into alveolar ducts, and a series of studies has shown that the effects of inhaled O3 on lung structure is also greatest in this region. Barry et al. (1985) showed that significant changes occurred in the alveoli just distal to the terminal bronchioles in rats exposed for 12 h/day for 6 or 12 weeks to 120 or 240 ppb O3. In both juvenile and adult rats there were significant increases in the numbers of alveolar type I and type II epithelial cells and alterations in the interstitium and endothelium. From physiological studies of rats that were simultaneously exposed, Raub et al. (1983) reported that there were significant increases in the vital capacity and end-expiratory volume that suggested alterations in distensibility of the lung tissue. For the 6-week exposures at 250 ppb, Barry et al. (1988) reported that exposure to O3 produced alterations in the surface characteristics of ciliated and nonciliated (Clara) cells in both groups. There were significant losses (20–30%) of the surface area contributed by ciliated cells, the luminal surface of Clara cells was decreased by 16–25%, and the number of brush cells per square millimeter of terminal bronchiolar basement membrane was also decreased. Thus, the normal structure of terminal bronchiolar epithelial cells was significantly altered. No statistically significant interaction between the effects of O3 and the animal age at the beginning of the exposure was found. The series of inhalation studies was extended to include tests in which there was a daily cycle with a baseline of 60 ppb for 13 h with a 5-day/week broad peak for 9 h averaging 180 ppb, and containing a 1 h maximum of 250 ppb over a period of 3–12 weeks. Combining the results of these tests with the 6-week studies, Huang et al. (1988) and Chang et al. (1991) reported that hyperplasia of type I alveolar cells in the proximal alveoli was linearly related to the cumulative O3 exposure in the four groups. Thus, there is no threshold for cumulative lung damage. Rats exposed for 6 weeks to clean air or to O3 using the daily cyclic exposure regimen used by Huang et al. (1988) were exposed once for 5 h to an asbestos aerosol by Pinkerton et al. (1989). When they were sacrificed 30 days later, the fiber count in the lungs of the O3exposed animals was three times greater than in the sham-exposed animals. Thus, subchronic O3 exposure can increase the effective dose of insoluble particles, which may have toxic and/or carcinogenic effects. In rats exposed for 12 months by Grose et al. (1989) to the daily exposure cycle used by Huang et al. (1988), an increase in the rate of lung nitrogen washout was observed. Residual volume and total lung capacity were reduced. Glutathione peroxidase and reductase activities were increased, but pulmonary superoxide dismutase remained unchanged. aTocopherol levels were decreased in lung lavage supernatant but were unchanged in lavaged cells, while ascorbic acid and lavage fluid protein were increased. Immunological changes were not observed. Thus, 12 months of exposure to O3 caused (1) functional lung changes indicative of a “stiffer” lung; (2) biochemical changes suggestive of increased antioxidant metabolism; and (3) no observable immunological changes. In a follow-up study in which the same exposure cycle was extended for up to 78 weeks, Tepper et al. (1991) found small, but statistically significant, changes in breathing patterns and mechanisms in unanesthetized, restrained rats at weeks 1, 3, 13, 52, and 78 during postexposure challenge with 0, 4, and 8% CO2. The data indicate that O3 exposure caused an overall increase in expiratory resistance (Rc), but particularly at 78 weeks. The spontaneous frequency of breathing and CO2-induced hyperventilation were also reduced. The decrease

914

OZONE

in frequency depended on a significant increase in the inspiratory time relative to control without a change in the expiratory time. However, light microscopic evaluation of the lung did not reveal any lesions associated with O3 exposure. Chang et al. (1992) extended the analyses of animals exposed for 78 weeks to electron microscopic morphometry. Samples from proximal alveolar regions and terminal bronchioles were obtained by microdissection. Analysis of the proximal alveolar region revealed a biphasic response. Acute tissue reactions after 1 week of exposure included epithelial inflammation, interstitial edema, interstitial cell hypertrophy, and influx of macrophages. These responses subsided after 3 weeks of exposure. Progressive epithelial and interstitial tissue responses developed with prolonged exposure and included epithelial hyperplasia, fibroblast proliferation, and interstitial matrix accumulation. The epithelial responses involved both type I and type II epithelial cells. Alveolar type I cells increased in number, became thicker, and covered a smaller average surface area. These changes persisted throughout the entire exposure period and did not change during the recovery period, indicating the sensitivity of these cells to injury. The main response of type II epithelial cells was cell proliferation. The accumulation of interstitial matrix after chronic exposure consisted of deposition of both increased amounts of basement membrane and collagen fibers. Interstitial matrix accumulation underwent partial recovery during follow-up periods in air; however, the thickening of the basement membrane did not resolve. Analysis of terminal bronchioles showed that short-term exposure to O3 caused a loss of ciliated cells and differentiation of preciliated and Clara cells. The bronchiolar cell population stabilized on continued exposure; however, chronic exposure resulted in structural changes, suggesting injury to both ciliated and Clara cells. Thus, chronic exposure to low levels of O3 caused epithelial inflammation and interstitial fibrosis in the proximal alveolar region and bronchiolar epithelial cell injury. Studies at relatively low O3 concentrations have also been conducted on monkeys. Hyde et al. (1989) exposed them to O3 for 8 h/day for 6 or 90 days to 150 or 300 ppb O3. Responses included ciliated cell necrosis, shortened cilia, and secretory cell hyperplasia with less stored glycoconjugates in the nasal region. Respiratory bronchiolitis observed in 6 days persisted for 90 days of exposure. Even at the lower concentration of 150 ppb O3, nonciliated bronchiolar cells appeared hypertrophied and increased in abundance in respiratory bronchioles. For some chronic effects, intermittent exposures can produce greater effects than those produced by a continuous exposure regime that results in higher cumulative exposures. For example, Tyler et al. (1988) exposed two groups of 7-month-old male monkeys to 250 ppb O3 for 8 h/day either daily or, in the seasonal model, on days of alternate months during a total exposure period of 18 months. A control group breathed only filtered air. Monkeys from the seasonal exposure model, but not those exposed daily, had significantly increased total lung collagen content, chest wall compliance, and inspiratory capacity. All monkeys exposed to O3 had respiratory bronchiolitis with significant increases in related morphometric parameters. The only significant morphometric difference between seasonal and daily groups was in the volume fraction of macrophages. Even though the seasonally exposed monkeys were exposed to the same concentration of O3 for only half as many days, they had larger biochemical and physiological alterations and equivalent morphometric changes as those exposed daily. Lung growth was not completely normal in either exposed group. Thus, longterm effects of oxidant air pollutants that have a seasonal occurrence may be more dependent on the sequence of polluted and clean air than on the total number of days of pollution, and estimations of the risks of human exposure to seasonal air pollutants from effects observed in animals exposed daily may underestimate long-term pulmonary damage. The effects

CHRONIC EFFECTS OF AMBIENT OZONE EXPOSURES

915

observed may be considered directly relevant to human health, especially in view of our knowledge that humans receive even greater local doses of O3 in the vicinity of the acinus than do rats. A number of other interesting chronic exposure studies have been done in animals with O3 concentrations in the range of 300–1000 ppb. Those that appear to provide useful insights into mechanisms of toxic action will also be briefly reviewed. Sherwin and Richters (1985) exposed newborn Swiss–Webster mice to intermittent 300 ppb O3 for 7 h/day, 5 days/week for 6 weeks. O3 exposure increased cell and wall measurements. In contrast to results previously reported for adult animals (Sherwin et al., 1983), there was a greater increase in mean type II cell area than in numbers of type II cells. Effects on the type II cell population implicate damage to the type I alveolar lining cells. The increases in alveolar wall measurements that were found in both the adult and the developing mouse lung imply an alteration of the lung scaffolding and raising the question of impaired regeneration of the epithelial lining. The results of a chronic exposure study in rats by Gross and White (1987) illustrate the importance of exposure pattern on the magnitude of the response. They exposed F-344 rats to 500 ppb O3 for 20 h/day, 7 days/week for 52 weeks and produced only mild functional changes (functional residual capacity, residual volume, and DLco), which returned to normal during 3 months of recovery. Grose et al. (1989), using a 23 h exposure ranging from 60 ppb to a peak 1 h maximum of 250 ppb for 5 days/week, produced comparable functional changes in 1 year. Thus, as in the comparison by Tyler et al. (1988) in monkeys, intermittent exposures, modeled after realistic human exposure conditions, can produce much greater responses per unit dose than continuous exposure at high concentration. These results suggest that the damage results, at least in part, from the repeated attempts to adapt to the irritant challenge as well as to the direct effects of the irritant exposure. To characterize the response of respiratory bronchioles (RBs) to chronic high ambient levels of O3, Moffatt et al. (1987) exposed bonnet monkeys 8 h/day for 90 days to 400 or 640 ppb O3. Significant changes in RB following exposure included (1) a thicker wall and a narrower lumen; (2) a thicker epithelial compartment and a much thicker interstitial compartment; (3) shifts in epithelial cell populations with many more nonciliated bronchiolar epithelial cells and fewer squamous type I epithelial cells; (4) larger nonciliated bronchiolar epithelial cells with a larger compliment of cellular organelles associated with protein synthesis; (5) greater amounts of both interstitial fibers and amorphous ground substance; (6) greater numbers of interstitial smooth muscle cells per epithelial basal lamina surface area; and (7) greater volumes of interstitial smooth muscle, macrophages, mast cells, and neutrophils per epithelial basal lamina surface area. These observations imply that chronic O3 exposure causes a concentration-dependent reactive peribronchiolar inflammatory response and an adaptive response consisting of hypertrophy and hyperplasia of the nonciliated bronchiolar cell. Fujinaka et al. (1985) quantitated the response of RB epithelium and peribronchiolar connective tissue (PCT) to chronic exposure to high ambient levels of O3. Adult male bonnet monkeys were exposed 8 h daily for 1 year to either 640 ppb or filtered air. Significant exposure-related changes were greater volume of RB in the lung, smaller diameter of RB lumen, thicker media and intima of peribronchiolar arterioles, thicker RB epithelium, and thicker PCT. Cellular numerical density increased in cuboidal bronchiolar cells and decreased in type I pneumocytes. Cell volume increases occurred in cuboidal bronchiolar, ciliated, and type 2 cells. PCT changes included more amorphous extracellular matrix, neutrophils, and lymphocytes/plasma cells. It was concluded that centriacinar changes

916

OZONE

caused by long-term exposure to high ambient O3 in bonnet monkeys results in narrowing of RBs primarily by peribronchiolar inflammation (inflammatory cells, fibers, and amorphous matrix) and secondarily through hyperplasia of cuboidal bronchiolar cells. The effects of chronic O3 exposure on lung collagen cross-linking were investigated by Reiser et al. (1987) in two groups of juvenile cynomolgus monkeys exposed to 610 ppb for 8 h/day for 1 year. One group was killed immediately after the exposure period; the second exposed group breathed filtered air for 6 months after the O3 exposure before being killed. Previous studies of these monkeys had revealed that lung collagen content was increased in both exposed groups (Last et al., 1984) in this study. The changes in the group killed at the termination of exposure were characteristic of those seen in lung tissue in the acute stage of experimental pulmonary fibrosis. Although the changes seen in the postexposure group suggest that the lung collagen being synthesized at the time the animals were killed was normal, “abnormal” collagen synthesized during the period of O3 exposure was irreversibly deposited in the lungs. This study suggests that long-term exposure to relatively low levels of O3 may cause irreversible changes in lung collagen structure. Barr et al. (1988) exposed rats to either filtered air or 950 ppb O3 8 h daily for 90 days and examined the centriacinar region of lungs morphologically and morphometrically. After chronic O3 exposure, there was a decrease in terminal bronchiole luminal diameter but no change in total terminal bronchiole volume. The most notable change was a 3.4-fold increase in RB volume. They concluded that RB is formed from the centriacinar alveolar duct. Morphologic parameters supporting this conclusion included the presence of fused basement membrane beneath reactive bronchiolar epithelium in the RB, the presence of similar basal laminar changes in both the RB and the proximal alveolar duct septal tips, and the observation that most severe epithelial damage and inflammation occurred in the most proximal alveolar duct rather than in the terminal bronchiole. The severe injury within the acinus shifts distally as RB segments are formed. Hence, most of the damage occurs at the tips of alveolar septa at the RB alveolar duct junction. The issue of the effects of chronic O3 exposure during childhood on lung development was investigated by Tyler et al. (1987) in studies in 28-day-old rats exposed to filtered air or to 640 or 960 ppb O3, 8 h/night, for 42 nights. A second control group was fed ad libitum and exposed to only filtered air. Half the rats were studied at the end of the 42-night exposures, the rest after a 42-day postexposure period during which all rats were fed ad libitum and breathed filtered air. Rats examined at the end of the exposure period had larger saline and fixed lung volumes. These larger lungs had greater volumes of parenchyma, alveoli, and respiratory bronchioles. Some of these changes persisted throughout a 42-day postexposure period. Thus, O3 inhalation by young rats alters lung growth and development in ways likely to be detrimental, and these changes persist after O3 inhalation stops. In summary, chronic exposures to ambient air appear to produce a functional adaptation that persists for at least a few months after the end of the O3 season but dissipates by the spring. Several population-based studies of lung function indicate that there may be an accelerated aging of the lung associated with living in communities with persistently elevated ambient O3, but the limited ability to accurately assign exposure classifications of the various populations in these studies makes a cautious assessment of these data prudent. The plausibility of accelerated aging of the human lung from chronic O3 exposure is greatly enhanced by the results of a series of recent chronic animal exposure studies in rats and monkeys (especially those in rats of Huang et al. (1988) and Grose et al. (1989) using a daily cycle with a 180 ppb average over 9 h superimposed on a 13 h base of 60 ppb and those in monkeys of Hyde et al. (1989) and Tyler et al. (1988) using 8 h/day of 150 and 250 ppb).

AMBIENT AIR QUALITY STANDARDS AND GUIDELINES

917

The persistent cellular and morphometric changes produced by these exposures in the terminal bronchioles and proximal alveolar region and the functional changes consistent with a stiffening of the lung reported by Raub et al. (1983) and Tyler et al. (1988) are certainly consistent with the results of the epidemiological studies.

23.10 AMBIENT AIR QUALITY STANDARDS AND GUIDELINES The U.S. Occupational Safety and Health Administration’s permissible exposure limit (PEL) for O3 is 100 parts per billion (ppb), equivalent to 235 mg/m3, as a time-weighted average for 8 h/day, along with a short-term exposure limit of 300 ppb for 15 min (U.S. DOL, 1989). The American Conference of Governmental Industrial Hygienists (ACGIH, 2008) threshold limit value (TLV) for occupational exposure is 100 ppb as an 8 h time-weighted average for light work; 80 ppb for moderate work; and 50 ppb for heavy work. For a 2 h exposure at any workload, the TVL is 200 ppb. The initial primary (health-based) National Ambient Air Quality Standard, established by the Environmental Protection Agency in 1971, was 80 ppb of total oxidant as a 1 h maximum not to be exceeded more than once per year. The NAAQS was revised in 1979 to 120 ppb of O3 as a 1 h maximum not to be exceeded more than four times in 3 years. This initial revision was based on clinical studies by DeLucia and Adams (1977), showing that exercising asthmatic adults exposed for 1 h to 150 ppb in a test chamber had increased cough, dyspnea, and wheezing, along with small but nonsignificant reductions in pulmonary function (U.S. EPA, 1986). A small margin of safety was applied to protect against adverse effects not yet uncovered by research and effects whose medical significance is a matter of disagreement. EPA initiated a review of the 1979 NAAQS in 1983, completed a criteria document for ozone in 1986, and updated it in 1992 (U.S. EPA, 1992). EPA decided, in March 1993, to maintain the existing standard, but to proceed as rapidly as possible with the next round of review. This expedited review was completed with the publication of both a new criteria document (U.S. EPA, 1996) and staff paper (U.S. EPA, 1996) in 1996. In July 1997, the EPA administrator promulgated a revised primary O3 NAAQS of 80 ppb as an 8 h timeweighted average daily maximum, with no more than four annual exceedances, and averaged over 3 years. The change from one allowable annual exceedance to four was to minimize the designation of NAAQS nonattainment in a community that was triggered by rare meteorological conditions especially conducive to O3 formation. The switch to an 8 h averaging time recognized that ambient O3 in much of the United States has broad daily peaks and that human responses are more closely related to the total daily exposure than to brief peaks of O3 exposure. Since the 120 ppb, 1 h average, one exceedance NAAQS was approximately equivalent to an 8 h average, four exceedance NAAQS at a concentration a bit below 90 ppb in average stringency in the United States as a whole, the 1997 NAAQS represents about a 10% reduction in permissible O3 exposure. On March 12, 2008, the EPA administrator issued revised O3 NAAQS to replace the ones promulgated in 1997. The primary (health based) NAAQS, with an 8-hour averaging time, was reduced to 0.075 ppm, and once again the secondary (welfare-based) NAAQS was made equal to the primary NAAQS. Since the official interpretation of the NAAQS at 0.08 ppm was that an exceedance did not occur unless the concentration was greater than 84 ppb, the specification of the NAAQS to three significant figures meant a tightening of the daily NAAQS limit by about 10%. In promulgating the 2008 O3 NAAQS, the administrator disregarded

918

OZONE

the unanimous recommendations of his Clean Air Scientific Advisory Committee (CASAC) that the primary standard be set at a level between 0.060 to 0.070 ppm and that the secondary standard should be based on a seasonal form that better reflected foliar damage and productivity (Letter Report: EPA-CASAC-08-009). The World Health Organization (WHO) has developed Air Quality Guidelines for O3 to assist member states in establishing their own standards. 23.10.1

Guidelines

The second edition of the WHO AQG set the guideline value for O3 at 120 mg/m3 for an 8 h daily average. Since the mid-1990s, there has been no major addition to the evidence from chamber or field studies. There has, however, been a marked increase in health effects

TABLE 23.4

WHO Ozone Air Quality Guideline and Interim Target Daily Maximum 8 h Mean

High level

3

240 mg/m (120 ppb) 160 mg/m3 (80 ppb)

WHO interim target-1 (IT-1)

WHO Air Quality Guidelines (AQGs)

100 mg/m3 (50 ppb)

Effects at the Selected Ozone Level Significant health effects; substantial proportion of vulnerable population affected. Important health effects, an intermediate target for populations with 03 concentrations above this level. Does not provide adequate protection to public health. Rationale: . Lower level 6.6 h chamber exposures of healthy exercising young adults, who show physiological and inflammatory lung effects. . Ambient level at various summer camp studies showing effects on the health of children. . Estimated 3–5% increase in daily mortalitya (based on findings of daily time-series studies). . This concentration will provide adequate protection to public health, though some health effects may occur below this level. Rationale: Estimated 1–2% increase in daily mortalitya (based on findings of daily time-series studies). . Extrapolation from chamber and field studies based on the likelihood that real-life exposure tends to be repetitive and chamber studies do not study highly sensitive or clinically compromised subjects or children. . Likelihood that ambient O is a marker for 3 related oxidants. .

a Deaths attributable to ozone concentrations above estimated baseline of 70 mg/m3. Based on the range of 0.3–0.5%, increase in daily mortality for an increase of 10 mg/m3 8 h O3.

AMBIENT AIR QUALITY STANDARDS AND GUIDELINES

919

evidence from epidemiological time-series studies. Combined evidence from those studies show convincing, though small, positive associations between daily mortality and O3 levels, independent of the effects of particulate matter. Similar associations have been observed in both North America and Europe. These time-series studies have shown effects at O3 concentrations below the previous guideline of 120 mg/m3 (60 ppb) without clear evidence of a threshold. Evidence from both chamber and field studies also indicate that there is a considerable individual variation in response to O3. In view of these considerations, there is a good case for reducing the WHO AQG from the existing level of 120 mg/m3. In 2005, the WHO air quality guidelines for O3 was set at the level of ozone: 100 mg/m3 (50 ppb) for daily maximum 8 h mean (WHO, 2006). See Table 23.4. WHO acknowledged that it is possible that health effects will occur below this level in some sensitive individuals. Based on time-series studies, the number of attributable deaths brought forward can be estimated at 1–2% on days when O3 concentrations reaches this guideline level compared to the background O3 level. There is some evidence that O3 also represents unmeasured toxic oxidants arising from similar sources. Measures to control O3 are also likely to control the effects of these pollutants. Hemispheric background concentrations of tropospheric O3 vary in time and space, but can reach average levels of around 80 mg/m3 (40 ppb). These arise from both anthropogenic and biogenic emissions of O3 precursors and downward intrusion of stratospheric O3 into the troposphere. The proposed guideline value may occasionally be exceeded due to natural causes. There is some evidence that long-term exposure to O3 may have chronic effects, but it is not sufficient to recommend an annual guideline. As concentrations increase above the guideline value, health effects at the population level become increasingly numerous and severe. Such effects can occur in places where concentrations are currently high due to human activities or during episodes of very hot weather. The 8 h interim target-1 level has been set at 160 mg/m3 (80 ppb) at which measurable, though transient, changes in lung function and lung inflammation among some healthy young adults have been shown with intermittent exercise in controlled chamber tests. Although some would argue that these responses may not be adverse effects and that they were seen only with vigorous exercise, these views are counterbalanced by the possibility that there are substantial numbers of persons in the general population, including persons of different ages, pre-existing health status, and coexposures that might be more susceptible than the relatively young and generally healthy subjects who were studied. Furthermore, chamber studies provide little evidence about repeated exposure. The exposure to 160 mg/m3 is also likely to be associated with the same effects noted at 100 mg/m3. Based on time-series evidence, attributable deaths can be estimated at 3–5% for daily exposures above the estimated background. At concentrations exceeding 240 mg/m3 (120 ppb), important health effects are likely. This is based on findings from a large number of clinical inhalation and field studies. Both healthy adults and asthmatics would experience significant reductions in lung function as we all as airway inflammation that would cause symptoms and alter performance. There are additional concerns about increased respiratory morbidity in children. Based on time-series evidence, attributable deaths can be estimated at 5–9% for daily exposures above the estimated background. The WHO guidelines and interim targets for ozone are summarized in Table 27.4.

920

OZONE

23.11 SUMMARY AND CONCLUSIONS The apparently reversible effects that have followed acute exposures lasting from 5 min to 6.6 h include changes in lung capacity, flow resistance, epithelial permeability, and reactivity to bronchoactive challenges. These effects may persist for many hours or days after the exposure ceases. Repetitive daily exposures over several days or weeks can exacerbate and prolong these effects. Most of the data available on transient functional effects of O3 were obtained from controlled human exposure studies and field studies of limited duration. Such studies can provide information on chronic pollutant effects only to the extent that prior exposures affect the transient response to single-exposure challenges. Furthermore, interpretation of the results of such tests is limited by our generally inadequate ability to characterize the nature and/or magnitude of the prior chronic exposures. Most of the limited data we have on the effects of chronic O3 exposures on humans come from epidemiological studies. Epidemiological studies offer the prospect of establishing chronic health effects of longterm O3 exposure in relevant populations and offer the possibility that the analyses can show the influence of other environmental factors on responses to O3 exposure. However, the strengths of any of the associations may be difficult to firmly establish because of the complications introduced by uncontrolled cofactors that may confound or obscure the underlying causal factors. The most convenient and efficient way to study mechanisms and patterns of response to inhaled O3 and of the influence of other pollutants and stresses on these responses is by controlled exposures of laboratory animals. One can study the transient functional responses to acute exposures and establish the interspecies differences in response among different animal species and between them and humans similarly exposed. One can also look for responses that require highly invasive procedures or serial sacrifice and gain information that cannot be obtained from studies on human volunteers. Finally, one can use long-term exposure protocols to study cumulative responses and the pathogenesis of chronic diseases in animals. Other advantages of studies on animals are the ability to examine the presence of and basis for variations in response that are related to age, sex, species, strain, genetic markers, nutrition, the presence of other pollutants, and so on. Among the significant limitations to the use of exposure–response data from animal studies in human risk assessments is our quite limited ability to interpret the responses in relation to likely responses in humans who might be exposed to the same or lower levels. Controlled chronic exposure protocols can be very expensive, limiting the number of variables that can be effectively examined in any given study. For studies focused on the biochemical mechanisms of epithelial cells’ responses to O3, cells can be harvested from humans or animals and exposed to O3 in vitro. Interspecies comparisons of cellular response can often be made, and relatively few animals can provide much study material. However, our ability to interpret the results of in vitro assays in relation to likely effects in humans in vivo is quite limited, even when the studies are done with human cells. The cellular response in vitro may differ from that of the same cells in vivo, and the in vivo controls on cellular metabolism and function, which may play a significant role in the overall response, are absent. In terms of functional effects, we know that single O3 exposures to healthy nonsmoking young adults at concentrations in the range of 80–200 ppb produce a complex array of pulmonary responses including decreases in respiratory function and athletic

SUMMARY AND CONCLUSIONS

921

performance and increases in symptoms, airway reactivity, neutrophil content in lung lavage, and rate of mucociliary particle clearance. Responses to O3 in purified air in chambers occur at concentrations of 80 or 100 ppb when the exposures involve moderate exercise over 6 h or more and require concentrations of 180 or 200 ppb when the duration of exposure is 2 h or less. However, mean FEV1 decrements more than 5% have been seen at 100 ppb of O3 in ambient air for children at summer camps and for adults engaged in outdoor exercise for only 1/2 h. The apparently greater responses to O3 in ambient air may be related to the presence of, or prior exposures to, acidic aerosol, but further investigation of this hypothesis is needed. Further research is also needed to establish the interrelationships between small transient functional decrements, such as FEV1, PEFR, and mucociliary clearance rates, which may not, in themselves, be adverse effects, and changes in symptoms, performance, reactivity, permeability and neutrophil counts. The latter may be more closely associated with adversity or in the accumulation or progression of chronic lung damage. Successive days of exposure of adult humans in chambers to O3 at current high ambient levels leads to a functional adaptation in that the responses are attenuated by the third day and are negligible by the fifth day. However, a comparable functional adaptation in rats does not prevent the progressive damage to the lung epithelium. Daily exposures of animals also increase other responses in comparison to single exposures, such as a loss of cilia, a hypertrophic response of Clara cells, alterations in macrophage function, and alterations in the rates of particle clearance from the lungs. The clearest evidence that current ambient levels of O3 are closely associated with health effects in human populations comes from epidemiological studies focused on acute responses. The 1997 revision to the O3 NAAQS relied heavily for its quantitative basis on a study of emergency hospital admissions for asthma in New York City (Thurston et al., 1992) and its consistency with other time-series studies of hospital admissions for respiratory diseases in Toronto, all of Southern Ontario, in Montreal, Canada; and in Detroit and Buffalo in the United States. However, other acute responses, while less firmly established on quantitative bases, are also occurring. Chronic human exposures to ambient air appear to produce a functional adaptation that persists for at least a few months after the end of the O3 season but dissipates by the spring. Several population-based studies of lung function indicate that there may be an accelerated aging of the lung associated with living in communities with persistently elevated ambient O3, but the limited ability to accurately assign exposure classifications of the various populations in these studies makes a cautious assessment of these provocative data prudent. The plausibility of accelerated aging of the human lung from chronic O3 exposure is greatly enhanced by the results of a series of chronic animal exposure studies in rats and monkeys. There is little reason to expect humans to be less sensitive than rats or monkeys. On the contrary, humans have a greater dosage delivered to the respiratory acinus than do rats for the same exposures. Another factor is that the rat and monkey exposures were to confined animals with little opportunity for heavy exercise. Thus, humans who are active outdoors during the warmer months may have greater effective O3 exposures than the test animals. Finally, humans are exposed to O3 in ambient mixtures. The potentiation of the characteristic O3 responses by other ambient air constituents seen in the short-term exposure studies in humans and animals may also contribute to the accumulation of chronic lung damage from long-term exposures to ambient air containing O3.

922

OZONE

The lack of a more definitive database on the chronic effects of ambient O3 exposures on humans is a serious failing that must be addressed with a long-term research program. The potential impacts of such exposures on public health deserve serious scrutiny and, if they turn out to be substantial, strong corrective action. Further controls on ambient O3 exposure will be extraordinarily expensive and will need to be very well justified. In summary, this review has shown that (1) the control of ambient O3 to levels within the current NAAQS presents an intractable problem; (2) the current NAAQS contains little, if any, margin of safety against effects considered to be adverse; and (3) a large fraction of U.S. population resides in communities that exceed the O3 NAAQS. Thus, it is important that health scientists and control agency personnel understand the nature and extent of human exposures and the effects they produce in order to communicate health risks effectively to the public and to help prioritize feasible options for reducing exposures. ACKNOWLEDGMENT This research is part of a Center program supported by Grant ES 00260 from the National Institute of Environmental Health Sciences. REFERENCES Abbey DE, Lebowitz MD, Mills PK, Petersen FF, Beeson WL, Burchette RJ (1995) Long-term ambient concentrations of particulates and oxidants and development of chronic disease in a cohort of nonsmoking California residents. Inhal. Toxicol. 7:19–34. Abraham WM, Delehunt JC, Yerger L, Marchete B, Oliver W Jr (1984) Changes in airway permeability and responsiveness after exposure to ozone. Environ. Res. 34:110–119. ACGIH (2008) Threshold Limit Values and Biological Exposure Indices for 2008. Cincinnati: American Conference of Governmental Industrial Hygienists. Adams WC, Schelegle ES (1983) Ozone and high ventilation effects on pulmonary function and endurance performance. J. Appl. Physiol. Respir. Environ. Exercise Physiol. 55:805–812. Adams WC (2002) Comparison of chamber and face-mask 6.6-hour exposures to ozone on pulmonary function and symptoms responses. Inhal. Toxicol. 14:745–764. Adams WC (2003) Comparison of chamber and face-mask 6.6-hour exposures to 0.08 ppm ozone via square-wave and triangular profiles on pulmonary responses. Inhal. Toxicol. 15:265–281. Adams WC (2006) Comparison of chamber 6.6-hour exposures to 0.04–0.08 ppm ozone via squarewave and triangular profiles on pulmonary responses. Inhal. Toxicol. 18:127–136. Alexis N, Urch B, Tarlo S, Corey P, Pengelly D, O’Byrne P (2000) Cyclooxygenase metabolites play a different role in ozone-induced pulmonary function decline in asthmatics compared to normals. Inhal. Toxicol. 12:1205–1224. Altshuller AP (1977) Eye irritation as an effect of photochemical air pollution. J. Air Pollut. Control Assoc. 27:1125–1126. Altshuller AP (1987) Estimation of the natural background of ozone present at surface rural locations. J. Air Pollut. Control Assoc. 37:1409–1417. Amoruso MA, Goldstein BD (1988) Effect of 1, 3, and 6 hour ozone exposure on alveolar macrophages superoxide production. Toxicologist 8:197 Amoruso MA, Witz G, Goldstein BD (1981) Decreased superoxide anion radical production by rat alveolar macrophages following inhalation of ozone or nitrogen dioxide. Life Sci .12:2215–2221.

REFERENCES

923

Anderson HR, Ponce de Leon A, Bland MJ, Bower JS, Strachan DP (1996) Air pollution and daily mortality in London: 1987–1992. Brit. Med. J. 312:665–669. Aranyi C, Vana SC, Thomas PT, Bradof JN, Fenters JD, Graham JA, Miller FJ (1983) Effects of subchronic exposure to a mixture of O3, SO2, and (NH4)2SO4 on host defenses of mice. J. Toxicol. Environ. Health 12:55–71. Aris R, Christian D, Sheppard D, Balmes JR (1991) The effects of sequential exposure to acidic fog and ozone on pulmonary function in exercising subjects. Am. Rev. Respir. Dis. 143:85–91. Arsalane K, Broeckaert F, Knoops B, Clippe A, Buchet JP, Bernard A (1999) Increased serum and urinary concentrations of lung Clara cell protein in rats acutely exposed to ozone. Toxicol. Appl. Pharmacol. 159:169–174. Asplund PT, Ben-Jebria A, Rigas ML, Ultman JS (1996) Longitudinal distribution of ozone absorption in the lung: effect of continuous inhalation exposure. Arch. Environ. Health 51:431–438. Avol EL, Linn WS, Venet TG, Shamoo DA, Hackney JD (1984) Comparative respiratory effects of ozone and ambient oxidant pollution exposure during heavy exercise. J. Air Pollut. Control Assoc. 34:804–809. Avol EL, Linn WS, Shamoo DA, Valencia LM, Anzar UT, Hackney JD (1985) Respiratory effects of photochemical oxidant air pollution in exercising adolescents. Am. Rev. Respir. Dis. 132:619– 622. Avol EL, Linn WS, Shamoo DA, Spier CE, Valencia LM, Venet TG, Trim SC, Hackney JD (1987) Short-term respiratory effects of photochemical oxidant exposure in exercising children. J. Air Pollut. Control Assoc. 37:158–162. Avol EL, Gauderman WJ, Tan SM, London SJ, Peters JM (2001) Respiratory effects of relocating to areas of differing air pollution levels. Am. J. Respir. Crit. Care Med. 164:2067–2072. Balmes JR, Chen LL, Scannell C, Tager I, Christian D, Hearne PQ, Kelly T, Aris RM (1996) Ozoneinduced decrements in FEV1 and FVC do not correlate with measures of inflammation. Am. J. Respir. Crit. Care Med. 153:904–909. Barr BC, Hyde DM, Plopper CG, Dungworth DL (1988) Distal airway remodeling in rats chronically exposed to ozone. Am. Rev. Respir. Dis. 137:924–938. Barry BE, Miller EF, Crapo JD (1985) Effects of inhalation of 0.12 and 0.5 ppm ozone on the proximal alveolar region of juvenile and adult rats. Lab. Invest. 53:692–704. Barry BE, Mercer RR, Miller FJ, Crapo JD (1988) Effects of inhalation of 0.25 ppm ozone on the terminal bronchioles of juvenile and adult rats. Exp. Lung Res. 14:225–245. Bates DV (1989) Ozone-myth and reality. Environ. Res. 50:230–237. Bates DV (1995) Ozone: a review of recent experimental, clinical and epidemiological evidence, with notes on causation. Can. Respir. J. 2: 25–31 and 161–171. Bates DV, Sizto R (1989) The Ontario air pollution study: identification of the causative agent. Environ. Health Perspect. 79:69–72. Beckett WS, McDonnell WF, Horstman DH, House DE (1985) Role of the parasympathetic nervous system in acute lung response to ozone. J. Appl. Physiol. 59:1879–1885. Beckett WS, Freed AN, Turner C, Menkes HA (1988) Prolonged increased responsiveness of canine peripheral airways after exposure to O3. J. Appl. Physiol. 64(2):605–610. Bedi JF, Horvath SM, Drechsler-Parks DM (1989) Adaptation by older individuals repeatedly exposed to 0.45 parts per million ozone for two hours. J. Air Pollut. Control Assoc. 39:194–199. Bell ML, Dominici, F, Samet JM (2005) A meta-analysis of time-series studies of ozone and mortality with comparison to the national morbidity, mortality, and air pollution study. Epidemiology. 16:436–445. Bhalla DK, Young C (1992) Effects of acute exposure to O3 on rats: sequence of epithelial and inflammatory changes in the distal airways. Inhal. Toxicol. 4:17–31.

924

OZONE

Bhalla DK, Manniz RC, Lavan SM, Phalen RF, Kleinman MT, Crocker TT (1987) Tracheal and bronchoalveolar permeability changes in rats inhaling oxidant atmospheres during rest of exercise. J. Toxicol. Environ. Health 22:417–437. Brauer M, Blair J, Vedal S (1996) Effect of ambient ozone exposure on lung function in farm workers. Am. J. Respir. Crit. Care Med. 154:981–987. Braun-Fahrlander C, Kunzli N, Domenighetti G, Carell CF, Ackermann-Liebrich U (1994) Acute effects of ambient ozone on respiratory function of Swiss schoolchildren after a 10-minute heavy exercise. Pediat. Pulmo. 17:169–177. Broeckaert F, Arsalane K, Hermans C, Bergamaschi E, Brustolin A, Mutti A, Bernard A (2000) Serum Clara cell protein: a sensitive biomarker of increased lung epithelium permeability caused by ambient ozone. Environ. Health Perspect. 108:533–537. Brookes KA, Adams WC, Schelegle ES (1989) 0.35 ppm O3 exposure induces hyperresponsiveness on 24-h reexposure to 0. 20 ppm O3. J. Appl. Physiol. 66:2756–2762. Brown JS, Bateson TF, McDonnell WF (2008). Effects of exposure to 0.06 ppm ozone on FEV1 in humans: a secondary analysis of existing data. Environ. Health Perspect. 116:1023–1026. Brunekreef B, Hoek G, Breugelmans O, Leentvaar M (1994) Respiratory effects of low-level photochemical air pollution in amateur cyclists. Am. J. Respir. Crit. Care Med. 150:962–966. Burnett RT, Dales RE, Raizenne ME, Krewski D, Summers PW, Roberts GR, Raad-Young M, Dann T, Brook J (1994) Effects of low ambient levels of ozone and sulfates on the frequency of respiratory admissions to Ontario hospitals. Environ. Res. 65:172–194. Burnett RT, Brook JR, Yung WT, Dales RE, Krewski D (1997) Association between ozone and hospitalization for respiratory diseases in 16 Canadian cities. Environ. Res. 72:24–31. Burnett RT, Smith-Doiron M, Stieb D, Raizenne ME, Brook JR, Dales RE, Leech JA, Cakmak S, Krewski D (2001) Association between ozone and hospitalization for acute respiratory diseases in children less than 2 years of age. Am. J. Epidemiol. 153:444–452. Castillejos M, Gold DR, Damokosh AI, Serrano P, Allen G, McDonnell WF, Dockery DW, Velasco SR, Hernandez M, Hayes C (1995) Acute effects of ozone on the pulmonary function of exercising schoolchildren from Mexico City. Am. J. Respir. Crit. Care Med. 152:1501–1507. Chang L, Miller FJ, Ultman J, Huang Y, Stockstill BL, Grose E, Graham JA, Ospital JJ, Crapo JD (1991) Alveolar epithelial cell injuries by subchronic exposure to low concentrations of ozone correlate with cumulative exposure. Toxicol. Appl. Pharmacol. 109:219–234. Chang L-Y, Huang Y, Stockstill BL, Graham JA, Grose EC, Menache MG, Miller FJ, Costa DL, Crapo JD (1992) Epithelial injury and interstitial fibrosis in the proximal alveolar regions of rats chronically exposed to a simulated pattern of urban ambient ozone. Toxicol. Appl. Pharmacol. 115:241–252. Cody RP, Weisel CP, Birnbaum G, Lioy PJ (1992) The effect of ozone associated with summertime photochemical smog on the frequency of asthma visits to hospital emergency departments. Environ. Res. 58:184–194. Coffin DL, Gardner DE (1972) Interaction of biological agents and chemical air pollutants. Ann. Occup. Hyg. 15:219–235. Coffin DL, Blommer EJ, Gardner DE, Holzman R (1967) Effect of air pollution on alteration of susceptibility to pulmonary infection. In:Proceedings of the Third Annual Conference on Atmospheric Contamination in Confined Spaces. Report No. AMRL-TR-67-200. Aerospace Medical Research Laboratories, Wright-Patterson Air Force Base, OH. (Available from NTIS, Springfield, VA, AD-835008.) Coffin DL, Gardner DE, Holzman RS, Wolock FJ (1968) Influence of ozone on pulmonary cells. Arch. Environ. Health 16:633–636. Coleridge HM, Coleridge JCG, Ginzel KH, Baker DG, Banzett RB, Morrison MA (1976) Stimulation of “irritant” receptors and afferent C-fibres in the lungs of prostaglandins. Nature 264:451–453.

REFERENCES

925

Cosio MG, Hole KA, Niewohner DE (1980) Morphologic and morphometric effects of prolonged cigarette smoking on the small airways. Am. Rev. Respir. Dis. 122:265–271. Costa DL, Hatch GE, Highfill J, Stevens MA, Tepper JS (1989) Pulmonary function studies in the rat addressing concentration vs. time relationships for ozone. In: Schneider T, Lee SD, Wolters GJR, Grant LD, editors. Atmospheric Ozone Research and Its Policy Implications. Nijmegen, The Netherlands: Elsevier. Delfino RJ, Becklake MR, Hanley JA (1994a) The relationship of urgent hospital admissions for respiratory illnesses to photochemical air pollution levels in Montreal. Environ. Res. 67:1–19. Delfino RJ, Becklake MR, Hanley JA (1994b) Estimation of unmeasured particulate air pollution for an epidemiological study of daily respiratory morbidity. Environ. Res. 67:20–38. DeLucia AJ, Adams WC (1977) Effects of O3 inhalation during exercise on pulmonary function and blood biochemistry. J. Appl. Physiol. 43:75–81. Desqueyroux H., Pujet, J.-C., Prosper, M., LeMoullec, Y., Momas, I. 2002. Effects of air pollution on adults with chronic obstructive pulmonary disease. Arch. Environ. Health 57:554–560. Devlin RB, McDonnell WF, Mann R, Becker S, House DE, Schreinemachers D, Koren HS (1991) Exposure of humans to ambient levels of ozone for 6.6 hours causes cellular and biochemical changes in the lung. Am. J. Respir. Cell Mol. Biol. 4:72–81. Dockery DW, Pope CA III, Xu X, Spengler JD, Ware JH, Fay ME, Ferris BG Jr, Speizer FE (1993) An association between air pollution and mortality in six U.S. Cities. N. Engl. J. Med. 329:1753– 1759. Dowell AR, Lohrbauer LA, Hurst D, Lee SD (1970) Rabbit alveolar macrophage damage caused by in vivo ozone inhalation. Arch. Environ. Health 21:121–127. Drechsler-Parks DM, Bedi JF, Horvath SM (1987) Pulmonary function responses of older men and women to ozone exposure. Exp. Gerontol. 22:91–101. Driscoll KE, Vollmuth TA, Schlesinger RB (1986) Early alveolar clearance of particles in rabbits undergoing acute and subchronic exposure to ozone. Fund. Appl. Toxicol. 7:264–271. Driscoll KE, Vollmuth TA, Schlesinger RB (1987) Acute and subchronic ozone inhalation in the rabbit: response of alveolar macrophages. J. Toxicol. Environ. Health 21:27–43. Ehrlich R (1980) Interaction between environmental pollutants and respiratory infections. Environ. Health Perspect. 35:89–100. Ehrlich R, Findlay JC, Fenters JD, Gardner DE (1977) Health effects of short-term inhalation of nitrogen dioxide and ozone mixtures. Environ. Res. 14:223–31. Ehrlich R, Findlay JC, Gardner DE (1979) Effects of repeated exposures to peak concentrations of nitrogen dioxide and ozone on resistance to streptococcal pneumonia. J. Toxicol. Environ. Health 5:631–642. Emmons K, Foster WM (1991) Smoking cessation and acute airway response to ozone. Arch. Environ. Health 46:288–295. Farman CA, Pinkerton KE, Rajini P, Witschi H, Last JA (1997) Evolution of lung lesions in rats exposed to mixtures of ozone and nitrogen dioxide. Inhal. Toxicol. 9:647–677. Farrell BP, Kerr HD, Kulle TJ, Sauder LR, Young JL (1979) Adaptation in human subjects to the effects of inhaled ozone after repeated exposure. Am. Rev. Respir. Dis. 119:725–730. Folinsbee LJ, Hazuch MJ (1989) Persistence of ozone-induced changes in lung function and airway responsiveness. In: Schneider T, Lee SD, Wolters GJR, Grant L, editors Atmospheric Ozone Research and its Policy Implications. Nijmegen, The Netherlands: Elsevier. Folinsbee LJ, Bedi JF, Horvath SM (1980) Respiratory responses in humans repeatedly exposed to low concentrations of ozone. Am. Rev. Respir. Dis. 121:431–439. Folinsbee LJ, Bedi JF, Horvath SM (1984) Pulmonary function changes in trained athletes following 1-hour continuous heavy exercise while breathing 0.21 ppm ozone. J. Appl. Physiol. Respir. Environ. Exercise Physiol. 57:984–988.

926

OZONE

Folinsbee LJ, McDonnell WF, Horstman DH (1988) Pulmonary function and symptom responses after 6.6 hour exposure to 0.12 ppm ozone with moderate exercise. J. Air Pollut. Control Assoc. 38:28–35. Folinsbee LJ, Horstman DH, Kehrl HR, Harder S, Salaam SA, Ives PJ (1994) Respiratory response to repeated prolonged exposure to 0.12 ppm ozone. Am. J. Respir. Crit. Care Meditor. 149:98–105. Foster WM, Stetkiewicz PT (1996) Regional clearance of solute from the respiratory epithelia: 18– 20 hr postexposure to ozone. J. Appl. Physiol. 81:1143–1149. Foster WM, Costa DL, Langenback EG (1987) Ozone exposure alters tracheobronchial mucociliary function in humans. J. Appl. Physiol. 63:996–1002. Frager NB, Phalen RF, Kenoyer JL (1979) Adaptation to ozone in reference to mucociliary clearance. Arch. Environ. Health 34:51–57. Frampton MW, Morrow PE, Cox C, Levy PC, Condemi JJ, Speers D, Gibb FR, Utell MJ (1995) Sulfuric acid aerosol followed by ozone exposure in healthy and asthmatic subjects. Environ. Res. 69:1–14. Frampton MW, Morrow PE, Torres A, Cox C, Voter KZ, Utell MJ (1997) Ozone responsiveness in smokers and nonsmokers. Am. J. Respir. Crit. Care Med. 155:116–121. Frank R, Liu MC, Spannhake EW, Mlynarek S, Macri K, Weinmann GG (2001) Repetitive ozone exposures of young adults: evidence of persistent small airway dysfunction. Am. J. Respir. Crit. Care Med. 164:1253–1260. Frischer TM, Kuehr jPullwitt A, Meinert R, Forster J, Studnicka M, Koren H (1993) Ambient ozone causes upper airways inflammation in children. Am. Rev. Respir. Dis. 148:961–964. Fujinaka LE, Hyde DM, Plopper CG, Tyler WS, Dungworth DL, Lollini LO (1985) Respiratory bronchiolitis following long-term ozone exposure in bonnet monkeys: a morphometric study. Exp. Lung Res. 8:167–190. Gardner DE, Miller FJ, Illing JW, Kirtz JM (1977) Increased infectivity with exposure to ozone and sulfuric acid. Toxicol. Lett. 1:59–64. Gauderman WJ, McConnel R, Gilliland F, London S, Thomas D, Avol E, Vora H, Berhane K, Rapaport EB, Lurmann F, Margolis HG, Peters JM (2000) Association between air pollution and lung function growth in Southern California children. Am J. Respir. Crit. Care Med. 162:1383– 1390. Gauderman WJ, Gilliland FD, Avol E, Vora H, Thomas DT, Rappaport E, Berhane K, Lurmann F, McConnell R, Sram D, Perters JM (2002) Association between air pollution and lung function growth in Southern California children: results from a second cohort. Am. J. Respir. Crit. Care Med. 166 (1):74–84. Gauderman WJ, Avol E, Gilliland F, vora H, Thomas D, Berhane K, McConnel R, Kunzli N, Lurmann F, Rappaport E, Margolis H, Bates D, Peters M (2005) The effect of air pollution on lung development from 10–18 years of age. N. Engl. J. Meditor. 351:1057–1067. Gerrity TR, Wiester MJ (1987) Experimental measurements of the uptake of ozone in rats and human subjects. Preprint 87–99. 3.1987 Annual Meeting of the Air Pollution Control Association, New York, June. Gerrity TR, Weaver RA, Berntsen J, House DE, O’Neil JJ (1988) Extrathoracic and intrathoracic removal of O3 in tidal-breathing humans. J. Appl. Physiol. 65:393–400. Gerrity TR, Biscardi F, Strong A, Garlington AR, Brown JS, Bromberg PA (1995) Bronchoscopic determination of ozone uptake in humans. J. Appl. Physiol. 79:852–860. Gilliland FD, Berhane K, Rappaport E, Thomas DC, Avole E, Gauderman WJ, London SJ, Margolis HG, McConnel R, Islam T, Peters JM (2001) The effects of ambient air pollution on school absenteeism due to respiratory illnesses. Epidemiology 12:43–54. Gilmour MI, Selgrade MK (1993) A comparison of the pulmonary defenses against streptococcal infection in rats and mice following O3 exposure. Toxicol. Appl. Pharmacol. 123:211–218.

REFERENCES

927

Goldstein E, Tyler WS, Hoeprich PD, Eagle C (1971a) Ozone and the antibacterial defense mechanisms of the murine lung. Arch. Intern. Meditor. 128:1099–1102. Goldstein E, Tyler WS, Hoeprich PD, Eagle C (1971b) Adverse influence of ozone on pulmonary bactericidal activity of murine lungs. Nature 229:262–263. Goldstein BD, Hamburger SJ, Falk GW, Amoruso A (1977) Effect of ozone and nitrogen dioxide on the agglutination of rat alveolar macrophages by concanavalin A. Life Sci . 21:1637–1644. Gong H Jr, Bedi JF, Horvath SM (1988) Inhaled albuterol does not protect against ozone toxicity in nonasthmatic athletes. Arch. Environ. Health 43:46–53. Gordon T, Taylor BF, Amdur MO (1981) Ozone inhibition of tissue cholinesterase in guinea pigs. Arch. Environ. Health 36:284–288. Graham DE, Koren HS (1990) Biomarkers of inflammation in ozone-exposed humans: comparison of the nasal and bronchoalveolar lavage. Am. Rev. Respir. Dis. 142:152–156. Graham JA, Gardner DE, Blommer EJ, House DE, Menache MG, Miller FJ (1987) Influence of exposure patterns of nitrogen dioxide and modifications by ozone on susceptibility to bacterial infectious disease in mice. J. Toxicol. Environ. Health 21:113–125. Graham D, Henderson F, House D (1988) Neutrophil influx measured in nasal lavages of humans exposed to ozone. Arch. Environ. Health 43:228–233. Grievink L, Zijlstra AG, Ke X, Brunekreef B (1999) Double-blink intervention trial on modulation of ozone effects on pulmonary function by antioxidant supplements. Am. J. Epidemiol. 149: 306–314. Grose EC, Gardner DE, Miller FJ (1980a) Response of ciliated epithelium to ozone and sulfuric acid. Environ. Res. 22:377–385. Grose EC, Richards JH, Illing JW, Miller FJ, Davies DW, Graham JA, Gardner DE (1980b) Pulmonary host defense responses to inhalation of sulfuric acid and ozone. J. Toxicol. Environ. Health 10:351–362. Grose EC, Stevens MA, Hatch GE, Jaskot RH, Selgrade MJK, Stead AG, Costa DL, Graham JA (1989) The impact of a 12 month exposure to a diurnal pattern of ozone on pulmonary function, antioxidant biochemistry and immunology. In: Schneider T, Lee SD, Wolters GJR, Grant LD, editors. Atmospheric Ozone Research and its Policy Implications. Nijmegen, The Netherlands: Elsevier. Gross KB, White HJ (1987) Functional and pathologic consequences of a 52 week exposure to 0.5 ppm ozone followed by a clean air recovery period. Lung 165:283–295. Gryparis A, Forsberg B, Katsouyanni K, Analitis A, Touloumi G, Schwartz J, Samoli E, Medina S, Anderson H R, Niciu EM, Wichmann HE, Kriz B, Kosnick M, Skorkovsky J, Vonk JM, Dortbudak Z (2004) Acute effects of ozone on mortality from the “air pollution and health: a European approach” project. Am. J. Resp. Crtit. Care Med. 170:1080–1087. Guth DJ, Warren DL, Last JA (1986) Comparative sensitivity of measurements of lung damage made by bronchoalveolar lavage after short-term exposure of rats to ozone. Toxicology 40:131–143. Hackney JD, Linn WS, Mohler JG, Collier CR (1977) Adaptation to short-term respiratory effects of ozone in men exposed repeatedly. J. Appl. Physiol. 43:82–85. Hammer DI, Hasselblad V, Portnoy B, Wehrle PF (1974) Los Angeles student nurse study: daily symptom reporting and photochemical oxidants. Arch. Environ. Health 28:255–260. Hatch GE, Slade R, Stead AG, Graham JA (1986) Species comparison of acute inhalation toxicity of ozone and phosgene. J. Toxicol. Environ. Health 19:43–53. Hatch GE, Koren H, Aissa M (1989) A method for comparison of animal and human alveolar dose and toxic effect of inhaled ozone. Health Phys . 57(S1):37–40. Hayes SR, Moezzi M, Wallsten TS, Winkler RL (1987) An Analysis of Symptom and Lung Function Data from Several Human Controlled Ozone Exposure Studies (Draft Final Report) San Rafael, CA:Systems Applications, Inc.

928

OZONE

Hazucha MJ (1987) Relationship between ozone exposure and pulmonary function changes. J. Appl. Physiol. 62:1671–1680. Hazucha MJ, Bates DV, Bromberg PA (1989) Mechanism of action of ozone on the human lung. J. Appl. Physiol. 67:1535–1541. Hazucha MJ, Folinsbee LJ, Seal E Jr (1992) Effects of steady-state and variable O3 concentration profiles on pulmonary function. Am. Rev. Respir. Dis. 146:1487–1493. Hernandez-Garduno E, Perez-Neria J, Paccagnells AM, Pina-Garcia MA, Mungufa-Castro M, Catalan-Vazquez M, Rojas-Ramos M (1997) Air pollution and respiratory health in Mexico City. J. Occup. Environ. Med. 39:299–307. Higgins ITT, D’Arcy JB, Gibbons DI, Avol EL, Gross KB (1990) Effect of exposures to ambient ozone on ventilatory lung function in children. Am. Rev. Respir. Dis. 141:1136–1146. Highfill JW, Hatch GE, Slade R, Crissman KM, Norwood J, Devlin RB, Costa DL (1992) Concentration–time models for the effects of ozone on bronchoalveolar lavage fluid protein from rats and guinea pigs. Inhal. Toxicol. 4:1–16. Hoek G, Brunekreef B (1995) Effect of photochemical air pollution on acute respiratory symptoms in children. Am. J. Respir. Crit. Care Med. 151:27–32. Hoek G, Fischer P, Brunekreef B, Lebret E, Hofschreuder P, Mennen MG (1993) Acute effects of ambient ozone on pulmonary function of children in The Netherlands. Am. Rev. Respir. Dis. 147:111–117. Holguin AH, Buffler PA, Contant CF Jr, Stock TH, Kotchmar D, His BP, Jenkins DE, Gehan BM, Noel LM, Mei M (1985) The effects of ozone on asthmatics in the Houston area. Trans. Air Pollut. Control Assoc. TR-4:262–280. Holz D, J€orres A, Timm P, M€ucke M, Richter K, Koschyk S, Magnussen H (1999) Ozone-induced airway inflammatory changes differ between individuals and are reproducible. Am. J. Respir. Crit. Care Meditor. 159:776–784. Holtzman MJ, Cunningham JH, Sheller JR, Irsigler GB, Nadel JA, Boushey H (1979) Effect of ozone on bronchial reactivity in atopic and nonatopic subjects. Am. Rev. Respir. Dis. 120:1059–1067. Horstman D, Roger LJ, Kehrl H, Hazucha M (1986) Airway sensitivity of asthmatics to sulfur dioxide. Toxicol. Ind. Health 2:289–98. Horstman DH, Folinsbee LJ, Ives PJ, Abdul-Salaam S, McDonnell WF (1990) Ozone concentration and pulmonary response relationships for 6.6-hour exposures with five hours of moderate exercise to 0.08, 0.10, and 0.12 ppm. Am. Rev. Respir. Dis. 142:1158–1163. Horvath SM, Gliner JA, Folinsbee LJ (1981) Adaptation to ozone: duration of effect. Am. Rev. Respir. Dis. 123:496–499. Huang Y, Chang LY, Miller FJ, Graham JA, Ospital JJ, Crapo JD (1988) Lung injury caused by ambient levels of oxidant air pollutants: extrapolation from animal to man. Am. J. Aerosol Med. 1:180–183. Hurst DJ, Coffin DL (1971) Ozone effect on lysosomal hydrolases of alveolar macrophages in vitro. Arch. Intern. Med. 127:1059–1063. Hurst DJ, Gardner DE, Coffin DL (1970) Effect of ozone on acid hydrolases of the pulmonary alveolar macrophage. J. Reticuloendothel. Soc. 8:288–300. Hyde DM, Plopper CG, Harkema JR, St. George JA, Tyler WS, Dungworth DL (1989) Ozone-induced structural changes in monkey respiratory system. In: Schneider T, Lee SD, Wolters GJR, Grant LD, editors. Atmospheric Ozone Research and its Policy Implications. Nijmegen, The Netherlands: Elsevier. Illing JW, Miller FJ, Gardner DE (1980) Decreased resistance to infection in exercised mice exposed to NO2 and O3. J. Toxicol. Environ. Health 6:843–851. Ito K, DeLeon SF, Lippmann M (2005) Associations between ozone and daily mortality: analysis and meta-analysis. Epidemiology 16:446–457.

REFERENCES

929

Jones GL, Lane CG, O’Byrne PM (1990) Effect of thromboxane antagonists on ozone-induced airway responses in dogs. J. Appl. Physiol. 69:880–884. Jorres R, Nowak D, Magnussen H, Speckin P, Koschyk S (1996) The effect of ozone exposure on allergen responsiveness in subjects with asthma or rhinitis. Am. J. Respir. Crit. Care Med. 153:56–64. Kagawa J (1984) Exposure–effect relationship of selected pulmonary function measurements in subjects exposed to ozone. Int. Arch. Occup. Environ. Health 53:345–358. Keefe MJ, Bennett WD, DeWitt P, Seal E, Strong AA, Gerrity TR (1991) The effect of ozone exposure on the dispersion of inhaled aerosol boluses in healthy human subjects. Am. Rev. Respir. Dis. 144:23–30. Kehrl HR, Vincent LW, Kowalsky RJ, Horstman DH, O’Neil JJ, McCartney WH, Bromberg PA (1987) Ozone exposure increases respiratory epithelial permeability in humans. Am. Rev. Respir. Dis. 135:1174–1178. Kenoyer JL, Phalen RF, Davis JR (1981) Particle clearance from the respiratory tract as a test of toxicity: effect of ozone on short and long term clearance. Exp. Lung Res. 2:111–120. Kinney PL, Ware JH, Spengler JD (1988) A critical evaluation of acute ozone epidemiology results. Arch. Environ. Health 43:168–173. Kinney PL, Nilsen DM, Lippmann M, Brescia M, Gordon T, McGovern T, El-Fawal H, Devlin DB, Rom WN (1996) Biomarkers of lung inflammation in recreational joggers exposed to ozone. Am. J. Respir. Crit. Care Med. 154:1430–1435. Kleeberger SR (1995) Genetic susceptibility to ozone exposure. Toxicol. Lett. 82/83:295–300. Kleeberger SR, Zhang L-Y, Jakab GJ (1997) Differential susceptibility to oxidant exposure in inbred strains of mice: nitrogen dioxide versus ozone. Inhal. Toxicol. 9:601–621. Kleinman MT, Phalen RF, Mautz WJ, Mannix RC, McClure TR, Crocker TT (1989) Health effects of acid aerosols formed by atmospheric mixtures. Environ. Health Perspect. 79:137–145. Koenig JQ, Covert DS, Marshall SG, van Belle G, Pierson WE (1987) The effects of ozone and nitrogen dioxide on pulmonary function in healthy and in asthmatic adolescents. Am. Rev. Respir. Dis. 136:1152–1157. Koenig JQ, Covert DS, Smith MS, van Belle G, Pierson WE (1988) The pulmonary effects of ozone and nitrogen dioxide alone and combined in healthy and asthmatic subjects. Toxicol. Ind. Health 4:521–532. Koenig JQ, Covert DS, Hanley QS, VanBelle G, Pierson WE (1990) Prior exposure to ozone potentiates subsequent response to sulfur dioxide in adolescent asthmatic subjects. Am. Rev. Respir. Dis. 141:377–380. Koren HS, Devlin RB, Graham DE, Mann R, McGee MP, Horstman DE, Kozumbo WJ, Becker S, House DE, McDonnell WF, Bromberg PA (1989) Ozone-induced inflammation in the lower airways of human subjects. Am. Rev. Respir. Dis. 139:407–415. Korrick SA, Neas LM, Dockery DW, Gold DR, Allen GA, Hill LB, Kimball KO, Rosner BA, Speizer FE (1998) Effects of ozone and other pollutants on the pulmonary function of adult hikers. Environ. Health Perspect. 106 (2):93 99. Kull TJ, Sauder LR, Kerr HD, Farrell BP, Bermel MS, Smith DM (1982) Duration of pulmonary function adaptation to ozone in humans. Am. Ind. Hyg. Assoc. J. 43:832–837. Kulle TJ, Sauder LR, Hebel JR, Chatham MD (1985) Ozone response relationships in healthy nonsmokers. Am. Rev. Respir. Dis. 132:36–41. Kunzli N, Lurmann F, Segal M, Ngo L, Balmes J, Tager IB (1997) Association between lifetime ambient ozone exposure and pulmonary function in college freshmen: results of a pilot study. Environ. Res. 72:8–23. Kunzli N, McConnell R, Bates D, Bastain TM, Hricko A, Lurmann F, Avol E, Gilliland FD, Peters JM (2003) Breathless in Los Angels: the exhausting search for clean air. Am. J. Pub. Health 93 (9):1494–1499.

930

OZONE

Larsen RI, McDonnell WF, Horstman DH, Folinsbee LJ (1991) An air quality data analysis system for interrelating effects, standards, and needed source reductions: Part 11. A lognormal model relating human lung function decrease to O3 exposure. J. Air Waste Manage. Assoc. 41:455–459. Last JA (1989) Effects of inhaled acids on lung biochemistry. Environ. Health Perspect. 79:115–119. Last JA, Reiser KM, Tyler WS, Rucker RB (1984) Long-term consequences of exposure to ozone. I. Lung collagen content. Toxicol. Appl. Pharmacol. 72:111–118. Lee LY (1990) Inhibiting effect of gas phase cigarette smoke on breathing: role of hydroxyl radical. Respir. Physiol. 82:227–238. Lee LY, Dumont C, Djokie TD, Menzel TE, Nadel JA (1979) Mechanism of rapid shallow breathing after ozone exposure in conscious dogs. J. Appl. Physiol. 46:1108–1114. Lee JS, Mustafa MG, Afifi AA (1990) Effects of short-term, single and combined exposure to low-level NO2 and O3 on lung tissue enzyme activities in rats. J. Toxicol. Environ. Health 29:293–305. Leikauf GD, Driscoll KE, Wey HE (1988) Ozone-induced augmentation of eicosanoid metabolism in epithelial cells from bovine trachea. Am. Rev. Respir. Dis. 137:435–442. Levy JI, Chemerynski SM, Sarnat JA (2005) Ozone exposure and mortality: an empiric Bayes metaregression analysis. Epidemiology 16:458–468. Linder J, Herren D, Monn C, Wanner HU (1988) Die Wirkung von Ozon auf die korperliche Leistungsfahigkeit (The effect of ozone on physical activity). Schweiz. Z. Sport-med. 36: 5–10. Linn WS, Jones MP, Bachmeyer EA, Spier CE, Mazur SF, Avol EL, Hackney JD (1980) Short-term respiratory effects of polluted air: a laboratory study of volunteers in a high-oxidant community. Am. Rev. Respir. Dis. 121:243–252. Linn WS, Shamoo DA, Venet TG, Spier CE, Valencia LM, Anzar UT, Hackney JD (1983) Response to ozone in volunteers with chronic obstructive pulmonary disease. Arch. Environ. Health 38:278– 283. Linn WS, Avol EL, Shamoo DA, Peng RC, Valencia LM, Little DE, Hackney JD (1988) Repeated laboratory ozone exposures of volunteer Los Angeles residents: an apparent seasonal variation in response. Toxicol. Ind. Health 4:505–520. Lioy PJ, Dyba RV (1989)The dynamics of human exposure to tropospheric ozone. In: Schneider T, Lee SD, Wolters GJR, Grant LD, editors. Atmospheric Ozone Research and its Policy Implications. Nijmegen, The Netherlands: Elsevier. Lioy PJ, Vollmuth TA, Lippmann M (1985) Persistence of peak flow decrement in children following ozone exposures exceeding the national ambient air quality standard. J. Air Pollut. Control Assoc. 35:1068–1071. Lipfert FW, Hammerstrom T (1992) Temporal patterns in air pollution and hospital admissions. Environ. Res. 59:374–399. Lippmann M (1987) Role of science advisory groups in establishing standards for ambient air pollutants. Aerosol Sci. Tech. 6:93–114. Lippmann M (1988) Health significance of pulmonary function tests. J. Air Pollut. Control Assoc. 38:881–887. Lippmann M (1989) Effects of ozone on respiratory function and structure. Annu. Rev. Public Health 10:49–67. Lippmann M (1989b) Effective strategies for population studies of acute air pollution health effects. Environ. Health Perspect. 81:115–119. Lippmann M (1991) Morbidity associated with air pollution. In: The Handbook of Environmental Chemistry, Vol. 4: Part C: Air Pollution. Hutzinger O, editor. Heidelberg: Springer. 31–71. Lippmann M (1993) Health effects of tropospheric ozone: implications of recent research findings to ambient air quality standards. J. Exp. Anal. Environ. Epidemiol. 3:103–129.

REFERENCES

931

Lippmann M, Lioy PJ, Leikauf G, Green KB, Baxter D, Morandi M, Pasternack B, Fife D, Speizer FE (1983) Effects of ozone on the pulmonary function of children. Adv. Mod. Environ. Toxicol. 5:423–446. Lippmann M, Gearhart JM, Schlesinger RB (1987) Basis for a particle size-selective TLV for sulfuric acid aerosols. Appl. Ind. Hyg. 2:188–199. Makino K, Mizoguchi I (1975) Symptoms caused by photochemical smog. Nippon Koshu Eisei Zasshi 2:421–430. McAllen SJ, Chiu SP, Phalen RF, Rasmussen RE (1981) Effect of in vivo ozone exposure on in vitro pulmonary alveolar macrophage mobility. J. Toxicol. Environ. Health 7:373–381. McDonnell WF, Horstman DH, Hazucha MJ, Seal E, Haak ED, Salaam SA, House DE (1983) Pulmonary effects of ozone exposure during exercise: dose–response characteristics. J. Appl. Physiol. 54:1345–1352. McDonnell WF, Horstman DH, Abdul-Salaam S, House DE (1985a) Reproducibility of individual responses to ozone exposure. Am. Rev. Respir. Dis. 131:36–40. McDonnell WF III, Chapman RS, Leigh MW, Strope GL, Collier AM (1985b) Respiratory responses of vigorously exercising children to 0.12 ppm ozone exposure. Am. Rev. Respir. Dis. 132:875–879. McDonnell WF, Horstman DH, Abdul-Salaam S, Raggio LJ, Green JA (1987) The respiratory responses of subjects with allergic rhinitis to ozone exposure and their relationship to nonspecific airway reactivity. Toxicol. Ind. Health 3:507–517. McDonnell WF, Kehrl HR, Abdul-Salaam S, Ives PJ, Folinsbee LJ, Devlin RB, O’Neil JJ, Horstman DH (1991) Respiratory response of humans exposed to low levels of ozone for 6.6 hours. Arch. Environ. Health 46:145–150. McDonnell WF, Muller KE, Bromberg PA, Shy CM (1993) Predictors of individual differences in acute response to ozone exposure. Am. Rev. Respir. Dis. 147:818–825. McDonnell WF, Stewart PW, Andreoni S, Smith MV (1995) Proportion of moderately exercising individuals responding to low-level, multi-hour ozone exposure. Am. J. Respir. Crit. Care Med. 152:589–596. McDonnell WF, Stewart PW, Andreoni S, Seal E Jr, Kehrl H, Horstman DH, Folinsbee LJ, Smith MV (1997) Prediction of ozone-induced FEV1 changes: effects of concentration, duration, and ventilation. Am. J. Respir. Crit. Care Med. 156:715–722. McDonnell WF, Abbey DE, Nishino N, Lebowitz MD (1999) Long-term ambient ozone concentrations and the incidence of asthma in nonsmoking adults: the AHSMOG study. Environ. Res. 80:110–121. McConnell R, Berhane K, Gilliland F, London SJ, Islam T, Gauderman WJ, Avol E, Margolis HG, Peters JM (2002) Asthma in exercising children exposed to ozone: a cohort study. Lancet 359 (9304):386–391. McConnell R, Berhane K, Gilliland FD, Molitor J, Thomas D, Lurmann F, Avol E, Gauderman WJ, Peters JM (2003) Prospective study of air pollution and bronchitic symptoms in children with asthma. Am. J. Respir. Crit. Care Med. 168 (7):790–797. McKittrick T, Adams WC (1995) Pulmonary function response to equivalent doses of ozone consequent to intermittent and continuous exercise. Arch. Environ. Health 50:153–158. Mendelsohn R, Orcutt G (1979) An empirical analysis of air pollution dose–response curves. J. Environ. Econ. Management 6:85–106. Messineo TD, Adams WC (1990) Ozone inhalation effects in females varying widely in lung size: comparison with males. J. Appl. Physiol. 69:96–103. Miller FJ, Menzel DB, Coffin DL (1978b) Similarity between man and laboratory animals in regional pulmonary deposition of ozone. Environ. Res. 17:84–101. Miller FJ, Illing JW, Gardner DE (1978a) Effects of urban ozone levels on laboratory-induced respiratory infections. Toxicol. Lett. 2:163–169.

932

OZONE

Moffatt RK, Hyde DM, Plopper CG, Tyler WS, Putney LF (1987) Ozone-induced adaptive and reactive cellular changes in respiratory bronchioles of bonnet monkeys. Exp. Lung Res. 12:57–74. National Research Council.(1977) Toxicology. In: Ozone and Other Photochemical Oxidants. Washington, DC: National Academy of Sciences, Committee on Medical and Biologic Effects of Environmental Pollutants. Neas LM, Dockery DW, Koutrakis P, Tollerud DJ, Speizer FE (1995) The association of ambient air pollution with twice daily peak expiratory flow rate measurements in children. Am. J. Epidemiol. 141:111–122. Nightingale JA, Rogers DF, Chung KF, Barnes PJ (2000) No effect of inhaled budesonide on the response to inhaled ozone in normal subjects. Am. J. Respir. Crit. Care Meditor. 161:479–486. O’Bryne PM, Waters EH, Gold BD, Aizawa HA, Fabbri LM, Alpert SE, Nadel JA, Holtzman MJ (1984a) Neutrophil depletion inhibits airway hyperresponsiveness induced by ozone exposure. Am. Rev. Respir. Dis. 130:214–219. O’Byrne PM, Waters EH, Aizawa H, Fabbri LM, Holtzman MJ, Nadel JA (1984b) Indomethacin inhibits the airway hyperresponsiveness but not the neutrophil influx induced by ozone in dogs. Am. Rev. Respir. Dis. 130:220–224. O’Connor GT, Neas L, Vaughn B, Kattan M, Mitchell H, Mastin P, Crain EF, Evans R, Gruchalla R, Morgan W, Stout J, Lippmann M (2006) Acute respiratory health effects of air pollution on asthmatic children in US inner cities. N. Engl. J. Med. (Submitted) Osebold JW, Gerschwin LJ, Zee YC (1980) Studies on the enhancement of allergic lung sensitization by inhalation of ozone and sulfuric acid aerosol. J. Environ. Pathol. Toxicol. 3:221–234. Ostro BD, Rothschild S (1989) Air pollution and acute respiratory morbidity: an observational study of multiple pollutants. Environ. Res. 50:238–247. Ostro BD, Lipsett MJ, Mann JK, Krupnick A, Harrington W (1993) Air pollution and respiratory morbidity in Southern California. Am. J. Epidemiol. 137:691–700. Overton JH, Miller FJ (1987) Modelling ozone absorption in lower respiratory tract. Paper presented at 1987 Annual Meeting of Air Pollution Control Association, New York, June. Overton JH, Graham RC (1989) Predictions of ozone absorption in human lungs from newborn to adult. Health Phys. 57(S1):29–36. Peden DB, Setzer RW Jr, Devlin RB (1995) Ozone exposure has both a priming effect on allergeninduced responses and an intrinsic inflammatory action in the nasal airways of perennially allergic asthmatics. Am. J. Respir. Crit. Care Med. 151:1336–1345. Phalen RF, Kenoyer JL, Crocker TT, McClure TR (1980) Effects of sulfate aerosols in combination with ozone on elimination of tracer particles inhaled by rats. J. Toxicol. Environ. Health 6:797– 810. Phipps RJ, Denas SM, Siekzak MW, Wanner A (1986) Effects of 0.5 ppm ozone on glycoprotein secretion, ion and water fluxes in sheep trachea. J. Appl. Physiol. 60:918–927. Pinkerton KE, Brody AR, Miller FJ, Crapo JD (1989) Exposure to low levels of ozone results in enhanced pulmonary retention of inhaled asbestos fibers. Am. Rev. Respir. Dis. 140:1075–1081. Pope CA III, Burnett RT, Thun MJ, Calle EE, Krewski D, Ito K, Thurston GD (2002) Lung cancer, cardiopulmonary mortality, and long-term exposure to fine particulate matter. JAMA 287:1132– 1141. Raizenne ME, Burnett RT, Stern B, Franklin CA, Spenger JD (1989) Acute lung function responses to ambient acid aerosol exposures in children. Environ. Health Perspect. 79:179–185. Rao ST (1988) Prepared discussion: ozone air quality models. J. Air Pollut. Control Assoc. 38:1129– 1135. Raub JA, Miller FJ, Graham JA (1983) Effects of low level ozone exposure on pulmonary function in adult and neonatal rats. Adv. Mod. Environ. Toxicol. 5:363–367.

REFERENCES

933

Reisenauer CS, Koenig JQ, McManus MS, Smith MS, Kusic G, Pierson WE (1988) Pulmonary response to ozone exposures in healthy individuals aged 55 years or greater. J. Air Pollut. Control Assoc. 38:51–55. Reiser KM, Tyler WS, Hennessy SM, Dominguez JJ, Last JA (1987) Long-term consequences of exposure to ozone. I. Structural alterations in lung collagen of monkeys. Toxicol. Appl. Pharmacol. 89:314–322. Rigas ML, Ben-JebriaUltman JS (1997) Longitudinal distribution of ozone absorption in the lung: effects of nitrogen dioxide, sulfur dioxide, and ozone exposures. Arch. Environ. Health 52:173– 178. Rigas ML, Catlin SN, Ben-JebriaUltman JS (2000) Ozone uptake in the intact human respiratory tract: relationship between inhaled dose and actual dose. J. Appl. Physiol. 88:2015–2022. Rivas-Arancibia S, Vazquez-Sandoval R, Gonzalez-Kladiano D, Schneider-Rivas S, Lechuga-Guero A (1998) Effects of ozone exposure in rats on memory and levels of brain and pulmonary superoxide dismutase. Environ. Res. 76:33–39. Roberts AM, Schultz HD, Green JF, Armstrong DJ, Kaufman MP, Coleridge HM, Coleridge JCG (1985) Reflex tracheal contraction evoked in dogs by bronchodilator prostaglandins E2 and I2. J. Appl. Physiol. 58:1823–1831. Rombout PJA, vanBree L, Heisterkamp SH, Marra M (1989)The need for an eight hour standard. In: Schneider T, Lee SD, Wolters GJR, Grant LD, editors. Atmospheric Ozone Research and its Policy Implications.Nijegen, The Netherlands: Elsevier. Rombout PJA, Lioy PJ, Goldstein BD (1986) Rationale for an eight-hour ozone standard. J. Air Pollut. Control Assoc. 36:913–917. Romieu I, Meneses F, Ruiz S, Sienta JJ, Huerta J, White MC, Etzel RA (1996) Effects of air pollution on the respiratory health of asthmatic children living in Mexico City. Am. J. Respir. Crit. Care Med. 154:300–307. Romieu I, Sienra-Monge JJ, Ramirez-Aguilar M, Tellez-Rojo MM, Moreno-Macias H, Reyes-Ruiz NJ, del Rio-Navarro BE, Ruiz-Navarro MX, Hatch G, Slade R, Hernandez-Avila M (2002) Antioxidant supplementation and lung functions among children with asthma exposed to high levels of air pollutants. Am. J. Respir. Crit. Care Med. 166:703–709. Samet JM, Hatch GE, Horstman D, Steck-Scott S, Arab L, Bromberg PA, Levine M, McDonnell WF, Devlin RB (2001) Effect of antioxidant supplementation on ozone-induced lung injury in human subjects. Am. J. Respir. Crit. Care Med. 164:819–825. Sartor F, Demuth C, Snacken R, Walckiers D (1997) Mortality in elderly and ambient ozone concentration during the hot summer, 1994, in Belgium. Environ. Res. 72:109–117. Sartor F, Snacken R, Demuth C, Walckiers D (1995) Temperature, ambient ozone levels, and mortality during summer, 1994, in Belgium. Environ. Res. 70:105–113. Savin W, Adams W (1979) Effects of ozone inhalation on work performance and VO2 max. J. Appl. Physiol. 46:309–314. Scannell C, Chen L, Aris RM, Tager I, Christian D, Ferrando R, Welch B, Kelly T, Balmes JR (1996) Greater ozone-induced inflammatory responses in subjects with asthma. Am. J. Respir. Crit. Care Med. 154:24–29. Schelegle ES, Adams WC (1986) Reduced exercise time in competitive simulations consequent to low level ozone exposure. Med. Sci. Sports Exerc. 18:408–414. Schelegle ES, Adams WC, Siefkin AD (1987) Indomethacin pretreatment reduces ozone-induced pulmonary function decrements in human subjects. Am. Rev. Respir. Dis. 136:1350–1354. Schlesinger RB, Driscoll KE (1987) Mucociliary clearance from the lungs of rabbits following single and intermittent exposures to ozone. J. Toxicol. Environ. Health 20:125–134. Schoettlin CE, Landau E (1961) Air pollution and asthmatic attacks in the Los Angeles area. Public Health Rep .76:545–548.

934

OZONE

Schwartz J (1989) Lung function and chronic exposure to air pollution: a cross-sectional analysis of NHANES II. Environ. Res. 50:309–321. Schwartz J (1994a) Air pollution and hospital admissions for the elderly in Birmingham, Alabama. Am. J. Epidemiol. 139:589–598. Schwartz J (1994b) Air pollution and hospital admissions for the elderly in Detroit, Michigan. Am. J. Respir. Crit. Care Med. 150:648–655. Schwartz J (1994c) PM10, ozone, and hospital admissions for the elderly in Minneapolis, MN. Arch. Environ. Health 49:366–374. Schwartz J, Zeger S (1990) Passive smoking, air pollution, and acute respiratory symptoms in a diary study of student nurses. Am. Rev. Respir. Dis. 141:62–67. Schwartz J, Dockery DW, Neas LM, Wypij D, Ware JH, Spengler JD, Koutrakis P, Speizer FE, Ferris BG Jr (1994) Acute effects of summer air pollution on respiratory symptom reporting in children. Am. J. Respir. Crit. Care Med. 150:1234–1242. Seal E Jr, McDonnell WF, House DE, Salaam SA, Dewitt PJ, Butler SO, Green J, Raggio L (1993) The pulmonary response of white and black adults to six concentrations of ozone. Am. Rev. Respir. Dis. 147:804–810. Seltzer J, Bigby BG, Stulbarg M, Holtzman MJ, Nadel JA (1986) O3-induced changes in bronchial reactivity to methacholine and airway inflammation in humans. J. Appl. Physiol. 60:1 321–1326. Selwyn BJ, Stock TH, Hardy RJ, Chan FA, Jenkins DE, Kotchmar DJ, Chapman RS (1985) Health effects of ambient ozone exposure in vigorously exercising adults. In Evaluation of the Scientific Basis for Ozone/Oxidants Standards: Proceedings of an APCA International Specialty Conference (TR4) Lee SD, editor. Pittsburgh, PA: Air Pollution Control Association. pp. 281–296. Shephard RJ, Urch B, Silverman F, Corey PN (1983) Interaction of ozone and cigarette smoke exposure. Environ. Res. 31:125–137. Sherwin RP, Richters V (1985) Effect of 0.3 ppm ozone exposure on type II cells and alveolar walls of newborn mice: An image-analysis quantitation. J. Toxicol. Environ. Health 16:535–546. Sherwin RP, Richters V (1991) Centriacinar region (CAR) disease in the lungs of young adults: a preliminary report. In: Berglund RL, Lawson DR, McKee DJ, editors. Tropospheric Ozone and the Environment (TR-19). Air and Waste Management Association: Pittsburgh, PA. pp. 178–196. Sherwin RP, Richters V, Okimoto D (1983) Type 2 pneumocyte hyperplasia in the lungs of mice exposed to an ambient level (0.3 ppm) of ozone. Adv. Mod. Environ. Toxicol. 5:289–297. Solic JJ, Hazucha MJ, Bromberg PA (1982) The acute effects of 0.2 ppm ozone in patients with chronic obstructive pulmonary disease. Am. Rev. Respir. Dis. 125:664–669. Spektor DM, Lippmann M, Lioy PJ, Thurston GD, Citak K, James DJ, Bock N, Speizer FE, Hayes C (1988a) Effects of ambient ozone on respiratory function in active normal children. Am. Rev. Respir. Dis. 137:313–320. Spektor DM, Lippmann M, Thurston GD, Lioy PJ, Stecko J, O’Connor G, Garshick E, Speizer FE, Hayes C (1988b) Effects of ambient ozone on respiratory function in healthy adults exercising outdoors. Am. Rev. Respir. Dis. 138:821–828. Spektor DM, Thurston GD, Mao J, He D, Hayes C, Lippmann M (1991) Effects of single- and multiday ozone exposures on respiratory function in active normal children. Environ. Res. 55:107–122. Spengler JD, Keeler GJ, Koutrakis P, Ryan PB, Raizenne M, Franklin CA (1989) Exposures to acidic aerosols. Environ. Health Perspect. 79:43–51. Stacy RW, Seal E Jr, House DE, Green J, Roger LJ, Raggio L (1983) A survey of effects of gaseous and aerosol pollutants on pulmonary function of normal males. Arch. Environ. Health 38:104–115. Stern BR, Raizenne ME, Burnett RT, Jones L, Kearney J, Franklin CA (1994) Air pollution and childhood respiratory health: exposure to sulfate and ozone in 10 Canadian rural communities. Environ. Res. 66:125–142.

REFERENCES

935

Stieb DM, Burnett RT, Beveridge RC, Brook JR (1996) Association between ozone and asthma emergency department visits in Saint John, New Brunswick, Canada. Environ. Health Perspect. 104:1354–1360. Tepper JS, Weiss B, Wood RW (1985) Alterations in behavior produced by inhaled ozone or ammonia. Fundam. Appl. Toxicol. 5:1110–1118. Tepper JS, Costa DL, Lehmann JR, Weber MF, Hatch GE (1989) Unattenuated structural and biochemical alterations in the rat lung during functional adaptation to ozone. Am. Rev. Respir. Dis. 140:493–501. Tepper JS, Wiester MJ, Weber MF, Fitzgerald S, Costa DL (1991) Chronic exposure to a stimulated urban profile of ozone alters ventilatory responses to carbon dioxide challenge in rats. Fundam. Appl. Toxicol. 17:52–60. (1997) Testimony submitted to U.S. Senate Committee on Environment and Public Works, Subcommittee on Clean Air, Wetlands, Private Property, and Nuclear Safety. February 1977. Thurston GD, Ito K, Kinney PL, Lippmann M (1992) A multi-year study of air pollution and respiratory hospital admissions in three New York State metropolitan areas: results for 1988 and 1989 summers. J. Expo. Anal. Environ. Epidemiol. 2:429–450. Thurston GD, Ito K, Hayes CG, Bates DV, Lippmann M (1994) Respiratory hospital admissions and summertime haze air pollution in Toronto, Ontario: consideration of the role of acid aerosols Environ. Res. 65:271–290. Thurston GD, and Ito K (2001) Epidemiologic studies of acute ozone exposures and mortality. J. Expos. Anal. Environ. Epidemiol. 11:286–294. Torres A, Utell MJ, Morrow PD, Voter KZ, Whitin JC, Cox C, Looney RJ, Speers DM, Tsai Y, Frampton MW (1997) Airway inflammation in smokers and nonsmokers with varying responsiveness to ozone. Am. J. Respir. Crit. Care Med. 156:728–736. Touloumi G, Katsouyanni K, Zmirou D, Schwartz J, Spix C, Ponce de Leon A, Tobias A, Quennel P, Rabczenko D, Bacharova L, Bisanti L, Vonk JM, Ponka A (1997) Short-term effects of ambient oxidant exposure on mortality: a combined analysis within the APHEA project. Am. J. Epidemiol. 146:177–185. Tyler WS, Tyler NK, Last JA, Barstow TJ, Magliano DJ, Hinds DM (1987) Effects of ozone on lung and somatic growth. Pair fed rats after ozone exposure and recovery periods. Toxicology 46:1–20. Tyler WS, Tyler NK, Last JA, Gillespie MJ, Barstow TJ (1988) Comparison of daily and seasonal exposures of young monkeys to ozone. Toxicology 50:131–144. U.S. DOL(1989) Air contaminants: Final Rule. 29CFR, Part 1910. Federal Register 54(12):2332– 2983. U.S. EPA(1986) Air Quality Criteria for Ozone and Other Photochemical Oxidants, Vol. II. EPA/600/ 8-84/020F, ECAO. NTIS, Springfield, VA-PB87-142949. U.S. EPA(1992) Summary of Selected New Information on Effects of Ozone on Health and Vegetation: Supplement to 1986 Air Quality Criteria for Ozone and Other Photochemical Oxidants. EPA/600/8-88/105F, ECAO. NTIS, Springfield, VA-PB92-235670. U.S. EPA (1996a) Air Quality Criteria for Ozone and Other Photochemical Oxidants. United States Environmental Protection Agency, Research Triangle Park, NC. EPA Publication 600/P-93/004F. U.S. EPA (1996b) Review of National Ambient Air Quality Standards for Ozone: Assessment of Scientific and Technical Information. OAQPS Staff Paper. EPA-452/A-96-007. U.S. EPA-OAQPS, Research Triangle Park, NC 27711. U.S. EPA (2006) Air Quality Criteria for Ozone and Related Photochemical Oxidants. EPA/600/R-05/ 004a,b,c. Utell MJ, Morrow PE, Speers DM, Darling J, Hyde RW (1983) Airway responses to sulfate and sulfuric acid aerosols in asthmatics. Am. Rev. Respir. Dis. 128:444–450.

936

OZONE

Vagaggini B, Taccola M, Cianchetti S, Carnevali S, Bartoli ML, Bacci E, Dente FL, DiFranco A, Giannini D, Paggiaro PL (2002) Ozone exposure increases eosinophilic airway response induced by previous allergen challenge. Am. J. Respir. Crit. Care Med. 166:1073–1077. Valentine R (1985) An in vitro system for exposure of lung cells to gases: effects of ozone on rat macrophages. J. Toxicol. Environ. Health 16:115–126. Van Bree L, Rombout PJA, Rietjens IMCM, Dormans JAMA, Marra M (1989) Pathobiochemical effects in rat lung related to episodic ozone exposure. In: Schneider T, Lee SD, Wolters GJR, Grant LD, editors. Atmospheric Ozone Research and Its Policy Implications. Nijmegen, The Netherlands: Elsevier. Verhoeff AP, Hoek G, Schwartz J, vanWijnen JH (1996) Air pollution and daily mortality in Amsterdam. Epidemiology 7:225–230. Wayne WS, Wehrle PF, Carroll RE (1967) Oxidant air pollution and athletic performance. JAMA 199:901–904. Weinmann GG, Bowes SM, Gerbase MW, Kimball AW, Frank R (1995) Response to acute ozone exposure in healthy men. Am. J. Respir. Crit. Care Med. 151:33–40. Weisel CP, Cody RP, Lioy PJ (1995) Relationship between summertime ambient ozone levels and emergency department visits for asthma in central New Jersey. Environ. Health Perspect. 103:97– 102. White MC, Etzel RA, Wilcox WD, Lloyd C (1994) Exacerbations of childhood asthma and ozone pollution in Atlanta. Environ. Res. 65:56–68. Whittemore A, Korn E (1980) Asthma and air pollution in the Los Angeles area. Am. J. Public Health 70:687–696. WHO-EURO (2003) Health Aspects of Air Pollution with Particulate Matter, Ozone, and Nitrogen Dioxide. Report of a WHO Working Group. Bonn, Germany. WHO Regional Office for Europe, DK 2100, Copenhagen, Denmark. WHO-EURO (2006) Air Quality Guidelines: Global update 2005. Particulate Matter, Ozone, Nitrogen Dioxide and Sulfur Dioxide. WHO, Copenhagen, Denmark. Witz G, Amoruso MA, Goldstein BD (1983) Effect of ozone on alveolar macrophage function. Membrane dynamic properties. Adv. Mod. Environ. Toxicol. 5:263–272. Yang W, Jennison BL, Omaye ST (1997) Air pollution and asthma emergency room visits in Reno, Nevada. Inhal. Toxicol. 9:15–29. Yang Q, Chen Y, Shi Y, Burnett RT, McGrail KM, Krewski D (2003) Association between ozone and respiratory admissions among children and the elderly in Vancouver, Canada. Inhal. Toxicol. 15:1297–1308. Ying RL, Gross KB, Terzo TS, Eschenbacher WL (1990) Indomethacin does not inhibit the ozoneinduced increase in bronchial responsiveness in human subjects. Am. Rev. Respir. Dis. 142:817– 821.

24 PESTICIDES Philip J. Landrigan and Luz Claudio

Synthetic pesticides are a diverse group of chemical compounds, most of them derived from petroleum. Pesticides are used to control insects, unwanted plants, fungi, rodents, and other pests (Hayes and Laws, 1991). Approximately 900 pesticide active ingredients including insecticides, herbicides, rodenticides, and fungicides are currently registered for use (California Department of Pesticide Regulation, 2005). These compounds are mixed with each other and are also blended with “inert” ingredients to produce more than 20,000 commercial pesticide formulations. The United States Environmental Protection Agency (EPA) estimates that in 2001, the most recent year for which data are available, the United States spent $11 billion for over 1.2 billion pounds of pesticide active ingredients (U.S. EPA Office of Prevention, Pesticides and Toxic Substances, 2001). There are also naturally occurring pesticides produced by plants and other organisms; these compounds are discussed in Chapter 20. Pesticides are used in an extraordinarily wide range of settings. In the homes, they control mice, termites, and other rodents. In gardens and lawns as well as along highways and under power-line right-of-ways, pesticides control the growth of unwanted plants. By controlling agricultural pests, pesticides have contributed to dramatic increases in crop yields and in the quantity and variety of the diet (National Research Council, 1993). The agricultural sector is the primary consumer of pesticides, accounting for 76% of use by volume. Industrial, commercial, and governmental users (13%), and home and garden users (11%) account for the remainder (U.S. EPA Office of Prevention, Pesticides and Toxic Substances, 2001). Herbicides for weed control account for the largest volume of agricultural pesticide use (59%) and are applied primarily on corn and soybeans. Insecticides are the next major category of agricultural pesticides (21% of volume) and are used primarily on corn, cotton, and soybeans (Schierow, 1996). Because pesticides create risks as well as benefits, their use poses a perennial problem for public policy and regulation (National Research Council, 1993). Pesticides are specifically designed to be toxic to certain species, and this toxicity is the basis of their utility (Hayes and

Environmental Toxicants, Third Edition Edited by Morton Lippmann Copyright Ó 2009 John Wiley & Sons, Inc.

937

938

PESTICIDES

Laws, 1991). However, many pesticides are also toxic to species beyond those targeted and have, consequently, caused severe damage to ecosystems. Humans are among the species at risk of such unintended effects, and pesticides have been shown to cause a wide range of adverse effects on human health including acute and chronic injury to the nervous system, lung damage, injury to the reproductive organs, dysfunction of the immune and endocrine systems, birth defects, and cancer. Children are especially susceptible to the effects of pesticides on health (National Research Council, 1993). Recognition of the toxicity of pesticides has stimulated enactment of a vast body of protective laws and regulations in nations around the world. It has also fostered the development of a continuing series of newer and less toxic pesticides, and these actions have helped to control the toxic hazards. The National Academy of Sciences estimates that since 1954, a 90% reduction in the toxicity of pesticides applied to food crops has occurred in the United States (National Research Council, 1996).

24.1 EVOLVING PATTERNS OF PESTICIDE USE The era of modern pesticides began in the nineteenth century, when sulfur compounds were developed as fungicides. In the late nineteenth century, arsenical compounds were introduced to control the insects that attack fruit and vegetable crops, for example, lead arsenate was used widely on apples and grapes. All these substances were acutely toxic (Schuman and Simpson, 1997). In the 1940s, the chlorinated hydrocarbon pesticides, most notably DDT (dichlorodiphenyltrichloroethane), were introduced. For a time, DDT and similar chemicals were used extensively in agriculture and for the control of malaria and other insect-borne diseases. Because these pesticides had little or no immediate toxicity, they were widely hailed and initially believed to be safe (Wargo, 1996). The publication of Rachel Carson’s Silent Spring (1962) brought to the attention of the American public the potential of the chlorinated hydrocarbon pesticides for long-term accumulation and toxicity in the food chain. Carson documented that DDT had caused widespread reproductive failure and near extinction of bald eagles and ospreys, two species that had accumulated large quantities of DDT because of their high position in the food chain. In 1972, the newly created EPA banned DDT in the United States. Today, the principal classes of insecticides used in most industrialized countries are organophosphates, carbamates, and pyrethroids. Unlike the chlorinated hydrocarbons, these compounds are short-lived in the environment and do not bioaccumulate. The organophosphates and carbamates are, however, neurotoxicants, and the human nervous system can be affected by some of these compounds, causing serious acute and chronic toxicity (Blondell, 1997). Other commonly used pesticides in current use are known or suspected to be carcinogens, reproductive toxicants, or toxicants to the endocrine system (Costa, 1997; Zahm et al., 1997). Insecticide use has declined in recent years, reflecting in part the adoption of programs such as integrated pest management (IPM)(Benbrook, 1996). These programs emphasize the use of nonchemical means of pest control to replace and complement pesticide use. Elements of IPM in the home include the cleanup of food residues, the sealing of foundation cracks, and good maintenance.Forexample,theproject“GrowingUpHealthyinEastHarlem”,developedbythe Mount Sinai Children’s Environmental Health and Disease Prevention Research Center in partnership with two neighborhood health centers, illustrated the successful implementation of

EXPORT OF HAZARDOUS PESTICIDES

939

IPM with the reduction of organophosphate use in an urban setting (Brenner et al., 2003). In agriculture, integrated pest management may involve crop rotation and the use of resistant plant strains (Baker et al., 2002). Fungicide use has remained steady and herbicide use has increased substantially, whereas the use of insecticides has begun to decline.

24.2 EXPORT OF HAZARDOUS PESTICIDES The export of highly toxic pesticides manufactured but banned in industrially developed nations to developing countries for use remains a major issue in pesticide regulation. Between, 1997 and 2000, the United States exported nearly 65 million pounds of either forbidden or severely restricted pesticides; 57% of which were shipped to developing nations, while the remaining 43% went to ports in the Netherlands and Belgium (Smith, 2001), presumably for transshipment. While application of highly toxic pesticides continues in many nations and especially in less developed countries, the past three decades has witnessed a concerted effort by the international community to monitor, restrict, and eliminate the use of hazardous compounds. The United Nations Environment Programme has provided an international framework for the management of hazardous chemicals and pesticides through the adoption of the Rotterdam Convention on the Prior Informed Consent Procedure for Certain Hazardous Chemicals and Pesticides in International Trade, and also the Stockholm Convention on Persistent Organic Pollutants (POPs). Adopted in 1998, The Rotterdam Convention covers pesticides and industrial chemicals that have been banned or severely restricted for health or environmental reasons and established the “Prior Informed Consent” (PIC) procedure, which mandates that export of a chemical covered by the Convention can only take place with the prior informed consent of the importing party (United Nations Environment Programme, 1998). The Stockholm Convention, in effect since May 2004, seeks the elimination or restriction of production and use of all intentionally produced POPs and targets 12 persistent organic pollutants including nine pesticides aldrin, chlordane, DDT, dieldrin, endrin, heptachlor, hexachlorobenzene, mirex, and toxaphene (United Nations Environment Programme, 2004). Vigorous debate continues on whether DDT should continue to be used, despite the Stockholm Convention, for malaria control in developing countries (Wikipedia DDT, 2006). The main argument in favor of continued use of DDT is its low economic cost. The main arguments against continuing use are(1) the effects of DDT on the environment; and (2) increasing resistance of mosquitoes to the compound. A further argument against continuing us e of DDT for malaria control is that it may result in contamination of food crops intended for export to developed nations, a situation that may cause these products to be rejected by their intended purchasers (Benbrook, 2002).

24.3 EXPOSURE TO PESTICIDES Pesticide exposure may be percutaneous, by inhalation, or by ingestion. Exposure may occur via the diet, in the workplace, in the yard or home and in the community. In assessing exposure, it is important to understand that persons may simultaneously be exposed to multiple pesticides through several routes and that the effects of these multiple exposures may be additive or even synergistic (National Research Council, 1993) (Table 24.1).

940

PESTICIDES

TABLE 24.1 Pesticide Residues on Fruits and Vegetables Heavily Consumed by Young Children Supermarket Warehouse Data 1990–1992

Food Apples Bananas Broccoli Cantaloupes Carrots Cauliflower Celery Cherries Grapes Green beans Leaf lettuce Oranges Peas Peaches Pears Potatoes Spinach Strawberries Tomatoes Total

Number of Samples

Detected Number With One or More Pesticides

Percent With One or More Pesticides Detected

Number of Different Pesticides Detected

542 368 63 225 252 65 114 90 313 249 201 237 191 246 328 258 163 168 395 4,468

425 134 16 78 125 26 85 72 192 95 136 190 87 194 240 120 88 138 203 2,644

78% 36% 25% 35% 50% 40% 75% 80% 61% 38% 68% 80% 46% 79% 73% 47% 54% 82% 51% 59%

25 9 9 19 12 13 13 13 22 20 22 25 19 20 11 17 19 17 22 81

Source: Environmental Working Group (Wiles and Campbell, 1993). Compiled from U.S. EPA. Office of Planning, Policy, and Evaluation, Pesticide Food Residue Database, Anticipated Pesticide Residues in Food.

24.3.1

Occupational Exposure

Occupational exposure to pesticides occurs among manufacturers and formulators; during transport and storage; among mixers, loaders, and applicators working in fields, greenhouses, parks, and residential buildings; among vector control and structural applicators; and among farm workers entering fields or greenhouse workers handling foliage previously sprayed by pesticides (Blondell, 1997; McConnell, 1994). Crop duster aviation mechanics have also been reported to be at high risk for pesticide poisoning. Other groups occasionally exposed include emergency crews or sewer workers involved in cleanup. In developed countries, a very large exposed group consists of building maintenance workers who apply insecticides in public and private housing, schools, hospitals, and commercial structures. 24.3.2

Environmental Exposure

Environmental exposure to pesticides can occur through consumption of contaminated water, ingestion of pesticide residues in food, inhalation of airborne spray drift, exposure to pesticides applied in the home, school or community, or from exposure to improperly disposed hazardous waste. Heaviest use of pesticides in the home has been found to occur in inner-city neighborhoodsforthecontrolofroachesinapartments.InNewYorkState,heaviestuseofpesticidesinall counties statewide occurred in Manhattan and Brooklyn (Landrigan et al., 1999). Seasonal contamination of drinking water by herbicides is reported each spring in the American Midwest, a pattern that coincides with annual application of these compounds,

EXPOSURE TO PESTICIDES

941

atrazine in particular, prior to spring planting. Although nonoccupational exposure is usually at a low level, numerous episodes of acute illness have resulted from environmental exposure to pesticides (National Research Council, 1993). 24.3.3

Pediatric Exposures

Children are a group at particular risk of exposure to pesticides (National Research Council, 1993). A major route of children’s exposure is through their diet. They may also be exposed to pesticides applied in homes or schools, on lawns and in gardens. Children employed in agriculture or living in migrant farm worker camps are particularly at high risk of exposure to pesticides and of suffering from acute pesticide intoxication (McConnell, 1994). Children’s tissues and organs are rapidly developing, and at various stages in early development these growth processes create windows of great vulnerability to pesticides and other environmental toxicants. An analysis undertaken by the National Academy of Sciences (1993) has established that the unique vulnerability of infants and children to pesticides and other environmental toxicants is based on the following four factors: 24.3.3.1 Children have Greater Exposures to Environmental Toxicants Than Adults Pound for pound of body weight, children drink more water, eat more food, and breathe more air than adults. For example, children in the first 6 months of life consume seven times as much water per unit body weight as does the average American adult. Consequently, children are more heavily exposed to toxicants in air, food, and water than adults. Two behavioral characteristics of infants and children further magnify their exposures: their normal hand-to-mouth activity, and their play close to the ground. And lastly, dermal absorption by young children to certain environmental toxicants such as pesticides is higher than in adults, because of their greater surface area relative to body weight and greater skin permeability. 24.3.3.2 Children’s Metabolic Pathways, Especially in the First Months After Birth, are Immature Compared to Those of Adults In some instances, children are actually better able than adults to cope with environmental toxicants, because they are unable to metabolize toxicants to their active form (Kimmel, 1992). More commonly, however, children are less able to detoxify chemicals such as organophosphate pesticides, and thus are more vulnerable to them (Mortensen et al., 1996; Peto et al., 1991; Bearer, 1995). 24.3.3.3 Infants and Children are Growing and Developing, and Their Delicate Developmental Processes are Easily Disrupted Many organ systems in infants and children, the nervous system in particular, undergo extensive growth and development throughout the prenatal period and the first months and years of extrauterine life. If cells in an infant’s brain are injured by neurotoxic pesticides; or if reproductive development is diverted by endocrine-disrupting pesticides, the resulting dysfunction can be permanent and irreversible (National Research Council, 1993). 24.3.3.4 Because Children have More Future Years of Life Than Most Adults, They have More Time to Develop Chronic Disease That may be Initiated by Early Exposures Exposures sustained early in life, including prenatal exposures, appear more likely to lead to disease than similar exposures encountered later. Also deficits sustained early on may be persistent throughout their lives (National Research Council,

942

PESTICIDES

1993). Early exposures to pesticides have been linked in toxicological studies to the subsequent development of Parkinsonian symptoms (Cory-Slechta et al., 2005). Surveys of foods commonly consumed by children have shown that a high proportion of them contain pesticide residues, and that these foods also frequently contain residues of multiple pesticides. A 2005 study designed to measure dietary organ phosphorus pesticide exposure in children found that the most likely route of exposure to these pesticides was through the diet (Wiles and Campbell, 1993). These observations on children’s exposures to pesticides in food were complemented by a 1995 study that found 16 different pesticides present in some of the baby foods most commonly sold in the U. S. (Wiles and Davies, 1995). These pesticide residues included eight that have been shown to be toxic to the nervous system, five that affect the endocrine system, and eight that are potential carcinogens. Consumption of a diet rich in organically grown fruits and vegetables has been shown to be highly effective in reducing dietary exposure to pesticides. Children who consumed a largely organic diet had dramatically lower levels in urinary pesticide residue levels as compared to classmates who consumed a conventional American diet (Lu et al., 2006). In the past decade, since passage of the Food Quality Protection Act in 1996 (see below), levels of pesticide residues in domestically grown foods in the United States have steadily declined. However, residue levels in imported foods have increased during the same period, with the result that the aggregate U.S. exposure to pesticides in foodstuffs has remained relatively constant. Imported winter fruits, such as table grapes, are the most heavily contaminated imported foodstuffs (Groth et al., 2000).

24.4 EPIDEMIOLOGY OF ACUTE PESTICIDE POISONING Data on pesticide poisonings are sparse, and there is serious underreporting of even acute, life-threatening episodes (Blondell, 1997). The best information on occupational pesticide poisoning in the United States comes from California, where physicians are required by law to report all incidents of pesticide intoxication. In the mid-1990s, the average annual number of occupational pesticide poisoning cases reported in California was approximately 1,500, of which 54% occurred in agriculture (Blondell, 1997). Organophosphates were the class of compounds most frequently involved. By extrapolating California data to the nation, it has been estimated that there are between 10,000 and 20,000 cases of physician-diagnosed pesticide poisoning in the United States per year (U.S. EPA, 1992). Data on nonoccupational pesticide poisonings in the United States are collected by the Consumer Product Safety Commission (CPSC) based on a statistical sample of emergency rooms in 6,000 selected hospitals (Blondell, 1997). In 1990–92, there were an estimated 20,000 emergency room visits resulting from pesticide exposure. Incidence was disproportionately high in children, who accounted for 61% of cases. 24.4.1

Pesticide Epidemiology in the Third World

Pesticide use in developing nations, although rapidly increasing, accounts for only 25% of the 3 million tons of pesticides produced worldwide each year (21). Nonetheless, 90% of the estimated 3 million yearly poisonings worldwide, and more than 99% of the 220,000 deaths, occur in developing countries. Examples of major outbreaks of pesticide-associated disease include an estimated 60,000 illnesses and 2000 to 2500 deaths from exposure to methyl

TOXICITY OF PESTICIDES

943

isocyanate used as an intermediary in insecticide manufacture in an accidental release in Bhopal, India, and an epidemic of acute parathion poisoning in Jamaica (with 17 deaths) caused by consumption of contaminated, imported wheat flour (Melius, 1998; Diggory et al., 1977). In surveys of agricultural workers in several Asian countries, and of cotton workers in Mexico, 3% to 7% and 13% to 15%, respectively, reported poisoning in the previous year. In addition, pesticide poisoning may be more likely to be underdiagnosed in these nations than in the developed countries (Jeyaratnam, 1990).

24.5 TOXICITY OF PESTICIDES Because the chemistry of pesticides is highly diverse, they are capable of causing a wide range of adverse health effects. Depending on the pesticide, or combination of pesticides, to which an individual or a population is exposed, these effects can involve virtually every organ system in the body. Pesticides can produce acutely toxic effects, delayed effects, and chronic effects. Also, some pesticides are developmental toxicants and others are carcinogens and reproductive toxicants. This review summarizes data on the toxic effects of the major classes of agents. 24.5.1

Insecticides

24.5.1.1 Cholinesterase Inhibitors the carbamates.

This class includes both the organophosphates and

24.5.1.2 Acute Clinical Effects of Organophosphates and Carbamates The toxicity of these two classes of insecticides is similar, and both inhibit neuronal acetylcholinesterase (Costa, 1997). Acute poisonings by organophosphates and carbamates account for the majority of systemic pesticide poisoning cases seen each year in the United States (Blondell, 1997). Inhibition of acetylcholinesterase results in an increase in acetylcholine, with resultant overstimulation of the postsynaptic receptors in the cholinergic nervous system. These effects can be differentiated toxicologically into overstimulation of (1) the central nervous system; (2) the nicotinic receptors (skeletal muscle and autonomic ganglia); and (3) the muscarinic receptors (secretory glands and postganglionic fibers in the parasympathetic nervous system). The nicotinic effects generally appear later, and only in the more severe cases. Headache, anxiety, and sleep disturbance, often accompanied by salivation and anorexia, are common early symptoms of mild overexposure to cholinesterase inhibitors (McConnell, 1994). Chest tightness is a symptom of moderately severe poisoning after dermal exposure, but it can be an early symptom if there is significant exposure via the respiratory tract. Seizures and impaired consciousness occur in the most severe cases. Death can follow respiratory arrest. Other reported acute effects include myopathy (including myocardopathy), hypothermia, liver dysfunction, brady- or tachyarrhythmias, leukocytosis, and acute psychosis. 24.5.1.3 Delayed or Chronic Effects of Organophosphates Chronic low-level exposure to organophosphates may result in weakness and malaise, often accompanied by headache and light-headedness, but without other specific symptoms (McConnell, 1994). Also,

944

PESTICIDES

several additional complications may occur in patients poisoned with organophosphates as follows: 1. Psychosis: There have been case reports of acute and persistent psychotic illness associated with poisoning by organophosphates. 2. Intermediate Syndrome: A recently reported (and rare) so-called intermediate syndrome, characterized by the paralysis of the proximal musculature and muscles of respiration and of cranial motor nerves, occurs one to four days after acute poisoning. Although the mechanism of this intermediate syndrome is not known, it is not thought to result from inhibition of acetylcholinesterase. 3. Organophosphate-induced delayed polyneuropathy (OPIDP): This distal dyingback axonopathy is characterized clinically by cramping muscle pain in the legs, often followed by paresthesia and motor weakness. The onset occurs 10 days to 3 weeks after severe poisoning (Lotti, 1992). There may be marked foot drop and weakness of the distal upper extremities. The pathophysiology of the disease probably requires the inhibition and subsequent aging of a poorly characterized intraneuronal esterase known as neuropathy target esterase (NTE), which is distinct from acetyl cholinesterase (Costa, 1997). 4. Developmental Neurotoxicity: Certain organophosphates, of which chlorpyrifos (CP) has been the most extensively studied, have the ability to cause developmental neurotoxicity when exposures occur during windows of vulnerability during prenatal or early postnatal life, when the nervous system is actively developing and differentiating. Early postnatal exposure of rat pups to low doses of CP has been shown to produce reduction in the number of brain cells, learning deficits, and behavioral difficulties. These functional deficits have been shown to persist into adult life (Slotkin and Seidler, 2007). The developmental toxicity of the organophosphate pesticides appears to be quite disjunct from their systemic toxicity (Slotkin et al., 2006). Parathion, for example, has great systemic toxicity, but relatively little developmental toxicity. Systemic toxicity results principally from cholinesterase inhibition. By contrast, a key pathophysiological mechanism underlying the developmental effects of CP appears to be binding of CP to the nicotinic cholinergic receptor during early brain development (Slikker et al., 2005). Acting through this mechanism, CP has been shown to disrupt the basic cellular machinery that controls the patterns of neural cell maturation and the formation of synapses. These effects do not depend on the inhibition of cholinesterase. These mechanisms of developmental toxicity are likely to be shared by other organophosphates, but the potential developmental toxicity of most of these compounds has not been evaluated in detail (Groth et al., 2000). Recent prospective epidemiological studies of birth cohorts have found human correlates to the developmental neurotoxicity of OPs observed in rodents. Human infants exposed in utero to CP have been found to have smaller head circumference at birth than unexposed babies; this effect is most pronounced in infants born to mothers with low expression levels of the OP-metabolizing enzyme paraoxonase, apparent evidence for a novel gene–environment interaction (Berkowitz et al., 2004). Followup studies of children with biochemically documented exposures to CP in utero have found evidence for slowed reflexes at birth (Young et al., 2005), developmental delays (Engel et al., 2007; Fenster et al., 2007) and increased prevalence of attention deficit/hyperactivity disorder (ADHD) (Rauh et al., 2006).

TOXICITY OF PESTICIDES

945

24.5.1.4 Pyreththrum and Pyrethroids In the mid-1800s, pyrethrum extracts of chrysanthemum flowers (containing pyrethrin and other active ingredients) were found to be effective insecticides (Hayes and Laws, 1991). They are relatively less toxic than other commonly used insecticides. Pyrethrum may cause asthma or allergic rhinitis and contact dermatitis is frequent. Although pyrethroid insecticides are closely related chemically to the naturally occurring pyrethrin, those in use today are all synthetic. They are more stable in the natural environment than the natural compounds and are used in agriculture and in household pest control. Ingestion of large doses of pyrethroids causes salivation, nausea, vomiting, diarrhea, irritability, tremor, incoordination, seizures, and death. Toxicity is thought to be mediated through delay in the closing of sodium channels after discharge of an action potential, which results in repetitive neuronal discharge (Narahashi, 1985). 24.5.1.5 Organochlorine Insecticides This class of insecticides includes DDT, lindane, and the cyclodienes: aldrin, dieldrin, endrin, and heptachlor. Most uses of these chemically diverse insecticides have been banned or restricted in the developed world because of their environmental persistence and bioaccumulation, two properties that have led to great damage to wildlife (Carson, 1962). Fat levels of DDT in human surveys have decreased markedly since DDTwas banned in the United States in 1973. In many nations of the developing world, concentrations of DDT in human fat and in milk from humans and cows continue to be high. 24.5.1.6 Acute Effects Of Organochlorines Dermal absorption of the cyclodienes, chlordecone, and lindane is high. Most occupational poisonings within this class result from the acute toxicity of chlordecone, endrin, aldrin, and dieldrin (Blondell, 1997). Most other acute poisonings result from the ingestion of these insecticides. Acute toxicity reflects poorly understood neuronal hyperactivity in the central nervous system. Sudden seizures (especially from aldrin, dieldrin, endrin, lindane, and toxaphene) may occur up to 48 h after exposure and may be relatively intractable. Headache, nausea, dizziness, incoordination, confusion, tremor, and paresthesia are common. Tremor is characteristic of poisoning with DDT and chlordecone. Abnormal liver enzyme levels and renal tubular abnormalities may be seen. 24.5.1.7 Chronic Effects of Organochlorines Epidemic occupational chlordecone (Kepone) poisoning in a manufacturing facility in Virginia was characterized by anxiety and tremor, opsoclonus, personality change, oligospermia, pleuritic and joint pains, weight loss, and liver disease. The effects were chronic. In addition, two workers’ wives were poisoned by contact with contaminated work clothes, and a portion of the Chesapeake Bay was polluted by discharge from the plant (Cannon et al., 1978). Almost all organochlorine insecticides have been found to be carcinogenic in at least one species of rodent. Idiosyncratic cases of aplastic anemia have been reported anecdotally in association with exposure to organochlorines, especially to chlordane and lindane (McConnell, 1994). 24.5.2

Herbicides

Herbicides are the most important class of pesticides in terms of U.S. market share. Their use in agriculture and elsewhere is increasing steadily. This diverse class includes atrazine; 2,4-D; 2,4,5-T; glyphosate; and paraquat. Most of these compounds, paraquat excepted, have

946

PESTICIDES

low acute toxicity. Some herbicides, such as alachlor and atrazine, are important groundwater contaminants, and are animal carcinogens. 24.5.2.1 Chlorphenoxy Herbicides Chlorphenoxy herbicides have been used widely in the United States and elsewhere against broad-leaved plants. During the Vietnam War, 2,4-D and 2,4,5-T were applied together, in a 50:50 mixture, by American forces for deforestation and to destroy food crops, in a formulation known as Agent Orange. 2,4,5-T that is no longer marketed in most industrialized countries may become contaminated, during its manufacture, with dioxins, of which the most toxic and intensely studied is the tetrachlorinated 2,3,7,8-tetrachloro-dibenzodioxin (TCDD) isomer. An increased incidence of cancer has been observed in workers chronically and heavily exposed to TCDD. The largest published cohort study of heavily exposed herbicide production workers demonstrated an association between exposure, non-Hodgkin’s lymphoma, and soft tissue sarcoma (Fingerhut et al., 1991). To assess the hazards of exposure to dioxin in Agent Orange among Vietnam War Veterans, the United States National Academy of Sciences convened an expert panel in 1991. After evaluating the available epidemiological and toxicological data, this committee concluded that there was strong evidence for a positive association between dioxin-contaminated herbicide exposure in Vietnam and soft-tissue sarcoma, nonHodgkin’s lymphoma and Hodgkin’s disease. The Committee also concluded that there was strong evidence for a link between Agent Orange and chloracne, as well as for a link with porphyria cutanea tarda. In addition, the Committee found weaker evidence of association between exposure to dioxin-contaminated herbicides and cancer of the lungs, larynx and trachea, as well as prostate cancer and multiple myeloma (National Academy of Sciences, 2006). Further discussion of the toxicity and epidemiology of dioxins and related compounds is presented in Chapter 20. 24.5.2.2 Bipyridils (Diquat and Paraquat) High acute toxicity, lack of an effective antidote, and ready availability (because of low cost and herbicidal efficacy) have contributed to the notoriety of paraquat. 24.5.2.3 Acute Effects of Bipyridils Painfulburnsandbleedingofthegastrointestinaltract are common following acute exposure to paraquat. Approximately 20% of ingested paraquat is absorbed systemically, where it may cause acute hepatic necrosis and renal disease. The most distinctive aspect of paraquat poisoning is delayed pulmonary toxicity. Paraquat is concentrated from the systemic circulation into the lungs, where pulmonary edema may develop two to four days after ingestion. The mortality rate of pulmonary toxicity induced by paraquat is greater than 50% in most case series. The lethal dose in 50% of rabbits dosed dermally is only 4.5 mg/kg/day. Occupational poisoning resulting from dermal absorption is more likely to occur in developing countries, where applicators may carry and mix concentrated paraquat in leaky backpack sprayers. Splashes of paraquat may cause corneal opacification; nosebleeds, rashes, burns, and loss of fingernails are common local effects (McConnell, 1994). 24.5.2.4 Chronic Effects of Bipyridils survivors of acute paraquat poisoning.

Pulmonary fibrosis has been reported among

TOXICITY OF PESTICIDES

947

Paraquat is structurally similar to 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP), an illicit drug produced as a heroin substitute that caused an outbreak of acuteonset Parkinson’s disease. Although it has been argued that paraquat does not cross the blood–brain barrier, ecologic studies and one case–control study suggest an association between herbicide exposure and Parkinson’s disease (Costa, 1997). Recently, combined exposure in early postnatal life to paraquat and the fungicide Maneb, both of which adversely affect dopamine systems, has been linked in toxicological studies to the development of Parkinsonian symptoms (Cory-Slechta et al., 2005). 24.5.3

Fungicides

Fungicides are applied to seeds, crops, and gardens to prevent growth of fungi. This class of synthetic pesticides encompasses a wide variety of chemicals, including copper, cadmium, organomercury and organotin compounds, substituted benzene, dithiocarbamates, thiophthalimides mancozeb, maneb, pentachlorophenol, and zineb. 24.5.3.1 Dimethyldithiocarbamates The dimethyldithiocarbamates ziram, ferbam, and thiram inhibit acetaldehyde dehydrogenase and have a disulfiram (Antabuse) effect. There have been reports of illness consistent with disulfiram reactions (nausea, vomiting, headache, diaphoresis, thirst, chest pain, and vertigo) among thiram-exposed workers who subsequently consumed ethanol. Ziram and ferbam are irritants and hemolysis occurred in one case of ziram poisoning. All dithiocarbamates are metabolized to carbon disulfide, which may explain similarities in the symptoms of poisoning. 24.5.3.2 Ethylenebisdithiocarbamates The ethylenebisdithiocarbamates (maneb, mancozeb, and zineb) are metabolized to ethylene thiourea, a potent animal carcinogen. This metabolite may also account for the antithyroid effects that occur in animals dosed with these compounds. Ethylene thiourea may concentrate on cooked food previously treated with ethylenebisdithiocarbamates. 24.5.3.3 Alkyl Mercury Alkyl mercury, although little used now in the United States, has resulted in occupational poisoning in seed treating facilities and has produced epidemics of poisoning and death including catastrophic fetotoxicity among farm families in New Mexico and Iraq who consumed treated seed grain or meat from animals that had consumed mercurytreated grain (Pierce et al., 1972; Marsh et al., 1980). Acute poisoning is manifested by headache, metallic taste, paresthesia, tremor, incoordination, slurred speech, constricted visual fields, hearing loss, loss of position sense, and spasticity. Among survivors, permanent neurologic effects are common. 24.5.3.4 Hexachlorobenzene Hexachlorobenzene is a potent inhibitor of uroporphyrinogen decarboxylase, resulting in increases in photosensitive porphyrins. Consumption of hexachlorobenzene-treated seed grain resulted in thousands of cases of poisoning that resembled porphyria cutanea tarda in Turkey in the 1950s. Bullous dermatitis, liver damage, hypertrichosis, and arthritis were permanent in many cases. Hexachlorobenzene levels can be measured in blood and metabolites can be measured in urine. 24.5.3.5 Pentachlorophenol Pentachlorophenol, used to treat lumber, is well absorbed through the skin. It produces an uncoupling of oxidative phosphorylation from oxidative

948

PESTICIDES

metabolism with resultant hyperthermia (McConnell, 1994). Numerous occupational deaths have occurred from overexposure. Symptoms include fever, thirst, sweating, weakness, tachycardia, other arrhythmias and tachypnea. Restlessness, anxiety, and dizziness reflect injury to the central nervous system. 24.5.4

Fumigants and Nematocides

Fumigants and nematocides are a chemically diverse group of pesticides, characterized by vapor pressures that are sufficiently high at room temperature to create airborne concentrations that cause acute toxicity. Inhalation is the principal route of absorption. 24.5.4.1 Acute Effects of Fumigants and central nervous system depression.

Many fumigants cause acute pulmonary edema

24.5.4.2 Methyl Bromide Methyl bromide vapors cause respiratory irritation, pulmonary edema, anorexia, nausea, vomiting, headache, visual disturbances, agitation, dizziness, tremor, incoordination, myoclonus, and muscle weakness. Liquid methyl bromide is absorbed dermally, and causes severe skin burns. Under(1) the Montreal Protocol on Substances that Deplete the Ozone Layer and (2) the Clean Air Act of 1990, production of methyl bromide was to be phased out by January of 2005. However, its production and use still continues under critical use exemptions (CUE). 24.5.4.3 Aluminum Phosphide Aluminum phosphide slowly releases phosphine upon contact with water in air; the release is more rapid in a moist environment. Phosphine is both a mucous membrane- and a respiratory-irritant that produces nausea, vomiting, diarrhea, headache, vertigo, fatigue, paresthesia, cough, dyspnea, chest tightness, and pulmonary edema. Nephro-, hepato-, cardio-, and central nervous system toxicity are common. Patients may smell of rotten fish or garlic. Deaths have occurred among people living near recently fumigated granaries, as a result of early reentry into fumigated structures and aboard grainhauling ships (Wilson et al., 1980). 24.5.4.4 Chronic Effects of Fumigants Survivors of acute methyl bromide poisoning may be left with organic brain damage, seizures, and personality disorders. Ethylene dibromide isa potentanimal carcinogen and,intheoccupational setting,ahumanspermatotoxin. Residues have been found in food and as a fumigant, it is no longer used in the United States. 24.5.4.5 Dibromochloropropane Dibromochloropropane (DBCP) causes decreased sperm counts and testicular atrophy (Babich and Davis, 1981). In 1977, almost one half of a group of poorly protected production workers exposed to DBCP in a plant in California were demonstrated to be azospermic or oligospermia. Recovery was better among initially oligospermic than among azospermic workers (Whorton et al., 1979). DBCP is an animal carcinogen. It has been removed from the continental United States market. 24.5.5

Other Pesticides

24.5.5.1 N,N-diethyltoluamide In the United States, N,N-diethyltoluamide (DEET) is the most commonly used insect repellent. Repellents are purposely applied topically or to

PESTICIDES AND ENDOCRINE/REPRODUCTIVE TOXICITY

949

clothing. Severe irritation of the eyes, irritation of the skin and exacerbation of pre-existing skin disease has been reported. Application of DEET formulations with high concentrations (more than 75% active ingredient) in tropical climates has resulted in severe dermatitis in antecubital and popliteal fossae. It is rapidly absorbed dermally and by ingestion. Behavioral effects have been demonstrated in chronic feeding studies in rats and several idiosyncratic cases of encephalopathy and death have been reported among children heavily treated with DEET. In one study of exposed workers, there was an increased prevalence of symptoms associated with impaired cognitive function and sleep disturbances and a trend toward poorer objectively measured neurobehavioral performance among highly exposed workers (McConnell et al., 1986). 24.5.6

Inert Ingredients

Pesticidal active ingredients represent only a portion of most pesticide formulations. In most countries, the nonpesticide components of commercial pesticide formulations are listed on the label as “inert ingredients.” These materials include solvents, emulsifiers, spreaders, stickers, penetrants, and anticaking agents. The term “inert” is misleading, and reflects only the lack of toxicity of these agents to pests; some inert ingredients are known or suspected to be human toxicants. Volatile mixtures of aliphatic and aromatic hydrocarbons are the most common inert ingredients. Some hydrocarbon agents used to facilitate the penetration of active ingredients to the interior of plants (penetrants) may be skin and eye irritants. The identity of these ingredients, some of which have acute and chronic toxicities, are available in most jurisdictions only at the discretion of the manufacturer (Vacco, 1996). EPA categorizes inert ingredients into four groups: (1) substances known to cause longterm health damage and to harm the environment; (2) those suspected of causing such health and environmental effects; (3) chemicals of unknown toxicity; and (4) those of minimal concern.

24.6 PESTICIDES AND ENDOCRINE/REPRODUCTIVE TOXICITY Concern has arisen in recent years that certain pesticides may have adverse effects on the endocrine system. It has been found that certain organochlorine pesticides such as DDT can interfere with the effects of estrogen. Indeed, it was the study of estrogenic effects of DDT in eagles and ospreys that led to Rachel Carson’s original recognition of the ecotoxicology of the persistent chlorinated hydrocarbon compounds (Carson, 1962). Many pesticides have been found to be estrogenic, including dieldrin, toxaphene, chlordane, DDT, and endosulfan (Soto et al., 1994). The estrogenic and antiestrogenic properties of pesticides may be examined in cell culture models, such as normal human mammary epithelial cells and human breast cancer cells. When these cells are exposed to an estrogenic substance they divide and grow more rapidly, and these effects can be quantified. Because hormones play critical roles in the early development of the immune, nervous, and reproductive systems, even low-dose exposure to endocrine disrupting pesticides during early windows of developmental vulnerability can have devastating effects on fetal life. The developmental effects of exposure to endocrine disrupters will vary depending on age at exposure and sex. It has been proposed that increased exposure to these agents may be the cause of an observed doubling in the incidence of undescended testes in male infants, which

950

PESTICIDES

has been reported since 1960. Effects have also been documented in the sexual development of wildlife in areas where there are elevated levels of pesticides such as DDT in the environment (Guillette et al., 1994). Studies of these effects in humans are in their early stages and are urgently needed. Preliminary analyses suggest associations between reduced sperm counts and elevated levels of pesticides in men’s urine (Swan, 2006). The Food Quality Protection Act of 1996 (see later) now requires that pesticides be tested for potential endocrine toxicity. For further information on endocrine disruptors.

24.7 PESTICIDES AND CHILDHOOD CANCER Steady increases have occurred over the past three decades in the incidence of the two most common pediatric malignancies—leukemia and brain cancer. The cumulative increase in the incidence of primary brain cancer since 1972 has been approximately 40%. During the same time, an increase of more than 50% has occurred in the incidence of testicular cancer among adolescents and young men (National Cancer Institute, SEER Database, 2006; Devesa et al., 1995). To investigate the possible contribution to these trends of prenatal or early postnatal exposures to pesticides, a series of epidemiologic studies have been undertaken. These studies have found consistent, modest associations between early pesticide exposure and childhood cancer. The strongest associations have been found for childhood leukemia and brain cancer. Risk estimates appeared most robust when exposure was characterized in the greatest detail. Highest risks were associated with frequent parental occupational exposure to pesticides and home pesticide use (Daniels et al., 1997). Clearly, more work is required in this area using prospective study designs and precise, real-time measures of exposure.

24.8 LEGISLATIVE FRAMEWORK The regulation of pesticides has been contingent on an assumed need to balance the economic benefits of these compounds against the risks associated with their use (National Research Council, 1993). That is the principle that underlay the Federal Insecticide, Fungicide and Rodenticide Act (FIFRA) of 1947, the first federal statute controlling pesticide use in the United States. This law, which has undergone several revisions, regulated pesticides on the basis that they would not cause “unreasonable adverse effects”. Unreasonable adverse effects were defined in the statute as “any unreasonable risk to man or the environment taking into account the economic, social, and environmental costs and benefits of its use.” This statement implies that some adverse effects were considered reasonable under FIFRA when weighed against potential economic benefits (Schierow, 1996). Due to rising concern about the health effects of pesticides in foods, particularly regarding the potential of some agents to cause cancer, Congress passed the Delaney Clause in 1958, as an amendment to the Federal Food, Drug, and Cosmetic Act. This Clause banned any pesticide that had been shown to cause cancer in humans or animals from processed foods. From the beginning, the Delaney Clause was highly controversial. Representatives of the pesticide and food industries argued that the law was too inflexible as it prohibited processed foods from containing even the smallest amount of carcinogenic chemicals. Environmentalists, in contrast, considered the Delaney Clause a bulwark of public health and environmental protection.

LEGISLATIVE FRAMEWORK

951

The Delaney Clause posed a policy dilemma by establishing a standard for pesticides in processed foods much stricter than that established for pesticides in raw foods—this was termed the “Delaney Paradox” (National Research Council, 1987). To copewith this paradox, the EPA for many years did not strictly enforce the Clause and allowed very low, or “de minimis,” levelsof carcinogenic pesticides in processed foods. In the Agency’sopinion, these levels posed no more than a “minimal risk” to health. In 1992, however, the United States Court of Appeals ruled that this approach contravened the intent of the Delaney Clause, and thus was not legal. This decision set the stage for passage in 1996 of the Food Quality Protection Act (FQPA), the major statute regulating pesticides in the United States today. The central premise of the Food Quality Protection Act is that there must be “a reasonable certainty that no harm will result from aggregate exposure to the pesticide chemical residue, including all anticipated dietary exposures and all other exposures for which there is reliable information”. In terms of public health, this is unquestionably a more protective standard than the previous risk–benefit standard based “no unreasonable adverse effect.” The intellectual foundation for the FQPA was provided by a landmark report released in 1993 by the National Academy of Sciences. This report found that the then current laws and regulations did not adequately protect children from the risks of pesticides in foods (National Research Council, 1993). The report recommended that Congress enact a legislation that would specifically consider the effects of pesticide residues on children’s health. Following that guidance, the FQPA requires that . .

. . .

Regulation of pesticides be based on health effects One uniform health-based standard be applied to pesticide residues in both processed and raw foods The latest scientific data be used for the assessment of health risks The regulation of “reduced risk” pesticides be streamlined Pollution prevention be promoted through integrated pest management practices.

Four provisions in the FQPA are aimed specifically at increasing protection for infants and children. Accordingly, the Environmental Protection Agency is directed to 1. Consider children’s sensitivities and unique exposure patterns to pesticides in setting pesticide standards. 2. Explicitly determine that tolerance levels are safe for children. 3. Adopt an additional safety factor of up to tenfold to account for uncertainty in the database relative to children, and to reflect children’s greater exposure and greater susceptibility to pesticides, unless there is reliable evidence that a different factor should be used. Consider sources of pesticide exposure in addition to diet when performing risk assessments and in setting tolerances. Enforcement of the provision of FQPA calling for the imposition of a child protective safety factor, in the setting of pesticide tolerance levels in foods, has been uneven. A “tolerance” is defined as the highest permissible level of a pesticide residue in a particular food. Traditionally, two tenfold safety factors have been employed in setting pesticide tolerance levels. These safety factors are based on the no observed effect level (NOEL), the lowest dose at which effects are seen in toxicological studies of pesticides conducted in

952

PESTICIDES

animals. The first of the two traditional tenfold safety factors is used to account for the extrapolation of toxicological data from animal species to humans. The second accounts for variation among humans. When both are applied in setting a tolerance, the tolerance is set at a level of 1/100th of the NOEL. The new, third, child protective factor was recommended by the National Academy of Sciences committee in their report Pesticides in the Diets of Infants and Children (1993). It is based on the increased biological susceptibility of children as compared to adults and on the substantial differences in exposure that have been demonstrated to exist between children and adults. The third tenfold safety factor is intended in the law as a “default provision”. It is intended to be automatically utilized, unless “reliable data” exist to show that there is no difference in susceptibility between children and adults. To date, the third tenfold safety factor has been applied to the tolerances for only 6 (12%) of 49 organophosphate pesticides that have been examined since the passage of FQPA. A threefold, child protective safety factor has been applied in an additional 6 (12%) instances (Benbrook, 2005b). Thus, for 75% of organophosphate pesticides, despite a widespread lack of data on developmental neurotoxicity, and substantial suspicion that such toxicity exists for many of these compounds, no child protective safety factor has been imposed in setting tolerances (Slotkin, 2004).

24.9 CONCLUSION: ISSUES FOR THE FUTURE Achieving full implementation of the Food Quality Protection Act (FQPA). A major provision of the FQPA is a requirement that a third tenfold, child protective safety factor be applied in setting tolerances for pesticide residues in food. This is meant to protect children against the noncarcinogenic effects of pesticide exposure, especially neurodevelopmental toxicity (Slotkin, 2004). As was noted in the preceding section, enforcement of this requirement has been uneven. Enhanced enforcement will be needed in the future if infants and children are to be adequately protected against the toxic effects of pesticides, especially developmental toxicity. A critical decision now confronting EPA is how to define “reliable data” under FQPA. The traditional toxicologic tests that have been used to assess differences in susceptibility between adult and young animals are relatively insensitive (Tilson, 1995). They tend not to show differences in susceptibility between infants and adults even when such differences may exist, unless the differences lead to gross anatomical defects (Rodier, 1995). Relying on such traditional tests, the EPA failed to apply a third tenfold safety factor in more than twothirds of the first 80 pesticides to come before the Agency for reassessment of tolerance in 1996 and 1997. Concern exists in the environmental community that even chemicals such as lead and PCBs, which are known to be functional neurotoxins but do not cause anatomical birth defects, would be judged “innocent” in this schema. This is an issue that will require close and continuous scrutiny in the years ahead. 24.9.1

Lifetime Toxicity Testing

Too few chemicals are tested for chronic or developmental neurotoxicity (Slotkin, 2004), and those that are examined are typically studied under test protocols in which the chemicals are administered during adolescence and the animals sacrificed and studied 12–24 months later. Functional assessment of neurological function is often not included. This approach

REFERENCES

953

misses the opportunity to study neurodevelopmental neurotoxicity and to examine possible late effects of early exposures. One approach to overcoming these limitations in design would be to require a lifetime toxicological approach, as has been used by the European Ramazzini Institute for the testing of the carcinogenicity of chemicals (Soffriti et al., 2005). In this approach, duration of toxicity testing protocols would be extended to incorporate administration of chemicals in early life— ideally in utero or even before conception—coupled with lifelong follow-up to natural death. Such expanded protocols may also incorporate functional neurobehavioral test batteries as well as neuropathologic examinations of relevant areas of the brain (Landrigan et al., 2005). 24.9.2

Addressing Pesticide Export Issues

Earlier in this chapter, the concept of the “circle of poison” was discussed. This phenomenon refers to the export of hazardous pesticides from the developed nations, in which they are manufactured but banned, followed by the subsequent return of those highly toxic pesticides on food grown in developing nations and imported to the industrialized world (20). Legislative pressure to resolve this problem has surfaced periodically in the past. To date, this legislative concern has not resulted in the passage of protective legislation. However, it seems likely that renewed legislative concern over this major unresolved issue will rise again. 24.9.3

Ending DDT Manufacture

DDT and other highly persistent and bioaccumulative pesticides continue to be manufactured and used in certain developing nations. A major international effort seeks to bring this manufacture to an end. An economic issue that must be addressed is that the chlorinated hydrocarbon compounds are relatively less expensive to manufacture than the organophosphates, carbamates, and other newer generation pesticides that have been introduced as substitutes. On the other hand, the increasing resistance of insects to DDT and other traditional pesticides may force this transition. 24.9.4

Risk Assessment Versus Pollution Prevention

For the past 20 years, risk assessment has been the predominant paradigm utilized by regulatory agencies to control exposure to pesticides. While risk assessment has had its successes, it is an inherently slow process that typically proceeds by considering only one chemical compound at a time. Moreover, a perennial problem is that pesticides under assessment are presumed innocent until proven to cause injury, and thus they typically remain on the market until the risk assessment is completed. A more effective paradigm for controlling exposures to pesticides consists of pollution prevention. Under this approach, described sometimes as the “precautionary principle”, numerical targets are set for reducing pesticide use over a span of years. One approach to pesticide use reduction is integrated pest management (13).

REFERENCES Babich A, Davis DL (1981) Dibromochloropropane (DBCP): a review. Sci. Total Environ. 17: 207–221.

954

PESTICIDES

Baker BP, Benbrook CM, Benbrook Groth E (2002) Pesticide residues in conventional, integrated pest management (IPM)- grown and organic foods: insights from three US data sets. Food Adit. Contam. 19:427–446. Bearer C (1995) How are children different from adults? Environ. Health Perspect. 103 (Suppl 6): 7–12. Benbrook CM (2005) Tracking the Impacts of the FQPA on Pesticide Dietary Risks: A Preliminary Assessment. 7-7-2005b. Consultant Report to the EPA Office of Inspector General. Berkowitz GS, Wetmur JG, Birman-Deych E, Obel J, Lapinski RH, Godbold JH, Holzman IR, Wolff MS (2004) In utero pesticide exposure, maternal paraoxonase activity, and head circumference. Environ. Health Perspect. 112 (3):388–91. Benbrook C (1996) Pest Management at the Crossroads. Yonkers NY: Consumers Union. Benbrook CM (2002) Organochlorine residues pose surprisingly high dietary risks. J. Epidemiol. Community Health 56:822–823. Blondell J (1997) Epidemiology of pesticide poisonings in the United States, with special reference to occupational cases. Occup. Med. State Art Rev. 12:209–220. Brenner BL, Markowitz S, Rivera M, Romero H, Weeks M, Sanchez E, Deych E, Garg A, Godbold J, Wolff MS, Landrigan PJ, Berkowitz G (2003) Integrated pest management in an urban community: a successful partnership for prevention. Environ. Health Perspect. 111(13):1649–53. California Department of Pesticide Regulation. Newly Registered Active Ingredients. (2005) Available at http://www.cdpr.ca.gov/docs/registration/ais/newreg/2005.pdf Accessed 14 June 2006. Cannon SB, Veazey JM, Jackson RS, Burse VM, Hayes C, Straub WE, Landrigan PJ (1978) Epidemic kepone poisoning in chemical workers. Am. J. Epid. 107:529–537. Carson R (1962) Silent Spring. Cambridge: Riverside Press. Cory-Slechta DA, Thiruchelvam M, Barlow BK, Richfield EK (2005) Developmental pesticide models of the Parkinson disease phenotype. Environ. Health Perspect. 113(9):1263–70. Costa L (1997) Basic Toxicology of Pesticides. Occup. Med. State Art Rev. 12:251–268. Daniels JL, Olshan AF, Savitz D (1997) Pesticdes and childhood cancers. Environ. Health Perspect. 105:1068–1077. DDT chapter. Wikipedia. Accessed 21 July 2006. Devesa SS, Blot WJ, Stone BJ, Miller BA, Tarone RE, Fraumeni JF Jr (1995) Recent cancer trends in the United States. J. Nat. Cancer Inst. 87:175–182. Diggory P, Landrigan PJ, Latimer KP, Ellington AC, Kimbrough RD, Liddle JA, Cline AF, Smrek AL (1977) Fatal parathion poisoning caused by contamination of flour in international commerce. Am. J. Epidemiol. 106:145–153. Engel SM, Berkowitz GS, Barr DB, Teitelbaum SL, Siskind J, Meisel SJ, Wetmur JG, Wolff MS (2007) Prenatal organophosphate metabolite and organochlorine levels and performance on the Brazelton neonatal behavioral assessment scale in a multiethnic pregnancy cohort. Am. J. Epidemiol. Fenster L, Eskenazi B, Anderson M, Bradman A, Hubbard A, Barr DB (2007) In utero exposure to DDT and performance on the Brazelton neonatal behavioral assessment scale. Neurotoxicology. Fingerhut MA, Halperin WE, Marlow DA (1991) Cancer mortality in workers exposed to 2,3,7,8tetrachlorodibenzo-p-dioxin. N. Engl. J. Med. 324:212–218. Groth E, Benbrook CM, Benbrook KL (2000) Pesticide Residues in Children’s Food. Yonkers, NY: Consumers Union. Guillette LJ Jr, Gross TS, Masson GR, Matter JM, Percival HF, Woodward AR (1994) Developmental abnormalities of the gonad and abnormal sex hormone concentrations in juvenile alligators from contaminated and control lakes in Florida. Environ. Health Perspect. 102: 680–688. Hayes WJ, Laws ER (1991) Handbook of Pesticide Toxicology. San Diego: Academic Press.

REFERENCES

955

Jeyaratnam J (1990) Acute pesticide poisoning: a major global health problem. World Health Stat. Q 43:139–144. Kimmel C (1992) Animal models for assessing developmental toxicity. Similarities and Differences Between Children and Adults. Washington, DC: ILSI Press. pp.43–65. Landrigan PJ, Claudio L, Markowitz SB, Berkowitz GS, Brenner BL, Romero H, Wetmur JG, Matte TD, Gore AC, Godbold JH, Wolff MS (1999) Pesticides and inner-city children: exposures, risks, and prevention. Environ. Health Perspect. (Suppl 3):431–437. Landrigan PJ, Sonawane B, Butler RN, Trasande L, Callan R, Droller D (2005) Early environmental origins of neurodegenerative disease in later life. Environ. Health Perspect. 113:1230–1233. Lotti M (1992) The pathogenesis of organophosphate neuropathy. Crit. Rev. Toxicol. 21:465–487. Lu C, Toepel K, Irish R, Fenske RA, Barr DB, Bravo R (2006) Organic diets significantly lower children’s dietary exposure to organophosphorus pesticides. Environ. Health Perspect. 114 (2):260–263. Marsh DO, Myers GJ, Clarkson TW, Amin-Zaki L, Tikriti S, Majud M (1980) Fetal methyl mercury poisoning: clinical and toxicological data on 29 cases. Ann. Neurol. 7:348–353 McConnell R (1994) Pesticides and related compounds. In: Rosenstock L, Cullen M, editors. Textbook of Clinical Occupational and Environmental Medicine. Philadelphia: WB Saunders Co. McConnell R, Fidler A, Chrislip D (1986) Health Hazard Evaluation Report. Everglades National Park, Everglades, Florida. HETA 83-085-1757. Cincinnati, Ohio: National Institute for Occupational Safety and Health. Melius JM (1998)The Bhopal disaster. In: Rom WN, editor. Environmental and Occupational Medicine. 3rd edn Boston: Little, Brown. Mortensen SR, Chanda SM, Hooper MJ, Padilla (1996) Maturational differences in chlorpyrifosoxonase activity may contribute to age-related sensitivity to chlorpyrifos. J. Biochem. Toxicol. 11:279–287. Narahashi T (1985) Nerve membrane ionic channels as the primary target of pyrethroids. Neurotoxicology 2:3–22. National Academy of Sciences (2006) Health Risks from Dioxin and Related Compounds: Evaluation of the EPA Reassessment. Washington: The National Academies Press. National Cancer Institute. SEER Database.http://seer.cancer.gov/. Accessed 14 February 2006. National Research Council (1996) Carcinogens and Anticarcinogens in the Diet. Washington: National Academy Press. National Research Council (1993) Pesticides in the Diets of Infants and Children. Washington: National Academy Press. National Research Council (1987) The Delaney Paradox. Washington: National Academy Press. Peto R, Gray R, Brantom P, Grasso P (1991) Dose and time relationships for tumor induction in the liver and esophagus of 4080 inbred rats by chronic ingestion of N-nitrosodiethylamine or N-nitrosodimethinalmine. Cancer Res. 51:6452–6469. Pierce PE, Thompson JF, Likosky WH (1972) Alkyl mercury poisoning in humans: a report of an outbreak. JAMA 220:1439–1442. Rauh VA, Garfinkel R, Perera FP, Andrews HF, Hoepner L, Barr DB, Whitehead R, Tang D, Whyatt RW (2006) Impact of prenatal chlorpyrifos exposure on neurodevelopment in the first 3 years of life among inner-city children. Pediatrics 118: (6). Rodier PM (1995) Developing brain as a target of toxicity. Environ. Health Perspect. 103:73–76. Schierow LJ (1996) Pesticide Policy Issues. Washington: The Library of Congress, Congressional Research Service, Environment and Natural Resources Policy Division (IB95016). Schuman SH, Simpson WM Jr (1997) A clinical historical overview of pesticide health issues. Occup. Med. State Art Rev. 12:203–207.

956

PESTICIDES

Slikker W, Xu ZA, Levin ED, Slotkin TA (2005) Mode of action: disruption of brain cell replication, second messenger, and neurotransmitter systems during development leading to cognitive dysfunction—developmental neurotoxicity of nicotine. Crit. Rev. Toxicology 35:703–711. Slotkin TA, Levin ED, Seidler FJ (2006) Comparative developmental neurotoxicity of organophosphate insecticides: effects on brain development are separable from systemic toxicity. Environ. Health Perspect. 114 (5):746–51. Slotkin TA (2004) Guidelines for developmental neurotoxicity and their impact on organophosphate pesticides: a personal view from an academic perspective. Neurotoxicology 25:631–640. Slotkin TA, Seidler FJ (2007) Prenatal chlorpyrifos exposure elicits presynaptic serotonergic and dopaminergic hyperactivity at adolescence: critical periods for regional and sex-selective effects. Reprod Toxicol. 23(3):421–7. Smith C (2001) Pesticide exports from U.S. ports. 1997–2000. Int. J. Occup. Environ. Health. 7 (4):266–274. Soffriti M, Belpoggi F, Degli Espositi D, Lambertini L (2005) Aspartame induces lymphomas and leukaemias in rats. Eur. J. Oncol. 10:107–116. Soto AM, Chung KL, Sonnenschein C (1994) The pesticides endosulfan, toxaphene, and dieldrin have estrogenic effects on human estrogen-sensitive cells. Environ. Health Perspect. 102:380–383. Swan SH (2006) Does our environment affect our fertility? Some examples to help reframe the question. Semin. Reprod. Med. 24:142–146. Tilson HA (1995) The concern for developmental neurotoxicology: is it justified and what is being done about it?. Environ Health Perspect 103(Suppl 6):147–151. U.S. EPA Office of Prevention, Pesticides, and Toxic Substances. Pesticides Industry Sales and Usage: 2000 and 2001 Market Estimates. Available at http://www.epa.gov/oppbead1/pestsales/01pestsales/market_estimates2001.pdf. Accessed 18 June 2006. U.S. Environmental Protection Agency. (1992) Regulatory Impact Analysis of Worker Protection Standards for Agricultural Pesticides. Washington: EPA. United Nations Environment Programme. Division of Technology, Industry and Economics. PIC Rotterdam Convention. Available at http://www.pic.int/index.html Accessed: 10 July 2006. United Nations Environment Programme. Stockholm Convention on Persistent Organic Pollutants (POP). Available at http://www.pops.int/documents/background/hcwc.pdf Accessed: 10 July 2006. Vacco DC (1996) The Secret Hazards of Pesticides: Inert Ingredients. Albany, NY: New York Department of Law, Environmental Protection Bureau. Wargo J (1996) Our Children’s Toxic Legacy. New Haven:Yale University Press. Whorton D, Milby TH, Krauss RM, Stubbs HA (1979) Testicular function in DBCP exposed pesticide workers. J. Occup. Med. 21:161–166. Wiles R, Davies E (1995) Pesticides in Baby Food. Washington:Environmental Working Group. Wiles R, Campbell C (1993) Pesticides in Children’s Food. Washington:Environmental Working Group. Wilson R, Lovejoy FH, Jaeger RJ, Landrigan PJ (1980) Acute phosphine poisoning aboard a grain freighter: epidemiologic, clinical, and pathological findings. JAMA 244:148–1150. Young JG, Eskenazi B, Gladstone EA, Bradman A, Pedersen L, Johnson C, Barr DB, Furlong CE, Holland NT (2005) Association between in utero organophosphate pesticide exposure and abnormal reflexes in neonates. Neurotoxicology. 26 (2):199–209. Zahm SH, Ward MH, Blair A (1997) Pesticides and cancer. Occup. Med. State Art Rev. 12:269–289.

25 SULFUR OXIDES—SO2, H2SO4, NH4HSO4, AND (NH4)2SO4 Morton Lippmann

Avariety of gaseous and particulate chemicals in the ambient air are acidic, and some of them are known to have produced health effects in exposed populations. The best documented of the effects in laboratory studies are attributable to strong acids in aerosol form, that is, sulfuric acid (H2SO4) and ammonium bisulfate (NH4HSO4). Among the weak acids, sulfur dioxide (SO2) and its hydrolysis products have been associated with both acute bronchoconstriction and elevated morbidity and mortality rates. With respect to these elevated rates, SO2 may have been a surrogate for its oxidation products and other particulate matter (PM) pollutants, which often coexist with SO2. This chapter summarizes and discusses the health effects attributable to SO2 and acidic aerosols at concentrations that have been monitored or measured in community and industrial atmospheres. The role of acidic aerosols in the health effects associated with ambient air PM is also discussed in Chapter 10. This chapter does not discuss effects associated with occupational exposures to these sulfur oxide or other acidic pollutants at much higher concentrations because of their lack of relevance to subject at issue, that is, the health effects of atmospheric acidity.

25.1 SOURCES AND EXPOSURES 25.1.1

Sources of Sulfur Oxides

Most of the sulfur in fossil fuel is converted into SO2 in the combustion zone, and it is vented to the atmosphere with the other products of combustion. A small fraction of the sulfur, generally less than 10%, is emitted as H2SO4, with some of it forming a surface film on ultrafine-sized mineral ash particles. When the discharge point is a tall stack, most of the SO2 escapes local deposition on terrestrial surfaces and is gradually (1–10%/h) converted into SO3, a highly hygroscopic vapor. The SO3 rapidly combines with water vapor to Environmental Toxicants, Third Edition Edited by Morton Lippmann Copyright
2009 John Wiley & Sons, Inc.

957

958

SULFUR OXIDES—SO2, H2SO4, NH4HSO4, AND (NH4)2SO4

produce ultrafine droplet aerosols of H2SO4. The H2SO4 is then gradually neutralized by ammonia, first to the strong acid ammonium bisulfate (NH4HSO4) and then to ammonium sulfate ((NH4)2SO4), a nearly neutral salt. Rates of ammonia neutralization vary widely, depending on emission rates from ground-based sources. Neutralization rates are high over cities and agricultural areas, low over forests, and virtually nil over deepwater bodies. The ratios between SO2, H2SO4, and total particulate sulfate (SO42 ) in the atmosphere are highly variable in space and time. While ambient concentration data are relatively plentiful for SO2 and, to a lesser extent, for SO42 , they are, unfortunately, very sparse for H2SO4 and NH4HSO4. These strong acid aerosols account for much of the mortality and morbidity historically associated with mixtures of SO2 and PM. SO2 is a very poor surrogate index for ambient concentrations of acid aerosols, but SO42 can often serve as an excellent surrogate in some parts of the United States. One of the few significant indoor sources of SO2 and H2SO4 is the unvented kerosene space heater. The sulfur content of kerosene is generally within the ASTM D-3699 standard of 0.04%. Leaderer et al. (1990) studied pollutant emissions from four portable kerosene space heaters using kerosene containing 0.039% sulfur. The heaters were operated in a 34 m3 room at 1.4 air changes per hour. Background chamber pollution levels were low. On a mass balance basis, SO42 accounted for 2–26% of the sulfur in the fuel, with the balance emitted as SO2. Sulfate concentrations ranged from 33 to 693 mg/m3, and acidic particulates, as H2SO4, ranged from 1.3 to 75 mg/m3. Since the sulfur content of kerosene has been reduced since 1990, these concentrations may represent upper limits. 25.1.2

Exposures to Sulfur Oxides

Current U.S. ambient air levels of SO2 are generally well within the current primary National Ambient Air Quality Standard (NAAQS) of 80 mg/m3 for an annual average and 365 mg/m3 for a 24 h maximum. There is an additional special concern for asthmatics’ peak exposures to SO2 while performing outdoor exercise. It has been estimated that the size of the asthmatic population with peak 5–10 min exposures at concentrations >0.2 ppm (520 mg/m3) during light to moderate exercise, who may exhibit a bronchoconstrictive response, varies from 5000 to 50,000. For acidic aerosols, there is a very limited ambient concentration database. Data on annual average acidic aerosol concentrations in U.S. communities in the 1980s were reported by Spengler et al. (1989). In the four eastern U.S. communities studied, the annual average ranged up to 1.8 mg/m3 (as H2SO4). In the 1980s, levels of acidic aerosol in excess of 20–40 mg/m3 (as H2SO4) were observed for time durations ranging from 1 to 12 h. These were associated with high but not necessarily the highest atmospheric SO42 levels. Exposures (concentration–time product) of 100–900 mg/m3 h were calculated for the acid events that were monitored. In the 1990s, mandated reductions of 50% in SO2 emissions resulted in comparable reductions in strong acid aerosols. By contrast, studies in London in the early 1960s indicated that acidity in excess of 100 mg/m3 (as H2SO4) was present in the atmosphere, and exposures >2000 mg/m3 h were possible. Brauer et al. (1989) measured exposures to acidic and basic vapors and aerosols with a personal annular denuder/filter pack sampler and compared the results to those measured at a centrally located monitoring site in the metropolitan Boston, MA, area. Personal exposures to aerosol Hþ were only slightly lower than the concentrations at the central monitor, and

SOURCES AND EXPOSURES

959

personal SO42 was similar to central site values. By contrast, SO2 and nitric acid (HNO3) vapors were much lower for personal exposures than at the central site. Meteorology and regional transport are extremely important to acid sulfate concentrations. Keeler et al. (1991) measured elevated levels of ambient Hþ simultaneously during a regional episode at multiple sites located from Tennessee to Connecticut, and Lamborg et al. (1992) measured Hþ concentrations to investigate the behavior of regional and urban plumes advecting across Lake Michigan. Their results suggested that aerosol acidity is maintained over long distances in air masses moving over large bodies of water (up to 100 km or more). The conversion of SO2 to acidic aerosols takes place as the prevailing winds carry the precursors from the source region in the midwest, northeast to the northeastern United States, and southwestern Canada. This type of northeasterly wind flow occurs on the backside (western side) of midlatitude anticyclones (high-pressure systems). Highest atmospheric acidity is associated with: (1) slow westerly winds traversing westward SO2 source areas; (2) local stagnation; or (3) regional transport around to the backside of a high-pressure system. Low acidity is associated with fast-moving air masses and with winds from the northerly directions; upwind precipitation also played a moderating role in air parcel acidity. Much of the SO2 and aerosol Hþ originates from coal-fired power plants. Size distributions of aerosol Hþ and SO42 are alike, with MMED 0.7 mm, in the optimum range for efficient light scattering and inefficient wet/dry removal. Thus, light scattering and visual range degradation are attributable to the acidic SO42 aerosol. Due to the inefficient removal of aerosol Hþ, strong acids may be capable of long-distance transport in the lower troposphere. Water associated with the acidic aerosol was shown to account for much of the light scattering. A study of acid aerosols and ammonia (Suh et al., 1992) found no significant spatial variation of Hþ at Uniontown, a suburb of Pittsburgh, PA. Measurements at the central monitoring site accounted for 92% of the variability in outdoor concentrations measured at various homes throughout the town. There was no statistical difference (p > 0.01) between concentrations of outdoor Hþ among five sites (a central site and four satellite sites) in Newtown, CT (Thompson et al., 1991). However, there were differences in peak values, which were probably related to the proximity of the sampling sites to ammonia sources. These studies suggest that while peak values may differ significantly, long-term averages should not substantially differ across a suburban community. Outdoor concentrations of Hþ in small suburban communities are fairly uniform, suggesting that minor differences in population density do not significantly affect outdoor Hþ or NH3 concentrations (Suh et al., 1992). In urban areas, however, both Hþ and NH3 exhibit significant spatial variation. Waldman et al. (1990) measured ambient concentrations of Hþ, NH3, and SO42 at three locations in metropolitan Toronto. The sites, located up to 33 km apart, had significant differences in outdoor concentrations of Hþ. Waldman and coworkers reported that the sites with higher NH3 measured lower Hþ concentrations. An intensive monitoring study was conducted during the summers of 1992 and 1993 in Philadelphia (Suh et al., 1995). Twenty-four hour measurements of aerosol acidity (Hþ), sulfate, and NH3 were collected simultaneously at seven sites in metropolitan Philadelphia and at Valley Forge, 30 km northeast of the city center. They reported that SO42 was evenly distributed throughout the measurement area but Hþ concentrations varied spatially within metropolitan Philadelphia, related to local variations in NH3 concentrations (Fig. 25.1). The amount of NH3 available to neutralize Hþ increased with population

960

SULFUR OXIDES—SO2, H2SO4, NH4HSO4, AND (NH4)2SO4

0.5

5

120 SO4–

0.4

4

0.2

3

2

SO4–, H+(nmol/m3)

Ratio

0.3

NH3 (ppb)

90 NH3

60 H+/SO4–

30 0.1

1 H+

0

0

0 0

5000 10,000 15,000 20,000 Population density (persons/sq. mile)

FIGURE 25.1 Mean air pollutant concentrations for days when winds were from the southerly direction, plotted versus population density. The solid line represents Hþ concentrations, the long dashed line represents SOV concentrations, the dashed and dotted line represents the radio of Hþ to SOV levels, and the dotted line represents NH3 concentrations. All data collected in Philadelphia during the summers of 1992 and 1993 (Source: Adapted from Suh et al., 1995).

density, resulting in lower Hþ concentrations in more densely populated areas. The extent of the spatial variation in Hþ concentrations did not appear to depend on the overall Hþ concentration. It did, however, show a strong inverse association with local NH3 concentrations. An analysis of results from Harvard’s 24-city study (Thompson et al., 1991; Spengler et al., 1996), which measured acid aerosol concentrations at eight different small cities across North America each year during a 3-year period, revealed that the summer Hþ mean concentrations were significantly higher than the annual means at all sites. The results showed that at the sites with high Hþ concentrations, approximately two-thirds of the aerosol acidity occurred from May through September. Wilson et al. (1991) examined concentration data for Hþ, NH3, and SO42 from the Harvard 24-city study for the evidence of diurnal variability (Fig. 25.2). A distinct diurnal pattern was found for Hþ concentrations and the Hþ/SO42 ratio, with daytime concentrations being substantially higher than nighttime levels. Both Hþ and SO42 concentrations peaked between noon and 6:00 p.m. No such diurnal variation was found for NH3. They concluded that the diurnal variation in Hþ was probably due to atmospheric mixing. Air containing high concentrations of Hþ and SO42 mixes downward during daylight, when the atmosphere is unstable and well mixed. During the night, ammonia emitted from groundbased sources neutralizes the acid in the nocturnal boundary layer, the very stable lower part of the atmosphere, but a nocturnal inversion prevents the ammonia from reacting with the acid aerosols aloft. Then, in the morning as the nocturnal inversion dissipates, the acid aerosols mix downward again as the process begins anew. Spengler et al. (1996) also noted diurnal variations in SO42 and H2SO4 concentrations and suggested atmospheric dynamics as the cause. The diurnal variation in SO42 has been observed by other workers and discussed in terms of atmospheric dynamics by Wolff et al. (1979) and Wilson and Stockburger (1990).

HEALTH EFFECTS

961

5 Sulfate Hydrogen ion nmol/m3 (thousand)

4

3

2

1

0 0

FIGURE 25.2 et al., 1991).

20

40

60

80

100 Hour

120

140

160

180

200

Diurnal pattern of sulfate and hydrogen ion at Harriman, Tennessee (Source: Wilson

25.2 HEALTH EFFECTS 25.2.1

Health Effects of SO2

25.2.1.1 Dosimetry of Inhaled SO2 As a water-soluble acidic vapor, SO2 is efficiently captured in the upper respiratory tract during inhalation, and virtually none penetrates to the lungs during normal, quiescent breathing. However, during vigorous physical activity, there is less residence time in the upper airways and, in humans, a shift to oronasal breathing involving partial flow through the less efficient oral passages. Under exercise conditions, some inhaled SO2 can penetrate to the smaller conductive airways of the lungs and perhaps beyond them. Skornik and Brain (1990) showed that hamsters exposed to SO2 while running had reduced pulmonary macrophage endocytosis of particles in comparison to shamexposed animals. 25.2.1.2 Acute Bronchoconstrictive Effects of SO2 in Humans For asthmatics and others with hyperreactive airways exposed to SO2 at 0.25–0.50 ppm and higher while exercising, the most striking acute response is rapid bronchoconstriction (airway narrowing), usually evidenced in increased airway resistance, decreased expiratory flow rates, and the occurrence of symptoms such as wheezing and shortness of breath. Similar responses can be produced in healthy persons, but require exposure concentrations about an order of magnitude higher and outside the range of ambient levels. The penetration of SO2 to sensitive portions of respiratory tract is largely determined by the efficiency of the oral or nasal mucosa in absorbing SO2, which in turn depends on the mode of breathing (nasal, oral, or oronasal) and the rate of airflow. Controlled SO2 exposure studies on asthmatics show that at comparable SO2 concentrations, bronchoconstrictive effects increase with increased ventilation rates and with the relative contribution of oral ventilation to total ventilation (Bethel et al., 1983; Roger et al., 1985). Increased oral ventilation not only allows more direct penetration of SO2, but may also result in airway

962

SULFUR OXIDES—SO2, H2SO4, NH4HSO4, AND (NH4)2SO4

drying and changes in surface liquids, affecting SO2 absorption and penetration. Evaporation of airway surface liquid and perhaps convective cooling of the airways caused by cold, dry air can act as direct bronchoconstrictive stimuli in asthmatics (Deal et al., 1979; Strauss et al., 1977; Anderson, 1985). The combined effect of SO2 and cold, dry air further exacerbates the asthmatic response (Bethel et al., 1984; Sheppard et al., 1984; Linn et al., 1985). The bronchoconstrictive effects of SO2 are reduced under warm, humid conditions (Linn et al., 1985). To determine whether bronchoconstriction induced by SO2 can be predicted by the airway response to inhaled histamine, Magnussen et al. (1990) exposed 46 patients with asthma to air or 0.5 ppm SO2 for two days. The exposure protocol consisted of 10 min of tidal breathing followed by 10 min of isocapnic hyperventilation at a rate of 30 L/min. Airway response was measured before (baseline) and after hyperventilation in terms of specific airway resistance, SRaw. Exposure to air increased baseline mean SRaw by 45%, whereas exposure to SO2 with hyperinflation increased mean baseline SRaw by 163%. When evaluated individually, 26 and 34 of the 46 patients showed an airway response to hyperventilation of air and SO2, respectively. The airway response after SO2 and histamine showed a weak but significant correlation (R ¼ 0.48), whereas the responses to hyperventilation and SO2 did not correlate. Thus, the mechanisms by which histamine and SO2 exert their bronchomotor effects are different, and the risk of SO2-induced asthmatic symptoms is poorly predicted by histamine responsiveness. The response to inhaled SO2 can also be exacerbated by prior exposure to ozone (O3). Koenig et al. (1990) exposed eight male and five female adolescent asthmatics during intermittent exercise to a sequence of atmospheres, with 45 min to one followed by 15 min to the other. The combinations were: (1) air–100 ppb SO2; (2) 120 ppb O3–120 ppb O3; and (3) 120 ppb O3–100 ppb SO2. Air–SO2 and O3–O3 did not cause significant changes in function. By contrast, O3–SO2 produced significant changes, that is, an 8% decline in FEV1, a 19% increase in total flow resistance, and a 15% decrease in Vmax50. Little time is required for SO2 exposure to elicit significant bronchoconstriction in exercising asthmatics; exposure durations as short as 2 min at 1.0 ppm have produced significant responses (Horstman et al., 1988). Little enhancement of response is apparent on prolonged exposure beyond 5 min, although some suggestion of an increase is seen with continuous exercise between 10 and 30 min (Kehrl et al., 1987). Following a single SO2 exposure during exercise, airway resistance in asthmatics appears to require a recovery period of 1–2 h (Hackney et al., 1984). The magnitude of response induced by any given SO2 concentration is variable among asthmatics. Exposures to SO2 concentrations of 0.25 ppm or less, which do not induce significant group mean increases in airway resistance, also do not cause symptomatic bronchoconstriction. On the contrary, exposures to 0.40 ppm SO2 or greater (combined with moderate to heavy exercise), which induce significant group mean increases in airway resistance, also cause substantial bronchoconstriction in some individual asthmatics. This bronchoconstriction is often associated with wheezing and the perception of respiratory distress, sometimes necessitating the discontinuance of the exposure and the provision of medication. The significance of these observations is that some SO2-sensitive asthmatics are at risk of experiencing symptomatic bronchoconstriction requiring termination of activity and/or medical intervention when exposed to SO2 concentrations of 0.40–0.50 ppm (1040–1300 mg/m3) or greater when this exposure is accompanied by at least moderate activity. These concentrations can occur downwind of point sources as 10 min averages. Various studies have examined exposure–response relationships over various concentration and ventilation ranges. Some examined the influence of various subject-related

HEALTH EFFECTS

963

and environmental factors. Since the individual studies used different conditions of airway entry, ventilation rate, concentration, and so on, it is difficult to compare directly the results from different investigations. An approach used by Kleinman (1984) and Linn et al. (1983) normalizes studies according to effective oral dose rate. They showed that reasonably consistent results are derived from the various controlled SO2 asthmatic studies when adjustments are made for differences in ventilation rates and oral/nasal breathing patterns. 25.2.1.3 Associations Between SO2 and Ambient Mortality and Morbidity Evidence for effects of SO2 other than short-term bronchoconstriction is less direct. There is a considerable body of epidemiological evidence demonstrating statistically significant associations between SO2 and rates of mortality and morbidity. However, it is less likely that SO2 was a causal factor than that it was serving as a surrogate exposure index for other pollutants in the sulfur oxide–PM complex deriving from fossil fuel combustion. 25.2.1.4 Multipollutant Epidemiology that Includes SO2 Older epidemiological studies (up to about the mid-1980s) assessing the health effects of air pollution, including those caused by SO2, have not been considered as providing reliable evidence for the independent effects of SO2. Rather, they assessed the effects of the traditional pollutant mixture produced by fossil fuel combustion processes, which included particulate matter and SO2 as primary pollutants plus secondary particles, including acid aerosols. Although epidemiological studies of air pollution exposure have the advantage of studying the populations of interest (including sensitive individuals) exposed at the usual ambient pollutant levels and monitoring relevant outcomes (transient or irreversible), they have the drawback that they inevitably study exposure to a pollutant mixture. In recent years, however, more sophisticated statistical methodology has allowed partial separation of the effects of individual pollutants via modeling. Furthermore, a large number of published studies allow an overall evaluation of the effects of SO2 in situations with varying pollutant mixes, and in particular with different levels of PM. The Ozkaynak and Spengler (1985) reanalysis of 14 years of New York City data (1963–1976) found significant associations between excess daily mortality and airborne particulate matter, SO2, and temperature. Differences in the rate of change of SO2 and PM indicators during the study period allowed estimation of their separate effects. In joint regression analysis across all years, PM indicators (coefficient of haze and visibility extinction coefficient) together accounted for significantly greater excess mortality than did SO2. The main focus of the air pollution epidemiological studies in the past decade has been on the health effects of PM. However, numerous studies have also examined SO2 and other gaseous pollutants as potential confounders of PM’s effects. Thus, a large number of risk estimates for SO2 have accumulated, providing a more comprehensive assessment of relative importance of the classical air pollutants. While these observational studies have not resolved the issue of confounding between SO2 and PM or other pollutants and have not systematically examined the synergistic effects, they are still generally useful in assessing the potential adverse health impacts of SO2. When multiple pollutants were evaluated, PM has tended to be more strongly associated with mortality or morbidity outcomes than has SO2, but there were exceptions. The discussion focuses on studies published in and after 1997. To minimize the potential influence of bias due to the software convergence issue of that confounded analyses using the Generalized Additive Model (GAM) (Dominici et al.,

964

SULFUR OXIDES—SO2, H2SO4, NH4HSO4, AND (NH4)2SO4

2002; Ramsay et al., 2003), the discussion focuses on those studies that were unaffected or have been reanalyzed. Short- and long-term effects are considered separately. 25.2.1.5 Short-Term Effects In the past decade, there have been nearly 200 mortality and morbidity time-series studies that examined short-term impacts of PM, and about 60% of these studies also examined the impacts of SO2. There have also been several multicity studies of mortality and morbidity in Europe, the United States, and Canada that also examined SO2. These multicity studies have advantages over a collection of single-city studies because they analyze data from many cities using consistent methodology and attempt to explain variations in the risk estimate using city characteristics (differences in weather, poverty, etc.), and this discussion focuses on the results from the multicity studies. 25.2.1.6 Mortality Studies A series of studies from Air Pollution and Health: A European Approach (APHEA) project examined mortality effects of air pollution in multiple cities. The APHEA 1 project (Katsouyanni et al., 1997) reported total nonaccidental mortality risk estimates for SO2 and PM in 12 European cities. It noted that the effects of these two pollutants were “mutually independent,” and were stronger during the summer. The observed associations were stronger in western European cities than in central and eastern European cities (see Table 25.1). An examination of cause-specific mortality in a 10-city subset found that estimated risks were larger for cardiovascular and respiratory categories than those for total nonaccidental mortality (Zmirou et al., 1998). Samoli et al. (2001) (reanalysis by Samoli et al., 2003a, 2003b) applied an alternative model (a more flexible smoothing model to adjust for seasonal cycles) to the 12 cities data and also conducted subset analyses for moderate SO2 levels (less than 200 and 150 mg/m3). They found that both the alternative model and the restriction of the data to lower SO2 levels produced higher SO2 risk estimates and reduced the contrast between western, central, and eastern risk estimates, though the differences still remained. The APHEA 2 project expanded the number of cities to 29, increasing the statistical power to explain possible city-to-city variations in air pollution mortality effects. However, its published mortality studies’ focus has been on either PM indices (Katsouyanni et al., 2001; reanalysis by Katsouyanni et al., 2003; Aga et al., 2003), NO2 (Samoli et al., 2003a, 2003b), or O3 (Gryparis et al., 2004), and no mortality risk estimates were reported for SO2. The PM effects analyses reported that PM risk estimates were not affected by including SO2 in the models. The PM analyses, in their second stage regressions, also found that NO2 was an important effect modifier of PM (i.e., the cities with higher NO2 levels showed larger PM risk estimates) in total mortality (Katsouyanni et al., 2001, 2003) and in elderly mortality (Aga et al., 2003). Although they did not report numerical results, the results imply that the difference in SO2 levels across cities did not alter the PM risk estimates. A Spanish multicity study (Spanish multicenter study on air pollution and mortality, or EMECAM) analyzed short-term associations between mortality and SO2 and PM in 13 Spanish cities (Ballester et al., 2002). They examined both 24 h average and daily 1 h maximum SO2 levels. The estimated mortality risks for the 24 h average SO2 were greatly reduced when two-pollutant models with PM were performed, but the estimates for 1 h maximum SO2 were not attenuated by PM. They concluded that peak rather than the daily average concentrations of SO2 were related to mortality.

HEALTH EFFECTS

965

TABLE 25.1 Estimated Total (Nonaccidental) Mortality Percent Excess Deaths (95% CI in Parenthesis) Per 50 mg/m3 Increase in SO2 Reported in Recent Multicity Time-Series Studies and Meta-Analyses Study APHEA 1 (Katsouyanni et al., 1997), 12 European cities

APHEA 1 (Samoli et al., 2001, 2003a, 2003b), 12 European cities: using natural splines rather than sine/cosine to adjust for temporal trends EMECAM (Ballester et al., 2002; GAM study), 13 Spanish cities NMMAPS (Samet et al., 2000; Dominici et al., 2003), 90 largest U.S. cities Stieb et al. (2002, 2003) meta-analyses

Estimate Western Europe: 2.9% (2.3, 4.6) at the best lag between 0 and 3 days for each city Central eastern Europe: 0.9% (0.2, 1.5) Western Europe: 2.6% (2.1, 3.1) Central eastern Europe: 0.7% (0.0, 1.4)

Comment The effects of SO2 and PM were “mutually independent”

Restricting data range below 150 mg/m3 or 200 increased SO2 risk estimates

2.5% (0.3, 4.9), average of lag 0 and 1 day 1.1% (0.5, 1.7) at lag 1 day

Adding copollutants reduced the estimate by 20% and widened confidence bands

Non-GAM: single pollutant (29 studies): 1.7% (1.2, 2.3) With copollutant(s) (10 studies): 1.6% (0.6, 2.5) GAM: single pollutant (17 studies): 2.0% (1.3, 2.6) With copollutant(s) (11 studies): 1.6% (0.8, 2.4)

The largest U.S. multicity mortality study, the National Morbidity, Mortality, and Air Pollution Study (NMMAPS), had PM10 as its main focus. However, it also analyzed SO2 and other gaseous pollutants in the 90 largest U.S. cities (Samet et al., 2000; Dominici et al., 2003). In their reanalysis, Dominici et al. noted that the results did not indicate significant associations for SO2 (or for NO2 or CO) with total mortality. These three pollutants were generally less strongly associated with mortality than PM10 or O3. The combined estimates across cities for SO2 (and for NO2 and CO) were positive and significant at lag 1 day in singlepollutant models and remained positive (though not significant because of larger confidence intervals (CIs)) with additions of other pollutants. The estimated excess total mortality risk estimate per 50 mg/m3 was smaller than those estimated in the APHEA 1 studies. The results from the Canadian eight-city study (Burnett et al., 2000, GAM-affected; reanalyzed by Burnett and Goldberg, 2003, but SO2 and other gaseous pollutants were not reanalyzed) indicate that while SO2 was significantly associated with total mortality in a single-pollutant model at lag 1 day, adding PM2.5 into the regression model reduced the SO2 risk estimates and SO2’s association with total mortality was generally weakest among the pollutants. In the Canadian 11-city study (Burnett et al., 1998, GAM-affected and not

966

SULFUR OXIDES—SO2, H2SO4, NH4HSO4, AND (NH4)2SO4

reanalyzed), among the gaseous pollutants (PM was not analyzed), the estimated excess mortality risk for SO2 at the mean level (15 mg/m3), 1.4%, was smaller than those for NO2 (4.1%) or O3 (1.8%). Stieb et al. (2002, 2003) (reanalysis to evaluate the impact of GAM-affected studies) conducted meta-analyses of air pollutants by extracting results from 109 time-series mortality studies undertaken worldwide. For SO2, there were 46 studies (29 non-GAM and 17 GAM studies) that reported single-pollutant estimates and 21 studies (10 non-GAM and 11 GAM studies) that reported estimates with copollutant(s) in the model. As shown in Table 25.1, the impact of GAM as well as an inclusion of copollutants appears to be small. There are several single-city studies that warrant attention. Hoek et al. (2000, 2001) (reanalysis by Hoek, 2003) analyzed associations between air pollution and total mortality as well as deaths from specific cardiovascular causes in the entire Netherlands. PM10, black smoke (BS), SO2, O3, NO2, and CO were analyzed in single- and two-pollutant models in these studies. Essentially, all the pollutants were significantly associated with total mortality in single-pollutant models. In two-pollutant models with SO2 and each of the PM indices (PM10, black smoke, sulfate, and nitrate), SO2 was more strongly associated with total mortality than the PM indices. Wichmann et al. (2000) (reanalysis by Stolzel et al., 2003) examined the mortality effects of fine and ultrafine particles in Erfurt, Germany. The number and mass concentrations of several size ranges of ultrafine particles, as well as PM2.5, PM10, TSP, SO2, NO2, and CO, were analyzed. Among the various PM indices, the strongest associations were found for the number concentrations in the 0.01–0.03 mm range and mass concentrations in the 0.01– 2.5 mm size range. SO2 was associated with mortality more strongly than any of the fine, ultrafine particulate indices and other gaseous pollutants. In two-pollutant models with PM indices, SO2 remained more strongly associated with mortality than the PM indices. However, the authors stated that “the persistence of the SO2 effect was interpreted as an artifact, because the SO2 concentration was much below the levels at which effects are usually expected.” Although the time-series studies provide estimates of excess deaths from regression models, there remains the question of whether a reduction in SO2 actually results in a reduction in deaths. A sudden change in regulation can provide a basis for treating the results as coming from an “intervention study.” Such a situation occurred in Hong Kong, China, in July, 1990 when a restriction was introduced over one weekend that required all power plants and road vehicles to use fuel oil with a sulfur content of not more than 0.5% by weight (Hedley et al., 2002). In another recent “intervention study,” in Dublin, Ireland (Clancy et al., 2002), the ban on coal sales led to 70% reduction in black smoke, but only a 34% reduction in SO2. In the Hong Kong case, SO2 levels after the intervention declined about 50%, while PM10 levels did not change. In the Hedley et al. study, the average annual trend in death rate significantly declined after the intervention for all causes (2.1%), respiratory (3.9%) causes, and cardiovascular causes (2.0%). It should also be noted that a time-series mortality study in Hong Kong (Wong et al., 2001) suggested that SO2 was the pollutant most consistently associated with mortality, whereas PM10’s association with mortality was only marginal. This appeared to support the case for SO2, not PM, being the more influential air pollutant in this locale. Thus, the Hong Kong case suggested that a reduction in SO2 emissions led to an immediate reduction in deaths. However, where Hedley et al. (2002) analyzed the elemental composition of the PM, they showed that nickel (Ni) and vanadium (V), but not other elements, also had sudden and prolonged concentration drops that could have accounted for the reduction in mortality (see Chapter 10).

HEALTH EFFECTS

967

25.2.1.7 Morbidity Studies The focus of air pollution acute morbidity studies in the past decade has also been on PM, but there have been several multicity studies (mostly APHEA projects) that examined SO2 either as a potential confounder for PM or as a pollutant of primary interest. In the following, we will not only summarize the results of the multicity studies but also describe several important single-city studies. APHEA 1 examined associations between emergency hospital admissions for asthma and black smoke, SO2, NO2, and O3 in four European cities (Sunyer et al., 1997). Pediatric (age <15 years) and adult (age 15–64 years) subjects were analyzed. They found associations between asthma admissions and NO2 in adults and SO2 in children. These associations were reported to be independent of black smoke. It should be noted that the associations between SO2 and pediatric asthma admissions were seen in London and Paris, but not in Helsinki. Another APHE A1 project examined associations between hospital admissions for chronic obstructive pulmonary disease (COPD) and SO2, black smoke, TSP, NO2, and O3 in six European cities (Anderson et al., 1997). In the combined estimates across cities, SO2 was not as strongly associated with COPD admissions as the other pollutants, but this appeared to be at least partly due to the larger heterogeneity of SO2 estimates across cities; the SO2 risk estimates were significantly positive in Paris, Milan, and Barcelona, but negative in Amsterdam and Rotterdam. Other pollutants showed more consistent estimates across cities, resulting in overall statistically significant estimates. Their analysis by season found that SO2–COPD admission associations were stronger in warm seasons. Spix et al. (1998) summarized associations between air pollution and hospital admissions for respiratory diseases by age (15–64 and >65 years) in five west European cities as part of the APHEA 1 project. In this study, the most consistent associations for both adult and elderly respiratory admissions were found with O3. The authors concluded that “no consistent evidence of an influence on respiratory admissions was found” for SO2. However, they also noted that the heterogeneity of estimated SO2 effects across the cities was best explained by the number of stations providing data (i.e., larger effects for cities with more monitoring stations). Thus, the exposure estimation error associated with SO2 may have affected the results. The combined effect estimate for elderly admissions was positive and significant. In the APHEA 2 project, Atkinson et al. (2001) (reanalysis of GAM in Atkinson et al., 2003) investigated acute effects of PM on respiratory admissions in eight European cities, but SO2 was examined only for its influence on PM risk estimates in two-pollutant models and the risk estimates for SO2 were not reported. Asthma (age 0–14 and 15–64 years), COPD, and all-respiratory causes (age >65 years) were examined. PM, especially PM10, was associated with these outcomes, and O3 was suggested as a potential effect modifier of the PM effects. The inclusion of SO2 in the models only modified (reduced) PM10–asthma associations in the 0–14-year age group. Sunyer et al. (2003a) (a GAM study) specifically examined the effects of SO2 on the respiratory admissions in the seven APHEA 2 cities. The respiratory categories examined were the same as those analyzed by Atkinson et al. above. SO2 was associated with asthma admissions in children, but not with other respiratory diseases in other age groups. The authors also noted that the SO2 risk estimates were sensitive to the inclusion of PM10 or CO in the models. Due to relatively high correlations among these pollutants, the issue of potential confounding could not be resolved. As part of the APHEA 2 project, Le Tertre et al. (2002) (reanalysis in Le Tertre et al., 2003) examined the association between PM10 and black smoke and hospital admissions for cardiovascular causes in eight European cities. Hospital admissions for total cardiovascular, cardiovascular for age >65 years, ischemic heart disease (IHD) for age 0–64 years, IHD for age >65 years, and stroke for age >65 years were analyzed. They did not specifically

968

SULFUR OXIDES—SO2, H2SO4, NH4HSO4, AND (NH4)2SO4

estimate SO2 effects, but examined the sensitivity of PM risk estimates when SO2 and other gaseous pollutants were added. Adding SO2 in the regression models did not affect PM risk estimates, but adding CO and especially NO2 greatly reduced PM risk estimates. The authors concluded that the primary effect was likely attributable to diesel exhaust. Sunyer et al. (2003b) (a GAM analysis) analyzed the same outcomes as those analyzed by Le Tertre et al. in seven cities (Barcelona excluded) and provided the combined SO2 risk estimates across seven cities. Single-pollutant models resulted in positive and significant SO2 risk estimates for all of the cardiac outcomes except stroke. However, these estimates were reduced when CO, NO2, black smoke, or PM10 were included in the models except for IHD admissions for ages below 65 years. The authors noted that SO2 could be a surrogate of urban pollution mixtures, which in some cases is more strongly associated with cardiovascular hospitalizations than particles. The NMMAP analysis of elderly respiratory and cardiovascular hospital admissions from 14 U.S. cities focused on PM10 effects. SO2 was analyzed only to examine its influence on PM10 risk estimates in the second-stage regression (Samet et al., 2000; reanalysis by Schwartz et al., 2003). The authors concluded that there was little evidence of PM10 effects being confounded by SO2. There were several other smaller scale studies that suggested the roles of SO2 in respiratory and cardiovascular outcomes. A 6-month follow-up of 84 asthmatic children in Paris found an association between air pollution and increased asthma attacks and symptoms in mild asthmatic children (Segala et al., 1998). The strongest association was found for the risk of asthma attacks and SO2 on the same day. A comparison of air pollution effects on respiratory and cardiovascular hospital admissions in Hong Kong and London found that SO2 was associated with cardiac admissions after adjusting for other pollutants (Wong et al., 2002; a GAM analysis). The Hong Kong “intervention” event described earlier also provided an opportunity to investigate health end points other than mortality. Wong et al. (1998) compared the effects of the intervention on bronchial responsiveness in primary schoolchildren living in two districts (polluted versus less polluted) in Hong Kong. Bronchial hyperreactivity (BHR) and bronchial reactivity (BR) slope were used to estimate responses to a histamine challenge. They found a greater decline in both BHR and BR slope in the polluted district than in the less polluted district. The results suggest that the reduction in SO2 emissions was associated with reduction in bronchial hyperresponsiveness in schoolchildren. 25.2.1.8 Long-Term Multipollutant Effects Studies Earlier studies on the chronic effects of air pollutants relied on cross-sectional comparisons that could be subject to ecologic confounding. More recent studies often involve investigations of large cohorts for which detailed individual-level information is collected to adjust for confounding. The air pollution exposure estimates in these studies are still “ecologic” in the sense that all the subjects in a community are assigned the same community average air pollution level, but the ability to adjust for potential confounders (smoking, diet, body mass index, occupational exposures, etc.) on the individual level is a major advantage over purely ecologic studies. Hence, this type of study is called “semi-individual” (Kunzli and Tager, 1997). Since these are prospective cohort studies, they require extended periods and resources, and thus, there have not been many such studies. Krewski et al. (2000) reanalyzed two large U.S. cohort studies, the Harvard six-city study (Dockery et al., 1993) and the American Cancer Society (ACS) data (Pope et al., 1995). Their replication analyses confirmed the original investigators’ findings of PM effects, and their

HEALTH EFFECTS

969

additional analyses of the ACS data reported several interesting observations. Of the gaseous pollutants examined (SO2, NO2, O3, and CO), only SO2 showed positive and significant associations with all-cause mortality. This association appeared to be robust against adjustment for other variables including fine particles and sulfate. The risk estimates for fine particles and sulfate were reduced when SO2 was jointly included in the models. These findings are not too surprising in that the high SO2 areas overlap the areas of high sulfate and fine particles in these U.S. data, and therefore “independent” mortality associations of these variables may be difficult to infer from statistical analyses alone. However, these findings suggest the impact of air pollution sources that emit SO2. There have been two updated analyses of the ACS cohorts. In the analysis by Pope et al. (2002), the follow-up data of approximately half a million subjects during 1982–1998 were linked to fine particles, sulfate, and gaseous pollutants data. Fine particles were associated with deaths due to all, cardiopulmonary, and lung cancer causes. SO2 was the only gaseous pollutant associated with mortality. This was consistent with Krewski et al.’s extended analysis of the original ACS data (1982–1988 follow-up period). The Pope et al. (2004) study analyzed more specific cardiovascular causes from the 1982–1998 follow-up data and found associations with PM2.5 and IHD, dysrhythmias, heart failure, and cardiac arrest, but SO2 and other pollutants were not examined. Another large U.S. cohort study, the Adventist Health Study of Smog (AHSMOG), followed a cohort of over 6000 nonsmoking Californian Seventh-Day Adventists since 1977. The AHSMOG study (Abbey et al., 1999) analyzed the 1977–1992 follow-up period. PM10 was associated with nonmalignant respiratory disease as well as lung cancer in males. SO2 was associated with lung cancer for both males and females. However, the number of cases for lung cancer in this study was relatively small (18 for males and 12 for females). Therefore, interpretation of these results requires caution. 25.2.2

Panel Studies

25.2.2.1 Morbidity The studies of Lawther and colleagues (Lawther, 1958; Lawther et al., 1970) showed associations between 24 h average concentrations of SO2 of 0.18 ppm (500 mg/m3), in association with BS of 250 mg/m3, and a worsening of health status among chronic bronchitis patients in London in the 1950s and 1960s. Schenker et al. (1983) reported that wheeze was more prevalent in nonsmoking women living downwind from mine-mouth coal-burning electric utility plants than among women in control communities with lesser exposures to the effluents. There was a significant association with SO2, the only effluent measured, and the highest exposure group had 24 h and annual average SO2 levels that were between 100 and 125% of the U.S. standards. For an acidic pollution episode in January 1985 in the Ruhr district in West Germany in which average concentrations of SO2 and suspended particles were 800 and 600 mg/m3, Wichmann et al. (1989) reported significant increases in deaths, hospital admissions, outpatient visits, and ambulance deliveries to hospitals in comparison to those in a less polluted control area. Baskurt et al. (1990) studied the hematogical and hemorheological effects of an air pollution episode in Ankara, Turkey, using SO2 as a surrogate of the pollution mixture. The blood measurements were made on 16 young male military students. The mean SO2 levels at a station proximal to the campus where the students lived were 188 and 201 mg/m3 during first and second blood measurements, respectively. During the period between the two measurements, the mean SO2 level was 292 (mg/m3). Significant erythropoiesis was

970

SULFUR OXIDES—SO2, H2SO4, NH4HSO4, AND (NH4)2SO4

indicated by increased erythrocyte counts and hemoglobin and hematocrit levels. Methemoglobin percentage was increased to 2.37  0.47% (mean  standard error) from 0.51  0.23%. Sulfhemoglobinemia was present in six subjects after the period of pollution, but it was not present in any student prior to this period. Significant increases in erythrocyte deformability indexes were observed after the period of pollution, that is, from 1.13  0.01 to 1.21  0.02, implying that erythrocytes were less flexible, which might impair tissue perfusion. Other short-term responses to PM–SO2 mixtures have been seen in children. Repeated measurements of lung function by Dockery et al. (1982) in schoolchildren in Steubenville, Ohio, in 1978–1980 showed statistically significant but physiologically small and apparently reversible declines of FVC and FEV0.75 levels to be associated with short-term increases in PM and SO2. The highest 24 h average PM and SO2 concentrations were 422 and 455 mg/m3, respectively. The small, reversible decrements persisted for up to 3–4 weeks after episodic exposures. A study of the association between episodic exposures to PM and SO2 and pulmonary function in children was conducted in the Netherlands by Dassen et al. (1986) producing results similar to those of Dockery et al. (1982). Pulmonary function values measured during an air pollution episode in which both 24 h average PM and SO2 levels reached 200–250 mg/m3 were significantly lower (3–5%) than baseline values measured 1–2 months earlier in the same group of Dutch schoolchildren. Lung function parameters that showed significant declines included FVC and FEV1 as well as measures of small airway function. Declines from baseline were observed 2 weeks after the episode in a different subset of children, but not after 3.5 weeks in a third subgroup. Studies of associations between chronic exposure to SO2 and PM and long-term changes in respiratory function in children have also been performed. Arossa et al. (1987) reported on the changes in baseline lung function between 1981 and 1983 in 1880 schoolchildren living in or near Turin, Italy. During that interval, annual average SO2 in central Turin decreased from 200 to 110 mg/m3, and total suspended particulate matter dropped from 150 to 100 mg/m3. During the same period, SO2 in a suburban area declined from 70 to 50 mg/m3. A group of 162 children from the suburban area served as controls. In the first survey, FEV1, FEF25–75, and MEF50 of children from urban areas were significantly lower, while in the second survey they were not significantly different from those of the controls. The slopes over time of FEV1, FEF25–75, and MEF50, adjusted for sex and anthropometric variables, were closely related to the decrease of pollutant concentrations, suggesting that the decrease of air pollution produced an improvement of baseline lung function. The effects of ambient air pollution on cardiac function in recent years has focused on PM, and several research groups in North America have exposed groups of volunteer subjects to concentrated ambient air particles and they have reported effects on heart rate and heart rate variability, as discussed in Chapter 10 on ambient particulate matter. The first controlled acute human exposure to SO2 involving cardiac function measurements was reported by Tunnicliffe et al. (2001). It involved electrocardiogram recordings made for 12 normal and 12 asthmatic young adults. Exposures were of 1 h duration, double blind, in random order, >2 weeks apart with clean air and 0.2 ppm of SO2. Spectral analyses of R–R intervals were performed. The SO2 exposures were associated with statistically significant increases in high-frequency (HF) and low-frequency (LF) power in the normal subjects and reductions of comparable magnitude in HF and LF in the asthmatic subjects. No pulmonary function changes or symptom frequency changes were observed in either group

HEALTH EFFECTS

971

of subjects. These results suggest that SO2 exposures at concentrations that are frequently encountered during air pollution episodes can influence the autonomic nervous system. This may help in elucidating the mechanisms involved in the induction of bronchoconstriction and the cardiovascular effects of ambient air pollution. Summary of Health Effects of SO2: In summary, the more quantitative epidemiological evidence from London suggests that effects may occur at SO2 levels at or above 0.19 ppm (500 mg/m3), 24 h average, in combination with elevated particle levels. Additional early evidence suggested the possibility of short-term, reversible declines in lung function at SO2 levels above 250–450 mg/m3 (0.10–0.18 ppm). The results of more recent multipollutant epidemiology studies suggest mortality and morbidity effects of SO2 at much lower concentrations. These effects could be due to SO2 alone, formation of sulfuric acid or other PM components or peak SO2 values well above the daily mean, but the relative roles of these factors cannot be determined at this time. We do know that the capacity of fog particles to “carry” untransformed SO2 is limited. Thus, it appears more likely that the role of SO2, in the presence of smoke, involved transformation products such as acidic fine particles. There is little evidence for associations between annual average levels of SO2 and chronic disease end points. To an even greater extent than the more acute response associations, they are likely to be artifacts of colinear associations between SO2 and fine particles from combustion processes. 25.2.3

Health Effects of Acidic Aerosols

25.2.3.1 Deposition, Growth, and Neutralization Within the Respiratory Tract The deposition pattern within the respiratory tract depends on the size distribution of the droplets. Acidic ambient aerosol typically has a mass median aerodynamic diameter (MMAD) of 0.3–0.6 mm, whereas industrial aerosols can have an MMAD as large as 14 mm (Williams, 1970). With hygroscopic growth in the airways, submicrometer-sized droplets can increase in diameter by a factor of 2–4, and still remain within the fine particle range that deposits preferentially in the distal lung airways and airspaces. As droplet sizes increase above about 3 mm MMAD, deposition efficiency within the airways increases, with more of the deposition taking place within the upper respiratory tract, trachea, and larger bronchi (Lippmann et al., 1980). For larger droplets, the residence time in the airways is too short for a large growth factor. Some neutralization of inhaled acidic droplets can occur before deposition, due to the normal excretion of endogenous ammonia into the airways (Larson et al., 1977). Once deposited, free Hþ reacts with components of the mucus of the respiratory tract, changing its viscosity (Larson et al., 1977). Unreacted Hþ diffuses into surrounding tissues. The capacity of the mucus to react with Hþ depends on the Hþ absorption capacity, which is reduced in acidic saturated mucus as found in certain disease states, for example, asthma (Holma, 1985). 25.2.4 25.2.4.1

Effects on Experimental Animals Short-Term Exposures

Respiratory Mechanical Function Alterations of pulmonary function, particularly increases in pulmonary flow resistance, occur after acute exposure. Reports of the irritant potency of various sulfate species are variable, due in part to differences in animal species and strains, and also due to differences in particle sizes, pH, composition, and solubility

972

SULFUR OXIDES—SO2, H2SO4, NH4HSO4, AND (NH4)2SO4

(U.S. EPA, 1986). H2SO4 is more irritating than any of the sulfate salts in terms of increasing airway resistance. For short-term (1 h) exposures, the lowest concentration shown to increase airway resistance was 100 mg/m3 (in guinea pigs). The irritant potency of H2SO4 depends in part on droplet size, with smaller droplets having more effect (Amdur et al., 1978). Animal inhalation studies by Amdur et al. (1986) are of interest to this discussion because they demonstrate that effects produced by single exposures at very low acid concentrations can be persistent. Guinea pigs were exposed by inhalation for 3 h to the diluted effluent from a furnace that simulates a model coal combuster. Pulverized coal yields large particle mineral ash particles and an ultrafine (<0.1 mm) condensation aerosol. The core of the ultrafine particles consists of oxides of Fe, Ca, and Mg, covered by a layer containing Na, As, Sb, and Zn. Zn generally has the highest concentration on the surface of the solidified particle. As the particles cool further, there is surface formation and/or condensation of a layer of H2SO4. In a single 3 h exposure, it produced significant decrements in lung diffusing capacity (DLco). At 1 h after exposure, there was an increase in lung permeability. At 12 h after exposure, there was distention of perivascular and peribronchial connective tissues, and an increase in lung weight. The alveolar interstitium also appeared distended. At 72 h after exposure, total lung capacity (TLC), vital capacity (VC), and functional residual capacity (FRC) had returned to baseline levels, but DLco was still significantly depressed. Based upon prior experience with pure SO2 and H2SO4 exposures in the guinea pig model, Amdur et al. (1986) concluded that the humid furnace effluent effect was an acid aerosol effect because of its persistence. In subsequent tests, 3 h exposures to the acid-coated ZnO aerosol were given on five successive days. Significant depressions of DLco were produced on the second and subsequent days for 30 mg/m3 of H2SO4, while 20 mg/m3 produced significant depressions on the fourth and fifth days. The most sensitive response was a change in airway reactivity, where a significant response was produced by a single 1 h exposure to 20 mg/m3 H2SO4 as a surface coating on the ZnO (Amdur, 1989, 1989b). The persistent changes in function and morphological changes following exposure to very low levels of acidic aerosol suggest that repetitive exposures could lead to chronic lung disease. However, the implications of these changes in guinea pigs to human disease remain highly speculative. Particle Clearance Function Donkeys exposed by inhalation for 1 h to 0.3–0.6 mm H2SO4 at concentrations ranging from 100 to 1000 mg/m3 exhibited slowed bronchial mucociliary clearance function at concentrations 200 mg/m3, whereas, as shown in Fig. 25.3, rabbits undergoing similar exposures exhibited an acceleration of clearance at concentrations between 100 and 300 mg/m3, and a progressive slowing of clearance at 500 mg/m3 (Schlesinger, 1985). Schlesinger (1989) examined the relative roles of concentration (C) and daily exposure (T) to H2SO4- induced changes in particle clearance from the gas exchange region of rabbit lungs. Exposures were for 1–4 h/day for 14 days at concentrations ranging from 250 to 1000 mg/m3. In a follow-up study, Schlesinger (1990) extended the concentrations downward to 50 mg/m3. The results are summarized in Fig. 25.4. The acceleration in clearance produced by 4 h at 50 mg/m3 is essentially the same as that produced by 2 h at 100 mg/m3 and 1 h at 250 mg/ m3, indicating that cumulative exposure, rather than concentration, governs the response, at least within the ranges of concentration and time evaluated. The results are similar to those for mucociliary clearance in the sense that relatively low levels of exposure produce an acceleration of clearance, but clearance retardation occurs at higher levels of exposure.

HEALTH EFFECTS

+300

*

Rabbit +200 ∆ MRT (min)

973

*

+100

95% CL

0

*

*

-100

∆ T50% (min)

-200

+100

Human

*

95% CL

0

*

-100 60

100

200

400 1000 (H2SO4) µg/m3

2000

4000

FIGURE 25.3 Exposure-dependent changes in characteristic bronchial mucociliary clearance times in rabbits (mean residence time) and humans (clearance half-time) from 1 h exposures to submicrometer-sized H2SO4 aerosols. Each point represents the average for the group, with the vertical bars indicating 1 SE. The horizontal bands represent the mean 1 SE of the measurements for the sham-exposure controls. Asterisks indicate a significant change (p < 0.05) at an individual concentration (paired t-test, one-tailed) (reproduced from Lippmann, 1986).

Cellular Function Schlesinger et al. (1990) examined the comparative effects of exposure to the two main ambient acidic sulfates, H2SO4 and NH4HSO4, using the phagocytic activity of alveolar macrophages as the end point. Rabbits were exposed to 250–2000 mg/m3 H2SO4 (as SO42 ) and 500–4000 mg/m3 NH4HSO4 (as SO42 ) for 1 h/day for 5 days; +200 0.25 – 1.0 mg/m3 0.1 mg/m3 0.05 mg/m3

*

%∆ t1/2

+100

0

* * -100 0.01

*

1.0 10 0.1 C × T [(mg/m3) • h] * Exposures were for 14 days

FIGURE 25.4 Mean percentage change in half-time (%Dtm) of a clearance of tracer particles from the respiratory region of the lungs following exposure to H2SO4 as a function of the product of exposure concentration (C) and exposure time (T) (n ¼ 5 rabbits for each point). Positive changes indicate slowing of clearance; negative changes indicate speeding. Shading represents 95% confidence interval for (%Dt1/2) ¼ 0 (based on sham-control exposures). Data for 0.25–1.0 mg/m3 are from Schlesinger (1989); *p < 0.05 compared to control.

974

SULFUR OXIDES—SO2, H2SO4, NH4HSO4, AND (NH4)2SO4

bronchopulmonary lavage was then performed for recovery of free lung cells. Phagocytosis, measured by uptake of opsonized latex spheres in vitro, was altered by exposure to H2SO4 at concentrations 500 mg/m3 and to NH4HSO4 at 2000 mg/m3. Assessment of results in terms of the calculated hydrogen ion concentration in the exposure atmosphere showed that identical levels of Hþ produced different degrees of response depending on whether exposure was to H2SO4 or NH4HSO4. On the contrary, macrophages incubated in acidic environments in vitro responded similarly, regardless of whether H2SO4 or NH4HSO4 was used to adjust the pH. Thus, the response may relate more to the local pH change in the vicinity of the depositing droplet than to the total Hþ delivered. Pollutant Interactions Osebold et al. (1980) exposed antigenically sensitized mice to 500 ppb O3 for 3 days, with and without concurrent exposure 1 mg/m3 of submicrometer H2SO4 droplets. There was an increase in atopic reactivity that was greater than that for each pollutant alone. Kleinman et al. (1989) reported that lesions in the gas exchange region of the lung of rats exposed to O3 were greater in size in rats exposed to mixtures containing H2SO4 or NO2 as well as O3. Last (1989) reported significant increases in lung protein content in rats exposed for 9 days to 200 ppb O3 plus 20 mg/m3 H2SO4 over those in rats exposed to 200 ppb O3 alone, as well as a trend toward increased protein in rats exposed to 200 ppb O3 and 5 mg/m3 H2SO4. 25.2.5

Subchronic Exposures

25.2.5.1 Particle Clearance Function Donkeys exposed for 1 h/day (5 days per week) for 6 months to an aerosol (0.3–0.6 mm) of H2SO4 at a concentration of 100 mg/m3 developed highly variable clearance rates and a persistent shift from baseline rate of bronchial mucociliary clearance during the exposures and for 3 months after the last exposure. Two animals had much slower clearance than their baseline during the 3 months of followup, but two had faster than baseline rates (Schlesinger et al., 1979). Rabbits exposed for 1 h/ day (5 days per week) for 4 weeks to 0.3 mm H2SO4 at 250 mg/m3 developed variable mucociliary clearance rates during the exposure period, and their clearance during a 2-week period following the exposures was substantially faster than their baseline rates (Schlesinger et al., 1983). For a group of rabbits undergoing daily exposures via the nose at 250 mg/m3 for 1 year, Fig. 25.5 shows that bronchial mucociliary clearance was consistently slowed after

∆ Retention (%)

400 300

Acid exposed Sham exposed

Acid stopped

200 100 0 -100 0

20

40 Time (week)

60

FIGURE 25.5 Mean change in percentage retention of tracer particles (SD) during intermittent exposure to H2SO4 in acid- and sham-exposed animals from that established in pre-exposure control tests (reproduced from Lippmann et al., 1987).

HEALTH EFFECTS

975

the first few weeks and became even slower during a 3-month period following the end of acid exposures (Gearhart and Schlesinger, 1988). During the course of a 1-year series of 1 h/day, 5 days per week nasal exposures to submicrometer H2SO4 at 250 mg/m3, groups of rabbits were exposed on three occasions to 85 Sr-tagged latex aerosols for determination of the rates of clearance from the nonciliated alveolar region (Schlesinger and Gearhart, 1986). The latex aerosols were inhaled on days 1, 57, and 240 following the start of the H2SO4 exposures, and particle retention was followed for 14 days after each latex administration. As compared to baseline rates of clearance in control animals, early alveolar clearance was accelerated to a similar degree in all three tests performed during the chronic H2SO4 exposures. 25.2.5.2 Airway Hyperresponsiveness The effects of daily 1 h exposures to 250 mg/m3 of H2SO4 on bronchial responsiveness of rabbits was assessed at the end of 4, 8, and 12 months by in vitro administration of double doses of acetylcholine and measurement of pulmonary resistance (RL), as shown in Figure 25.6 (Gearhart and Schlesinger, 1986). Dynamic compliance (Cdyn) and respiratory rate (f) were also measured following agonist challenge. Those animals exposed for 4 months showed increased sensitivity to acetylcholine (i.e., the dose required to produce a 150% increase in RL), and there was an increase in reactivity (i.e., the slope of dose versus change in RL) by 8 months, with a leveling off of the response after this time. No changes in Cdyn or f were noted at any time. Thus, repeated exposures to H2SO4 resulted in the production of hyperresponsive airways in previously healthy animals. This has implications for the role of nonspecific irritants in the pathogenesis of airway disease. 25.2.5.3 Histology In the study of Schlesinger et al. in which rabbits were exposed to 250 mg/m3 for 4 weeks and sacrificed 2 weeks later, histological examination showed increased numbers of secretory cells in distal airways and thickened epithelium in airways extending from midsized bronchi to terminal bronchioles (Schlesinger et al., 1983). There were no corresponding changes in the trachea or other large airways. In the follow-up study, in which rabbits received daily exposures for 1 year via the nose at 250 mg/m3, the secretory cell density was elevated in some lung airways at 4 months and in all lung airways at 8 months (Gearhart and Schlesinger, 1986, 1988; Schlesinger and Gearhart, 1986). At

Pulmonary resistance (RL) (%)

400

300

200

100

0 0

Control Acid –– 4 months Acid –– 8 months Acid –– 12 months

1.3

10 20 2.5 5 Acetylchlorine (µg/kg/min)

40

FIGURE 25.6 Effect of bronchoprovocation challenges on pulmonary resistance (RL). The abscissa is expressed in terms of doubling doses of acetylcholine. Data are expressed as the group mean (SD) percentage of baseline RL at each dose (n = 12) for control; n = 4 for each acid group. (Reproduced from Gearhart and Schlesinger, 1986).

976

SULFUR OXIDES—SO2, H2SO4, NH4HSO4, AND (NH4)2SO4

12 months, the increased density remained in small and midsized airways, but not large airways. Partial recovery was observed at 3 months after the last exposure. In a study in which dogs were exposed daily for 5 years to 1100 mg/m3 SO2 plus 90 mg/m3 H2SO4 and then allowed to remain in unpolluted air for 2 years, there were small changes in pulmonary functions during the exposures, which continued following the termination of exposure. Morphometric lung measurements made at the end of a 2-year postexposure period showed changes analogous to an incipient stage of human centrilobular emphysema (Stara et al., 1980). 25.2.6 25.2.6.1

Effects on Humans Acute Effects: Controlled Exposures

Respiratory Mechanical Function H2SO4 and other sulfates have been found to affect both sensory and respiratory function in humans. Respiratory effects from exposure to H2SO4 (350–500 mg/m3) have been reported to include increased respiratory rates and tidal volumes (Amdur et al., 1952; Ericsson and Camner, 1983). However, other studies of pulmonary function in nonsensitive healthy adult subjects indicated little effect on pulmonary mechanical function when subjects were exposed to submicrometer H2SO4 at 10–1000 mg/m3 for 10–120 min (Fig. 25.7). In a study, the bronchoconstrictive action of carbachol was potentiated by 0.8 mm H2SO4, and by other sulfate aerosols, more or less in relation to their acidity (Utell et al., 1984). Asthmatics are substantially more sensitive in terms of changes in pulmonary mechanics than healthy people, and vigorous exercise potentiates the effects at a given concentration. The lowest demonstrated effect level was 68 mg/m3 of 0.6 mm H2SO4 via mouthpiece inhalation in exercising adolescent asthmatics (Koenig et al., 1989) with somewhat greater responses at 100 mg/m3 (Koenig et al., 1983a). The effects disappeared within about 15 min. In adult asthmatics undergoing exposure to 0.8 mm H2SO4 for 2 h, the lowest observed effect level was 75 mg/m3 (Bauer et al., 1988). Spengler et al. (1989) concluded that these results are consistent when the exposure metric is total amount of H2SO4 inhaled rather than the concentration. By contrast, Avol et al. (1990) found no significant functional responses to H2SO4. They exposed 32 asthmatic volunteers, 8–16 years of age, in a chamber to clean air and to sulfuric acid aerosol at a “low” concentration (46  11 mg/m3; mean  SD) and at a “high” concentration (127  21 mg/m3). Acid aerosols had mass median aerodynamic diameters near 0.5 mm with geometric standard deviations near 1.9. Temperature was 21 C, and relative humidity was near 50%. Subjects were exposed with unencumbered oronasal breathing for 30 min at rest plus 10 min at moderate exercise (ventilation rate 20 L/min m2 of body surface). A subgroup (21 subjects) was exposed similarly to clean air and to “high” acid (134  20 mg/m3) with 100% oral breathing. Increased symptoms and bronchoconstriction were found after exercise under all exposure conditions. For the group, symptom and lung function responses were not statistically different during control and during acid exposures with unencumbered breathing or with oral breathing. Aris et al. (1990) exposed nonsmoking adult volunteers in chambers for 1 h to fogs containing hydroxymethanesulfonate (HMSA), the bisulfate adduct of formaldehyde, and a common constituent of California acid fogs. The droplet size was 7 mm, the HMSA concentration was 260 mg/m3, and the H2SO4 content of the aerosol was 1.1 mg/m3. A control exposure was to H2SO4 only. Both acid fogs produced slight increases in respiratory symptoms, but no changes in airway resistance. Thus, HMSA did not produce a specific

HEALTH EFFECTS

(a)

977

Tc

100

7.6 µm Fe2O3

80

Tracheobronchial percent retained

60

Control 967 µg/m3

40 20

104 µg/m3

0 (b) 100

4.2 µm Fe2O3

80 60

108 µg/m3 983 µg/m3

40 20 0 0

Control

1

2

3 Hour

4

5

6

FIGURE 25.7 Tracheobronchial retention of 99mTc-tagged Fe2O3 microspheres in healthy adult volunteers as a function of time after a 1 min tagged aerosol inhalation for particles with aerodynamic diameter of (a) 7.6 mm or (b) 4.2 mm. Submicrometer-sized droplets of H2SO4 are inhaled via nasal mask during three 20 min intervals as indicated by cross-hatched boxes. The solid line indicates retention for sham exposure, the long dash–dot line for about 100 mg/m3 H2SO4 and the short dash line for about 1000 mg/m3 H2SO4 (reproduced from Leikauf et al., 1984).

bronchoconstrictor effect at a concentration about three times greater than the highest ambient measurements. The effects of acid fog droplets on respiratory function and symptoms were studied by Avol et al. (1988). They exposed both normal and mild asthmatic adult volunteers for 60 min to 8 mm MMAD fog droplets containing 0, 150, and 680 mg/m3 of H2SO4, with alternating 10 min periods of rest and heavy exercise. Both normals and asthmatics reported more symptoms with increasing concentration, and the asthmatics showed an increase in airway resistance at the higher acid concentration. There were no significant differences in either forced expiratory function or airway reactivity to methacholine (MC) between the sham and acid exposures. Linn et al. (1989) exposed both healthy and asthmatic volunteers for 1 h with intermittent exercise to H2SO4 at 2000 mg/m3, with droplet sizes of 1, 10, and 20 mm. Healthy subjects had no significant changes in lung function or bronchial reactivity to methacholine, but did show irritant symptoms with the 10 and 20 mm aerosols. By contrast, the asthmatics had significant decrease in function and increase in airway resistance and also in symptoms for all three droplet sizes.

978

SULFUR OXIDES—SO2, H2SO4, NH4HSO4, AND (NH4)2SO4

Raizenne et al. (1996) examined the effects of exposure to acidic airborne particles on respiratory function among 8–12 year-old children living in 22 communities in the United States and Canada. Air quality and meteorology were measured in each community for the year preceding the pulmonary function tests. Forced vital capacity (FVC) and forced expiratory volume (FEV) in 1 s (FEV1.0) of 10,251 white children, adjusted for age, sex, height, weight, and sex–height interaction, were examined. A 52 nmol/m3 difference in annual mean particle strong acidity was associated with a 3.5% (95% CI, 2.0–4.9) decrement in adjusted FVC and a 3.1% (95% CI, 1.6–4.6) decrement in adjusted FEV1.0. The FVC decrement was larger, although not significantly different, for children who were lifelong residents of their communities (4.1%, 95% CI, 2.5–5.8). The relative odds for low lung function (i.e., measured FVC less than or equal to 85% of predicted) were 2.5 (95% CI, 1.8–3.6) across the range of particle’s strong acidity exposures. These data suggested that long-term exposure to ambient particle strong acidity may have a deleterious effect on lung growth, development, and function. Particle Clearance Function In healthy nonsmoking adult volunteers exposed to 0.5 mm H2SO4 at rest at 100 mg/m3 for 1 h, there was an acceleration of bronchial mucociliary clearance of tracer particles (7.6 mm), which deposited primarily in the larger bronchial airways, and a slowing of clearance when the exposure was raised to 1000 mg/m3 (Figs 29.3 and 29.7) (Leikauf et al., 1981). For tracer particles (4.2 mm), which deposited primarily in midsized–small conducting airways, there was a small but significant slowing of clearance at 100 mg/m3 H2SO4 and a greater slowing at 1000 mg/m3 (Leikauf et al., 1984). These changes are consistent with the greater deposition of acid in midsized–smaller airways. Exposures to 100 mg/m3 for 2 h produced slower clearance than the same exposure for 1 h, indicating a cumulative relationship to dose (Spektor et al., 1989). The results of these studies were used by Yu et al. (1986) to construct a model for the effects of surface deposition of acidic droplets on mucus transport velocity along the tracheobronchial airways. Based on this model, mucus velocities are increased when less than about 10 7 g/cm2 of H2SO4 is deposited, while clearance is retarded when the acid deposition exceeds this limit. The effects of a 1 h inhalation of submicrometer H2SO4 aerosols via nasal mask on tracheobronchial mucociliary particle clearance and respiratory mechanics were studied by Spektor et al. (1985) in subjects with histories of asthma. A brief inhalation of tagged aerosol preceded the 1 h H2SO4 or a sham exposure. Respiratory function was measured 15 min before and 3 h after the H2SO4 or sham exposure. After exposure to 1000 mg/m3 of H2SO4, the six subjects not on routine medication exhibited a transient slowing of mucociliary clearance and also decrements in specific airway conductance (SGaw), forced expiratory volume in 1 s (FEV1), midmaximal expiratory flow rate (MMEF), and flow rate at 25% of total lung capacity (V25) (p < 0.05) in both sets of measurements. The four asthmatics on daily medication exhibited stepwise mucociliary clearance that was too variable to allow detection of any H2SO4 effect on clearance. Mucociliary clearance rates in both groups in the sham exposure tests were significantly slower than those of healthy nonsmokers studied previously by Leikauf et al. (1984) using the same protocols. The extent of mucociliary clearance slowing following the 1000 mg/m3 exposure in the nonmedicated subjects was similar to that in the healthy nonsmokers. This similar change, from a reduced baseline rate of clearance, together with the significant change in respiratory function, indicates that asymptomatic asthmatics may respond to H2SO4 exposures with functional changes of greater potential health significance than do healthy nonsmokers.

HEALTH EFFECTS

979

Effects of Ambient Air Exposures in Population Studies: There are numerous studies of associations between SO42 and various health effect indices in polluted communities, and a more limited number of studies with measurements of Hþ. The earliest direct association between measured acidity and human health was Gorham’s highly significant (p < 0.01) correlation between mortality rates for bronchitis in 53 UK metropolitan areas in the period 1950–1954 and the pH of winter precipitation in these areas (Gorham, 1958). There was also a correlation with SO42 at the 5% level of significance. When multiple regression analyses were performed, pH remained significant at the 1% levels, but SO42 lost its statistical significance. An association based on some limited, but direct measurements was reported by Kitagawa (1984) who identified sulfuric acid as the probable causal agent for approximately 600 cases of acute respiratory disease in the Yokkaichi area in central Japan between 1960 and 1969. The patients’ residences were concentrated within 5 km of a titanium dioxide plant with a 14 m stack that emitted 100,000–300,000 kg/month of H2SO4 in the period 1961–1967. The average concentration of SO3 in February 1965 in Isozu, a village 1–2 km from the plant, was 130 mg/m3, equivalent to 159 mg/m3 of H2SO4. Kitagawa estimated that the peak concentrations might be up to 100 times as high with a north wind. Electrostatic precipitators were installed to control aerosol emissions in 1967, and after 1968 the number of newly found patients with “allergic asthmatic bronchitis” or “Yokkaichi asthma” gradually decreased. Although Kitagawa’s quantitative estimates of exposure to H2SO4 and the criteria used to describe cases of respiratory disease may differ from current methods, the unique aspect of this report is the identification of H2SO4 as the likely causal agent for an excess in morbidity. In an independent analysis of mortality from asthma and chronic bronchitis associated with changes in sulfur oxide air pollution in Yokkaichi from 1963 to 1983, Imai et al., 1986 correlated mortality with sulfation index (lead peroxide candle measurements), and focused on reductions in SOx emissions from a petroleum refinery in the harbor area in 1972. Thus, it is not clear from their analysis what the SO2 or H2SO4 exposures to the population from these emissions were. In any case, mortality rates for bronchial asthma were significantly elevated in Yokkaichi in the period 1967–1970, and the mortality rates due to chronic bronchitis were significantly elevated for the periods 1967–1970 and 1971–1974. There was a greater lag between the reduction in SOx pollution and reduction in mortality rate for chronic bronchitis than for bronchial asthma. Although sulfuric acid aerosol was believed, by some, to be a likely causal factor for excess mortality and morbidity during and following the December 1952 smog episode in London, the only air quality available for that time was for BS and SO2. During the late 1950s, a monitoring method was developed by Commins and Waller (1963) to measure H2SO4 in urban air, and they used it to make daily measurements of H2SO4 at St. Bartholomew’s Hospital in Central London during the December 1962 episode. As shown in Chapter 10 (Fig. 13.6), the airborne H2SO4 rose rapidly during the 1962 episode, with a relative increase greater than that for BS or SO2. Using the method of Commins and Waller (1963), daily measurements of aerosol strong acid (Hþ) were made at a Central London site (St. Bartholomew’s Medical School) between 1965 and 1972. The December 1962 London fog episode was the last to produce a clearly evident increase in the number of daily deaths, albeit a much smaller one than in December 1952. The UK Clean Air Act of 1954 had led to the mandated use of smokeless fuels, and annual mean smoke levels had declined by 1962 to about one-half of the 1958 level. The annual average SO2 concentrations had not declined by 1962, but dropped off markedly

980

SULFUR OXIDES—SO2, H2SO4, NH4HSO4, AND (NH4)2SO4

thereafter, along with a further marked decline in BS levels. For the period between 1964 and 1972, the measured levels of H2SO4 followed a similar pattern of decline. The daily concentration data have been correlated with concurrent daily records of mortality in several studies. Based on an initial time-series analysis of the winter data (Thurston et al., 1989), Hþ appeared to be more strongly associated with total daily mortality than either BS or SO2. However, a more detailed analysis of the full-year data set by Ito et al. (1993), involving statistical “prewhitening,” did not indicate that Hþ had a greater degree of association with daily mortality than BS or SO2. In the Ito et al. (1993) analysis, temperature had the greatest influence in all seasons, and all three of the pollution variables (same day and lagged one or two days) were significantly associated with daily mortality. However, there were limitations imposed on the analysis in terms of the limitation of the Hþ data to only one monitoring site, the limited precision of the measurement of Hþ, and the selection of filters for controlling the confounding long-wave influences. In a further exploratory analysis of the Central London data set, Lippmann and Ito (1995) developed an alternate approach for separating the effects of season and temperature on daily mortality from those of pollution. They analyzed the data for each season separately, and within each season, they restricted analyses to those days in which ambient temperature had little, if any, influence on mortality. Regressions were performed for BS, SO2, and aerosol Hþ. For the winter period (November–February), mortality was most closely correlated with Hþ, while for the rest of the year, it was most closely correlated with SO2. In all seasons, the correlation was poorest for BS. The results for winter and summer are illustrated in Figs 25.8 and 25.9. Ostro et al. (1989, 1991) correlated data on aerosol Hþ, SO42 , NO32 , and FP, as well as gaseous SO2 and CO, with daily symptom, medication usage, and other variables for a panel of about 200 adults with moderate to severe asthma in Denver, CO, between November 1987 and March 1988. The Hþ concentrations ranged from 2 to 41 neq/m3 (0.04–0.84 mg/m3 of H2SO4 equivalent), and were significantly related to both the proportion of the survey respondents reporting a moderate or worse overall asthma condition and the proportion reporting a moderate or worse cough. Of all the pollutants considered, Hþ displayed the strongest association with asthma and cough. The magnitudes of the effects were compared by computing elasticities or the percent change in the health effect due to a given percent change in the pollutant. Using asthma as an example, the results indicate elasticities with respect to SO42 , FP, and Hþ of 0.060, 0.055, and 0.096, respectively (Ostro et al., 1989). This indicates that a 10% change in the concentrations of Hþ could increase the proportion reporting a moderate or worse asthma condition by 0.96%. In their follow-up report on this study, Ostro et al. (1991) examined evidence for lagged effects and concluded that contemporaneous measures of Hþ concentration provided the best associations with asthma status and that meteorological variables were not associated with the health effects reported. They also examined the effects of exposure to Hþ, adjusting for time spent outdoors, level of activity, and penetration of acid aerosol indoors. Based on the adjusted exposures, the effect of Hþ on cough increased 43%, suggesting that dose– response estimates that do not incorporate behavioral factors affecting actual Hþ exposures may substantially underestimate the impact of the pollution. Some morbidity studies have utilized daily measurements of Hþ concentrations. In the six-city study (Speizer, 1989; Damokosh et al., 1993), aerosol Hþ had a closer association with parent-reported bronchitis symptoms in children than any of the other PM or gas-phase components. This was also true in the Harvard-Health Canada study in 22 North American towns (Dockery et al., 1996), and in this study aerosol Hþ was also the most closely related

HEALTH EFFECTS

981

FIGURE 25.8 Mortality in London: 1965–1972, winter days with temperatures between 5 and 10 C (from Lippmann and Ito, 1995).

pollutant to baseline lung function (Raizenne et al., 1996). Although pulmonary function differences in the six-city study were not statistically significant, the direction and magnitude of the differences were consistent with the results reported by Raizenne et al. (1996). In a supplemental acute–response study in Uniontown, PA, one of the 22 communities in the Harvard-Health Canada study, Neas et al. (1995) had a stratified sample of 83 children and reported twice daily peak expiratory flow rate (PEFR) measurements on 3582 child-days during the summer of 1990. Upon arising and before retiring, each child recorded PEFR and the presence of cold, cough, or wheeze symptoms. The session-specific average deviation was then calculated across all the children. A 12 h Hþ exposure to a 125 nmol/m3 increment was associated with a 2.5 L/min deviation in the group mean PEFR (95% CI, 4.2 to 0.8) and with increased cough incidence (odds ratio (OR) ¼ 1.6, 95% CI, 1.1–2.4). Studnicka et al. (1995) studied three consecutive panels of children participating in a summer camp in the Austrian Alps. For 47, 45, and 41 subjects, daily FEV1, FVC, and peak PEFR were recorded; 15, 11, and 5% of participants, respectively, reported current asthma medication. Mean levels of ambient pollutants were approximately 15% higher for the first panel than for the other two panels, but the Hþ component was twice as high for panel1. The maximum Hþ exposure during panel 1 was 84 nmol/m3 (4 mg/m3 H2SO4 equivalent).

982

SULFUR OXIDES—SO2, H2SO4, NH4HSO4, AND (NH4)2SO4

FIGURE 25.9 Mortality in London: 1965–1972, summer days with temperatures between 13 and 18 C (from Lippmann and Ito, 1995).

For FEV1 in panel 1, a significant decrease of 0.99 mL per nmol/m3 Hþ (p ¼ 0.01) was observed. For panel 2, the FEV1/Hþ coefficient was found to be similar ( 0.74 mL per nmol/m3 Hþ; p ¼ 0.28), while for panel 3 it was in the opposite direction (0.10 mL per nmol/m3 Hþ; p ¼ 0.83). The decrease in FEV1 observed in panel 1 was more pronounced when the mean exposure during the previous 4 days was considered ( 2.99 mL FEV1 per nmol/m3 Hþ; p ¼ 0.004). Thurston et al. (1997) studied 52, 58, and 56 children (aged 7–13) attending a summer “asthma camp” in the Connecticut River Valley during the last week of June in 1991, 1992, and 1993, respectively. Most of the subjects had moderate to severe asthma. Daily records were kept of the environmental conditions, as well as of subject medication use, lung function, and respiratory symptoms. Hþ and SO42 were found to be significantly and consistently correlated with acute asthma exacerbations and chest symptoms. Lung function decrements were consistently associated with O3, but not with SO42 and Hþ. The prospective cohort mortality study of Dockery et al. (1993) reported that Hþ correlated less well with mortality than PM2.5 or SO42 , but they only had 9–12 months of Hþ data in each city, compared to 14–16 years of data on the other pollutant variables, and many of the daily concentrations of Hþ were below the detection limit. A similar limitation was present in the analysis of Dockery et al. (1992) of air pollution and daily mortality rates.

HEALTH EFFECTS

983

25.2.7 Implications of the Effects of Acidic Aerosols on Respiratory Function in the Exacerbation of Asthma and Chronic Bronchitis The studies of Utell et al. (1982, 1984 demonstrate that brief exposures to acidic aerosols reduce airway conductance in healthy humans, and that asthmatic subjects are more sensitive than healthy individuals. The lowest concentration that produced a significant response in the group as a whole was 450 mg/m3. Koenig et al. (1983a) reported a 40% increase in total airway resistance in a group of exercising asthmatic adolescents when they were exposed to 100 mg/m3 of H2SO4, and lesser but still significant effects at 68 mg/m3 (Koenig et al., 1989). The responses were similar to those reported by Koenig et al. (1981, 1983b for the same protocols and kinds of subjects for exposure to 0.5 ppm of SO2 (1300 mg/m3). Thus, when SO2 is oxidized and hydrolyzed, the resulting H2SO4 is 10–20 times more potent. While the average increases in airway resistance were small, the populations studied were small, and some subjects had much greater responses than the average. Also, the populations were carefully selected and did not include the more unstable and potentially more reactive asthmatics in the population. Moderate exercise appears to enhance the response by increasing the dose of irritant delivered to epithelial surfaces. With increasing exercise, more pollutant is inhaled. The greater inspiratory flow rates also act to increase the percentage of the highly soluble SO2 vapor that can penetrate beyond the upper airways into those bronchial airways where reflex responses are most likely to be initiated. The greater flow rate also produces a thinner boundary layer around the airway bifurcations, enhancing “hot spots” of deposition of particles and vapors from the airstream. Thus, exercise results in increased deposition in this region for the submicrometer-sized H2SO4 droplets that would have minimal deposition in such airways at lower flow rates. The irritant dose delivered to the larger bronchial airways is greater in patients with asthma and bronchitis than in healthy individuals because the former groups have airways with smaller diameters. This may account for some, or perhaps all, of the greater responsiveness of asthmatics to inhaled irritants. They may also have a greater responsiveness at the sites of deposition to the delivered dose, but this has not been clearly established in in vivo tests. In any case, an irritant-induced narrowing of the conducting airways of the lung can increase the surface deposition of subsequently inhaled irritant, resulting in further airway constriction. The subjective responses to inhaled acid aerosols may include a feeling of chest tightness, and the work of breathing is increased. For individuals with chronic respiratory disease, any increment of work in breathing may be considered an adverse effect. For asthmatic individuals, the major concern is the induction of bronchospasm. The few clinical laboratory studies on carefully selected asthmatics cannot be expected to generate data on the exact conditions that provoke bronchospasm and acute respiratory insufficiency. It would be highly desirable to have an animal model for bronchial asthma so that this important issue could be systematically studied.

25.2.8 Implications of the Effects of Acidic Aerosols on Mucociliary Clearance in the Pathogenesis of Chronic Bronchitis Schlesinger et al. (1983) studied the effects of concentration and duration of exposure to submicrometer-sized H2SO4 aerosols on tracheobronchial and alveolar rates of particle clearance in rabbits and found that the effects increased with duration and concentration.

984

SULFUR OXIDES—SO2, H2SO4, NH4HSO4, AND (NH4)2SO4

Spektor et al. (1989) reported a similar finding for tracheobronchial particle clearance in humans. The altered clearance rates during and after the exposure period may be an adaptive response of the mucociliary system to acid exposures. On the contrary, they may be early stages in the progression toward more serious dysfunctions, for example, those found in chronic bronchitis, which may result from continued irritant exposures. A mechanistic basis for the linkage between chronic exposure to H2SO4 and the pathogenesis of chronic bronchitis lies in the series of studies involving chronic exposures of animals and persistent histological alterations in lung structure. As noted by Lippmann et al. (1987), these structural changes in the rabbit model have correlates in terms of clearance function changes. These, in turn, are indicative of changes in mucus secretion leading to mucus stasis, a hallmark of bronchitic disease. The animal studies can be related to human responses in two ways. One is the concordance in functional and morphometric responses of animals to H2SO4 and cigarette smoke, a known causal factor for human chronic bronchitis. The other is that humans, rabbits, and donkeys all have essentially the same transient mucociliary clearance function responses to single 1 h exposures to H2SO4. The fact that daily 1 h H2SO4 exposures in rabbits and donkeys produce persistent changes in clearance function, making it very likely that humans would also show these effects if similarly exposed. Furthermore, a comparison of the human and rabbit responses to single exposures indicates that humans respond at lower concentrations than do rabbits (Schlesinger, 1986). The effects of H2SO4 on the airways are very likely to be cumulative during each exposure day, at least in part. Thus, the daily 1 h exposures at 250 mg/m3 in the rabbits may be equivalent to <50 mg/m3 for 7–8 h a day and to a still lower concentration for equivalent effects in humans. On the contrary, the effects produced by the 1-year series of exposures in the rabbits were less severe than the condition corresponding to a clinical diagnosis of chronic bronchitis in humans. Unfortunately, there are few data concerning the response of the human mucociliary clearance system under prolonged insult by potentially harmful pollutants such as H2SO4. The most direct evidence for an association between chronic bronchitis and exposure to H2SO4 comes from occupational exposures, but these were at high levels. Williams (1970) observed an excess incidence of chronic bronchitis in workers occupationally exposed to H2SO4 levels above 1 mg/m3 (probable diameter ¼ 14 mm); however, the excess was actually in increased incidence of episodes in affected workers rather than an increase in the number of workers affected. Although available evidence suggests that exposure to H2SO4 may exacerbate disease, it has not been clearly established that it can initiate it. Some limited evidence indicates that it can. For example, in two previously healthy human subjects, Sim and Pattle (1957) found the development of what appeared to be long-lasting symptoms of bronchitis as a result of repeated exposure to H2SO4 lasting 1 h and given no more than twice a week, with at least 24 h between exposures. Concentrations were, however, high, ranging from 3 to 39 mg/m3. The suggestion for the role of H2SO4 in the development of chronic bronchitis is given added strength when results of studies of submicrometer H2SO4 or whole fresh cigarette smoke exposures, both conducted at New York University Medical Center with laboratory animals and humans, are compared (Lippmann et al., 1982). Cigarette smoke is an agent known to be involved in the etiology of human chronic bronchitis and, as shown in Fig. 25.10, the effects of smoking two cigarettes on the mucociliary clearance of tracer particles are essentially the same in humans and donkeys in terms of a transient acceleration

HEALTH EFFECTS

985

FIGURE 25.10 (a) Tracheobronchial particle retention versus time for donkey Gus in a control test, and in tests involving exposure to whole fresh cigarette smoke from the indicated number of cigarettes. (b) Tracheobronchial particle retention versus time for a 38-year-old nonsmoking man for two tagged aerosols inhaled 2.5 h apart on the same day. Smoke from two cigarettes, which was inhaled beginning about 1 h after the inhalation of the second tagged aerosol, accelerated the clearance of both (reproduced from Lippmann et al., 1982).

of clearance in single low-dose exposures. Both agents produce a transient slowing of mucociliary particle clearance following single high-dose exposures (Lippmann et al., 1982). Furthermore, alterations in clearance rates persist for several months, which followed multiple exposures to both agents (Fig. 25.11). Although direct evidence for an association between intermittent low-level exposures to H2SO4 and chronic bronchitis is lacking, the similarity in response between H2SO4 and cigarette smoke exposures suggests that such an association is likely. Human chronic bronchitis is a clinically diagnosed disease, but it is the one that is characterized by certain morphological changes associated with these clinical symptoms (Lourenco, 1969; Reid, 1963; Mitchell, 1967; Suhs et al., 1969; Jefferey, 1982). One of the

FIGURE 25.11 Effect of exposures to the whole fresh smoke from 30 cigarettes, three times per week, on the mean residence time for tagged particles on the tracheobronchial airways in three donkeys. The dashed lines indicate the range of the three control tests for each animal, which preceded the smoke exposures (reproduced from Lippmann et al., 1982).

986

SULFUR OXIDES—SO2, H2SO4, NH4HSO4, AND (NH4)2SO4

basic stigmata is an increase in the number and/or size of epithelial mucus secretory cells in both proximal bronchi as well as in peripheral airways, where such cells are normally absent or few in number; this change is accompanied by an increase in the volume of secretion (Reid, 1963). In the subchronic as well as chronic rabbit studies, an increase in epithelial secretory cell proportions in smaller airways was noted (Schlesinger et al., 1983; Gearhart and Schlesinger, 1989). The appearance of persistently increased secretory cell number in peripheral airways as a result of H2SO4 is a finding of major importance, since excessive mucus production in small airways, which is consistent with an increase in the propagation of secretory cells, may be an early feature in the pathogenesis of bronchitis (Hogg et al., 1968). Furthermore, it demonstrates an underlying histological change consistent with the observed physiological effects of the H2SO4, that is, altered mucociliary clearance. In addition to a change in the relative number of secretory cells in different airway levels of acid-exposed rabbits, two other changes were noted after H2SO4 exposures. There was an increase in epithelial thickness and a decrease in airway diameter. A significant increase in epithelial thickness of small bronchi and bronchioles occurred in rabbits exposed orally at approximately 250 mg/m3 and nasally at approximately 500 mg/m3. In addition, in the oral exposure series, the lumen diameter of the smallest airways was significantly less than in the sham controls. In human chronic bronchitis as well as in experimental bronchitis in laboratory animals, an initial change in secretory cell number or size is followed by intrabronchial narrowing, especially in small bronchi and bronchioles, in part due to a thickening of the bronchial wall (Matsuba and Thurlbeck, 1973; McKenzie et al., 1969). In summary, the first stage of effect of acid exposure may be a change in secretory cell proportions in the airways, and thickening of the epithelium may occur later. Thus, the studies of Schlesinger et al. (1983) and Gearhart and Schlesinger (1989) provide further support for the role of H2SO4 in the pathogenesis of chronic bronchitis via effects on the mucociliary clearance system. However, the progression of clearance dysfunction in the pathogenesis of chronic bronchitis is not known. The Albert et al. (1973) schema describing the pathogenesis of chronic bronchitis in man from cigarette smoking is illustrated in Fig. 25.12. It may also apply to repeated exposures to other irritants such as H2SO4. According to this schema, irritant inhalation initially results in a tendency toward some acceleration of clearance, as excess mucus is produced but mucosal damage has not occurred. The H2SO4 dose delivered to the rabbit in nasal breathing at 250 mg/m3 may have been enough to initiate this first stage. An increase in the number of airways containing epithelial secretory cells is consistent with increased mucus production. However, the degree of clearance rate change could vary with the individual rabbit and with the time after exposure at which the clearance was measured. This may account for the fact that a significant acceleration was often observed when clearance was measured immediately after H2SO4 exposure, whereas retardation was more commonly observed when clearance was measured 1 day after the last H2SO4 exposure (Gearhart and Schlesinger, 1988). In the next stage of the Albert et al. (1973) schema, a further increase in the level of secretion, coupled with some mucosal damage, results in an overloading of transport mechanisms; the result is a retardation of clearance. Such a retardation was observed in the study by Schlesinger et al. (1979) involving 6 months of daily 1 h H2SO4 exposures at 100 mg/m3 in donkeys and in the study of Gearhart and Schlesinger (1988) involving daily 1 h H2SO4 exposures in rabbits for 1 year.

HEALTH EFFECTS

Asymptomatic Compensated bronchorrhea

Symptomatic Decompensated bronchorrhea

987

Disability Airway obstruction

Mucus production

tran spo rt

Stasis and refluxing

Th

a ro

le tc

n ari

gc

nc cie effi tion i s o p ol de Aeros

y

gh ou D ys pn e

Rate

Mu coc ilia ry

a

Cigarette smoking

FIGURE 25.12 Tentative scheme for the pathogenesis of obstructive lung disease resulting from exposure to inhaled irritants (reproduced from Albert et al., 1973).

Since H2SO4 produces essentially the same sequence of effects on mucociliary bronchial clearance as cigarette smoke following both short-term and chronic exposures, it may be capable of contributing to the development of bronchitis. But the question still remains whether variable clearance rates and persistent clearance rate changes merely predispose to chronic bronchitis or are the actual initiating events in a pathogenic sequence leading to its development. Furthermore, the response of the mucociliary clearance system observed in the rabbits may be adaptive rather than pathological. Many irritants may stimulate clearance at low doses or after exposure for a short time and then retard it at higher doses or with prolonged exposures (Wolff et al., 1981). An increase in secretory cell proportions is consistent with hypersecretion. Thus, low-level exposures may initially increase secretion, which can be coped with, and may even be protective. However, pathological changes appear when adaptive capacity is overloaded. Thus, with increasing exposure time or dose, the degree of enhanced secretion may be too great, resulting in overwhelming clearance and leading to retardation and, eventually, bronchitis (Schlesinger et al., 1983). Airborne Acidity and Cancer: There is a possibility that exposure to acidic aerosols may play a role in carcinogenesis. Soskolne et al. (1989) published a review that examined a broad array of epidemiological and toxicological literature. They concluded that there was support for the hypothesis that acidic pollutants contribute to carcinogenesis in humans. They examined possible biologic mechanisms for such a contribution, including pH modulation of toxicity of xenobiotics and pH-induced changes of cells involving mitotic and enzyme regulation. The strongest epidemiological evidence for an effect of acid mists on lung cancer comes from a follow-up study through 1986 of 1165 steelworkers exposed to acid mists at steel pickling operations at concentrations on the order of 200 mg/m3 (droplet size not specified). The results are summarized in Table 25.2. Since the H2SO4 concentrations in ambient air are about two orders of magnitude lower, there is no strong evidence at this time to link ambient air exposures to carcinogenesis. Public health authorities should consider the Soskolne et al. hypothesis as a margin-of-safety factor in the establishment of air quality guidelines and/or

988

SULFUR OXIDES—SO2, H2SO4, NH4HSO4, AND (NH4)2SO4

TABLE 25.2

Lung Cancer and Acid Mists Standardized Mortality Ratioa

Disease Lung cancer (without adjustment for smoking, n ¼ 1156) Lung cancer (with adjustment for smoking, n ¼ 752) Heart disease (n ¼ 1156) All causes (n ¼ 1156)

Overall

20 Years Since First Exposure

1.55 (1.12–2.11)

1.72 (1.21–2.39)

1.36 (0.97–1.84)

1.50 (1.05–2.27)

0.92 (0.77–1.09) 0.92 (0.83–1.01)

Source: Steenland and Beaumont (1989). Note: The cohort was 1156 male steelworkers. Exposure was to an average H2SO4 concentration of 0.2 mg/m3 based on 49 air samples collected in the late 1970s. Average duration of employment (and exposure) was 8.8 years. a

With 95% confidence intervals.

standards based on the more clearly established health effects associated with exposure to acidic aerosols. 25.2.9

Summary of Health Effects of Acidic Aerosols

The human health effects of major concern with respect to the inhalation of acidic aerosols are bronchospasm in asthmatics and chronic bronchitis in all exposed persons. The former relates to acute exposure, whereas the latter can be related more closely to chronic or cumulative exposures. In either case, the effects are produced by droplets depositing on the surface of the conductive airways of the lungs. Studies related to the provocation of bronchospasm show evidence for increased airway resistance in exercising mild to moderate asthmatics. Koenig et al. (1983a, 1989 reported increased airway resistance following H2SO4 exposures for 30 min at rest and 10 min of exercise for adolescents at 68 or 100 mg/m3 (0.6 mm diameter droplets). Bauer et al. (1988) saw similar effects in adults exposed for 2 h at 75 mg/m3 while exercising or for 16 min at 450 mg/m3 while at rest (0.8 mm). Koenig et al. (1989) also showed that the effects previously seen at 68 mg/m3 were increased when there was coexposure to 0.1 ppm (260 mg/m3) of SO2. Evidence for synergism is also to be found in the work of Amdur and Chen (1989), who showed that sulfuric acid (1 h at 20 mg/m3) on ultrafine ZnO particles that simulate coal combustion effluent, when present in a mixture with SO2, produces increased lung reactivity responses about 10-fold greater than those produced by pure droplets of H2SO4 of comparable size. A causal basis for the historic association between acidic aerosols and the prevalence of bronchitis in humans has been established, at least in part, by the acute and chronic exposure studies in which rates of particle clearance from the lungs have been measured. The shortterm effects are cumulative over at least several hours, and total exposures at the equivalent of current peak ambient levels produce similar transient changes in mucociliary clearance rates in rabbits, donkeys, and humans. Repetitive daily exposures of rabbits and donkeys, at comparable rates of exposure, produce both transient and persistent clearance abnormalities that are essentially the same as those produced by whole fresh cigarette smoke, a known causal factor for chronic bronchitis in humans.

AMBIENT AIR QUALITY STANDARD AND GUIDELINES

989

Conclusive evidence for human health responses to ambient acidic aerosols is lacking, and inferences have to be drawn from associations between health and SO42 . Of particular interest are the studies of Dockery et al. (1982) and Dassen et al. (1986) showing the effects of episodic overexposures to ambient mixtures containing concentrations of SO2 and PM close to current U.S. standards. These exposures produced an apparent continuum of response, with a substantial fraction (perhaps 25% or more) of children having at least a small loss of lung function persisting for at least several weeks. A smaller percentage may have persistent functional decrements exceeding 10%. At concentrations that are about twice the current U. S. standards, an episode in western Germany in 1985 produced increases in deaths, hospital admissions, outpatient visits, and ambulance deliveries to hospitals (Wichmann et al. 1989). More direct semiquantitative evidence for a causal role for acidic aerosols comes from the studies of Kitagawa (1984) and Thurston et al. (1989). Kitagawa showed an association between acid aerosols and morbidity in Japan, and Lippmann and Ito (1995) showed that total daily mortality in London in the period 1965–1972 was more closely associated with daily measured acid aerosol than with BS or SO2. Epidemiological studies that directly address the role of acidic aerosols on human health are beginning to produce results. The reports of Raizenne et al. (1996) and Ostro et al. (1989, 1991 are both encouraging and dismaying. They are encouraging in that they directly address serious public health concerns. The initial findings are, however, quite disturbing. Dockery et al. (1996) showed that the prevalence of bronchitic symptoms in schoolchildren varies from about 3.5% to 10% as annual average Hþ concentrations (expressed as H2SO4 equivalent) varied from 0.4 to 1.8 mg/m3. Ostro et al. (1989, 1991 reported that responses among adult asthmatics were more closely associated with Hþ than with any other pollution variable for concentrations (expressed as H2SO4 equivalent) ranging from 0.04 to 0.84 mg/m3. These are commonly encountered levels in the United States and well below historic ambient levels in the United States and Europe. More direct measurement data need to be made and coupled to health effects studies. The implications of the preliminary data to public health and, ultimately, to the health costs of fossil fuel consumption need to be much better documented.

25.3 AMBIENT AIR QUALITY STANDARD AND GUIDELINES 25.3.1

Sulfur Dioxide

The initial primary (health-based) NAAQS for SO2 was established by the U.S. Environmental Protection Agency (EPA) in 1971, with a concentration of 365 mg/m3 (140 ppb) for a 24 h maximum not to be exceeded more than once per year and with 80 mg/m3 (30 ppb) as an annual arithmetic mean. EPA initiated a review of the 1971 SO2 NAAQS in 1980, and reaffirmed the levels in 1987. A further review was initiated in 2005, and the outcome of that review is not known at this time. The World Health Organization has developed Air Quality Guidelines for SO2 to assist member states in establishing their own standards. The latest version was prepared in 2006 (WHO, 2006). 25.3.2

WHO Guidelines

Short-Term Exposures: Controlled studies with exercising asthmatics indicate that some of them experience changes in pulmonary function and respiratory symptoms after periods

990

SULFUR OXIDES—SO2, H2SO4, NH4HSO4, AND (NH4)2SO4

of exposure as short as 10 min. Based on this evidence, it is recommended that a value of 500 mg/m3 (0.175 ppm) should not be exceeded over averaging periods of 10 min. Because exposure to sharp peaks depends on the nature of local sources and meteorological conditions, no single factor can be applied to this value to estimate corresponding guideline values over somewhat longer periods, such as an hour. 25.3.3

Exposure over a 24 h Period and Long-Term Exposure

Day-to-day changes in mortality, morbidity, or lung function related to 24 h average concentrations of sulfur dioxide are necessarily based on epidemiological studies in which people are in general exposed to a mixture of pollutants, with little basis for separating the contributions of each to the effects, which is why guideline values for sulfur dioxide were linked before 1987 with corresponding values for particle matter. This approach led to a guideline value of 125 mg/m3 (0.04 ppm) before 1987 as a 24 h average, after applying an uncertainty factor of 2 to the lowest observed adverse effect level. In the 2000 revision, it was noted that recent epidemiological studies showed separate and independent adverse public health effects for particulate matter and sulfur dioxide, and this led to a separate AQG for SO2 of 125 mg/m3 (0.04 ppm) as a 24 h average. Now recent evidence, beginning with the Hong Kong study (Hedley et al., 2002), of a major reduction in sulfur content in fuels over a very short period of time shows an associated substantial reduction in health effects (childhood respiratory disease and all age mortality outcomes). In time-series studies on hospital admissions for cardiac disease, there is no evidence of a concentration threshold within the range of 5–40 mg/m3 in both Hong Kong and London (Wong et al., 2002). Daily SO2 was significantly associated with daily mortality in 12 Canadian cities with an average concentration of only 5 mg/m3 (Burnett et al., 1998). If there were an SO2 threshold for either the Burnett et al. (1998) study of daily mortality or the annual mortality study of Pope et al. (1995, 2002), they would have to be very low. For the significant associations in the ACS cohort for the 1982–1998 period in 126 U.S. metropolitan areas, the mean SO2 was 6.7 mg/m3 (Pope et al., 2002). Nevertheless, there is still considerable uncertainty as to whether sulfur dioxide is the pollutant responsible for the observed adverse effects or rather a surrogate for ultrafine particles or some other correlated substance. In considering: (1) the uncertainty of SO2 in causality, (2) the practical difficulty of reaching levels that are certain to be associated with no effects, and (3) the need to provide greater degrees of protection than those provided by the current WHO Guidelines and assuming that reduction in exposure to a causal and correlated substance is achieved by reducing sulfur dioxide concentrations, there is a basis for revising the 2000 24 h guideline downward for sulfur dioxide, and the following guideline is recommended as a prudent precautionary level: 24 h:20 mg/m3. An annual guideline is not needed, since compliance with the 24 h level will assure low levels for the annual average. For the 24 h guideline, which may be quite difficult for some countries to achieve in the short term, we suggest a stepped approach using high and intermediate goals, as shown in Table 25.3. For instance, a country could move toward guideline compliance by controlling emissions from one major source at a time, selecting among motor vehicle sources, industrial sources, and power sources, for the greatest effect on SO2 at the lowest cost, and monitor public health and SO2 levels for health effect gains. Demonstrating health benefits will provide an incentive to mandate controls for the next major source category.

REFERENCES

991

TABLE 25.3 Existing WHO Guideline—Reason to Change. Controlled Studies on Exercising Asthmatics 24 h Average SO2 WHO Historic WHO Intermediate

Guideline

125 mg/m3 2000 WHO level 50 mg/m3 Intermediate goal based on controlling either (1) motor vehicle, (2) industrial emissions, and/or (3) power production; this would be a reasonable and feasible goal within the next few years for some developing countries and lead to significant health improvements that would justify further improvements (such as aiming for the guideline) 20 mg/m3 Based on Hong Kong intervention study, Hong Kong and London comparison study, and time-series studies (median value) Coherence with ACS mortality studies with annual mean media value of 6.7 mg/m3

10 min Average SO2

500 mg/m3

These recommended guideline values for sulfur dioxide are not linked with guidelines for particles. 25.3.4

Other Sulfur Oxides

There are no established standards or guidelines for ambient air concentration limits for H2SO4 or its ammonium salts. ACKNOWLEDGMENTS This research was supported, in part, by EPA (CR818325 and 822050), and by Grant ES 04612 from the National Institute of Environmental Health Sciences (NIEHS). It is part of a Center Program supported by Grant ES 00260. REFERENCES Abbey DE, Nishino N, McDonnell WF, Burchette RJ, Knutsen SF, Beeson I, Yang JX, (1999). Longterm inhalable particles and other air pollutants related to mortality in nonsmokers. Am. J. Resp. Crit. Med. 159:337–382. Aga E, Samoli E, Touloumi G, Anderson HR, Cadum E, Forsbergz B, Goodman P, Goren A, Kotesovec F, Kriz B, Macarol-Hiti M, Medinazz S, Paldy A, Schindler C, Sunyer J, Tittanen P, Wojtyniak B, Zmirou D, Schwartz J, Katsouyanni K (2003) Short-term effects of ambient particles on mortality in the elderly: results from 28 cities in the APHEA2 project. Eur. Respir. J. 21 (Suppl. 40):28s–33s. Albert RE, Lippmann M, Peterson Jr HT, Berger JM, Sanborn K, Bohning DE (1973) Deposition and clearance of aerosols. Arch. Intern. Med. 131:115–127.

992

SULFUR OXIDES—SO2, H2SO4, NH4HSO4, AND (NH4)2SO4

Amdur MO (1989) Sulfuric acid: the animals tried to tell us. Appl. Ind. Hyg. 4:189–197. Amdur MO, Chen LC (1989) Furnace generated acid aerosols: speciation and pulmonary effects. Environ. Health Perspect. 79:147–150. Amdur MO, Silverman L, Drinker P (1952) Inhalation of H2SO4 mist by human subjects. Arch. Ind. Hyg. Occup. Med. 6:305–313. Amdur MO, Dubriel M, Creasia D (1978) Respiratory response of guinea pigs to low levels of sulfuric acid. Environ. Res. 15:418–423. Amdur MO, Sarofim AF, Neville M, Quann RJ, McCarthy JF, Elliot JF, Lam HF, Rogers AE, Connor MW (1986) Coal combustion aerosols and SO2: an interdisciplinary analysis. Environ. Sci. Technol. 20:138–145. Anderson SD (1985) Issues in exercise-induced asthma. J. Allergy Clin. Immunol. 76:763–772. Anderson HR, Spix C, Medina S, et al. (1997) Air pollution and daily admissions for chronic obstructive pulmonary disease in 6 European cities: results from the APHEA project. Eur. Respir. J. 10:1064–1071. Aris R, Christian D, Sheppard D, Balmes JR (1990) Acid fog-induced bronchoconstriction. Am. Rev. Respir. Dis. 141:546–551. Arossa W, Spinaci S, Bugiani M, Natale P, Bucca C, deCandussio G (1987) Changes in lung function of children after an air pollution decrease. Arch. Environ. Health 42:170–174. Atkinson RW, Anderson HR, Sunyer J, et al. (2001) Acute effects of particulate air pollution on respiratory admissions: results from APHEA 2 project. Air pollution and health: a European approach. Am. J. Respir. Crit. Care Med. 164:1860–1866. Atkinson RW, Anderson HR, Sunyer J, et al. (2003) Acute effects of particulate air pollution on respiratory admissions. In: Revised Analyses of Time-Series Studies of Air Pollution and Health. Special Report. May 16, 2003. Available at http://www.healtheffects.org/pubs-special.htm.Boston, MA:Health Effects Institute. pp.81–84. Avol EL, Linn WS, Wightman LH, Whynot JD, Anderson KR, Hackney JD (1988) Short-term respiratory effects of sulfuric acid in fog: a laboratory study of healthy and asthmatic volunteers. J. Air Pollut. Control Assoc. 38:258–263. Avol EL, Linn WS, Shamoo DA, Anderson KR, Peng RC, Hackney JD (1990) Respiratory responses of young asthmatic volunteers in controlled exposures to sulfuric acid aerosol. Am. Rev. Respir. Dis. 142:343–348. Ballester F, Saez M, Perez-Hoyos S, et al. (2002) The EMECAM project: a multicentre study on air pollution and mortality in Spain: combined results for particulates and for sulfur dioxide. Occup. Environ. Med. 59:300–308. Baskurt OK, Levi E, Caglayan S, Dikmenoglu N, Kutman MN (1990) Hematological and hemorheological effects of air pollution. Arch. Environ. Health 45:224–228. Bauer MA, Utell MJ, Speers DM, Gibb FR, Morrow PE (1988) Effects of near ambient levels of sulfuric acid aerosol on lung function in exercising subjects with asthma and COPD. Am. Rev. Respir. Dis. 137:A167. Bethel RA, Erle DJ, Epstein J, Sheppard D, Nadel JA, Boushey HA (1983) Effect of exercise rate and route of inhalation on sulfur-dioxide-induced bronchoconstriction in asthmatic subjects. Am. Rev. Respir. Dis. 128:592–596. Bethel RA, Sheppard D, Epstein J, Tam E, Nadel JA, Boushey HA (1984) Interaction of sulfur dioxide and cold dry air in causing bronchoconstriction in asthmatic subjects. J. Appl. Physiol. Respir. Environ. Exercise Physiol. 57:419–423. Brauer M, Koutrakis P, Spengler JD (1989) Personal exposures to acidic aerosols and gases. Environ. Sci. Technol. 23:1408–1412. Burnett RT, Goldberg MS (2003) Size-fractionated particulate mass and daily mortality in eight Canadian cities. In: Revised Analyses of Time-Series Studies of Air Pollution and Health. Special

REFERENCES

993

Report. May 16, 2003. Available at http://www.healtheffects.org/pubs-special.htm.Boston, MA: Health Effects Institute. pp.85–90. Burnett RT, Cakmak S, Brook JR (1998) The effect of the urban ambient air pollution mix on daily mortality rates in 11 Canadian cities. Can. J. Public Health 89:152–156. Burnett RT, Brook J, Dann T, et al. (2000) Association between particulate- and gas-phase components of urban air pollution and daily mortality in eight Canadian cities. Inhal. Toxicol. 12 (Suppl. 4):15–39. Clancy L, Goodman P, Sinclair H, Dockery DW (2002) Effect of air-pollution control on death rates in Dublin, Ireland: an intervention study. Lancet 360:1210–1214. Commins BT, Waller RE (1963) Determination of particulate acid in town air. Analyst 88:364–367. Damokosh AI, Spengler JD, Dockery DW, Ware JH, Speizer FE (1993) Effects of acidic particles on respiratory symptoms in 7 U.S. communities. Am. Rev. Respir. Dis. 147:A632. Dassen W, Brunekreef B, Hoek G, Hofschreuder P, Staatsen B, deGroot H, Schouten E, Biersteker K (1986) Decline in children’s pulmonary function during an air pollution episode. J. Air Pollut. Control Assoc. 36:1223–1227. Deal EC, McFadden ER, Ingram RH, Jaeger JJ (1979) Hyperpnea and heat flux: initial reaction sequence in exercise induced asthma. J. Appl. Physiol. 46:476–483. Dockery DW, Ware JH, Ferris BG Jr, Speizer FE, Cook NR, Herman SM (1982) Change in pulmonary function in children associated with air pollution episodes. J. Air Pollut. Control Assoc. 32:937–942. Dockery DW, Schwartz J, Spengler JD (1992) Air pollution and daily mortality: associations with particulates and acid aerosols. Environ. Res. 59:362–373. Dockery DW, Pope CA III Xu X, Spengler JD, Ware JH, Fay ME, Ferris BG Jr, Speizer FE (1993) An association between air pollution and mortality in six U.S. cities. N. Engl. J. Med. 329: 1753–1759. Dockery DW, Cunningham J, Damokosh AI, Neas LM, Spengler JD, Koutrakis P, Ware JH, Raizenne M, Speizer FE (1996) Health effects of acid aerosols on North American children: respiratory symptoms. Environ. Health Perspect. 104:500–505. Dominici F, McDermott A, Zeger S , et al. (2002) On generalized additive models in time-series studies of air-pollution and health. Am. J. Epidemiol. 156:1–11. Dominici F, McDermott A, Daniels M , et al. (2003) Mortality among residents of 90 cities. In: Revised Analyses of Time-Series Studies of Air Pollution and Health. Special Report. May 16, 2003. Available at http://www.healtheffects.org/pubs-special.htm.Boston, MA:Health Effects Institute. pp.9–24. Ericsson G, Camner P (1983) Health effects of sulfur oxides and particulate matter in ambient air. Scand. J. Work Environ. Health 9 (S3):1–52. Gearhart JM, Schlesinger RB (1986) Sulfuric acid-induced airway hyperresponsiveness. Fund. Appl. Toxicol. 7:681–689. Gearhart JM, Schlesinger RB (1988) Response of the tracheobronchial mucociliary clearance system to repeated irritant exposure: effect of sulfuric acid mist on function and structure. Exp. Lung Res. 14:587–605. Gearhart JM, Schlesinger RB (1989) Sulfuric acid-induced changes in the physiology and structure of the tracheobronchial airways. Environ. Health Perspect. 79:127–137. Gorham E (1958) Bronchitis and the acidity of urban precipitation. Lancet 2:691–693. Gryparis A, Forsberg B, Katsouyanni K , et al. (2004) Acute effects of ozone on mortality from the “Air Pollution and Health: A European Approach” Project. Am. J. Respir. Crit. Care Med. 170:1080–1087. Hackney JD, Linn WS, Bailey RM, Spier CE, Valencia LM (1984) Time course of exercise-induced bronchoconstriction in asthmatics exposed to sulfur dioxide. Environ. Res. 34:321–327.

994

SULFUR OXIDES—SO2, H2SO4, NH4HSO4, AND (NH4)2SO4

Hedley AJ, Wong CM, Thach TQ, Ma S, Lam TH, Anderson HR (2002) Cardiorespiratory and allcause mortality after restrictions on sulphur content of fuel in Hong Kong: an intervention study. Lancet 360:1646–1652. Hoek G (2003) Daily mortality and air pollution in the Netherlands. In: Revised Analyses of Time-Series Studies of Air Pollution and Health. Special Report. May 16, 2003. Available at http://www.healtheffects.org/pubs-special.htm.Boston, MA:Health Effects Institute. pp.133–142. Hoek G, Brunekreef B , et al. (2000) Daily mortality and air pollution in the Netherlands. J. Air Waste Manage Assoc. 50:1380–1389. Hoek G, Brunekreef B, Fischer P, et al. (2001) The association between air pollution and heart failure, arrhythmia, embolism, thrombosis, and other cardiovascular causes of death in a time series study. Epidemiology 12:355–357. Hogg JC, Macklem PT, Thurlbeck WM (1968) Site and nature airway obstruction in obstructive lung disease. N. Engl. J. Med. 278:1355–1360. Holma B (1985) Influence of buffer capacity and pH-dependent rheological properties of respiratory mucus on health effects due to acidic pollution. Sci. Total Environ. 41:101–123. Horstman DH, Seal Jr E, Folinsbee LJ, Ives P, Roger LJ (1988) The relationship between exposure duration and sulfur dioxide-induced bronchoconstriction in asthmatic subjects. Am. Ind. Hyg. Assoc. J. 49:38–47. Imai M, Yoshida K, Kitabatake M (1986) Mortality from asthma and chronic bronchitis associated with changes in sulfur oxides air pollution. Arch. Environ. Health 41:29–35. Ito K, Thurston GD, Hayes C, Lippmann M (1993) Associations of London, England, daily mortality with particulate matter, sulfur dioxide, and acidic aerosol pollution. Arch. Environ. Health 48:213–220. Jefferey PK (1982) The effects of irritation on the structure of bronchial epithelium. In: Bonsignore G, Cummings G,editors. The Lung in the Environment. New York: Plenum Press. pp.303–313. Katsouyanni K, Touloumi G, Spix C, Schwartz J, Balducci F, Medina S, Rossi G, Wojtyniak B, Sunyer J, Bacharova L, Schouten JP, Ponk A, Anderson HR (1997) Short-term effects of ambient sulfur dioxide and particulate matter on mortality in 12 European cities: results from time series data from the APHEA project. BMJ 314:1658–1663. Katsouyanni K, Touloumi G, Samoli E, Gryparis A, Le Tertre A, Monopolis Y, Rossi G, Zmirou D, Ballester F, Boumghar A, Anderson HR, Wojtyniak B, Paldy A, Braunstein R, Pekkanen J, Schindler C, Schwartz J (2001) Confounding and effect modification in the short-term effects of ambient particles on total mortality: results from 29 European cities within the APHEA 2 project. Epidemiology 12:521–531. Katsouyanni K, Touloumi G, Samoli E, Petasakis Y, Analitis A, Le Tertre A, Rossi G, Zmirou D, Ballester F, Boumghar A, Anderson HR, Wojtyniak B, Paldy A, Braunstein R, Pekkanen J, Schindler C, Schwartz J (2003)Sensitivity analysis of various models of short-term effects of ambient particles on total mortality in 29 cities in APHEA 2. In: Revised Analyses of Time-Series Studies of Air Pollution and Health. Special Report. May 16, 2003. Available at http://www.healtheffects.org/pubs-special.htm.Boston, MA:Health Effects Institute. pp.157–164. Keeler GJ, Spengler JD, Castillo RA (1991) Acid aerosol measurements at a suburban Connecticut site. Atmos. Environ. Part A 25:681–690. Kehrl HR, Roger LJ, Hazucha MJ, Horstman DH (1987) Differing response of asthmatics to SO2 exposure with continuous and intermittent exercise. Am. Rev. Respir. Dis. 135:350–355. Kitagawa T (1984) Cause analysis of the Yokkaichi asthma episode in Japan. J. Air Pollut. Control Assoc. 34:743–746.

REFERENCES

995

Kleinman MT (1984) Sulfur dioxide and exercise: relationships between response and absorption in upper airways. J. Air Pollut. Control Assoc. 34:32–37. Kleinman MT, Phalen RF, Mautz WJ, Mannix RC, McClure TR, Crocker TT (1989) Health effects of acid aerosols formed by atmospheric mixtures. Environ. Health Perspect. 79:137–145. Koenig JQ, Pierson WE, Horike M, Frank R (1981) Effects of inhaled SO2 plus NaCl aerosol combined with moderate exercise on pulmonary function in asthmatic adolescents. Environ. Res. 25:340–348. Koenig JQ, Pierson WE, Horike M, (1983a) The effects of inhaled sulfuric acid on pulmonary function in adolescent asthmatics. Am. Rev. Respir. Dis. 128:221–225. Koenig JQ, Pierson WE, Horike M, Frank R (1983b) A comparison of the pulmonary effects of 0.5 ppm versus 1.0 ppm sulfur dioxide plus sodium chloride droplet in asthmatic adolescents. J. Toxicol. Environ. Health 11:129–139. Koenig JQ, Covert DS, Pierson WE (1989) Effects of inhalation of acidic compounds on pulmonary function in allergic adolescent subjects. Environ. Health Perspect. 79:173–178. Koenig JQ, Covert DS, Hanley QS, van Belle G, Pierson WE (1990) Prior exposure to ozone potentiates subsequent response to sulfur dioxide in adolescent asthmatic subjects. Am. Rev. Respir. Dis. 141:377–380. Krewski D, Burnett RT, Goldberg MS, et al. (2000) Reanalysis of the Harvard Six Cities Study and the American Cancer Society Study of Particulate Air Pollution and Mortality. Special Report. Cambridge, MA:Health Effects Institute. Kunzli N, Tager IB (1997) The semi-individual study in air pollution epidemiology: a valid design as compared to ecologic study. Environ. Health Perspect. 105:1078–1083. Lamborg C, Keeler GJ, Evans G (1992) Atmospheric Acidity Measurements During the Lake Michigan Urban Air Toxics Study. EPA Report No. EPA/600/A-92/250. Available from NTIS, Springfield, VA, pB93-121069. Research Triangle Park, NC: U.S. Environmental Protection Agency, Atmospheric Research and Exposure Assessment Laboratory. Larson TV, Covert DS, Frank R, Charlson RJ (1977) Ammonia in the human airways: neutralization of inspired acid sulfate aerosols. Science 197:161–163. Last JA (1989) Effects of inhaled acids on lung biochemistry. Environ. Health Perspect. 79:115–119. Lawther PJ (1958) Climate, air pollution and chronic bronchitis. Proc. R. Soc. Med. 51:262–264. Lawther PJ, Waller RE, Henderson M (1970) Air pollution and exacerbations of bronchitis. Thorax 25:525–539. Le Tertre A, Medina S, Samoli E, et al. (2002) Short-term effects of particulate air pollution on cardiovascular diseases in eight European cities. J. Epidemiol. Comm. Health 56:773–779. Le Tertre A, Medina S, Samoli E, et al. (2003)Short-term effects of particulate air pollution on cardiovascular diseases in eight European cities. In: Revised Analyses of Time-Series Studies of Air Pollution and Health. Special Report. May 16, 2003. Available at http://www.healtheffects.org/ pubs-special.htm.Boston, MA: Health Effects Institute. pp.173–176. Leaderer BP, Boone PM, Hammond SK (1990) Total particle, sulfate, and acidic aerosol emissions from kerosene space heaters. Environ. Sci. Technol. 24:908–912. Leikauf G, Yeates DB, Wales KA, Albert RE, Lippmann M (1981) Effects of sulfuric acid aerosol on respiratory mechanics and mucociliary particle clearance in healthy nonsmoking adults. Am. Ind. Hyg. Assoc. J. 42:273–282. Leikauf GD, Spektor DM, Albert RE, Lippmann M (1984) Dose-dependent effects of submicrometer sulfuric acid aerosol on particle clearance from ciliated human lung airways. Am. Ind. Hyg. Assoc. 45:285–292. Linn WS, Venet TG, Shamoo DA, Valencia LM, Anzar UT, Spier CE, Hackney JD (1983) Respiratory effects of sulfur dioxide in heavily exercising asthmatics: a dose–response study. Am. Rev. Respir. Dis. 127:278–283.

996

SULFUR OXIDES—SO2, H2SO4, NH4HSO4, AND (NH4)2SO4

Linn WS, Shamoo DA, Anderson KR, Whynot JD, Avol EL, Hackney JD (1985) Effects of heat and humidity on the responses of exercising asthmatics to sulfur dioxide exposure. Am. Rev. Respir. Dis. 131:221–225. Linn WS, Avol EL, Anderson KR, Shamoo DA, Peng RC, Hackney JD (1989) Effect of droplet size on respiratory responses to inhaled sulfuric acid in normal and asthmatic volunteers. Am. Rev. Respir. Dis. 140:161–166. Lippmann M, Ito K (1995) Separating the effects of temperature and season on daily mortality from those of air pollution in London: 1965–1972. Inhal. Toxicol. 7:85–97. Lippmann M, Yeates DB, Albert RE (1980) Deposition, retention, and clearance of inhaled particles. Br. J. Ind. Med. 37:337–362. Lippmann M, Schlesinger RB, Leikauf G, Spektor D, Albert RE (1982) Effects of sulphuric acid aerosols on respiratory tract airways. Ann. Occup. Hyg. 26:677–690. Lippmann M, Gearhart JM, Schlesinger RB (1987) Basis for a particle size-selective TLV for sulfuric acid aerosols. Appl. Ind. Hyg. 2:188–199. Lourenco RV (1969) Distribution and clearance of aerosols. Am. Rev. Respir. Dis. 101:460–461. Magnussen H, Jorres R, Wagner HM, von Nieding G (1990) Relationship between the airway response to inhaled sulfur dioxide, isocapnic hyperventilation, and histamine in asthmatic subjects. Int. Arch. Occup. Environ. Health 62:485–491. Matsuba K, Thurlbeck WM (1973) Disease of the small airways in chronic bronchitis. Am. Rev. Respir. Dis. 107:552–558. McKenzie HI, Glick M, Outhred KG (1969) Chronic bronchitis in coal miners: antemortem/ postmortem correlations. Thorax 24:527–535. Mitchell RS (1967) Clinical and morphologic correlations in chronic airway obstruction. Am. Rev. Respir. Dis. 97:54–62. Neas LM, Dockery DW, Koutrakis P, Tollerud DJ, Speizer FE (1995) The association of ambient air pollution with twice daily peak expiratory flow rate measurement in children. Am. J. Epidemiol. 141:111–122. Osebold JW, Gershwin LJ, Yuan CZ (1980) Studies on the enhancement of allergic lung sensitization by inhalation of ozone and sulfuric acid aerosol. J. Environ. Pathol. Toxicol. 3:221–234. Ostro B, Lipsett M, Wiener M, Selner JC (1989) A panel study of the effect of acid aerosols on asthmatics. Paper presented at Annual Meeting of the Air and Waste Management Association, Anaheim, CA, June. Ostro BD, Lipsett MJ, Wiener MB, Selner JC (1991) Asthmatic responses to airborne acid aerosols. Am. J. Public Health 81:694–702. Ozkaynak H, Spengler JD (1985) Analysis of health effects resulting from population exposures to acid precipitation precursors. Environ. Health Perspect. 63:45–55. Pope CA, IIIThun MJ, Namboodiri M, Dockery DW, Evans JS, Speizer FE, Heath CW Jr ,(1995) Particulate air pollution is a predictor of mortality in a prospective study of U.S. adults. Am. J. Respir. Crit. Care Med. 151:669–674. Pope CA III, Burnett RT, Thun MJ , et al. (2002) Lung cancer, cardiopulmonary mortality, and longterm exposure to fine particulate air pollution. JAMA 287:1132–1141. Pope CA IIIBurnett RT, Thurston GD , et al. (2004) Cardiovascular mortality and long-term exposure to particulate air pollution: epidemiological evidence of general pathophysiological pathways of disease. Circulation 109:71–77. Raizenne M, Neas LM, Damokosh AI, Dockery DW, Spengler JD, Koutrakis P, Ware J, Speizer FE (1996) Health effects of acid aerosols on North American children: pulmonary function. Environ. Health Perspect. 104:506–514.

REFERENCES

997

Ramsay T, Burnett R, Krewski D (2003) The effect of concurvity in generalized additive models linking mortality and ambient air pollution. Epidemiology 14:18–23. Reid L (1963) An experimental study of the hypersecretion of mucus in the bronchial tree. Br. J. Exp. Pathol. 44:437–445. Roger LJ, Kehrl HR, Hazucha M, Horstman DH (1985) Bronchoconstriction in asthmatics exposed to sulfur dioxide during repeated exercise. J. Appl. Physiol. 59:784–791. Samet J, Zeger S, Dominici F, et al. (2000) The National Morbidity, Mortality, and Air Pollution Study Part II: Morbidity, Mortality, and Air Pollution in the United States.Cambridge, MA: Health Effects Institute. Samoli E, Schwartz J, Wojtyniak B, Touloumi G, Spix C, Balducci F, Medina S, Rossi G, Sunyer J, Bacharova L, Anderson HR, Katsouyanni K (2001) Investigating regional differences in short-term effects of air pollution on daily mortality in the APHEA project: a sensitivity analysis for controlling long-term trends and seasonality. Environ. Health Perspect. 109:349–353. Samoli E, Schwartz J, Analitis A, Petasakis Y, Wojtyniak B, Touloumi G, Spix C, Balducci F, Medina S, Rossi G, Sunyer J, Anderson HR, Katsouyanni K (2003a) Sensitivity analyses of regional differences in short-term effects of air pollution on daily mortality in APHEA cities. In: Revised Analyses of Time-Series Studies of Air Pollution and Health. Special Report. May 16, 2003. Available at http://www.healtheffects.org/pubs-special.htm.Boston, MA:Health Effects Institute. pp.205–210. Samoli E, Touloumi G, Zanobetti A, Le Tertre A, Schindler C, Atkinson R, Vonk J, Rossi G, Saez M, Rabczenko D, Schwartz J, Katsouyanni K (2003b) Investigating the dose–response relation between air pollution and total mortality in the APHEA-2 multicity project. Occup. Environ. Med. 60:977–982. Schenker MB, Samet JM, Speizer FE, Gruhl J, Batterman S (1983) Health effects of air pollution due to coal combustion in the Chestnut Ridge region of Pennsylvania: results of cross-sectional analysis in adults. Arch. Environ. Health 38:325–330. Schlesinger RB (1985) Effects of inhaled acids on respiratory tract defense mechanisms. Environ. Health Perspect. 63:25–38. Schlesinger RB (1986)The effects of inhaled acids on lung defenses. In: Lee SD, Schneider T, Grant LD, Verkerk PJ,editors. Aerosols. Chelsea, MI: Lewis Publishers. Schlesinger RB (1989) Factors affecting the response of lung clearance systems to acid aerosols: role of exposure concentration, exposure time, and relative activity. Environ. Health Perspect. 79:121–126. Schlesinger RB (1990) Exposure–response pattern for sulfuric acid-induced effects on particle clearance from the respiratory region of rabbit lungs. Inhal. Toxicol. 2:21–27. Schlesinger RB, Gearhart JM (1986) Early alveolar clearance in rabbits intermittently exposed to sulfuric acid mist. J. Toxicol. Environ. Health 17:213–220. Schlesinger RB, Halpern M, Albert RE, Lippmann M (1979) Effect of chronic inhalation of sulfuric acid mist upon mucociliary clearance from the lungs of donkeys. J. Environ. Pathol. Toxicol. 2:1351–1367. Schlesinger RB, Naumann BD, Chen LC (1983) Physiological and histological alterations in the bronchial mucociliary clearance system of rabbits following intermittent oral or nasal inhalation of sulfuric acid mist. J. Toxicol. Environ. Health 12:441–465. Schwartz J, Zanobetti A, Bateson T (2003) Morbidity and mortality among elderly residents of cities with daily PM measurements. In: Revised Analyses of Time-Series Studies of Air Pollution and Health. Special Report. May 16, 2003. Available at http://www.healtheffects.org/pubs-special. htm. Boston, MA: Health Effects Institute. pp.25–58.

998

SULFUR OXIDES—SO2, H2SO4, NH4HSO4, AND (NH4)2SO4

Segala C, Fauroux B, Just L, Pascual L, Grimfeld A, Neukirch F (1998) Short-term effect of winter air pollution on respiratory health of asthmatic children in Paris. Eur. Respir. J. 11:677–685. Sheppard D, Eschenbacher WL, Boushey HA, Bethel RA (1984) Magnitude of the interaction between the bronchomotor effects of sulfur dioxide and those of dry (cold) air. Am. Rev. Respir. Dis. 130:52–55. Sim VM, Pattle RE (1957) Effect of possible smog irritants on human subjects. JAMA 165:1908–1913. Skornik WA, Brain JD (1990) Effect of sulfur dioxide on pulmonary macrophage endocytosis at rest and during exercise. Am. Rev. Respir. Dis. 142:655–659. Soskolne CL, Pagano G, Cipollaro M, Beaumont JJ, Giordano GG (1989) Epidemiologic and toxicologic evidence for chronic health effects and the underlying biologic mechanisms involved in sublethal exposures to acidic pollutants. Arch. Environ. Health 44:180–191. Speizer FE (1989) Studies of acid aerosols in six cities and in a new multicity investigation: design issues. Environ. Health Perspect. 79:61–67. Spektor DM, Leikauf GD, Albert RE, Lippmann M (1985) Effects of submicrometer sulfuric acid aerosols on mucociliary transport and respiratory mechanics in asymptomatic asthmatics. Environ. Res. 37:174–191. Spektor DM, Yen BM, Lippmann M (1989) Effect of concentration and cumulative dose of inhaled sulfuric acid on lung clearance in humans. Environ. Health Perspect. 79:167–172. Spengler JD, Keeler GJ, Koutrakis P, Ryan PB, Raizenne M, Franklin CA (1989) Exposures to acidic aerosols. Environ. Health Perspect. 79:43–51. Spengler JD, Koutrakis P, Dockery DW, Raizenne M, Speizer FE (1996) Health effects of acid aerosols on North American children: air pollution exposures. Environ. Health Perspect. 104:492–499. Spix C, Anderson HR, Schwartz J ,et al.(1998) Short-term effects of air pollution on hospital admissions of respiratory diseases in Europe: a quantitative summary of APHEA study results. Air pollution and health: a European approach. Arch. Environ. Health 53:54–64. Stara JF, Dungworth DL, Orthoefer JG, Tyler WS (1980) Long-Term Effects of Air Pollutants in Canine Species. EPA-600/8-80-014.Research Triangle Park, NC:U.S. Environmental Protection Agency. Steenland K, Beaumont J (1989) Further follow-up and adjustment for smoking in a study of lung cancer acid mists. Am. J. Ind. Med. 16:347–354. Stieb DM, Judek S, Burnett RT (2002) Meta-analysis of time-series studies of air pollution and mortality: effects of gases and particles and the influence of cause of death, age, and season. J. Air Waste Manag. Assoc. 52:470–484. Stieb DM, Judek S, Burnett RT (2003) Meta-analysis of time-series studies of air pollution and mortality:update in relation to the use of generalized additive models. J. Air Waste Manag. Assoc. 53:258–261. Stolzel M, Peters A, Wichmann HE (2003)Daily mortality and fine and ultrafine particles in Erfurt, Germany. In: Revised Analyses of Time-Series Studies of Air Pollution and Health. Special Report. May 16, 2003. Available at http://www.healtheffects.org/pubs-special.htm. Boston, MA:Health Effects Institute. pp.231–240. Strauss RH, McFaddenJrER, IngramJrRH, Jaeger JJ (1977) Enhancement of exercise-induced asthma by cold air. N. Engl. J. Med. 297:743–747. Studnicka MJ, Frischer T, Meinert R, Studnicka-Benke A, Hajek K, Spengler JD, Neumann MG (1995) Acidic particles and lung function in children. A summer camp study in the Austrian Alps. Am. J. Respir. Crit. Care Med. 151:423–430. Suh RH, Lumen JL, Leppe MH (1969) An experimental immunologic approach to the induction and perpetuation of chronic bronchitis. Arch. Environ. Health 18:564–573. Suh HH, Spengler JD, Koutrakis P (1992) Personal exposures to acid aerosols and ammonia. Environ. Sci. Technol. 26:2507–2517.

REFERENCES

999

Suh HH, Allen GA, Koutrakis P, Burton RM (1995) Spatial variation in acidic sulfate and ammonia concentrations within Metropolitan Philadelphia. J. Air Waste Manag. Assoc. 45:442–452. Sunyer J, Spix C, Quenel P , et al. (1997) Urban air pollution and emergency room admissions for asthma in four European cities: the APHEA Project. Thorax 52:760–765. Sunyer J, Atkinson R, Ballester F, Le Tertre A, Ayres JG, Forastiere F, Forsberg B, Vonk JM, Bisanti L, Anderson RH, Schwartz J, Katsouyanni K (2003a) Respiratory effects of sulphur dioxide: a hierarchical multicity analysis in the APHEA 2 study. Occup. Environ. Med. 60 (8):e2. Sunyer J, Ballester F, Tertre AL, Atkinson R, Ayres JG, Forastiere F, Forsberg B, Vonk JM, Bisanti L, Tenias JM, Medina S, Schwartz J, Katsouyanni K (2003b) The association of daily sulfur dioxide air pollution levels with hospital admissions for cardiovascular diseases in Europe (The Aphea-II study). Eur. Heart J. 24 (8):752–760. Thompson KM, Koutrakis P, Brauer M, Spengler JD, Wilson WE, Burton RM (1991) Measurements of aerosol acidity: sampling frequency, seasonal variability, and spatial variation. Presented at the 84th Annual Meeting of the Air and Waste Management Association, June 1991, Vancouver, BC, Canada. Pittsburgh, PA: Air and Waste Management Association. Paper No. 91-89.5. Thurston GD, Ito K, Kinney PL, Lippmann M (1992) A multi-year study of air pollution and respiratory hospital admissions in three New York State metropolitan areas: results for 1988 and 1989 summers. J. Expos. Anal. Environ. Epidemiol. 2:429–450. Thurston GD, Lippmann M, Scott M, Fine JM (1997) Summer time haze pollution and children with asthma. Am. J. Respir. Crit. Care Med. 155:654–660. Tunnicliffe WS , et al. (2001) The effect of sulphur dioxide on indices of heart rate variability in normal and asthmatic adults. Eur. Respir. J. 17:604–608. U.S. EPA (1986) Second Addendum to Air Quality Criteria for Particulate Matter and Sulfur Oxides (1982). EPA/600/8-86/020F. Washington, DC:Office of Health and Environmental Assessment. Utell MJ, Morrow PE, Hyde RW (1982) Comparison of normal and asthmatic subjects’ response to sulfate pollutant aerosols. Ann. Occup. Hyg. 26:691–697. Utell MJ, Morrow PE, Hyde RW (1984) Airway reactivity to sulfate and sulfuric acid aerosols in normal and asthmatic subjects. J. Air Pollut. Control Assoc. 34:931–935. Waldman JM, Lioy PJ, Thurston GD, Lippmann M (1990) Spatial and temporal patterns in summer time sulfate aerosol acidity and neutralization within a metropolitan area. Atmos. Environ. Part B 24:115–126. WHO (2006) Air Quality Guidelines: Global Update 2005. Copenhagen; Denmark: World Health Organization, pp. 448. Wichmann HE, Mueller W, Allhoff P, Beckmann M, Bochter N, Csicsaky MJ, Jung M, Molik B, Schoeneberg G (1989) Health effects during a smog episode in West Germany in 1985. Environ. Health Perspect. 79:89–99. Wichmann HE, Spix C, Tuch T, Wolke G, Peters A, Heinrich J , et al. (2000) Daily mortality and fine and ultrafine particles in Erfurt, Germany. Part I: role of particle number and particle mass. Res. Rep. Health Eff. Inst. 98:5–86. Williams MK (1970) Sickness, absence and ventilatory capacity of workers exposed to sulphuric acid mist. Br. J. Ind. Med. 27:61–66. Wilson WE, Stockburger L (1990)Diurnal variations in aerosol composition and concentration. In: Masuda S, Takahashi K, editors. Aerosols:Science, Industry, Health and Environment. Oxford: Pergamon. pp.962–965. Wilson WE, Koutrakis P, Spengler JD (1991) Diurnal variations of aerosol acidity, sulfate, and ammonia in the atmosphere. Presented at the 84th Annual Meeting and Exhibition, June, Vancouver, BC, Canada. Pittsburgh, PA:Air and Waste Management Association. Paper No. 91-89.9. Wolff GT, Manson PR, Ferman MA (1979) On the nature of the diurnal variation at rural sites in the eastern United States. Environ. Sci. Technol. 13:1271–1276.

1000

SULFUR OXIDES—SO2, H2SO4, NH4HSO4, AND (NH4)2SO4

Wolff RK, Mauderly JL, Pickrell JA (1981) Chronic bronchitis and asthma: biochemistry, rheology and mucociliary clearance. In:Pickrell JA,editor. Lung Connective Tissue: Location, Metabolism and Response to Injury. Boca Raton, FL:CRC Press. pp.169–183. Wong CM, Lam TH, Peters J, Hedley AJ, Ong SG, Tam AY, Liu J, Spiegelhalter DJ (1998) Comparison between two districts of the effects of an air pollution intervention on bronchial responsiveness in primary school children in Hong Kong. J. Epidemiol. Commun. Health 52:571–578. Wong CM, Ma S, Hedley AJ, Lam TH (2001) Effect of air pollution on daily mortality in Hong Kong. Environ. Health Perspect. 109:335–340. Wong CM, Atkinson RW, Anderson HR , et al (2002) A tale of two cities:effects of air pollution on hospital admissions in Hong Kong and London compared. Environ. Health Perspect. 110:67–77. Yu CP, Hu JP, Yen BM, Spektor DM, Lippmann M (1986) Models for mucociliary particle clearance in lung airways. In:Lee SD, Schneider T, Grant LD, Verkerk PJ,editors. Aerosols.Chelsea, MI: Lewis Publishers. pp.569–578. Zmirou D, Schwartz J, Saez M, Zanobetti A, Wojtyniak B, Touloumi G, Spix C, de Leon AP, Le Moullec Y, Bacharova L, Schouten J, Ponka A, Katsouyanni K (1998) Time-series analysis of air pollution and cause-specific mortality. Epidemiology 9:495–503.

26 MICROWAVES AND ELECTROMAGNETIC FIELDS David H. Sliney and Francis Colville

Few occupational or environmental exposures have received more press than electromagnetic fields. Alleged health hazards include possible carcinogenicity, teratogenic effects, and a host of stress-type syndromes. However, when most scientific panels meet to recommend exposure control standards, much of the supporting epidemiological and laboratory data are found to be nonexistent, very weak, or inconclusive. Present standards are based on wellrecognized biological effects of induced currents or heating in exposed biological tissue. Thermal effects dominate at high frequencies, and induced current effects dominate at lower frequencies. Occupational exposure limits (OELs) or guidelines for electromagnetic (EM) fields have been promulgated by different professional, governmental, and standards organizations over the past 30 years. Most attention has been focused on the microwave (0.3–300 GHz) region of the spectrum. Initially, the primary attention paid to this health issue was limited to the communications industry and military establishments, and this was true worldwide. However, with the rapid expansion of microwave cooking ovens in the 1970s, and portable cellular devices since the 1980s, many other organizations became interested in the question of potential health hazards from EM fields (Schwan, 1982; NCRP, 1993; Harlen, 1982; ICNIRP, 1998). The possibility of occupational health hazards associated with the use of radiofrequency (RF) inductive heaters and heat sealers led standards setting groups to re-examine the frequency range below 100 MHz. Their concern was focused on the well-known ability of devices at these low frequencies to produce localized heating in biological tissue. For example, RF medical diathermy is a widely used therapeutic modality. Through coupling of RF energy (usually at 13.5 and 27 MHz), it produces a significant temperature rise directly into a selected region of the body (Hill, 1989). Coincidentally, the EM bioeffects community’s interest grew in the specific absorption rate (SAR) of the human body, particularly as affected by its size relative to the wavelength Environmental Toxicants, Third Edition Edited by Morton Lippmann Copyright
2009 John Wiley & Sons, Inc.

1001

1002

MICROWAVES AND ELECTROMAGNETIC FIELDS

of the incident EM energy. This is known as the “resonance” absorption phenomenon. The practical extent of the frequency region for this to occur is 30 to 70 MHz (for a standing adult). This resonance region of enhanced absorption of EM energy shifts slightly to shorter wavelengths for a seated individual or for a child. In addition, there is a growing concern today about possible adverse effects of exposure to even lower frequency EM energy associated with video display units, electric blankets, and power transmission lines. This has raised the question of whether there is a need for standards down to the 50/ 60 Hz region even though thermal damage mechanisms are not being suggested at these frequencies. Limits are also proposed, or being considered, at even lower frequencies (below 10 Hz), to prevent induced pseudodirect currents in the body. * Note: The opinions or assertions herein are those of the authors and do not necessarily reflect the official position of the Department of the Army or Department of Defense. At these frequencies, the body can misinterpret them as direct currents and interact adversely with the very low direct-current levels functioning in the neurological circuits of the central nervous system. The exposure limits (ELs) being considered in the extremely low frequency (ELF) region (see Table 26.1) are several orders of magnitude higher than those based on thermal effects at 10 kHz. As a result of the interest in lower frequency exposures, concerns that had once been limited to occupational exposure have expanded to environmental exposure. Thus, some groups have demanded environmental exposure limits (EELs) (Bernhardt, 1979, 1986). Previously, most EM standards setting groups in the United States and around the world had concentrated their attention on the regions above 10 kHz. The consensus was that the SAR of biological tissue was so low that bioeffects at lower frequencies would not be of concern. Currently, the issue has been increasingly focused on much lower frequencies (Schwan, 1982; NCRP, 1993; Harlen, 1982; UNEP/WHO/IRPA). The issue of potential adverse health effects of EM fields has sometimes been clouded by special interests concerned with community appearance. The siting of large structures such as microwave communication, radar, high-voltage power lines, and cellular phone towers has been their main concern. Their effort has been to block the construction of what were deemed “unsightly” structures. However, cellular phone towers are being installed in both residential and business localities. Some community groups and legal counsel have sought to convince authorities that the mere possibility of adverse health effects should prohibit such TABLE 26.1 Electromagnetic Energy Bands and Their Associated Frequency and Wavelength Ranges Bands Extremely low frequency Superlow frequency Ultralow frequency Very low frequency Low frequency Medium frequency High frequency Very high frequency Ultrahigh frequency Superhigh frequency Extremely high frequency

Frequency

Wavelength (m)

1–30 Hz 30–300 Hz 300 Hz–3 kHz 3–30 kHz 30–300 kHz 300 kHz–3 MHz 3–30 MHz 30–300 MHz 300 MHz–3 GHz 3–30 GHz 30–300 GHz

108–107 107–106 106–105 105–104 104–103 103–100 100–10 10–1 1–0.1 10 1–10 2 10 2–10 3

BACKGROUND

1003

construction near inhabited areas, even when present exposure standards would not be exceeded. In such instances, lawyers and journalists began to debate the scientific conclusions of specific biological studies, often without the rigor or balance that is customary in normal scientific research. The task for the health professional to maintain a balanced judgment then becomes a special challenge. 26.1 BACKGROUND Before discussing ELs for EM fields, it is necessary to briefly review the physical units and quantities used to express such limits. Electromagnetic waves are periodic and are characterized by their frequency and wavelength. Frequency is the wave’s rate of generation, and wavelength is the distance from the start of one cycle to the start of another. Frequency is generally expressed in hertz (Hz, cycles per second). The Greek symbol l (lambda) represents the wavelength, normally in meters. The relationship of the frequency and wavelength for all EM waves depends on the speed of light, such that frequency (f in hertz) multiplied by the wavelength (l in meters) is equal to the speed of light (c in meters per second): fl ¼ c

or

f ¼ c=l or l ¼ c=f :

The EM spectrum is divided into several regions exhibiting common properties. Each region is designated in terms of wavelength or frequency band. Each band in the EM spectrum is additionally identified by a name, as shown in Table 26.1, for frequencies less than 300 GHz (Dolezalek, 1990). Electromagnetic field intensities are characterized in two ways: either as power density (power per unit area) or as a field strength component (voltage or current per unit length). The most conventional unit for expressing power density is milliwatts per square centimeter (mW/cm2) and for field strength volts per meter (V/m) for an electric (E) field and amperes per meter (A/m) for the magnetic (H) field. The selection of a measurement quantity is determined by the kind of EM field that is being measured and its location in space. If the EM field is coupled to a free radiating wave, then the E and H field vectors of the radiating wave are at right angles to each other and at right angles to the direction of propagation (so-called transverse EM propagation mode, TEM). In this region, fields decrease proportionately with distance following the inverse square law. With a TEM wave, the E and H field vectors are constantly related by the impedance of the propagation medium (120p or 377 W for free space). For this reason, the E or H field vectors can be measured to characterize the EM energy at the wave front. Normally, the E field is measured, and because we know that E/H ¼ 377 W, the power (E  H) can be automatically calculated and presented on a meter as power density. In the radiating field condition, either field strength component or the power density can be determined by a simple field component measurement. Power density for the radiating field condition may be calculated from the output power, radiating frequency, the antenna gain, and the distance to the point of interest. In the radiating field region, EM energy transfer into the tissue is based on the power density of the wave front and the cross-sectional area of the tissue exposed (IEEE, 2005; Johnson and Jasik, 1984). Situations in which the E and H fields are not related by a constant factor require independent measurements of both the E and H fields. The measurement of only one field vector cannot characterize the EM energy available at that point in space. This most often

1004

MICROWAVES AND ELECTROMAGNETIC FIELDS

occurs in the region close to an antenna element or an open transmission line. This area is called the reactive near-field region (usually closer than l/2p from such a source). In this reactive near-field region, energy transfer occurs by a mechanism called capacitive or inductive coupling. In the reactive near-field region, both the E and H fields must be measured to determine exposure levels.

26.2 PHILOSOPHICAL APPROACHES OELs are intended for a narrow segment of the population who are involved in working with equipment or machinery that produce EM energy. Some examples of this type of workforce include military personnel and police officers who use radar guns and medical technicians who use diathermy devices. In the United States, OEL and EEL guidelines for EM fields were first developed by the American National Standards Institute (ANSI) and the Institute for Electrical and Electronic Engineers (IEEE) in ANSI/IEEE C95.1 (1992). A number of national and international groups have recommended human exposure limits (HELs) for RF radiation (i.e., microwaves and lower frequency radiant energy). In the United States, the American Conference of Governmental Hygienists (ACGIH) issues HELs, which it refers to as threshold limit values (TLVs) and these are issued yearly, so that there is an opportunity for a yearly revision (ACGIH, 2006; Sliney et al., 1985; Sliney et al., 2000). The IEEE Standards Committee, known as the “International Committee on Electromagnetic Safety,” also issues standards, which have in the past been adopted as ANSI Standards after additional procedures (ANSI, 1999; IEEE, 2005). The current OEL and EEL guidelines for EM fields were developed in IEEE C95.1 (2005). ACGIH (2006) differs only slightly in its OELS and EELs. The IEEE (2004) expanded its frequency range down to 3 kHz and modified its OELs and EELs to harmonize them with the IEEE 0–3 kHz standard. The ACGIH has traditionally adopted the ANSI/IEEE standards for inclusion in the ACGIH TLVs, and it is expected that they will adopt the 2005 IEEE document in the next revision of their standard. The International Commission on NonIonizing Radiation Protection (ICNIRP), a successor to the International Radiation Protection Association’s International Non-Ionizing Radiation Committee (IRPA/INIRC), has developed OEL and EEL guidelines that differ slightly from the current IEEE standards. The biological database required to derive the ELs was the same for the occupational and environmental conditions. The differences in the standards and guidelines are based on the level of awareness of the population about the EM energy around it. OELs are invoked for personnel who are aware of the potential of EM exposure as a concomitant of employment who have to traffic an EM area. EELs address the general population, those individuals who have no knowledge or awareness of an EM environment. In considering the need for an exposure standard or guideline for the E or H fields of ELF to very low frequency (VLF) EM fields, one must first question whether sufficient knowledge of potential injury thresholds exists. The promulgation of an EL standard suggests that even more knowledge is available for a guideline. In addition, promulgation of a standard often has the effect of curtailing much support of further biological studies necessary to determine thresholds of injury and interaction mechanisms. The opponents of standard setting have argued that history can show that premature standards stifle the technological and bioelectromagnetic studies that lead to the understanding of injury thresholds and mechanisms. In the HF frequency region there has been a lack of knowledge about any interaction mechanisms, other than thermal, for biological systems. Aside from such thermal effects

STANDARDS DEVELOPMENT

1005

and RF shock and burns, any other potentially adverse effect is purely speculative and subject to current research. At these frequencies, even extremely high external fields induce only minute electrical currents within the body. The magnitude of these currents can be calculated (Bernhardt, 1986); however, the biological effects, either helpful or adverse, of such currents are still being debated. It is worth noting that some early OEL proposals for static magnetic fields originated from engineers and scientists who were attempting to allay concerns about potential hazards. They proposed working levels that could readily implemented in industry and the military environments without disrupting current workforce operations (Sliney, 1985). Alternatively, in some countries, such as those in the former Soviet Union, the dynamics of standards setting led to more conservative OELs. Among the several reasons, it has been alleged that very conservative levels were supported by workers’ organizations to obtain extra benefits if they worked in “hazardous environments.” Different philosophies of setting standards in the East and West also explain much of the historical basis for the variation in recommended OELs and EELs despite the use of the same biological database (GOST, 1982; ACGIH, 1990; Sliney et al., 1985; Minnin, 1962; Magnuson et al., 1964). For example, it has been argued that the Soviets had set OELs at which no known biological effects of potential concern existed, and generally well below any adverse effect levels. Indeed, it was argued that the Soviets, in practice, allowed exposure above their guidelines, since they knew that it was not seriously hazardous (Sliney et al., 1985; Minin, 1962; Magnuson et al., 1964). By contrast, the United States and most Western countries had set limits closer to known thresholds for injury. Exceeding a threshold exposure level will more likely result in injury. Hence, more serious protective efforts than those in the East are made to limit exposures levels to below the EL (Sliney et al., 1985; Minin, 1962). It is important to add, however, that the ELs in the West do not actually border the threshold of known injury level but are usually several-fold, even 10 fold, lower than any known risk of adverse health effect. Health scientists from different countries differ on these interpretations, but it is good to remember that one may be comparing “apples with oranges” when simply comparing numerical values of OELs or EELs published by different countries. At an international symposium devoted to the setting of OELs (Copenhagen, Denmark, 23–26 April 1985), the philosophies of setting health standards were reviewed. Although most of the presentations at that symposium related to setting limits for airborne chemical contaminants, the proceedings of the symposium are worthwhile to review to obtain a broad perspective on this issue (ACGIH, 1985).

26.3 STANDARDS DEVELOPMENT Over the last decade, a quest for orderliness and completeness led the standards setting community to extend earlier RF limits into the HF to VLF region and even the ELF and superlow frequency (SLF) frequency region. The SARs are generally considered insignificant at these frequency regions without direct contact with a current-carrying conductor. There were few or no biological data on which to base a standard in these frequency regions (ICNIRP, 1998; ACGIH, 2006). Some industrial and user groups, however, favored EL development for HF to VLF regions to calm the concerns from users who had argued, “if there are no standards for safe exposure, then any exposure should be considered potentially hazardous.” These proponents for standards felt that any limit could be lived with and would be better than allowing a vacuum to exist.

1006

MICROWAVES AND ELECTROMAGNETIC FIELDS

The OELs for lower frequencies required separate treatments of the E fields and the H fields, since almost all potential exposures of individuals were within the reactive field where a simple power-density specification was not meaningful (Schwan, 1982; Harlen, 1982; ICNIRP, 1998, 2006). During the past 15–20 years, standards committees of governmental and professional organizations (e.g., the Committee on Threshold Limit Values for Physical Agents of ACGIH) have periodically reviewed the state of knowledge regarding the biological effects of ultralow frequency (ULF), SLF, and ELF fields and pulsed E or H fields with an eye on establishing EL guidelines and relevant control measures. However, until this decade, all these groups consistently came to the conclusion that there was an absence of scientific information pointing to pathological or other adverse effects on which to base such OELs or EELs (Harlen, 1982; ICNIRP, 1998; ACGIH, 2006; Repacholi, 1985; Sliney, 1985). Even at the very lowest frequencies (less than 10 Hz), adverse physiological effects of currents induced in the human body are believed to occur only at levels higher than those ever likely to be experienced in a realistic environment (Bernhardt, 1979, 1986). ACGIH proposed revised EM TLVs in 1982 that extended, for the first time, into the VLF region, and TLVs for the ELF region were first proposed in 1990. These TLVs extended down to 1 Hz, and this low-frequency extension was prompted mostly by what the committee considered to be unwarranted concerns about health hazards from VDTs. It was felt that TLVs based on theoretical biophysical principles and limited biological data would be well above those produced by most electrical and electronic equipment and thereby put to rest many unwarranted concerns. The biophysical principle used was to keep TLVs low enough at the frequencies involved so that the induced currents would be lower than those that exist naturally in the body. This same principle was used initially to set the thermal SAR limit at higher frequencies. That earlier application of the principle fixed a standard (0.4 W/kg SAR) that is still considered to be conservative 35 years later (ACGIH, 1982). In 1980 and 1983, ACGIH published a statement that insufficient knowledge existed to set guidelines for exposure to static or slowly varying H fields, but in 1987 and 1990, ACGIH published Notices of Intent to Establish TLVs for both ELF and static H fields, and these have been only slightly modified since then. The ELs are based in large part on animal studies and from data collected from human accidental injuries (Schwan, 1982; van der Rosen, 2006; ACGIH, 2006; Sliney et al., 1985; Sliney et al., 2000; ANSI, 1999; IEEE, 2005). A number of studies of ocular effects have been carried out, and the data were applied when deriving the ELs. Only excessively high radiated fields could produce sufficient heating of ocular tissues to produce cataracts or other effects—and then only at microwave frequencies where tissue absorption is far greater than at 400–500 kHz (Schwan, 1982; ACGIH, 2006; Sliney et al., 1985; ANSI, 1999; IEEE, 2005; Elder, 2003). The ELs also have an underlying assumption that sufficient energy must be absorbed to produce a hazardous condition for the eye (Elder, 2003). The ELs for RF radiation exposure are normally identical to those of the IEEE Standard C95.1 (2005); however, some updates in the TLVs were planned for 2007 (ICNIRP, 1998; Matthes et al., 1999; UNEP/WHO/ICNIRP, 1993; NRPB, 1994; EU, 1999; NCRP, 1993; IEC, 2005). 26.3.1

Current OELs and EELs for RF Radiation

Tables 26.2 and 26.3 show the principal, well-known EM OELs published in the U.S. literature. Tables 26.4 and 26.5 show the EM EELs for the United States. The difference

STANDARDS DEVELOPMENT

TABLE 26.2

1007

IEEE and ACGIH Occupational Exposure Limits IEEE OELs RMS Electric Field Strength (E) (V/m)

Frequency Range (MHz) 0.1–1.0 1.0–30 30–100 100–300 300–3 000 3 000–15 000 15 000–300 000

RMS Magnetic Field Strength (H) (A/m)

RMS Power Density (S) E-Field, H-Field (mW/cm2)

Averaging Time (min)

16.3/fM 16.3/fM 16.3/fM 0.163

(9000, 100 000/fM2) (9000/fM2, 100 000/fM2) (10, 100 000/fM2) 1.0 fM/30 100 100

6 6 6 6 6 19.63/fG1.079 2.524/fG0.476

1842 1842/fM 61.4 61.4

fM is frequency in MHz; fG is in gighertz. These plane-wave equivalent power-density values are commonly used as a convenient comparison with OELs at higher frequencies and are displayed on some measuring instruments.

between the two ELs is the level of awareness of the personnel. Figure 26.1 provide the log– log representation of the OELs and EELs of ACGIH, IEEE, and ICNIRP expressed as equivalent far-field (radiating field) power densities. Figures 26.2 and 26.3 provide the log– log representations as a function of frequency of the E field and the H field, respectively. The low-frequency region (less than 3 MHz) in Fig. 26.1 has rather large ELs for all three standards. This is due to the poor absorption efficiency of the human body at low frequencies. As stated earlier, most of the energy is sensed as a form of neurological stimulation. The region between approximately 30 and 3 GHz defines the SAR boundaries. This is the frequency band where the human body most efficiently absorbs EM energy. Therefore, the ELs in this region are significantly lower to compensate for this condition.

TABLE 26.3

ICNIRP Occupational Exposure Limits ICNIRP OELs

Frequency Range (MHz) Up to 1 Hz 1–8 Hz 8–25 Hz 0.025–0.82 kHz 0.82–65 kHz 0.065–1 MHz 1–10 MHz 10–400 MHz 400–2,000 MHz 2–300 GHz

Electric Field Strength (E) (V/m) – 20,000 20,000 500/f 610 610 610/f 61 3f 1/2 137

Magnetic Field Strength (H) (A/m) 1.63  105 1.63  105/f 2 2  104/f 20/f 24.4 1.6/f 1.6/f 0.16 0.008f 1/2 0.36

Power Density (S) E-Field, H-Field (W/m2) – – – – – – – 10 f/40 50

f s indicated in the frequency range column. For frequencies between 100 kHz and 10 GHz. S, E, and H are averaged over a 6 min period. For frequencies exceeding 10 GHz. S, E, and H are to be averaged over any 68/f 1.05 min period (f in GHz).

1008

MICROWAVES AND ELECTROMAGNETIC FIELDS

TABLE 26.4

IEEE Environmental Exposure Limits IEEE EELs

RMS Electric RMS Magnetic Field Strength Field Strength Frequency (H) (A/m) (E) (V/m) Range (MHz) 0.1–1.34 614 1.34–3.0 823.8/f 3–30 823.8/f 30–100 27.5 100–400 27.5 400–2000 2000–5000 5000–30 000 30 000–100 000 100 000–300 000

16.3/f 16.3/f 16.3/f 158.3/f 1.668 0.0729

RMS Power Density (S) E-Field, H-Field (mW/cm2) (1000, 100 000/f 2)2 (1800/f 2, 100 000/f 2)2 (1800/f 2, 100 000/f 2)2 (2, 9 400 000/f 3.336)2 2 fM/200 10 10 10 (90fG–7000)/200

Averaging Time |E|2, |H|2, S 66 f 2/0.3 6 30 6 30 0.0636f 1.337 30 30 30 30 150fG 25.24/f G 0.476 5048/[(9fG–700) fG0.476]

f is frequency in MHz. These plane-wave equivalent power-density values are commonly used as a convenient comparison with OELs at higher frequencies and are displayed on some measuring instruments.

In the region above approximately 3 GHz, most thermal energy is deposited on or near the skin. There is very little penetration of EM energy into the body. The IEEE, the ACGIH, and the ICNIRP all have about the same ELs in the middle (SAR) region. In the upper frequency region, the ANSI/IEEE and the ACGIH closely match each other and both have adopted the similar EM ELs while the ICNIRP EL is set at half the IEEE/ACGIH ELs. The IEEE and ACGIH EELs are set at one-fifth of the OELs. This is an additional level of protection provided to the general population. It is not based on any additional scientific studies or TABLE 26.5

ICNIRP Environmental Exposure Limits ICNIRP EELs

Frequency Range (MHz) up to 1 Hz 1–8 Hz 8–25 Hz 0.025–0.8 kHz 0.8–3 kHz 3–150 kHz 0.15–1 MHz 1–10 MHz 10–400 MHz 400–2,000 MHz 2–300 GHz

Electric Field Strength (E) (V/m) – 10,000 10,000 250/f 250/f 87 87 87/f1/2 28 1.375f1/2 61

Magnetic Field Strength (H) (A/m) 3.2  104 3.2  104/f2 4 000/f 4/f 5 5 0.73/f 0.73/f 0.073 0.0037f1/2 0.16

Power Density (S) E-Field, H-Field (W/m2) – – – – – – – 2 f/200 10

1. f as indicated in the frequency range column. 2. For frequencies between 100 kHz and 10 GHz. S, E, and H are averaged over a 6 min period. 3. For frequencies exceeding 10 GHz. S, E, and H are to be averaged over any 68/f 1.05 min period (f in GHz).

STANDARDS DEVELOPMENT

1009

FIGURE 26.1

Occupational exposure limits and environmental exposure limits for power density.

FIGURE 26.2

The Occupational exposure limits and environmental exposure limits for electric fields.

1010

MICROWAVES AND ELECTROMAGNETIC FIELDS

FIGURE 26.3

Occupational exposure limits and environmental exposure limits for magnetic fields.

investigations, but on the premise that the general population requires an extra margin of safety due to their lack of EM awareness. The IEEE, the ACGIH, and the ICNIRP guidelines all provide an important caveat: higher levels are permitted if one wishes to go to the trouble of calculating the actual SAR and showing that an individual is actually absorbing less than an average SAR of 0.4 W/kg. This permits a user to follow much less stringent limits; however, safety procedures must be followed to limit “startle reaction” to RF or shock from large ungrounded metal structures such as truck or school bus bodies immersed in the field (IEEE, 2005; ACGIH, 2006; Repacholi, 1985; DIN, 1984; ICNIRP, 1998).

26.4 CURRENT DEVELOPMENTS 26.4.1

The Lowest Frequencies

All of the initial EM standards concentrated on the microwave radiation frequencies where biological effects could be clearly demonstrated as a result of overt heating of tissue (Schwan, 1982). As time passed, OELs at lower frequencies were promulgated. Concerns about potential adverse effects of RF heat sealers forced most groups to recommend standards or guidelines in the frequency range down to 10 MHz or lower, and essentially all standards reach down to 100–300 kHz (IEEE, 2005; ACGIH, 2006; ICNIRP, 1998). As seen in Fig. 26.2, EELs are more stringent for the E field because the E field-induced currents were thought to account for more absorbed power than the H field at lower frequencies (Bernhardt, 1979, 1986; NRPB, 1988; ICNIRP, 1998; Ghandi et al., 1980; Deno, 1974). This distinction was also made by IEEE.

CURRENT DEVELOPMENTS

1011

The limits at 100 kHz can vary by orders of magnitude among the all standards. This can be explained by the significant lack of biological data showing any adverse effect on which to base a limit more realistically. At frequencies below 100 KHz, there is virtually no absorption of energy in the human body. This led to the ever-ascending OEL and EEL of the standards; however, concerns about “startle-reaction” RF shocks led ANSI/IEEE and ACGIH to place a cap on their ELs (ANSI/IEEE, 1992; ACGIH, 1997). The IEEE standard for safety levels with respect to human exposure to electromagnetic fields in the frequency range 0–3 kHz (2004) was adopted to address OEL and EEL concerns in this region. The ICNIRP standard (ICNIRP, 1998), as shown in Tables 26.3 and 26.5 are the result of certain modifications of the IRPA/ INIRC (1984) and IRPA/INIRC (1988) standards. Although the basic limit of an SAR of 0.4 W/kg over the whole body was retained, the SAR limit is now 2 W per 0.1 kg for partialbody exposure of the extremities (hands, feet, etc.). All SAR values are averaged over any 6 min. These partial-body SAR limits overcome the possibly excessive SARs that could occur for exposures in small regions (wrists or ankles) of the body. Below 10 MHz, it is found that the EM H field limit between 0.1 and 10 MHz now remains at the same value. 26.4.2

Us Government Activities

In 1996, the Federal Communications Commission (FCC) circulated proposed EM ELs to other governmental agencies and outside scientific experts for comments. These proposals were somewhat different and in some cases more conservative than those used by ANSI/ IEEE, ACGIH, or OSHA. Both OSHA and ACGIH had adopted the ANSI/IEEE standards. The written reviews included considerable criticism from various government agencies, including the Department of Defense (DOD), which had already adopted the ANSI/IEEE standards. The FCC draft standards included parts of the ANSI/IEEE standards along with parts of the National Council on Radiation Protection (NCRP) standards. The primary objections to FCC standard was that it was not sufficiently comprehensive and it was not a general consensus standard. Despite its shortcomings, the FCC guidelines were accepted and released on August 1, 1996. Since these guidelines were accepted by the FCC and published in the Federal Register, there is expected to be increased pressure on the National Telecommunications and Information Administration (NTIA) to require other federal agencies to comply with the new FCC guidelines. 26.4.3

Alternatives To OELs: Do They Exist?

Some occupational health professionals argue that there appears to be little or no reason for concern regarding conventional occupational exposure to most EM transmitter sources operating in the HF bands (other than RF heat sealers). They ask whether there is any justification for protective measures. In the absence of knowledge, one can always encourage avoidance of needless exposure and then use protective measures to reduce human exposure where practical and where little added cost is involved. The problem with this approach is that a controversy ensues as to what is practical, realistic, and reasonable in terms of added cost. One needs only observe the impact of recent developments from a similar approach tried in the field of ionizing radiation protection to see the potential problem. In the ionizing radiation protection field, the idealistic philosophy of as low as reasonably achievable (ALARA) was inserted into regulations. Many legal and regulatory controversies ensued. The ALARA principle unsettled the orderliness provided by the existence of a fixedvalue EL. The fundamental problem is that almost anything is achievable by engineering

1012

MICROWAVES AND ELECTROMAGNETIC FIELDS

means and enough financial support; however, everyone differs with regard to what is “reasonable.” The lesson here is that idealistic goals should not be regulated. For the reasons illustrated above, many health professionals welcomed the promulgation of VLF to HF OELs as an alternative to reduce the level of controversy, even though little scientific basis for the standard truly existed. When control measures are employed, most are inexpensive. An exception has been on naval vessels where many antennae exist near walkways. It is worth noting that to reduce electromagnetic interference (EMI), other electronic means such as shielding, and signal reduction have often been employed in the past, independent of any motivation for reducing potential health hazards. 26.4.4

Epidemiological Studies

As noted earlier, there has been a degree of controversy surrounding the interpretation of epidemiological studies that appeared to relate the incidence of childhood and other cancers with low-frequency (60 Hz) EM fields. One of the most referenced early studies suggesting this correlation was that of Wertheimer and Leeper (1979). This study was highly criticized, and a re-evaluation at the same location with an eye on reducing confounders still showed a weaker correlation with increased cancer risk (Savitz et al., 1988). The critics argued that many other factors (confounders) may explain these and other findings; for example, heavier roadway traffic densities near the same areas where wiring codes would indicate higher currents. A more recent study by the National Research Council (NRC, 1998) revisited the controversy of power lines causing certain types of cancers. The NRC concluded that there is no clear, convincing evidence to show that residential exposures to electric and magnetic fields (EMFs) are a threat to human health. The NRC also tried to duplicate the 1979 findings linking EMFs to childhood leukemia. Although the NRC found a weak but statistically significant correlation between childhood leukemia (which is rare) and electrical wire configurations, it could not determine if the correlation was based on another confounder or EMF. It has never been demonstrated that this apparent association was caused by exposure to electromagnetic fields. Controversy can be expected to continue with epidemiological studies such as those appear to show correlations between illness and exposure despite a lack of theoretical basis for linking a given disease to EM fields (Lin et al., 1985; Logue et al., 1985). Even the studies themselves concluded that the correlation was uncertain (Wertheimer and Leeper, 1979; Fulton et al., 1980; Juutilainen et al., 1990). Most studies like these are limited by their exposure assessment and control of factors other than EM fields as a cause for cancer. Using similar controls and configurations as those that find possible correlations, other studies have found “no relationship between leukemia and power lines” (Fulton et al., 1980). Although conclusions as to whether EM fields have adverse or positive health effects have not been achieved, studies do support the need for more research, both epidemiological and theoretical (Savitz et al., 1988; Monson, 1990; Juutilainen et al., 1990).

26.5 PROTECTIVE MEASURES Protective measures for industrial and scientific EM exposure control can be conveniently separated into three categories: the use of engineering, installation, or system-design controls to keep the exposure threat away from people; the use of range controls, that is, the principle of using separation or distance to keep the people away from the threat; and the

PROTECTIVE MEASURES

1013

use of administrative controls to assure that all of the EM control measures are understood and observed. A fourth category of exposure control involves the issue of personal protective equipment such as special garments, goggles, or face masks to protect those who must enter intense fields (e.g., to replace aircraft warning lights on active broadcast towers or to repair online military broadcast antennae). These have been tried in the past (Czerski et al., 1974), but are generally too cumbersome for continuous use or have proven ineffective against the pervasiveness of the EM fields at the lower frequency (broadcast) portion of the spectrum. Insulated gloves or similar outer clothing can protect against EM shock such as from physical contact with an active emitter or physical contact with a metallic body (fence, tower, vehicle, etc.) carrying induced EM energy from high-power sources. Such insulated protection is commonly used around antennae and open transmission lines. The effect of damaging levels of induced currents in metallic surgical and dental implants or implanted electronic devices (e.g., cardiac pacemakers) is also important to consider in assessing the need for “control measures.” 26.5.1

Engineering/Installation/Design Controls

The most dependable control measures for providing protection against exposure are those that isolate the EM energy from people. In the original design of a device that incorporates an EM source, interlocks can be included to control the direction and intensity of radiation. Also, components should be selected (e.g., coaxial cable versus) that reduce the possibility of inadvertently leaking EM energy into areas that might be occupied. Installation control involves mounting radiating systems on inaccessible places like towers or rooftops. Electromechanical limit switches should be designed into the EM beam-directing mechanisms of radar systems to ensure that the energy does not radiate into occupied places. Careful design of industrial RF heat sealers can greatly reduce potential exposure to the operator. Another engineering control method involves the use of metal enclosures (such as a Faraday cage) to reduce stray E fields around sources that emit EM energy. External enclosures of ferromagnetic materials can also “capture” magnetic flux lines and reduce external magnetic flux densities around some low-frequency equipment. Such shielding measures applied to large equipment are expensive and are primarily designed to reduce the effects of EM interference on sensitive industrial, scientific, or medical instrumentation. Some basic form of such shielding, usually at the component level, is required for all electronic devices that produce EM emissions (computers, TVs, radios, remote controllers, etc.) to ensure that normal EM emissions from such devices are not a source of EMI to other nearby electronic systems. These EMI controls are mandated by the FCC, and although they are not designed for personnel exposure control, these controls do provide assurance that the device does not and will not exceed certain FCC-approved EM emission levels during its lifetime. Responding to the heightened concerns among VDT users regarding EM exposure, some computer manufacturers have included magnetic shielding in their VDT designs. These shields are supposed to reduce EM levels even lower than the levels required by FCC, stating that it was a business decision rather than one based on any serious concern borne of actual existing health threats. Some electric blanket manufacturers have redesigned the heating-wire routing in their products to minimize stray H fields, also for business reasons. The modifications are designed to position the wires so that the alternating EM fields’ negative minimum and positive maximum cancel one another. This results in local minima that approach zero field

1014

MICROWAVES AND ELECTROMAGNETIC FIELDS

strength within the exposure area. Recent studies on adverse effects of electric blanket exposure have been inconclusive. Any EL considered for their frequency region has recommended controls for exposure levels greater than those produced by electric blankets. However, the public’s perceived threat of EM fields from electric blankets is economically real for manufacturers (Florig and Hoburg, 1990; Verreault et al., 1990). 26.5.1.1 Range Controls If the EM energy cannot be kept entirely away from inhabited areas by means of engineering/design/installation, then ways must be found to keep people away from the source. Because there is always a decrease in EM energy level as the separation (distance or range) from the source is increased, the exposure control can be accomplished by keeping the people far enough away from the source so that the EM level is below the OEL or EEL. Fences or other barricades can be erected to prevent personnel entry into areas where the EM levels exceed the OELs or EELs. Simple warning signs are adequate if the EM levels are only marginally excessive. “Walk-through” areas can be designated (painted on roofs or decks, walkways, etc.) around certain EM sources, where the maximum exposure will not be an “overexposure,” if the time spent in the overexposure zone is controlled (thus “walk-through,” as opposed to areas where stopping or other delay is permitted). Warning lights, buzzers, horns, and so on are also effective in alerting people to the need for control and the necessary distances around high-power sources that need to be controlled. 26.5.1.2 Administrative Controls As the name suggests, this category is always a necessary part of exposure control. The types of administrative controls include all of the management, information, and documentation, features of the protection program. Some key elements are the inventory of all controlled items, a list of people who are known to be at risk of overexposure, the record of all required personnel controls, the standing operating procedures specified for controlling each source, identification of the persons responsible for administering the programs, and procedures to follow in case of a suspected overexposure. Individuals with special needs, such as those with medical conditions or electronic implants, must be identified and warned about any threat to them from entry into areas of high EM field strengths. Special initial and routine EM safety briefings should be required, to inform all appropriate persons of the exposure threats and the control programs. Normally, records need to be kept of persons attending such briefings and of the materials covered. In the end, there can be no substitute for properly informing and training those personnel who must work around sources of potential EM overexposure, whether whole systems or portions of systems, such as antennae, transmission lines, transmitters, and so on. The need for all controls intensifies in proportion to the degree of the overexposure threat.

26.6 CONCLUSIONS Finally, comprehensive standards for controlling either occupational or general public exposure in all frequency bands of EM energy are not considered likely in the near term. Reasonable agreement (within an order of magnitude) exists for exposure control in the microwave frequency region, at least for short-term exposures. Only interim guidelines are realistic regarding possible adverse health effects associated with exposure to VLF and ELF fields. There are still too many unanswered questions for the standards setting community to

GLOSSARY

1015

define a level and say definitively that above that level all effects are adverse, or even below that level all effects are not adverse. Even the choice of parameters to be controlled is uncertain, whether it should be E or H field strength, duration of exposure, kinds of exposure (short pulse, continuous wave, pulse trains, peak field limits, duty cycle limits, etc.), or combinations of some or all of these. The guidelines that are available at this time were based on dependable physiological principles. Also, these guidelines are acceptable in present-day occupational and environmental settings; that is, the interim guidelines can be enforced without requiring significant change in most installations. The guidelines should be applied and monitored continuously against ongoing developments in the bioelectromagnetics and epidemiological research communities. Unless some special need exists for relief in the OEL guidelines (e.g., to permit “hot” work on 60 Hz high-voltage transmission lines), change in the direction toward more relaxed limits should not be encouraged. If, however, scientific research and dependable epidemiology demonstrate a need for a more conservative standard, then that change should be made at that time. U.S. governmental ELs have been proposed by the EPA and NIOSH, but because of some controversies, these may not become official or binding in the near future, if ever. Even without such standards, practical applications and exposure situations (radar and communications operators, and certainly VDT users) for actual human exposure levels are very low. By comparison with RF levels used in biological experiments, one can safely assure the user that there are no known adverse effects from such uses based on the available biomedical knowledge and benefit–risk estimates (Cox, 1980; Weiss and Peterson, 1979; Wolbarsht et al., 1979; Stuchly et al., 1983).

26.7 GLOSSARY EELs Environmental exposure limits are for those areas where access is not controlled to exclude persons less than 140 cm (55 in.) in stature—common ground where any individual may be found. The RF exposures in these areas do not exceed the permissible exposure limits. Generally, these locations represent living quarters, workplaces, or public access areas where personnel would not expect to be exposed to the level of RF and energy. E field Electric field, a fundamental component of electromagnetic waves that exists when there is a voltage–potential difference between two points in space. Electric field strength (E) The magnitude of the electric field expressed in volts per meter (V/m). EM radiation (fields) Electromagnetic radiation, the propagation of energy in the form of varying electric and magnetic fields through space at the velocity of light. H field Magnetic field, a fundamental component of electromagnetic waves produced by a moving electric charge. Magnetic field strength (H) The strength of the magnetic field expressed in amperes per meter (A/m). OELs Occupational exposure limits are for those areas to which access is controlled for the purposeofexcludingentryofpersonslessthan140 cm(55 in.)instature.LocationswhereRF exposures may be incurred by workers who knowingly accept potential exposure as a condition of employment or duties, exposure of individuals who knowingly enter areas

1016

MICROWAVES AND ELECTROMAGNETIC FIELDS

where RF exposure is to be expected, and incidental exposure during transient passage throughareaswhereanalysis shows that exposurelevels maybeabovethoseset butnot above the permissible exposure limit. Overexposure Any human exposure to RFR that exceeds the established permissible exposure limit (PEL). Permissible exposure limit The maximum level expressed as either root mean square or peak electric and magnetic field strengths, plane-wave equivalent power densities, or induced body and contact currents, with added safety factor values, to which an individual may be exposed without expectation of any harmful effects. RF Radio frequency. Although the RF spectrum is formally defined in terms of frequency as extending from 0 to 3000 Ghz, commonly it is electromagnetic energy in the frequency region useful for radio transmission. The present practical limits of radio frequency are roughly 10–300 Ghz. SAR Specific absorption rate, the time rate at which RFR energy is imparted to an element of biological body mass. Average SAR in a body is the time rate of total energy absorption divided by the total mass of the body. SAR is usually expressed in units of watts per kilogram (W/kg). SA Specific absorption refers to the amount of energy absorbed over the exposure time and is usually expressed in units of J/cm2.

Q1 Q2

REFERENCES ACGIH (American Conference of Governmental Industrial Hygienists) (1985) Proceedings of an International Symposium on Setting Occupational Exposure Limits, Annals of the ACGIH, Vol. 12. Cincinnati: ACGIH. ACGIH (American Conference of Governmental Industrial Hygienists) (2006) Threshold Limit Values and Biological Exposure Indices for 1998. Cincinnati:ACGIH. ACGIH (American Conference of Governmental Industrial Hygienists) (1997) Threshold Limit Values and Biological Exposure Indices for 1998. Cincinnati: ACGIH. ACGIH (American Conference of Governmental Industrial Hygienists) (1992) Documentation for Threshold Limit Values (TLVs) for Chemical Substances and Physical Agents in the Work Environment with Intended Changes. Cincinnati: ACGIH. ANSI (American National Standards Institute) (1982) American National Standard C95.1, Safety Levels with Respect to Human Exposure to Radio Frequency Electromagnetic Fields (300 kHz–100 Ghz). New York: ANSI. ANSI (American National Standards Institute) (1991) American National Standard Safety IEEE Standard for Levels with Respect to Human Exposure to Radio Frequency Electromagnetic Fields, 3 kHz to 300 GHz. New York: ANSI. Austria (1990) Mikrowellen und Hochfrequenzfelder Zulassige Expositionswert. Vienna: Osterreichisches Normungsinstitut. Bernhardt J (1986) Assessment of experimentally observed bioeffects in view of their clinical relevance and the exposure at workplaces. In: Kaul A, Bernhardt J,editors.Proceedings of the Symposium on the Biological Effects of Static and Extremely Low Frequency Magnetic Fields. Munich: MMV Medizin Verlag. Bernhardt J (1979) The direct influence of electromagnetic fields on nerve and muscle cells of man within the frequency range of 1 Hz to 30 MHz. Rad. Environ. Biophys. 16:309–320.

REFERENCES

1017

Blank M (1995) Electromagnetic Fields: Biological Interactions and Mechanisms. Washington, DC: American Chemical Society Press. Cox EA (1980) Radiation emissions from visual display units. In: Health Hazards of VDUs? Loughborough, England: HUSAT Research Group. Czerski P, OstrowskiK Shore ML, Silverman C, Suess MJ, a Waldeskog B, editors.(1974) Biological effects and health hazards of microwave radiation. Proceedings of an International Symposium, Warsaw, 15–18 October, 1973. Warsaw: Polish Medical Publishers. Deno DW (1974) Calculating electrostatic effects of overhead transmission lines. IEEE Trans. Power Appl. Sys. PAS-93:1458–1471. Deutsche Institut fur Normeln (DIN) (1984) Hazards for electromagnetic fields, protection of persons in the frequency range from 10 kHz to 300 Ghz, DIN 578.848, Teil 2. Berlin: DIN (German Standards Institute). Dolezalek H (1990) Classification of electromagnetic radiation. In: CRC Handbook of Chemistry and Physics. 71st ed. Boca Raton: CRC Press. Durney CH, Massoudi H, Iskander MF (1986) Radiofrequency Radiation Dosimetry Handbook, 4th edn. Report USAFSAM-TR-85-73. Brooks AFB, TX: USAF School of Aerospace Medicine. Elder JA (2003) Ocular effects of radiofrequency energy. Bioelectromagnetics 6:S148–S161. European Community (EU) (July, 12, 1999) Council recommendation on the limitation of exposure of the general public to electromagnetic fields (0 Hz to 300 GHz). OJEC Florig HK, Hoburg Jf (1990) Power-frequency magnetic fields from electric blankets. Health Phys. 58:493–502. FRG (1986) (Federal Republic of Germany), Hazards from Electromagnetic Fields. Protection of Persons in the Frequency Range from 0 Hz to 3000 GHz, Deutsche Norm (New DIN BDE No. 0848, August). Berlin: Beuth Verlag. (In German) FRG (1986) (Federal Republic of Germany), Safety at Electromagnetic Fields: Limits of FieldStrength for the Protection of Persons in the Frequency Range from 30 kHz to 3000 GHz, Deutsche Norm (New DIN BDE No. 0848, Part 4). Berlin: Beuth, Verlag. (In German) Fulton JP, Cobb CS, Preble L, Leone L, Forman E (1980) Electrical wiring configurations and childhood leukemia in Rhode Island. Am. J. Epidemiol. 11:292–296. Ghandi OP, Chatterjee I, Wu D, Gu Y-G (1980) Likelihood of high rates of energy deposition in the human legs at the ANSI recommended 3–30 MHz RF safety levels. Proc. IEEE 68:104–113. Germany (1990) Sicherheit in elektromagnetichen Feldern. Frankfurt: Institut fur Rundfunktechnik. GOST (State Standards Institute) (1982) Occupational Safety Standards System: Electromagnetic Fields of Radiofrequency, General Safety Requirements. GOST-12, 1006–76. Moscow: State Standards Institute (originally issued in 1976 and amended in 1982). Harlen F (1982) Criteria for setting exposure limits for RF and MW. Gig. Ital. Med. Lab. 4:39–48. Hill CR (1989) Physical Principles of Medical Ultrasonics. New York: Ellis Horwood. Hitchcock RT, Patterson RM (1995) Radio-Frequency and ELF Electromagnetic Energies: A Handbook for Health Professionals. New York: John Wiley. IEEE (Institute of Electrical and Electronic Engineers) (2005) IEEE Standard for Safety Levels with Respect to Human Exposure to Radio Frequency Electromagnetic Fields, 3 kHz to 300 GHz. New York: IEEE. IEEE (Institute of Electrical and Electronic Engineers) (1999) IEEE Standard for Safety Levels with Respect to Human Exposure to Radio Frequency Electromagnetic Fields, 3 kHz to 300 GHz. New York: IEEE. IEEE (Institute of Electrical and Electronic Engineers) (2004) IEEE Standard for Safety Levels with Respect to Human Exposure to Radio Frequency Electromagnetic Fields, 0–3 kHz. New York: IEEE.

1018

MICROWAVES AND ELECTROMAGNETIC FIELDS

IEC (International Electrotechnical Commission) (2005) Standard IEC 60601-1, Medical Electrical Equipment—Part 1: General Requirements for Safety and Essential Performance. Geneva: International Electrotechnical Commission. IRPA (1988a) Guidelines on limits of exposure to radiofrequency electromagnetic fields in the frequency range from 100 kHz to 300 GHz. Health Phys. 54:115–123. IRPA (1988b) Alleged radiation risks from visual display units. Health Phys. 54:231–232. IRPA (1990) Environmental Health Criteria Document 16, Revised Version, Electromagnetic Fields in the Range of 300 Hz to 300 GHz. Rome, February 1990. ICNIRP (International Commission on Non-Ionizing Radiation Protection) (1998) Guidelines for Limiting Exposures to Time-Varying Electric, Magnetic, and Electromagnetic Fields (Up to 300 GHz). Health Phys. 74:494–522. ITU. (1982) Radio Regulations. Geneva: General Secretariat of the International Telecommunication Union. Japan (1990) Outline of Japanese Radio Frequency Protection Guide, Organization of Deliberations. Tokyo: Telecommunications Technology Council of the Minister of Posts and Telecommunications. Johnson RC, Jasik H (1984) Antenna Engineering Handbook. 2nd edn. New York: McGraw-Hill. Juutilainen J, Laara E, Pukala E (1990) Incidence of leukaemia and brain tumours in Finnish workers exposed to ELF magnetic fields. Int. Arch. Occup. Environ. Health 62:289–295. Kandioan AG, editor. (1968) Reference Data for Radio Engineers. Itt, Indianapolis: Howard W. Sams. Kitchen R (2001) RF and Microwave Radiation Safety Handbook. 2nd edn. p.448 Lin RS, Dischinger PC, Conde J, Farrell KP (1985) Occupational exposure to electromagnetic fields and the occurrence of brain tumors. J. Occup. Med. 27 (6):413–419. Logue JN, Hamburger S, Silverman PM, Chiacchierini RP (1985) Congenital abnormalities and paternal occupational exposure to shortwave, microwave, infrared, and acoustic radiation. J. Occup. Med. 27(6):451–452. Magnuson HJ, Fassett DW, Gerarde HW, Rowe VK, Henry MS, Smyth F Jr, Stokinger HE (1964) Industrial toxicology in the Soviet Union: theoretical and applied. Am. Ind. Hyg. Assoc. J. 25 (2):185–197. Matthes R, van Rongen E, Repacholi MH, editors. Health effects of electromagnetic fields in the frequency range 300 Hz to 10 MHz. Proceedings of the International Seminar on Health Effects of Electromagnetic Fields in the Frequency Range 300 Hz to 10 MHz, Maastricht, The Netherlands, June 7 and 8, 1999, Munich, International Commission on Non-Ionizing Radiation Protection. McKinlay AF, Anderson JB, Bernhardt JH, Grandolfo M, Hossmann K-A, Mild KH, Swerdlow AJ, Van Leeuwen M, Verschaeve L, Veyret B (1996) Radiotelephones and human health–proposal for a European Commission Expert Group, Brussels, European Commission Directorate General XIII. Minnin BA (1962) Microwaves and Human Safety (in Russian). Moscow: Radio Publication. (Available in translation as Report JPR-655506-1 annnd 0- from NTIS, Springfield, VA). Monson RR (1990) Editorial commentary: epidemiology and exposure to electromagnetic fields. Am. J. Epidemiol. 131:774–775. National Academy of Science/National Research Council (1996) Possible health effects of exposure to residential electric and magnetic fields. Washington, DC: National Academy Press. National Council on Radiation Protection and Measurements (NCRP) (1993) A Practical Guide to the Determination of Human Exposure to Radiofrequency Fields. NCRP Report No. 119. Washington, DC: NCRP. NRPB (National Radiological Protection Board) (1982) Proposals for the Health Protection of Workers and Members of the Public Against the Dangers of Extra Low Frequency, Radiofrequency and Microwave Radiations. Harwell, UK: NRPB.

REFERENCES

1019

NRPB (National Radiological Protection Board) (1986) Advice on The Protection of Workers and Members of the General Public from the Possible Hazards of Electric and Magnetic Fields with Frequencies Below 300 GHz: A Consultative Document. Harwell, Oxfordshire, UK: NRPB. NRPB (National Radiological Protection Board) (1988) Guidance on Standards: Guidance as to Restriction on Exposures to Time Varying, Electromagnetic Fields and the 1988 Recommendation of the International Non-Ionizing Radiation Committee. Harwell, UK: NRPB. NRPB (National Radiological Protection Board) (1994a) Health Effects Related to the Use of Visual Display Units. Report by the Advisory Group on Non-ionising Radiation. NRPB Documents 5 (2). Chilton, UK: National Radiological Protection Board. NRPB (National Radiological Protection Board) (1994b) Electromagnetic fields and the risk of cancer. Supplementary report by the Advisory Group on Non-ionising Radiation of 12 April 1994. Radiat. Prot. Bull. 154:10–12. Polk C, Postow E (1998) Biological Effects of Electromagnetic Fields. 2nd edn. Boca Raton, FL: CRC Press. Repacholi MR (1985) Limits of human exposure to magnetic fields. In: Kaul A, Bernhardt J,editors. Proceedings of the Symposium on the Biological Effects of Static and Extremely Low Frequency Magnetic Fields. Munich: MMV Medizin Verlag. Repacholi MR (1998) Low-level exposure to radiofrequency fields: health effects and research needs. Bioelectromagnetics 19:1–19. Savitz DA, Wachtel H, Barnes FA, John EM, Tvrdik JG (1988) Case–control study of childhood cancer and exposure to 60-Hz magnetic fields. Am. J. Epidemiol. 128:21–38. Schwan HP (1982) Physical properties of biological matter: some history, principles, and applications. Bioelectromagnetics 3:3–14. Sliney DH (1985) Does the basis for a standard exist? In: Kaul A, Bernhardt J, editors.Proceedings of the Symposium on the Biological Effects of Static and Extremely Low Frequency Magnetic Fields. Munich: MMV Medizin Verlag. Sliney DH, Wolbarsht JL, Mue AM (1985) Editorial: Differing radiofrequency standards in the microwave region-implications for future research. Health Phys. 49 (5):677–683. Sliney DH, Colville F (2000) Microwaves and electromagnetic fields. In: Lippmann M,editor. Environmental Toxicants. New York: Wiley-Interscience. pp.577–593. Stuchly MA, Lecuyer DW, Mann RD (1983) Extremely low frequency electromagnetic emissions from video display terminals and other devices. Health Phys. 45:713–722. Tenforde TS (1996) Interaction of ELF magnetic fields with living systems. In: Polk C, Postow E, editors. Biological Effects of Electromagnetic Fields. Boca Raton, FL: CRC Press. pp.185– 230. Ueno S (1996) Biological Effects of Magnetic and Electromagnetic Fields. New York: Plenum Press. UNEP/WHO/IRPA (United Nations Environmental Programme/World Health Organization/International Radiation Protection Association) (1990) Electromagnetic Fields in the Range of 300 Hz to 300 GHz. Rome: WHO. UNEP/WHO/IRPA (United Nations Environmental Programme/World Health Organization/International Radiation Protection Association) (1990) Electromagnetic Fields in the Range of 200 Hz to 200 GHz. Rome: WHO. UNEP/WHO/IRPA (United Nations Environmental Programme/World Health Organization/International Radiation Protection Association) (1993) Magnetic Fields: Environmental Health Criteria. Geneva: World Health Organization.137. USSR (1984b) Temporary Safety Standards and Regulations on Protection of the General Public Against the Effects of Electromagnetic Fields Generated by Radio-Transmitting Equipment, ct, No 2963-84. Moscow, USSR: Ministry of Public Health (in Russian).

1020

MICROWAVES AND ELECTROMAGNETIC FIELDS

USSR (1986a) Standard of the Soviet Council for Mutual Economic Assistance, Protection of workers against Radiofrequency Electromagnetic Fields, ST-SEV 5801-86, Moscow (in Russian). USSR (1986b) USSR Standard for Occupational Exposure, Amendment N.1 of January 1st 1982 to State Standard. 12.0.006-76. USSR (1987) Health Recommendations for the General Public Referring to Exposure Limits to Electromagnetic Fields of Different Frequencies Emitted by TV Stations. Moscow: USSR Ministry of Public Health (in Russian). van der Vorst A, Rosen A, Kotsuka Y (2006) RF/microwave interaction with biological tissues. In: Wiley Series in Microwave and Optical Engineering. 1st edn. p.344 Verreault R, Weiss NS, Hollenbach KA, Strader CH, Daling JR (1990) Use of electric blankets and risk of testicular cancer. Am. J. Epidemiol. 131:159–162. Weiss MW, Peterson RC (1979) Electromagnetic radiation emitted from video computer terminals. Am. Ind. Hyg. Assoc. J. 40:300–309. Wertheimer N, Leeper E (1979) Electrical wiring configurations and childhood cancer. Am J. Epidemiol. 109:273–284. Wolbarsht ML, O’Foghludha FA, Sliney DH, Guy AH, Smith AA, Johnson GA (1979) Electromagnetic emission from visual display units: a nonhazard. Proc. Soc. Photo-Opt. Instrum. Eng. 229:185–195.

27 SOURCES, LEVELS AND EFFECTS OF MANMADE IONIZING RADIATION AND RADIOACTIVITY John J. Mauro and Norman Cohen

27.1 SOURCE DOCUMENTS The characteristics and significance of the sources, levels and effects of both naturally occurring and manmade ionizing radiation and radioactivity are provided in several authoritative reports published by the National Academy of Sciences (NAS), the National Council on Radiation Protection and Measurement (NCRP), the International Commission on Radiation Protection (ICRP), the United Nations Scientific Committee on the Effects of Atomic Radiation (UNSCEAR), and the International Atomic Energy Agency (IAEA). In addition, the U.S. Nuclear Regulatory Commission (NRC), which regulates source, byproduct, and special nuclear material, maintains a library of technical reports, regulatory products, and databases on the commercial nuclear power industry, and the use of radioactive material in education, medicine and industry. The Office of Radiation and Indoor Air (ORIA) of the U.S. Environmental Protection Agency (EPA) has developed a repository of information, in the form of technical support documents, background information documents, environmental impact statements, and regulatory impact analyses, on radioactivity in the environment. Finally, the U.S. Department of Energy (DOE) serves as a national repository of information on all aspects of the use of radioactive materials in research and defense. In addition to these government funded publications, Merril Eisenbud and Thomas Gesell’s fourth edition of Environmental Radioactivity from Natural, Industrial, and Military Sources (1997) provides a comprehensive review of many of the topics addressed in this chapter. This chapter summarizes some of this large and continually expanding body of knowledge, collects and supplements these sources of information with new data, and presents this information in a manner that captures the potential magnitude of the exposures

Environmental Toxicants, Third Edition Edited by Morton Lippmann Copyright
2009 John Wiley & Sons, Inc.

1021

1022

SOURCES, LEVELS AND EFFECTS OF MANMADE IONIZING RADIATION

to this class of environmental toxicants and its significance in terms of its potential effects on public health and safety. The readers are encouraged to refer to the original source documents cited in this chapter, and their reviews as provided in the UNSCEAR publications and by Eisenbud and Gesell (1997). In addition, because of the vast body of literature describing the current status and radiation exposures associated with the nuclear fuel cycle (NFC), the weapons complex, and naturally occurring radioactivity, much of the most current information is not contained in published reports, but is maintained current on web sites maintained by the NRC DOE, EPA, and the Centers for Disease Control and Prevention (CDC), to name just a few of the federal agencies in the U.S. that are involved in gathering and publishing data and information pertinent to manmade and naturally occurring radiation in the workplace and the environment. The reader is encouraged to explore the web sites for these organizations and many of the other organizations and programs described in this chapter.

27.2 SPECIAL UNITS The levels of environmental toxicants are expressed in terms of the mass of the material (e.g., milligrams) per unit measure of the medium in which it is contained (e.g., kilogram of soil, liter of water, or cubic meter of air) or related expression (e.g., parts per million). In addition, exposure to most toxicants is expressed as the rate of intake of the material per unit time (e.g., milligrams per day) or per unit body weight (e.g., milligrams per day per kilogram). These expressions of levels and exposures are intuitively understandable. However, the units used to characterize the quantities of radioactive material in the environment and exposures of humans and organisms other than humans to radiation and radioactive material are unique, and require some discussion prior to addressing the sources, levels, and effects of radioactivity and radiation. The quantities of radioactive material can be, but are not, expressed in units of mass. Instead, they are expressed in terms of the number of atoms undergoing radioactive transformation (referred to as radioactive decay) per unit time. Units of decay rate instead of mass are used to quantify the amount of radioactive material partly because the effects of radioactive materials are related more to the decay rate of the material than to its mass. For example, 1 g of radium-226 (226 Ra) has a decay rate of 3.7  1010 transformations (also referred to as disintegrations)/s, while 1 g of cesium-137 (137 Cs) has a decay rate of 3.2  1012 transformations/s. Since the energy emitted by the radioactive material during radioactive decay is usually of public health concern, and generally not the chemical properties of the radioactive material, it is more convenient to quantify radioactive material according to decay rate. In addition, radioactive materials are detected and quantified by their types and amounts of disintegrations, and not by their unique chemistry, as is the case for nonradioactive material. For these reasons, the quantity of radioactive material is expressed in units of decay rate. The “Curie,” named after the discoverer of radium, Marie Curie, has been the unit most commonly used to define the quantity of radioactive material. It is defined as 3.7  1010 disintegrations/s, which corresponds to 1 g of 226 Ra. Since this is a relatively large quantity of radioactive material from the perspective of environmental radioactivity, it is common to express environmental levels of radioactive material in terms of picocuries (pCi), where “pico” refers to 10 12. Hence, 1 pCi is one trillionth of a Curie, which corresponds to 0.037 disintegrations/s. The Curie is being replaced by “Becquerel,” abbreviated Bq and

SPECIAL UNITS

1023

named after Antoine Henri Becquerel, the discoverer of the natural radioactive properties of uranium. The Becquerel has now been formally adopted to quantify environmental levels of radioactivity. A Bq is defined as one disintegration per second (i.e., 1 Ci ¼ 3.7  1010 Bq). It is instructive to note that a typical level of naturally occurring 226 Ra in soil is approximately 1 pCi/g, which corresponds to 1  10 12 g of 226 Ra/g of soil, and that a typical level of Cs137 in soil due to weapons testing fallout is about 0.1 pCi/g, which corresponds to 1.1  10 15 g of Cs-137/g of soil. At these extremely low mass concentrations, the potential chemical toxicity of radioisotopes is virtually nonexistent relative to the potential radiological toxicity. An important exception to this general rule is uranium, which can be both chemically and radiologically toxic because of its extremely long half-life. Exposure to radioactive materials, in addition to being expressed in terms of intake quantities per unit time (e.g., Bq/day), is also expressed in terms of the amount of energy deposited either in a unit mass of tissue per unit time due to the decay of radioactive atoms that have been deposited within tissue (i.e., internal radiation exposure), or due to the decay of radioactive atoms that are external to the exposed individual (external radiation exposure). One of the units that is commonly used to express the amount of energy deposited in an absorbing medium (referred to as the absorbed dose), such as tissue, is the rad. The rad is defined as 100 ergs/g of tissue.1 For example, it is not uncommon for the decay of a single atom to emit about 1 MeV or 1.6  10 6 erg of ionizing radiation. The amount of energy required to break a single chemical bond in a biological molecule is about 30–40 eV, depending on the molecule (Morgan and Turner, 1973). Hence, the disintegration of a single radioactive atom deposited in tissue can result in on the order of 10,000 ionizations. Theoretically, a sufficient number of these ionizations can damage or kill a cell, or result in chemical changes in biological molecules, which can lead to a carcinoma or mutagenic effect. Because of repair mechanisms, it is not certain whether there exists a dose or dose rate threshold below which no clinical effects of exposure to radiation are expressed. This is an area of active research and debate in the radiological and radiation protection sciences. In order to err on the safe side, radiation protection standards and the practice of health physics assume that biological damage is directly related to the amount of energy deposited in living tissue. Notwithstanding the uncertainties in the effects of radiation, it is convenient to express exposure to radiation in units corresponding to deposited energy per unit mass of living tissue (i.e., the absorbed dose) Some sources of radiation exposure are not due to the decay of radioactive atoms but originate from electronic devices, such as X-ray machines. In these cases, the source of the radiation is always external to the exposed individual, and radiation exposures are often expressed in terms of the amount of energy deposited in air; namely, Roentgens per second (R/s), named after the discoverer of X-rays, Wilhelm Konrad Roentgen. As a general rule of thumb, a radiation field in air of 1 R (which is defined as 1 electrostatic unit/cm3 of air at standard temperature and pressure, or 2.58  10 4 C/kg) is equivalent to about 1 rad. Throughout this chapter, radiation exposure is expressed in units of rem (Roentgen equivalent man). In addition, the expression is further qualified by the term “effective dose equivalent” (EDE). For the purpose of this chapter, the unit “rem/year EDE” (or mrem/year EDE) is used to normalize all exposures to the equivalent, in terms of potential adverse health consequences, of 1 rad (i.e., 100 erg/g of tissue) of uniform whole body exposure to either 1

Absorbed dose is also expressed in units of Gray (Gy) ¼ 1 J/kg ¼ 100 rad. This unit, as well as the Sievert (Sv) was developed as a means to express absorbed dose (and absorbed dose equivalent) in standard international (SI) units.

1024

SOURCES, LEVELS AND EFFECTS OF MANMADE IONIZING RADIATION

gamma or beta radiation.2 In this way, it doesn’t matter whether the exposures are caused by external or internal radiation, by alpha, beta, or gamma emitters, by neutrons, or if the energy of radioactive decay is deposited uniformly throughout the body or is limited to a given organ. No matter what the form of the exposures, the rem EDE allows the exposures to be intercompared on a potential health equivalent basis. The reader may notice that radiobiological and health physics publications also express the dose equivalent in units of the Sievert (Sv). The Sv ¼ 100 rem, and it was adopted as a means for expressing dose equivalent in standard international (SI) units. The units used to define the quantity of radioactive material and exposure to radiation and radioactive material have evolved and are undergoing continuous refinement to more precisely characterize the interaction of radiation with matter and the biological effects of exposure to ionizing radiation. The International Commission on Radiation Units and Measurement (ICRU)3 is the standard setting body that establishes the formal definition of radiation units. Brodsky (1978) provides a history of the evolution of radiation units.

27.3 SOURCES OF MANMADE RADIOACTIVITY AND RADIATION The actual and/or potential sources of exposures to manmade radioactivity and ionizing radiation can be categorized according to eight sources of exposures, whether the exposures occur as a result of normal or accident conditions associated with source, and whether the persons receiving exposures are members of the general public or radiation workers (Fig. 27.1). In addition, there is a consideration (or fourth dimension) of time, whereby the radiation exposures to workers and the public can and have changed over time and are projected to continue to change in the future. Each source category can cause exposure to the people who work in the industries that are responsible for creating the sources of exposure. In all categories, with the possible exception of the category entitled sources of naturally occurring radioactive materials, these individuals are referred to as “radiation workers” and are subject to strict radiation protection

FIGURE 27.1

Sources and manmade radioactivity and radiation.

2 One of the more confusing aspects of the radiological sciences is the periodic revision of nomenclature used to express dose. For example, the effective dose equivalent is now simply referred to as the effective dose. 3 ICRU is located at 7910 Woodmont Avenue, Suite 800, Bethesda, MD 20814-3095 (301-675-2652).

NUCLEAR FUEL CYCLE

1025

controls. Each source category is also responsible, to varying degrees, for exposures to members of the general public. For each source category, and for both radiation workers and members of the general public, radiation exposures occur as a result of normal operation of the facility or use of the sources of radioactivity, and also as a result of off-normal conditions and/or accidents.

27.4 NUCLEAR FUEL CYCLE As reported by the NRC, as of August 2008, there were 104 nuclear reactors licensed to operate in 31 states generated approximately one-fifth of the Nation’s electricity (NRC, 2008). Although no new nuclear plants have been ordered since about the time of the accident at the Three Mile Island-2 site in 1979, many nuclear utilities are seeking 20-year extensions to their original 40-year operating licenses because of the economic advantages associated with license renewal. In addition, there is increasing speculation that circumstances today may favor a reconsideration of nuclear power as a viable source of electrical energy. These circumstances include the rapidly rising price of fossil fuel, concerns about pollution and global warming from combustion technologies, streamlining the nuclear plant licensing process (see 10 CFR Part 52), and advances in reactor technology. The NRC reports (NRC, 2008) that it expects to receive a significant number of new combined construction and operating license (COL) applications for reactors over the next several years. The nuclear fuel cycle, as discussed here, refers to the totality of those activities surrounding the central defining process of controlled nuclear fission with the ultimate goal of producing energy required for the generation of electricity (see Fig. 27.2). However, the nuclear fuel cycle, as described here, also applies to the production of radionuclides necessary for a variety of medical and commercial applications, for carrying out scientific research, for creating locomotive power for naval vessels, and for producing nuclear weapons for the military (Cochran and Tsoulfanidis, 1990). At the outset, it should be noted that the designation “NFC” may give the incorrect impression of a self-sustaining, closed-loop process. In 1978, President Carter ordered a moratorium on reprocessing of spent fuel as a means of reducing the potential for weapons proliferation (i.e., reprocessing of spent fuel produces plutonium and enriched uranium for recycling back into fuel, but it is also then available for weapons). Hence, for the U.S. light– water reactor program, where recovery of fissionable fuel from spent nuclear fuel (SNF) elements, namely reprocessing and recycling have not been allowed options since 1978, the cycle is left “open.”4 It becomes reasonable, therefore, to speak of each stage of the fuel cycle as part of “front and back ends” with the operating nuclear fission reactor providing the central focus for both fuel production and waste management planning. This section briefly summarizes and reviews the various stages of the NFC in general terms, considering each aspect’s potential to act as a source of exposure to ionizing radiation. (Each stage of the NFC has some potential for exposing populations to external radiation and to the possible uptake of radionuclides either directly or from related transportation and waste disposal activities.) Thus, this review is not intended to be comprehensive in scope, and the reader is directed to the references cited for a more detailed accounting of any of the specific topics.

4

As described later in this chapter, this moratorium was recently lifted.

1026

SOURCES, LEVELS AND EFFECTS OF MANMADE IONIZING RADIATION

FIGURE 27.2

27.4.1

The nuclear fuel cycle (from NCRP, 1987a).

Exploration and Mining

After exploration for suitable mining areas of naturally high concentrations of uranium ore (for a variety of the most effective uranium exploration techniques, see DOE/NFC, 1990), the only concern noted for possible radiation exposure was the possibility of ground water contamination resulting from “improperly plugged boreholes” (NCRP, 1987b). Once suitable sources of uranium ore are discovered, the front end of the cycle begins with the extraction of the ore from the ground. Most often, this process takes place either relatively close to the original surface, that is, open pit mining (usually within 120 m of the surface), or from deep within the ground. A third method of uranium mining, “in situ leaching” or solution mining, is an economic alternative to underground mining (Cochran and Tsoulfanidis, 1990). Deep underground hard-rock mining, in particular, is considered a significant potential radiation hazard (although to a far lesser extent at present than in the past when mine ventilation was minimal), primarily because of the possibility of the inhalation of radon

NUCLEAR FUEL CYCLE

FIGURE 27.3

1027

Principal decay scheme of the uranium series.

and its short-lived daughters, that is, decay products. The detrimental health effects to the lungs from this occupational exposure were recognized in the early 1940s (Hueper, 1942; Lorenz, 1944), but it was not until about 1957 (Holaday et al., 1957) that the increased incidence of lung cancer was definitively attributed to high levels of exposure to these particulate radionuclides (Fig. 27.3). Generally speaking, exposures of the nonoccupational public to radioactive emissions resulting from underground mining are not considered to represent a significant risk (Blanchard et al., 1982). 27.4.2

Milling and Refining

After mining, the ore (usually containing on the average about 0.1–0.2% U3O8 as disseminated pitchblende) is sent to a processing mill where it is first crushed and then ground. A concentration process is then begun as part of the milling and refining stage, in which uranium is selectively extracted, converted by chemical techniques, and dried. At the final

1028

SOURCES, LEVELS AND EFFECTS OF MANMADE IONIZING RADIATION

stages of this process, the chemical composition of uranium is a complex mixture of uranium compounds including diuranates, uranyl sulfates, and U3O8, which, in combination, is generally designated as “yellowcake” (Momeni et al., 1979). The mixture is then converted at a chemical refinery to almost pure U3O8. In these compounds, uranium still has the isotopic composition given in Table 27.1 for “natural uranium” (Unat). To define a source term for assessing the ecological and human exposures and doses resulting from the releases of drying and packaging operations, it is first necessary to establish the temporal variability of the release rates as well as the characterization of the effluent by particle size, density, transportability, and bioavailability. At this stage of the fuel cycle, significant potential for exposure to radiation occurs to both workers and the general public. Occupationally, radiation exposure can occur internally from radioisotopes deposited within the body, or externally from one or more of the radionuclides of the uranium chain that may be present in radioactive equilibrium (i.e., in equal amounts to uranium activity) in the crushed ore, in concentrates, or in purified products. External exposures result primarily from the more penetrating gamma-rays, although energetic beta-particles from those nuclides closer to the body will also contribute to the external dose rate. Internal exposures result from the inhalation and/or ingestion of radionuclides, which then proceed to produce a radiological dose resulting from their radioactive decay, leading to energy absorption within specific tissues. The major difference between internal and external exposures is that, with internal exposures, there is the

TABLE 27.1 Nuclide 233

U U 235 U 238 U 234

Uranium: Radiological Dataa T1/2 (years) 5

Series

Nuclide S.A. (Ci/mg)

1.62  10 2.47  105 7.10  108 4.50  109

Neptunium Uranium Actinium Uranium

9.48 6.19 0.00214 0.000333

Nat. abun.

S.A. (Ci/mg)

Ci/mg of Unat

Specific activity for Unat (Ci/mg) 138 U 0.993 0.000333 235 U 0.007 0.00214 234 6.19 U 5.5  10 5 Specific activity for DU (Ci/mg) 138 U 0.997 0.000333 235 U 0.003 0.00214 234 U 3.4  10 5 6.19

3.31  10 4 0.14  10 4 3.40  10 4 Unat ¼ 6.86  10 3.32  10 4 0.06  10 4 2.10  10 4 DU ¼ 5.48  10

Activity (%) Nuclide 138

U U 234 U 235

a

Unat 48.25 2.19 49.56

DU 61.18 1.18 38.70

Unat, natural uranium; DU, depleted uranium; S.A., specific activity.

Nat. abun. (mass%) – 0.0055 0.7 99.3

4

4

NUCLEAR FUEL CYCLE

1029

possibility of energy deposition at biologically sensitive sites by highly energetic and damaging alpha-particles. Externally, alpha-particles cannot penetrate the dead layer of the skin. What is termed “uranium refining” is essentially the chemical conversion of the mill concentrates to either uranium metal or intermediate compounds such as UO3 (orange oxide) or UF4 (green salt). Once again, these operations have the potential for exposing workers internally to alpha-emitting uranium dusts and externally to beta- and gamma-radiations. Eisenbud and Gesell (1997) noted that, although uranium was released to the environment by the older refineries during World War II, today’s more modern plants “are equipped with filtration equipment that effectively removes uranium dust,” which, when properly maintained and operated, prevents any significant release to the environment. During the early years, when the technology for chemically separating and refining the uranium from the other radionuclides in uranium ore was in its infancy, radiation exposures to workers were relatively high and poorly monitored. For example, workers employed by the Uranium Division of Mallinckrodt Chemical Works, Destrehan Street Facility, during the period from 1942 through 1948, were recently classified as a special exposure cohort (SEC) under the Energy Employees Occupational Illness Compensation Program Act of 2000 (EEOICPA). Under this statute, and the regulation that implement this statute (i.e., 42 CFR Part 83), workers are designated as a SEC if the Department of Health and Human Services (DHHS) determines that it is not feasible to estimate with sufficient accuracy the radiation dose that the workers at a given site experienced and if it has been determined that there is reasonable likelihood that such radiation exposures may have endangered the health of the workers. As members of a SEC, these workers are entitled to special compensation under the statute. The residual mill tailings are the major waste products that can enter the exposure pathways of the nonoccupational environment, primarily through incorporation into the food web (i.e., ingestion) and by particulate matter resuspension and 222 Rn emission (i.e., inhalation). These tailings are comprised predominantly of silica compounds, for example, quartz grains, feldspars, rock fragments, and a variety of interstitial clay minerals (NRC/ NAS, 1986), and radionuclides from the uranium decay chain. Of these nuclides, 232 Th, 226 Ra, and the short-lived decay products of 222 Rn (Fig. 27.3) present the greatest risk of radiation exposure. The most dramatic example of environmental contamination and population exposure to radiation as a result of the existence of mill tailings piles occurred in Grand Junction, Colorado prior to 1966. It is reported (Eisenbud and Gesell, 1997) that a number of the local residents of this town used the then readily available uranium mill tailings sands in home construction as backfill and foundations and in concrete mixes. Since the tailing pile sands contained significant concentrations of 226 Ra and its decay products, which were left behind after uranium was chemically extracted, a constant flux of 222 Rn was produced from whatever home structural material had incorporated the tailings. Consequently, in addition to higher than normal external gamma-ray exposures to the residents, the excessively elevated airborne concentrations of 222 Rn and its short-lived daughters, created, in many of the homes, a significant risk of bronchial carcinoma from the inhalation of the radon daughters. The magnitude of the problem was greatly reduced by 1978, when more than 700 buildings were renovated at government expense (DOE, 1980). Since mill tailings can represent a significant source of exposure to radionuclides, the Uranium Mill Tailings Radiation Control Act (UMTRA) was enacted in 1978, which (1) directs the Federal government to undertake the elimination of the hazards associated with

1030

SOURCES, LEVELS AND EFFECTS OF MANMADE IONIZING RADIATION

abandoned inactive tailings piles; and (2) clarifies and strengthens the authority of the NRC to insist on proper tailings management by its uranium mill licensees and those states that license such milling operations (NRC/NAS, 1986). Since the enactment of this statue, the NRC, DOE, EPA, and state authorities have been monitoring the stability of the tailings impoundments and the groundwater resources in the vicinity of these sites in order to ensure that the facilities and sites meet the provision of the statute and its implementing regulations. 27.4.3

Enrichment

The next stage in the process of producing usable nuclear fuel is the final chemical conversion of the refined yellowcake, using hydrofluoric acid and fluorine, to uranium hexafluoride (UF6) gas. It is the UF6 gas that is then “enriched.” The enrichment process increases the concentration of the fissile uranium isotope 235 U from the 0.72% level found in natural uranium to the 2–4% needed to sustain a fission reaction with slow neutrons in light– water power reactors, and to greater than 90% in some production-type reactors for nuclear weapons manufacture. Since only small quantities of uranium hexafluoride escape to the air or are discharged as liquid waste in the enrichment process by gaseous diffusion, centrifugal techniques, or atomic vapor laser isotope separation (AVLIS), the potential for radiation exposure is considered minimal to both the plant worker and the general public (NCRP, 1987a). Furthermore, the stringent controls on emissions necessitated by the very high cost of enriched uranium and the antiproliferation safeguards imposed in the name of national security make the possibility of widespread environmental contamination from these plants unlikely (Eisenbud and Gesell, 1997). Though the potential for exposures to members of the general public from the uranium enrichment process is small, the exposures to workers at these facilities, especially in the early years, was not insignificant. This is evidenced by the fact that many former workers at the gaseous diffusion plants in Paducah, Kentucky, Portsmouth, Ohio, and Oak Ridge, Tennessee have been designated as members of a special exposure cohort under the EEOICPA (see 42 CFR Part 83) and are being compensated for the health detriment they have experienced as a result of the radiation exposures they may have experienced while working at these facilities. 27.4.4

Fuel Conversion and Fabrication

The last step in the production of nuclear reactor fuel (Fig. 27.4) involves the chemical and physical conversion of enriched UF6 to uranium dioxide UO2 fuel pellets. The very limited potential for environmental exposure associated with this process is attributed to the minute quantities of radionuclides released to the environment as gaseous, liquid, and solid wastes. However, the potential for significant historical occupational exposures at fuel fabrication plants is currently undergoing investigation by the National Institute of Occupational Safety and Health as part of the EEOICPA. 27.4.5

Reactor Operation

Most commercial utility-operated nuclear power reactors in the United States are of the light–water variety in which the uranium fuel is enriched to about 3% and the water serves

NUCLEAR FUEL CYCLE

FIGURE 27.4

1031

Simplified uranium fuel production scheme.

as both coolant and moderator. Of the three major categories of radionuclides routinely produced during the normal operation of these reactors (i.e., fission products, actinides, and neutron activation products), fission products (primarily noble gases) are the principal radionuclides that are routinely released to the environment in any significant amount (BNL, 1995). Various Federal and state agencies use an extremely effective and stringent system to control and regulate minimal allowable releases of radioactivity to the environment. On behalf of the NRC, Brookhaven National Laboratory has, in the past, published an annual report on the quantities of radioactive material released from nuclear power plants (BNL, 1995). These release estimates have been used to derive the potential radiation doses to the general public living in the vicinity of the plants (PNL, 1995). In 1989, the EPA published a comprehensive assessment of the potential doses and risks associated with the routine atmospheric releases of radionuclides from nuclear facilities (EPA, 1989). The results revealed that the potential whole body radiation doses to the

1032

SOURCES, LEVELS AND EFFECTS OF MANMADE IONIZING RADIATION

maximally exposed members of the public have been less than 1 mrem/year (well within the radiation protection standards) and much lower than the variability in doses from natural sources of radiation. The principal factors that account for the low exposures resulting from normal reactor operation are (1) a multiple barrier system that effectively limits the amounts of radionuclides leaving their points of origin; (2) effective treatment, that is, processing technology, for removing activity prior to its release as effluent; and (3) the release into systems that guarantees large dilution factors and minimal environmental transport in both air and water before reaching unrestricted areas. Estimates of dose have been derived from calculations of normal reactor effluent releases in conjunction with considerations of a variety of pathway parameters leading to ingestion from the food web, inhalation intakes, and/or external radiation exposures. All-in-all, normal reactor operation and its associated effluent represent the lowest potential for radiation exposure of any of the fuel cycle phases. Natural whole-body background radiation in the U.S. is currently given as 3.0 mSv/year (300 mrem/year) EDE, with the dose caused by the inhalation of radon and its short-lived decay products accounting for about two-thirds of this total and most of the variability (NCRP, 1987c). It is often assumed that radiation exposures to the general public as a result of any type of reactor accident are always of significant consequence and perhaps even potentially disastrous. However, the extremely low probability of the occurrence of a serious nuclear accident in this country (NRC, 1989), and the experience gained from the Three Mile Island accident, indicate that, in terms of the overall population’s radiation exposure, reactor accidents have not been as catastrophic as originally feared. No reactor accident in this country has caused quantities of radioactive material to be released that resulted in significant population exposures. The accident that occurred at the Three Mile Island-2 plant in 1979 resulted in the partial destruction of the reactor core. However, the designed safety features prevented all but a very small fraction of the core inventory of radionuclides from escaping into the environment. Furthermore, the highest doses to individuals living near the site were generally less than the average natural radiation dose received annually in the U.S. (Gerusky, 1981). The Chernobyl accident, at a reactor lacking a containment structure, demonstrates that a serous reactor accident can be disastrous. The Chernobyl #-4 reactor accident, which occurred in the former Soviet Union on April 26, 1986, released essentially all of its noble gases, 60% of its radioiodine isotopes, 40% of its radiocesium, 10% of its radiotellurium, and about 1% or less of the more refractory radioactive elements (Gudiksen et al., 1989). The geographical and temporal patterns of exposures, the exposure pathways, and the actual and potential health consequences of the accident are extremely complex. The “The Chernobyl Papers” edited by Merwin and Balonov (1993) and Eisenbud and Gesell (1997) provide an overview of the accident and its consequences. Eisenbud and Gesell divided the impacts of the accident into five geographical locations, and summarized the impacts as follows: (1) (2) (3) (4) (5)

On-site personnel. Environs within 30 km. European portion of the USSR beyond 30 km from the accident. European countries outside the USSR. North America and Asia.

NUCLEAR FUEL CYCLE

1033

Approximately 1000 cases of acute radiation poisoning occurred, including 28 early deaths, where doses were as high as 1600 rad. During the first year, approximately 200,000 workers were employed in the cleanup. Those individuals were estimated to receive an average dose of 20 rem. The approximately 45,000 persons within 3 km of the reactor, in the town of Pripyat, were exposed to an external radiation field of about 500–1500 mR/h prior to being evacuated. Within 10 days, about 135,000 persons were evacuated within a 30 km radius of the accident. The mean whole body dose to the evacuees was estimated to be 1.5 rem. The thyroid doses were estimated to be 100 rem for children and 70 rem for adults. A summary of the consequences of the Chernobyl accident is provided in Health Risks from Exposure to Low Levels of Ionizing Radiation, BEIR VII-Phase 2, National Research Council, National Academy of Sciences, Washington, DC, 2005 (referred to as the BEIR VII report, NAS, 2005). The scale of the long-term collective public health detriment on the evacuees, the cleanup workers (referred to as “liquidators”), and the nearby and regional population have been summarized in BIER VII to be in excess of 100,000 person Sieverts (10 million person-rem) collective effective dose. The World Health Organization predicts that in the Gomel region of Belarus alone, 50,000 children will develop thyroid cancer due to the accident, and the incidence of leukemia has already increased 50% in children and adults (see www.chernobyl-international.org/ facts.html). Beyond 30 km, the principal sources of exposure were 137 Cs and radioiodine deposited on the ground and in food. Estimates of the doses to individuals and the collective doses in different regions of the former Soviet Union and Europe are geographically and radioecologically complex, and are the subject of numerous reports, peer-reviewed publications, workshops, and symposia. In North America and Asia the exposures were small because the accident did not contribute to stratospheric fallout. 27.4.6

High-Level Nuclear Waste

27.4.6.1 Production and Management When neutrons are captured, uranium atoms are, by virtue of their resulting nuclear instability, split (fissioned) into atoms of lower atomic number and mass. The energy generated by the fission process is quantified by the term “burnup,” which has units expressed as “megawatt-days per metric ton of fuel.” After enough 235 U has been so fissioned, “spent” fuel assemblies must be replaced; that is, the reactor is refueled. These spent fuel rod assemblies containing unburned fuel, fission products, actinides, and other activation products are designated, in toto, as high-level waste (HLW) (DOE, 1996). Typically, about one-third of the fuel (i.e., uranium enriched to about 3%) of a pressurized water reactor (PWR) and one-fourth of that for a boiling water reactor (BWR) is replaced after every 12–18 months of reactor operation. The spent fuel is then stored (usually on site or at independent spent fuel storage facilities at the present time) under water or in dry casks for cooling and to allow for the initial decay of the relatively short-lived nuclides prior to the fuel rods being shipped to projected storage sites. As of 2002, about 46,000t of spent fuel from nuclear power plants were in storage in the United States and the amount is increasing at about 2000t per year. The vast majority of spent fuel is stored in water pools, and a small percentage is stored in dry casks. If all the spent fuel assemblies in storage were assembled in one place, they would only cover a football field about 51=2 yards high (NRC, 2002).

1034

SOURCES, LEVELS AND EFFECTS OF MANMADE IONIZING RADIATION

Under the Nuclear Waste Policy Act of 1982 (Public Law 97–425), as amended in 1992, the DOE entered into contracts with all utilities operating nuclear power plants to accept their spent fuel for disposal by 1998, but missed this deadline. As a result, the spent fuel being generated at nuclear power plants is continuing to be held in storage. The NRC estimated that, in 2005, 52,000 metric tons of spent fuel were in storage (NRC, 2008). In 2002, the Secretary of the DOE recommended and the President and Congress approved Yucca Mountain in Nevada as the site of the spent fuel and high-level radioactive waste repository. Engineering characteristics of the planned geological repository are not described in this chapter, other than to note that design specifications for the facility have been set so as to insure waste isolation for exceedingly long periods of time (in excess of 10,000 years), with the goal of minimizing the potential release fission and activation products to the biosphere. Regulatory requirements are being developed by the EPA, in accordance with the Energy Policy Act of 1992, to limit the dose to the general public from the spent fuel and HLW received and stored in a deep geological repository to a few mrem/year to no more than a few tens of mrem/year. According to the NCRP, “current technology appears to be capable of maintaining doses to a fraction of this requirement” (NCRP, 1987a). According to an EPA proposed rule issued in 2005 (i.e., proposed 40 CFR 197), the repository will be required to ensure that the dose to the general public will not exceed 15 mrem/year over the first 10,000 years of the life of the proposed repository, and no more than 350 mrem/year in the year in which the peak exposure is projected to occur. Though the Administration is committed to licensing and operating a repository at Yucca Mountain, the design, regulatory requirements, and schedule is uncertain, especially in light of the passage of the FY 2006 Energy and Water Appropriations Bill, which mandates spent fuel reprocessing and recycling. The approval of this bill in 2005 represented a major change in administration policy, which previously banned the reprocessing and recycling of commercial spent fuel. This change in policy has the potential to profoundly affect the future of nuclear energy and the entire nuclear fuel cycle in the U.S.* The projected exposures and health risks associated with the disposal of high-level waste and spent fuel in a repository have been the subject of numerous investigations performed by the DOE 1979, 1983, 2002a). DOE has electronically archived large numbers of documents presenting the results of its dose assessments, referred to as Total System Performance Assessments (TSPAs), at two web sites: http://www.ocrwm.doe.gov and http://search. lsnext.us/search/doesearch. The TSPAs predictively model, the radiation exposures that individuals living down-gradient from a repository might experience over a 1 million year time period. Because of the enormous uncertainties associated with modeling the performance of a repository over such long time periods, these predictive models employ Monte Carlo methods, which assign uncertainty distributions, many of which are empirically based, to most of the important modeling assumptions, scenarios, and parameters. The results of these analyses to date reveal that the highest plausible doses to members of the public are on the order of 1 rem/year EDE (not including direct inadvertent or deliberate intrusion into the repository) and are projected to occur well beyond 10,000 years following disposal of the waste (DOE, 2002b). The TSPAs are continually being revised. Further complicating the future of a waste repository is the passage of the Fiscal Year 2006 Energy and Water Appropriations bill, which will require that a spent fuel recycling technology plan be developed by the DOE. The implications are that the design of the proposed repository will need to meet a changing design basis. * At the time of the preparation of this chapter, the DOE application for a license for the proposed Yucca Mountain repository was in preparation, and EPA regulations pertaining to the repository were in preparation.

NUCLEAR FUEL CYCLE

FIGURE 27.5 DOE, 1998).

27.4.7

1035

Volume of low-level redioactive waste shipped for disposal in the United States (from

Low-Level Radioactive Waste (LLW)

Low-level waste (LLW) is generated by government facilities, utilities, industries, and institutional facilities. In addition to 35 major DOE facilities, over 20,000 commercial users of radioactive materials generate some amount of LLW. LLW generators include approximately 100 operating nuclear power reactors, associated fuel fabrication facilities, and uranium fuel conversion plants, which together are known as nuclear fuel-cycle facilities. Hospitals, medical schools, universities, radiochemical and radiopharmaceutical manufacturers and research laboratories are other users of radioactive materials which produce LLW. The cleanup of contaminated buildings and sites will generate more LLW in the future (see http://www.epa.gov/radiation/docs/radwaste/402-k-94001-llw.htm). Fig. 27.5 presents the volume of low-level radioactive waste shipped for disposal from 1985 to 1998 (from NRC licensees and does not include DOE waste). Table 27.2 presents a

TABLE 27.2 Low-Level Radioactive Waste Received at Commercial Disposal Sites in 2004 (cubic meters) (35.3 ft3/m3) (from DOE, 2005) Source Academic Government (from DOE) Government (non-DOE) Industry Medical Utility Government mixed waste (from DOE) All other mixed waste Total NRC licensees Grand total

Class A

Class B

Class C

Total

28 258,000 17,613 35,491 1.6 55,391 8,900 273 108,7798

0 0 20 7 0 385 0 0 412

1.5 0 26 15 0.7 447 0 0 490

29 258,000 17,659 35,513 2 56,223 8,900 273 109,699

376,000

412

490

337,000

1036

SOURCES, LEVELS AND EFFECTS OF MANMADE IONIZING RADIATION

detailed breakdown of the sources, types, and volumes of LLW received at commercial disposal sites in 2004. A distinction is made between Class A, B, and C LLW because they represent an increasing radionuclide concentrations and increasing potential long-term hazards. A distinction is also made between LLW and mixed LLW because mixed waste contains chemically hazardous or toxic material along with LLW, and must be disposed of in a manner that meets not only NRC requirements related to the disposal of LLW, but also EPA and state requirements applicable to the disposal of chemically hazardous waste. Mixed waste has become a contentious disposal issue because the regulations governing the two different types of waste have certain fundamental differences that create regulatory obstacles to its disposal. In general, the volume of LLW disposed at licensed LLW facilities has generally declined since the early 1980s, primarily due to a concerted effort by NRC licensees to reduce the volume of LLW. This volume reduction initiative was the result of the decreasing availability of low-level waste disposal sites in the U.S., and the associated increasing cost of waste disposal over this time period, which is generally attributed to the Low-Level Radioactive Waste Policy Act of 1980. A great deal of information on the history of this Act and its effects on the management and costs of low-level radioactive waste and mixed waste in the U.S. is available on the web. The EPA has evaluated the potential doses to members of the general public in the vicinity of a licensed low-level radioactive waste disposal facility and has determined that the doses would be a very small fraction of the radiation protection standards (EPA, 1988e), which have been established by the NRC at 25 mrem/year EDE (see 10 CFR 61). 27.4.7.1 Reprocessing Reprocessing is the chemical extraction and recovery of residual and newly formed fissile material, primarily 235 U and 239 Pu, from spent reactor fuel for possible recycling, thereby “closing” the nuclear fuel cycle. There has been only one reprocessing operation for commercial power reactor fuel in the United States, the Nuclear Fuel Services Plant in West Valley, New York, which was closed in 1972. Since there is no commercial reprocessing plant currently operating within the U.S., excluding those reprocessing nuclear fuels from naval propulsion and special government research and test reactors, estimates of public exposures do not exist except as projected models (NCRP, 1987a). However, the Energy and Water Appropriations Act for Fiscal Year 2006 mandates that a spent fuel recycling processing plan be developed by the DOE. The passage of this bill is likely to usher in a rebirth of nuclear power in the U.S., including reprocessing and recycling spent fuel. Some of the potential advantages of this new initiative are reduced dependence on foreign oil, reduction in the production of greenhouse gases, and reduction in the quantity of long-lived transuranic waste. In addition, some of the technologies under consideration can use weapons grade uranium and plutonium as part of the recycling process, thereby reducing the global inventory of weapons grade material (Stanford, 2001). 27.4.7.2 Transportation Routine truck shipments of uranium concentrates are not, under ordinary circumstances, associated with more than negligible radiation exposures to the driver and the general public. It has been reported, for example, that exposure rates in the truck cab and 2 ft from a loaded trailer rarely reach levels greater than 5 Sv/h (0.5 mR/h) (Miller and Scott, 1981). It is conceivable, however, that exposure problems could result after a vehicle accident that causes spillage of the truck contents.

NUCLEAR FUEL CYCLE

1037

Because of their composition and characteristically greater activity levels, transport of spent fuel and high-level waste could present considerably greater risks of exposure than spills from the transport of low-specific-activity material from the front end of the NFC. However, the stringent regulations governing the design of shipping containers for spent fuel and high-level radioactive greatly reduce the probability that a severe transportation accident will result in the release of radioactive material during transport. 27.4.7.3 Decommissioning and Decontamination An additional aspect of the NFC that needs consideration for future exposure prediction is reactor decommissioning. Undoubtedly, many of the early nuclear power plants will be reaching their useful lifetimes within the next 10–20 years, and the dismantling of these facilities is bound to produce some risk of exposure to radionuclides. In addition, the NRC’s Site Decommissioning Management Plan lists 28 sites that require special attention to resolve decommissioning policy and regulatory issues. To facilitate decommissioning and license termination, the NRC has issued its final rule on radiological criteria for license termination. The rule requires, in part, that the applicant for a license termination and release of the property for unrestricted use must demonstrate that the radiation exposures associated with the free release of the property would not exceed 25 mrem/year EDE for all pathways combined, and the exposures will be as low as reasonably achievable (FR, Vol. 62, No. 139, 7/21/97). The EPA continues to develop public health radiation protection criteria for residual radioactivity following cleanup of contaminated lands and facilities. The purpose of such criteria is to assure protection of public health and the environment after reactor facilities are shut down and decommissioned (Health Physics Society, 1986; EPA, 1993c). In light of the fact that both agencies share authority over the protection of the public from radiation and radioactive material, and in recognition of the fact that NRC cleanup criteria set forth in 10 CFR Part 20 and EPA cleanup criteria applicable to Superfund sites (40 CFR Part 300) are not entirely compatible, the two agencies have entered into a Memorandum of Understanding whereby EPA defers to the NRC in matters related to the cleanup of facilities and sites licensed by the NRC. However, EPA retains the right to participate in the license termination process under conditions where ground water resources may be impacted in excess of EPA’s Maximum Permissible Concentration (MPCs) found in 40 CFR 141 and matters involving hazardous materials (MOU dated September 30, 2002 and signed by Christine Whitman, Administrator EPA and Richard A. Meserve, Chairman NRC). Both the NRC and EPA recognize that very large volumes of steel, concrete, and soil are associated with the decontamination and decommissioning of NRC licensed facilities, DOE and DOD facilities, and sites being remediated under the EPA’s Superfund program. Rather than requiring these very large volumes of material, most of which contain very low residual levels of radioactive contamination, to be disposed of at licensed radioactive waste disposal facilities, both agencies are considering alternative strategies that would permit the disposal of this material at facilities other than licensed low-level radioactive waste disposal facilities, such as facilities permitted under the Resource Conservation and Recovery Act (RCRA). Such strategies would reduce the cost of disposal of this material and reserve licensed lowlevel radioactive waste disposal facilities for the disposal of waste containing higher concentrations of radioactive material (EPA Publication of Advanced Notice of Proposed Rulemaking Regarding the Disposal of Low-Activity Radioactive Waste: request for Comment, Federal Register: November 18, 2003, Vol. 68, No. 222, page 64993).

1038

SOURCES, LEVELS AND EFFECTS OF MANMADE IONIZING RADIATION

27.5 DISCUSSION OF RADIATION DOSES FROM THE NUCLEAR FUEL CYCLE The nuclear fuel cycle, as defined in this chapter, has been described for the most prevalent commercial fission reactors in the U.S., namely, light–water PWR and BWR facilities. It has also been noted that nuclear energy generates about 20% of the electricity in this country and more than 50% of the electricity production in New England (NRC, 2008). Although this is not as high as it is in some European countries, for example, 78% for France and 49% for Sweden (NRC, 2008), it is a clear indication that nuclear energy continues to exist and can make an even more significant contribution to the future generation of energy in this country. It is essential that the nuclear fuel cycle be considered as a source of radiation exposure to U.S. workers and the general public 27.5.1

Exposures to Radiation Workers

The NRC publishes an annual report summarizing the occupational radiation exposure at commercial nuclear power reactors and other facilities (NRC, 2004). Table 27.3 summarizes these exposures for major components of the NFS. The collective exposures to workers from these facilities has been about 15,000–16,000 person-rem/year since about 1998 and about between 20,000 to 25,000 person-rem/year prior to 1997. The reduction in the collective doses is due to a reduction in the total number of monitored individuals and a reduction in the average annual dose per worker, presumably due to increasing attention to sound ALARA principles. Using a risk coefficient of 5  10 4 fatal cancers per rem (see the discussion of risk coefficients at the end of this chapter), the more recent exposures may be associated with approximately 5–10 fatal cancers per year in a worker population of about 120,000 people. This theoretical risk may be compared to the risk of fatal industrial accidents with an overall average of 5 fatalities per 100,000 workers in 1993, and a range of 2 for business services to 94 for agricultural services per 100,000 workers in 1993 (DOL, 1995). Hence, the theoretical risk of worker fatalities due to radiation exposures at commercial nuclear power plants is comparable to the risks of industrial accidents. Bear in mind that the radiation risks are based on conservative (mathematical) theoretical models relating radiation exposure to risk, while the latter risks are based on actuarial data; that is, real verifiable risks. .

.

.

These categories consist only of NRC licensees. Agreement State licensed organizations are not required to report occupational exposure data to the NRC. As of 1999, these are no longer any NRC licensees involved in this activity. All lowlevel waste disposal facilities are now located in Agreement States and no longer report to the NRC. Includes all LWRs in commercial operation for a full year of the years indicated. Reactor data have been corrected to account for the multiple counting of transient reactor workers.

27.5.2

Exposures to the General Public

Table 27.4, which was taken from a review of the fuel cycle in this country as performed by the NCRP 1987a, indicates that, based on the radioactive effluents produced by the various stages of the fuel cycle activities, the milling operation is responsible for the greatest potential doses to regional populations. When expressed on the basis of dose to the

1039

1994 1995 1996

2 2 2

36 38 33 31 39 39 36 29 23

1995 1996 1997 1998 1999 2000 2001 2002 2003

Low-level waste disposalb

44

1994

Manufacturing and distribution

139 149 148 142 132 129 124 100 86

1994 1995 1996 1997 1998 1999 2000 2001 2002 2003

Calendar Year

Number of Licensees Reporting

202 212 165

2666 2631 1154 1986 2181 2461 1862 1437 1849

2941

2886 3761 3766 3570 4952 3837 3368 3780 3420 2649

Number of Monitored Individuals

83 56 67

1222 1241 665 654 836 1188 1211 1052 1459

1251

2007 2651 2639 2574 3446 2827 2542 3161 2842 2319

22 8 8

595 556 397 402 419 415 351 328 394

580

1415 1443 1449 1356 1863 1551 1528 2111 1729 1504

Number of Workers with Measurable Collective TEDE TEDE (person-rem)

0.11 0.04 0.05

0.22 0.21 0.34 0.20 0.19 0.17 0.19 0.23 0.21

0.20

0.49 0.38 0.38 0.38 0.38 0.40 0.45 0.56 0.51 0.57

Average TEDE (rem)

(continued)

0.27 0.15 0.12

0.49 0.45 0.60 0.61 0.50 0.35 0.29 0.31 0.27

0.46

0.71 0.54 0.55 0.53 0.54 0.55 0.60 0.67 0.61 0.65

Average Measurable TEDE per Worker (rem)

Average Annual Exposure data for Annual Occupational Exposures from Selected Nuclear Fuel Cycle Facilities for 1994–2003 (NRC,

Industrial radiography

NRC Licensee Categorya

TABLE 27.3 2004)

1040 3596

4106 4369 11214 10684 9693

1 1 1 1 2 2 2 2 2 8

8 8 10 10 9

1995 1996 1997 1998 1999 2000 2001 2002 2003

1994 Fuel cycle licensee including fabrication, processing, and enrichment 1995 1996 1997 1998 1999

104 97 55 53 86 146 154 75 55

158

1

Independent spent fuel storage

185 27

Number of Monitored Individuals

1994

Number of Licensees Reporting 2 1 0

Calendar Year

(Continued)

1997 1998 1999

NRC Licensee Categorya

TABLE 27.3

2959 3061 3910 3613 3927

2847

49 53 24 21 33 83 107 67 46

89

50 13

1217 878 1006 950 1020

1147

51 54 6 3 5 6 13 6 3

42

5 1

Number of Workers with Measurable Collective TEDE TEDE (person-rem)

0.30 0.20 0.09 0.09 0.11

0.32

0.49 0.56 0.11 0.05 0.06 0.04 0.08 0.08 0.05

0.27

0.03 0.05

Average TEDE (rem)

0.41 0.29 0.26 0.26 0.26

0.40

1.04 1.02 0.24 0.12 0.16 0.07 0.12 0.09 0.06

0.47

0.11 0.10

Average Measurable TEDE per Worker (rem)

1041

303 305 306 303 290 286 283 275 243 223

109 109 109 105 104 104 104 104 104

1995 1996 1997 1998 1999 2000 2001 2002 2003 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003

109

1994

9 9 8 8

149173 143115 137430 142959 132069 129951 125868 118869 120769 122281

132226 126402 126781 114367 114154 110557 104928 107900 109990

139390

9336 8145 7937 7738

77890 77758 75366 75595 65213 66839 65695 60751 62307 63424

70821 68305 68372 57466 59216 57233 52292 54460 55967

71613

4649 3980 3886 3633

24910 25003 21828 19919 16406 16661 15940 14746 14850 14413

21688 18883 17149 13187 13666 12652 11109 12126 11956

21704

1339 1162 661 556

0.17 0.17 0.16 0.14 0.12 0.13 0.13 0.12 0.12 0.12

0.16 0.15 0.14 0.12 0.12 0.11 0.11 0.11 0.11

0.16

0.14 0.14 0.08 0.07

0.32 0.32 0.29 0.26 0.25 0.25 0.24 0.24 0.24 0.23

0.31 0.28 0.25 0.23 0.23 0.22 0.21 0.22 0.21

0.30

0.29 0.29 0.17 0.15

b

These categories consist only of NRC licensees. Agreement State licensed organizations are not required to report occupational exposure data to the NRC. As of 1999, these are no longer any NRC licensees involved in this activity. All low-level waste disposal facilities are now located in Agreement States and no longer report to the NRC. c Includes all LWRs in commercial operation for a full year of the years indicated. Reactor data have been corrected to account for the multiple counting of transient reactor workers.

a

Grand totals and averages

Commercial light water reactorsc

2000 2001 2002 2003

1042

SOURCES, LEVELS AND EFFECTS OF MANMADE IONIZING RADIATION

TABLE 27.4 Summary of Collective Effective Dose Equivalents to Regional Populations Caused by Radioactive Effluents from Fuel Cycle Facilities Facility

Collective Effective Dose Equivalent (person-rem/year)

Basis of Estimate

Mining Open pit-air Open pit-water Underground-air Underground-water Milling from model mill

1.0 0.2 10 21 62

Airborne effluent

Conversion Wet Dry Enrichment Fabrication

0.4 2.9 0.002–0.4 0.01–0.7

Plant airborne effluent in 1980 Plant airborne effluent in 1980 3 plants, airborne effluents in 1981 7 plants, airborne effluents in 1980

Reactor Air Water Low-level waste storage

0.003–13 0–40 <4

47 plants in 1980 47 plants in 1980 Maxey flats, estimate

Model mine Model mine

Source: NCRP (1987a).

maximally exposed individual as a result of airborne radioactive effluent from fuel cycle facilities (Table 27.5), the mining operations are responsible for the highest estimated doses of all NFC stages. In all cases, however, reactor operation represents only a small fraction of the total estimated radiation dose to either an individual worker or the general population.

TABLE 27.5 Summary of Radiation Doses to the Maximally Exposed Individual Caused by Airborne Radioactive Effluents from Fuel Cycle Facilities Facility

Effective Dose Equivalent (mrem/year)

Basis of Estimate

Mining Open pit Underground Milling

26 61 0.4–260

Model mine Model mine Eight typical mills

Conversion Wet Dry Enrichment Fabrication

0.8 3.2 <0. 1–0.4 <0.1–0.7

Plant, calculation Plant, calculation All three plants All seven plants

Reactor PWR BWR Low-level waste storage Transportation

0.6 0.1 <1 –

Model reactor Model reactor Maxey flats, estimate Not reported

Source: NCRP (1987a).

NUCLEAR WEAPONS COMPLEX

1043

The collective dose to the general public from all commercial reactor operations in 1991 was 88 person-rem (PNL, 1995).

27.6 NUCLEAR WEAPONS COMPLEX The Nuclear Weapons Complex is an industrial complex consisting of a collection of enormous factories devoted primarily to research, development, and production of nuclear weapons. It has resulted in a legacy of radioactive contamination and radiation exposure to radiation workers and the general public that began in the 1940s, and continues today. In many respects, the nuclear weapons production process is similar to the nuclear fuel cycle; that is, it involves the mining and milling of uranium, uranium enrichment, fuel fabrication, operation of production reactors, and the management of spent fuel, high-level waste, and low-level waste. Some of the important differences between the commercial nuclear fuel cycle and weapons production include the following: (1) The weapons production program continued throughout the cold war, and national security was an important consideration in the management of the program. (2) The program required a large investment in research and development, which greatly exacerbated the potential for worker exposure and environmental contamination. (3) By its very nature, weapons production is associated with the production of large quantities of very long-lived transuranic wastes, which must be isolated from the accessible environment virtually indefinitely. (4) The program required weapons testing, discussed in the next section, which resulted in elevated levels of radiation exposure near the test sites and globally as a result of global fallout, traces of which are still detectable in the environment. (5) Our knowledge of the fate and effects of radioactivity was limited until the program matured, which, in retrospect, resulted in elevated exposures and environmental contamination that could have been minimized or avoided. With the end of the cold war, an enormous amount of information became available characterizing the nature and extent of the radioactive contamination at the weapons complex sites and the levels of exposure that were experienced by both workers and members of the general public as a result of this legacy. To an extent, some of these exposures continue today due to limited ongoing weapons related activities, worker and public exposures associated with the cleanup of these sites, and exposure of the public to small amounts of residual radioactivity in the environment associated with historical operations. Prior to the end of the cold war, most of this information was classified. A better understanding of the legacy is also unfolding as a result of efforts to clean up and release many of these sites for other uses and to better understand past exposures of radiation workers and the general public. Some of the core documents, which capture the expanse of this legacy, are as follows: “Complex Cleanup: The Environmental Legacy of Nuclear Weapons Production,” Congress of the United States (Office of Technology Assessment, 1991). “Closing the Circle on Splitting of the Atom: The Environmental Legacy of Nuclear Weapons Production in the United States and What the Department of Energy is Doing About It,” U.S. Department of Energy, Office of Environmental Management, 1996.

1044

SOURCES, LEVELS AND EFFECTS OF MANMADE IONIZING RADIATION

“Estimating the Cold War Mortgage: The 1995 Baseline Environmental Management Report,” U.S. Department of Energy, Office of Environmental Management, March 1995. “Radiation Dose Reconstruction for Epidemiologic Uses,” National Academy of Sciences, National Research Council, 1995.

In addition, a more complete understanding of the exposures experienced by civilian workers at the weapons complex is emerging as a result of investigations mandated by the Energy Employees Occupational Illness Compensation Act (EEOICPA) of 2000. On July 31, 2001, the Department of Labor began processing claims filed under a new compensation program created by the EEOICPA. As part of this program, NIOSH has been preparing site profiles for each major DOE and weapons complex facility. These documents, which are on the NIOSH web site, contain an enormous amount of information characterizing the historical exposures experienced by DOE workers and atomic weapons employees at the weapons complex.

27.6.1

Extent of Environmental Contamination from the Weapons Complex

The weapons complex, depicted in Fig. 27.5, consists of 137 DOE sites in 33 states and 1 territory. In the past, the DOE published an annual report, Environmental Management (EM) (e.g., DOE, 1996b), which provided an overview of the status of environmental restoration and waste management for these sites. In addition, DOE had also published an annual report entitled Integrated Data Base—1995: U.S. Spent Fuel and Radioactive Waste Inventories, Projections, and Characteristics (e.g., DOE, 1996a), which described the quantities of the different types of waste attributable to the complex. These reports are no longer published by DOE, due to the rapidly changing status of the complex. In their place, an array of web sites can be accessed which describe the status of cleanup of each site. The reader is directed to DOE web sites for two of its offices, the Office of Environmental Management and the Office of Legacy Management (LM). EM is responsible for the cleanup of each site, and its web site describes the status of these cleanup activities. LM is responsible for the implementation of postclosure activities that are designed to ensure the protection of human health and the environment following site cleanup. LM has control and custody for legacy land, structures, and facilities and is responsible for maintaining them at levels suitable for their long-term use. Reports on the overall status of DOE’s Environmental Management Program are available in its annual report to Congress. The latest report available on the web is Topto-Bottom Review of Environmental Management Program: Status of Implementation, October 2003. In addition, a more recent comprehensive status report on the management of radioactive waste in the U.S. is available in United States of America Second Annual Report for the Joint Convention on the Safety of Spent Fuel Management and on the Safety of Radioactive Waste Management, United States Department of Energy, DOE/EM-0654, Rev 1, October 2005. Table 27.6 presents the projected cumulative inventories of radioactive waste at the DOE complex as of December 31, 1995. This table provides a useful baseline that represented DOE’s understanding of the inventory of the different types of waste at the beginning of the program. The current quantities and status of these wastes can only be determined by a careful review of annual reports issues by each site.

NUCLEAR WEAPONS COMPLEX

1045

TABLE 27.6 Types and Quantities of Radioactive Wastes Associated with DOE Weapons Complex and R&D Facilities (DOE, 1996a, Except Where Otherwise Indicated) Category of RadioActive Waste

Definition

Spent nuclear fuel

Irradiated fuel discharged from a nuclear reactor

High-level waste

Waste generated by the chemical reprocessing of spent research and production reactor fuel, irradiated targets, and naval propulsion fuel Radioactive waste not classified as high-level waste, transuranic waste, spent nuclear fuel, or byproduct material (e.g., not uranium or thorium tailings and waste) Waste containing elements with atomic numbers greater than 92 and half-lives greater than 20 years, in concentrations greater than 100 nCi/g of alphaemitting isotopes Very large volumes of bulk material containing low levels of radioactivity and which will be generated during the environmental restoration of the sites. Considered a type of LLW. Wastes that are associated with in situ management, where the material is not removed from its location, are not included Soil at DOE sites that contains very low levels of radioactivity and may not be excavated but left in place, perhaps with institutional controls. There may be some overlap with environmental restoration waste

Low-level waste

Transuranic waste (tru waste)

Environmental restoration waste

Large volumes of very low levels of contaminated soil

Cumulative Quantities Projected Through 2035 2,721.56 metric tons heavy metal (MTHM). Information on Curie inventory not available 18,948 m3 of HLW stored as glass in canisters. Peak radionuclide inventory of 425.75 million curies in year 2026 5.5 million m3 projected to be disposed through 2030. 12.55 million Ci inventory in 1995

1.46E5 m3 of as generated retrievably stored waste (does not include irretrievable TRU waste buried prior to the early 1970s) 43 million m3

107–108 m3a

a

Estimates of the volume of contaminated soil at DOE sites are provided in a draft EPA report entitled “Radiation Site Cleanup Regulations: Technical Support Document for the Development of Radionuclide Cleanup Levels for Soil,” EPA 402-R-96-011, September 1994. This report is part of the docket for EPA’s planned cleanup rule (40 CFR 196). Copies of the report are available from EPA Office of Radiation and Indoor Air.

The final disposition of SNF, HLW, LLW, and TRU waste is to isolate the waste from the environment accessible to man through the use of carefully sited and highly engineered waste isolation systems, which will be effective for as long as the sources of the waste are potentially hazardous. For many sites, cleanup has been completed, while at other sites,

1046

SOURCES, LEVELS AND EFFECTS OF MANMADE IONIZING RADIATION

cleanup is underway. The low-level waste has either been stabilized in place, and is part of legacy management, or disposed of at any one of several commercial or DOE low-level radioactive waste disposal facilities. Transuranic waste is being disposed of at the Waste Isolation Pilot Plant in Carlsbad, New Mexico. Like the commercial sector, the final disposition of DOE SNF and HLW is in question due to delays in the licensing, construction, and operation of the proposed Yucca Mountain repository. 27.6.2 Past, Present, and Future Occupational Exposures Attributable to the Weapons Complex and Its Legacy The DOE periodically prepares a report entitled “DOE Occupational Radiation Exposure,” the most recent of which is on the World Wide Web and covers the period from 2000 to 2004, but also provides individual and collective dose trends from 1974 to 2004. Table 27.7 presents the individual and collective doses to DOE workers, contractors, and subcontractors. As a point of comparison, from 1974 to 1986, the collective doses ranged from about 10,000 to 8,000 person-rem, and the individual average doses ranged from about 0.3 rem to 0.2. In general, the individual and collective doses began to decline sharply after 1986. The decline is attributed to the end of the cold war, the associated reduction in weapons production activities, and transition of the complex to a cleanup mode. These doses are relatively low as compared to the radiation protection standard of 5000 mrem/year for radiation workers (10 CFR 835) and the fact that the average member of the general public receives about 300 mrem/year from exposure to natural background radiation (NCRP, 1987c) and 40 mrem/year from diagnostic X-ray examinations (NCRP, 1989). Notwithstanding these relatively low overall occupational exposures, it is now believed that there are subgroups within the worker population that may have experienced relatively high levels of exposure. For example, Wiggs et al. (1991) reported on workers at the Mound TABLE 27.7 Individual and Collective Doses Received by DOE Employees, Contractors and Subcontractors (from DOE, 2004) Parameter

2000

2001

2002

2003

2004

DOE and contractor 129,653 130,884 133,703 136,710 136,353 workforce Number of workers monitored 102,881 97,818 100,221 102,509 100,011 Percent of workers monitored 79 75 76 75 73 Number monitored with 15,983 16,687 17,051 17,484 15,739 measurable dose Percent monitored with 16 17 17 17 16 measurable dose 1,267 1,232 1,360 1,445 1,094 Collective dose (person-rem, total effective dose commitment equivalent, TEDE) Average measurable TEDE 0.075 0.074 0.080 0.083 0.070 (rem) Number of Individuals 3 0 0 2 0 exceeding 5 rem TEDE

NUCLEAR WEAPONS COMPLEX

1047

facility that received elevated internal exposures from Po-210 exceeding 100 rem. NIOSH is currently investigating these sources of exposures under the EEOICPA. 27.6.3 Past, Present, and Future Public Exposures Attributable to the Weapons Complex and Its Legacy Exposures of the general public in the past from weapons complex facilities include exposures associated with routine airborne emissions and episodic events. Because of the size of the DOE facilities, the offsite exposures to the public from routine operations have been small. For example, a compilation of exposure data from airborne emissions for 27 DOE facilities in 1986 revealed organ doses ranging from a (small) fraction of a mrem/ year to a maximum of 25 mrem/year. The collective organ doses ranged from less than 1 person-rem/year to a maximum of 670 person-rem/year (EPA, 1989a). Notwithstanding these relatively low doses (as compared to natural background and the radiation protection standards) to the members of the public living in the vicinity of DOE facilities, there has been concern that, in the past, many of the weapons complex facilities had relatively large chronic and episodic releases that resulted in significantly elevated exposures of the nearby populations (Miller and Smith, 1996). In response to this concern, the DOE and the CentCDC implemented several environmental dose reconstruction studies, including dose reconstructions in the vicinity of the Feed Materials Production Center (FMPC) in Fernald, Ohio (Meyer et al., 1996), the Marshall Islands (Simon and Graham, 1996), Oak Ridge National Laboratory (Widner et al., 1996), the Nevada Test Site (Whicker et al., 1996), Rocky Flats (Mongan et al., 1996), and Hanford (Shipler et al., 1996). These dose reconstructions largely have been completed and CDC is determining whether follow-up epidemiologic studies or other public health activities should be taken. For example, the primary result of the Fernald dose and risk assessment project performed by CDC showed that the number of lung cancer deaths occurring within the community surrounding the FMPC site from 1951 through 2088 may be increased by 1–12% as a result of FMPC-related radiation exposures. Risk estimates provided by this project will also be a key component in assessing the feasibility of conducting in-depth epidemiologic investigations at Fernald. This finding and the status of the dose reconstruction research at many of the weapons complex facilities are provided on the web site for the Radiation Studies Branch of CDC. The current exposures to the general public from the weapons complex are small primarily because (1) the physical size of the sites places the public at large distances from the sources of emissions; (2) most of the sites are located in remote areas; and (3) most of the cold war research, development, and production activities have ceased. Each year, each site issues an annual environmental report, which presents estimates of the radiation doses to members of the public. For example, each year the Westinghouse Savannah River Company (WSRC) publishes “Savannah River Site—Environmental Report.” These annual reports present estimates of releases of radioactivity to the environment, the results of environmental radiological surveillance programs, and estimates of the doses to members of the general public. In the latest annual report for 2004 (Westinghouse, 2004), the WSRC estimated that the dose to the hypothetical maximally exposed member of the public from the liquid pathway, including drinking water and the ingestion of fish was 0.09 mrem in 2004. The estimated dose from atmospheric releases to the maximally exposed individual in 2004 was 0.06 mrem. The collective population dose in 2004 was 2.9 person-rem, as compared to natural sources, which was 214,000 person-rem (Westinghouse, 2004).

1048

SOURCES, LEVELS AND EFFECTS OF MANMADE IONIZING RADIATION

Notwithstanding these small doses, there is concern that existing inventories of radioactive material could be inadvertently released or that contamination that has entered the ground water on site could eventually migrate to wells offsite and expose large populations to elevated levels of radioactivity. In response to this concern, the weapons complex sites have been undergoing remediation under DOE’s Environmental Management program. Sites where remediation has been completed are under DOE’s Legacy Management program. The status of the remediation programs at each site is provided in “Top-to-Bottom Review of Environmental Management Program: Status of Implementation, Report to Congress.” The most recent published report is for 2003. In addition to the annual report to Congress, the DOE maintains a web site where detailed current information is provided on the status of each site.

27.7 LOCAL, TROPOSPHERIC, AND GLOBAL FALLOUT Between 1945 and 1980, 520 atmospheric weapons tests were carried out with a total yield of 545 megatons (Mt) equivalent of TNT. The tests were conducted by the United States, the USSR, the United Kingdom, France, and China. These tests introduced radioactive debris into the atmosphere, which resulted in local, tropospheric, and global fallout. Local fallout, which contributed to about 12% of the total radioactive fallout, is comprised of relatively large particles that were deposited in the immediate vicinity of the tests and within a few hours following the tests. Tropospheric fallout, which contributed to about 10% of the total fallout, consisted of smaller particles of radioactive debris that was introduced into the troposphere, which extends from ground level to a height of about 10 km. Tropospheric fallout occurred within days following the tests and in bands extending tens to hundreds to over 1600 km downwind in the direction of the prevailing winds (depending on the yield of the device, the elevation of the test, and the prevailing meteorological conditions) and within the latitudes within which the tests were conducted. Stratospheric fallout, which contributed to about 78% of the total fallout, originated from radioactive debris that was injected into the stratosphere, which extends from a height of about 10–30 km. Weapons tests with yields exceeding about 100 kt have sufficient energy to inject radioactive debris into the stratosphere. Debris that is introduced into the stratosphere is deposited globally over several months following the detonation (Eisenbud and Gesell, 1997; UNSCEAR, 1993). Retrospective dose reconstruction studies have attempted to reconstruct the radiation doses to members of the public living in the vicinity of the Nevada Test Site (Whicker et al., 1996; Thompson and McArthur, 1996) and the Marshall Islands (Simon and Graham, 1996) at the time of, and subsequent to, the tests. The dose reconstruction efforts for both the Marshall Islands and Nevada Test Site make use of radiological surveillance data, meteorological data, land use and demographic data, and mathematical models to reconstruct the exposures to members of the nearby populations at the time of the tests and for extended periods of time following the tests. In addition, follow-up epidemiological investigations have attempted to discern a statistically significant increase in adverse health effects in the exposed populations. A total of 66 tests were conducted in the Bikini and Enewetak Atolls in the Republic of the Marshall Islands (RMI). Assessments of the doses to individual populations from selected important tests and exposure pathways have been published, along with medical follow-up studies (Cronkite et al., 1997). For example, Cronkite et al. (1997) report on the accidental exposures and the medical follow-up following the Bravo test, which resulted in very high

LOCAL, TROPOSPHERIC, AND GLOBAL FALLOUT

1049

exposures (over 100 rem) to navy personnel and Marshallese inhabitants. These individuals experienced varying degrees of skin burns and symptoms of acute radiation exposure. Latent effects of radiation exposure were also subsequently observed in the exposed populations. The effects included thyroid dysfunction and thyroid cancer. These and other studies are described in the July 1997 issue of Health Physics, which was dedicated to providing an overview of the RMI studies performed to date. In addition, the predicted distribution of doses among adult women who might inhabit Rongelap Island in 1995 has been reported. The projected doses range up to 200 mrem/year, with a median between 75 and 100 mrem/ year EDE (Simon and Graham, 1996). The purpose of this study was to evaluate the degree to which resettlement of Rongelap island could result in radiation exposures that exceed the dose action level of 100 mrem/year above natural background. Issues surrounding the just compensation of the people of the Republic of the Marshall Islands for damage to their lands, the proper cleanup of the northern atolls that still contain significant levels of residual radioactivity in the soil, the loss of use of their land, and concerns about historical and continuing health impacts resulting from weapons test remain to this day. A process is under way to air these grievances before the Congress of the United States (RMI, 2000). These issues are addressed in detail in the transcripts of the public hearings held in Majuro, the capitol of the Marshall Islands, by the Nuclear Claims Tribunal, an adjudicatory body authorized un Public Law 99–239, to rule on compensation claims associated with weapons testing in the Marshall Islands. In addition, on May 25, 2005, a hearing was held before the Committee on Resources and the Subcommittee on Asia and the Pacific of the Committee on International Relations United States House of Representatives. At that hearing, Dr. Andre Bouville gave testimony on behalf of the National Cancer Institute (NCI) that approximately 500 additional cancers may develop in the population of the Marshall Islands as a result of radiation exposures associated with weapons testing in the Marshall Islands (Bouville, 2005). An enormous body of literature exists on the environmental radiation exposures from nuclear testing at the Nevada Test Site. The November 1990 issue of Health Physics was dedicated to this subject. The October 1996 issue of Health Physics provided several papers on this subject, and, in December 1996, a comprehensive report was published by the University of Nevada, Las Vegas entitled “Preliminary Risk Assessment DOE Sites in Nevada.” Estimates of the time integrated unshielded exposure at 34 locations in the vicinity of the tests range from 0.1 to 23.2 R (Thompson and McArthur, 1996). Follow-up epidemiologic studies have been performed (Simon et al., 1995; Till et al., 1995). In addition to exposures of the general public in the vicinity of the Nevada Test Site, there is a great deal of concern that civilian and military personnel may have experienced excessive exposures during and immediately following weapons testing at the Nevada Test Site, and that cancers that were experienced by many of these individuals may have been caused by those exposures. In order to more fully characterize and quantify the radiation exposures and health risks experienced by these individuals and compensate those individuals whose illnesses were more likely than not caused by these radiation exposures, the government implemented the Nuclear Test Personnel Review Program (under the direction of the Defense Treat Reduction Agency, DTRA) to adjudicate claims by veterans. In addition, the Congress has initiated the EEOICPA to adjudicate claims by civilians, under the Congressionally mandated direction of the Department of Labor and NIOSH. A vast body of literature characterizing the radiation exposures in the vicinity of the tests is on the DTRA and NIOSH web sites. Global fallout due to weapons testing has resulted in slightly elevated exposures above background to very large populations located primarily in the 40 to 50 latitude band in the

1050

SOURCES, LEVELS AND EFFECTS OF MANMADE IONIZING RADIATION

northern and southern hemispheres. The time-integrated effective dose from global fallout is estimated to be 370 mrem, and the collective dose is estimated to be 3.0E9 person-rem (UNSCEAR, 1993). These values can be placed into perspective considering that everyone receives approximately 300 mrem/year EDE from natural background radiation. As discussed in greater detail at the end of this chapter, these exposures could result in close to 0.5 million fatal cancers. A report by the NCI, entitled, “Estimated Exposures and Thyroid Doses Received by the American People from Iodine-131 in Fallout Following Nevada Atmospheric Nuclear Bomb Tests” is available on the NCI web site. The report contains results from a study to assess Americans’ exposures to radioactive iodine-131 fallout from atmospheric nuclear bomb tests carried out at the Nevada Test Site in the 1950s and 1960s. This report was mandated by Congress under Public Law 97-414. Results of the study show that, depending on their age at the time of the tests, where they lived, and what foods they consumed, particularly milk, Americans were exposed to varying levels of I-131 for about 2 months following each of the 90 tests. The results of the study revealed that the average cumulative thyroid exposure for all Americans from radioactive iodine in weapons testing fallout was about 2 rads, and more heavily affected areas experienced thyroid doses ranging from 9 to 15 rads. The NCI report estimates between 10,000 and 75,000 Americans may develop thyroid cancer during their lifetime because as children they were exposed to radioactive iodine fallout from nuclear weapons tests. Notwithstanding these projections, which are based on fallout measurements and theoretical mathematical models that relate radiation exposure to cancer risk, epidemiological investigations were not able to establish an empirically based correlation between areas with predicted high radioiodine exposures and areas with elevated rates of thyroid cancer. These relationships are difficult to establish because predicted increases in the incidence of thyroid cancers are obscured by the large and variable normal incidence of thyroid cancer relative to the predicted effects, or the theoretical dose/response models could be incorrect.

27.8 MEDICAL EXPOSURES The use of radioactive materials, radiation sources, and X-ray machines for medical purposes results in the exposure of the patients under the physicians care, the physicians and medical technicians that administer the care, and, to a limited degree, members of the public exposed to the radioactive waste products generated as a result of medical treatment. 27.8.1

Exposure of Patients

NCRP (1989a) and UNSCEAR (1993) presented overviews of the radiation exposures associated with the diagnostic and therapeutic use of radiation and radioactive materials. The diagnostic use of radiation includes primarily medical and dental X-rays and the diagnostic use of radiopharmaceuticals. The therapeutic use of radiation and radioactive materials includes primarily teletherapy, brachytherapy, and the therapeutic use of radiopharmaceuticals. Teletherapy is the use of strong, highly directed external sources of radiation, such as 60 Co units, to destroy deep-seated tumors. Brachytherapy is the use of sealed radioactive sources inserted into the body or placed on the surface of the skin to destroy tumors. Therapeutic nuclear medicine is the internal administration of radiopharmaceuticals, such as I-131, to treat various diseases. Diagnostic X-rays are the predominate contributor to the collective dose to the general public due to the medical uses of radiation.

MEDICAL EXPOSURES

1051

UNSCEAR (1993) estimates that there were approximately 800 medical diagnostic X-ray examinations and 402 dental X-ray examinations per year per 1000 people in the U.S. from 1986 to 1990. The radiation dose per examination varied widely depending on the type of examination. For example, the effective dose equivalent for medical diagnostic examinations ranged from 7 mrem for chest radiography to 460 mrem for lower GI tract examinations. The effective dose equivalent from dental X-ray examinations is relatively small, about 3 mrem per examination. In the U.S., the average effective dose equivalent is about 50 mrem per medical (nondental) examination and about 40 mrem/year per capita. Worldwide, the total collective dose associated with medical X-ray examinations is 1.6E8 person-rem/year. The number of nuclear medicine examination in countries with advanced medical care programs is estimated to be 16.4 per 1000 people, with an average effective dose per examination of 570 mrem, that is, about 10 times higher than the average dose per X-ray examination. However, because of the lower frequency of use, the average annual dose per person from nuclear medicine examinations in advanced health care nations is 9.4 mrem (UNSCEAR, 1993). In countries with advanced health care programs, the frequency of radiotherapy treatments by teletherapy and brachytherapy is estimated to be 2.4 per 1000 people. The doses to individual organs and localized tissues are extremely high (in excess of 1000 rem), and no attempt is made here to estimate the effective dose equivalent per treatment. The worldwide collective dose associated with the therapeutic use of radiation and radioactive materials is estimated to be 1.5E8 person-rem/year, or about 80% of the collective doses associated with the diagnostic use of radiation (UNSCEAR, 1993). NCRP Report No. 100 (NCRP, 1989a) reports on exposures of the U.S. population from diagnostic medical radiation and updates some of the information provided in the 1980 survey. The report concludes that the annual per caput effective dose equivalent to the U.S. population from diagnostic X-rays is 40 mrem, and that from diagnostic nuclear medicine is 14 mrem, for a total of 54 mrem EDE. UNSCEAR (1993) estimates that the dose per capita from all sources of medical exposures is 60 mrem/year, and the total, worldwide collective effective dose equivalent is 3.3E8 person-rem/year. This is as compared to 1.3E9 person-rem/year effective dose equivalent from natural background radiation, including radon. With respect to future trends, it is projected that the total use of diagnostic X-radiation will increase because of the increasing proportion of older people in the population and increasing urbanization. Because of the important contribution of the diagnostic use of medical X-rays to the individual and collective dose to the U.S. population, the Conference of Radiation Control Program Directors (CRCPD), in collaboration with the Food and Drug Administration, documents the state of the diagnostic use of X-rays with its annual Nationwide Evaluation of X-ray Trends (NEXT) survey program (CRCPD, 2005). Based on a survey 239 chest X-rays performed in 2001, the mean entrance skin exposure was found to be 12 mrad per examination with a standard deviation of 7 mrad. The maximum observed entrance skin dose was 46 mrad. In the 1980s, the medical profession began to make widespread use of computed tomography (CT) as a diagnostic tool. It’s value as a diagnostic tool over conventional Xrays is its ability to create high resolution 3-dimensional images for the detection of coronary artery calcification, lung cancer, colon polyps or masses, and other abnormalities, such as benign or malignant tumors (HPS, 2003). The technique is an extremely effective diagnostic tool for symptomatic patients. However, it is being used increasingly as a screening tool for healthy, asymptomatic patients. The use of CT technology in this manner became of concern to the Health Physics Society because one of its disadvantages it that it is associated with

1052

SOURCES, LEVELS AND EFFECTS OF MANMADE IONIZING RADIATION

relatively high radiation doses. On average, the effective dose equivalent was reported as 320 mrem per examination in 1987 (UNSCEAR, 1993). The European Commission reported effective dosesof230 mrem,800 mrem,and1000 mrem perCTexaminationofthehead,chest, andabdomen orpelvis,respectively (EC, 2000).Inresponsetothis concern,in 2003,the Health Physics Society issued a position paper recommending against the use of CT for screening purposes since its risks are believed to outweigh its benefits when used in this manner. A new publication in Health Physics (Mettler, et. al. 2008) reveals that the per capita annual dose from medical exposures (not including dental or radiotherapy) had increased almost 600% to about 3.0 mSv (300 mrem) and the collective dose had increased over 700% to about 900,000 person Sv (90 million person-rem). The largest contributors and increases have come primarily from CT scanning and nuclear medicine. Mettler, Fred A., Thomadsen, B. R, Bhargavan, M, Gilley, D.B., Gray, J.E., Lipoti, J. A., McCrohan, J, Yoshizumi, T.T., and Mahesh, M. (2008). Medical Radiation Exposure in the U.S. in 2006: Preliminary Results. Health Physics 95(5): 502–507. 27.8.2

Occupational Exposure to Medical Technicians

NCRP 1989a and EPA (1984) presented compilations of occupational radiation exposures for 1980. The information in these reports is dated, and is currently undergoing revision by the EPA. However, the information is useful for providing a general overview of the magnitude of the occupational exposures for medical technicians. Table 27.8 presents the occupational exposure for medical technicians for 1980. NCRP Report No. 124 (NCRP, 1996) describes the sources and magnitudes of occupational and public exposures from nuclear medicine procedures. Exposures are associated with the receipt and handling of the radioactive material and the practice of nuclear medicine. The annual dose to technical personnel working with the material and the patients (and who are under radiation protection controls) is reported to be 400 mrem/year (a typical value for radiation workers). The dose to the members of the public in close contact with a patient being treated with radiopharmaceuticals is estimated to be <1 mrem/year to 21 mrem per procedure. The per capita dose to the American public from exposure to patients treated with radiophamaceuticals is considerably less than 1 mrem/year. TABLE 27.8 EPA, 1984)

Occupational Radiation Exposures to Medical Technicians in 1980 (from

Number of Workers (thousands)

Number of Workers who Received a Measurable Dose

Mean Annual Dose (mrem/year) to the Workers with MeaSurable Levels of Exposure

Collective Dose (person-rem/year) (thousands)

Dentistry Private practice Hospital Vetinary Chiropractic Podiatry

259 155

82 87

70 180

5.6 16

126 21 15 8

86 12 6 3

200 110 80 30

17 1.3 0.5 0.1

Total

584

277

Occupational Category

41

MEDICAL EXPOSURES

27.8.3

1053

Exposure to the General Public from Misplaced Brachytherapy Sources

The New York State Department of Health (New York, 1982) reported on a widespread problem associated with exposure to jewellery containing elevated levels of radioactivity. These exposures occurred as a result of the reuse of “seeds” cut or crimped from hollow gold tubes through which radon had been passed of hollow gold tubes and used as implants for the irradiation of diseased tissue. The NYS study surveyed 160,000 pieces of jewellery and found about 170 pieces to be radioactive. Nine individuals were identified who developed squamous cell carcinoma as a result of the long-term exposure to the radioactive jewellery. Radon seeds are no longer manufactured in the U.S. (Lubenau and Nussbaumer, 1986). The mishandling of radium needles, also used as medical implants, has also resulted in environmental contamination and elevated radiation exposures to members of the public (Belanger and Jasnosik, 1990; Googins, 1990). 27.8.4

Exposure to the General Public from Effluents

The medical use of radioisotopes is associated with the discharge of gaseous, liquid, and solid radioactive waste, which have some, but limited potential to cause radiation exposures of members of the general public. The discharges are under the direct regulatory control of the U.S. NRC or State Authorities. Of the approximate 3680 hospitals in the U.S., half handle radiopharmaceuticals for radionuclide imaging and to aid in diagnosis of disease, a smaller number use radionuclides for therapeutic purposes. Most hospitals are located in highly populated areas. A survey of 100 hospitals revealed that the primary gaseous emissions are Xe-133 and I-131, at a rate of 1.0 and 0.01 Ci/year, respectively (EPA, 1989a). The highest radiation doses to members of the general public in the vicinity of the hospitals is estimated to range from <1 mrem/year to a high of 8 mrem/year (EPA, 1993b). Accordingly, this source of exposure is relatively small. Small quantities of radioactive liquid waste may be discharged to sewage under current NRC regulations. There has been concern that these discharges may accumulate in the sludge at sewage treatment plants, and result in exposures of workers at sewage treatment plants and members of the public associated with the practice of applying the sludge to soil as a form of fertilizer. The work performed to date on this concern indicates that any exposures are extremely small (Shearer et al., 1995; Larsen et al., 1995). In response to concern over the presence of radioactive materials in water treatment plants, the Interagency Standards Committee on Radiation Standards (ISCORS) performed a series of investigations into this issue and published three reports (ISCORS, 2003, 2004a, 2004b). From 1998 to 2000, ISCORS conducted a joint survey to collect information on radioactivity in sludge and ash from sewage treatment plants referred to in the industry as publicly owned treatment works (POTWs). Questionnaires were sent to 631 POTWs requesting information related to wastewater sources, wastewater and sludge treatment processes, and sewage sludge disposal practices. A total of 420 questionnaires were returned, and out these, 313 POTWs were samples for the radionuclide content in sludge and ash samples. In general, the results of the analyses revealed that samples contained primarily naturally occurring radioactivity, such as radium, and that the levels are generally comparable to what is found in other media, such as soil and fertilizer. Using the results of the sampling analyses, an assessment of the possible doses to workers and the general public was performed. The results revealed that, depending on the exposure scenario and pathway, the effective dose equivalent ranged from less than 1 mrem/year to a theoretical high-end value

1054

SOURCES, LEVELS AND EFFECTS OF MANMADE IONIZING RADIATION

of 260 mrem/year. The highest doses were to theoretical members of the general public that might reside on property where POTW sludge is deposited for over a 100 year time period. The principal radionuclides of concern are radon and 226 Ra. These radionuclides are primarily associated with the treatment of groundwater with elevated levels of naturally occurring radionuclides, as opposed to the processing wastewater discharges from NRC licensed or DOE facilities. These investigations are continuing. Finally, the licensed disposal of the solid radioactive waste produced by the medical community provides a high level of assurance that the exposures to the public will be relatively small. For example, analyses of the performance of low-level radioactive waste disposal units have revealed maximum doses less than a few mrem/year (EPA, 1988e).

27.9 INDUSTRIAL USES (OTHER THAN THE NUCLEAR FUEL CYCLE) Radioactive materials are used in a wide range of industrial applications that are associated with exposures to workers in the industry and members of the public. 27.9.1

Worker Exposures

The principal industrial exposures are associated with the use of sealed sources for industrial radiography and the manufacture and distribution of radioisotopes and devices for industrial purposes, research, medical uses, and in commercial products, such as smoke detectors. Table 27.9 presents the occupational exposures for industrial radiographers and manufacturers and distributors for 1995 (NRC, 1997). 27.9.2

Exposures of the General Public

Exposure of the general public as a result of the routine manufacture, use, and disposal of industrial sources is small. However, occasionally, a source will be abandoned, lost, or stolen and create the potential for widespread contamination and exposure of the public. The most serious of these accidents, which involved abandoned teletherapy units, occurred in Juarez, Mexico in 1983 and Goiania, Brazil in 1987. A summary of these accidents is provided in Eisenbud and Gesell (1997). In the Juarez accident, some individuals received doses as high as 700 rem, while several thousand people received some elevated doses. Those individuals

TABLE 27.9

Occupational Exposures of Industrial Workers in 1995 (NRC, 1997) Average Workers with Measurable Measurable Dose (mrem/ year) Dose

Collective Dose (person-rem/ year)

Number of Licensees

Number of Monitored Workers

Industrial radiographers Manufacturers and distributors

139

3530

2465

540

1338

36

2666

1222

490

595

Total

175

6196

3687

524

1933

Industrial Category

CONSUMER PRODUCTS

1055

that received the higher doses experienced mild symptoms of radiation exposure, but there were no acute radiation exposure fatalities, primarily because the exposures were protracted over a period of time. The Goiania accident was more serious in that it resulted in the deaths of four individuals due to acute radiation injury. Yusko (1995) reports on some of the more serious accidents, including exposure to a sealed source in Estonia in 1993, which resulted in a fatality, and exposures in buildings in Taiwan and Japan where the structural reinforcing rods in the concrete contained 60 Co causing exposures in excess of 100 rem. Misplaced sources are often detected at scrap handling facilities and steel mills and reported to State authorities. Yusko (1995), who maintains a database of such reports, found that most incidents are the result of the detection of elevated levels of naturally occurring radioactivity, which have relatively little potential for causing significant exposures (exposures to elevated levels of naturally occurring radioactivity are discussed in Section 11). However, about one-eighth of the recorded incidents involve sealed sources, primarily of 226 Ra, 60 Co, and 137 Cs, which can cause dangerously high exposures, as evidenced by the Juarez, Goiania, Taiwan, and Japan experiences. The number of recorded incidents is increasing, and, at the time of the publication of his paper, Yusko recorded over 1200 such incidents. In response to this serious problem, the Institute of Scrap Recycling Industries (ISRI), the CRCPD, EPA, NRC, and other federal agencies and trade organizations have been aggressively addressing this problem (Yusko, 1995). In addition, the NRC is providing funding to CRCPD to establish, implement, and manage an orphan source program and is participating in cooperative agreements with the EPA, DOE, and IAEA (Travers, 2000). Also, the NRC issued a policy statement to establish base civil penalties for loss, abandonment, or improper transfer or disposal of sources (Federal Register Vol. 65, No. 243, pages 79139–19140, December 18, 2000.). 27.10 CONSUMER PRODUCTS NCRP has dedicated its Report No. 95 to this topic (NCRP, 1987d). Examples of consumer products that are widely distributed and can result in radiation exposures of members of the general public include . . . . . . . .

smoke detectors, airport luggage inspection systems, radioluminous products, static eliminators, televisions and video display terminals, plutonium powered cardiac pacemakers, lighting rods, and a wide variety of products that contain discrete enhanced sources of naturally occurring radioactivity.

Some of these products are designed and intended to generate a radiation field, while the radioactivity or radiation fields associated with others are an inadvertent by-product of their design and/or operation. Table 27.10 summarizes the individual and collective doses

1056

SOURCES, LEVELS AND EFFECTS OF MANMADE IONIZING RADIATION

TABLE 27.10 Radiation Exposure from Consumer Products (Not Including Diffuse NORM, Derived from NCRP, 1987d)

Source Electronic products Televisions Video display terminals Airport luggage inspection Luminous watches and clocks H-3 activated watches H-3 activated clocks Pm-147 luminous watches Pm-147 luminous clocks Static eliminators Electron tubes Smoke detectors Check sources Dental prosthesis Ophthalmic glass Thorium products Gas mantles Tungsten welding rods Fluorescent lamp starters Aircraft transport of radioactive materials Total

Number of People Exposed in the U.S.

Average Annual Effective Dose Equivalent to the Exposed Population (mrem/year)

Annual Collective Effective Population Dose Equivalent (person-rem/ year)

1 mrem

28,000

1 mrem

60

230,000,000 50,000,000 30,000,000

<1 mrem

1,520

10,000,000 750,000 3,700,000 4,400,000 40,000 230,000,000 100,000,000 800,000 45,000,000 50,000,000

<1 mrem 1 mrem 1 mrem <1 mrem 1 mrem <1 mrem

50,000,000 300,000

<1 mrem 16 mrem

50,000,000

1 mrem

14,000,000

<1 mrem

Less than 1 to a few mrem from all sources combined for the vast majority of people

13 1,000 800 <800 3,000 <20,000 8,600 5,000 <1 3,000

20,000–50,000

OVERVIEW OF POTENTIAL HEALTH IMPACTS OF NATURAL AND MANMADE

1057

associated with consumer products and other miscellaneous sources, including enhanced sources of discrete naturally occurring radioactivity. 27.10.1 Exposures to Diffuse, Enhanced Sources of Naturally Occurring Radioactive Material (NORM) and Radiation The average total effective dose equivalent for a member of the population in the U.S. and Canada from natural background radiation is about 300 mrem/year (NCRP Report No. 94, 1987). This includes exposure to cosmic rays, cosmogenic radionuclides produced by the interaction of cosmic rays with atoms in the atmosphere or on the earth, terrestrial radiation from naturally occurring radionuclides in soil and rock, the inhalation of radon and its progeny, and naturally occurring radionuclides deposited internally into the body. Table 27.11 presents a summary of the annual background radiation exposures in each state of the U.S. from terrestrial radiation, cosmic radiation, and indoor radon. The dose equivalents for terrestrial and cosmic radiation for each state obtained from Bogen and Goldin (1981) were added to the average dose equivalents for indoor radon derived from the EPA’s National Residential Radon Survey for each state (EPA, 1993d). The indoor radon data, which were originally published in units of pCi/L, were converted to effective whole body dose equivalent by assuming 200 mrem/year EDE/pCi/L of measured indoor radon, in accordance with NCRP 1987b (Table 2.4). In additional to these natural sources of radiation exposure, the population of the U.S. is also exposed to enhanced sources of naturally occurring radionuclides and radiation due to anthropomorphic activities. The sources of these elevated exposures and their magnitude are described in several publications, including UNSCEAR (1993) and Eisenbud and Gesell (1997), and several NCRP reports (NCRP 1984, 1987b, 1987c, 1987d, 1993 NCRP Reports No. 77 (1984), 93 (1987b), 94 (1987c), 95 (1987d), and 118 (1993)). In addition, because of the large number of people that are being exposed to elevated levels of naturally occurring radioactivity resulting from man’s activities, the EPA has been investigating this issue (SCA, 1993). Table 27.12 presents the sources of enhanced levels of naturally occurring radiation, the levels of exposures to individual, and the collective exposures. A distinction is made here between specific consumer products and electronic devices, as discussed above, and exposure to large volumes of bulk material enhanced in natural radioactivity due to man’s activities. The management of the very large volumes of this material is being evaluated by the EPA. These evaluations include the life cycle costs and benefits associated with alternative management strategies, including the no action alternative.

27.11 OVERVIEW OF POTENTIAL HEALTH IMPACTS OF NATURAL AND MANMADE SOURCES OF RADIOACTIVITY The production and use of radioactive materials and a wide variety of human endeavor are associated with increased levels of radiation exposure. Table 27.13 provides an overview of the magnitude of these exposures and compares them to natural background. Radiobiologists and epidemiologists have been involved in ongoing studies to determine if these relatively low exposures have an adverse impact on public health, have no impact, or possibly have some net beneficial or hormetic effect on the exposed

TABLE 27.11 Total Average Annual Doses (mrem/year EDE) from Cosmic Radiation, Terrestrial Radiation, and Indoor Radon Alabama Alaska Arizona Arkansas California Colorado Connecticut Delaware District of Columbia Florida Georgia Hawaii Idaho Illinois Indiana Iowa Kansas Kentucky Louisiana Maine Maryland Massachusetts Michigan Minnesota Mississippi Missouri Montana Nebraska Nevada New Hampshire New Jersey New Mexico New York North Carolina North Dakota Ohio Oklahoma Oregon Pennsylvania Rhode Island South Carolina South Dakota Tennessee Texas Utah Vermont Virginia Washington West Virginia Wisconsin Wyoming National Average 1 National Average 2

1058

27.1 26.6 31.5 27.5 26.8 47.5 26.4 26.3 26.4 26.2 27.6 26.3 36.8 27.4 27.6 28.3 29.2 27.7 26.6 26.8 26.4 26.4 27.6 28.5 26.6 27.6 36.3 29.3 36.6 27.3 26.2 45.7 26.5 27.8 29.9 27.7 29.0 27.4 27.2 26.3 25.9 30.7 27.6 28.1 41.8 27.3 27.2 26.9 28.9 27.8 50.4 29.5 29.5

22.5 29.2 29.2 19.1 23.2 42.6 32.7 20.1 22.7 14.3 25.7 29.2 29.2 26.6 28.7 29.2 29.2 27.8 14.6 29.2 20.7 29.0 29.2 25.1 14.6 28.7 29.2 29.2 21.2 29.2 28.0 33.7 28.8 24.4 29.2 28.0 28.8 29.2 23.2 27.4 23.4 29.2 25.1 18.2 29.2 29.2 21.4 29.2 29.9 29.2 29.2 26.6 26.6

170 97 250 142 126 610 180 112 No data 91 273 No data 342 343 401 727 474 470 No data 286 476 228 226 383 160 350 No data 361 164 378 98 269 223 268 730 417 247 99 293 No data No data 903 511 165 196 No data 260 79 197 293 260 303 294

219.6 152.8 310.7 188.6 176 700.1 239.1 158.4 Not enough 131.5 326.3 Not enough 408 397 457.3 784.5 532.4 525.5 Not enough 342 523.1 283.4 282.8 436.6 201.2 406.3 Not enough 419.5 221.8 434.5 152.2 348.4 278.3 320.2 789.1 472.7 304.8 155.6 343.4 Not enough Not enough 962.9 563.7 211.3 267 Not enough 308.6 135.1 255.8 350 339.6 359.1 350.1

data

data

data

data

data data

data

1059

Oil and gas scale and sludge

Coal ash, fly ash, bottom ash and slag

Phosphate fertilizers

Phosphate waste, phosphogypsum, slag, scale

Uranium mining overburden

Source The overburden, low grade ore, and spoils associated with uranium mining contains slightly elevated levels of naturally occurring radioactivity The mining of phosphate rock for phosphate for fertilizers, detergent, and numerous phosphate products generates huge volumes of tailings containing elevated levels of naturally occurring radioactivity Fertilizer contains elevated levels of naturally occurring radionuclides Very large volumes of coal ash are produced each year containing elevated levels of naturally occurring radionuclides Large volumes of scale and sludge are produced in the oil and gas industry

Description

About 260,000 Mt per year are produced with an average Ra226 concentration of about 90 pCi/g

About 5 million Mt are produced per year containing an average of 8.3 pCi/g of Ra-226 About 61 million Mt are produced per year containing an average of about 3.7 pCi/g of Ra-226

About 50 million Mt per year containing an average Ra-226 content of about 35 pCi/g with some scale containing over 1000 pCi/g

About 38 million Mt per year is produced containing an average of 25 pCi/g of Ra-226

Average Radionuclide Concentration (pCi/g) a

a

a

a

a

a

a

a

a

Collective Exposures (personrem/year EDE)

a

Individual Exposures (mrem/year)

(continued)

SCA, 1993

SCA, 1993

SCA, 1993

SCA, 1993

SCA, 1993

References

TABLE 27.12 Exposures to Enhanced Sources of Naturally Occurring Radioactive Material (NORM) and Radiation—Sources, Quantities, Characteristic, and Exposures

1060 The overburden, low grade ore, and spoils associated with metal mining contains slightly elevated levels of naturally occurring radioactivity

Metal mining and processing, rare earths, Zr, Ha, Ti, Sb, and large volume industries

Tobacco products

Naturally occurring Pb210 and Po-210 accumulate on tobacco leaves and are transported to and deposited in the lungs, where they deliver relatively high localized radiation doses

Sludges produced in water treatment systems contain elevated levels of naturally occurring radionuclides

Water treatment, sludges, radium selective resins

Geothermal energy wastes

Description

Source

TABLE 27.12 (Continued)

About 54,000 Mt per year of geothermal waste is produced containing an average Ra-226 concentration of 132 pCi/g NA

About 300,000 Mt are produced per year containing an average Ra-226 concentration of 16 pCi/g. However, radium selective resins can have Ra226 concentrations as high as 35,000 pCi/g About 1E9 Mt are produced per year containing an average Ra226 concentration of about 5 pCi/g, with some material containing 900 pCi/g

Average Radionuclide Concentration (pCi/g)

a

a

NC

a

a

About 16,000 mrem/year to lung tissue of smokersb

a

Collective Exposures (personrem/year EDE)

a

Individual Exposures (mrem/year)

NCRP, 1987

SCA, 1993

SCA, 1993

SCA, 1993

References

1061

Mining and agricultural products

Road construction materials

Building materials

Air travel

Elevated exposure to pilots, flight attendants, and passengers due to the higher levels of cosmic radiation at high altitudes Living in structures made with naturally occurring uranium, thorium, and potassium in wall board, cement and other building products Exposure of travelers to roadways made of material with elevated levels of natural radioactivity Fertilizer contains elevated levels of naturally occurring radionuclides, and people who handle fertilizer can receive elevated exposures. Also the fertilizer distributed in the environment can increase external exposures and internal exposures from radionuclides in food NC

NA

2E4 person-rem/ year EDE

<2E5 person-rem/ year

0.5–5 mrem/year EDE

8.4E5 person-rem/ year EDE

4 mrem/year EDE

About 0.6 mrem/h of flight at 35,000 ft. During solar flares, 10 mrem/h of flight at 41,000 ft. Air crews receive 20– 910 mrem/year 7 mrem/year EDE

NCRP, 1987d

NCRP, 1987d

NCRP, 1987d

DOT, 1990

1062 Coal and coal ash contain elevated levels of natural radionuclides, which are distributed into the environment in the airborne and liquid effluents and in solid wastes. Oil represents a much smaller source of exposure. Natural gas use in the home represents a source of indoor radon, but which is small compared to radon normally present in homes

Combustible fuels: coal, oil, natural gas, and LPG

About 9 pCi/g of U-238 in coal fly ash

Average Radionuclide Concentration (pCi/g) <1–1 mrem/year EDE

Individual Exposures (mrem/year) About 1E5 personrem/year

Collective Exposures (personrem/year EDE)

NCRP, 1987d

References

Dose and risk assessments associated with these large volumes of bulk material containing slightly elevated levels of naturally occurring radionuclides, primarily Ra-226, are being performed by EPA. SCA (1993) is a contractor report that is undergoing review and revision as part of the EPA rulemaking investigations pertaining to NORM. b NCRP Report No. 95 does not convert this lung dose to an effective dose equivalent dose, but does report on others that have made the conversion and estimate a dose of 1300 mrem/year EDE per smoker.

a

Description

Source

TABLE 27.12 (Continued)

OVERVIEW OF POTENTIAL HEALTH IMPACTS OF NATURAL AND MANMADE

1063

TABLE 27.13 Summary of Radiation Doses from the Principal Sources Source of Exposure

Individual Dose or Dose Rate

27 mrem/year 28 mrem/year 1 mrem/year 200 mrem/year 40 mrem/year 300 mrem/year

Commercial nuclear fuel cycle

Radiation workers

NRC, 1997; NCRP, 1987a; PNL, 1995 About 300 mrem/year

Members of the general public <1 to a few mrem/year from reactor effluents to 61 mrem/year from mine effluents

The weapons complex Occupational exposure General public (routine emissions) General public (episodic releases)

About 23,000 person-rem/year in U.S. 75–1800 personrem/year from all reactor effluents combined in the U.S. from 1995 to 1991

65 mrem/year (1994)

1,643 person-rem/ DOE, 1994 year (1994) <1 to 25 mrem/year organ <1 to 670 person- EPA 89 dose rem/year depending on site Under investigation by CDC

Fallout

Local at time of test at Republic of the Marshall Islands Local at test site at Republic of the Marshall Islands Local at test site Nevada Test Site Global

References

7.5E7 person-rem/ NCRP No. 94 year in U.S. (based on a U.S. population of 250 million people)

Natural background

Cosmic rays (external) Terrestrial (external Cosmogenic (internal) Inhaled (radon, internal) Internal terrestrial Total

Collective Dose

3E9 person-rem globally timeintegrated

Cronkite et al., 1997; Simon and Graham, 1996; Thompson and McArthur, 1996; UNSCEAR, 1993

Over 100 rem to some individuals Up to 200 mrem/year 0.1–23.2 R 370 mrem time integrated

(continued)

1064

SOURCES, LEVELS AND EFFECTS OF MANMADE IONIZING RADIATION

TABLE 27.13 (Continued) Individual Dose or Dose Rate

Source of Exposure

Collective Dose

References

Medical exposures

UNSCEAR, 1993; EPA, 1993a, 1993b; NCRP, 1989a

Patients medical diagnostic exposures* Occupational exposures

60 mrem/year per caput*

Public exposure from misplaced sources Public exposures from routine effluents

Very high potential doses

140 mrem/year

3.3E8 person-rem/ year globally 41,000 person-rem/ year in U.S in 1980

<1 to 8 mrem/year

Industrial uses

NRC, 1997; Yusko, 1995

Radiation workers

524 mrem/year

1933 person-rem/ year

General public routine General public misplaced sources

Negligible Potential for very high doses

Consumer products Consumers

<1 to a few mrem/year

20,000–50,000 person-rem/year

NCRP, 1987

NORM

Under study as EPA

Under study at EPA

population. The most authoritative advisory and regulatory bodies on this subject have made the following observations and conclusions regarding exposure to low levels of radiation: ICRP (1991) cites the following risk coefficients for low dose and low dose rate irradiation. Nominal Probability Coefficients for Stochastic Effects (Detriment Expressed in Units of 10 2/Sv)a Exposed Population Adult workers Whole population a b

Fatal Cancerb

Nonfatal Cancer

Severe Hereditary Effects

Total

4.0 5.0

0.8 1.0

0.8 1.3

5.6 7.3

Rounded values, a Sv is equal to 100 rem. For fatal cancer, the detriment coefficient is equal to the probability coefficient.

“. . .the use of a nominal value of 5%/Sv for mortality due to leukemia and solid cancers from irradiation at low doses for a population of all ages (4%/Sv for an adult working population) still seems valid to the Committee.” (UNSCEAR, 1994) *

These estimates have recently been increased substantially (see Mettler et al 2008).

OVERVIEW OF POTENTIAL HEALTH IMPACTS OF NATURAL AND MANMADE

1065

“Of the various types of biomedical effects that may result from irradiation at low doses and low dose rates, alteration of genes and chromosomes remain the best documented. . .. It is estimated that at least 1 Gy (100 rad) of low dose rate, low LET radiation is required to double the mutation rate in man. . .. The population-weighted average lifetime excess risk of death from cancer following an acute dose equivalent to all body organs of 0.1 Sv (0.1 Gy of low LET radiation) is estimated to be 0.8%.” (NAS, 1990) “. . .the calculated risk of a premature cancer death attributable to uniform, whole body, lowLET irradiation is about 5.1  10 2/Gy. The corresponding incidence risk (neglecting nonfatal skin cancer) is about 7.5  10 2/Gy.” (EPA, 1994c)

In referring to the data from Japan, the ICRP concluded “Although the study group is large (about 80,000), excess numbers of malignancies, statistically significant at the 95% level, can be found only at doses exceeding 0.2 Sv. . .. It must also be borne in mind that all the doses to the Japanese study group were incurred at very high dose rates. . .” (ICRP, 1991). “The Committee’s estimates of radiation exposure and its estimates of risk of exposure indicate that radiation is a weak carcinogen. About 4% of the deaths due to cancer can be attributed to ionizing radiation, most of which comes from natural sources that are not susceptible to control by man. Nevertheless, it is widely (but wrongly) believed that all the cancer deaths at Hiroshima and Nagasaki are the result of the atomic bombings. The studies in the two cities have included virtually all of the heavily exposed individuals and have shown that of 3,350 cancer deaths, only about 350 could be attributed to radiation exposures from the atomic bombings.” (UNSCEAR, 1993)

In 2005, The National Academy of Sciences published “Health Risks from Exposure to Low Levels of Ionizing radiation” BEIR VII-Phase 2. This report is the seventh in a series of reports from the National Research Council prepared to advise the U.S. government on the relationship between exposure to ionizing radiation and human health. Table 27.14 presents the results of those investigations.

TABLE 27.14 The Committees Preferred Estimates of the Lifetime Attributable Risk (LAR) of Incidence and Mortality for All Solid Cancers and for Leukemia with 95% Subjective Confidence Interval All Solid Cancers

Excess cases (including nonfatal cases) from exposure to 0.1 Gy (19 rad) Number of cases in the absence of exposure Excess deaths from exposure to 0.1 Gy (10 rad) Number of deaths in the absence of exposure

Leukemia

Males

Females

Males

Females

800 (400, 1600)

1300 (690, 2500)

100 (30, 300)

70 (20, 250

45,500

36,900

830

590

410 (200, 830)

610 (300, 1200)

70 (20, 220)

50 (10, 190)

22,100

17,500

710

530

Number of cases or deaths per 100,000 persons exposed (from Table ES-1 of NAS, 2005).

1066

SOURCES, LEVELS AND EFFECTS OF MANMADE IONIZING RADIATION

TABLE 27.15 Potential Fatal Cancers Due to Exposure to Manmade Sources of Radiationa Source Actual incidence of fatal cancer Natural background Nuclear fuel cycle in U.S. Weapons complex Fallout Medical exposures Industrial uses in U.S. Consumer products in U.S. Enhanced NORM

Upper end estimate of fatal cancers due to exposure to radiation About 500,000 per year in the U.S. (National Cancer Institute Statistics) 37,500 per year in U.S. About 10 per year in U.S. (mostly radiation workers) About 1 per year in U.S. (predominantly workers) 1.5 million time-integrated, global 165,000 per year globally (due to diagnostic X-rays) 1 per year in U.S. from other than fuel cycle facilities, primarily workers 25 per year in U.S. to consumers Undetermined but relatively large

a Except for the actual incidence of fatal cancer, which is based on actuarial data, the estimates of radiogenic fatal cancers are theoretical upper end values that were derived using the collective doses presented in Table 27.11 multiplied by a fatal cancer risk coefficient of 5E4 risk per person-rem.

The collective doses presented in Table 27.13 could be converted to potential fatal cancers by multiplying the collective doses by a nominal risk coefficient of 5  10 4/rem. As such, an upper end estimate of the potential numbers of fatal cancers associated with exposure to manmade radiation is summarized in Table 27.15. These values are compared to the actual number of fatal cancers in the U.S. from all causes and the theoretical number of fatal cancers from natural background radiation. REFERENCES Belanger W, Jasnosik V (1990) The lansdown radiation site. In: EPA Workshop on Radioactively Contaminated Sites. EPA 520/1-90-009, March 1990. Blanchard RL, Fowler TW, Horton TR, Smith JM (1982) Potential health effects of radioactive emissions of active surface and underground uranium mines. Nucl. Saf. 23:439–50. Bogen KT, Goldin AS (1981) Population exposures to external natural radiation background in the Unites States. U.S. Environmental Protection Agency, ORP/SEPD-80-12. Bouville A (2005) NCI’s development of baseline cancer and radiation-related illness rates relating to nuclear weapons testing in the Marshall Islands. Testimony Before the Committee on Resources and the Subcommittee on Asia and the Pacific of the Committee on International Relations United States House of Representatives. May 25, 2005. BNL (1995) Brookhaven National Laboratory. Radioactive materials released from nuclear power plants, Annual Report 1993. Prepared by Tichler J, Doty K, Lucadamo K, Brookhaven National Laboratory, prepared for the U.S. Nuclear Regulatory Commission, NUREG/CR-2907, BNLNUREG-51581, Vol. 14, December 1995. Brodsky A (1978) Handbook of Radiation Measurement and (Dtection) Detection. CRC Press. Cochran RG, Tsoulfanidis N (1990) The Nuclear Fuel Cycle: Analysis and Management. La Grange Park, IL: American Nuclear Society. CRCPD (2005)Conference on Radiation Control Program Directors, “Nationeide Evaluation of XRay Trends (NEXT)—Tabulation and Graphical Summary of 2001 Survey of Adult Chest Radiography,” CRCPD Publication E-05-2, September 2005. Cronkite EP, Conrad RA, Bond VP (1997) Historical events associated with fallout from Bravo shotoperation castle and 25 Y of medical findings. Health Phys.73 (1):176–186.

REFERENCES

1067

DOE (1979) Technology for Commercial Radioactive Waste Management, Department of Energy Report DOE/ET-0028. Springfield, VA: National Technical Information Service. DOE (1980) Grand Junction Remedial Action Program. DOE/-EV/01621-T1. Washington, DC: DOE. DOE (1983) Spent fuel and radioactive waste inventories, projections and characteristics. Department of Energy Report DOE/NE-0017/2. Springfield, VA: National Technical Information Service. DOE (1994) U.S. Department of Energy. DOE Occupational Radiation Exposure Report, 1992–1994 DOE (1996a) U.S. Department of Energy. Integrated Data Base—1995: U.S. Spent Fuel and Radioactive Waste Inventories, Projections, and Characteristics. DOE (1996b) U.S. Department of Energy. Environmental Management, Progress & Plans of the Environmental Management Program. DOE/EM-0317, November 1996. DOE (2002a) U.S. Department of Energy. Final Environmental Impact Statement for a Geologic Repository for the Disposal of Spent Nuclear Fuel and High Level Radioactive Waste at Yucca Mountain, Nye County, Nevada. DOE/EIS-0250. Las Vegas, Nevada. DOE (2002b) U.S. Department of Energy. Yucca Mountain Science and Engineering “Report,” Rev 1. DOE/RW-0539-1. Las Vegas, Nevada. DOE (2004) U.S. Department of Energy. DOE Occupational Radiation Exposure—2004 Report. DOE (2005) U.S. Department of Energy. United States of America Second National Report for the Joint Convention on the Safety of Spent Fuel Management and on the Safety of radioactive Waste Management, DOE/EM-0654, Rev 1, October 2005. DOE/NFC (1990) Nuclear Fuel Cycle: Current Abstracts. DOE/NFC-90/12 (PB 90-913412). Washington, DC: Office of Scientific and Technical Information. DOL (1995) U.S. Department of Labor, Bureau of Labor Statistics. Fatal Workplace Injuries in 1993: A Collection of Data and Analysis. Report 891, June 1995. DOT (1990) U.S. Department of Transportation. Radiation Exposure of Air Carrier Crew Members. Advisory Circular No. 120–52, March 5, 1990. Eisenbud M, Gesell T (1997) Environmental Radioactivity—From Natural, Industrial and Military Sources 4th edn. Academic Press. EC (2000) European Commission. Referral Guidelines for Imaging. Radiation protection 118. EPA(1984) U.S. Environmental Protection Agency. Occupational Exposure to Ionizing Radiation in the United States, A Comprehensive Review of the Year 1980 and a Summary of the Trends for the Years 1960–1985. EPA 520/1-84-005, September 1984. EPA (1988e) U.S. Environmental Protection Agency. Low-level and NARM Radioactive Wastes— Draft Environmental Impact Statement for Proposed Rules—Background Information Document. EPA 520/1-87-012, June 1988. EPA (1989) U.S. Environmental Protection Agency. Risk Assessments, Environmental Impact Statement, NESHAPS for Radionuclides, Background Information Document. EPA/520/1-89006, September 1989. EPA (1993a) U.S. Environmental Protection Agency. External Exposure to Radionuclides in Air, Water, and Soil. Federal Guidance Report No. 12, EPA 402-R-93-081, September 1993. EPA (1993b) U.S. Environmental Protection Agency. Background Information Document to Support NESHAPS Rulemaking on Nuclear Regulatory Commission and Agreement State Licensees other than Nuclear Power reactors. EPA 520/1-92-XXX, July 20, 1993. EPA (1993c) U.S. Environmental Protection Agency. Issues paper on Radiation Site Cleanup Regulations. EPA 402-R-93-084, September 1993. EPA (1993d) U.S. Environmental Protection Agency. National Radon Database, Vol. 6: National Residential Radon Survey. EPA 402-R-93-013, January 1993. EPA (1994c) U.S. Environmental Protection Agency. Estimating Radiogenic Cancer Risk. EPA 402R-93-076, June 1994.

1068

SOURCES, LEVELS AND EFFECTS OF MANMADE IONIZING RADIATION

Gerusky TM (1981) Three mile island: assessment of radiation exposures and environmental contamination. Ann. N. Y. Acad. Sci. 365:54–62. Googins SW (1990) Radium chemical company site summary. In: EPA Workshop on Radioactively Contaminated Sites, EPA 520/1-90-009, March 1990. Gudiksen PH, Harvey TF, Lange R (1989) Chernobyl source term, atmospheric dispersion, and dose estimation. Health Phys. 57:697–706. Health Physics Society (1986) Health physics considerations in decontamination and decommissioning. In: Proceedings of the Nineteenth Midyear Topical Symposium. CONF-860203. Springfield, VA: National Technical Information Service. Holaday DA, Rushing DE, Coleman RD, Woolrich RE, Kusnetz HL, Bale WE (1957) Control of radon and daughters in uranium mines and calculations on biologic effects. Public Health Service Publ. No. 494. Washington, DC: U.S. Government Printing Office. HPS (2003) Health Physics Society. Whole Body Computerized Tomography for Screening Should Not Be performed. Position Statement of the Health Physics Society. February 2003. Hueper WC (1942) Occupational Tumors and Allied Diseases. Springfield, IL: Charles C. Thomas. ICRP International Commission on Radiological Protection. 1990 Recommendations of the International Commission on Radiological Protection. ICRP-60. Pergamon Press. ISCORS (2003) Interagency Steering Committee on Radiation Standards. ISCORS Assessment of Radioactivity in Sewage Sludge: Radiological Survey Results and Analysis. NUREG-1775, EPA 832-R-03-002, DOE/EH-0069, ISCORS Technical Report 2003-02. ISCORS (2004a) Interagency Steering Committee on Radiation Standards. ISCORS Assessment of Radioactivity in Sewage Sludge: Modeling to Assess Radiation Doses. NUREG-1738, EPA 832-R03-002A. ISCORS Technical Report 2004-03. ISCORS (2004b) Interagency Steering Committee on Radiation Standards. ISCORS Assessment of Radioactivity in Sewage Sludge: Recommendations on Management of Radioactive Materials in Sewage Sludge and Ash at Publicly Owned Treatment Works. DOE/EH-0668, EPA 832-R-03002B. ISCORS Technical Report 2004-04. Larsen IL, Stetar EA, Glass KD (1995) In-house screening for radioactive sludge. Rad Prot. Manag. 12 (4):29–38. Lorenz E (1944) Radioactivity and lung cancer: a critical review in miners of Schneeberg and Joachimstahl. J. Natl. Cancer Inst. 5:1–15. Lubenau JO, Nussbaumer DA (1986) Radioactive contamination of manufactured products. Health Phys.51 (4): 409–429. Miller CW, Smith JW (1996) Why should we do environmental dose reconstructions. Health Phys. 71 (4):420–424. Merwin SE, Balonov MI (editors)(1993) The Chernobyl Papers.Richland, Washington: Research Enterprises Publishing Segment. Meyer KR, Viollrgue PG, Schmidt DW, Rope SK, Killough GC, Schleien B, Moore RG, Case MJ, Till JE (1996) Overview of the fernald dosimetry reconstruction project and source term estimates for 1951–1988. Health Phys. 71 (4): 425–43. Miller HT, Scott LM (1981)Radiation exposures associated with exploration, mining, milling and shipping uranium In: Gomez M, editor. Radiation Hazards in Mining: Control, Measurement and Medical Aspects. Golden, CO: Society of Mining Engineers of American Institute of Mining, Metallurgical and Petroleum Engineers, Inc., New York, NY. Momeni MH, Kisieleski WE, Rayno DR, Sabau CS (1979) Radioisotopic composition of yellowcake: an estimation of stack release rates. NUREG/CR-1216, ANL/ES-84. Argonne, IL: Argonne National Laboratory. Mongan TR, Ripple SR, Brorby GP, Ditimmaso DG (1996) Plutonium release from the 1957 fire at rocky flats. Health Phys. 71 (4):510–521.

REFERENCES

1069

Morgan KZ, Turner JE (editors)(1973) Principles of Radiation Protection—A Textbook of Health Physics. Huntington, New York:Robert E Krieger Publishing Company. NAS (1990) National Academy of Sciences, National Research Council. Health Effects of Exposure to Low Levels of Ionizing Radiation, BEIR V.Washington D.C.: National Academy Press. NAS (2005) National Academy of Sciences, National Research Council. Health Risks from Exposure to Low Levels of Ionizing Radiation. BEIR VII-Phase 2.Washington D.C.: The National Academy Press. NCRP (1984) National Council on Radiation Protection and Measurement. Exposures from the Uranium Series with Emphasis on Radon and its Daughters. NCRP Report No. 77, 1984. NCRP (1987a) National Council on Radiation Protection and Measurement. Public Radiation Exposure from Nuclear Power Generation in the United States. NCRP Report No. 92, 1987. NCRP (1987b) National Council on Radiation Protection and Measurement. Ionizing Radiation Exposure of the Population of the United States. NCRP Report No. 93, 1987. NCRP (1987c) National Council on Radiation Protection and Measurement. Exposure of the Population in the United States and Canada from Natural Background Radiation. NCRP Report No. 94, 1987. NCRP (1987d) National Council on Radiation Protection and Measurement. Radiation Exposure of the U.S. Population for Consumer Products and Miscellaneous Sources. NCRP Report No. 95, 1987. NCRP (1989) National Council on Radiation Protection and Measurement. Exposure of the U.S. Population from Diagnostic Medical Radiation. NCRP Report No. 100, 1989. NCRP (1993) National Council on Radiation Protection and Measurement. Radiation Protection in the Mineral Extraction Industry. NCRP Report No. 118, 1993. NCRP (1996) National Council on Radiation Protection and Measurement. Sources and Magnitude of Occupationaland Public Exposures fromNuclearMedicine Procedures. NCRPReportNo. 124,1996. New York (1982) New York State Department of Health, Report to the Governor & legislature, Radioactive Gold Jewellery, September 1982. NRC (1989) Nuclear Regulatory Commission. Severe Accident Risk: An Assessment of Accident Risks in US. Commercial Nuclear Power Plants. NUREG-1150. Washington, DC: NRC. NRC (1997) U.S. Nuclear Regulatory Commission. Occupational Radiation Exposure at Commercial Nuclear Power Plants and Other Facilities 2003. NUREG-0713, Vol. 17, January. NRC (2002) U.S. Nuclear Regulatory Commission. Radioactive Waste: Production, Storage, Disposal. NUREG/BR-0216-Rev 2. May 2002. NRC (2004) U.S. Nuclear Regulatory Commission. Occupational Radiation Exposure at Commercial Nuclear Power Plants and Other Facilities 2003. NUREG-0713, Vol. 25, October 2004. NRC (2008) U.S. Nuclear Regulatory Commission. Information Digest, 2008–2009 edn. NUREG1350, Vol. 20, August 2008. NRC/NAS (1986) Uranium Mill Tailings Study Panel: Scientific Basis for Risk Assessment and Management of Uranium Mill Tailings.Washington, DC:National Academy Press. Office of Technology Assessment (1991) Complex Cleanup: The Environmental Legacy of Nuclear Weapons Production.Washington, DC:OTA. PNL (1995) Pacific Northwest Laboratory. Dose Commitment due to Radioactive releases from Nuclear Power Plant Sites in 1991. Prepared by D.A. Backer of Pacific Northwest Laboratory for the U.S. Nuclear Regulatory Commission, NUREG/CR-2859, PNL-4221, Vol. 13, April 1995. SCA (1993) Sanford Cohen & Associates and Rogers & Associates Engineering Corporation. Diffuse NORM Wastes–Waste Characterization and Preliminary Risk Assessment. Prepared for U.S. Environmental Protection Agency, Office of radiation and Indoor Air, Contract No. 68D20155, Work Assignment No. 1–16, EPA Work Assignment Manager William E. Russo, May 1993.

1070

SOURCES, LEVELS AND EFFECTS OF MANMADE IONIZING RADIATION

Shearer DR, McCullough P, North D (1995) Radioactivity content of sewage and sludge from sewage plants. Health Phys. 68 (6 Suppl.): S25 Shipler DB ,et al.(1996) Hanford environmental dose reconstruction project—an overview. Health Phys. 71 (4):532–544. Simon SL, Till JE, Lloyd RD, Kerber RL, Thomas DC, Lyon JL, Preston-Martin S (1995) The Utah thyroid cohort study: dosimetry, methodology, and results. Health Phys.68: 460–471. Simon SL, Graham JC (1996) Dose assessment activities in the Republic of the Marshall Islands. Health Phys. 71 (4):438–456. Stanford GS (2001) Integral Fast reactors: Source of Safe, Abundant, Non-Polluting Power. National Policy Analysis, A Publication of the National Center for Public Policy Research. # 378, December 2001.http://www.nationalcenter.org/NPA378.html. Thompson CB, McArthur RD (1996) Challenges in developing estimates of exposure rate near the Nevada Test Site. Health Phys. 71 (4):470–476. Till JE, Simon SL, Kerber R, Lloyd RD, Stevens W, Thomas DC, Lyon JL, Preston-Martin S (1995) The Utah thyroid cohort study: analysis of the dosimetry results. Health Phys. 68:472–483. Travers WD (2000) Staff Progress on Orphan Source Issues: Follow-Up to SECY-99-038, SECY-000184, August 29, 2000. UNSCEAR (1993) United Nations Scientific Committee on the Effects of Atomic Radiation. Sources and Effects of Ionizing Radiation. United Nations, New York, 1993. UNSCEAR (1994) United Nations Scientific Committee on the Effects of Atomic Radiation. Sources and Effects of Ionizing Radiation. United Nations, New York, 1994. Westinghouse (2004) Environmental Report for 2005. WSRC-TR-2005–00005, Westinghouse Savannah River Company, Savannah River Site, Aiken, South Carolina. Whicker WF, Kirchner TB, Anspaugh LR, Ng YC (1996) Ingestion of Nevada Test Site fallout: internal dose estimates. Health Phys.71 (4): 477–486. Widner TE, Ripple SR, Buddenbaum JE (1996) Identification and screening evaluation of key historical materials and emission sources at the Oak Ridge Reservation. Health Phys. 71 (4):457–469. Wiggs LD, Cox-DeVore CA, Voelz GL (1991) Mortality among a cohort of workers monitored for Po210 exposure: 1944–1972. Health Phys. 61 (1):71–76. Yusko JG (1995) Radiation in the scrap recycling stream. In: Proceedings of The Third Annual Conference on the Recycle and Reuse of Radioactive Scrap Metal Beneficial Reuse ’95. Sponsored by the University of Tennessee’s Energy, Environment, and Resources Center and Oak Ridge National Laboratory’s Center for Risk Management, July 31–August 3, 1995, Knoxville, Tennessee.

28 NOISE: ITS EFFECTS AND CONTROL Arline L. Bronzaft

When discussing noise, we too often hear people commenting that music to one person’s ears may be noise to another’s ears. This oft-cited maxim has been used to justify inaction to limit or control noise. With respect to controlling noise, a better guiding principle would be that one’s right to enjoy sounds that may be unpleasant to another’s ears should be limited to the ears of the person controlling the sound. While in practice, this might be difficult to do, the idea that one’s right to make noise stops at another’s ears would create an environment where greater effort would be exerted to lessen the background din. Before we discuss the effects of noise and ways to limit and control it, we need to define sound and noise.

28.1 DEFINITIONS OF SOUND AND NOISE Sound and noise, though often used synonymously, are different. Sound starts as a physical phenomenon derived from audible pressure waves in the air that reach the human ear. To hear, the ear has to convert these sound pressure waves into nerve impulses for the brain to interpret. The outer portion of the external ear receives and transmits the sound waves through the auditory canal to the tympanic membrane or eardrum. The vibrations of the tympanic membrane are then sent to the three bones of the middle ear, which then amplifies the vibrations of the eardrum and passes them on to the hair cells lining the fluid-filled inner ear or cochlea. The hair cells of the inner ear respond to different patterns of incoming sounds and carry them through specific codes to the temporal lobe of the brain. In the brain, sound takes on meaning and emotional significance. The brain deems certain sounds to be wanted and others unwanted and annoying, and similarly judges some sounds as pleasant and others as unpleasant. Sounds become noise when they are judged unwanted, intrusive, annoying, unpredictable, and uncontrollable. Sound is characterized as having two distinct physical properties: the speed or frequency at which sound waves vibrate and the intensity or sound pressure level of each vibration. The

Environmental Toxicants, Third Edition Edited by Morton Lippmann Copyright Ó 2009 John Wiley & Sons, Inc.

1071

1072

NOISE: ITS EFFECTS AND CONTROL

ANSI weights

10 0

Attenuation

-10

C

-20

A

-30 -40 -50 -60 -70 10

100

FIGURE 28.1

1,000 Frequency

10,000

100,000

A-weighting and C-weighting curves.

frequency is interpreted by human ears as pitch, and the intensity accounts primarily for the psychological response of loudness, although frequency does contribute to the interpretation of loudness, with higher pitched sounds perceived as being louder. Humans are generally capable of sensing frequencies between 20 and 20,000 Hz (cycles/s, a measurement of pitch). Loudness of sound is measured on a logarithmic scale in decibels (dB), but to compensate for the effect of pitch on loudness, a modified scale, namely, the A-weighting network, has been developed to assess human response to sound (Fig. 28.1). The A scale, dBA, ranges from a low of “0” to the threshold of pain at 140 dBA to levels over 150 dBA. Soft leaves blowing in the wind may measure 20 dBA, conversation at 50 dBA, telephone ringing at 75 dBA, traffic at 80 dBA, and a jet takeoff at 150 dBA (Fig. 28.2). Sounds over 60 dBA are viewed as intrusive, over 80 dBA as annoying, and over 100 dBA as extremely bothersome; thus, turning sound into noise because these sounds are being judged psychologically. Yet, youngsters who listen to their ipods at over 85 dBA or their headsets at levels measured as high as 110 dBA (Madell, 1986) or play video games at arcades that have been measured as high as 111 dBA (Plakke, 1983) would undoubtedly judge these loud sounds as pleasant and definitely not as noise. Although the human auditory system generally responds to sounds between 20 and 20,000 Hz, there is growing evidence that low-frequency sounds, those between 10 and 200 Hz containing both infrasound and some audible sound, may still be perceived physically. Low-frequency sounds are not measured by the A-scale and, as a result, conventional methods of measurement have ignored their impact, which will be discussed later on in this chapter.

28.2 NOISE EXPOSURE IS WIDESPREAD AND ANNOYING Noise exposure is growing rapidly worldwide, but Zaner (1991) has reminded us that noise is not a new environmental pollutant. She cites accounts of noisy delivery wagons on the ancient streets of Rome and Old Testament stories of very loud music. Dr. Zaner concluded: “The growth and utilization of noise-producing and noise-related technology in modern civilization are proceeding at such an accelerated rate that it is practically impossible to compile a catalogue of noise sources that will not quickly become out-dated.” We should

NOISE EXPOSURE IS WIDESPREAD AND ANNOYING

180

160

Threshold of lung injury

Region of chest vibrations

Threshold of injury from gunfire Threshold of pain 120

120

100

100

80 Phone

80

60

60 l th rma No

40

40

ld ho res

Sound pressure level (db) (Re. 20 mPa)

140

1073

h of

20

ea

20

rin

g

0

-20

10

50 100

500 1000

5000 10000

Frequency (Hz)

FIGURE 28.2

Envelope of human hearing and sensation.

remember that her chapter was written before the proliferation of boomcars, ipods, and cell phones. Although large cities have long been associated with noise, noise is not limited to urban areas. Visitors to national parks complain about overhead plane and helicopter flights; jet skis disturb residents living near large water bodies; and small town community dwellers are increasingly complaining about the numbers of boomcars traveling through their neighborhoods. Although sporting events have been traditionally associated with the shouts and cheers of fans encouraging their teams, more recently loud sounds, or noise to many sports fans and players, have been introduced through loudspeakers blaring out chants and loud music. At some football games, fireworks explode when teams take to the field. Football games are consistently loud enough to harm hearing, and measurements at one Redskins/Giants game found that, for about a third of the time, the sound levels exceeded the workplace limit of 90 dBA (Epstein, 2000). Football coaches have complained that they cannot talk to their own quarterbacks because of the noise—at the Metrodome in Minneapolis and the Mile High Stadium in Denver (Bloomberg, 2000). Baseball and basketball games are also accompanied by loud music and scoreboard messages urge fans to make noise and some of the players have complained about these intrusive sounds. Bernie Williams of the New York Yankees had

1074

NOISE: ITS EFFECTS AND CONTROL

120

B-52

Turbofet

Noise level (EPNdB) 1500 ft sideline

707–100

Normalized to 445 KN thrust

CV-990A

720

CV-880–22

Comet 4 DC8-20

110

BAC-111 DC9-10 737–100 737–200

DC8-61 707–300B Caravelle

100

First generation turbofan

90 1950

1960

727–100

Noise levels are for airplane/engine configurations at time of initial service

Second generation turbofan

727–200 747–100 MD-80 A320–100 A320–100 A321 737–400 DC10–10 A300B2–101 747–400 737–300 767–300 DC10–30 757–200 A330 A340 BAe-146–200 L-1011 737–400 MD90-30 MD-11

747–200

1970

747–300 767–200

1980

1990

1995

Year of initial service

FIGURE 28.3

Progress in noise reduction (Stephens and Crazier, 1996).

asked the Yankees’ scoreboard operators not to play music when he came up to bat, preferring silence (Curry, 2002). The boomcars coming through the neighborhood, the aircraft flying above, the neighbor’s loud music, and the screeching elevated train cars adjacent to apartment buildings are just some examples of noises that are annoying, disturbing, and bothersome. How many people are annoyed by noise? U.S. Environmental Protection Agency’s (EPA’s) Office of Noise Abatement (1977) reported in its document The Urban Noise Survey, based on questionnaire responses from residents in seven cities, that 46% of the respondents claimed to be bothered by neighborhood noise. Zaner (1991) cited a 1977 National Academy of Sciences (NAS) report that over 40 million Americans are bothered by traffic noise and that about 14 million are disturbed by aircraft noise, and these numbers would be higher today. Despite efforts to replace noisier jets with quieter ones, the increase in air travel has more than likely exposed greater numbers of residents to aircraft noise (Stenzel, 1996) (Fig. 28.3). Furthermore, a rise in air travel has been accompanied by an increase in vehicles to and from the airports, and the concomitant increase in traffic noise (Bronzaft et al., 1998; Cohen et al., 2008) is especially annoying to people living near airports. Passchier-Vermeer and Paschieer (2000) have estimated that “ Because of road, railway, and aircraft traffic noise, most of the urban population in industrialized countries are exposed to outdoor Ldn levels exceeding 50 dBA.” The Ldn metric, or day–night level, is used to estimate exposure over a 24 h period, adjusting for evening and night noises that can be more annoying than daytime noises. The Leq (Equivalent Sound Level) is obtained for the period of 7 a.m. to 10 p.m. and then 10 dB is added to the Leq for the period from 10 p.m. to 7 a.m. The assumption here is that a sleeping person is entitled to 10 dB of quiet nighttime respite from noise. The two numbers, one for the day and the other for the evening, are added to obtain the day–night level. Over 30 years ago, the U.S. EPA (1974) estimated that 100 million American residents are exposed to more than 55 dBA in their homes and 33 million to more than 65 dBA. These figures are undoubtedly higher today. In Europe, Chepesiuk (2005) has estimated that about 65% of the population is exposed to ambient sound levels above 55 dBA and 17% to levels

EFFECTS OF LOUD SOUNDS AND NOISE ON HEARING

1075

exceeding 65 dBA. Chepesiuk also noted that the many public loudspeakers and other forms of noise in Tokyo have led many of its citizens to wear earplugs as they go about their daily business. Traditionally, noise was viewed as being annoying. Looking at aircraft and surface transportation annoyance, Schultz (1978) developed a descriptive dosage–effect relationship based on the results of numerous social surveys, which was used to formulate government policy, largely concerning airport expansions. However, noise does more than annoy—noise can be hazardous to health and well-being. Sound government policy can only follow when there is an appreciation of the adverse health impacts of noise.

28.3 EFFECTS OF LOUD SOUNDS AND NOISE ON HEARING Even when sound is desirable, it can harm the ear. A direct physical consequence of exposure to loud sounds is an increment of permanent damage to the inner ear’s hair cells, which results in hearing impairment. One single loud blast of sound near the ear may cause immediate transient damage as well as some permanent damage. Repeated exposure to very loud sounds (over 85 dBA)—resulting from working in industry, attending discos and rock concerts, and living in densely populated noisy cities—can bring about permanent hearing loss over time. For 19 years, the League for the Hard of Hearing collected hearing acuity data on older New Yorkers and Bat-Chava and Schur (2000) found, after examining these data, that a higher percentage of the seniors failed the hearing screening test with each passing year. Dr. Bat-Chava attributed this increase in the number of older citizens suffering some hearing loss to growing noise levels in New York City. Dr. Bat-Chava’s contention that New York is noisier today than it was 19 years earlier cannot be substantiated with actual data, but she can find some support for her statement from the fact that noise ranks first in number of complaints to the New York City’s 311 Quality of Life hotline (Hu, 2003). It is estimated that about 28 million Americans suffer from some hearing loss, with approximately “10 million of these impairments at least partially attributable to damage from exposure to loud sounds” (National Institutes of Health, 1990). There is no doubt that there is a relationship between hearing loss and exposure to loud sounds or noise (Fay, 1991; Kryter, 1994; Passchier-Vermeer and Passchier, 2000). Of particular concern is the finding that 12.5% of American children between 6 and 19 years of age suffer from noise-related hearing loss (Niskar et al., 2001). Lipscomb was quoted (Cherry, 1986) as having found that twothirds of the more than 14,000 college freshmen that he tested suffered from some hearing loss. The Royal National Institute for the Deaf found a very significant number of young people who attend clubs with loud music experience signs of hearing damage including ringing in their ears and dullness of hearing (BBC News, May 2003). The evidence linking hearing loss to loud sounds or noise is plentiful, but convincing people, especially young people, to protect their ears is still a difficult task. The Federal Occupational Safety and Health Administration (OSHA) requires that employees working in environments with sound pressure levels averaging over 90 dBA for 8 h/day wear hearing protection. Unfortunately, hearing protection is not always provided, and even when it is, too many workers refuse to wear it. Some rock musicians may wear earplugs, but few of their fans come to their concerts similarly protected. NASCAR fans, and the children they bring along to their races, are exposed to exceedingly loud roars as the cars circle the track. Movies, especially the coming attractions, expose customers to extremely loud sounds. When advertisers and magazines, such as Car Audio and Car Sound and Performance, promote

1076

NOISE: ITS EFFECTS AND CONTROL

the notion that loud and noisy is fun and exciting, is it any wonder that youngsters in boomcars are exposing their ears to stereo systems that can reach 140 dBA? It is of paramount need to educate people to the dangers of loud sounds and noise and to inform them of ways to protect their ears, but it is also important to develop and apply noise control technology to lessen the overall din (as discussed later in this chapter).

28.4 NOISE AS A STRESSOR Noise, defined as unwanted sound, does not have to be loud to have an adverse effect on a person’s physical or mental health. However, these unwanted, unpleasant, and annoying sounds, now identified as noise, can create stress on the body and, like other stressors, can bring about a complex set of physiological changes. Noise, as a stressor, can have indirect effects on health. Physiological reactions can include rise in blood pressure, slowing down of digestion, drying of the mouth, increase in hormone secretions, and change in heart rhythm. If the individual is continuously exposed to the noise source, and experiences more and more stress, then permanent damage may be caused to the circulatory, cardiovascular, neuroendocrine, or gastrointestinal systems. Studies examining nonauditory effects of noise are often conducted on workers in noisy work environments or on residents living near highways, railroads, or airports. The results from these studies are then generalized to large populations similarly exposed and to populations chronically exposed to noise from other sources, for example, motor raceways, recreational arcades, and noisy neighbors. Tomei et al. (1995) found that individuals occupationally exposed to noise are at a higher risk for developing certain cardiovascular disorders. Melamed et al. (1997) reported an association between high industrial noise and higher serum lipid levels in younger men and Melamed et al. (2001) found that workers performing complex occupational jobs showed elevated blood pressure over time. However, more research is needed before firm conclusions can be drawn that noise in the workplace adversely affects physiological systems. The World Health Organization’s (WHO) guidelines for community noise (Berglund et al., 1999) caution that noise can be detrimental to health. A growing body of scientific studies links noise to health, with the strongest evidence being for cardiovascular disorders (Ising and Kruppa, 2004; Passchier-Vermeer and Vermeer, 2000; Fay, 1991; Kryter, 1985). The findings that children’s cardiovascular systems may be adversely affected by high road traffic and aircraft noise are especially disturbing (PasschierVermeer and Vermeer, 2000). The WHO acknowledgment that there are sufficient data to warn people about the dangers of noise came 21 years after the U.S. Environmental Protection Agency booklet Noise: A Health Problem (1978) concluded that “It is finally clear that noise is a significant hazard to pubic health. Truly, noise is more than just an annoyance.” In the years that followed, the position of the United States government was to seek additional research to affirm the relationship between noise and health before taking strong actions to control noise, despite the statement in 1969 by the then Surgeon General Dr. William H. Stewart that in protecting health, absolute proof comes too late and to wait for it may prolong suffering unnecessarily. The WHO definition of health is not merely the absence of symptoms but a state of complete physical, mental, and social well-being. By relying too heavily on actual physiological symptoms to determine the effect of noise on health, we may be ignoring

NOISE AND MENTAL HEALTH

1077

the impact of noise on a “good state of health.” In other words, people may be suffering physically from noise but have not yet evidenced actual physical symptoms. In their study on the impacts of aircraft noise on health, Bronzaft et al. (1998) found that residents complained that the noise interfered with their right to open their windows, enjoy their outdoor areas, listen to radio and television, and converse with family members. The quality of life for these residents was diminished as it was for the Okinawa residents (Hiramatsu, 1999) who complained that aircraft noise disturbed their daily activities. When impacts of noise are extended to include quality of life, in keeping with WHO definition of health, it is clear that noise is more deleterious to health than might be concluded from studies linking noise to a particular ailment.

28.5 NOISE AND SLEEP INTERFERENCE In both the Bronzaft et al. (1998) and Hiramatsu (1999) investigations, residents complained that aviation noise intruded on their sleep. People expect sleep will provide them with needed rest as well as the ability to awaken hardy enough to carry out the activities of the next day. The body certainly needs sleep for recuperative functions. Two reviews of epidemiologic studies (Passchier-Vermeer and Vermeer, 2000; Health Council of the Netherlands, 2004) reported that nighttime noises disrupt sleep stages, increase awakenings, and affect heart rate. The secondary effects of loss of sleep (Pollak, 1991) might include dependency on tranquilizers to induce sleep, poor performance the next day, and less attentiveness to cues of danger.

28.6 NOISE AND MENTAL HEALTH Early studies on mental health and noise suggested a link between psychiatric admissions to a hospital and airport noise exposure but later studies refuted this relationship (Stansfeld et al., 1993). More recently, Hiramatsu (1997) surveyed residents living near an air base, who were exposed to aviation noise, and found greater mental instability and nervousness in this population. Evans (2003) reported on studies showing that children living near noise sources experience more psychological distress, but also cited other data that did not support these findings. That noise distresses and bothers people is underscored by the number of groups dealing with noise that have come into being in both the United States and abroad in the past few years. Some of these groups focus on one kind of noise only, such as aircraft or traffic noise, but most groups deal with noise in general. The Noise Pollution Clearinghouse (www.nonoise.org), NoiseOff (www.noiseoff.org), League for the Hard of Hearing (www.lhh.org/noise), and United Kingdom Noise Association (www.ukna.org.uk) are examples of such groups. These organizations respond to telephone calls, e-mails, and letters from individuals exposed to noise in their neighborhoods. When representatives of the Federal Aviation Administration (FAA) hold hearings on airport expansions or aircraft reroutings, citizens speak out vociferously about the noises from overhead jets, expressing a great deal of anguish and upset. Noise complaints head the list of complaints to New York City’s Quality of Life Hotline attesting to the stress that noise causes. When Bronzaft et al. (2000) queried individuals as to their emotional responses to noise, nearly 50% noted that noise makes them angry. Anger is often elicited by noise and, in some cases, can lead to aggression towards the noisemaker. Media accounts of fights between a noisemaker and someone bothered by noise are not uncommon.

1078

NOISE: ITS EFFECTS AND CONTROL

It should be noted that another emotional response may accompany unhappiness about noise problems, namely, learned helplessness. When a noise problem is not ameliorated, despite many complaints to higher authorities, the individual might give up hope about resolving noise problems, accept that nothing can be done, and feel helpless. This feeling of helplessness may exacerbate the stress brought about by the noise itself.

28.7 NOISE AFFECTS CHILDREN’S COGNITIVE, LANGUAGE AND LEARNING SKILLS Federal Interagency Committee on Aviation Noise (September 2000) concluded, after a careful review of the literature, that aircraft noise can interfere with reading, speech acquisition, and memory. For the impact on reading, the Committee noted more than 20 studies, including the study by Bronzaft and McCarthy (1975), which found that sixth-grade children in classrooms exposed to noise from adjacent elevated train tracks were nearly a year behind in reading when compared to children attending class on the quiet side of the school building. A later study by Bronzaft (1981) found that after the Transit Authority installed rubber resilient pads to lessen the noise from the train tracks, and the Board of Education installed acoustical tiles in ceilings of affected classrooms, the children in the classrooms facing the tracks were reading at the same level as the children attending class on the quiet side. A recent investigation conducted on school children in three countries, the Netherlands, Spain, and the United Kingdom (Stansfeld et al., 2005) reached the conclusion that aircraft noise could impair cognitive development in children, specifically reading comprehension. Thus, noise disrupts teaching and learning. Noise interferes with speech reception and, in noisy classrooms, children will very likely not hear the teacher over the passing elevated trains or overhead jets. Teachers often need to shout to compete with the noise, and many students choose to tune out all the sounds, from the teacher, and from the train or plane. Many classrooms are internally noisy as well—noise from hallways, from nearby classrooms, and from overcrowded classes. The American National Standards Institute (ANSI) promulgated acoustical standards for classrooms, hoping that this would result in quieter classroom environments (ANSI, 2002ANSI S12.60-2002). Wachs and Gruen (1982) reported that a noisy home can also disrupt a child’s development and parents need to be warned about shouting and playing televisions and stereos too loudly. They should also be told, as suggested by Wachs and Gruen, that noisy households are homes where parents and children interact little and that such interaction is helpful when children are developing speech and cognitive skills. By contrast, Bronzaft (1996) noted, in her research on high academic achievers, that they were reared in homes that respected quiet and that their parents provided the quiet times required for reading, studying, and learning. These achievers also reported that their parents did not tend to use shouts and loud voices to discipline them but rather did so with lowered, stern voices. There is little doubt that lessening train noise, rerouting aircraft away from schools and children’s homes, and diminishing the noises in the home would benefit children’s development. Government policies are needed to support the decrease in transportation noises. Also important is a recognition by parents that homes should be quieter and that their children should be encouraged to engage in quieter activities—less time watching loud videos, listening to blaring music, and playing high volume computer games and more time with books, puzzles, and board games.

CIVILITY, RESPONSIBILITY, AND NOISE

1079

28.8 IMPACTS OF LOW-FREQUENCY NOISE Research examining the impacts of noise on health and well-being are generally conducted on sound frequency within the range of hearing, but there is a growing body of studies examining the impact of low-frequency noise, between 20 and 200 Hz, on health. Lowfrequency sounds may be perceived as rumbles, with the eardrum able to pick up vibrations. For many people, these low-frequency sounds are not audible but are still experienced as feelings of discomfort. Low-frequency sounds, emitted by heating or cooling units, can cause distress for individuals who are particularly sensitive to such sounds, and there is a growing interest in the impacts of low-frequency sound (Findeis and Peters, 2004; Leventhall, 2004; Persson Waye, 2004). Castelo Branco and Alves-Pereira (2004) have been examining the physiological effects of low-frequency noise on workers for many years, with special emphasis on the aircraft industry. They have identified a set of symptoms in aircraft technicians that they have named vibroacoustic disease (VAD), which is a whole-body pathology brought about by their excessive exposure to low-frequency noise (LFN). For workers on the job for several years, the initial symptoms are mood swings, indigestion, and heartburn, but workers on the job for more than 10 years develop more severe symptoms such as ulcers, varicose veins, spastic colitis, and muscular pain. Despite years of research on vibroacoustic disease, Branco and Alves-Pereira concluded that “VAD is not acknowledged as a pathological entity.” Summing up the effects of low-frequency noise, the WHO (Berglund et al., 2000) noted that low-frequency noise can disturb rest and sleep, that A-weighted scales cannot assess the impact of low-frequency noise, and that further research on impacts of low-frequency noise is warranted.

28.9 CIVILITY, RESPONSIBILITY, AND NOISE Zaner (1991) listed the major sources of noise—transportation, recreational, industrial, home appliances, and emergency signals. Even though she did not include the most recent sources, for example, cell phones and leaf blowers, she acknowledged that new sources would be added to her list. However, one source she neglected to include has increasingly been the bane of many who complain about noise, that is, people who believe they have the right to impose their sounds on others. The youngsters in their boomcars, the neighbors playing loud music, and the people next door partying until four in the morning believe it is their right to make noise. In fact, these people speak of noise making as a right. Businesses are also guilty of imposing their sounds on nearby residents, for example, bar and restaurant owners, nightclubs, and dirt bike raceways. “Neighbor music, TV and radio” ranks amongst the top 10 sources of noise when individuals are asked to rank different types of noise (Bronzaft and Van Ryzin (2004). This source, as well as some of the other noises that were ranked even higher, that is, boomcars, rowdy passersby, reflects a disregard for another person’s rights. It is difficult to legislate respect and responsibility but laws and ordinances have been passed, as will be discussed, in an attempt to reign in this lack of civility. Education in the home and in the school on the “good neighbor policy” would also be helpful. Noise would be lessened if people learned to cooperate, to respect each other, and were taught to assume some responsibility for their behavior.

1080

NOISE: ITS EFFECTS AND CONTROL

28.10

CONTROLLING NOISE

28.10.1

Government Responsibility

The Noise Control Act, passed by Congress in 1972, essentially stated that Americans were entitled to be free from noises that jeopardize their health and welfare. The Office of Noise Abatement and Control (ONAC) in the EPA was charged with the responsibility of carrying out the mandate of the Act. ONAC prepared excellent brochures educating people to the dangers of noise and suggesting ways to lessen noise. Federal EPA was responsible for setting noise limits for trains, trucks, and machinery, and was working toward requiring labels on consumer products that would have allowed consumers to compare products on noise levels. EPA worked with the Department of Labor in providing information on the effects of noise and, although it was not responsible for overseeing aircraft noise, it was able to recommend aircraft noise regulations to the FAA. Under the 1978 Quiet Communities Act, EPAwas directed to increase assistance to states and localities in developing and carrying out their own noise control programs. However, in the early 1980s funding for ONAC ended. The termination of federal funds did not result in increased spending at the local levels; quite the opposite. In his report entitled The Dormant Noise Control Act, Shapiro (1991) concluded that Congress’ decision to go along with the cuts sent a signal to states that noise was not important. OSHA continues to monitor impacts of noise on hearing in the occupational setting and the FAA sets noise standards for certification of aircraft. The FAA actions have resulted in a significant decrease in the decibel levels of sound emitted by aircraft and the latest models of Stage-3 aircraft are considerably quieter, about 15 dB, than older Stage-1 and Stage-2 aircraft. The introduction of quieter aircraft has resulted in fewer people living within the 65 dB Ldn area (area designated by FAA as being most impacted by noise), but the increase in total numbers of flights at airports has increased the exposure of single-event noises that are especially disturbing to individuals living near airports (Stenzel, 1996). In an analysis of the comments received by the FAA concerning its draft Noise Abatement Policy 2000, people living near airports reported that the noise problem has grown. Furthermore, 96% of the more than 1,000 people who submitted comments to the FAA believe the draft Noise Abatement Policy will not adequately protect citizens from aviation noise (Blomberg and Sharp, 2002). Smaller airports have grown in the last few years, many catering to private company aircraft, and they are having difficulty phasing out noisier Stage-1 and Stage-2 aircraft (Aviation and Environment News, 2005). Without federal funding, states and cities had to pass their own noise ordinances, and many were slow to do this. Atypical was New York City in that it had introduced its own Noise Code in 1972. However, as the years passed, the code became less effective in coping with the increased noises of the City. New York State enacted the Rapid Rail Transit Noise Code in 1982 to lessen noise in the New York City rail system and the Transit Authority listed significant accomplishments in its noise abatement efforts but this law expired in 1994 and the subway and elevated trains have grown noisier again. The advent of anti-noise groups across the United States during this past decade has resulted in increased interest in noise control at the local level. Florida recently passed a law penalizing noise from automobile stereos (Luscombe, 2005) and Jupiter, Florida, reminds motorcyclists that loud bikes—exhaust systems without mufflers—are not permitted (Bradshaw, 2005). Judges in Louisiana can take a driver’s license for up to 30 days when car music is too loud (Liggett, 2005) and, in Loveland, Colorado, police officers can cite

CONTROLLING NOISE

1081

drivers for loud music or motorcyclists for loud exhaust systems (Carter, 2005). New York City’s Revised Noise Code, prepared by the City’s Department of Environment Protection, went into effect in 2007. These are examples of actions against noise pollution being implemented in cities across the United States. Noise pollution has attracted the attention of political leaders around the globe. The European Union has encouraged member countries to develop strategies to cope with increased noise problems. For example, London has already developed a road noise map to assist the city in designing appropriate mitigation. The Greater London Authority (Higgitt et al., 2004) had a report prepared entitled Quiet Homes for London: Review of Options that addressed noise problems in homes, including poor home insulation and protection from noisy entertainment centers. Newcastle in the United Kingdom has trained officers to respond to complaints from loud music, dogs barking, and other loud noise during the nighttime hours. As part of its World Environment Day activities in June 2005, Bangladesh announced it would formulate rules to lower the decibel level in its cities and towns. The Environmental Protection Agency of Taiwan in July 2005 recognized that low-frequency noises can be harmful to health and has set regulations against low-frequency noises emitted from machinery, air conditioners and ventilators, musical establishments, and so on. In June 2005, Delhi strengthened its ban on music in commercial and private vehicles and launched a campaign against use of air horns on commercial vehicles. The United States Federal government, with the passing of the Noise Control Act in 1972 and the activities of the Office of Noise Abatement and Control in the 1970s, had once led the way in noise control but, today, it lags behind those other nations that have demonstrated a greater concern for limiting noise pollution. Repeated attempts by a group of Congress people and Senators to revive ONAC have not yet been successful, while there has been renewed interest by local authorities to pass ordinances limiting noise, federal legislation is still needed; for example, to set noise standards on products and for the control of aircraft noise. Shapiro (1991) summed up the sentiments of many American citizens when he labeled Congress irresponsible with respect to noise control. 28.10.2

Controlling Noise at the Source

Undoubtedly, the best way to control aircraft noise is to reduce jet engine noise. Indeed, the last 30 years have seen a drop in jet noise by about 20 dB, with intentions on part of the airlines and the National Aeronautics and Space Administration to decrease the noise even further. BBC News announced in August 2005 that Cambridge University engineers are working on the design of a “silent aircraft.” They believe that much of the noise is caused at landing and takeoff, and are working on a design that would be inaudible once it took off from the airport. Their design involves mounting the engines on top of the aircraft to direct noise upward. However, testing of models is still years off. Just as air conditioners and refrigerators have been made quieter, so too can hair dryers and leaf blowers. Highway traffic can also be made quieter by using materials for roadways that lessen the sounds of the vehicles passing over, and many European cities have been using such materials for their roadways. ONAC had intended to introduce a program of Buy Quiet whereby manufacturers would be required to label the decibel levels of their products. Although ONAC was defunded before this program was instituted, several manufacturers have quieted down products such as vacuum cleaners and dishwashers, and Consumer Products identifies the noise levels of many products. In

1082

NOISE: ITS EFFECTS AND CONTROL

fact, some products, for example, dishwashers, are being promoted for their quiet operation. At one time, people viewed train whistles as pleasant sounds in the distance. However, in more recent years, residents living adjacent to our railroad system began to protest the loud train horns at crossings. In response to this outcry, the Federal Railroad Administration (FRA) limited the loudness of train horns and directed train engineers to sound the horns when they are closer to the crossings. The FRA believes the number of people affected by train horns will be lessened considerably (Wald, 2003). Just as some sounds are being lowered, new products have been introduced that are proving to be disturbing to large numbers of people, for example, cell phones. Not only is the ring intrusive but also people often have to shout into the phones to be heard. Cell phone manufacturers are seeking ways to correct the problem of poor reception so that voices on cell phones could be lowered. Auditory warning signals or back-up beeps on trucks, while necessary safety devices, are a source of noise pollution for nearby residents. The bbs-tek back up alarm developed by Brigade Electronics PLC (2000) (New York City) uses a broadband multifrequency sound, rather than the traditional narrow band frequency, resulting in a warning device that localizes the direction of sound, dissipates it more quickly, and is heard only by those in the hazardous area. The bbs-tek reversing alarms are being promoted in cities in the United States. New York City’s Sanitation Department is testing the appropriateness of these alarms for some of its sanitation trucks. However, the Sanitation Department did decide, in 2004, to install ambient noise sensitive back-up alarms in its new vehicles and replace older vehicles with the new device when a replacement is needed, as a means of responding to a common New York City noise complaint. Unfortunately, products that would be tolerable at lower decibel levels, for example, stereos for automobiles, are being advertised for their upper limits. When Sony advertised their amplifiers and speakers with a “Disturb the Peace” ad, the Noise Pollution Clearinghouse went into action urging people to write to Sony protesting an ad that promoted incivility. Sony got the point, but unfortunately ads such as “Play it Loud” and “Disturb The Neighbors” still exist. With ordinances being introduced to restrict stereo systems at high levels, it would be wise for advertisers to cease promoting “loud as fun.” Technological advancements that have allowed for noise cancellation has been particularly effective on noise signals below 1000 Hz, for example, on exhaust systems. While not cutting noise at its source, these microchips that cancel out noises can be inserted into earmuffs to protect people who are subjected to low-frequency sounds, for example, helicopter pilots. Some car makers have installed “noise cancellation” technology into new cars to drown out outside traffic noises. Designers of tools, home appliances, and machinery must consider noise in the design of their products. 28.10.3

Other Ways to Lessen Noise

Architects, engineers, and developers have to be as aware of the acoustical environment of their projects as they are of the visual. Tall buildings should be designed so that sounds within and between apartments are muffled. Large apartment complexes should have space allocated to meeting rooms and play areas so that there is less congregation in stairwells and hallways and as a result less noise that would be disturbing to others. With the U.S. Census Bureau reporting that noise is a top complaint that people have about their neighborhoods, home builders should consider using quieter products in construction and

CONTROLLING NOISE

1083

consider installing double-glazed windows to protect residents from outside sounds. Use of sound absorbing materials, such as acoustical tiles in ceilings, soft materials on walls, drapes, and floor coverings, would make apartments and offices quieter. Hospitals can be very noisy places, and ways to quiet them have included double-glazed windows, acoustical tiles hiding ducts and wiring, cushioned linoleum, televisions for patients with individual headsets, and chairs that make less noise as they are pulled across the floors. The placement of nurses’ stations can be a major factor in disturbing patients and designers must keep this in mind. Noisy cleaning equipment and food delivery trucks still need to be quieter. Quiet signs, which have disappeared in hospitals, are coming back, reminding visitors, as well as doctors and nurses, to lower their voices. Medical equipment designers must be cognizant of acoustics and the harmful effects of loud sounds and noises, especially in intensive care units where the beeps monitoring vital signs such as heart rate, blood pressure, and breathing are operating continuously through the day and night. Especially disturbing is a recent article (Raeburn, 2005 ) documenting that premature babies are still being cared for in noisy neonatal intensive care units (NICUs). “In the typical NICU their senses are overwhelmed by ringing telephones, screeching alarms . . .” This article also notes that a researcher, who was studying EEGs of premature babies, had to record for 3 h to get a 6 min quiet sleep EEG because of continuous sound interruption that aroused the babies. With so many studies demonstrating the harmful effects of noise on children’s learning, it is especially important that schools create an environment that enhances, rather than impedes, learning. ANSI recommends that an empty classroom should not exceed 35 dBA, recognizing that classrooms will get noisier once children are in them. Also, noises from heating and ventilation systems, ducts in the walls that allow sounds to travel from class-to-class, and doors that are not properly sealed to keep out sounds from the hallways, add to the sound level of the classroom. In addition, transportation noises coming in from the outside make the classroom even noisier. The FAA and the Port Authority of New York and New Jersey have participated in a soundproofing program for schools near New York’s airports. The program started in 1983 and provides noise abatement funds under its Airport Improvement Program. Noise levels were cut down significantly in those schools where such funding provided for acoustical windows and new ventilation systems. In February 2001, $30 million was approved to quiet classrooms in 33 schools in New York and New Jersey (The Port Authority of NY and NJ, 2001). The Noise Pollution Clearinghouse, supported by the Acoustical Society of American, has devoted a section on its website www.nonoise.org to quieting classrooms that is directed at principals, teachers and parents. The Council on the Environment of New York City has included noise pollution in its environmental school program and the New York City Department of Environmental Protection has an educational director who visits public schools to discuss noise. Noise from vehicular traffic has been one of the major sources of noise and over the past 30 years, and barriers to shield residents from the noise emanating from highways have been built along highways throughout the United States. According to the United States Federal Highway Administration, over 1,620 miles of noise barriers have been built in the United States and Puerto Rico since 1970. Communities have been critical of the appearance of many of the barriers constructed, but the last few years have seen more attractive designs and more recently technologies have also resulted in more effective abatement (Beck, 2001).

1084

NOISE: ITS EFFECTS AND CONTROL

28.11

EDUCATION AND PUBLIC AWARENESS

That noise is an annoyance, dangerous to our hearing, and hazardous to our health is a message that must be more widely disseminated amongst people worldwide. The EPA’s ONAC was noted for the excellent educational materials it produced. The pamphlets and brochures distributed by ONAC clearly stated the dangers of noise to hearing and health. Today, there are agencies that alert citizens through their web sites. The American Hearing Research Foundation, Institute for Occupational Safety and Health, and the National Institute on Deafness and Other Communicative Disorders are three such organizations, but their focus is primarily on hearing loss. These organizations urge people to protect their ears by wearing hearing protection when exposed to loud sounds, for example, rock concerts, discos, and playing instruments. The web site for the San Francisco-based organization, Hearing Education and Awareness for Rockers (HEAR) was established by Kathy Peck to educate musicians and music lovers about the importance of protecting their ears from loud music and to inform them that proper ear plugs still allow them to enjoy the music without harming the ear. Individuals should also be informed that they have to protect their ears in recreational activities, for example, shooting range, attendance at NASCAR races, and so on. Unfortunately, too many people do not wear hearing protection, believing it is not “macho” to do so or, as in the cases of some women in the workplace, their hair styling may be disrupted. The Toronto, Canada Public Health Department prepared a brochure on noise and children, recognizing the dangers of noise to children. The Council on the Environment of New York City has a brochure on noise pollution, as does the League for Hard of Hearing in New York City. Web sites of the Noise Pollution Clearinghouse, League for the Hard of Hearing and Citizens Coalition Against Noise Pollution provide information on hearing loss as well as information on the adverse impacts to mental and physical well-being. These organizations also provide information on what citizens can to do to reduce noise in their communities, especially those who live near highways and airports. These web sites also suggest ways to involve legislators, both at federal and at local levels, in working toward less noise pollution.

28.12

SUMMARY

That noise is harmful to health and well-being is supported by scientific evidence, even though additional research is needed to solidify the noise–health link. Unquestionably, individuals should be warned to protect themselves from the dangers of noises. In addition, the United States federal government should once again involve itself in protecting citizens against the harmful effects of noise. In the EPA booklet Noise: A Health Problem (U.S. EPA, 1978), the federal government took the following position on noise pollution: “It is finally clear that noise is a significant hazard to public health. Truly, noise is more than an annoyance.” Now over 30 years later the federal government has retreated from its early position. Although the United States held a leadership role in noise abatement in the 1970s, it has now taken a backseat to the European Union, which has directed member nations to proceed more aggressively in reducing noise. Noise abatement should not exclusively be the responsibilities of American cities and states, especially when transportation and industrial noises are two major sources of noise; thus dependent on some federal oversight. However, until the American government places greater emphasis on noise control, Americans should educate themselves about the hazards of noise and learn how to protect themselves from

REFERENCES

1085

noise. To a large extent, the ways to reduce noise pollution do exist, but the will to act to lessen the growing decibel level is lacking. The author acknowledges that by writing this article she hopes to encourage its readers to advocate for a quieter and healthier environment.

REFERENCES ANSI (American National Standards Institute) (2002) American National Standard Acoustical Performance Criteria, Design Requirements, and Guidelines for Schools. ANSI S12. 60-2002. New York: ANSI. Aviation and Environment News (2005) Boca Raton joins national effort to phase out Stage 1 and 2 aircraft. Glenside, PA: Great Circle Communications LLC. July 2005. Bat-Chava Y, Schur K (2000) Longitudinal trends in hearing loss: nineteen years of public screenings. 128 Annual Meeting of American Public Health Association, November 2000, Boston. BBC News (2003) Health clubbers risk premature deafness, May 8, 2003. Beck DE (2001) Restoring the peace. Http://www.cenews.com/article/asp?id=245. Berglund B, Lindvall T, Schwela D, Goh KT (2000) Guidelines for Community Noise. Geneva: World Health Organization. Berglund B, Lindvall T, Schwela DH (1999) Guidelines for Community Noise. Geneva: World Health Organization. Blomberg L, Sharp J (2002) The failure of America’s aviation noise abatement policy. Montpelier, VT: Noise Pollution Clearinghouse. Bloomberg (2000) Giants stadium gets no votes as “toughest.” New York Post, p. 73. Bradshaw K (2005) Police warn bikers: There’s a noise law in town. Treasure Coast and Palm Beaches; July 13, 2005.www.1tcpalm.com/tcp.local_news/article Brigade Electronics PLC (2000) Brigade sound back-up alarms and movement alarms. New York; December 24. Bronzaft AL, Van Ryzin G (2004) Neighborhood noise and its consequences. New York: NYC Council on the Environment. Bronzaft AL, Deignan E, Bat-Chava Y, Nadler NB (2000) Intrusive community noises yield more complaints. Noise Rehabil. Q. 25: 16–22, 34. Bronzaft AL, Ahern KD, McGinn R, O’Connor J, Savino B (1998) Aircraft noise: a potential health hazard. Environ. Behavior 30:101–113. Bronzaft AL, (1996) Top of the Class. Greenwich, CT: Ablex. Bronzaft AL (1981) The effect of a noise abatement program on reading ability. J. Environ. Psychol. 1: 215–222. Bronzaft AL, McCarthy D (1975) The effect of elevated train noise on reading ability. Environ. Behavior 7:517–528. Carter R (2005) Police may take matters into their own hands. New noise ordinance would rely on officers over machines. The Daily Reporter Herald; July 17, 2005. Castelo Branco NAA, Alves-Pereira M (2004) Vibroacoustic disease. Noise Health 6: 3–20. Chepesiuk R (2005) Decibel hell. Environ. Health Perspect. 113: A35–A41. Cherry L (1986) The hidden menace to your health: noise. Glamour; December 1986; 168–172. Cohen BS, Bronzaft AL, Heikkinen M, Goodman PE, Nadas A (2008) Airport-related air pollution and noise. J. Occup. Environ. Hyg. 5:119–129. Curry J (2002) Williams prefers to bat without music blaring. New York Times Website. Accessed 2 October 2002.

1086

NOISE: ITS EFFECTS AND CONTROL

Epstein K (2000) Sound offense, no defense. The Washington Post. Health section; December 19, 2000. Evans GW (2003) The built environment and mental health. J. Urban Health 80:536–554. Fay TH (1991) Noise and Health. New York: The New York Academy of Medicine. Federal Interagency Committee on Aviation Noise (FICAN) (2000) FICAN Position on Research into Effects of Aircraft Noise on Classroom Learning.Washington, DC: Federal Interagency Committee on Aviation Noise. Findeis H, Peters E (2004) Disturbing effects of low frequency sound immissions and vibrations in residential buildings. Noise Health 6:29–35. Health Council of the Netherlands (2004). The influence of night-time noise on sleep and health. Report No. 2004/14E. The Hague. Higgitt J, Whitfield A, Groves R (2004) Quiet Homes for London: Review of Options, an Initial Scoping Study. UK: Greater London Authority; July, 2004. Hiramatsu K (1997) A survey on health effects due to aircraft noise on residents living around Kadena Air Base in the Ryukyus. J. Sound Vibr. 205:451–460. Hiramatsu K (1999) A report on the aircraft noise as a public health problem in Okinawa. Okinawa Prefectural Government: Department of Culture and Environmental Affairs. Hu W (2003) New Yorkers Love to Complain, and Hot Line Takes Advantage. The New York Times; December, 2003 A1, B6. Ising H, Kruppa B (2004) Health effects caused by noise: evidence in the literature from the past 25 years. Noise Health 6:5–13. Kryter KD (1985) The Effects of Noise on Man, 2nd edn, Orlando: Academic Press. Kryter KD (1994) The Handbook of Hearing and the Effects of Noise. San Diego: Academic Press. Leventhall HG (2004) Low frequency noise and annoyance. Noise Health 6:59–72. Liggett B (2005) New Law Targets Music that is Too Loud. Judges can Take Drivers’ Licenses for Up To 30 days. The Opelousas Daily World; July 10, 2005. Luscombe R (2005) Florida Tries to Quiet “Boom Cars.” The Sunshine State Passes a Tough New Law as the National Fight to Reduce Noise Rumbles on. The Christian Science Monitor; July 13, 2005. Madell JR (1986) A report on noise. Hear. Rehabil. Q. 11: 4–13. Melamed S, Gofer D, Ribak J (1997) Industrial noise exposure, noise annoyance, and serum lipid levels in blue-collar workers: The Cordis study. Arch. Environ. Health 52:292–298. Melamed S, Fried Y, Froom P (2001) The interactive effect of chronic exposure to noise and job complexity on changes in blood pressure and job satisfaction: a longitudinal study of industrial employees. J. Occup. Health Psychol. 6:182–195. National Institutes of Health (1990) Noise and hearing loss: consensus conference. JAMA 263:3185– 3190. Niskar AS, Kieszak SM, Holmes A, Esteban E, Rubin C, Brody DJ (2001) Estimated prevalence of noise-induced hearing threshold shifts among children 6 to 19 years of age: the third national health and nutrition examination survey, 1988–1994. Pediatrics 108:40–43. Passchier-Vermeer W, Passchier WF (2000) Noise exposure and public health. Environ. Health Perspect. 108: 123–131. Persson Waye K (2004) Effects of low frequency noise on sleep. Noise Health 6:87–91. Plakke BL (1983) Noise levels of electronic arcade games: a potential hearing hazard to children. Ear Hear. 4:202–203. Pollak CP (1991) The effects of noise on sleep. In: Fay TH, editor. Noise and Health. New York: The New York Academy of Medicine.

REFERENCES

1087

Port Authority of NY and NJ (2001) Record 30 million school soundproofing program approved by Port Authority Board of Commissioners. New York, February 22, 2001. Raeburn P (2005) A Second Womb. New York Times; August 14, 2005. Schultz TJ (1978) Synthesis of social surveys on noise annoyance. J. Acoustical Soc. Am. 64: 377– 405. Shapiro SA (1991) The dormant noise control act and options to abate noise pollution. Report for the Administrative Conference of the United States. Washington, DC. Stansfeld SA, Berglund B, Clark C, Lopez-Barrio I, Fischer P, Ohrstrom E, Haines MM, Hygge S, van Kamp I, Berry BF (2005) Aircraft and road traffic noise and children’s cognition and health: a cross-national study. Lancet 365:1942–1949. Stansfeld SA, Sharp DS, Gallacher J, Babisch W (1993) Road traffic noise, noise sensitivity, and psychological disorder. Psychol. Med. 23:977–985. Stenzel J (1996) Flying off Course: Environmental Impacts of America’s Airports.New York:Natural Resources Defense Council. Stephens, DG, Crazier, FW (1996) NASA noise reduction program for advanced subsonic transports. Noise Control Eng. J. 44:135-144. Tomei F, Tomao E, Papaleo B, Baccolo TP, Cirio AM, Alfi P (1995) Epidemiological and clinical study of subjects occupationally exposed to noise. Int. J. Angiol. 4:117–121. United States Environmental Protection Agency (1974) Information on levels of environmental noise requisite to protect health and welfare with an adequate margin of safety (EPA/ONAC report 55/9– 74-004). United States Environmental Protection Agency, Washington, DC. United States Environmental Protection Agency, Office of Noise Abatement and Control (1977) The Urban Noise Survey. United States Environmental Protection Agency. Washington, DC. United States Environmental Protection Agency, Office of Noise Abatement and Control (1978) Noise: A Health Problem. United States Environmental Protection Agency. Washington, DC. Wachs T, Gruen G (1982) Early Experience and Human Development. New York, NY: Plenum. Wald ML (2003) Under Rules, Train Whistles will Lose Some of Their Blare. New York Times; December 18, 2003, p. A38. Zaner A (1991) Definition and sources of noise. In: Fay TH, editor. Noise and Health. New York: New York Academy of Medicine.

29 RADON AND LUNG CANCER Naomi H. Harley

29.1 RADON AND LUNG CANCER Radon ð222 RnÞ is present in all environments, resulting from the decay of radium, one of the Earth’s natural radionuclides. The entire chain consists of 13 radionuclides and is supported by the parent 238 U (half-life ¼ 4.5  109 years). The National Council on Radiation Protection and Measurements (NCRP, 1984) was the first to link the evidence from underground miner’s lung cancer risk from radon exposure and to project the risk to the U.S. population. At that time, it was clear that most underground mines had high radon concentrations and that home exposure was generally lower than in mines. However, a few homes had equal or even higher measured radon concentrations than some mines; therefore, residential studies became a priority. Lung cancer is the only observed health effect from exposure to radon. Several risk projection models were developed as residential measurements escalated (NCRP, 1984; ICRP, 1987; NAS/NRC, 1988, 1999) because it was necessary to extrapolate from high to low exposures. The pooling of 11 major epidemiological studies in underground mines, from nine countries, showed conclusively that exposure to radon and its short-lived decay products cause lung cancer in excess of that expected (NIH/NCI, 1994; NAS/NRC, 1999a). Tobacco smoke is a confounder and other substances in mine air influence lung cancer induction somewhat; however, the effect of radon is clearly seen. The ores mined in these follow-up studies include uranium, iron, lead, tin, and fluorspar. Some mines have air concentrations of arsenic. However, radon was the primary mine carcinogen with a dose– response relationship in essentially all the 11 underground cohorts. Most countries with significant health programs have now completed residential surveys of radon concentrations. There are 50 countries with surveys that show the average levels in homes to be about 40 Bq/m3 (1 pCi/L) with a few percent of homes showing measured concentrations over 740 Bq/m3 (20 pCi/L). Some differences do exist between radon exposure in the home and in an underground mine, and this is described in detail in Section 29.7. Generally, miners were exposed to

Environmental Toxicants, Third Edition Edited by Morton Lippmann Copyright
2009 John Wiley & Sons, Inc.

1089

1090

RADON AND LUNG CANCER

concentrations about 100 times that found in an average home and the exposures were for relatively few years (from 3 to 20), whereas home exposure is for a full lifetime. The inhaled aerosol particle characteristics differ somewhat, and breathing rates in homes are generally lower than in an occupational setting. Additional mining populations may be added to the database as the demand for uranium expands, and there is a need for long-term follow-up of the existing miner cohorts. The most recent 222 Rn epidemiology is directed toward residential case–control studies. As of now, over 80 residential studies are completed, of which 24 are case–control studies. At present, the existing residential case–control studies are sufficient to show the lung cancer risk directly. The data from the better studies are either pooled or a meta-analysis is performed (using published data) to yield risk estimates that are more precise at present than those obtained from the miner studies. These are discussed in more detail in section 29.6 on Residential Epidemiology. The major confounder in assessing lung cancer risk is always smoking. This is because of the high background lung cancer rate in all smoking populations. The risk from smoking and 222 Rn exposure in the miner populations is much greater than the risk from 222 Rn exposure alone. However, as yet, the residential studies show no difference in the relative risk for smokers and nonsmokers. This still indicates the risk to smokers is much greater because the baseline lung cancer rate for smokers is much higher. This is discussed in Section 29.11. This chapter describes the background information concerning exposures in the mines, exposure to both indoor and outdoor radon concentrations, differences in lung dosimetry between mining and environmental populations, the predictive risk models from the pooled miner follow-up studies, and the residential studies that directly observe risk from radon exposure at home. 29.1.1

Units of Radon Exposure

The decay scheme for the 238 U series is shown in Fig. 29.1 (NCRP, 1988). In soil, below about 1 m, the entire decay chain of radionuclides is in equilibrium (equal activity for each

FIGURE 29.1

Principle decay scheme of the uranium series.

1091

RADON AND LUNG CANCER

member). Radon, being a noble gas, breaks the chain in the top layer of soil and diffuses into the atmosphere. The conventional unit for exposure to airborne radioactivity is Bq/m3 (or historically pCi/L). One Bq/m3 equals 1 disintegration/s/m3 and 1 pCi/L equals 2.22 disintegrations/m/L. Although radon exposure is cited as posing lung cancer risk, it is actually from the solid, short-lived alpha-emitting decay products of radon, 218 Po and 214 Po, which deposit on the bronchial airways and deliver the carcinogenic dose to target cells lining the airways. The targets are the basal and mucous cells that lie in the thin (35–40 mm thick) layer of epithelial tissue lining the airways (Harley et al., 1996a,b; NCRP, 1984a, 1987a, 1987b; NRC, 1988; NIH/NCI, 1994; NAS/NRC, 1999a). The alpha dose to cells from radon gas itself is small compared to that from the decay products. In air, the short-lived decay products (effective T1/2 ¼ 33 m) are never in equilibrium or steady state with the parent 222 Rn gas (T1/2 ¼ 3.8 days) or with each other. This is because, as solids, the decay products readily plate out or attach to available surfaces as they are formed. Radon gas is always present to support the decay product mixture. However, the measurement of the individual nuclides, the actual carcinogens, is usually not done. Thus, to express the relevant concentration of the three radionuclides in terms of the usual Bq/m3 is a problem. To circumvent this, historic mining exposures were documented in terms of the working level (WL), which was easy to measure. A 5 m filtered air sample was taken in the mine and alpha counted after a 40 m waiting period (NCRP, 1988). A simple calculation correcting for the counter efficiency, and dividing by a count/WL factor, obtained the WL rapidly. The WL is defined as the total decay energy of the decay products in equilibrium wit 100 pCi/L radon or any combination of the short-lived daughters in 1 L of air resulting in the release of 1.3  105 MeV of alpha energy. In the historic mines, the dose to the lung was thought to be the dose to the lower or pulmonary lung where decay products deposit and decay completely. Therefore, the working level was thought to be related directly to dose. This was found later not to be true and is discussed in detail in Section 29.7. Cumulative exposure was recorded for workers in units of the working level month (WLM) that was exposure in WL times the number of work months (equal to hours worked/ 170 h). One WL is equal to 3700 Bq/m3 (100 pCi/L) at equilibrium; however, equilibrium with the decay products is never attained. The exact relationship between WL and the shortlived decay products is WL ¼ 0:00103ð218 PoÞ þ 0:00507ð214 PbÞ þ 0:00373ð214 BiÞ;

ð29:1Þ

where 218 Po, 214 Bi, and so on are concentrations of the decay products in pCi/L, or WL ¼ 2:78  10 5 ð218 PoÞ þ 1:37  10 4 ð214 PbÞ þ 1:01  10 4 ð214 BiÞ; where 218 Po, 214 Bi, and so on are concentrations of the decay products in Bq/m3: WLM ¼ WLðhours exposed=170Þ:

ð29:2Þ

The relationship between radon concentration and the short-lived daughters in WL depends mainly upon the ventilation rate, surfaces available for plate out, and the atmospheric particle concentration in the space considered. Plate out on walls or any surface can remove the decay products and air movement or ventilation enhances the plate out. The concentration of submicrometer (<1 mm) particles is important because the freshly formed decay products quickly get attached to these particles. This is known as the attached fraction. Their motion is now controlled by these larger sized particles, which inhibits diffusion and plate out. A small

1092

RADON AND LUNG CANCER

fraction of the decay products remain as small clusters of atoms. This is called the unattached (or ultrafine) fraction. In mines, homes, and outdoors, the approximate relationship between radon and decay products is 1 WL ¼ 300 pCi/L 1 WL ¼ 250 pCi/L 1 WL ¼ 160 pCi/L

Mine Home Outdoors

A simple way of expressing the relationship between the parent products is in terms of the equilibrium factor, F. This is

Rn222 gas Rn222 gas Rn222 gas 222

Rn and the decay

F ¼ 100ðWLÞ=ðpCi=L222 Rn gasÞ:

ð29:3Þ

For the cases above, F, the equilibrium factor, would be 0.33, 0.40, and 0.60 for mines, homes, and outdoors, respectively. The equilibrium factor, F, is very important because the measurement of home exposure is usually to radon gas concentration using alpha track detector measurements over a full year or a large fraction of a year. The decay product exposure is estimated from the radon gas concentration multiplied by the equilibrium factor, F. 29.1.2

Units of Radiation Dose

Radiation dose is defined as the energy deposition per unit mass. The older systems of dose units are the rad and rem, the new system is the gray and sievert. The rad is equal to 100 ergs per gram and the gray is equal to 1 joule per kilogram: 100 rad ¼ 1 Gy: The rem and sievert attempt to correct for the different biological responses from alpha, beta, and gamma radiations. The dose, corrected by the radiation weighting factor, Wr, is called the equivalent dose: rem ¼ Wr ðradÞ; sievert ¼ Wr ðGyÞ; for beta and gamma radiation; Wr ¼ 1; for alpha particle radiation; Wr ¼ 20: These numerical values of Wr are based on the energy transfer per unit path length in water. Future research should establish more realistic values for specific tissues. Because it is the decay products that deliver the dose, the dose factors are usually given for a radon concentration in equilibrium with the decay products, not the gas concentration. This is called the equilibrium equivalent concentration (EEC). Indoors, the equilibrium factor, F, is adopted as 0.4 and outdoors as 0.6 (UNSCEAR, 2000): Rn concentration ðEECÞ ¼ F ðradon gas concentrationÞ: The dose factors are usually given in terms of effective dose (sieverts or rem) not equivalent dose. This requires another factor to account for the fact that the lung dose is a partial body exposure. The tissue weighting factor, Wt, normalizes to an equivalent whole

OUTDOOR RADON

1093

body dose in terms of cancer risk. This is called the effective dose, that is, an effective whole body dose: effective dose ¼ Wr Wt ðrad or GyÞ: In some publications, the use of rem or sieverts neglects to state the factors used, and there is confusion about which dose has been calculated, equivalent or effective dose. It is unfortunate that the rem and sievert are units for both equivalent dose and effective dose. There are several published effective dose factors for radon and a central value is effective dose factor ¼ 15 nSv per Bq=m3 h ðradon Bq=m3 in EECÞ: For example, the annual effective dose indoors, from the U.S. average home radon concentration, 46 Bq/m3, assuming 70% of time is spent indoors, is annual effective dose indoors ¼ ð15 nSv per Bq=m3 hÞ 46  0:4Þ ð0:7Þ ð8760 hÞ ¼ 1:6  106 nSv or 1:6 mSv: This dose calculation uses the calculated radon (EEC) concentration and the dose factor incorporates the weighting factors, Wr and Wt, to obtain the effective dose.

29.2 OUTDOOR RADON Uranium-238 is a trace element in all materials native to the Earth. Radon-222 is the gaseous decay product of 226 Ra, which in turn is a member of the 238 U series. The half-life of 222 Rn is sufficiently long (3.82 days) so that it can escape from the surface of most substances with an efficiency of a few percent. Therefore, 222 Rn exists in all atmospheres on the planet. The entire 238 U series is shown in Fig. 29.1 (NCRP, 1988). Some long-term outdoor measurements of 222 Rn are available. Fisenne (Fisenne, 1988; Fisenne and Keller, 1996; NCRP, 1987a) has reported 222 Rn concentrations in New York City and Chester, NJ, over a 9-year period. These are shown in Fig. 29.2. Harley (1997) has

FIGURE 29.2 Chester, NJ.

Outdoor radon averaged monthly over a 9-year period in New York City and

1094

RADON AND LUNG CANCER

FIGURE 29.3

Monthly average outdoor radon at a suburban New Jersey home 1990–1996.

measured outdoor concentrations in a suburban New Jersey location for 7 years, and these data are shown in Fig. 29.3. Harley (1990) summarized the long-term outdoor data from all countries. These data are shown in Table 29.1. The U.S. Environmental Protection Agency (EPA 1992) assessed outdoor or ambient 222 Rn for the purpose of assessing guidelines for indoor 222 Rn and 222 Rn released from groundwater use in homes. The original EPA measurements and all publications on outdoor concentrations were summarized in the NAS/NRC report on the risk from radon in drinking water (NAS/NRC, 1999b). The published outdoor concentrations for each state in the United States is shown in Table 29.2. Outdoor 222 Rn in the U.S. averages 15 Bq/m3 (0.4 pCi/L) for locations where no elevated sources are present. In the western United States, the potential for mineralized uranium deposits occurs and local outdoor concentrations could increase. However, significant horizontal transport of 222 Rn occurs and measurements at a particular site do not exclusively measure 222 Rn emanated locally. Average wind speed at ground level is 5 km/h and this brings 222 Rn from significant distances to the site of measurement (Harley, 1974). Large short-term differences can exist in outdoor 222 Rn concentration, depending upon local circumstances. For example, in Fig. 29.2, data from Chester, NJ, show 9 years of data where annual averages have a range of only 30%. If hourly data were compared, differences of a factor of 10 can be observed (NCRP, 1987; UNSCEAR, 2007). Radon concentration is highest during the night and lowest in late afternoon. A nighttime inversion occurs (air temperature increasing rather than decreasing with height), as the soil surface cools by radiative heat transfer, resulting in very stable air. There is little upward dilution of radon emanating from the soil surface and resulting high surface 222 Rn. Solar heating at the soil surface during the day causes large turbulent (eddy) diffusion in the atmosphere and dilution of the radon emanating from the soil surface. Vertical wind profiles also mix 222 Rn to great altitude. Radon originating at the soil surface is transported to various heights in the atmosphere depending upon these factors, which are not constant. Therefore, models of 222 Rn with height cannot be precise. Figure 29.4 shows the measured falloff of radon concentration with height (Fisenne et al., 2005). A rough generality for greater heights is that the radon is reduced from ground level concentration by one-half for each 700 m altitude.

OUTDOOR RADON

TABLE 29.1 Location Manila New York City Chicago Netherlands Chester, NJ 1977 1978 1979 1980 1981 1982 1983 1984 1985 New York City 1983 1984 1985 1986 1987 1988

1095

Average Outdoor Radon Concentrations from Extended Measurements Method

pCi/m3

Reference

Charcoal ion chamber 13 L ion chamber Charcoal scintillation cell Karlsruhe alpha track

70 54 (10–120) (50–1600) 90

Wright and Smith (1915) Hess (1953) Moses et al. (1963) Put and deMeijer (1988)

Continuous two-filter

230 230 190 240 220 220 250 200 200

Fisenne (1988)

Continuous two-filter

125 115 130 135 80 90

Fisenne (1990)

200 200 180 140

Harley (1990)

Northern New Jersey 1986 Continuous radon monitor 1987 1988 1989 Range in parentheses. Source: Harley (1990).

Very little 226 Ra exists in seawater and, therefore, 222 Rn over the oceans should be low if no nearby land source can be accessed by wind transport. Table 29.3 (Harley, 1990) is a summary of 222 Rn measured over oceans. It can be seen that the Mediterranean is the exception (4 Bq/m3) to the average values of 0.1 Bq/m3 because wind transport from land is significant. Harley (1974) modeled global 222 Rn concentration based on an average emanation from soil of 20 Bq/m2/s (0.5 pCi/m2/s), the land and water areas of the globe and a half-depth in the atmosphere of 700 m. The estimate of ground level concentration was 8 Bq/m3 (0.2 pCi/L), in good agreement with measurements. The lung cancer risk from exposure to outdoor radon is estimated from current models. This is possible only by assuming a linear, no-threshold model for carcinogenesis. This is done in section 29.8 on risk models. Outdoor radon is a small fraction of indoor exposure and is, at most, 30% of the total exposure (Harley et al., 1991; NAS/NRC, 1999b). For this reason, outdoor 222 Rn is sometimes neglected when calculating health effects from lifetime exposure.

TABLE 29.2

U.S.EPA Outdoor Radon Survey in the United States: Average of All Quarters

State AL AK AR AZ CA CO CT DE FL GA HI IA ID IL IN KY KS LA MA MD ME MI MN MO MS MT NC ND NE NH NJ NM NV NY OH OK OR PA RI SC SD TN TX UT VA VT WA WI WV WY Average Source: NAS/NRC (1999).

1096

Average of All Quarters (Bq/m3) 14.3 10.2 16.4 16.7 14.4 12.8 14.6 13.6 13.7 15.7 7.8 18.8 15.1 18.5 15.2 17.4 20.7 9.8 14.7 16.5 16.4 13.7 13.5 18.5 13.3 17.1 12.0 19.5 20.1 14.8 15.4 5.3 8.1 11.5 13.9 14.6 11.7 17.4 10.7 16.5 21.7 16.7 13.8 10.2 17.2 13.4 17.3 14.8 18.4 9.9 14.8

INDOOR RADON

1097

Upper air radon all data 100000

222Rn (mBq/m3)

10000

1000

100

10

1

0

10

20

30

Sampling altitude (km)

FIGURE 29.4 Stratospheric radon measurements in three North American locations, Alaska, Southwest United States, and Canal Zone.

29.3 INDOOR RADON Indoor radon was first measured in Sweden by Hultqvist (1956) and in the United States by Glauberman and Breslin (1957). Hultqvist found a few homes with concentrations up to 800 Bq/m3 (20 pCi/L) but the significance of the early measurements was not known. This was because quantitative risk estimates for lung cancer in underground miners were not yet available. Beginning in mid-1975, several studies of indoor radon appeared in the literature (Rundo et al., 1979; George and Breslin, 1980; Sachs et al., 1982; Nero, 1983). These indicated that measurements of 222 Rn concentrations in a sample of homes were distributed lognormally. The geometric mean 222 Rn concentration was generally 60 Bq/m3 (1.5 pCi/L) with a geometric standard deviation of 2.0–2.5. In December 1984, a home occupied by the now famous Watras family was found to have 222 Rn concentrations of 110,000 Bq/m3 (3000 pCi/L). The Pennsylvania Department of Environmental Resources moved the family (NCRP, 1989; Reilly, 1990). Stanley Watras, an TABLE 29.3 Ocean Mediterranean North Atlantic South Pacific Antarctic

Radon Concentration over the Oceans Concentration (pCi/m3) 100 6 1.5 0.8

Range in parentheses. Source: Harley (1990).

Reference Servant (1966)

1098

RADON AND LUNG CANCER

engineer at a nuclear power facility, showed significant external contamination of his clothing that was not related to any workplace exposure. Investigation showed that the radioactive contamination was derived from his home. The short-lived decay products of 222 Rn formed in air deposit on any surface, including people and clothing. Lung cancer risk estimates for exposure to environmental concentrations were available at this time (NCRP, 1984b) and the family was moved because of the public health implications. This incident was a watershed for residential 222 Rn in that it was clear that homes could attain 222 Rn concentrations equal to the highest historical values observed in mines. Surveys have now been carried out in the living space of homes in 50 countries. These are summarized in Fig. 29.5 (UNSCEAR, 2000 and updated). In the United States, a stratified random sample of 5694 dwellings was conducted by the EPA to determine average exposure. Over 15,000 alpha track detectors were placed in all levels of the homes for 1 year. The measured median concentration in the United States was 25 Bq/m3 (0.67 pCi/L) with an average of 46 Bq/m3 (1.25 pCi/L) About 6% of homes had measured concentrations greater than 150 Bq/m3 (4 pCi/L). Some individuals, for example Dr. Bernard Cohen of the University of Pittsburgh, have measured 222 Rn for various scientific purposes. Dr. Cohen has measured 222 Rn in 175,000 homes and obtained data in living areas for most of the counties in the United States (Cohen, 1992). Nero et al. (1986) utilized published data to arrive at an estimate of the typical distribution of 222 Rn exposures in living areas. Many surveys, such as the state surveys supported by EPA, measure 222 Rn in the basement, if a basement is available. Basement measurements are not indicative of actual exposure and overestimate actual exposure, on average, by factors of 2–5. Personal exposure measurements in 52 homes and 88 occupants in a study conducted near Chicago showed there was a consistent relationship between first floor living levels and

FIGURE 29.5

Average radon concentrations indoors from residential surveys in 50 countries.

INDOOR RADON

1099

actual personal exposure (Harley et al., 1991). The personal exposure was on average 0.71 of the measured first floor concentration with good correlation (R2 ¼ 85%). Correlation of personal exposure with basement 222 Rn measurements was poor. Exposure outside the home in public places and offices is generally lower than home exposure and near to outdoor levels due to differences in construction and ventilation. Thus, from information available from many countries, the average indoor exposure is about 40 Bq/m3 (1 pCi/L). The model estimates for lung cancers for average exposure, ranges from 9000 to 15,000 per year in the United States (NCRP, 1984a), NAS/NRC, 1988; Lubin and Boice, 1989; NIH/NCI, 1994; NAS/NRC, 1999a). The average nonsmoker in the United States has a calculated risk of about 1 in 2000 of dying from 222 Rn-induced lung cancer and smokers 1 in 200. Of importance is the distribution of exposures, with perhaps 5% of the population having exposures of greater than 150 Bq/m3 (>4 pCi/L) as estimated by the various individuals and organizations (NCRP, 1984b; Nero et al., 1986; NAS/NRC, 1988; NIH/NCI, 1994; NAS/NRC, 1999a). Using the variables known to control 222 Rn entry into a building, models can adequately predict the variability of indoor concentrations (Harley and Chittaporn, 1993). The major variables are (1) the baseline 222 Rn concentration established by diffusion, radium concentration of soil, and soil temperature; (2) indoor/outdoor temperature difference, which in turn causes a differential pressure known as the stack effect; (3) the air exchange rate in the home; and (4) barometric pressure changes. To a lesser extent, wind and rain affect the indoor concentration. With a few basic measurements in a home, the variability can be modeled; however, it is not as yet possible to predict the concentration in a home without baseline measurements. The source of indoor 222 Rn is the 226 Ra bearing soil under any home. Another potential source, described in the next section, is groundwater when it is used as the residential water source. A clear decrease in 222 Rn concentration can be seen with height above ground level, with an average factor of 2 for the ratio of basement to first floor concentrations (Cohen, 1992; Harley et al., 1991; UNSCEAR, 2000). The only exception to the soil as a source of 222 Rn is in high-rise apartments. In this case, the soil is too remote to dominate as the source for apartments above about the third floor. Construction materials such as wallboard play a role, as well as outdoor 222 Rn exchanging with indoor air. Harley (1991) summarized the data for high-rise apartments and showed the average indoor concentration to be a factor of 2 higher than outdoor concentration. A detailed knowledge of the geological properties of a particular region could help to identify geographic areas of potentially high indoor radon in single-family dwellings. EPA has produced a map of radon zones in the United States, based on aerial surveys. These data can be accessed at www.epa.gov/radon/zonemap.html. In general, 222 Rn is measured in homes during a real estate transaction and is mandatory in some states because of EPA guidelines. Radon-resistant construction is included in some new homes. It is necessary for the U.S. EPA to update frequently the results of the remediation due to their guidelines, for example, overall radon reduction. 29.3.1

Ground Water as a Source of Indoor 222

222

Rn

In some cases, indoor Rn is increased substantially by the use of groundwater. Radon, like any gas, is somewhat soluble in water. Well water or groundwater in close contact with rock and soil can exhibit typical concentrations of 222 Rn range from 20 to 200 Bq/L. Some wells in the State of Maine, for example, have measured 222 Rn concentrations of 3.7  104 Bq/L

1100

RADON AND LUNG CANCER

(106 pCi/L) (Hess et al., 1985, 1987). Surface water, that is, from reservoirs, is very low in 222 Rn concentration because the gas is readily removed by surface agitation. The dose to internal organs from drinking water with even high concentrations is very low (Harley and Robbins, 1994). However, the 222 Rn released to indoor air through normal use, such as showering, can increase the overall indoor concentration significantly depending upon the ground water concentration (Hess et al., 1985, 1987; Chittaporn and Harley, 1994; NAS/NRC, 1999b). Airborne 222 Rn and decay products derived from water cannot be distinguished from 222 Rn entering the dwelling from the usual soil source. Estimates of the contribution of groundwater to indoor air based on all published data find a factor of 1/10000, that is, 10,000 units (pCi/L or Bq/L) in ground water will yield on average 1 unit in indoor air (NAS/NRC, 1999b). The full data set in the report “Risk Assessment of Radon in Drinking Water” may be accessed at www.nap.edu. On October 1, 1997, the Swedish Government, by request from the National Food Administration, set a limit for 222 Rn in public water supplies of 100 Bq/L (2700 pCi/L). The radon must be reduced if the concentration is >100 Bq/L, and public drinking water cannot be delivered for use if the concentration exceeds 1000 Bq/L. The U.S. EPA has the authority to set a 222 Rn limit for drinking water supplied for public use (>25 households). Because 222 Rn is a known carcinogen, this limit can be set to zero. However, as zero concentration is not attainable by any present technology, EPA has proposed a limit of 300 pCi/L (11 kBq/m3), on the basis of the practical detection limit for the measurement of 222 Rn in water. There is considerable pressure to increase this proposed limit because of the high cost of remediating water to this limit, and because of the negligible health consequences implied from air concentrations of 222 Rn associated with water use near this limit. The final EPA decision has not yet been made.

29.4 THE OTHER RADON,

220

Rn, Thoron

The radioactive 232 Th (half-life ¼ 1.4  1010 years) series also has a gaseous isotope of radon, 220 Rn, thoron. The half-life of 220 Rn is short (55 s), air concentrations are generally lower, and the equilibrium factor is small (0.02), so the dose from thoron and its decay products should be lower than 222 Rn. Demonstrable health effects have not, as yet, been shown for thoron. An emerging issue is that many measurements made for radon gas also unknowingly include thoron gas. Some of the risk estimates based on gas measurements, without correction, are therefore in error. Yamada et al. (2005), for example, showed the correction for thoron to the measurements in the Chinese Loess Plateau epidemiological data decreased the original dose estimates by more than a factor of 2.

29.5 RADON EPIDEMIOLOGY IN UNDERGROUND MINES The pooling of the follow-up studies of 11 underground miner cohorts provides the best data to date for the risk to miners, and also for modeling the features of lung cancer appearance. The excess risk in miners was observed at high exposures and has been extrapolated to the risk at environmental 222 Rn concentrations (NCRP, 1984; NAS/NRC, 1988; NIH/NCI, 1994; NAS/NRC, 1999a). The underground mining populations are from China, Czechoslovakia, Canada, Colorado, New Mexico, France, Australia, and Sweden.

RADON EPIDEMIOLOGY IN UNDERGROUND MINES

1101

ERR/WLM

10

1

ne d C om bi

h or ad o O N ew nta ri fo un o dl an d Sw e N ew den M e B ea xico ve r Po lod ge rt R ad iu m Fr an ce ol

ze c

C

C

C

hi

na

0.1

FIGURE 29.6 Excess relative risk per WLM exposure (ERR/WLM) in 10 underground mining cohorts. From NIH/NCI (1994) with updated data, Radium Hill data omitted.

Although the results from 11 studies were pooled, it is clear that the exposure assessments for the various cohorts were of extremely varied quality. Fig. 29.6 shows the risk and the large error margins. In the Eldorado mine (Canada, Beaverlodge mine), for example, it is clear that exposures in mines other than the Beaverlodge contribute to lung cancer in these miners. The additional exposure was not documented in the miner’s work record (Chambers et al., 1990; UNSCEAR, 2007). In the Australian Radium Hill cohort (Woodward et al., 1991), the authors state the increased rate of lung cancer in this workforce may be due wholly or partly to factors other than radiation. Cigarette smoking is potentially the most important confounding factor. It remains, however, that lung cancer above that expected from smoking, the major carcinogen, is reported in all these mines, and radon and radon decay product exposure is the primary mine carcinogen, shown by the clear dose–response relationship. The detailed information concerning the 11 mining cohorts can be found in NIH/NCI (1994), NAS/NRC (1999), and UNSCEAR (2007). Miners were exposed for a period of 3 to over 20 years. Radon exposures were made in most of the mines after 1950, but measurements were generally sparse, and data on individuals, and sometimes entire mines, had to be interpolated. In all mines, the weakest part of the study was the poor quality of the exposure estimates. The excess relative risk of lung cancer per WLM exposure (ERR/WLM) for 10 of these 11 cohorts is shown in Fig. 29.6 (NIH/NCI, 1994, with updated data). Radium Hill data are omitted because of the very large confidence intervals. The observed lung cancer, as a function of cumulative exposure in WLM for the pooled data from all 11 cohorts, is shown in Fig. 29.7 (NIH/NCI, 1994). One of the best mining populations for establishing risk from environmental radon is the Colorado Plateau uranium miner follow-up. This is the only study, so far, to have essentially complete smoking information on each miner. Other studies rely on a small subgroup within the cohort for smoking information. The Colorado uranium miner follow-up does not show excess lung cancers below a cumulative exposure of 120 WLM. Unfortunately, complete information to project miners’ risk to residential radon exposure is lacking. Home exposure for an average of 150 Bq/m3 (4 pCi/L) yields an equivalent

1102

RADON AND LUNG CANCER

FIGURE 29.7 Relative risk of lung cancer as a function of exposure in WLM. All 11 underground mining cohorts. From NIH/NCI (1994).

lifetime exposure of about 40 WLM. One important mining study showing excess lung cancer risk, the Ontario uranium miners, has documented low exposures, given as a range from 33 to 74 WLM (Muller et al., 1983; Senes Final Report, 2006). So far, no threshold for lung cancer is proposed and, if one exists, it is very near the low end of residential exposures.

29.6 RESIDENTIAL EPIDEMIOLOGY Lung cancer excess (above that expected from smoking) in the underground miner populations has been demonstrated conclusively. Combining this, and the knowledge that some homes have radon and daughter concentrations above those found in historical mines, it is virtually certain that environmental radon is responsible for some lung cancer in the general population. There are more than 80 published environmental epidemiological studies of radon exposure to determine whether health effects can be documented directly. Of these, 24 are case–control studies summarized (either using pooled or meta-analyses) by Lubin and Boice (1997), Pavia et al. (2003), Krewski et al. (2005), and Darby et al. (2005). Most non-case–control studies show either a slight positive or negative correlation between measured radon in the home and lung cancer mortality. The majority of studies are ecological exercises that relate lung cancer mortality in a region with indoor radon concentration. In some cases, the radon is not measured but estimated as high or low depending upon the type of house. The ecological studies are unsatisfying, because no attempt is made to determine actual exposure to individuals in the area of study, and no correction can be made for smoking, the strongest confounder for lung cancer. Several case–control studies attempted to verify past exposure accurately using various methods. The New Jersey study, Schoenberg et al. (1990), analyzed data for 400 lung cancer cases and 400 controls. The individuals had to be living in their present residence for at least 10 years. Radon measurements were made in the current residence on most living levels, and exposure measurements in past residences were also performed. The Iowa study, Field et al. (2000), selected subjects living in the same home for the past 20 years and measured radon concentrations using at least one alpha track detector on each living level for 1 year.

RESIDENTIAL EPIDEMIOLOGY

1103

The largest case–control study to date was performed in Sweden (Pershagen et al., 1994). There were 1360 cases and 2847 controls. The exposure was assessed with 3-month measurements during the heating season, retrospectively assessing homes lived in for more than 2 years, beginning in 1947, up to 3 years prior to diagnosis of cancer. The lung cancer excess was statistically significant only for smokers or nonsmokers with a concentration over 400 Bq/m3 (>10 pCi/L) for over 32 years. A weak point in all studies is accurate radon exposure history over the relevant period for lung cancer, that is, from 5 to 30 years prior to the onset of the disease. This accuracy is not yet possible and there are many attempts to find a technical solution. One is by measuring the long-lived radon decay products, 210 Pb and 210 Po, embedded on glass surfaces that are more than 20 years old (Samuelsson, 1988, 1996; Walsh and McLaughlin, 2001). So far there is no technique to determine accurate measurements of 30-year past exposure, but support is given by the various supplementary methods. The first meta-analysis was performed using results from eight published case–control residential studies. A meta-analysis uses published results and does not have access to the original data. The lung cancer excess was not statistically significant but the trend with increasing concentration in the homes was significant (Lubin and Boice, 1997). The graph of the eight studies from Lubin and Boice (1997) is shown in Fig. 29.8. There are technical difficulties with meta-analyses or pooling of the residential studies to obtain a risk estimate. These are described by Neuberger et al. (1996). Many of the protocols are disparate, and the dose response varies. In spite of the technical problems, there are now six meta- or pooled analyses of the residential data, and these provide the best estimate of residential risk. A summary of the risk (OR ¼ odds ratio) from the six meta- and pooled studies is shown in Fig. 29.9. One of the best pooling results is that of Darby et al. (2005). This included 13 European studies and accounted for errors in measurement. The best present estimate of residential risk is an OR of 1.16 (95% CI: 1.06–1.28) for an exposure of 100 Bq/m3. There was no difference in the risk estimate for smokers or nonsmokers. This differs from the mining studies where smokers had a risk of about three times that of nonsmokers. Because of the baseline differences, the lung cancer risk for smokers is still much higher than for nonsmokers. See Section 29.11 for an example calculation.

FIGURE 29.8 Boice (1987).

Meta-analysis of eight residential radon epidemiologic studies. From Lubin and

1104

RADON AND LUNG CANCER

Pooled or meta-analyses of residential radon studies 1.40 95% Confidence intervals

1.35

Odds ratio (OR)

1.30 1.25 1.20 1.15 1.10 1.05 1.00 0.95 North Meta (Lubin Meta (Pavia European et al., 2003) (Darby et al., American and Boice, 1997) (Krewsky 2005) et al., 2005)

FIGURE 29.9 cancer.

China (Lubin, 2003)

Germany (Wichmann et al., 2005)

Pooled or meta-analyses including 24 indoor residential studies of radon and lung

29.7 LUNG DOSIMETRY Lung cancer due to radon exposure is due to the inhalation of the particulate short-lived daughters of radon 218 Po, 214 Pb, and 214 Bi–214 Po (the half-life of 214 Po is so short that it is always in steady-state equilibrium with 214 Bi). As each 218 Po atom is formed from the decay of 222 Rn, it grows by condensation of water and other molecules, such as SO2, to a few nanometers diameter. In ordinary indoor atmospheres, there are abundant submicrometersized particles (from 1000 to 50,000 cm 3 in the 50–200 nm size range) present from various sources such as energy use, influx of outdoor aerosol, and so on. Subsequent to formation from 226 Ra, most atoms of 218 Po rapidly attach to this larger size aerosol. As the 218 Po atom decays through the chain—214 Pb to 214 Bi–Po—these subsequent radioactive atoms (mostly) remain on the carrier aerosol particle. The 218 Po and some 214 Pb not attached to the carrier aerosol are called the unattached fraction, while the decay products attached to the carrier aerosol are called the attached fraction of 222 Rn daughters (NCRP, 1988). In homes, there are normally fewer aerosol particles per unit volume than in underground mines. Therefore, the fraction of the radioactivity that is unattached is smaller in mines than in homes, because there is a greater probability for attachment to the abundant number of aerosol particles present in mine atmospheres. Inhaled decay products of radon are primarily on 50–200 nm diameter carrier particles. A varying percentage (from a few to over 50%) of the 218 Po radioactivity exist as the original ultrafine particle of a few nanometers size. The actual size of the nanometer aerosol depends on the trace gas composition, as well as other factors not well defined at present. The greater the concentration of carrier aerosol, the smaller the unattached fraction of 218 Po. This is because rapid diffusion of the unattached species allows interaction with the larger aerosol particles and an interaction will remove the unattached particle to become part of the larger particle size population.

LUNG DOSIMETRY

1105

Polonium-218 emits an alpha particle and it is possible that the recoil energy is large enough to dislodge or free the daughter 214 Pb atom from the larger particle to which it is attached so that it also is in the unattached form. Some unattached 218 Po atoms can decay to 214 Pb before attachment to larger particles, and also contribute to airborne unattached 214 Pb. The ratio of unattached 218 Po=214 Pb is considered to be 10/1 for dosimetric situations. As the mixed aerosol of unattached and attached 218 Po=214 Po is inhaled and exhaled, radon decay products are deposited by diffusion deposition on the airway lumen. All of the aerosol particles are too small for sedimentation or impaction to play a significant role in airway deposition. The primary carcinogenic dose is then delivered by alpha particles from the decay of 218 Po and 214 Po residing on the bronchial airways. These alpha particles easily reach the target cells in bronchial epithelium (basal and/or mucous cells) as their ranges in tissue are 47 and 70 mm. The basal and mucous target cells are located below the epithelial surface at an average depth of 27 and 18 mm, respectively. Figure 29.10 shows a section of human bronchial epithelium with the target basal and mucous cells identified (Robbins, 1990). These stem cells are the only targets for carcinogenesis because they are capable of division to maintain and replace terminally differentiated epithelial cells normally lost throughout life. The lifespan of most terminally differentiated epithelial cells is about 4 months.

FIGURE 29.10 A transmission electron micrograph of a bronchus from the uninvolved portion of the lung of a 69-year-old male smoker with adenocarcinoma. Among the ciliated cells of the epithelial lining are mucous (M), basal (B) intermediate/indeterminate (I), and granulated endocrine or APUD (amine precursor uptake and decarboxylation) (G) cells. Beneath the thick basement membrane (Bm) is an incomplete layer of fibroblasts, one of which is sectioned through its nucleus (F). Within the connective tissue of the lamina propria are capillaries (C), a mast cell (Ma) and an eosinophil (E). Magnification 2500. Bar ¼ 5 mm. (Robbins, 1990).

1106

RADON AND LUNG CANCER

The inhalation and deposition of any particles on the bronchial airways is accompanied by mucociliary clearance of particles up the tracheobronchial tree to the pharynx. This is the major defense of the lung against inhaled material deposited on the conducting airways. A steady-state activity through inhalation, deposition, and clearance is attained for each of the radioactive daughter products, 218 Po,214 Pb, and 214 Bi=214 Po. Diffusion deposition onto the airway wall increases as the particle diameter decreases. Therefore, the steady-state radioactivity on the bronchial tree is considerably higher for small-sized aerosol particles. For example, open flame burning or space heating with a kerosene heater can produce an aerosol with a median diameter of 30 nm or a factor of about 4 smaller than that considered in a normal room aerosol, that is, 120 nm diameter (Tu and Knutson, 1988; NAS/NRC, 1991, 1999a; Harley et al., 2005a, 2005b). Diffusional deposition on the bronchial airways has now been measured exhaustively in casts of the human lung. The deposition is well described by the equations below (Cohen and Asgharian, 1990) and varies as nd ¼ aðDÞÞb ;

ð29:4Þ

where a and b are constants, D ¼ BLD/(4Q), L is the airway length, D is the diffusion coefficient, and Q is the flow rate. This compensating factor, inverse proportionality with flow rate, along with the higher values of unattached fraction in the home versus mines, actually leads to similar values of the dose conversion factor in both homes and mines. This is an important feature because the risk estimates for 222 Rn exposure obtained from mining exposures may be transported to residential situations other factors remaining the same. There are actually three factors operating in the dosimetry of mines versus homes that equalize the dose. These are (1) similarity in particle size spectra in mines and homes (George and Hinchliffe, 1972; George and Breslin, 1980); (2) the breathing rate under working conditions in a mine is higher than in homes, leading to somewhat lower dose per unit exposure than in homes; and (3) the unattached fraction of the radioactivity is higher in homes and lower in mines (Ruzer, 1995). Figure 29.11 shows the complete modeled relationship between alpha dose to cells in bronchial epithelium located at a mean depth of 27 mm below the epithelial surface and the median particle size of a carrier aerosol (geometric standard deviation of 2). The depth of stem cells in the bronchial airway epithelium has been studied in over 10,000 tissue sections by Robbins (1990) and Robbins and Meyers (1995), and average depths summarized in Harley et al. (1996b). Figure 29.11 shows the model calculations for different breathing rates and unattached fractions (UNSCEAR, 2008). The dose is essentially the same for different ratios of individual decay products, expressed in terms of potential energy (WLM). Thus, disequilibrium of the decay products in any atmosphere does not lead to significantly different bronchial dose for the same total potential energy (WL or WLM) content. There are only a few moderately large-scale studies of the particle size of indoor aerosols (George and Breslin, 1980; Reineking and Porstendorfer, 1986; Reineking et al., 1990; Hopke et al., 1995; summary in NAS/NRC, 1999a; Harley et al., 2005). Because the dose for the same radioactivity content in an atmosphere depends mainly upon particle size, there is a great need for more detailed information concerning the indoor aerosol. In addition to 222Rn dosimetry, a study of the indoor aerosol particle size and composition is important because its overall impact on respiratory disease is poorly understood.

LUNG CANCER MODELS FOR HUMANS

1107

Dose factors for selected breathing rates fpot 0.05

nGy per Bq/m3 (EEC Rn)

100

1.2 m3/h nasal 1.2 m3/h oral 0.6 m3/h nasal 0.3 m3/h nasal

75

50

25

0 0

200

400

600

800

1000

1200

Median aerosol diameter (nm) Medianσg = 2

FIGURE 29.11 Dose factors for radon (EEC) as a function of breathing rate and particle size. From UNSCEAR, 2000, 2008.

29.8 LUNG CANCER MODELS FOR HUMANS The first risk projection model for residential exposure was developed in 1984 by NCRP (NCRP, 1984). Subsequent to the NCRP model, two other models based on the mining epidemiology came to be accepted as the best available risk estimates (ICRP, 1987; NAS/ NRC, 1988, BEIR IV). At present, two additional models are widely accepted. The first was developed by the NIH (1994) with pooling of 11 underground mining cohorts. The BEIRVI Committee (NAS/ NRC, 1999a) used the same data and refined the models somewhat. Once the residential epidemiologic studies became available to determine risk without projection models, it became clear that the NIH and BEIR VI projection models are conservative. Both the NIH and BEIR VI models generally overestimate residential lung cancer risk by factors of 2–3 (NIH/NCI, 1994; NAS/NRC, 1999a). The models are still useful in that important features of lung cancer risk, such as decreasing risk with time since exposure, can be observed in the mining cohorts. This is not possible in the low-exposure residential studies. The historic model derived by the National Council on Radiation Protection and Measurements (NCRP, 1984) is discussed briefly, as this was the first to show the reduction of risk with time since exposure. This risk reduction has been seen in all subsequent models. 29.8.1

NCRP Model

The National Council on Radiation Protection and Measurements was the first to propose a model for residential lung cancer risk based on the miner data (NCRP, 1984a). The model was cognizant of the fact that miners exposed for the first time at ages above 40 appeared to have a higher lifetime risk of lung cancer than miners exposed for the first time in their 20s. Because age, as such, was not thought to confer greater risk for lung cancer, this effect was

1108

RADON AND LUNG CANCER

assumed to be due to a reduction in risk with time. Thus, earlier exposures was assumed to diminish because of cell death or repair of cells transformed by earlier exposure. This halflife for repair (or loss) was assumed to be 20 years. This was the first model to recognize risk reduction with time postexposure. One key factor noted by NCRP was that lung cancer is a rare disease before age 40, regardless of the population considered. A miner exposed at young ages did not generally appear as a lung cancer case until the usual cancer ages were attained (50–70 years). This would account for an apparent increase in lifetime lung cancer risk at older ages because there would be a shorter time for the loss of transformed cells to occur compared to a person exposed at young age, Miners were exposed, on average, for less than 10 years in the Colorado, Ontario, and Czech cohorts. The total time for follow-up was 20 or more years so that the apparent reduction of risk with time from exposure could be observed. The NCRP model took the form of an exponential reduction with time from exposure, with the stipulation that there was a minimum latent period (the time between exposure and a frank cancer) of 5 years. Also, lung cancer could not appear before age 40. This model is known as a modified absolute risk model. Risk is expressed following exposure regardless of other risks of lung cancer such as smoking, but risk is modified by time from exposure. Mathematically, the risk is expressed for each year’s exposure by Rðt; TÞ ¼ CEF½expð 0:345ðt TފL=L0 ; F¼0

ð29:5Þ

if t < 40;

F ¼ 0 if t  40; ðt TÞ must be  5; where R(t, T) is the lung cancer mortality at age t, from an exposure at age T; C is the risk coefficient, lung cancer rate per year per WLM exposure; E is the exposure in 1 year’s time in WLM; and L/L0 is the life table correction for other causes of death (persons alive at age t, divided by persons alive at age T per 100,000 born). To express lifetime risk following a single exposure, it is necessary to sum the risk in Equation 29.5 over the number of years of life following exposure, taken as age of exposure to age 85: LRðTÞ ¼

85 X

Rðt; TÞ:

ð29:6Þ

T

The lifetime risk, TR, from continuous exposure was expressed as the sum of lifetime risk for a single year’s exposure, Equation 29.6, over the total exposure interval considered: TR ¼

85 X

LRðTÞ;

ð29:7Þ

t0

where t0 is the age at first exposure. At the time the model was developed, there was not enough information on the risk of smoking and 222 Rn exposure combined to separate an additional effect from this carcinogen. It was stated that the risk coefficient, C, could be modified when sufficient data were available. Numerical values of lifetime risk for different exposure protocols are shown in Table 29.4.

1109

0.90 0.62 0.56 1.1

1.8

2.0

0.7 2.3 1.2 0.8 (100 Bq/m3) 0.55 (100 Bq/m3) Range 0.5–2.3%

ICRP (1987) ICRP (1987) ICRP (1993) BEIR IV (NAS/NRC, 1988)

NIH/NCI (Lubin and others 1994)

BEIR VI (NAS/NRC, 1999a)

Meta-analysis (Lubin and Boice, 1997) EPA (2003)

Pavia et al. (2003) Darby et al. (2005)

Krewski et al., 2005

Overall

Source: NAS/NRC (1999b) with updated data.

0.70

Lifetime Risk (percent)

NCRP (1984)

Model

Residential; Conditional likelihood regression

Miner based; modified NAS/NRC BEIR VI model Residential; meta-analysis 17 studies Residential; two models used

Miner based; modified absolute risk; two-parameter model Miner based; constant relative risk Miner based; constant additive risk Miner based; single value risk per WLM Miner based; two-parameter model; two time-modified relative risks; two time windows Miner based; modified relative risk; three time windows, age, and exposure rate; three-parameter model Miner based; modified relative risk; three time windows, age, and exposure rate; three-parameter model Residential; observed mortality

Model Type

Pooled odds ratio, 17 residential studies Pooled 13 European studies; no difference with smoking, age, or sex Pooled seven North American studies

Adopted lifetime risk per WLM Risk decreases with time since exposure and decreases with very high exposure Risk decreases with time since exposure and decreases with very high exposure Risk decreases with time since exposure and decreases with very high exposure Linear regression fit to data from eight residential studies Assumes continuous exposure

Risk decreases with time since exposure

Comment

TABLE 29.4 Lung Cancer Risk for Continuous Whole Life Exposure to 148 Bq/m3 (4 pCi/L Unless Noted), at Indoor Conditions as Predicted by Various Underground Mining or Residential Models

1110

RADON AND LUNG CANCER

29.8.2

BEIR IV Model

The National Research Council, Biological Effects of Ionizing Radiation IV (BEIR IV) Committee, produced a report on the health risks of radon and other internally deposited radionuclides (NAS/NRC, 1988). The committee was given the raw data or selected parts of the original data from four mining cohorts, the United States (Colorado), Canadian (Ontario and Eldorado), and Malmberget (Swedish). Reanalysis was performed using the program AMFIT developed for analysis of the Japanese A-bomb survivor data. The program estimates parameters based on Poisson regression. The data were analyzed using both internal and external cohorts for a control population. The BEIR IV committee stated that a relative risk model fit the observed mortality well. The relative risk model assumes that radon exposure increases the known baseline age-specific lung cancer mortality rate in the population by a constant fraction per WLM exposure. However, in all cohorts, there was an obvious reduced lung cancer mortality with time from exposure. The relative risk model was modified to reduce risk with time since exposure. The BEIR IV Committee called their modified relative risk model a time since exposure (TSE) model. Smoking was examined as a confounder. The only study with complete smoking history on the miners is the Colorado study. The effect was tested using a hybrid relative risk model, which incorporated a mixing parameter for smoking. A parameter value of zero fit an additive effect of smoking and 222 Rn interaction, while a value of 1 fit a multiplicative model best. A maximum log-likelihood test was applied to the data, and it was found that the best parameter fit was between 0 and 1 indicating the combined risk but was more than additive but less than multiplicative. That is, the lifetime risk of lung cancer from radon exposure did not simply add to the lifetime risk of lung cancer from smoking. Neither did the risks multiply. The risk of radon and smoking appeared to be between these two extremes. However, in the final BEIR IV model, the risks of radon and smoking were treated as though they were multiplicative because there was no methodology to treat the hybrid model. The TSE model was given the mathematical form rðaÞ ¼ r0 ðaÞ½1þ0:025ðaðWLM1 þ0:5WLM2 ފ;

ð29:8Þ

where r(a) is the age-specific lung cancer mortality rate, r0(a) is the baseline age-specific lung cancer mortality rate, a is 1.2 for age <55, 1.0 for age 55–64, and 0.4 for age 65, WLM1 is the WLM incurred between 5 and 15 years before age a, and WLM2 is the WLM incurred 15 or more years before age a. The exposure data for the Eldorado cohort were not considered carefully by the BEIR IV Committee. A reported Eldorado mining exposure of 1 WLM gave a 50% excess lung cancer mortality, clearly an erroneous value. It is known that the Eldorado miners had prior exposure in other mines, but this additional exposure was not added to the Eldorado exposure (Chambers et al., 1990). The exaggerated risk per WLM in this study for the 1 WLM exposure cohort had a significant influence on the overall BEIR IV model in that this exposure group included a large number of person-years; therefore, when combining the four cohorts to yield a best estimate of the relative risk coefficient, the 1 WLM group and its erroneously high risk carried a substantial weight. If this data point was omitted, the risk coefficient in the model would be 0.015/WLM instead of 0.025/WLM, as used in the final BEIR IV model. Thus, considering all the inaccuracies incorporated into the BEIR IV model, the calculated risk estimates for both smokers and nonsmokers at environmental exposures are overestimates.

LUNG CANCER MODELS FOR HUMANS

1111

The value of lifetime risk as calculated by the BEIR IV TSE model is shown in Table 29.4. 29.8.3

NIH/NCI Model

The National Cancer Institute coordinated an effort to pool the epidemiological data from 11 underground mining studies. The pooled results are reported in NIH/NCI (1994) and were the most complete analysis of the health detriment to underground miners at that time. The analysis was published in the document “Radon and Lung Cancer Risk: A Joint Analysis of 11 Underground Miners Studies” (NIH/NCI, 1994). This work brought together the investigators from each of the 11 mining groups, and their data were analyzed jointly to provide the best information for estimating the lung cancer risk from exposure to 222 Rn and decay products. There were 2701 lung cancer deaths among 68,000 miners accumulating about 1.2 million person-years of exposure. In all the 11 cohorts, the excess relative risk (ERR) of lung cancer (the fractional increase in lung cancer) was linearly related to the cumulative exposure estimated in WLMs. The Colorado uranium miner data from NIH/NCI (1994) is shown as a typical example of the 11 cohorts (Fig. 29.12). The ERR/WLM for all of the studies is shown in Fig. 29.6. One important aspect of the data shown in Fig. 29.12 is that the ERR at high exposures tends to flatten out. This observation is erroneously called the inverse exposure effect. It is usually stated that the lung cancer risk per unit exposure increases for low exposures compared to high exposures. The flattening of the response curve is typical in all studies with animals and high exposure and is likely the result of cell killing due to multiple traversals of cell nuclei. Therefore, the effect is a reduced response at high exposure, not an increased response at low exposure. This terminology has caused considerable misinterpretation with the implication that residential exposure can somehow be more dangerous than mine exposure. This is not true, and it has been demonstrated that no additional risk above the linearity shown in all cohorts is present in residential exposures. The residential studies now confirm this fact.

FIGURE 29.12 Relative risk of lung cancer in U.S. Colorado uranium miners as a function of exposure in WLM. From NIH/NCI (1994).

1112

RADON AND LUNG CANCER

TABLE 29.5 Lifetime lung cancer risk estimates for residential exposure from both BEIR VI models (NAS/NRC 1999a). Male Ever-Smokers

Male Never-Smokers

Female Ever-Smokers

Female Never-Smokers

Exposure–age–concentration model 100 Bq/m3 1.318

1.775

1.352

1.821

Exposure–age–duration model 100 Bq/m3 1.214

1.518

1.235

1.547

The main features of the lung cancer risk model derived from the jointly analyzed data are as follows: 1. There is a reduction in risk subsequent to cessation of mining. This is called the time since exposure (TSE FACTOR) effect. 2. There appears to be no clear effect of age at start of exposure, that is, the age at the start of mining is not a factor. However, the age attained after the start of mining is a factor, and there is decreased risk with older age subsequent to exposure (AGE FACTOR). 3. Longer duration (DUR FACTOR) or lower 222 Rn concentration (WL FACTOR), gives rise to larger risk. As this is the way the model parameters are derived, it is the reason for the so-called inverse exposure effect. The two models derived from the joint analysis are considered equally likely as a fit to the observations. A striking feature of the data is the time since exposure effect, with three time windows modeled for the joint analysis, versus two time windows modeled in the BEIR IV report, when four cohorts were available for analysis. 1. TSE/AGE/WL: RR ¼ 1 þ bðw5--14 þ q2 w15--24 þ q3 w25þ Þðfage Þðg WL Þ; where w5–14, w15–24 and so on denote the exposures in WLM 5–14 years, 15–24 years, and so on before the current age; b ¼ 0.0611, q2 ¼ 0.81, and q3 ¼ 0.40; fage ¼ 1 for age <55, 0.65 for age 55–65, 0.38 for age 65–75, and 0.22 for age >75; and g WL ¼ 1.0 for WL < 0.5, 0.51 for 0.5 < WL < 1.0, 0.32 for 1.0 < WL < 3.0, 0.27 for 3.0 < WL < 5.0, 0.13 for 5.0 < WL < 15.0, and 0.10 for WL > 15. 2. TSE/AGE/DUR model: RR ¼ 1 þ bðw5--14 þ q2 w15--24 þ q3 w25þ Þðfage Þðg DUR Þ; where b ¼ 0.0611, 22 ¼ 0.81, 23 ¼ 0.40; w5–14 and so on as above; fage ¼ 1 for age <55, 0.57 for age 55–65, 0.34 for age 65–75, and 0.28, age >75; and g DUR ¼ 1.0 for DUR < 0.5, 3.17 for 3.17 < DUR < 15, 5.27 for 5.27 < DUR < 25, 9.08 for 9.08 < DUR < 35, and 13.6 for DUR > 35. The combined effect of smoking and 222 Rn exposure could not be determined quantitatively. The pooled analysis showed a linear increase in risk of about a factor of about 3 for smokers versus never-smokers.

CHILDHOOD EXPOSURE

1113

The NIH report summarized the calculated deaths in the U.S. population from the assumed exposure of 46 Bq/m3 (1.25 pCi/L). Their calculated value was 15,000 deaths per year with 10,000 deaths in smokers and 5000 in nonsmokers. The BEIR VI report changed these values slightly. 29.8.4

BEIR VI

The National Research Council revised and updated the BEIR IV report published in 1988. The BEIR VI report was published in 1999 (NAS/NRC, 1999a) and had longer term followup in some cases. The model to project lung cancer from 222 Rn decay product exposure is essentially identical to that reported in NIH/NCI (1994). The BEIR VI committee analyzed the same 11 underground mining cohort data with slight modification, and updating of the data, and so the models produce similar results. The BEIRVI committee decided that the risk at the lower mining exposures could be used to estimate environmental risk. At very high exposures, the risk decreases per unit exposure as discussed earlier but residential exposures rarely if ever attain these high values. Two types of models were used similar to those in the NIH/NCI (1994) report. One model was an exposure/age/duration model and the other an exposure/age/concentration model. The two models predict 15,400 and 21,800 lung cancer deaths per year, respectively, in the United States from 222 Rn exposure in the homes. An uncertainty analysis suggested the number of calculated cases could range from 3000 to 33,000. The lifetime relative risk (LRR) estimated for residential exposure by the BEIR VI Committee is described in section 29.11 on Radon and Smoking. The long-term follow-up of the mining populations was strongly recommended by BEIR VI to more precisely assess the magnitude of lung cancer due to radon.

29.9 CHILDHOOD EXPOSURE It is apparent from the studies of A-bomb survivors that leukemia and breast cancer are elevated for radiation exposure in childhood versus exposure in adulthood. Concern has been expressed that the same might be true for lung cancer derived from exposure to radon and its decay products at early ages. There is a suggestion in the literature that the alpha dose per unit exposure is somewhat larger at about 10 years of age because of the smaller dimensions of the child’s bronchial tree with breathing rates similar to those of an adult (Harley, 1984a; NCRP, 1984; NAS/NRC, 1991). Some human data are available for occupational exposure to elevated radon in childhood. Lubin et al. (1990) analyzed data from tin mines in the Yunnan province, China. Thirty-seven percent of exposed workers started employment under the age of 13. For this group, the risk coefficient was 1.2%/WLM, while for those first employed above age 18, the risk coefficient was 2.9%/WLM. Although this difference was not statistically significant, the results suggest that children are not a particularly sensitive population. The studies of adult underground miners show a significant reduction in risk with time since exposure. Tentatively, the lower risk coefficient for children compared to that for adults reported by Lubin et al. (1990) suggests that the reduction in lung cancer risk with time since exposure is also effective for short-term childhood exposure. The fact that lung cancer does not appear at a significant rate in any population before the age of 40 permits a substantial time interval for risk reduction in an exposed child.

1114

RADON AND LUNG CANCER

29.10 ANIMAL STUDIES The animal studies have been supportive of the human epidemiology and they are considered briefly. Studies with SPF Wistar rats at Battelle Pacific Northwest Laboratories and with SPF Sprague–Dawley rats at COGEMA laboratories in France have produced lung tumors at total exposures as low as 200 and 20–50 WLM, respectively (NCRP, 1984a, 1989). Gilbert et al. (1996) and Monchaux et al. (1999) reported the lung cancer in rats exposed to cumulative exposures as low as 25–100 WLM was comparable to that in humans. Monchaux et al. (1999) reported a decreasing lung cancer risk with decreasing exposure rate. Detailed tracheobronchial dosimetry in the rat lung shows that the alpha dose per unit exposure in the rat and the humans is similar (Harley, 1988). The dose conversion factor for both is about 6–9 mGy/WLM depending upon the model chosen. However, the distribution of tumors differs in that the rat has a predominance of tumors that are in the distal airways, essentially in the pulmonary parenchyma, while in humans the tumors appear in the upper or proximal airways. The reason for this is not known; however, there are significant differences between cell types and cell turnover rates in the rat and human bronchial airways. In the distal airways of the rat, Clara cells predominate and there are no basal cells. In the humans, basal cells persist down to the distal airways (bronchioles). Human tumors occur in the distal airways but are preferentially found in the most proximal airways, airway branching generations 1 to about 10 (Saccomanno et al., 1996). Cellular turnover rates most likely play a key role. One biological model of 222 Rn related lung cancer showed that the reduced number of stem cells with age may be responsible for the time since exposure decrease in lung cancer observed in the miners (Harley et al., 1996b).

29.11 SMOKING AND RADON One of the most critical issues concerning lung cancer and radon exposure concerns the effect of cigarette smoking on lung cancer. Both tobacco and radon are carcinogens but tobacco far outweighs radon as a lung carcinogen. Underground miners were mostly smokers in all the cohorts studied. Smoking miners, exposed to high radon concentrations, show a significant excess of lung cancer above that expected due to smoking alone. Regardless of the type of mine exposure, the increased or excess lung cancer is linearly related to total or cumulative radon exposure (NIH/NCI 1994; NAS/NRC, 1999a). This is convincing evidence that radon decay products are a carcinogen even if there is the potential influence of other inhaled mine agents such as silica, ore dust, arsenic, and so on. There are some nonsmokers in the exposed mining and residential populations studied (Radford and Renard, 1984; Samet et al., 1984; Roscoe et al., 1989, 1995; Sevc et al., 1988; Neuberger and Gesell, 2002). Both the residential and miner studies of nonsmokers show increased risk of lung cancer with radon exposure. The model parameters used in the underground miners’ studies for risk projection either for miners or residential exposure are critical. The NAS BEIR IV, NIH/NCI (1994), and BEIRVI reports all state that the data were better fit by a relative or multiplicative risk model than an absolute additive model, but that a relative risk model between additive and multiplicative (submultiplicative) provides the best fit. Although the miner studies show an OR greater for smokers, the residential studies show no difference. If the increase in lung cancer per unit exposure is the same for smokers and nonsmokers, and a relative risk model applies, then the number of lung cancers in a group of smokers will be much greater per unit

SUMMARY

1115

exposure than for the same size group of nonsmokers because of the large difference in baseline cancer mortality. The calculation of LRR derived from the NAS/NRC (1999) or NIH/NCI (1994) model parameters can differ substantially depending upon smoking status. Estimated LRR of lung-cancer risk associated with lifetime indoor exposure to radon BEIR VI submultiplicative model (from NAS/NRC, 1999a, p. 87). Lubin and Boice (1989), using the NAS/NRC BEIR IV model, estimated that 13,000 deaths per year occur in the United States from environmental radon exposure, and that these are apportioned 9000 smokers versus 4000 nonsmokers. The NAS/NRC BEIR VI model increased this to 15,400–21,000 deaths per year, with 2100–2900 estimated for neversmokers. The NCRP model, on the other hand, with preliminary data, could not differentiate between smokers and nonsmokers and estimated that 9000 radon-related deaths per year could occur. The best estimate of radon-related lung cancer risk in homes is now documented in the pooled and meta-analyses described in Section 29.6. An OR of 1.16 for a lifetime residential exposure to 100 Bq/m3 implies about 10,000 annual radon-related lung cancer deaths in the United States for the average exposure of 40 Bq/m3. A comparison of the residential studies OR of 1.16 with the above table shows the BEIR VI estimates of LRR from the projection models are very conservative. Nonetheless, the models based on the underground miner data are important to understand the factors affecting lung cancer appearance over time. As an example, the best estimate of lifetime risk (RR) of lung cancer from residential exposure to 100 Bq/m3 is 1.16 for both smokers and nonsmokers (Fig. 29.9). This means that there is a 16% increase over the normal lung cancer risk. BEIR VI (NAS/NRC, 1999a) estimates male smokers’ and nonsmokers’ baseline lifetime lung cancer risk as 12% and 0.9%, respectively, and female smokers’ and nonsmokers’ risk as 6.8% and 0.6%, respectively. A 16% increase in these rates for both means that the lifetime risk is 2% ( ¼ 0.16  0.12) and 0.1% ( ¼ 0.16  0.9) for male smokers and nonsmokers, respectively, and 1.0% and 0.1% for female smokers and nonsmokers, respectively. For the average radon concentration in a U.S. home, 46 Bq/m3, the lifetime risk would be about one-half of these values.

29.12 SUMMARY There is a large body of evidence demonstrating that 222 Rn and its short-lived decay products are the carcinogens responsible for lung cancer in underground miners in excess of that expected due to smoking. The miner studies with documented exposure and mortality data show that for exposures as low as 50 WLM, excess lung cancer is observed. This exposure is approximately the same as lifelong exposure in a home with a 222 Rn concentration at the current EPA guideline of 4 pCi/L. The only difference between themining and home exposures is that in underground mines the exposure was generally for a short period of time (<20 years), whereas in the home it is lifelong exposure to perhaps accumulate the same total dose. Studies of the average 222 Rn exposure in homes have been conducted in 50 countries. The average radon concentration ranges from 7 to 140 Bq/m3 (0.2–4 pCi/L). The U.S. average based on an EPA survey is 46 Bq/m3. This leads to an average annual dose of 1.6 mSv or more than half of the total background radiation dose to perhaps the population. There are many dosimetric models to calculate the alpha dose to the target cells in the bronchial airways, the site of most radon-related human lung cancer. The dose to the target

1116

RADON AND LUNG CANCER

cells is the same per unit exposure in both homes and mines, so that risk can be transported from mines to homes with proper adjustment of factors such as smoking prevalence. There are now 80 published epidemiologic studies of residential radon exposure and lung cancer. Twenty-three are case–control studies. The case–control studies have either been pooled or a meta-analysis was performed to yield a best estimate for the OR of 1.16 (95% CI: 1.06–1.28) for residential lung cancer risk for lifetime exposure to 100 Bq/m3 (2.5 pCi/L). The risk projections for residential exposure based on the miner studies are conservative, and overestimate risk, but the models are useful to understand the factors affecting risk. It is thus reasonable to believe that residential exposure to 222 Rn should be reduced, especially in homes with high concentrations. The epidemiologic studies for smoking miners show greater risk of lung cancer than nonsmoking miners given the same total exposure. However, the same odds ratio or relative risk for smokers and nonsmokers is seen in the residential studies. Because the baseline risk for smokers is much greater than that for nonsmokers, the calculated lung cancer risk for smokers is still much higher than that for nonsmokers. Stopping tobacco use is strongly encouraged. Good construction practices in new homes, that is, radon-resistant techniques, and the identification and remediation of existing homes with high radon concentration should reduce population risk. The other radon (thoron) is an emerging problem. Radon-222 is a decay product in the naturally occurring 238 U series. The radioactive 232 Th series also has a gaseous isotope of radon, 220 Rn, thoron. The half-life of 220 Rn is short (55 s), and the equilibrium factor is small (0.02), so concentrations of thoron and its decay products are usually lower than 222 Rn. The lung dose from thoron per unit exposure is small compared to that from 222 Rn. Demonstrable health effects have not yet been shown for thoron. Many measurements made for radon gas also include thoron gas. Because the dose is different for the two gases, some of the risk estimates based on gas measurements are somewhat in error and the magnitude is not known at this time. Thoron ð220 RnÞ dosimetry, measurement, and risk are ongoing research priorities.

REFERENCES Chambers DB, Lowe LM, Reilly PM, Duport P (1990) Effects of exposure uncertainty on estimation of radon risks. 29th Hanford Symposium on Health and the Environment—Indoor Radon and Lung Cancer: Reality or Myth. Columbus, OH: Battelle Press. Chittaporn P, Harley NH (1994) Water use contribution to indoor 222 Rn: Health Phys. S29. Cohen BL (1992) Compilation and integration of studies of radon levels in U.S. homes by states and counties. Crit. Rev. Environ. Control 22:243–364. Cohen BS, Asgharian B (1990) Deposition of ultrafine particles in the upper airways: an empirical analysis. J. Aerosol Sci. V 21:789. Darby S, Hill D, Auvinen A, Barros-dios JM, Baysson H, Bochicchio F, Deo H, Falk R, Forastiere F, Hakama M, Heid I, Kreienbrock L, Kreuzer M, Lagarde F, Makelainen I, Muirhead C, Oberaigner W, Pershagen G, Ruano-ravina A, Ruosteenoja E, Rosario AS, Tirmarche M, Tomasek L, Whitley E, Wichmann HE, Doll R (2005) Radon in homes and risk of lung cancer: collaborative analysis of individual data from 13 European case–control studies. Br. Med. J. 330:23. DOE/CEC (1989) International Workshop on Residential Radon Epidemiology CONF-8907178. Springfield, VA: NTIS. U.S.EPA (1992) National Radon Survey Summary Report. U.S. Environmental Protection Agency. EPA 402-R-92-011.

REFERENCES

1117

EPA (2003) EPA assessment of risks from radon in homes. EPA 402-R-03-003 (June) Office of Radiation Protection and Indoor Air, United States Environmental Protection Agency, Washington, DC. Field RW, Steck DJ, Smith BJ, Brus CP, Fisher EL, Neuberger JS, Platz CE, Robinson RA, Woolson RF, Lynch CF (2000) Residential radon gas exposure and lung cancer. Am. J. Epidemiol. 151:1091–1102. Fisenne IM (1988) Radon-222 measurements at Chester, NJ through July 1986. 1985–1986 Biennial Report of the EML Regional Baseline Station at Chester, NJ. Environmental Measurements Laboratory Report EML-504. New York: EML. Fisenne IM, Keller HW (1996) Continuous indoor and outdoor measurements of 222 Rn in New York City: city as a source. Environ. Int. 22:S131–S138. Fisenne IM, Machta L, Harley NH (2005) Stratospheric radon measurements in three North American locations. In: McLaughlin JP, Simopoulos SE, Steinhausler F, editors. The Natural Radiation Environment VII. New York: Elsevier. George AC, Breslin AJ (1980) The distribution of ambient radon and radon daughters in residential buildings in the New Jersey–New York area. In: Gesell TF, Lowder WM, editors. Natural Radiation Environment III CONF-780422. Washington: U.S. DOE. George AC, Hinchliffe L (1972) Measurements of uncombined radon daughters in uranium mines. Health Phys. 23:791–803. Gilbert ES, Cross FT, Dagle GE (1996) Analysis of lung tumor risks in rats exposed to radon. Radiat. Res. 145:350–360. Glauberman H, Breslin AJ (1957) Environmental radon concentrations. U.S. Atomic Energy Commission Health and Safety Laboratory Report NYO-4861. New York: HASL. Harley JH (1974) Environmental radon. In: Stanley RE, Moghissi AA, editors. Noble Gases CONF730915. Washington: U.S. EPA. Harley NH (1984) Comparing radon daughter dose: environmental versus underground exposure. Radiat. Prot. Dosim. 7:371. Harley NH (1988) Radon daughter dosimetry in the rat tracheobronchial tree. Radiat. Prot. Dosim. 24:457. Harley JH (1990) Radon is out. Presented at the 29th Hanford Symposium on Health and the Environment: Indoor Radon and Lung Cancer—Reality or Myth. PL Columbus, OH: Battelle Press. Harley NH (1991) Radon levels in a high-rise apartment. Health Physics 61:263–265. Harley NH, Chittaparn P, Sylvester J, Roman M (1991) Personal and home 222Rn and gamma-ray exposure measured in 52 dwellings. Health Phys. 61:737–744. Harley NH, Chittaporn P (1993) Modeled versus measured indoor 222Rn. Health Physics 64: S20. Harley NH (1997) Outdoor radon measurements at two New Jersey sites. Private communication. Harley NH, Robbins ES (1994) A biokinetic model for 222 Rn gas distribution and alpha dose in humans following ingestion. Environ. Int. 20:605–610. Harley NH, Chittaporn P, Heikkinen MSA, Medora R, Merrill R (2005a) Airborne particle size distribution measurements at USDOE Fernald. In: Berkey E, Zachry T, editors. American Chemical Society Monograph ACS Symposium Series 904, pp. 342–350. Harley NH, Chittaporn P, Merrill P, Medora R, Wanitsooksumbut W (2005b) Thoron versus radon: measurement and dosimetry. In: Sugahara T, editors. High Levels of Natural Radiation and Radon Areas: Radiation Dose and Health Effects. International Congress Series 1276. Elsevier, pp. 72–75. Harley NH, Cohen BS, Robbins ES (1996a) The variability in radon decay product bronchial dose. Environ. Int. 22:S959–S964. Harley NH, Meyers OA, Chittaporn P, Robbins ES (1996b) A biological model for lung cancer risk from 222 Rn exposure. Environ. Int. 22:S977–S989.

1118

RADON AND LUNG CANCER

Hess VF (1953) Radon and thoron and their decay products in the atmosphere. J. Atmos.Terrestr. Phys. 3: 172. Hess CT, Korsah JK, Einloth CJ (1985) Radon in houses due to radon in potable water. In: Hopke P, editor. Radon and Its Decay Products. ACS Symposium, Washington, DC. Hess CT, Vietti MA, Mager DT (1987) Radon from drinking water: evaluation of waterborne transfer into house air. Environ. Geochem. Health 8:68. Hopke PK, Jensen B, Li CS, Montassier N, Wasoliek P, Cavallo A, Gatsby K, Socolow R, James AC (1995) Assessment of the exposure to and dose from radon and decay products in normally occupied homes. Environ. Sci. Technol. 29:1359–1364. Hultqvist B (1956) Studies on naturally occurring ionizing radiations. Thesis (in English), Kgl Svenska Vetenkaps Handl. #3, Series 4. ICRP (1987) Lung Cancer Risk from Indoor Exposures to Radon Daughters. International Commission on Radiological Protection Publication 50. Oxford: Pergamon Press. Krewski D, Lubin Jay H, Zielinski JM, Alavanja M, Catalan VS, Field RW, Klotz JB, Letourneau EG, Lynch CF, Lyon JI, Sandler DP, Schoenberg JB, Steck DJ, Stolwijk JA, Weinberg CE, Wilcox H, Omer B (2005) Residential radon and risk of lung cancer: a combined analysis of 7 North American case–control studies. Epidemiology 16:137–145. Lubin JH (2003) Studies of radon and lung cancer in North America and China. Radiat. Prot. Dosim. 104:315–319. Lubin JH, Boice JD (1989) Estimating Rn-induced lung cancer in the United States. Health Phys. 57:417. Lubin JH, Boice JD (1997) Lung cancer risk from residential radon meta-analysis of eight epidemiologic studies. J. Natl. Cancer Inst. 89:49–57. Lubin JH, You-Lin Q, Taylor PR, Shu-Xiang Y, Schatzkin A, Bao-Lin M, Jian-Yu R, Xiang-Zhen X (1990) Quantitative evaluation of the radon and lung cancer association in a case control study of Chinese tin miners. Cancer Res.V 50:174. Monchaux G, Morlier JP, Altmeyer S, Debroche M, Morin M (1999) Influence of exposure rate on lung cancer induction in rats exposed to radon progeny. Radiat. Res. 152:S137–S140. Moses H, Lucas HF, Zerbe GA (1963) T The effect of meteorological variables upon radon concentration three feet above the ground. J. Air Pollut. Control Assoc. 13:12. Muller J, Wheeler WC, Gentleman JF, Suranyi G, Kusiak RA (1983) Study of Mortality of Ontario Miners. Toronto: Ministry of Labour. NAS/NRC (1988) National Academy of Sciences/National Research Council Health Risks of Radon and Other Internally Deposited Alpha Emitters, Committee on Biological Effects of Ionizing Radiation, BEIR IV. Washington: National Academy Press. NAS/NRC (National Academy of Sciences/National Research Council) (1991) Comparative Dosimetry of Radon in Mines and Homes. Washington, DC: National Academy Press. NAS/NRC (1999a) Health Effects of Exposure to Radon. Committee on Health Risks of Exposure to Radon, Board on Radiation Effects Research Commission on Life Sciences. (National Research Council) BEIR VI. Washington, DC: National Academy Press. NAS/NRC (1999b) National Academy of Sciences/National Research Council Risk Assessment of Radon in Drinking Water (National Academy Press, Washington, DC). NCRP (1984a) Evaluation of occupational and environmental exposures to radon and radon daughters in the United States National Council on Radiation Protection and Measurements Report No 78. Bethesda, MD: NCRP. NCRP (1984b) Exposures From the Uranium Series with Emphasis on Radon and its Daughters National Council on Radiation Protection and Measurements Report No 77. Bethesda, MD: NCRP.

REFERENCES

1119

NCRP (1987a) Exposure to the Population in the United States and Canada from Natural Background Radiation National Council on Radiation Protection and Measurements Report No 94. Bethesda, MD: NCRP. NCRP (1987b) Ionizing Radiation Exposure of the Population of the United States National Council on Radiation Protection and Measurements Report No 93. Bethesda, MD: NCRP. NCRP (1988) Measurement of Radon and Radon Daughters in Air National Council on Radiation Protection and Measurements Report No 97. Bethesda, MD:NCRP. NCRP (1989) Radon Proceedings of the Twenty-Fourth Annual Meeting of the National Council on Radiation Protection and Measurements. Bethesda, MD: NCRP. Nero AV (1983) Indoor radiation exposures from Rn-222 and its daughters: a view of the issue. Health Phys. 45:277. Nero AV, Schwehr MB, Nazaroff WW, Revzan KL (1986) Distribution of airborne radon-222 concentrations in U.S. homes. Science 234:992–996. Neuberger JS, Gesell TF (2002) Residential radon exposure and lung cancer: risk in nonsmokers. Health Phys. 83:1–18. Neuberger JS, Harley NH, Kross BC (1996) Residential radon exposure and lung cancer potential for pooled or meta-analysis. J. Clean Technol. Environ. Occup. Med. 5:207–221. NIH/NCI 1994, Lubin J, Boice JD, Hornung RW, Edling C, Howe GR, Kunz E, Kusiak RA, Morrison HI, Radford EP, Samet JM, Tirmarche M, Woodward A, Xiang YS, Pierce DA (1994) Radon and Lung Cancer Risk: A Joint Analysis of 11 Underground Miner Studies NIH 94-3644. Washington, DC: National Institutes of Health, National Cancer Institute. Pavia M, Bianco A, Pileggi C, Angelillio I (2003) Meta-analysis of residential exposure to radon gas and lung cancer. Bull. World Health Organ. 81:732–738. Pershagen G, Ackerblom G, Axelson O, Clavensjo B, Damber L, Desai G, Enflo A, LaGarde F, Mellander H, Svartengren M, Swedjemark GA (1994) Residential radon exposure and lung cancer in Sweden. N. Eng. J. Med. 330:159–164. Put LW, deMeijer RJ (1988) Variation of time-averaged indoor and outdoor radon concentrations with time, location and sampling height. Rad. Prot. Dosimetry V 24:317. Radford EP, Renard KGS (1984) Lung cancer in Swedish iron miners exposed to low doses of radon daughters. N. Eng. J. Med. 310:1485–1494. Reilly MA (1990)The index house: Pennsylvania radon research and demonstration project, Pottstown PA 1986–1988. In: Majumdar SK, Schmaltz RF, Miller EW, editors. Environmental Radon Occurrence and Control. The Pennsylvania Academy of Sciences. Reineking A, Butterweck G, Kesten J, Porstendorfer J (1990) Unattached Fraction and Size Distribution of Aerosol-Attached Radon and Thoron Daughters in Realistic Living Atmospheres and Their Influence On Radiation Dose. Presented at the 29th Hanford Symposium on Health and the Environment: Indoor Radon and Lung Cancer: Reality or Myth. Reineking A, Porstendorfer J (1986) High-volume screen diffusion batteries and alpha-spectroscopy for measurement of the radon daughter activity size distributions in the environment. J. Aerosol Sci.873. Robbins ES (1990) Cellular morphometry in human bronchial epithelium. Private Communication. Robbins ES, Meyers OA (1995) Cycling cells of human and dog tracheobronchial mucosa: normal and repairing epithelium. Technol. J. Franklin Inst. 332 A:35–42. Roscoe RJ, Steenland K, Halperin WE, Beaumont JJ, Waxweiler RJ (1989) Lung cancer mortality among nonsmoking uranium miners exposed to Radon. JAMA V 262:629. Roscoe RJ, Deddens JA, Salvan A, Schnorr TM (1995) Mortality among Navajo uranium miners. Am. J. Public Health 85:535–540. Rundo J, Markun F, Plondke NJ (1979) Observation of high concentrations of radon in certain houses. Health Phys. 36:729.

1120

RADON AND LUNG CANCER

Ruzer LS, Nero AV, Harley NH (1995) Assessment of lung deposition and breathing rate in underground miners in Tadjikistan. Radiat. Prot. Dosimetry 58:261–268. Sachs HM, Hernandez TL, Ring JW (1982) Regional geology and radon variability in buildings. Environ. Int. 8: 97-103. Saccomanno G, Auerbach O, Kuschner M, Harley NH, Michaels RY, Anderson MW, Bechtel JJ (1996) A comparison between the localization of lung tumors in uranium miners and in nonminers from 1947 to 1991. Cancer 77:1278–1283. Samet JM, Kutvirt OM, Waxweiler RJ, Key CR (1984) Uranium mining and lung cancer in Navaho menN. Eng. J. Med.1481–1484. Samuelsson C (1988) Retrospective determination of radon in houses. Nature. 334:338–340. Samuelsson C (1996) Plateout and implantation of 222 Rn decay products in dwellingsEnviron. Int. 22 (Suppl):S839–S843. Senes Consultants Limited (2006) Final Report Eldorado nuclear epidemiology study update Eldorado uranium miners’ cohort part I of the saskatchewan uranium miners’ cohort study. Schoenberg JB, Klotz JB, Wilcox HB, Nicholls GP, Gil-del-Real MT, Stemhagen A, Mason J (1990) Case–control study of residential radon and lung cancer among New Jersey women Cancer Res. 50:6520. Servant J (1966) Temporal and spatial variations of the concentration of the short-lived decay products of radon in the lower atmosphere. Tellus 18:663. Sevc J, Kunz E, Tomasek L, Placek V, Horacek J (1988) Cancer in man after exposure to Rn daughters. Health Phys. 54:27–46. Steinhausler F (1975) Long-term measurements of 222Rn, 220Rn, 214Pb and 212Pb concentrations in the air of private and public buildings and their dependence on meteorological parameters. Health Phys. 29:705–713. Tu KW, Knutson EO (1988) Indoor outdoor aerosol measurements for two residential buildings in New Jersey. Aerosol Sci. Tech. 9:71. UNSCEAR (2000) Sources and Effects of Ionizing: United Nations Scientific Committee on the Effects of Atomic Radiation. Vol. I Sources. UNSCEAR (2008) Sources-to-Effects Assessment for Radon in Homes and Workplaces: United Nations Scientific Committee on the Effects of Atomic Radiation. New York: United Nations. Walsh C, McLaughlin JP (2001) Correlation of 210Po implanted in glass with radon gas exposure: sensitivity analysis of critical parameters using a Monte-Carlo approach. Sci. Total Environ. 272:195–202. Woodward A, Roder D, McMichael AJ, Crouch P, Mylvaganam A (1991) Radon daughter exposures at the Radium Hill uranium mine and lung cancer rates among former workers 1952–1987. Cancer Causes & Control 2:213–220. Wright JR, Smith OF (1915) The variation with meteorological conditions of the amount of radon emanation in the atmosphere, in the soil gas and in the air exhaled from the surface of the ground in Manila. Phys. Rev. 5:459. Yamada Y, Tokonami S, Zhou W, Yonehara H, Ishikawa T, Furukawa M, Fukutsu K, Sun Q, Hou C, Zhang S, Akiba S (2005) Rn–Tn discriminative measurements and their dose estimates in Chinese Loess plateau. High Levels of Natural Radiation and Radon Areas: Radiation Dose and Health Effects. International Congress Series 1276. San Diego, CA: Elsevier.

30 ULTRAVIOLET RADIATION Nigel Cridland and Colin Driscoll

30.1 INTRODUCTION Ultraviolet radiation (UVR) may be defined as that part of the electromagnetic spectrum (Fig. 30.1) with wavelengths between 100 nm (corresponding to a photon energy of approximately 12 eV) and 400 nm. Photon energies at the short wavelength end of this range are sufficient to cause ionization of many atoms but not of those that are the principal components of biological materials (hydrogen, carbon, oxygen, and nitrogen). Consequently, for the purposes of practical health protection, the short wavelength limit of the UVR region is generallytakenastheboundarybetween theionizing (wavelengths <100 nm)andnonionizing regions of the electromagnetic spectrum. In practice, the actual boundary is somewhat academic as wavelengths less than 180 nm are strongly absorbed in air and other common materials so that environmental exposures are unlikely to occur. Although penetration of UVR into biological tissues increases with wavelength, it is limited across the entire ultraviolet region of the spectrum and consequently the principal interactions occur in the surface tissues of the body. In discussing the resulting adverse health effects, it is usually convenient to subdivide the UV spectrum into three spectral regions (CIE, 1970), termed UVA (315– 400 nm), UVB (280–315 nm), and UVC (100–280 nm). A summary of quantities and units used to describe UVR is given in Table 30.1. The sun is the principal environmental source of UVR for most people (AGNIR, 2002), and exposure depends on many factors including geographical location, altitude, time of the day, season, and prevailing weather conditions, particularly cloud cover. Moreover, it has been predicted that depletion of the stratospheric ozone layer, which absorbs strongly at shorter wavelengths, will result in increased exposure. Nevertheless, for most people personal behavior is likely to be a major factor in determining exposure. For some people, artificial sources could also contribute significantly to their total UVR exposure. Exposure to some sources, especially sunbeds, may be elective, but for many, including welding arcs and high intensity lamps used in the workplace, exposure is likely to

Environmental Toxicants, Third Edition Edited by Morton Lippmann Copyright
2009 John Wiley & Sons, Inc.

1121

1122

ULTRAVIOLET RADIATION

TABLE 30.1

Radiometric Quantities and Units

Quantity

Unit

Radiant energy Radiant flux Irradiance Effective irradiance Radiant exposure Effective radiant exposure

FIGURE 30.1

Joule (J) Watt (W) Watt per square meter (W/m2) Watt per square meter effective (W=m2eff ) Joule per square meter (J/m 2) Joule per square meter effective (J=m2eff )

Electromagnetic spectrum.

be incidental to another activity and is therefore susceptible to management through implementation of appropriate controls. In addition, some patients may receive controlled exposures from artificial sources as part of their medical treatment. After introducing some of the key parameters that define UV exposures, this review summarizes the output characteristics of a variety of sources and then discusses the evidence for short-term and long-term adverse health effects. The main populations at special risk are identified. Current recommendations on restricting exposure are discussed in the context of practical protection measures, and finally the principal approaches to quantifying exposures are summarized.

30.2 PATHWAYS FOR HUMAN EXPOSURE As a consequence of its low penetration, the principal interactions of UVR occur in the surface tissues of the body. In practice, this means that the two organs most susceptible to damage are the skin and the eye. The human eye is a roughly spherical organ, deeply set in a bony orbital cavity. The bony ridge above the eyes, which is most pronounced in males, provides protection from both mechanical injury and overhead sunlight. Hence, ocular damage resulting from solar UVR is mainly associated with high levels of reflected solar radiation. The effectiveness of these protective arrangements may be reduced or eliminated during exposure to artificial sources positioned in the immediate field of view, or during deliberate viewing of the sun or sky. Skin damage usually occurs as a result of direct exposure to the source in question, regardless of whether it is the sun or an artificial source. Clothing, which generally provides a high degree of protection, may significantly modify exposure of skin. Thus, parts of the body

PATHWAYS FOR HUMAN EXPOSURE

1123

that are not normally clothed, such as face and hands, tend to receive higher exposures and in chronic exposure situations may accumulate much more damage. There may, however, be an important exception to this, as malignant melanoma appears to be associated with intermittent intense exposure of normally unexposed skin. 30.2.1

Characterizing Exposures

The typical emissions from various sources are discussed in more detail in the next section. As will be seen below, the biological effectiveness of incident UVR strongly depends on the spectral distribution of the radiated power. Consequently, assessment of the hazard presented by a particular source requires information on the spectral emission. Spectral irradiance (measured in W/m2 nm) at the exposure position may be measured, or less frequently calculated, and could be used to obtain the total irradiance (W/m2) by summing over all the emitted wavelengths. However, it is more useful to use it to determine the effective irradiance (W=m2effective ), which is derived by multiplying the spectral irradiance at each wavelength by a hazard weighting factor (which quantifies the relative efficacy at each wavelength for causing the effect) and summing over all wavelengths. This simple summation assumes that there are no synergistic interactions between the spectral components. The hazard weighting factors are obtained from action spectra, which are graphs of the reciprocal of the radiant exposure required to produce effects at each wavelength; all the data are normalized to the datum at the most efficacious wavelength(s). The action spectra for threshold erythema (McKinlay and Diffey, 1987) (Fig. 30.2) and photokeratitis have been used to derive hazard weighting factors in the ultraviolet region of the spectrum. Although many of the data on which these action spectra are based are fairly old and techniques have

FIGURE 30.2

Reference erythmal action spectrum (adapted from McKinlay and Diffey, 1987).

1124

ULTRAVIOLET RADIATION

TABLE 30.2 Skin Type

Information on Human Skin Type

Unexposed Skin Color

UVR Sensitivity

Minimum Erythemal UVR Dose (J=m2eff )

I

White

Very sensitive

150–300

II

White

Very sensitive

250–350

III

White

Sensitive

300–500

IV

Light brown

Moderately sensitive

450–600

V

Brown

600–1000

VI

Chocolate brown, black

Minimally sensitive Insensitive

1000–2000

Sunburn/tanning history Always burns easily, never tans Always burns easily, tans minimally Burns moderately, tans gradually (light brown) Burns minimally, tans well always (moderate brown) Rarely burns, tans profusely (dark brown) Never burns, deeply pigmented (black)

Source: Adapted from Pathak and Fanselow (1983).

moved on somewhat, a recent reevaluation of the data showed that the hazard functions remain valid (Chaney and Sliney, 2005). As discussed below, most of the known effects of UVR result from reversible photochemical changes in biochemical constituents of tissue and are characteristically cumulative with time in the short term. It is therefore useful to consider the effective radiant exposure (J=m2effective ) occurring in a given exposure period (usually taken as an 8 h day). This is obtained by summing the effective irradiance over the exposure period. Individuals show considerable variation in their susceptibility to erythema induction by UVR. In an attempt to try and allow for this in describing exposures, particularly in a medical context, the concept of a minimal erythemal dose (MED) was introduced. This is the radiant exposure of UVR that produces a just perceptible erythema in previously unexposed skin. It corresponds to a radiant exposure of monochromatic radiation at the maximum spectral efficacy for erythema (around 300 nm) of approximately 150–2000 J=m2effective , depending on skin type (Table 30.2). An effective radiant exposure of 200 J/m2 is often used as the value of 1 MED for white skin. This is twice the value of the standard erythema dose (SED) recommended by the CIE (CIE, 1996).

30.3 SOURCES OF ULTRAVIOLET RADIATION 30.3.1

The Sun

For most people, the main source of exposure to UVR is from the sun, and while many factors influence exposure, individual behavior is probably the most important with respect to personal risk. The broad spectrum and the intensity of UVR from the sun are due to the high temperature at its surface and its size. The sun approximates to a “blackbody” emitter with a temperature

SOURCES OF ULTRAVIOLET RADIATION

1125

TABLE 30.3 Spectral Distribution of Solar Radiation Prior to Attenuation by the Earth’s Atmosphere Wavelength Band UVC UVB UVA Total UV Visible and IR

Irradiance (W/m2) 6.4 21.1 85.7 113.2 1254

Percentage of Total 0.5 1.5 6.3 8.3 91.7

Source: Frederick et al. (1989).

of about 5900 K. The solar radiation reaching the top of the Earth’s atmosphere is affected by the solar output and the Earth–sun distance. Variations in solar output are much smaller than the variations caused by atmospheric attenuation factors. The mean solar irradiance just outside the Earth’s atmosphere is approximately 1370 W/m2 (the so-called “solar constant”). The spectral distribution is summarized in Table 30.3, but will vary with the exact distance of the Earth from the sun at a particular time. The extreme values associated with this variation are approximately 3.3% above and below the annual mean and occur in January and July, respectively. Solar UVR undergoes absorption and scattering as it passes through the Earth’s atmosphere, with absorption by molecular oxygen and ozone being the most important processes (Fig. 30.3). The ozone layer prevents almost all UVR of wavelengths <290 nm and a substantial fraction (in excess of 90% of the total energy) from 290 to 315 nm from reaching the Earth’s surface. Thus, the terrestrial environment is exposed to UVR between 290 and 400 nm. The total solar UVR reaching the Earth’s surface, termed global UVR, can be divided into two components: direct and diffuse. Global UVR reaching a horizontal surface is the quantity most often measured. On average, this comprises about 98% UVA and 2% UVB. However, the exact amount and spectral distribution of solar UVR irradiance reaching the Earth’s surface depends on a number of factors (WHO, 1994), including . . . . . . . . . .

Wavelength of the UVR Solar zenith angle, which depends on latitude, date of the year, and time of day Solar source spectrum incident at the top of the atmosphere Ozone column thickness and vertical distribution Molecular absorption and scattering (including localized gaseous pollutants) Aerosol absorption and scattering (including anthropogenic aerosols) Absorption, scattering, and reflection by clouds Reflectance characteristics (albedo) of the ground Shadowing by surrounding objects Altitude above sea level.

In January (in the Northern Hemisphere) or July (in the Southern Hemisphere) when the solar elevation is low, direct UVR travels a longer path through the atmosphere and a large amount of scattering occurs. In addition, much of the resultant scattered UVR propagates downward to the Earth’s surface at angles to the horizontal that are larger than

1126

ULTRAVIOLET RADIATION

FIGURE 30.3

Solar UVR attenuation by the earth’s atmosphere (adapted from UNEP, 1987).

the solar elevation, hence traveling a shorter and less absorptive path. This results in large ratios of scattered to direct UVR. During the summer, the ratio of diffuse to direct UVR is smaller. The maximum erythemally effective UVB irradiances are shown in Table 30.4 for the Northern Hemisphere as a function of latitude and time of year at sea level (Driscoll, 1992). There are strong seasonal and latitudinal variations in UVB. Under cloudless skies, the UVB is more intense in summer and at all times of year is greater at lower latitudes. Approximately 50% of the daily UVR is received during the middle 4 h around noon when the sun is high in the sky (Sliney, 1987). The number of MEDs in a 3 h exposure period centered around 12 GMT is shown for the Northern Hemisphere as a function of latitude and time of year in Table 30.5. Data for the Southern Hemisphere can be derived from these tables by advancing the data sets by 6 months so that June’s data becomes those of December and vice versa. The presence of cloud cover, air pollution, haze, or even scattered clouds plays a significant role in attenuating solar UVR. UVB and UVA irradiances are reduced due to scattering by water droplets and/or ice crystals in the clouds. Clouds can block a significant portion of the UVR that would have otherwise reached the surface. Cloud cover and type are highly variable. The transmission of UVR radiation through clouds depends on cloud height, type, and optical density. The resultant effect on UVR transmission is difficult to assess,

SOURCES OF ULTRAVIOLET RADIATION

TABLE 30.4 Hemisphere 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90

Jan 229 212 203 155 127 95 79 42 28 16 12 6 4 0 0 0 0 0 0

1127

Maximum Erythemally Effective UV-B Irradiances for the Northern Feb 251 232 220 199 183 134 110 80 61 33 22 14 10 6 3 0 0 0 0

Mar 260 242 229 201 191 165 157 114 93 66 54 33 21 14 10 6 3 2 0

Apr 229 235 248 227 215 185 176 146 139 105 85 65 53 35 23 15 11 7 4

May 210 214 226 221 234 203 192 160 152 126 119 92 75 59 48 30 20 14 10

Jun 207 213 225 226 239 227 215 188 178 158 150 130 105 85 70 53 33 23 16

Jul 214 220 232 234 247 219 208 182 172 153 145 114 93 76 62 39 26 18 13

Aug 233 241 255 240 227 200 190 166 158 128 104 81 67 42 29 20 14 9 5

Sep 278 258 244 223 212 194 186 148 121 93 77 48 31 21 15 9 5 2 0

Oct 248 227 215 198 184 143 117 93 73 42 28 18 12 7 4 0 0 0 0

Nov 226 207 198 155 127 101 85 50 33 21 15 8 5 2 0 0 0 0 0

Dec 221 206 175 135 111 84 56 31 21 13 8 4 2 0 0 0 0 0 0

Note: Maximum erythemally effective UV-B irradiances (mW/m2eff) are calculated as a function of latitude and time of year (21st of each month).

TABLE 30.5 Number of Minimum Erythemal Doses (MEDs) in a 3 h Exposure Period for a Sensitive Skin Type for the Calculated Maximum Erythemally Effective UV-B Irradiances Given in Table 30.4 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90

Jan 12 11 11 8 7 5 4 2 2 1 1 0 0 0 0 0 0 0 0

Feb 14 13 12 11 10 7 6 4 3 2 1 1 1 0 0 0 0 0 0

Mar 14 13 12 11 10 9 8 6 5 4 3 2 1 1 1 0 0 0 0

Apr 12 13 13 12 12 10 10 8 8 6 5 4 3 2 1 1 1 0 0

May 11 12 12 12 13 11 10 9 8 7 6 5 4 3 3 2 1 1 1

Jun 11 12 12 12 13 12 12 10 10 9 8 7 6 5 4 3 2 1 1

Note: 1 MED for sensitive skin types is taken as 200 J=m2eff .

Jul 12 12 13 13 13 12 11 10 9 8 8 6 5 4 3 2 1 1 1

Aug 13 13 14 13 12 11 10 9 9 7 6 4 4 2 2 1 1 0 0

Sep 15 14 13 12 11 10 10 8 7 5 4 3 2 1 1 0 0 0 0

Oct 13 12 12 11 10 8 6 5 4 2 2 1 1 0 0 0 0 0 0

Nov 12 11 11 8 7 5 5 3 2 1 1 0 0 0 0 0 0 0 0

Dec 12 11 9 7 6 5 3 2 1 1 0 0 0 0 0 0 0 0 0

1128

ULTRAVIOLET RADIATION

particularly in the case of partial cloudiness. The effect of cloudiness on the solar irradiation of a horizontal plane can be approximated by F ¼ 1 0:056C

where C is the total cloud index in tenths of sky covered from 0 to 10, 10 being complete sky cover. Thus, on this analysis for complete cloud cover, the transmitted UVR irradiance would decrease to 44% and for half cloud cover to 72% of the incident value. However, cloud cover has been shown to decrease the measured UVR irradiance by over 90% in extreme cases. Estimates of the average reduction of UVB due to clouds (relative to cloudless skies) based on satellite measurements of backscattered solar UVR are 30% at 60 latitude, 10% at 20 latitude, and 20% at the equator (WHO, 1994). Increased levels of solar UVR due to stratospheric ozone layer depletion may have serious consequences for living organisms. A 10% reduction in stratospheric ozone could lead to as much as a 15–20% increase in UVR effective exposure at the Earth’s surface depending on the state of the troposphere and the biological process being considered. 30.3.2

Artificial Sources

Ultraviolet radiation from artificial sources may contribute significantly to certain people’s exposure in leisure, occupational and medical contexts. Typical UVR sources used for specific applications are shown in Table 30.6, although more than one type of source may be suitable for a given application. Conversely, a particular type of source may be used in a variety of applications. When considering the exposures that are likely to result from a TABLE 30.6

Nonlaser Ultraviolet Radiation Sources and Some of their Applications

Source Type Incandescent

Subgroup Tungsten Tungsten halogen

Solid-state lamps

Open arcs Gaseous discharge

LEDs, electroluminescent lamps Various Low-pressure Na Low-pressure Hg UVA black light Hg High-pressure Na High-pressure Hg Special high-intensity discharge (HID) High-pressure Hg/Xe Xenon Pulsed Xe

a

Typical Application

ELVs Exceededa

General, display, and emergency lighting Spotlights, heating, and floodlighting Display, panel indicators, night lights

No

Welding General and industrial lighting General lighting, horticultural, germicidal sunbeds Fluorescence, medical Floodlighting Industrial, printing, curing, commercial lighting Polymerization, reprography, sunlamps Photochemical and projection Projection and photography Printing, medical

Yes No Yes

Indicates whether the relevant ELVs can or are likely to be exceeded for unshielded sources.

Yes No

Possibly No Yes Yes Yes Yes Yes

SOURCES OF ULTRAVIOLET RADIATION

TABLE 30.7 Lamp

1129

Summary of Available Optical Radiation Data for Common Lamps

a

UVR ELVsb

Xenon short arc

6700

HID (high-intensity discharge) (Hg and metal halide), 400 W HID (Hg and metal halide), 3000 W

10

Low-pressure Hg, 30 W Low-pressure Hg, 15 W Medium-pressure Hg, 275 W Medium-pressure Hg, 400 W (clear) Medium-pressure Hg, 400 W (fluorescent) High-pressure Hg, 400 W

Tungsten halogen, 1 kW High-pressure Na, 150 W (HID)

240

750 5760 (at 0.2 m) 380 (at 1 m) 4800 10

Typical applicationc Solar simulators, searchlights, cinema projectors, spotlights Highway, industrial Photochemical, Printing and graphic arts (photoresist, ink and paint drying, curing) Germicidal Germicidal Sunlamp Lighting

10

Lighting

<1–12 (at 2 m)

Lighting (with outer envelope) (without outer envelope) Stage lighting Roadway lighting

<1–3900 (at 2 m) 20 0

Source: Adapted from Sliney and Wolbarsht (1980) and McKinlay et al. (1988). a

For base lamps at 0.5 m unless otherwise stated. Number of ELVs possible in an 8 h working day. c The data presented in column 2 are for bare lamps and may not be representative of the typical applications listed. For most well-designed installations the ELV will not be exceeded due to adequate shielding of the lamps. b

particular application, it is therefore important to obtain information about the source characteristics. It is difficult to generalize this information as the range of variants on each type of source is extremely large and relatively few measurement data have been published. Table 30.6 gives a general assessment of compliance with the relevant exposure guidelines in terms of the exposure limit values (ELVs) for each unshielded source listed; more specific measurement data for a range of sources are given in Table 30.7. 30.3.3

Incandescent Lamps

When a material is heated, a large number of energy transitions occur within the molecules of the material and optical photons are emitted. An idealized (most efficient) radiator of such radiation is termed a “blackbody”. For a true blackbody emitter, the total radiant power and its spectral distribution depend only on its temperature. However, in practice, no real material emits radiation with a blackbody spectrum, although tungsten at high temperatures (such as used for the filaments of incandescent lamps) and molten metals approximate to blackbody emitters. The incandescent lamp is the oldest type of lamp still in common use. Its emission results from a tungsten filament, which for typical powers of up to 500 W is generally heated to between 2700 and 3000 K. In applications where more power or a physically smaller source

1130

ULTRAVIOLET RADIATION

TABLE 30.8 Luminaires

Summary of Measurement Data for Ultraviolet Radiation from Desktop

Type of Lampa Desktop tungsten halogen, 20 W Desktop tungsten halogen, 20 W Desktop tungsten halogen, 20 W Desktop tungsten halogen, 50 W “Cool white” general lighting fluorescent lamp

Effective Irradianceb (mW=m2eff ) ACGIHc 6.1 2.2 25.0 2.2 0.20

Illuminance (lux)b Erythemad 10.1 3.3 41.0 3.8 0.28

1400 7600 14000 1600 500

Source: McKinlay et al. (1989) and Whillock et al. (1988). a Incorporating tungsten halogen lamps and from general lighting fluorescent lamps. b All tungsten halogen lamp measurements performed at 0.3 m from lamps. c ACGIH occupational hazard weighted irradiance. The exposure limit is equivalent to an 8 h exposure at 1 mW=m2eff . d International Commission on Illumination reference action spectrum (McKinlay and Diffey, 1987).

is required, tungsten (quartz) halogen lamps are often used. The combination of filament temperatures which are likely to be in the range 2900–3450 K and quartz bulbs results in a significantly higher level of UVR emission compared to the more traditional tungsten filament lamps. This may be problematic if close access to the lamps is possible for extended periods, where the lamps are fitted into desktop luminaires, for example (Table 30.8). In this situation, consideration should be given to the incorporation of appropriate UVR absorption filters into the luminaire. Incandescent lamps, other than tungsten halogen, normally have sufficiently thick glass envelopes to completely preclude a UVR hazard. Tungsten halogen lamps are used in agricultural applications, such as in plant cultivation and stock raising, and in industrial applications, such as ink and paint drying. If they are not enclosed or filtered, they can represent a potential UVR hazard as can unfiltered tungsten halogen lamps used for stage, display, and heating applications. 30.3.4

Electrical (Gaseous) Discharge

Optical radiation can be generated by electrical excitation of a low-pressure gas or vapor. A current is passed through a gas (or gases) ionized to produce electrons and positive ions. A fluorescent lamp discharge is a typical example. The energetic electrons that produce the ionization also excite the electrons of the gas atoms, and these subsequently deexcite resulting in the emission of characteristic radiations. The emission at 253.7 nm from mercury vapor is used as a source of excitation of the phosphors of low-pressure fluorescent lamps. By raising the pressure of the discharge to a few atmospheres the emission lines are increasingly broadened, eventually effectively forming a continuum. In some cases the 253.7 nm line emission will be self-absorbed by the vapor of the discharge. 30.3.5

Low Pressure Gas Discharge Lamps

Low-pressure gas discharge lamps are usually filled with an inert gas. The commonest type of low-pressure (nonphosphor) discharge lamp is the “neon” lamp. As glass absorbs shorter

SOURCES OF ULTRAVIOLET RADIATION

1131

wavelengths in the UVC and UVB regions of the spectrum, many low-pressure discharge lamps, including many filled with sodium or mercury, emit little short wavelength UVR. Where the UVR output is specifically needed, it is usual to manufacture the lamps with quartz envelopes. Of the common low-pressure lamps, only unfiltered mercury lamps constitute a severe UVR hazard. For example, the high UVC output (95% of the radiant energy emitted at 253.7 nm) of low-pressure mercury lamps with quartz envelopes is very effective at killing microorganisms, and these are often referred to as germicidal lamps. Applications where this is useful include area of sterilization (in hospital hallways, intensive care wards, and operating room suites), sterilization of biological containment cabinets, and water and food sterilization. In addition, such lamps have many analytical applications in chromatography, document assessment, and mineral identification. Given the potential to cause injury in a relatively short time, germicidal lamps should, where possible, be contained within a shielded enclosure. Particular problems may arise because tubular germicidal lamps often have the same bi-pin fittings as other tubular low-pressure discharge lamps and care is required to ensure that they are not accidentally fitted as replacements for general lighting tubes. Variants on the low-pressure discharge lamps are commonly used in sunbeds and horticulture, where they may represent a UVR hazard. 30.3.6

Fluorescent Lamps

The most common application of the low-pressure discharge is in fluorescent lamps. The fluorescent lamp operates by means of a discharge between two electrodes through a mixture of mercury vapor and a noble gas, usually argon. Visible light is produced by conversion of the 253.7 nm mercury emission to longer wavelength radiations by means of a phosphor coating on the inside of the wall of the lamp. By varying both the phosphor and the envelope, it is possible to produce a wide range of spectral emissions covering the visible, UVA, and UVB regions. Emission spectra generally consist of a broad continuum characteristic of the phosphor, combined with narrow bandwidth peaks corresponding to the characteristic line emission spectrum of the low-pressure mercury vapor discharge. Fluorescent lamps are available in a range of sizes, powers, and phosphors, the latter including a large selection producing “near white” and “special color” outputs. As indicated above, the presence of the phosphor coating and the outer glass envelope modify the output spectrum to remove shorter ultraviolet wavelengths. Exposures from white fluorescent lamps used for general lighting in the United Kingdom are low at around 2–5 MEDs for 1500 h of exposure (Whillock et al., 1988), although higher exposures have been reported for lamps in use in the United States (Cole et al., 1986a). Recent years have seen a growth in the popularity of compact fluorescent lamps that are essentially low-wattage, small-diameter fluorescent tubes folded into a compact form. Measurements from compact fluorescent lamps, normalized to an illuminance of 500 lx, are presented in Table 30.9. The output of any fluorescent tube may be further modified by the presence of diffusers/controllers, and Table 30.10 provides data on some typical combinations. 30.3.7

High-Pressure Discharge Lamps

High-pressure discharge lamps (mercury, sodium) are widely used for industrial and commercial lighting, street lighting, display lighting, and floodlighting. The general construction of high-pressure mercury lamps is a fused silica (quartz) discharge tube containing the mercury/argon vapor discharge mounted inside an outer envelope of soda lime or

1132

ULTRAVIOLET RADIATION

TABLE 30.9 Measured UVR Irradiance from Compact Fluorescent Lamps, Normalized to an Illuminance of 500 lx Lamp Type

Irradiance

Luma LC7 (7 W) Luma LC7 (7 W) with diffuser Osram Dulux EL (11 W) Philips SL9 (9 W) Sylvania Lynx CFD (13 W) Thorn 2D (16 W) Tungsram Globulux (16 W)

UVA (W/m2) 47 0.2 38 37 43 54 0.7

UVB (mW/m 2) 0 0 0.1 0 31 2.5 0

Source: Whillock et al. (1990).

borosilicate glass. The outer glass envelope effectively absorbs UVR; consequently, the quantity of potentially harmful UVR emitted by such lamps depends critically on the integrity of this envelope (FDA, 2008) although, even when intact, secondary filtration may also be required. The family of “metal halide” lamps encompasses a number of different types of highintensity mercury lamps whose discharges all contain additives. The additives are most typically metal halides chosen to produce a strongly colored emission (usually a single halide), produce a more broadly spectrally uniform emission (multihalide), or enhance the UVR (most often UVA) emission. Compared to “ordinary” mercury high-intensity discharge lamps, the luminous efficacies of metal halide lamps are high. Special HID mercury and metal halide lamps are used for photopolymerization, drying of inks and resins, reprography, projection and studio lighting, graphic arts, and commercial and domestic sunlamps and emit large amounts of UVR (Table 30.11). HID lamps require a UVR hazard evaluation, and units employing them should be interlocked to prevent exposure of workers and have appropriate warning signs. High-pressure sodium lamps are similar in construction to high-pressure mercury lamps but generally have a smaller diameter. The inner tube is made of polycrystalline alumina and the outer envelope of borosilicate glass. The emissions are characteristic of the sodium vapor and as the vapor pressure is raised, the emissions broaden across the visible spectrum. TABLE 30.10 Measured UVR Irradiance from a White Fluorescent Lamp Fitted with Various Diffusers/Controllers Diffuser Type Bare lamp Clear acrylicb Clear styreneb Opal styrenec Opal polycarbonatec

Irradiance UVA (mW/m2) 22.3 16.4 2.87 0.92 0.2

UVB (mW/m2) 3.45 2.91 0 3  10 3 1.2  10 2

Source: McKinlay et al. (1988). a

Weighted using the ACGIH occupational hazard weighting factors. Surface figured with small prisms. c Reeded surface. b

UVReff (mW/m2eff)a 5.9  10 2 4.8  10 2 0 2  10 5 9  10 5

SOURCES OF ULTRAVIOLET RADIATION

1133

TABLE 30.11 UV Irradiances Measured at 1 m from Typical Metal Halide Mercury Lamps Used for Graphic Arts Applications Lamp Type

Power (W)

HPA 400 HPA 1000 HPA 2000

400 930 1750

Irradiance (W/m2) UVC 0.5 2.3 4.5

UVB 3.2 9.0 19.0

UVA 9.0 23.0 48.0

Source: McKinlay et al. (1988).

30.3.8

Black Lights

Black lights are high- or low-pressure mercury discharge tubes where the broad spectrum of the mercury discharge is heavily filtered at both short and long wavelengths to produce an output that is almost exclusively UVA. Filtration is typically achieved through the use of a Woods’ glass (nickel/cobalt oxide) filter either incorporated into the envelope or positioned directly in front of the tube in the luminaire. Woods’ glass effectively removes short wavelengths and is almost entirely opaque to visible light. In addition, low-pressure tubes often incorporate a phosphor, which is selected for absorption of short wavelengths and has a peak emission around 370 nm in the UVA. Black lights are used in many nondestructive testing applications (in combination with fluorescent powders), for chemical and biochemical analysis, in philatelic and mineralogical examinations, in medical applications, and to produce special effects in display and entertainment. They are also widely used for many applications in forensic examination, where the recent development of SmartWater has led to their widespread use by police forces for routine examination of detainees and recovered property. Black lights are not normally considered hazardous, but problems can arise when the filter is inadequate to remove all the short wavelength output from the discharge, or when unfiltered UVR leaks out of cooling louvers or cracks in the luminaire housing. Provided the filtration is adequate, lowpressure lamps where the filter is incorporated into the envelope are intrinsically safe, as any damage to the envelope that might affect filtration would result in catastrophic failure of the lamp. In contrast, units with separate filters are susceptible to problems of poor maintenance and inspection that can result in continued use of luminaires with damaged or missing filters. For example, there have been a number of documented cases where the operation of damaged units in nightclubs has led to patrons suffering the effects of excessive UV exposure. Problems may also arise where people who are photosensitive (e.g., as a result of taking photosensitizing medication) work with black lights, as severe skin reactions can develop. 30.3.9

Arc Lamps

The UVR emissions from carbon and short arc lamps (high-pressure xenon or mercury) can be very high, and exposure to such sources may require an extensive hazard evaluation. Carbon arcs are commonly used in welding processes. High-pressure xenon lamps are used, for example, as solar simulators and in projection units and searchlights. Low-pressure and pulsed xenon lamps are often used in printing. Effective controls are required in the use of these lamps. Where an optical source of very high radiance and of small size is required. a very highpressure arc lamp may be used. These have a filling gas of mercury vapor, mercury vapor plus

1134

ULTRAVIOLET RADIATION

xenon gas, or xenon gas. Metal halide types are also available. Two physical types are commonly used: the compact (or short) arc and the linear arc. Typical applications of compact arcs include projectors, searchlights, and solar radiation simulators. The spectral emission of xenon lamps, which at wavelengths shorter than infrared closely matches that of a “blackbody” radiator at about 6000 K, enables their use as solar radiation simulators. Their emission spectrum is continuous from the UVR through to the infrared region. Emission peaks in the infrared principally between 800 and 1000 nm may be effectively removed by filtration. The luminance of compact xenon arcs may approach that of the sun (approximately 109 cd/m2) and in some lamps with greater than 10 kW rating the luminance of the brightest spot may exceed 1010 cd/m2. 30.3.10

Gas and Arc Welding

Because of their comparatively low operating temperature, the optical radiation hazards associated with gas welding processes are minimal. The use of standard welding filters necessary for comfort of viewing will prevent injury. In comparison with gas welding processes, the emissions of optical radiations from arc welding are very high (Sliney and Wolbarsht, 1980) (Table 30.12). The most common injury is photokeratitis resulting from excessive UVR exposure of the cornea, which can occur relatively quickly given the high UVR irradiances that occur close to the arc. The aversion response will generally limit accidental viewing sufficiently to prevent retinal photochemical injury, although this could occur during long-term viewing of a welding arc, which is not unknown. Exposed skin may develop ultraviolet-induced erythema. Containment screens can be used to provide collective protection to those in the general environment, but given the nature of arc welding, operator protection normally depends on personal protective equipment.

TABLE 30.12 UVR Emission Data Measured at a Distance of 1 m from Electric Welding Arcs Process/Base Metal/Gasa

Arc Gap (mm)

TIG/steel/Ar TIG/steel/He TIG/steel/Ar TIG/Al/Ar MIG/MAG/steel/CO2 MIG/MAG/steel/O2 þ Ar MIG/MAG/Al/Ar FCAW/steel/CO2 PAC/steel/N2 PAW/steel/Ar MMA/steel/none

1.6 1.6 3.2 3.1 – 6.4–9.5 6.4–9.5 – 6.4–19 4.8 3.2

Current (A)

UVR ELVsb

50–300 40–450 50–300 50–265 90–150 150–350 130–300 175–350 300–1000 260 100–200

30–1000 400–4500 30–1200 30–600 490–1300 6500–23000 2000–12500 210–3000 440–2800 1000 400–4400

Source: Adapted from McKinlay et al. (1988). a Key: TIG, tungsten inert gas; MIG/MAG, metal inert/active gas; FCAW, flux core arc welding; PAC/W, plasma arc cutting/welding; MMA, manual metal arc. b Number of ELVs in an 8 h working day.

OCULAR EFFECTS

1135

30.4 BIOLOGICAL MECHANISMS LEADING TO HEALTH EFFECTS The biological effects of UVR have been extensively reviewed (IARC, 1992; WHO, 1994; NRPB, 2002; Cridland and Saunders, 1994; Saunders et al., 1997). Acute exposure to UVR may produce deterministic effects, such as increased pigmentation, skin reddening (erythema), swelling (edema), and corneal sensitivity (photokeratitis). In addition, chronic exposure can induce degenerative changes such as wrinkling and elastosis of the skin, and cataract formation in the eyes; retinal degeneration is also associated with solar exposure. Chronic exposure may also elicit stochastic effects including both melanoma and nonmelanoma skin cancers. The induction of antigen-specific immunosuppression, which may occur following acute exposure, could be relevant to skin carcinogenesis and infectious disease. While human volunteer studies are of most relevance to human health effects, they are limited by ethical considerations, and it is necessary to consider data from studies on other animals where the human data are inadequate. It is, however, essential to exercise due caution when extrapolating data from animals to humans. This is especially so in the case of UVR, where target cells may be shielded by overlying cells or tissues. For example, in the case of skin, target cells in the basal and suprabasal layers are shielded by overlying layers, particularly the stratum corneum. This is approximately 10 cell layers thick in humans, but only one to two cell layers thick in mice, suggesting that mice will be much more sensitive, an effect likely to be more pronounced at shorter wavelengths where the radiation is generally more strongly absorbed by overlying cells. Similarly, for the eye, differences in size and structure will affect the penetration of damaging UVR in a wavelength dependant manner, making it difficult to extrapolate rodent data to humans. The adaptation of the rodent eye to a nocturnal habit further limits its usefulness, as an animal model, in this context.

30.5 OCULAR EFFECTS The eye (see Fig. 30.4) is adapted to not only focus visible radiation, but also focus other optical radiations including UVR, and the consequent increase in irradiance within the eye increases its sensitivity to the harmful effects of exposure. For the purposes of discussing these effects, the eye may be conveniently divided into three compartments, the anterior eye, the lens, and the posterior eye. The anterior eye is composed of the conjunctiva and cornea, the aqueous humor, and the iris. The cornea absorbs UVR strongly, particularly at shorter wavelengths, with over 90% of incident radiation below 300 nm absorbed. Transmission of UVA by the human lens is greatest in young people, falling from about 75% under the age of 10 years to around 20% in adults (Lerman and Borkman, 1976). There is some evidence for a small peak in transmission between 310 and 340 nm, with maximum transmittance at 320 nm similar to that at 400 nm. The human lens also absorbs strongly throughout the blue light region of the spectrum. As the wavelength of the radiation increases, progressively more penetrates to the posterior eye, and in particular to the retina. Here, the effect of focusing by the optical components of the eye is most marked and results in an increase in irradiance of over 105-fold. The combination of carotenoid and other pigments and a highly oxygenated environment, particularly in the central macula, can result in damage as a consequence of the photosensitized generation of reactive oxygen species.

1136

ULTRAVIOLET RADIATION

FIGURE 30.4

30.5.1

Structure of the human eye (adapted from Davson, 1967).

Acute Effects

Anterior eye: Acute overexposure to UVR can induce photokeratitis and photoconjunctivitis, inflammatory responses of the cornea and conjunctiva, respectively. Photokeratitis is a painful condition said to give the sensation of sand in the eyes. Symptoms typically occur around 6–12 h after exposure and usually last for one or two days; the exact time course depends on the severity of the exposure. Photoconjunctivitis exhibits a similar time course, but the symptoms may be less obvious, as the conjunctiva is less richly innervated than the cornea. Action spectra for photokeratitis have been determined in rabbits, primates, and humans (Pitts et al., 1977), and all show maximum sensitivity at around 270 nm, where the threshold value for humans is about 40 J/m2. Photoconjunctivitis exhibits a similar action spectrum, but with lower thresholds below 310 nm; at 270 nm, the threshold value is around 30 J/m2 (Cullen and Perera, 1994). Retina: In general, ultraviolet wavelengths are strongly absorbed by the anterior eye and the lens so that exposure to broadband UVR is unlikely to result in high retinal irradiances. However, there is some transmission across the UVB and UVA regions of the spectrum, so that exposure to an intense source emitting within this wavelength range might result in a high retinal irradiance. Only sources with very intense emissions in this spectral region are likely to be hazardous, and in practice this probably means that this type of damage is most likely from an ultraviolet laser. Transmission increases through the UVA region of the spectrum so that sources with very high UVA output may also give rise to acute effects. For very short (submicrosecond) exposures producing extremely high retinal irradiance, the principal damage mechanism is thermoacoustic disruption of the retina. For longer exposures (in the range of microseconds to seconds) with irradiances high enough to elevate thermal temperatures by 10 C or more, the predominant mechanism is thermal injury. The very high irradiances required for these effects are typically associated with very small source sizes. This combined with the requirement for emission to occur within a relatively

OCULAR EFFECTS

1137

narrow wavelength range suggests that injuries of this type are most likely to occur following exposure to ultraviolet lasers, particularly those that are pulsed. Longer exposure at more moderate retinal irradiance can result in photochemical damage. For example, prolonged viewing of the sun may produce solar retinitis, a form of photochemical damage to the retina that occurs as a result of absorption of both UVR and blue light, and similar injuries can occur following prolonged exposure to welding arcs (Romanchuk et al., 1978). Thresholds for photochemical damage may be exceeded in as little as 90 s (even with a constricted pupil) when gazing directly at the sun (Sliney and Wolbarsht, 1980). The risk is generally higher in the young due to the greater UVR transmittance of the young lens and the greater sensitivity of the young retina (Sliney, 1986; Boettner and Wolter, 1962). It may be increased at all ages through the use of poorly designed sunglasses that filter out long-wavelength visible radiation but transmit UVA and blue light. A degree of recovery from this type of damage may be possible, although this depends on the severity of the injury. 30.5.2

Chronic Exposures

Anterior eye: Repeated exposure to high levels of solar UVR is associated with the development of climatic droplet keratopathy (CDK), pterygium, and, to a lesser extent, pinguecula. CDK is a condition that involves the accumulation of translucent yellow deposits in the cornea, which spread from the periphery toward the center (Gray et al., 1992). The deposits are thought to be derived from plasma proteins that diffuse into the normal cornea and are then photochemically degraded following excessive exposure to UVR. Progressive accumulation of these deposits in later life can lead to significant visual disability, and people leading an outdoor life, especially in areas with high levels of reflected UVR, are particularly at risk. Pterygium are wing-shaped overgrowths of the conjunctiva that arise on the nasal side of the cornea and spread over the corneal surface. In severe cases, these overgrowths may obstruct vision. Like CDK, pterygium appears to be associated with an outdoor life in areas with high levels of UVR, particularly UVA and UVB (Taylor et al., 1989). Pinguecula, which are fleshy lesions of the conjunctiva, appear to be associated with UVR exposure but have no effect on vision. Lens: Cataracts are opacities of the lens and are the leading cause of visual loss in the world. They may be classified into cortical, nuclear, and anterior or posterior subcapsular on the basis of anatomical location. They arise as a consequence of disruption of the highly ordered structure of the lens fibers and damage to crystallins, the major lens proteins. Animal studies provide clear evidence of anterior cortical cataract induction by both acute and chronic exposure to UVB, although the former requires extremely high irradiances. The experimental evidence for cataract induction by UVA is weaker. Similarly, epidemiological studies of human populations suggest that cumulative exposure to solar UVR is an important cause of cortical cataracts and that UVB appears to be the main causative agent. The role of UVR in the causation of other forms of cataract is less clear (AGNIR, 2002). Retina: Age-related macular degeneration (ARMD) is a major cause of blindness in the developed world. ARMD results from damage to the macula, the cone-rich central part of the retina that is responsible for high visual acuity and color vision. Although it has been proposed that chronic exposure to solar radiation, and in particular to UVA and blue light, is associated with the development of ARMD (Mainster, 1978; Young, 1988), the evidence is not conclusive (NRPB, 2002). It is generally accepted that the aetiology of this condition is multifactorial and complex, involving both age and genetic factors, although

1138

ULTRAVIOLET RADIATION

there is certainly some evidence for a small adverse risk from blue-light exposure (Seddon, 2000).

30.6 NONMALIGNANT SKIN EFFECTS UVR is generally strongly absorbed in tissues and consequently has limited penetration into the deeper layers of the skin. In particular, short wavelengths in the UVC region of the spectrum are absorbed predominantly in the epidermis with little penetration into the dermis. Penetration increases with increasing wavelength, and data from human skin in situ reveal that for all wavelengths above about 300 nm penetration to the basal layer of the epidermis is high, suggesting that most of the UVR present in sunlight should show significant penetration (Chadwick et al., 1995). Acute exposure: With the exception of vitamin D synthesis, the acute responses of skin to UVR may be broadly divided into inflammatory responses (sunburn), which include erythema and edema, reparative responses, and protective responses, which include not only the familiar tanning response, but also thickening of the epidermis and stratum corneum. One of the most obvious and therefore most intensively studied consequences of acute overexposure to UVR is sunburn, which is characterized by erythema and blistering. Erythema, or reddening of the skin, results from increased blood volume in the superficial and deep vascular plexi of the dermis (Gange and Parrish, 1983) and is thought to be a consequence of localized release of eicosanoids, particularly prostaglandins. Histological examination has revealed that this is associated with intracellular edema, cytolysis, and, most prominently, the presence of sunburn cells. These cells can appear as early as 1 h after exposure, are eliminated within a few days, and probably arise from basal keratinocytes as a consequence of apoptosis. The nature of the erythemal response is spectrally dependent, with longer wavelength UVA radiation penetrating further and thereby producing more severe damage to the dermis, which may include vascular damage (Hruza and Pentland, 1993). A reference erythemal action spectrum (Fig. 30.2) has been derived from a statistical analysis of post-1945 data, derived mostly from studies on volunteers with skin types I and II (McKinlay and Diffey, 1987). This has been adopted by the International Commission on Illumination (CIE), the International Commission on Nonionizing Radiation Protection (ICNIRP), and the International Electrotechnical Commission (IEC) for spectral weighting and evaluation of effective irradiances of sources. While this action spectrum accurately predicts the response of human skin at shorter wavelengths, there is evidence that it may underestimate the effectiveness of long-wavelength UVA (Diffey, 1994; Anders et al., 1995). In mice, the most commonly used species for studying skin photobiology, edema, is probably a more sensitive response to UVR than erythema and action spectra have been derived (Cole et al., 1986a; de Gruijl et al., 1993). The other well-characterized effect of acute overexposure is altered skin pigmentation or tanning. Normal skin color is primarily determined by the nature, amount, and distribution of melanin in the keratinocytes of the epidermis. Melanin is an extremely dense and virtually insoluble polymer that occurs in at least two forms: brown eumelanin predominates in darkly pigmented skins, while reddish pheomelanin is the dominant variety in lighter skin types. Melanin is synthesized in specialized cells called melanocytes, packaged into melanosomes, and exported to neighboring keratinocytes. Exposure to UVR transiently increases the level of pigmentation in the skin, and this is known as facultative or inducible skin color to

NONMALIGNANT SKIN EFFECTS

1139

distinguish it from that of unexposed skin, which is referred to as constitutive skin color. Inducible pigmentation occurs in two phases, immediate pigment darkening and delayed tanning. Immediate pigment darkening (IPD) begins during exposure to UVR and reaches a maximum immediately afterward. It is characteristically short lived and may fade within minutes or days depending on the constitutive skin color and the dose of UVR. It is thought that IPD results from the oxidation and redistribution of melanin already present in the skin and occurs mainly as a result of exposure to UVA and possibly visible light. Delayed tanning, or melanogenesis, does not become visible until about 72 h after exposure and results from increases in both the number and activity of melanocytes. The action spectrum for melanogenesis is broadly similar to that for erythema, although wavelengths shorter than 290 nm are less effective for the former (Parrish et al., 1982). Melanogenesis is often thought of as a protective response but is in reality relatively ineffective, providing sun protection factors that range from about 1 (skin type II) to 4 (skin types V and VI). 30.6.1

Chronic Exposure

With the exception of skin cancers, which will be discussed separately, the most important consequences of excessive chronic exposure of the skin to UVR are photoaging and solar keratoses. Severely photoaged skin is characterized by dryness, wrinkling, elastosis and telangiectasia, or dilated blood vessels. It has a leathery, thickened, yellowed appearance, with actinic lentigines and motled hyperpigmentation (including freckles) present. Both skin type and anatomical site can affect the clinical features of photoaged skin. Comparison of exposed and unexposed skin on the same individuals indicates that photoaging of skin is different from chronological aging. Histologically, photoaged skin is characterized by the presence, in the dermis, of tangled, degraded elastic fibers that eventually degenerate into an amorphous mass (Oikarinen, 1990; Oikarinen et al., 1991). This is associated with degeneration and loss of collagen fibers, and a marked increase in proteoglycans and glycosaminoglycans, the “ground substance” of the dermal matrix. There is an increase in the number of inflammatory cells, particularly mast cells, and photoaged skin thus has the appearance of being chronically inflamed. Damage to the microcirculation that results in the formation of telangiectasias may become severe, with almost complete destruction of the horizontal superficial plexus. Although normally associated with excessive exposure to UVR, there is evidence that similar changes may result from even suberythemal doses. In particular, suberythemal UVA (30–50 kJ/m2) produces both dermal and epidermal damage (Lowe et al., 1995), while suberythemal UVB (30–50 J/m2) induces the activity of metalloproteinases that can degrade both collagen and elastin (Fisher et al., 1996). The mouse is probably the most frequently used model system for studies of photoaging, and despite considerable differences in skin structure, the changes observed appear to be broadly consistent with those in human skin. Mice have been used to determine action spectra for a variety of end points associated with photoaging, including collagen damage, elastosis, epidermal thickening, dermal cellularity, dermal inflammation, and skin wrinkling, and almost all were broadly similar with peak effectiveness observed around 290– 300 nm, and sharp decline in the UVA region (Bissett et al., 1989). In contrast, the action spectrum for skin sagging appears to be very broad with maximum effectiveness observed at around 340 nm. It has been suggested that pigskin more closely resembles human skin, and

1140

ULTRAVIOLET RADIATION

may therefore be a better model. Certainly, chronic exposure to broadband UVR at suberythemal doses induces similar changes to those observed in photoaged human skin (Fourtanier and Berrebi, 1989). Chronic exposure to UVR is associated with the development of small scaly, erythematous lesions called solar keratoses. They consist of benign proliferations of epidermal keratinocytes and are common on exposed body sites in older Caucasians living in areas of high ambient solar irradiance. They are considered to be premalignant lesions, with the potential to progress to squamous cell carcinoma, although it appears that the probability of an individual lesion progressing is low. Moreover, many may actually regress if appropriate sun avoidance is adopted (Marks et al., 1986). Nevertheless, they usually occur as multiple lesions and their number on the skin is strongly associated with the risk of nonmelanoma skin cancer (Kricker et al., 1991).

30.7 SKIN CANCER The most serious health effects in humans for which exposure to UVR is a recognized risk factor are the cutaneous malignancies (skin cancers). They can be divided into two main types, nonmelanoma skin cancer (NMSC) and the malignant melanomas. 30.7.1

Nonmelanoma Skin Cancer

Although the nonmelanoma skin cancers are relatively common in white populations, they are rarely fatal, accounting for around 10% of registered malignancies in the United Kingdom, but under 0.5% of cancer deaths. There are two main types of NMSC, basal cell carcinomas (BCC), which account for around three-quarters of cases, and the less frequently observed squamous cell carcinomas (SCC). The former appear typically as raised translucent nodules, which develop slowly over a period of months or even years on the face, head, and neck and have a relentless capacity for local spread and destruction. Squamous cell carcinomas, which develop from slightly more differentiated epidermal cells, appear as persistent red, crusted lesions on sun-exposed skin, most commonly on the face and the scalp. Overall, the findings from both descriptive and analytical epidemiological studies support the hypothesis that, for both types of NMSC, exposure to UVR is a major causative factor (AGNIR, 2002). The evidence suggests that cumulative lifetime exposure is the main risk factor for SCC. However, the situation is more complex for the much more common BCC, and it now appears that above a certain level of dose the risk does not rise, and the degree of intermittency of exposure may be of importance. Most recent studies of photocarcinogenesis have used the hairless mouse model, employing either the albino (Skh:h-1) strain or the lightly pigmented (Skh:h-2) strain. Tumors have been induced following repeated exposure of these strains to simulated solar radiation and broadband UVR, as well as to radiations in each of the UVC, UVB, and UVA regions of the spectrum. The tumors induced are usually SCC, often preceded and accompanied by papillomas; keratoacanthomas and similar benign epidermal neoplasms found in humans have also been reported. However, the most common human malignant skin tumor, BCC, is rarely observed, having been identified in only a few studies with nude mice and one or two early studies with rats (IARC, 1992). Hence, the validity of the hairless mouse as a model of human skin carcinogenesis must be open to question. The induction of

SKIN CANCER

1141

squamous cell carcinoma in hairless mice provides most of the recent quantitative data available on experimental photocarcinogenesis in vivo. However, extrapolation to humans is not straightforward, since mouse skin differs from human skin in several respects. Mouse skin is much thinner, for example, and daily doses of UVR required for skin tumorigenesis in mice are usually well below those present outdoors, and most experiments have been conducted with UVB doses lower than those required to elicit acute reactions, such as erythema, in mice. The wavelength dependence of UVR-induced carcinogenesis is crucial to accurate risk assessment. Clearly, an action spectrum for skin cancer can only be obtained from animal models, such as the hairless mouse, and this effectively restricts the data to squamous cell carcinomas. The best spectrum currently available is the SCUP (Skin Cancer Utrecht Philadelphia) action spectrum (de Gruijl et al., 1993) derived by combining data on chronic UVR-induced tumors from experiments carried out at the University of Utrecht (de Gruijl and van der Leun, 1992) and at the former Skin and Cancer Hospital in Philadelphia (Cole et al., 1986b). It is based on data from a total of 1100 hairless albino Skh:h-1 mice exposed each day to radiation from one of the 14, mainly broadband, sources; in general, the use of narrowband sources presents logistical problems. Exposures in the UVA and UVB regions were below thresholds for acute effects, whereas UVC exposures exceeded them. The SCUP action spectrum (Fig. 30.5) extends between 270 and 400 nm but is best defined between 280 and 340 nm, with increased uncertainty outside this region. The interpretation of action spectra could be complicated by the possibility that UVR, of different wavelengths, might interact in the induction of skin tumors in a more complex way than simple addition. This has been the subject of much debate and investigation, and remains controversial (IARC, 1992). Difficulties with interpretation arise from differing spectral content of various sources of UVR, uncertainties in measurement, exposure to insufficient doses of UVA, and differences in epidermal transmission, absorption, and photoproduct formation across the UVR spectrum; it has also been suggested that UVB may be a more effective initiator and UVA a more effective promoter (Kelfkens et al., 1992). Overall, it appears that the interactions of UVR from different wavelength ranges,

FIGURE 30.5

SCUP action spectrum for mouse carcinogenesis.

1142

ULTRAVIOLET RADIATION

when given simultaneously, prior to, or immediately after each other, are either unproven or small (IARC, 1992).

30.8 MALIGNANT MELANOMA Although it is much less common than nonmelanoma skin cancer, cutaneous malignant melanoma (CMM) is the main cause of skin cancer death, particularly in young people; in the United Kingdom, CMM accounts for 1 in 11 cancers and 1 in 20 deaths between the ages of 20 and 39. There are four main clinical pathological types. Superficial spreading melanomas are by far the most common and are typically seen on the leg in women and the trunk in men, while nodular melanomas may be seen on any site. Lentigo maligna melanomas are typically found on sun-exposed skin in the elderly, and acral (or acral lentiginous) melanomas are found on the palms of the hands and soles of the feet. The incidence of CMM has increased substantially in white populations for several decades. For example, in the period between 1960–1964 and 1985–1990, incidence rates in Scotland quadrupled, while in the period between 1953–1954 and 1990–1993 mortality more than doubled (Swerdlow et al., 1998); similar large increases in incidence and mortality have been observed in England and Wales (Swerdlow et al., 2001) and in Sweden (Karlsson et al., 1998). The major risk factor appears to be exposure to solar radiation, although the exact relationship is unclear (AGNIR, 2002). Most of the epidemiological evidence fits best with intermittent intense exposure of untanned skin as the main aetiological exposure. This is particularly true for intermittently exposed skin, superficial spreading melanoma, and nodular melanoma. However, there is also some evidence that fits well with cumulative dose, particularly in relation to head and neck melanomas and lentigo maligna melanoma. There is little evidence on which to assess the contribution of solar radiation to cancers of rarely exposed sites, although a recent study has suggested that sun exposure may be important, presumably through a systemic effect. There is also a body of evidence suggesting that exposures during childhood and adolescence may be of special importance. The evidence that exposure to sunlamps and sunbeds may contribute to melanoma risk is suggestive but not definitive (AGNIR, 1995). A higher incidence is seen in people with large numbers of naevi (moles), atypical naevi, light skin, red or blond hair, blue eyes, a tendency to freckle, and a tendency to sunburn and not to tan on sun exposure. In a small number of cases (<5%), there is a strong family history of disease. Despite many concerted attempts, malignant melanomas have not been experimentally induced by UVR in any placental mammals, although pigmented melanoma-like tumors have been induced in mice by a combination of UVR and chemical carcinogens (see below). At present, the best animal models for UVR-induced melanoma are the South American opossum, Monodelphis domestica, and certain hybrid fish (platyfish and swordtail fish hybrids). However, melanomas induced in these animals are histologically different to human malignant melanomas. Hybrids of the fish Xiphophorus maculatus and X. Couchianus have been used to determine an action spectrum for melanoma induction (Setlow et al., 1993). The effectiveness of UVR in inducing melanoma appeared to decrease by less than two orders of magnitude between 302 and 435 nm, which contrasts strongly with action spectra for DNA damage and suggests the presence of an efficient sensitizer in the target melanocytes. However, these fish represent a specialized genetic model in which melanoma

MALIGNANT MELANOMA

1143

can be induced by inactivation of a single gene that may not be relevant to exposed human populations. 30.8.1

Mechanisms of UVR-Induced Carcinogenesis

Experimental protocols that involve sequential application of different carcinogens provide a means to investigate mechanistic aspects of carcinogenesis. This approach has been extremely informative in relation to skin carcinogenesis induced by exposure to chemical carcinogens and has been central to the development of the multistage model of carcinogenesis. It is currently believed that there are at least three main stages of skin carcinogenesis, initiation, promotion, and progression. Initiating events are thought to result from stable changes in the genetic information (DNA) carried by an affected cell. Initiation can occur following a single exposure to the initiating agent and is irreversible. Promotion is usually a more extended process than initiation, requiring prolonged or repeated exposure and may be at least partially reversible if the promoting agent is withdrawn. Promoting agents often induce hyperplasia, and it is thought that this increased proliferation permits initiated cells to multiply and form a clone of cells expressing an altered phenotype. However, while clonal expansion of initiated cells may be required for promotion, it is not sufficient, and agents that are capable of promoting carcinogenesis following a single application of an initiating agent usually induce additional changes in a process sometimes termed conversion. These may include stable genetic changes and probably result from oxidative stress. The third stage of carcinogenesis, progression, is the process by which initiated and promoted cells acquire increasingly malignant phenotypes. These changes are associated with increasing genetic instability, loss of growth control, and ultimately acquisition of invasive characteristics. As exposure to UVR alone is sufficient to induce tumors, it may be classified as a complete carcinogen (IARC, 1992). This implies that exposure to UVR has an effect on both of the first two stages of carcinogenesis, initiation and promotion. This conclusion is supported by evidence from traditional two-stage experimental carcinogenesis protocols. Studies designed to evaluate the action of UVR as a tumor initiator have employed a protocol involving irradiation of mice, followed by repeated application of known chemical tumor promoters such as phorbol esters. Irradiation has usually involved a single exposure to UVR, although in some cases the exposure has been fractionated over a period of days or even weeks; fractionation, particularly over a long period, may complicate interpretation of the results. In general, the results of these studies have shown that both UVC (Pound, 1970) and broadband UVB (Epstein and Roth, 1968; Stenb€ack, 1975) radiations are effective initiating agents. In addition to studies designed to assess whether UVR can act as a tumorinitiating agent, there have been a number of studies that have investigated whether known or suspect tumor-promoting agents can enhance tumor formation in mouse skin initiated with UVR. One agent that was found to promote UVR-initiated skin tumors was methyl ethyl ketone peroxide, a compound widely used in the polymer industry (Logani et al., 1984); the mechanism of action of this compound appeared to involve oxidative stress. In contrast, there was no evidence for promotion of UVR-initiated skin tumors by benzoyl peroxide, a known skin tumor promoter used in the treatment of acne and that also has applications in the polymer industry (Epstein, 1988; Iverson, 1986, 1988). Possible promoting actions of UVR have been investigated by painting mouse skin with the chemical initiator 7,12-dimethylbenz[a]anthracene (DMBA) prior to repeated exposure to UVR. In general, such studies have shown that the combination of chemical initiation and either UVB or UVA is more effective at inducing tumors than treatment with DMBA alone

1144

ULTRAVIOLET RADIATION

(Epstein, 1965; Epstein and Epstein, 1962; Epstein et al., 1967; Husein et al., 1991). Interestingly, when pigmented mice were used in these studies, it was found that both UVB and UVA promoted the development of chemically induced pigmented lesions, including melanoma-like tumors (Epstein et al., 1967; Husein et al., 1991). A novel approach has been used to demonstrate that UVB also promotes chemically initiated SCCs in human skin grafted onto immunocompromised mice. However, human skin was much less susceptible than the surrounding mouse skin (Soballe et al., 1996). 30.8.2

UVR-Induced DNA Damage

The transition from a normal cell to a malignant one is a complex process involving a number of changes, many of which involve the induction of mutations as a consequence of genetic damage. UVR is an effective DNA damaging agent (Cridland and Saunders, 1994), although the nature of the damage is wavelength dependant (Fig. 30.6). UVB and UVC radiations are absorbed directly by the bases of DNA resulting in the electronic excitation and the formation of photoproducts. The most common photoproducts are those formed between adjacent pyrimidines and include cyclobutane pyrimidine dimers and, less frequently, pyrimidine (6-4) pyrimidone adducts or “(6-4) photoproducts.” There has been much debate about the relative importance of these two photoproducts, and, overall, the available data suggest that they are probably of roughly equal importance. Action spectra for photoproduct formation have been determined in vitro, in cultured cells, and in skin (Fig. 30.7). Above about 300 nm, the absorption of DNA, and consequently the formation of pyrimidine photoproducts, falls rapidly. However, the presence of photosensitizing chemicals that can absorb UVR and transfer the energy to DNA may extend photoproduct

FIGURE 30.6 Summary of principal pathways of UVR-induced DNA damage. Only those pathways producing the most common photoproducts at each wavelength are shown: (6-4_PP, (6-4) photoproducts; Hyd, pyrimidine hydrates; CPD, cyclobutane pyrimidine dimers; SSB, single-strand breaks; DSB, double-strand breaks; PDC, protein–DNA cross-links; 8-HG, 8-hydroxyguanine.

MALIGNANT MELANOMA

1145

FIGURE 30.7 Action spectrum for cyclobutane pyrimidine dimer formation in epidermal DNA (&, data from Ley et al., 1983; ~, data from Freeman et al., 1989).

formation to longer wavelengths. Sensitized DNA damage may occur either by direct energy transfer from an excited sensitizer to DNA or through the generation of a chemically reactive intermediate, although the latter mechanism is probably the most important and certainly predominates at longer wavelengths (Tyrrell and Keyse, 1990). Typically, the intermediates formed are activated oxygen species, and such reactions are termed photodynamic; the energy required to form these species is very low and can correspond to wavelengths extending into the infrared region. The most important intermediate is probably singlet oxygen ð1 O2 Þ and, in mammalian cells, this appears to react principally with guanine to form products such as 8-hydoxyguanine (Epe, 1991; Piette, 1991; Kvam and Tyrrell, 1997a). Evidence for variation in the action spectrum in different cells (Kvam and Tyrrell, 1997b; Kielbassa et al., 1997) suggests that a number of different chromophores may be involved in mediating the formation of 8-hydroxyguanine, although its relevance to either mutagenesis (Douki et al., 1999) or carcinogenesis (Van Kranen et al., 1997) appears questionable. Breaks in the sugar–phosphate backbone of one (single-strand breaks or SSBs) or both (doublestrand breaks or DSBs) of the DNA strands are produced following irradiation at both short and long wavelengths. However, SSBs are produced much less efficiently than CPDs at short wavelengths and only constitute a major type of photodamage at wavelengths in the UVA and visible regions (Moan and Peak, 1989). Double-strand breaks are likely to show similar wavelength dependence, although they are produced at a lower frequency than single-strand breaks. The biological importance of both types of lesion is unclear. The third major lesion, which is produced by indirect sensitization is the DNA–protein cross-link (DPC) (Moan and Peak, 1989; Cridland and Saunders, 1994). Cellular DNA is intimately associated with protein to form chromatin, and reactions between excited bases and susceptible protein components in close proximity have been well established for some time. It is considered that these lesions may be biologically important at longer wavelengths. A number of compounds can act to sensitize UV-induced DNA damage. Cellular constituents capable of behaving in this way include riboflavin, bilirubin, porphyrins, nicotinamide coenzymes, and rare thiolated bases (Tyrrell and Keyse, 1990; Cridland and Saunders, 1994); some steroids, quinines, and carotenoids may exhibit similar properties. Synthetic chemicals that may act as photosensitizers include dyes, drugs, both oral and

1146

ULTRAVIOLET RADIATION

topical, antibiotics, ketones, polycyclic aromatic hydrocarbons, and some cosmetic and sunscreen preparations (Hawk, 1984; Cridland and Saunders, 1994; Saunders et al., 1997). In addition, natural products such as chlorophyll may also act as sensitizers.

30.9 IMMUNE SYSTEM EFFECTS Immunological responses protect individuals from infectious disease caused by invading microorganisms such as viruses, bacteria, and various single-celled or multicellular organisms. Responses can also be elicited by noninfectious foreign substances that enter the body, such as bacterial toxins, foreign proteins, and various organic and inorganic chemicals. In addition, tumor cells may express on their surfaces molecules that can induce immunological responses against them although, in general, for a number of different reasons, most tumors elicit only weak immune responses (Saunders et al., 1997). UVR exerts its effect on the immune system via its interaction with the skin. This is a major site of entry of foreign antigens into the body and is an active participant in immunological defense; the dermis and the epidermis contain cell populations able to initiate and support appropriate immunological responses. Many skin cells, especially the keratinocytes, secrete cytokines, which are important mediators in the control of immune and inflammatory responses. Effective immune responses additionally depend upon the clonal expansion and activation of appropriate antigen-specific lymphocytes. Antigenpresenting cells are particularly important in this context, and the epidermal Langerhans cells are the major antigen-presenting cells in the skin. Langerhans cells process antigen encountered in the skin and migrate via lymphatic vessels to the regional lymph node where they “present” the antigen to T helper lymphocytes, activating those specific to the antigen. Activated T helper cells undergo clonal expansion and differentiation, releasing cytokines that further activate cell types involved in the immune response. In general, the first application of an antigen results in sensitization, a process in which the immune system becomes poised to respond; a second, later, application of the antigen elicits a much stronger immune response from previously activated antigen-specific lymphocytes such as memory T cells, the predominant dermal T cell. The effects of UVR on the immune system, and the mechanisms that underlie them, have been reviewed widely (Kripke, 1993; Noonan and De Fabo, 1993; Cridland and Saunders, 1994; Vermeer and Hurks, 1994; Krutman and Elmets, 1995; Ullrich, 1995; Saunders et al., 1997). Animal and human studies have shown that the major effect of exposure to UVR on immune system responses is to suppress T cell-mediated responses such as delayed-type (type IV) hypersensitivity (DTH), the primary defense mechanism against intracellular bacteria and protozoa, and the antiviral and antitumor activities of cytolytic T cells and natural killer cells. This is believed to be a consequence of profound changes to the populations of epidermal antigen presenting cells and changes in cytokine secretion by epidermal cells. These are thought to result in the inactivation of the T helper cell subset that stimulates the T cell-mediated responses. Nevertheless, activation of the T helper cell subset mediating antibody responses such as immediate hypersensitivity appears to be retained. It appears that the principal chromophores mediating these responses are DNA and transurocanic acid (trans-UCA). One form of DTH that has been the subject of considerable investigation in relation to UVR effects is contact hypersensitivity (CHS), which is characterized by an eczematous reaction at the point of contact with the allergen; CHS is a significant cause of morbidity among

POPULATIONS AT SPECIAL RISK: OCULAR DAMAGE

1147

employees in industries that use highly reactive chemicals such as nickel, chromate, and rubber (chemical) accelerators. Volunteer studies have shown that exposure to UVR sufficient to induce erythema, followed by chemical sensitization on either exposed or unexposed skin, can result in a marked dose-dependent suppression of the CHS response to a subsequent challenge. Individual responses to UVR are, however, quite variable, and there is evidence to suggest that the susceptibility of the immune system to UVR exists as a balanced polymorphism within the human population; it is possible that UVR sensitivity may be an additional risk factor for skin cancer. Moreover, there is some evidence that the CHS response to previously encountered antigens may actually be enhanced following exposure to UVR. Pathogenic organisms differ greatly in their patterns of host invasion, colonization, and the nature of the antigenic signal that they generate and, not surprisingly, are able to elicit a very diverse number of immune responses involving both antibody-mediated and cellmediated effector mechanisms. Exposure to UVR has been shown to downregulate T cellmediated tuberculin-type DTH responses, characterized by an area of hardening and swelling, following intradermal challenge by a variety of pathogens (or antigenic preparations), mostly associated with intracellular skin infection. These include bacteria such as the leprosy and tuberculosis bacilli, protozoan parasites such as Leishmania and viruses such as the cold sore virus herpes simplex (HSV). Suppression of DTH responses to leprosy antigens and reactivation of latent HSVinfection following UVR exposure was seen in human studies. Support for some role of the immune system in tumor eradication in humans includes the occurrence of spontaneous tumor regression, the observation of lymphoid infiltration in many tumors (although this is not clearly related to patient prognosis), and a higher incidence of certain types of tumor in immune suppressed (e.g., transplant) patients. Strongly immunogenic tumors include the skin cancers induced by UVR and lymphocytic infiltration, and more rarely, spontaneous regression, have been observed. Moreover, the risk of skin cancer, especially squamous cell carcinoma, is elevated in immunosuppressed patients, particularly those living in areas of high insolation. Although animal studies have provided good evidence that UVR-induced immunosuppression plays a role in the growth of SCC, there is no animal model for basal cell carcinoma, and the evidence concerning melanomas induced by exposure to a combination of chemicals and UVR is less clear. 30.10 POPULATIONS AT SPECIAL RISK: OCULAR DAMAGE A number of people have a greater sensitivity to UVR-induced damage to the retina as a consequence of increased transmission through the anterior eye. In addition, there is a potential for overexposure as a consequence of clinical examinations and through accidental exposure to laser radiation. 30.10.1

Children and Young People

The risk of retinal damage is higher in the young because their lenses transmit more UVR and blue light. This is an area of particular concern as neonates and infants may be exposed to high blue-light radiances in neonatal intensive care units. 30.10.2

Aphakes

Those who have had the lens of the eye removed during cataract surgery are also a high-risk group for retinal injury. Replacement by a polymethyl methacrylate (PMMA) lens implant

1148

ULTRAVIOLET RADIATION

does not provide the same protection for the retina that results from the presence of the natural lens (Mainster, 1978; West et al., 1989). It is therefore important to incorporate appropriate filters for retinal protection, and these should reduce transmission of blue light as well as UVR. 30.10.3

Ophthalmic Patients

Patients undergoing ophthalmic examination with the slit lamp microscope or indirect ophthalmoscope may be exposed to high irradiances of visible radiation. In practice, extremes of exposure are probably avoided by the lack of a focused image and by constant movement of the source, but there is a need for practitioners to exercise care, especially in anaesthetized patients whose eyes are immobilized (AGNIR, 2002). 30.10.4

Laser Workers

Accidental injuries may occur in those working with high-power lasers, particularly during alignment procedures or as a result of reflections from neighboring surfaces (Sliney and Wolbarsht, 1980). In addition, bystanders observing laser ocular surgery may suffer injury as a result of reflections from contact lens used to direct radiation into the patients’ eyes.

30.11 POPULATIONS AT SPECIAL RISKS: SKIN EFFECTS Skin phenotype is probably the most important factor determining sun sensitivity in the general population. In addition, there are small numbers of people with clinical conditions that result in extreme sun sensitivity, while exposure to photosensitizing chemicals from a variety of sources can greatly modify the response to UVR. 30.11.1

Phenotype

Certain phenotypic traits such as pale skin, red or blond hair, and blue eyes are clearly associated with sun sensitivity and a tendency to burn easily and to tan poorly. In addition to predisposing to the acute effects of overexposure, such characteristics also carry an elevated risk of UV-induced skin cancers; for malignant melanoma, a tendency to freckle, large numbers of naevi and atypical naevi are also risk factors. 30.11.2

Clinical Conditions

A number of genetic disorders also predispose to sun sensitivity. For example, oculocutaneous albinism is an uncommon autosomal recessively inherited disorder of melanin synthesis that results in reduced or absent melanin pigmentation in the skin, hair, and eyes. Patients suffer from an increased frequency of sunburn, actinic damage, and NMSC but appear to have a low incidence of malignant melanoma. Predictably, genetic disorders that affect DNA repair also affect sun sensitivity. The best characterized of these is probably a rare autosomal condition called xeroderma pigmentosum. Sufferers typically exhibit sun sensitivity, abnormal erythemal responses, and changes in chronically exposed skin that may include the development of pigmented macules, achromic spots, telangiectasia, BCC, SCC, and malignant melanoma. Two other rare

POPULATIONS AT SPECIAL RISKS: SKIN EFFECTS

1149

recessive disorders of DNA repair, Cockayne syndrome and trichothiodystrophy, result in increased sun sensitivity but interestingly do not appear to confer an elevated risk of skin cancer. Other clinical conditions associated with increased sun sensitivity include lupus erythematosus and the porphyria group of disorders. Patients typically exhibit acute and prolonged UVR-induced erythema and even blistering following exposure to levels of UVR that would not affect most people. 30.11.3

Exposure to Photosensitizers

Exposure to photosensitizing chemicals may result in increased sun sensitivity, and these responses may be conveniently divided into phototoxic and photoallergic reactions (Cridland and Saunders, 1994; Saunders et al., 1997). Phototoxic reactions usually involve reactive oxygen species that damage cells in the exposed tissue; phototoxic reactions may be observed in both the skin and the eye. There are three main types of phototoxic response. The first is an immediate burning sensation with erythema and urticaria (wheal and flare), while the second is a delayed reaction resembling sunburn (erythema and edema). A third type of reaction involves increased pigmentation, which is often delayed in onset and confined to the exposed area. Photoallergic reactions are those in which an immune mechanism can be demonstrated. Rather than exaggerated sunburn, the clinical reaction more usually consists of an eczematous eruption or discrete papules and plaques. Because photoallergy depends upon individual immunologic reactivity, it occurs in only a small minority of exposed individuals. Moreover, the dose–response relationship is less evident than in phototoxic reactions, and small quantities of chemical or radiation may be adequate to elicit a response. In a minority of people, a photoallergic reaction persists after exposure to the photosensitizer has stopped. Whereas photocontact dermatitis is usually UVA induced, the action spectrum for persistent light reactivity alters to include UVB and the patient may be unable to tolerate even minimal exposure to sunlight. This syndrome of persistent light reactivity is now thought to be a form of chronic actinic dermatosis, an acquired idiopathic disorder. Other idiopathic dermatoses include solar urticaria, an immediate (type I) hypersensitivity that may occasionally be photosensitized by topical chemicals and drugs, and polymorphic light eruption, the commonest of these disorders for which, however, there is no evidence of prior chemical photosensitization. Photosensitizing chemicals may be present in coal tar extracts and their derivatives, plant extracts, pharmaceuticals, and cosmetics (Cridland and Saunders, 1994; Saunders et al., 1997). Thus, they may be encountered in a variety of situations, including the home (Harber and Levine, 1969), manufacturing industry (Crow, 1961), agricultural industry (Birmingham et al., 1961), and during recreational activities (Hjorth and M€ oller, 1976). 30.11.4

Exposure to Environmental Chemicals

There is evidence suggesting that environmental exposure to arsenic may act to increase the carcinogenicity of UVR (reviewed in Rossman et al., 2004). Arsenic contamination of drinking water is of concern in a number of countries and has been associated with an increased incidence of skin cancer (SCC and BCC, but not melanoma) in Taiwan, Chile, Argentina, Bangladesh, and Mexico. The increased risk has been generally attributed to arsenite, but at low concentrations there is little evidence that this is mutagenic, a complete

1150

ULTRAVIOLET RADIATION

carcinogen, a tumor initiator, or a tumor promoter. It therefore appears likely that arsenite may act in concert with another agent, and in relation to skin cancer, environmental exposure to UVR would be an obvious candidate. Recent studies on hairless mice exposed to mainly UVB radiation indicated that the presence of arsenic in drinking water enhanced UV carcinogenesis in a dose-dependent fashion (Rossman et al., 2001; Burns et al., 2004). A recent study from the same group has shown that the presence of chromium in drinking water also enhances the carcinogenic action of exposure to ultraviolet radiation (mainly UVB) in a dose-dependent manner (Davidson et al., 2004).

30.12 APPLICABLE STANDARDS AND EXPOSURE GUIDELINES Quantitative guidelines on limiting exposure are generally restricted to the avoidance of immediate effects of acute exposure, and this reflects the paucity of quantitative data in relation to the late effects resulting from either acute or chronic exposures. Maximum permissible exposures from coherent UVR (laser radiation) are published by the International Commission on Nonionizing Radiation Protection (ICNIRP, 1999) and have been used as the basis of the International Laser Safety Standard (IEC 60825-1: 2007). Similar limits are also published by the American Conference of Governmental Industrial Hygienists (ACGIH, 2003) and are used as the basis of the US ANSI standards. Laser technology and applications in the UVR region of the spectrum have been comprehensively reviewed elsewhere (Elliot, 1995). In relation to noncoherent UVR, comprehensive exposure guidelines have been published by ICNIRP and ACGIH (ICNIRP, 1999; ACGIH, 2003). Both sets of guidelines define similar exposure criteria for UVR, a set of exposure limit values (ELVs) below which it is expected that nearly all people may be repeatedly exposed without adverse effect. The guidelines are based on the available scientific data and no consideration is given to economic impact or other nonscientific priorities. The stated intention of the ICNIRP guidelines is to serve as guidance to the various international and national bodies or individual experts who are responsible for the development of regulations, recommendations, or codes of practice to protect workers and the general public from the potentially adverse effects of UVR. For example, within the United Kingdom, it is currently recommended (NRPB, 2002) that the ICNIRP ELVs should be applied in situations where quantitative limitation of exposure is practical. This will include workplace exposure to UVR from artificial sources and public exposure from artificial sources in a controlled environment. It is, however, recognized that members of the public may elect to be exposed to artificial sources such as tanning equipment, and in these situations the application of quantitative limits is not practical because the intended outcome of the exposure would not be achieved. Similarly, for elective sun exposure by members of the public, strict application of quantitative limits is simply not practical, and in this case public education is the most effective approach to exposure limitation. The situation is somewhat different for occupational exposure to solar UVR, as there is no intention to elicit specific responses such as tanning. Nevertheless, factors such as exposure geometry and behavior may make it impractical to apply quantitative limits and the recommended approach is to use available control measures to limit exposure without adhering to specific limits. The criteria for limiting exposure to sources of noncoherent UVR are summarized in Table 30.13.

APPLICABLE STANDARDS AND EXPOSURE GUIDELINES

1151

TABLE 30.13 Summary of Exposure Limit Values (ELVs) Applicable in Any 8 h Period Wavelength (nm)

Exposure Limit Value

Tissue Protected

180–400 (UVR) 315–400 (UVA)

30 J=m2eff 4 2

Skin and eye Eye

10 J=m

Source: ICNIRP (1999).

The ELVs discussed above are intended for use in the control of exposure to both pulsed and continuous noncoherent sources where the exposure duration is not less than 0.1 ms. They are used to evaluate potentially hazardous exposure from, for example, arcs, gas and vapor discharges, fluorescent lamps, incandescent sources, and solar radiation. The ELVs are below levels that are often used for the UVR exposure of patients required as part of medical treatment and below levels for elective cosmetic purposes. The limits apply to exposure directed perpendicular to those surfaces of the body facing the source of UVR, measured with an instrument having a cosine angular response. The measured irradiance (W/m2) and the radiant exposure (J/m2) should be averaged over the area of a circular measurement aperture not greater than 1 mm in diameter. A distinction is drawn between the ELVs for the skin, which are considered advisory due to the wide range of susceptibility to skin injury depending on skin type, and those for the eye, which should be treated as absolute limits. The values were developed by considering lightly pigmented populations (white Caucasian) having the greatest sensitivity and genetic predisposition to sunburn. ACGIH (2003) indicates that conditioned (tanned) individuals can tolerate skin exposure in excess of the exposure limits without erythemal effects but that conditioning may not protect individuals against skin cancer. For certain sources, the threshold for erythema may be exceeded in certain people without exceeding the exposure limit. In the United Kingdom, it has been recognized that there is a degree of uncertainty in the scientific evidence on which the guidelines are based and that they may not provide adequate protection against the long-term effects of exposure. Hence, it is currently recommended that exposure to artificial sources for nonelective purposes, which will include all occupational exposures, should comply with the ICNIRP ELVs, and in addition a balanced cautionary approach should be adopted to reduce exposures wherever possible (NRPB, 2002). The ELVs are not intended to apply to UVR exposure of photosensitive individuals, nor to individuals concomitantly exposed to photosensitizing agents. Similarly, they are not intended to protect neonates; separate weighting functions are available for aphakes (persons whose natural lens has been surgically removed) in the wavelength range 305–400 nm (ICNIRP, 1999; ACGIH, 2003). Many unfiltered or uncontrolled artificial optical radiation sources, particularly those used in industrial processes, have the potential to produce emissions that will exceed recommended ELVs under certain conditions. However, it is generally straightforward to adopt effective control measures that will reduce exposures to acceptably low levels. As for other hazards, it is appropriate to adopt a hierarchy of risk reduction, with avoidance of the hazard placed at the top, followed by controls that achieve collective protection in preference to personal protection. In general, although both engineering and administrative controls will provide collective protection, the former will be more robust and should be the primary method, with the latter used to control remaining risks. Personal protective equipment, including gloves, eyewear, and face shields, are acceptable control

1152

ULTRAVIOLET RADIATION

measures, but as they provide only personal protection, they should only be used as a last resort when risks cannot be adequately controlled by other means. It is relevant to note that the European Parliament has recently passed a Physical Agents Directive covering optical radiations, and this will require all member states of the European Union to introduce new legislation to restrict occupational exposures to artificial sources. The directive incorporates a set of limit values that are numerically equivalent to the ICNIRP MPEs for laser radiation and ELVs for noncoherent optical radiation. The effect of the directive will be to make these limits legally enforceable in the workplace and to require employers to introduce appropriate control measures wherever there is “any possibility that the limit values may be exceeded.” Moreover, employers will have to take into account of other relevant factors such as exposure to photosensitizing chemicals and clinical conditions that may increase the photosensitivity of particular individuals.

30.13 TECHNIQUES FOR EVALUATING ACTUAL OR POTENTIAL EXPOSURES The most accurate means of determining the biologically effective irradiance involves measuring the spectral distribution with a spectroradiometer, and then calculating the weighted integral. In some situations, it may be possible to estimate exposure using a broadband radiation detector whose sensitivity varies with wavelength according to the weighting function. However, great care is needed with this approach as deviations from the weighting function at wavelengths where it is strongly wavelength dependent can lead to errors of an order of magnitude or more. Personal monitoring may be the preferred approach for evaluation of individual exposure and can be used to assess the fraction and distribution of the daily radiant exposure (J=m2effective ) for people particularly at risk. The basic components of a spectroradiometer are shown in Fig. 30.8. Optical radiation is collected by the input optics, which should possess a 2p field of view and a cosine-weighted angular response. Two types of input optics achieve these requirements. A transmission diffuser at the entrance slit of the spectroradiometer can produce an approximate cosine response. Alternatively, an integrating sphere with a small entrance aperture and an internal diffuse coating, such as MgO or BaSO4, with a high UVR reflectance can produce a cosineweighted response. A second aperture in the integrating sphere provides input for the optical radiation via an entrance slit to a monochromator. The entrance slit of the monochromator is at the focal point of a collimating mirror, and optical radiation is reflected from the mirror as a parallel beam incident on a ruled diffraction grating. A second mirror collects optical radiation from the grating at a particular angle (hence wavelength) and focuses it onto the exit slit of the monochromator. Highperformance spectroradiometers, used for determining low irradiances of UVR, require low stray radiation levels and use a double grating monochromator. The emerging wavelength is altered by angular rotation of the diffraction grating. An appropriate detector, such as a photomultiplier tube or occasionally a photodiode, is mounted at the exit slit of the monochromator. The signal from the detector is integrated for a preselected time and then is transferred to a computer for storage and display. The wavelength drive and output integration are synchronized to provide a spectral scan of irradiance in equal wavelength intervals throughout a given optical spectrum. The optical radiation flux passing through the exit slit depends on factors such as angular dispersion and the square of the bandwidth. For a small entrance slit, the bandwidth (Dl) is

TECHNIQUES FOR EVALUATING ACTUAL OR POTENTIAL EXPOSURES

FIGURE 30.8

1153

Basic components of a scanning spectroradiometer.

equal to the width of the exit slit divided by the linear dispersion, and the intensity distribution across the exit slit corresponds to the wavelength distribution of the source. When the widths of the entrance and exit slits are comparable, a triangular distribution across the exit slit results, with the Dl defined at full-width, half-height. This can lead to inaccuracies in measurement, particularly when the spectral irradiance varies rapidly with wavelength, such as with the solar spectrum below 305 nm. When the monochromator slit widths are sufficiently narrow, the detected output signal S(l) is related to the measured spectral irradiance E(l) through the equation SðlÞ ¼ EðlÞRðlÞ;

where R(l) is the spectral responsivity, determined by measuring the response of the spectroradiometer to a source, such as a tungsten halogen or deuterium lamp, of known spectral irradiance and traceable to national standards. The biologically effective irradiance Eeff is obtained by summing over all wavelengths emitted by a source using the equation Eeff ¼ l SEðlÞBðlÞDlðW=m2effective Þ;

where B(l) is the appropriate biological weighting function at wavelength l, and Dl is the wavelength interval. Spectroradiometers are capable of high precision in the measurement of spectral irradiance, but they are expensive, and measurements are typically time consuming, making them unsuitable for rapidly changing sources such as welding arcs. A new generation of faster spectroradiometers has been developed in recent years. These typically use a single

1154

ULTRAVIOLET RADIATION

fixed grating to spread the available spectrum across a detector array. This means that all wavelength intervals can be measured simultaneously, and the instruments are small enough to be portable. However, these instruments tend to have lower resolution and are prone to ‘out-of-band’ responses. Nevertheless, when used with care they offer significant advantages for assessments in many ‘real-life’ situations where it would be impracticable to bring sources into a laboratory environment. Less precise estimates of E(l) or Eeff are achievable with less expensive broadband instruments, which integrate spectral irradiance over a range of wavelengths. Physical phenomena used in two measurement methods are either thermal, where energy absorption results in a measurable temperature change in a detector, or photoelectric, which involves a conversion of UVR into an electrical signal. Thermal detectors have a uniform response within their region of operation and are particularly useful for the absolute determination of irradiance and as such are often used as calibration instruments. There are two types of thermal detector for operation with UVR: (a) A thermopile, which depends on the Seebeck effect, whereby a voltage is generated when heat is applied to the junction of two dissimilar metals, (b) A pyroelectric detector, which depends on a voltage generated by a temperature change via a change in electrical polarization in a crystal, such as lithium tantalate. A fused silica window is used in a thermopile for UVR operation and the instrument will operate with a near-flat response over the wavelength range from 180 nm to 3.4 mm. Heat losses in a thermopile can result in nonlinearity in response, particularly at high irradiances (e.g., greater than 300 W/m2). Thermopiles should not be used for irradiances greater than 2000 W m 2. Pyroelectric detectors have a faster response than a thermopile, and a typical irradiance range from 10 4 to 106 W/m2. There is a range of devices employing photoelectric detectors for UVR measurement. Some of these devices provide a measurement of E(l) and some, with the addition of optical filters, provide an assessment of Eeff. The photoelectric detectors used in such devices include photomultiplier tubes, vacuum photodiodes, silicon photodiodes, and GaAsP photodiodes. Three commonly used devices that employ this technology for health hazard assessment and environmental monitoring in the UVR spectral region are the direct reading radiometer, the filtered UVR meter, and the Robertson-Berger (R-B) meter. The detecting head of a typical direct reading radiometer incorporates a quartz wide-angle diffuser, an interference filter, a blocking filter, and a ‘solar blind’ vacuum photodiode. Different detector heads are available for the assessment of different biological effects. Filtered UVR meters, which incorporate a diffuser and filters to provide measurements of global UVB at 306 nm and UVA at 360 nm, have been used in Sweden since 1989 (Wester, 1983). The R-B meter design of detector (Berger, 1976) is available commercially and is used as the erythemally weighted UVR detector elements in solar radiation measurement systems (Fig. 30.9). The detector is housed in an aluminum cylinder with a Vycor quartz dome to seal the input. UVR incident upon the dome passes through a UG11 black glass filter, which absorbs visible and infrared radiation. This is then wavelength shifted by a magnesium tungstate phosphor deposited on a Corning 4010 green filter, which absorbs UVR and the small amount of red light transmitted by the UG11 filter. The emission from the fluorescing

TECHNIQUES FOR EVALUATING ACTUAL OR POTENTIAL EXPOSURES

FIGURE 30.9

1155

HPA solar measurement system incorporating an R-B meter (from Dean et al., 1991).

phosphor in the wavelength range 400–600 nm is detected by a side window vacuum photodiode with an S4 bialkali photocathode. The relative spectral response of the R-B detector is an acceptable approximation to the spectral efficacy curve for UVR-induced erythema in human skin. Desiccants are used to control condensation and humidity in the housing. Temperature coefficients of between 0.3 and 0.8% per  C have been reported in the older design of R-B (model 500), mainly due to the temperature sensitivity of the phosphor in the temperature range 20–40 C. Broadband instruments employing diffusers and filters for measurement in the UVA are also commercially available. Personal monitoring may be the preferred measurement method to adequately evaluate individual exposures and to establish appropriate protective measures. By this method, the fraction and distribution of the daily radiant exposure (J=m2effective ) from an optical radiation source, such as the sun, can be assessed for people particularly at risk. Sometimes, the principles (such as the weighted response characteristic) of the broadband radiation detector are employed in a personal detector, but there is a range of other physical, chemical, and biological options (Moseley, 1988). The biologically effective UVR dose applied to the body over a time interval T (s) is defined as D ¼ t Sl SEðl; tÞBðlÞDlDt;

where E(l,t) is the spectral irradiance, and B(l) is the appropriate biological weighting factor at wavelength l. Changes in the optical properties of photosensitive films to incident UVR can be used in UVR personal and environmental dosimetry. These dosemeters provide a simple means of integrating UVR exposure continuously and because they are cheap and compact can be used for environmental measurements at many locations inaccessible to bulky instrumentation. The UVR absorption of polysulphone film increases when exposed to UVR of wavelength less than 330 nm. The increased absorbance is proportional to the UVR dose. The spectral

1156

ULTRAVIOLET RADIATION

response characteristics of 1 mm thick film matches closely the erythema action spectrum for wavelengths in the range 300–315 nm (Diffey, 1982). The spectral response of the more commonly used 40 mm thick film deviates from the erythema action curve, which can lead to significant errors if uncorrected. CR39 (allyl diglycol carbonate) plastic has a response that approximates the reference erythema action spectrum (Wong et al., 1992). After exposure to UVR, the plastic is etched in a concentrated caustic solution at an elevated temperature of 80 C for 3 h and then rinsed and dried. The change in absorbance in the exposed dosemeter is measured. The sensitivity of the dosemeter can be changed by the concentration of the etching solution and by the wavelength at which the absorbance is measured (400 nm for D less than 3000 J=m2effective and at 700 nm for higher doses). Overall, measurement uncertainties of 20% have been reported. A number of other personal UVR devices, including personal sun alarms, UVR-sensitive stick-on tapes, sun sensors, timers, and small monitors, can be bought at relatively low cost. Many of these are very poor indicators of UVR dose, have a restricted angular response, and are no substitute for common sense in the sun. Other devices, such as photoluminescent and thermoluminescent dosemeters and diazo film, have been reported (WHO, 1979).

30.14 SUMMARY The major source of personal exposure to UVR is the sun although exposure from other optical sources such as sunbeds, arcs, and medical sources may be significant. The major target organs are the eyes and the skin. The ocular effects of acute overexposure include photokeratitis and photoconjunctivitis, inflammation of the cornea and conjunctiva, respectively, solar retinitis, and photochemical damage to the retina. Chronic overexposure may contribute to corneal injuries, such as climatic droplet keratopathy and pterygium, lens opacities or cataracts, and age-related macular degeneration, a retinal condition that is a major cause of blindness in the developed world. Acute overexposure of the skin may result in sunburn, characterized by erythema, edema and, in severe cases, blistering; the response to skin damage includes skin thickening and increased pigmentation. Chronic overexposure results in photoaging of exposed skin and the development of solar keratoses, benign proliferations of epidermal keratinocytes that are considered to be premalignant lesions; chronic overexposure to UVA may result in increased skin fragility. The most serious health effect of chronic overexposure is the development of skin cancer. There are two main types, the nonmelanoma skin cancers, comprising basal cell carcinomas and squamous cell carcinomas, and the cutaneous malignant melanomas. Cumulative dose is a major factor determining risk for the nonmelanoma skin cancers. Although much less common than nonmelanoma skin cancers, malignant melanoma is the main cause of skin cancer death, particularly in the young. The epidemiological evidence suggests that while sun exposure is causally related to melanoma risk, the relationship is complex; both intense intermittent exposure and exposure in early life are implicated in melanoma aetiology. Phenotype is an important risk factor for the development of both nonmelanoma and melanoma skin cancers, with light skin, red or blond hair, blue eyes, a tendency to burn easily and to tan poorly all associated with increased risk. A tendency to freckle, large numbers of naevi, and atypical naevi are all additional risk factors for malignant melanoma. Guidelines exist that provide exposure limits to UVR, and these are based on avoidance of acute damage to the skin and eyes. However, the numerical exposure limit values will not

REFERENCES

1157

necessarily provide adequate protection against long-term damage, as there is currently insufficient information to derive appropriate limits. In the United Kingdom, it is recommended that for occupational exposures to artificial sources the exposure limit values should be observed and additionally exposures should be kept as low as reasonably practical. Further guidance in respect of occupational sun exposure, and elective exposures to both solar and artificial UVR has also been provided. Guidelines are not intended to provide protection from medical exposures as the risks and benefits from these are a matter of clinical judgment. Since it is relatively easy to shield UVR, limitation of occupational exposure from artificial sources is often fairly straightforward and can be achieved by simple engineering controls. Where this is not possible, reliance should be placed on administrative controls, with personal protective equipment only used as a last resort. Good education policy, involving easily understandable advice, is the key to reducing people’s overall exposure to solar UVR. This combined with a comprehensive range of effective personal protective measures and reliable measurement data pertaining to global, occupational, and personal exposures are fundamental to addressing the health effect issues raised by exposure to solar UVR. In addition, the development of inexpensive and reliable integrating personal dosimetry methods, which allow UVR baseline measurement data to be compared with UVR doses at specific body locations, is highly desirable.

REFERENCES ACGIH (2003) Threshold Limit Values for Chemical Substances and Physical Agents and Biological Exposure Indices.Cincinnati, OH:American Conference of Governmental Industrial Hygienists. AGNIR (1995) Health Effects from Ultraviolet Radiation, Docs. NRPB, 6 (2) AGNIR (2002) Health Effects from Ultraviolet Radiation, Docs. NRPB, 13(1), 5-276. Anders A, Altheide H-J, Kn€almann M, Tronnier H (1995) Action spectrum for erythema in humans investigated with dye lasers. Photochem. Photobiol. 61:200–205. Berger DS (1976) The sunburning ultraviolet meter: design and performance. Photochem. Photobiol. 24:587–593. Birmingham DJ, Key MM, Tubich GE, Perone VB (1961) Phototoxic bullae among celery harvesters. Arch. Dermatol. 83:73–87. Bissett DL, Hannon, DP, Orr, TV (1989) Wavelength dependence of histological, physical, and visible changes in chronically UV-irradiated hairless mouse skin. Photochem. Photobiol. 50:763–769. Boettner EA, Wolter JR (1962) Transmission of the ocular media. Invest. Ophthalmol. 1:776–783. Burns FJ, Uddin AN, Wu F, Nadas A, Rossman TG (2004) Arsenic-induced enhancement of ultraviolet radiation carcinogenesis in mouse skin: a dose–response study. Environ. Health Perspect. 112:599–603. CIE (1970) International lighting vocabulary, Paris, Bureau Central de la Commission Internationale de l’Eclairage 3,51. CIE (1996) Standard erythema dose. Commission Internationale de l’Eclairage draft standard DS 007.1/E. Chadwick CA, Potten CS, Nikaido O, Matsunaga T, Proby C, Young AR (1995) The detection of cyclobutane thymine dimers, (6–4) photolesions and the Dewar photoisomers in sections of UVirradiated human skin using specific antibodies, and the demonstration of depth penetration effects. J. Photochem. Photobiol. B 28:163–170.

1158

ULTRAVIOLET RADIATION

Chaney EK, Sliney DH (2005) Re-evaluation of the ultraviolet hazard action spectrum—the impact of spectral bandwidth. Health Phys. 89(4):322–332. Cole CA, Forbes PD, Davies RE, Urbach F (1986a) Effect of Indoor Lighting on Normal Skin. New York: New York Academy of Science. Cole CC, Forbes PD, Davies RE (1986b) An action spectrum for UV photocarcinogenesis. Photochem. Photobiol. 43:275–284. Cridland NA, Saunders RD (1994) Cellular and Molecular Effects of UVA and UVB. NRPB-R269. London:HMSO. Crow KD (1961) Photosensitivity due to pitch. Br. J. Dermatol. 73:220–232. Cullen AP, Perera SC (1994) Sunlight and human conjunctival action spectrum. SPEI Proceedings 2134B:24–30. Davidson T, Kluz T, Burns F, Rossman T, Zhang Q, Uddin A, Nadas A, Costa M (2004) Exposure to chromium (IV) in the drinking water increases susceptibility to UV-induced skin tumors in hairless mice. Toxicol. Appl. Pharmacol. 196:431–437. Davson H, editor. (1967)Vegetative Physiology and Biochemistry. The Eye, Vol. 1. New York: Academic Press. Dean SF, Rawlinson AI, McKinlay AF, Pearson AJ, Whillock MJ, Driscoll CMH (1991) NRPB solar radiation measurement system. Radiol. Protec. Bull. 124:6. de Gruijl FR, Sterenborg HJCM, Forbes PD, Davies RE, Cole C, Kelfkens G, van Weelden H, Slaper H, van der Leun JC (1993) Wavelength dependence of skin cancer induction by ultraviolet irradiation of albino hairless mice. Cancer Res. 53:53–60. de Gruijl FR, van der Leun JC (1992) Action spectra for carcinogenesis. In: Urbach F, editor. Biological Responses to Ultraviolet A Radiation. Proceedings of Second International Conference: Biological Effects of UVA Radiation, San Antonio, June, 1991. Overland Park: Valdenmar, pp.91– 97. Diffey BL (1982) Ultraviolet radiation in medicine, Medical Physics Handbook 11, Adam Hilger. Diffey BL (1994) Observed and predicted minimal erythema doses: a comparative study. Photochem. Photobiol. 60:380–382. Douki T, Perdiz D, Grof P, et al., (1999) Oxidation of guanine in cellular DNA by solar UV radiation: biological role. Photochem. Photobiol. 70 (2):184–190. Driscoll CMH (1992) Solar UV trends and distributions. Radiol. Protec. Bull. 137:7–13. Elliot DJ (1995) Ultraviolet laser technology and applications. Academic Press. Epe B (1991) Genotoxicity of singlet oxygen. Chem. Biol. Interact. 80:239–260. Epstein JH (1965) Comparison of the carcinogenic and cocarcinogenic effects of ultraviolet light on hairless mice. J. Natl. Cancer Inst. 34:741–745. Epstein JH (1988) Photocarcinogenesis promotion studies with benzoyl peroxide (BPO) and croton oil. J. Invest. Dermatol. 91:114–116. Epstein JH, Epstein WL (1962) Cocarcinogenic effect of ultraviolet light on DMBA tumour initiation in albino mice. J. Invest. Dermatol. 39:455–460. Epstein JH, Roth HL (1968) Experimental ultraviolet light carcinogenesis. J. Invest. Dermatol. 50:387–389. Epstein JH, Epstein WL, Nakai T (1967) Production of melanomas from DMBA-induced “blue nevi” in hairless mice with ultraviolet light. J. Natl. Cancer Inst. 39:19–30. FDA (2008). Food and Drug Administration Centre for Devices and Radiological Health. Performance Standards for Light Emitting Products: High Pressure Mercury Vapour Discharge Lamps. Code of Federal Regulations, Title 21, Volume 8, Section 1040.30. Fisher GJ, Datta SC, Talwar HS, Wang Z-Q, Varni J, Kang S, Voorhees JJ (1996) Molecular basis of sun-induced premature skin ageing and retinoid antagonism. Nature 379:335–339.

REFERENCES

1159

Fourtanier A, Berrebi C (1989) Minature pig as an animal model to study photoaging. Photochem. Photobiol. 50:771–784. Frederick JE, Snell HE, Haywood EK (1989) Solar ultraviolet radiation at the earth’s surface. Photochem. Photobiol. 50 (8):443–450. Freeman SE, Hacham H, Gange RW, Maytum DJ, Sutherland JC, Sutherland BM (1989) Wavelength dependence of pyrimidine dimer formation in DNA of human skin irradiated in situ with ultraviolet light. Proc. Natl. Acad. Sci. USA 86:5605–5609. Gange WR, Parrish JA (1983)Acute effects of ultraviolet radiation upon the skin. In: Parrish JA, Kripke ML, Morison WL, editors. Photoimmunology. New York: Plenum Press. Gray RH, Johnson GJ, Freedman A (1992) Climatic droplet keratopathy. Surv. Ophthalmol. 36:241– 253. Harber LC, Levine GM (1969) Photosensitivity dermatitis from household products. General Practice 39:95–100. Hawk JLM (1984) Photosensitising agents used in the United Kingdom. Clin. Exp. Dermatol. 9:300– 302. Hjorth N, M€oller H (1976) Phototoxic textile dermatitis (“bikini dermatitis”) Arch. Dermatology 112:1445–1447. Hruza LL, Pentland AP (1993) Mechanisms of UV-induced inflammation. J. Invest. Dermatol. 100:35S–41S. Husein Z, Pathak MA, Flotte T, Wick MM (1991) Role of ultraviolet light in the induction of melanocytic tumours in hairless mice following 7, 12–dimethylbenz-(a)anthracene application and ultraviolet irradiation. Cancer Res. 51:4964–4970. IARC (1992) International Agency for Research on Cancer. IARC Monographs on the evaluation of carcinogenic risks to humans, Volume 55, Solar and Ultraviolet Radiation. Lyon: IARC. ICNIRP (1999) Guidelines on limiting exposure to non-ionizing radiation. International Commission on Non-ionizing Radiation Protection, ICNIRP 7/99.207–226. Iverson OH (1986) Carcinogenesis studies with benzoyl peroxide (Panoxyl gel 5%). J. Invest. Dermatol. 86:442–448. Iverson OH (1988) Skin tumourigenesis and carcinogenesis studies with 7,12-dimethylbenz(a) anthracene, ultraviolet light, benzoyl peroxide (Panoxyl gel 5%) and ointment gel. Carcinogenesis 9:803–809. Karlsson P, Boeryd B, Sander B Westermark P, and Rosdahl I (1998). Increasing incidence of cutaneous malignant melanoma in children and adolescents 12–19 years of age in Sweden 1973– 92. Acta Derm Venereol (Stockh), 78, 289–92. Kelfkens G, de Gruijl FR, van der Leun JC (1992) Carcinogenesis by short- and long wave ultravioletA: papilloma versus squamous cell carcinomas. In: Urback F, editor. Biological Responses to Ultraviolet A Radiation. Proceedings of Second International Conference: Biological Effects of UVA Radiation, San Antonio, June, 1991.Overland Park: Valdenmar. pp.285–294. Kielbassa C, Rosa L, Epe B (1997) Wavelength dependence of oxidative DNA damage induced by UV and visible light. Carcinogenesis 18:811–816. Kricker A, Armstrong BK, English DR, Heenan PJ (1991) Pigmentary and cutaneous risk factors for non-melanocytic skin cancer: a case-control study. Int. J. Cancer 48:650–662. Kripke ML (1993)Immunosuppressive action of UV radiation. In: De Gruijl FR, editor. The Dark Side of Sunlight. Utrecht University, pp.77–86. Krutman J, Elmets CA, editors (1995) Photoimmunology. Oxford: Blackwell Science. Kvam E, Tyrrell RM (1997a) Artificial background and induced levels of oxidative base damage in DNA from human cells. Carcinogenesis 18(11):2281–2283.

1160

ULTRAVIOLET RADIATION

Kvam E, Tyrrell RM (1997b) Induction of oxidative base damage in human skin cells by UV and near visible radiation. Carcinogenesis 18(12):2379–2384. Lerman S and Borkman RF (1976). Spectroscopic evaluation and classification of the normal aging and cataractous lens. Ophthalmic Res. 8, 335–53. Ley RD, Peak MJ, Lyon LL (1983) Induction of pyrimidine dimers in epidermal DNA of hairless mice by UVB: an action spectrum. J. Invest. Dermatol. 80:188–191. Logani MK, Sambuco PD, Forbes PD, Davies RE (1984) Skin-tumour promoting activity of methyl ethyl ketone peroxide—a potent lipid periodizing agent. Food Chem. Toxicol. 22:879–882. Lowe NJ, Meyers DP, Wieder JM, Luftman D, Borget T, Lehman MD, Johnson AW, Scott IR (1995) Low doses of repetitive ultraviolet A induce morphologic changes in human skin. J. Invest. Dermatol. 105:739–743. Mainster MA (1978) Spectral transmittance of intraocular lenses and retinal damage from intense light sources. Am. J. Ophthalmol. 85:167–170. Marks R, Foley P, Goodman G, Hage BH, Slewood TS (1986) Spontaneous remission of solar keratoses: the case for conservative management. Br. J. Dermatol. 115:649–655. McKinlay AF, Diffey BL (1987) A reference action spectrum for ultraviolet induced erythema in human skin. CIE J. 6(1):17–22. McKinlay AF, Harlen F, Whillock MJ (1988) Hazards of Optical Radiation. A Guide to Sources, Uses and Safety. Bristol and Philadelphia: Adam Hilger. McKinlay AF, Whillock MJ Meulemans CCE (1989) Ultraviolet radiation and blue light emissions from spotlights incorporating tungsten halogen lamps. Chilton: National Radiological Protection Board (NRPB) Report R-228. Moan J, Peak MJ (1989) Effects of UV radiation on cells. J. Photochem. Photobiol. B 4:21–34. Moseley H (1988) Non-Ionising Radiation, Medical Physics Handbook 18, Chapter 8. Bristol:Adam Hilger. Noonan FP, De Fabo EC (1993) UV-induced immunosuppression. In: Young AR, Bj€orn LO, Moan J, Nultch W, editors. Environmental UV Photobiology. New York:Plenum Press, pp.113– 148. NRPB (2002) Advice on protection against ultraviolet radiation, Docs. NRPB, 13 (3): 1-35. Oikarinen A (1990) The aging of skin: chronoageing versus photoageing. Photodermatol. Photoimmunol. Photomed. 7:3–4. Oikarinen A, Peltonen J, Kallioinen M (1991) Ultraviolet radiation in skin ageing and carcinogenesis: the role of retinoids for treatment and prevention. Ann. Med. 23:497–505. Parrish JA, Jaenicke KF, Anderson RR (1982) Erythema and melanogenesis action spectra of normal human skin. Photochem. Photobiol. 36:187–191. Pathak MA, Fanselow DL (1983) Photobiology of melanin pigmentation: dose/response of skin to sunlight and its content. J. Am. Acad. Dermatol. 9:724–733. Piette J (1991) Biological consequences associated with DNA oxidation mediated by singlet oxygen. J. Photochem. Photobiol. B 11:241–260. Pitts DG, Cullen AP, Hacker PD (1977) Ocular effects of ultraviolet radiation from 295 to 365 nm. Invest. Ophthalmol. Vis. Sci. 16:932–939. Pound AW (1970) Induced cell proliferation and the initiation of skin tumour formation in mice by ultraviolet light. Pathology 2:269–275. Romanchuk KG, Pollak V, Schneider RJ (1978) Retinal burn from a welding arc. Can. J. Ophthalmol. 13:120–122. Rossman TG, Uddin AN, Burns FJ, Bosland MC (2001) Asenite is a cocarcinogen with solar ultraviolet radiation for mouse skin: an animal model for arsenic carcinogenesis. Toxicol. Appl. Pharmacol. 176:64–71.

REFERENCES

1161

Rossman TG, Uddin AN, Burns FJ (2004) Evidence that arsenite acts as a cocarcinogen in skin cancer. Toxicol. Appl. Pharmacol. 198:394–404. Saunders, RD, Cridland NA, Kowalczuk CI (1997) Animal and Human Responses to UVA and UVB. NRPB-R297. London:HMSO. Seddon JM (2000). Epidemiology of age-related macular degeneration. In: Albert DM, Jakobiec FA, editors. Principles and Practice of Ophthalmology, 2nd edn. Philadelphia, WB Saunders, pp 1039– 1050. Setlow, RB, Grist E, Thompson KWoodhead A (1993) Wavelengths effective in the induction of malignant melanoma. Proc. Natl. Acad. Sci. USA 90:6666–6670. Sliney DH (1986) Physical factors in cataractogenesis: ambient ultraviolet radiation and temperature. Invest. Ophthalmol. Vis. Sci. 27:781–790. Sliney DH (1987) Estimating the solar ultraviolet radiation exposure to an intraocular lens implant. J. Cataract Refract. Surg. 13(5):296–301. Sliney DH, Wolbarsht M (1980) Safety with Lasers and Other Optical Sources: A Comprehensive Handbook. New York: Plenum Press. Soballe PW, Montone KT, Satyamoorthy K, Nesbit M, Herlyn M (1996) Carcinogenesis in human skin grafted to SCID mice. Cancer Res. 56:757–764. Stenb€ack F (1975) Ultraviolet light irradiation as initiating agent in skin tumour formation by the twostage method. Eur. J. Cancer 11:241–246. Swerdlow AJ, dos Santos Silva I, Reid A, Qiao Z, Brewster DH, and Arrundale J (1998). Trends in cancer incidence and mortality in Scotland: description and possible explanations. Br. J. Cancer, 77, (Suppl 3), 1–54. Swerdlow AJ, dos Santos Silva I, and Doll R (2001). Cancer Incidence and Mortality in England and Wales: Trends and Risk Factors. Oxford University Press. Taylor HR, West SK, Rosenthal FS, Munoz B, Newland HS, Emmett EA (1989) Corneal changes associated with chronic UV irradiation. Arch. Ophthalmol. 107:1481–1484. Tyrrell RM, Keyse SM (1990) New trends in photobiology (invited review): the interaction of UVA radiation with cultured cells. J. Photochem. Photobiol. B 4:349–361. Ullrich SE (1995) Modulation of immunity by ultraviolet radiation: key effects on antigen presentation. J. Invest. Dermatol. 105:30S–36S. UNEP (1987) The ozone layer. United Nations Environment Programme. UNEP/GEMS Environmental Library No 2. Nairobi: UNEP. van Kranen HJ, de Laat A, van de Ven J, Wester PW, de Vries A, Berq RJ, van Kreijl CF and de Gruijl FR (1997). Low incidence of p53 mutations in UVA (365 nm)-induced skin tumours in hairless mice. Cancer Res. 57, 1238–1240. Vermeer M, Hurks H (1994) J. Photochem. Photobiol B 24:149–154. West SK, Rosenthal FS, Bressler SB, Munoz SB, Fine SL, Taylor HR (1989) Exposure to sunlight and other risk factors for age related macular degeneration. Arch. Ophthalmol. 107:875–879. Wester U (1983) Solar ultraviolet radiation—A method for measuring and monitoring, Report RI 1983-02, Swedish Radiation Protection Institute, Box 60204. Whillock MJ, Clark I, McKinlay AF, Todd C, Mundy S (1988) Ultraviolet radiation levels associated with the use of fluorescent general lighting, UVA and UVB lamps in the workplace and home. Chilton: National Radiological Protection Board (NRPB) Report R-221. Whillock MJ, McKinlay AF, Kemmlert J, Forsgren RG (1990) Ultraviolet radiation emissions from miniature (compact) fluorescent lamps. Lighting Res. Technol. 23(3):125–128. WHO (1979) Ultraviolet Radiation. Environmental Health Criteria 14. Geneva: World Health Organization.

1162

ULTRAVIOLET RADIATION

WHO (1994) Ultraviolet Radiation. Environmental Health Criteria 160. Geneva: World Health Organization. Wong CF, Fleming RA, Carter SJ, Ring IT, Vishvakarman D (1992) Measurement of human exposure to ultraviolet-B solar radiation using a CR-39 dosimeter. Health Phys. 63:457. Young RW (1988) Solar radiation and age-related macular degeneration. Surv. Ophthalmol. 32:252– 269.

INDEX

2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD), 633 Absorption, 816 Acceptable daily intake, 203 Accumulation, 21 Accumulation mode, 5, 319–320 Acidic aerosols, 958, 971 Administrative controls, 1014 Adversity, 871 Aerodynamic diameter, 4 Aerosol dispersion, 889 Aerosols, 2 Aflatoxins, 224 Ah receptor, 643 Air contaminants, 2 Air quality standards, 917, 989 Airway hyperresponsiveness, 975 Airway inflammation, 885–889 Airway permeability, 885, 900–901 Airway reactivity, 274, 278–279, 884–885, 907 Allergic airway disorders, 593 Allergic responses, 591 Alveolar macrophages, 832 Alveolus, 280 Ambient air, 11, 328 Amosite fibers, 419 Amphibole fibers, 405

Anemia, 512 Animal drug residues, 215 Anthophyllite fibers, 428 Asbestiform nature (fibers), 395 Asbestosis, 420 Athletic performance, 882, 900 Becquerel, 1022 Behavioral effects, 513 BEIR IV model, 1110 Bioavailability, 562 Biological mechanisms, 1135 Biological plausibility, 712 Biomarkers, 380, 664, 781 Biomonitoring, 112 Biphenyls, 634 Bladder cancer, 729 Brain tumors, 715 Bronchiole, 14 Bronchoconstrictive effects, 961 Bronchus, 13 Cancer, 561 Cancer hazard, 573–574, Cancer risk, 590–591 Carbamates, 943 Carcinogenesis, 1143

Environmental Toxicants, Third Edition Edited by Morton Lippmann Copyright Ó 2009 John Wiley & Sons, Inc.

1163

1164

INDEX

Carcinogenicity, 289, 482, 539, 641 Carcinogenicity bioassasys, 563 Cardiac function, 336 Cardiopulmonary effects, 512 Cardiovascular disease, 502, 515 Cardiovascular effects, 508 Cardiovascular responses, 95 Case-control, 582 Central nervous system, 481 Chemical exposure, 8 Childhood asthma, 719 Childhood cancer, 720, 960 Children’s cognitive, language, and learning skills, 1078 Cholinesterase inhibitors, 943 Chronic bronchitis, 985 Chronic exposures, 337 Chronic lung disease, 503 Chronic obstructive pulmonary disease (COPD), 513 Chrysotile fibers, 396 Clearance, 8, 831 Cleavage fragments, 396–398 Clinical approaches, 80 Clinical signs, 675 Coal smoke, 333–334 Cognition and behavior, 714 Coherence, 342 Cohort studies, 337, 582, 588–589 Color additives, 214 Concentration units, 2 Condensation nuclei (CN), 4 Consumer products, 1055 Contaminants, 233 Coronary heart disease, 730 Corrective action, 66 Counseling of patients, 97 Critical fiber dimensions, 420 Critical fiber parameters, 427 Critical fiber properties, 443 Crocidolite fibers, 403 Curie, 1022 Decommissioning and decontamination, 1037 Defense mechanisms, 279 Deposition, 8, 971 Dermal absorption, 771 Dermal effects, 19–20 Development of chronic disease, 912 Developmental changes, 685 Developmental toxicity, 642 Dibenzofurans (CDFs), 634

Dietary supplements, 228 Diethylstilbestrol (DES), 682 Diffusion, 19 Disease pathogenesis, 30 Disinfection, 122 Disinfection by-products, 122 Disintegration, 404–405 Disposition, 462, 470 Dissolution, 404 Dissolved contaminants, 6 Dissolved gases, 6 Dissolved solids, 6 Distribution, 816 DNA-Protein Cross-Links, 281 Dose, 8 Dose-response, 438, 482 Dose-response analysis, 43 Dosimetric models, 21–22 Dosimetry, 7–8, 500–501, 826, 846, 848, 875, 961 Drinking water disinfection, 126 Dust, 4 Dustfall, 5 Effects on hearing, 1075 Effects on the immune system, 515 Electrostatic deposition, 15 Elimination, 817 EM fields, 1001 Emission standards, 559 Endocrine activity, 662 Endocrine disruptors, 645 Endocrine/reproductive toxicity, 949 Engineering/installation/design/controls, 1011 Environmental contamination, 1029 Environmental exposure, 940 Environmental lung disease, 81 Environmental tobacco smoke (ETS), 705 Epidemiological approaches, 90 Epidemiological studies, 1012 Epidemiology, 827, 1100 Erionite fibers, 398 Essentiality, 375, 529 Estimated daily intake, 208 Exacerbation of asthma and chronic bronchitis, 983 Excess mortality, 342 Exercise tolerance, 512 Exposure assessment, 44, 87, 93, 485, 710, 825 Exposure guidelines, 356, 543, 1150 Exposure limits, 826 Exposure surrogates, 8

INDEX

Exposure-response relationships, 25, 376 Exposures, 531, 825, 848, 1038, 1053, 1190, 1123, 1152 Exposures to radiation workers, 1038 Fallout, 1048 Fetal development, 514 Fetal effects, 711–712 Fetal growth, 712–713 Fiber deposition, 402 Fiber dissolution, 407–412 Fiber dosimetry, 443 Fibrogenesis, 422 Fine particulate matter, 317 Fog, 4 Food additives, 210 Food contaminants, 7 Food-related health risks, 198 Food supplements, 232 Fume, 4 Fumigants, 948 Fungicides, 947 Gases and vapors, 2 Gastrointestinal absorption, 766, 769 Gastrointestinal tract, 18 Genotoxicity, 540 GRAS substances, 209 Ground water, 1099 Growth dysregulators, 645 Guidelines, 111, 219, 228, 251, 356, 543, 917, 989, 1001, 1150 Haloacetates, 151 Haloacetonitriles, 153 Hazard control strategies, 109 Hazard identification, 42, 482 Haze, 4 Health, 23 Health effects, 23 Health surveillance, 113 Hematological effects, 477 Herbicides, 945 High-level nuclear waste, 1033 Histology, 673, 975 Human genome, 29 Immune responses, 606 Immune system effects, 1146 Immunotoxicity, 536, 643 Impaction, 15 Indoor radon, 1097

1165

Industrial hygiene, 110 Infant mortality, 343 Inflammation, 596 Ingestion, 18 Inhalation, 416 Inhalation studies, 347, 349 Insecticides, 943 Instillation, 415 Integrated assessment, 71 Interception, 15 Intervention, 108 Intraperitoneal injection, 415 Involuntary smoking, 722 Ionizing radiation, 1021 Kinetics, 374, 774 Kinetics and metabolism, 816 Larynx, 14 Lead poisoning, 757 Leukemia, 712 Life cycle of a chemical, 108 Lifespan shortening, 338 Light scatter, 5 Lithium, 676 Longevity, 912 Low birth weight (LBW), 343, 712 Low-level radioactive wastes (LLW), 1035 Lung, 13 Lung cancer, 422, 586, 589, 722, 1089 Lung cancer models, 1107 Lung cancer risk, 574 Lung dosimetry, 1104 Lung function, 507, 734 Lung growth and development, 720 Lung infectivity, 904 Lung structure, 911 Lymphomas, 716 Macroingredients, 215 Malignant melanoma, 1140 Markers of exposure, 709 Mechanisms of action, 640 Mechanisms of toxicity, 471 Medical exposures, 1050 Mental health, 1077 Mesothelioma, 422, 432 Metabolism, 462 Middle ear disease in children, 720 Mineralogy, 395 Mist, 4 Molecular dosimetry, 265

1166

INDEX

Morbidity, 24, 335, 336, 894, 963, 967, 969 Mortality, 24, 329, 334, 335, 827, 892, 957, 963 Mucociliary transport, 831 Mucociliary particle clearance, 275, 276, 983– 984 Multiple chemical sensitivity, 246 Mutagenicity, 289, 540 Mycotoxins, 224 Nasopharynx, 13 Naturally occurring radioactive material (NORM), 1057 Nature of risk, 39 Nematocides, 948 Neoplastic effects, 478 Neurotoxic effects, 785 Nicotine, 706 Nitrosamines, 164 Noise as a stressor, 1076 Nuclear fuel cycle, 1025 Nuclear weapons complex, 1043 Nutritional status, 767 Occupation health, 111 Occupational exposure, 940 Occupational exposure limits (OELs), 1001 Ocular damage, 1147 Organic mercury, 819 Organochlorines, 945 Organophosphates, 943 Outcome assessment, 94 Overload, 406 Panel studies, 969 Particle clearance, 889, 972, 974, 978 Pathogenesis of chronic bronchitis, 983 Pathophysiological responses, 80 Pathways for human exposure, 1122 PCBs and congeners, 217, 634, 678 Pediatric exposures, 941 Perchlorates, 677 Perinatal effects, 514 Pesticides, 222, 937 Pharmacokinetics, 639 Photosensitizers, 1149 PM10, 317 Polychlorinated biphenyls (PCBs), 677 Polychlorinated dibenzo-P-dioxin (CDDs), 634 Polyhalogenated aromatic hydrocarbons (PHAHs), 633 Population exposures, 326 Populations of concern, 266, 876

Pregnancy outcomes, 507 Prevention, 820 Product stewardship, 114 Protective measures, 1012 Pulmonary function testing, 84 Pyrethroids, 945 Radiation dose, 1092 Radioactivity, 1021 Range controls, 1014 Reducing risk, 61 Reprocessing, 1036 Reproductive and developmental effects, 139, 481 Reproductive toxicity, 676 Respiratory function, 272, 279, 597, 844, 871, 878, 905, 970, 976, 983 Respiratory infection, 594 Respiratory symptoms, 597, 718, 704–705 Respiratory tract, 13 Respiratory tract biochemistry, 837 Respiratory tract defenses, 831 Respiratory tract illnesses, 716 Responsible care, 117 Responsiveness, 890, 901 Retention, 8, 404 Risk assessment, 56, 206, 482, 811 Risk characterization, 44, 486, 649 Risk communication, 44 Risk factor, 343 Risk management, 56 Risk reduction, 49 Sedimentation, 12 Sick building syndrome, 241, 245 Site remediation, 65 Size-classified fibers, 415 Skin cancer, 1140 Skin damage, 1122 Sleep interference, 1077 Smog, 4 Smoke, 4 Sound, 1071 Sources, 212, 242, 258, 318, 371, 459, 499, 531, 558, 634, 762, 812, 823, 879, 951, 957, 1024, 1059, 1104, 1124, 1128 Standards, 356, 371, 503, 543, 917, 988, 1005, 1150 Standards development, 1005 Sudden infant death syndrome (SIDS), 712 Susceptibility, 28, 380 Suspended particles, 6 Symptomatic responses, 274, 882

INDEX

Synthetic vitreous fibers (SVF), 396, 442 Systemic effects, 818 Target tissues, 8, 21 Teratogenic effects, 539 Thoracic particulate matter, 317 Time-series, 335 Tobacco smoke, 705 Total suspended particulate matter (TSP), 317 Toxic equivalency factors, 647 Toxicity test requirements, 203 Toxicology, 831, 846, 848

Trachea, 13 Transgenerational effects, 682 Translocation, 16, 404–405 Transportation, 1036 Treatment, 796 Tremolite fibers, 424, 432 Trihalomethanes, 150 Uptake and distribution, 534 Volatile organic compounds (VOCs), 241 Working level (WL), 1091

1167